Sie sind auf Seite 1von 325

Congestion Management based on Power Flow Contribution Factors

Vinod Kumar Pal, 2Ashwani Kumar


1,2
1

NIT Kurukshetra

fvinod.k.pal@rediffmail.com 2 ashwa_ks@yahoo.co.in

Abstract In this paper, an application of proportional sharing principle and power flow comparison method have been presented for congestion management deciding rescheduling of generators for congestion management. Contribution formulae based approach has been implemented and Power flow comparison method and contribution formulae have been presented for congestion management studies. The comparison of PFC and CF methods has been tested on IEEE-57 bus test system for congestion management. The comparison of PFC and CF based approaches has been implemented for 6-bus test system, IEEE-57 bus test system and 25-bus test system for congestion management. Keywords Congestion management, contribution formulae, proportional sharing principle, power flow comparison method.

I. INTRODUCTION Restructuring of the power industry aims at abolishing the monopoly in the generation and trading sectors, thereby, introducing competition at various levels wherever it is possible. Generating companies may enter into contracts to supply the generated power to the power dealers/distributors or bulk consumers or sell the power in a pool in which the power brokers and customers also participate. In a power-exchange, the buyers can bid for their demands along with their willingness to pay. Power generation and trading will, thus, become free from the conventional regulations and become competitive. According to Phillipson and Willis [1], Deregulation is a restructuring of the rules and economic incentives that governing authority sets up to control and drive the electric power industry. In a restructured electricity market environment, when the producers and consumers of electric energy desire to produce and consume in amounts that would cause the transmission network to operate at or beyond one or more transfer limits, the system is said to be congested [2]. The congestion in the system can not be allowed to persist for a long time, as it can cause sudden rise in the electricity price and threaten system security and reliability. Congestion management is one of the most challenging tasks of the SO in the deregulated environment. In different types of market, the method of tackling the transmission congestion differs. There are three different ways to tackle the network congestion: Price Area Congestion Management

Available Transfer Capability (ATC) based Congestion Management Optimal Power Flow (OPF) based Congestion Management The first method is used in Nordic pool; the second one in US and the third is employed in UK. In Nordic pool, which consists of Norway, Sweden, Denmark, and Finland, when congestion is predicted, the system operator declares that the system is split into price areas at the predicted congestion bottlenecks. Spot market bidders must submit separate bids for each price area in which they have generation or loads. In case of no congestion, the market will settle at one price and in case of congestion, the price areas are separated settled at prices that satisfy transmission constraints. Area with excess generation will have lower prices, and those with excess load will have higher prices. The US Federal Energy Regulatory Commission (FERC),[3] established a system, where each SO would be responsible for monitoring its own regional transmission system and calculating its ATC for potentially congested paths entering, leaving, and inside its network. The ATC values for next hour and for each hour in the future are placed on OASIS, operated by SO. Anyone wishing to do transaction would access OASIS web pages and use ATC information available there to determine if system could accommodate transaction. Details of the regional congestion management methodologies applied in Europian electricity market is presented in final report [21]. The report summarizes the congestion management methods and procedures adopted in Europian region. A report on congestion management in Nordic region details out the rules for congestion management and nodal and zonal pricing based methods [22]. In an optimal power flow, (OPF) is performed to minimize generators operating cost subject to set of constraints that represent a model of the transmission system within which the generator operate. The generator sends a cost function and those customers willing to purchase power send a bid function to the SO. The SO has a complete transmission model and performs OPF calculation. OPF solution gives cost/MW at each node of the system. In some countries zonal pricing method is followed in which the system is divided into various zones

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

on geographical basis. The zone prices obtained from OPF are used in the following manner: Generators are paid zone price of energy The loads must pay the zone price of energy In case of no congestion, the zone price will be same and the generators are paid the same price for their energy as the loads pay. When there is congestion, the zone prices will differ, each generator is paid its zone price and the loads also pay its zone price for the energy. Thus, the OPF, through different zonal pricing, performs the function of controlling the transmission flows and thus maintaining the transmission security. The goal of deregulation is to encourage lower electric utility rates by structuring an orderly transition to competitive bulk power markets. The key is to open up transmission services, the vital link between sellers and buyers. To achieve the benefits of robust, competitive bulk power markets, all wholesale buyers and sellers must have equal access to the transmission grid. Otherwise efficient trades cannot take place, and ratepayers will bear unnecessary costs. Problems arise because all transactions have to share the same transmission network simultaneously. What portion of the capacity of a particular transmission system is used by a certain transaction? How can a fair charge be calculated based on the capacity usage? Can some transactions be adjusted, or even canceled, when transmission congestion occurs? Which generator or transaction should be considered first for adjustment in congestion management, based on transmission capacity usage? To answer these questions, an algorithm was developed which can calculate the contribution from individual generator units to the flows and losses trough the transmission network and at the load centers. This is both an essential and challenging task. Scholars from England were the first to propose the Flow Based Proportional Sharing Method (PS), based on a strong proportional sharing assumption, which has not been proven either correct or incorrect at this time [4-8]. Among them, D. Kirschen proposed the famous Topological Trace Algorithm; J. Bialek proposed the Upstream-Looking and Downstream-Looking algorithms. R. Shoults, who proposed the Circuits Based Method to challenge the currently very popular Flow Based Proportional Sharing Method, believed that the correct method should be founded upon established circuits theories. The Circuits Based Method is comprised of two sub-methods: 1. Current Division method 2. Voltage Division method. A new method named the Power Flow Comparison Method (PFC) which makes an effort to conform to the physical concepts commonly understood and accepted by power system engineers. These methods are applied to an example of 6-bus power system, and the results are presented, compared and discussed to verify their correctness. Further applications in transmission charges

[9-11] and transmission congestion management [12, 13] show the great practical value of tracing the flow of power from source to load. The impact of corrective actions on one group of lines to other heavily loaded lines is also shown. II.POWER FLOW CONTRIBUTION FACTORS METHOD The Power Flow Comparison Method (PFC) is comprised of the following procedure to find the contribution of each generator to the line flows, losses and loads: i) Calculate the base case power flow. ii) For the generator of interest, remove generation and a corresponding load from the power system in even quantities. iii) Make this generator bus the swing bus and calculate the power flow again iv) Find the line flow difference on each transmission line by comparing the two power flow results above. An example 3-bus power system, shown in Fig 1, is provided to illustrate this algorithm. Line resistance is ignored, and distributed capacitance is considered. Part I of Fig 1 is the base case power flow result. Part II of the Fig 1 is obtained by removing the generation at bus A (150 MW) and its corresponding loads at bus C (150MW) from the base case power system. The contribution of generator A to the line flows can be found by subtracting the line flow results in Part II from those in Part I. Part III of the Figis obtained by removing the generation at bus B (50MW) and its corresponding loads at bus C (50MW) from the base case power system. The contribution of generator B to the line flows can be obtained by subtracting the line flow results in Part IV from those in Part I. The final results are listed in Part IV. From this illustration, the reader may be reminded of the distribution factors method, which shows the sensitivity of the line flows to changes in generation. It is always easy to remove a generator (in a simulation.) It is rather difficult to find the corresponding loads of this generator, which are different from the contract loads. The term corresponding loads is used to mean the load served by this particular generator. How they are calculated is summarized in the following paragraphs. It is assumed that the voltage at each network node will not change much when a quantity of generation and its corresponding loads are evenly removed from the network. This is the constant voltage assumption used in the paper.
150MW A B 50MW

200MW

Fig.1 (a): Base case power flow

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

0.0MW

50MW

50MW

Fig1 (b): Remove generator A and its load from base case power flow

150MW

0.0MW

150MW

Fig1(c): Remove generator B and its load from base case power flow.

150MW

A
A: 50.42MW B: 16.56MW

50MW

A: 99.58MW B: 16.56MW

A: 50.42MW B: 33.44MW

200MW

Fig1 (d): Power flow comparison methods

III. APPLICATION IN CONGESTION MANAGEMENT With the open access of transmission networks and the competitive bulk power market, transmission congestion has become a more and more serious problem in the deregulated world. However, based on the calculated contribution of line flows from each generator, the PFC Method and the Proportional Sharing (PS) Method can provide guidance to relieve these congestion problems. Test cases on the IEEE 57-bus system [17] are presented to show the details. Results of two test cases on the IEEE 57-bus system show the capability of the Power Flow Comparison (PFC) Method over the Proportional Sharing Method (PS) for solving the transmission congestion management problems. Key output data are given in the [18], which includes: Bus output of the load-flow study; Line flows of the load-flow study; Generator contributions to MW line flows based on the PFC Method; Generator contributions to MW line flows based on the PS Method.

Case #1: Line 8 is heavily loaded with a total line flow of 174.0983 MW. Based on the MW contribution of each generator to the line flow, transmission congestion management methods are suggested by both the PS Method and the PFC Method as shown in Table 1. The PS Method predicts that decreasing G3 is the only way to lower the line flow. The PFC Method, on the other hand, predicts that the line flow can be decreased with more flexibility. In tests 1, 2, and 6, the line flow is decreased without changing G3. The TESTS portion of Table 2.1 shows the sensitivities of line flow changes vs. the generation changes. Because line 15 is also heavily loaded, the sensitivities of line 15 are listed in the TESTS table as well so that a management method, with a minimum impact online 15, can be found. Tests 1 to 6 apply 1 MW to each generation change. Test 6 is able to lower the line flows at both lines 8 and 15. Test 5 is able to drop the line flow at line 8 dramatically while keeping the line flow increase at line 15 to a minimum. Case #2: Line 69 is loaded with a total line flow of 5.8089 MW. Based on the MW contribution of each generator to the line flow, the transmission congestion management suggested by the PS Method and the PFC 40 3.Method are quite different, as shown in the Table 2.The PS Method predicts that decreasing G3 is the only way to lower the line flow. Conversely, the PFC Method predicts that the line flow will be decreased, by increasing G3 and decreasing the power output of the other generators, or without changing G3. The TESTS table shows the sensitivities of line flow changes v/s the generation changes. Tests 1 to 8 apply 1 MW to each generation change. Tests 1 to 3 show that the PS Method leads to a change in the wrong direction because the line flow at line 69 is increased when G3 is decreased. Tests 4 to 8 show that the PFC Method provides more usable guides and more alternatives for transmission congestion management. IV. REAL AND REACTIVE POWER CONTRIBUTION FACTORS Algorithm for calculating the contribution factors of each generator to the line flows, losses and loads as proposed in [16] is given below: (a) Perform the base case Newton-Raphson power flow. (b) Compute the sensitivity Sijk of the real and reactive power flow Pij and Qij of a line connected between bus-i and bus-j to real and reactive power output PGk and QGk of generator-k. The fast-forward/fastbackward substitution method allows an efficient computation of the sensitivity[15]: 1 dPij Pij Pij g g k S ijp (1) k k k dPG PG PG

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

k S ijq

dQij
k dQG

Qij
k QG

Qij

g
k QG

u i

(2) (3) (4)

Pi m ,load
j
u i

m CFijp

(11)

k S ijp k S ijq

N ijp J N ijq J

1 1

Mk Mk

Qim ,load
j

m CFijq

(12)

where Nijp and Nijq are the sparse block vector with subvector [ -bij Vi Vj, 00] and [ bij Vi Vj, 00] and [ -gij Vi Vj, 00] and [ gij Vi Vj, 00] in the ith and jth position, respectively. Mk is the sparse block vector with sub-vector [-1, 0] in the kth position. [J] is the Jacobian power flow matrix. g is the power flow equation vector and is the voltage angle vector. (c) As proved in [16], the contribution factor of slack generator to real and reactive power flow of line i-j is:
NG

where iu is the set of nodes which are directly connected to node-i. The total contribution of all the generators to line i-j is equal to the real power flow of this line:
NG

Pij
m 1

m CF ijp

(13)

NG

Qij
m 1

m CF ijq

(14)

( Pij CFijp

k k 1 S ijp PG ) PG k 2 NG k PG k 1

(5)

NG

(Qij CFijq
k 2 NG

k k 1 S ijq QG )QG

(6)
k QG

k 1

where CFijp and CFijq are contribution factor of slack generator to the real and reactive power flow of line i-j. PG1 is the real power output of the slack generator. (d) The contribution factor of the mth generator (except slack generator-1) to the real power flow of line i-j is
NG NG k k S ijp PG k 2 NG k PG k 1 NG NG k k S ijq QG k 2 NG k QG k 1 m S ijq k 1 k m QG QG m S ijp k 1 k m PG PG

Pij
m CFijp

(7)

The contribution factors based on sensitivity of power flow are used to find the share of each generator on each line flow and consequently the generator share will help ISO to identify the generators to re-dispatch their generation in congestion management. This can be computed in advance and the SO can post the contribution of each generator to line flow, so that during the congestion hours the SO can send the signal in the market for rescheduling of real and reactive powers to manage congestion. An example 3-bus system is provided to illustrate the capability of the above formulas. Resistance is 0.01 p.u. Reactance is 0. 1 p.u. and the distributed capacitance is considered with a n-model in all transmission lines. Figs. 2 and 3 show the base case power flow result and contribution of each generator unit to each line flow in the above system, based on proposed CF formulas and the PFC method [20], respectively. The results of Fig 3.2 are similar to those of Fig 3.1, because the voltage at each network node does not change much when the quantity of generation and its corresponding loads arc 'evenly' removed from the network.

Qij
m CFijq

(8)

Where CFijpm and CFijqm is the contribution factor of generator-m (except slack generator) to real and reactive power flow of line i-j. (e) The contribution of each generator-m to losses of each line i-j can be calculated as: m m m Pij ,loss CFijp CF jip (9)
m Qij ,loss m CFijq
m

Fig 2: Illustration of CF formulas for 3 bus system

m CF jiq
m

(10)

Where Pij,loss and Qij,loss is the contribution factor of generator-m to loss of line i-j. (f) The contribution of generator-m to load at the bus-i, is given as:

Fig.3: Illustration of PFC method by 3 bus system

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

The proposed CF formulas could have applications in, computation of contribution of generation to a DC-line or DC-link of an AC-DC power system [19], computation of contribution of' reactive generation to reactive power flows when the real power is replaced by reactive power i n the above formulas, application to individual customers for the apportionment of the use-of-the system charges, calculation of the sensitivity regarding corrections, that will affect the flows in the critical lines. V. POWER FLOW CONTRIBUTION FACTORS METHOD Test results for 6 bus system An example 6-bus power system with three generators and five load centers are shown to verify and compare the proposed CF formulas with the PFC method. Fig 4 gives the topology of the 6-bus test system and the line flow results. The contribution of the generator units to line flows, resulting from the implementation of CF formulas and PFC method, is shown in Tables 3 and 4, respectively. The contribution of the generators to load centers, resulting from the implementation or CF formulas and the PFC method, is shown in Tables 5 and 6, respectively. Table 7 shows the voltage at each network node changes, when the PFC method is enforced and the generation and 'corresponding load' are removed. These results indicate that the 'constant voltage' assumption, which is applied in the PFC method, is not acceptable as accurate.
350.145MW 250MW 10 250MW 30 100MW 20 50MW

respectively. The results of these tests show the sensitivities of line-22 flow changes vs. the generation changes (
line.22/

G ).This tests shown are chosen to

decrease the flow of line-22 with more flexibility because they present the maximum value in the above sensitivities. Since line-21 (120 bus- 230 bus) is also heavily loaded (366.51 MW), the sensitivities of line-21 (
line.21/

G )

are shown in Tables 7 and 8 as well, so that a management method, with a minimum impact on line-21, can he found. Tests 1 to 5 apply 1 MW to each generation change. In tests 1, 3, 4 and 5, generators whose power output changes (1 MW) have the same effect on the sensitivities ( P /

G ) are indicated with '(or)'. Both methods predict

that the line flows can be decreased with more flexibility. The difference between these methods is the proposed scenario for solving the congestion problem, and that occurs because of the inaccuracy of the 'constant voltage' assumption of the PFC method [20]. IV. CONCLUSIONS In this paper, the Power Flow Comparison Method (PFC) and CF method has been analyszed and applied for congestion management by rescheduling the power available from the generators. Results of calculations applied to an example 6 bus power system and IEEE-57 bus test systems are presented, compared and discussed to verify the correctness of the methods. An application of PFC and CF approaches has been carried-out and compared for 25-bus system for congestion management studies. In particular, the following conclusions are made Changes in generation suggested by all the methods will affect flows in other lines. Hence, other heavily loaded lines should be considered as well when solving a transmission congestion problem. The PFC Method provides convenient facilities for doing this. The sensitivities used in the congestion management change very little when the generation is increased. According to the PFC Method, all generators have some contribution to each line flow. Based on their contribution, the order in which the generators should be adjusted can be determined. When a generator has a positive contribution to the line flow, the output of this generator needs to be decreased to lower the transmission congestion and vice versa. When two generators have conflicts in congestion management, the one with higher MW contribution has more impact on the line flow changes. The Proportional Sharing Method sometimes may lead to changes in the wrong direction. The CF based approach expressing the contribution of each generator unit to loads or flows or losses or lines has been implemented based on [16].The proposed CF formulas are applicable independently to active and

40

50 200 MW

200MW 60 300MW

Fig 4: 6-bus system

Application in Congestion Management In this Section, the implementation of the proposed CF formulas and PFC method [20] in the transmission congestion management are presented. The proposed CF formulas can provide guidance to relieve congestion problems. Results on the 25-bus test system show the ability and accuracy of CF over PFC for solving transmission congestion problems. In this test system, line-22 (130 bus-230 bus) is heavily loaded with a total line flow of 381.54MW. Based on the CF and PFC methods, transmission congestion management solutions are suggested as shown in Tables 8 and 9 respectively. Tables 8 and 9 show the contribution ( P) of each generator to line-22 based on the CF and PFC methods,

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

reactive loads, flows and losses replacing the real power by reactive power in the load flow scheme. These formulas and other state-of-the-art methods have been tested on various test systems to compare results. The results obtained from CF are similar as obtained from PFC method. Also, the proposed CF formulas can provide guidance to relieve the congestion problems REFERENCES
[1] [2] L.PHILIPSON and H.LEE WILLIS, Understanding Electric Utilities and Deregulation, Marcell Dekker Inc., NY 1998. R.S.FANG and A.K. DAVID, Transmission Congestion Management in Electricity Market, IEEE Transactions on Power Systems, Vol. 14, No. 3, August 1999, pp. 877-883. North American Electric Reliability Council (NERC), Available Transfer Capability Definition and Determination, NERC Report, June 1996. D. KIRSCHEN, R ALLEN and G. STRBAC, Contributions of Individual Generators to Loads and Flows. IEEE Transactions on Power Systems, Vol. 12, No. 1, pp. 52-60, February 1997. G. STRBAC, D. KIRSCHEN, S. AHMED, Allocating Transmission System Usage on the Basis of Traceable Contributions of Generators and Loads to Flows, paper PE-222PWRS-0-01-1997, IEEE, 1997. J. BIALEK, Topological Generation and Load Distribution Factors for Supplement Charge Allocation in Transmission Open Access, IEEE Transactions on Power Systems, Vol. 12, No. 3, August 1997. J. BIALEK, Identification of Source-sink Connections in Transmission Networks, Power Systems Control and Management, 16-18 April 1996. R. SHOULTS and L.D. SWIFT, A comparison Between Circuits Based Methods and Topological Trace Methods for Determining the contribution of Each Generator to Load and Line Flows, Proceedings of the workshop on Available Transfer Capability, pp. 167-182, University of Illinois at Urbana-Champaign, June 26-28, 1997. R. A. WAKEFIELD, J. S. GRAVES, A. F. VOJDANI, A Transmission Services Costing Framework, IEEE Transactions on Power Systems, Vol. 12, No. 2, May 1997.

[3]

[4]

[5]

[6]

[7]

[8]

[10] C. W. YU, A. K. DAVID, Pricing Transmission Services in the Context of Industry Deregulation, IEEE Transactions on Power Systems, Vol. 12, No. 1, February 1997. [11] H. H. HAPP, Cost of Wheeling Methodologies, IEEE Transactions on Power System, Vol. 9, No. 1, February 1994. [12] J. D. FINNEY and H. A. OTHMAN, Evaluating Transmission Congestion Constraints in System Planning, IEEE Transactions on Power System, Vol. 12, No. 3, August 1997. [13] H. SINGH, S. HAO and A. PAPALEXOPOULOS, Transmission Congestion Management in Competitive Electricity Markets, paper PE-543- P WRS-2-06-2997, IEEE, 1997. [14] R. A. WAKEFIELD, J. S. GRAVES, A. F. VOJDANI, A Transmission Services Costing Framework, IEEE Transactions on Power Systems, Vol. 12, No. 2, May 1997. [15] A.G. BAKIRTZIS: Sensitivity computation of power flow control in electric power systems. Electric Power System. I991. 22. pp. 77-84 [16] J.G.VLACHOGIANNIS: Accurate model for contribution of generation to transmission system effect on charges and congestion management. IEEE proceeding generation transmission vol.147 no.6 November 2000, pp342-348. [17] J. D. GLOVER and G. DIGBY, Software Manual, Power System Analysis and Design, Second Edition, PWS Publishing Company, Boston, 1994. [18] JIAN YANG, Power Systems Deregulation and Development of Powergraf, a PhD Dissertation, University of Missouri - Rolla, July, 1998. [19] J.G. VLACHOGIANNIS: Control Adjustment in Fast Decoupled load flow, Electric Power System Res. I994. 31..pp. 185-194. [20] J. YANG and M. D, ANDERSON,:Tracing the power flow of power in transmission networks foe use of transmission system charges and congestion management, IEEE Trans. Power System., 1998.13 (2), pp.399-405. [21] www. Constec.de, Towards a common coordinated congestion management methods in Europe, Final Report, 2007, DirectorateGeneral Energy Transport, 2007. [22] www.nordiceenergyregulators.org, Congestion management in the Nordic region: A common regulatory opinion on congestion management, A report, 2007.

[9]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

Method

LF_G 1 (MW) 0

TABLE. 1 : CASE #1 - LINE 8 IS HEAVILY LOADED, G3 MUST BE DECREASED LF_G2 LF_ G3 LF_G4 MW) 0 1.0689 174.0983 232.6123 (MW) (MW) 0 -36.2409 G3 G1 G4 G2 G3 G4 G1 G2

Production of management

PS PFC

-23.342

Test G1 1 2 3 4 5 6 +1 0 0 +1 0 -1 G2 -1 -1 0 0 +1 0

G3

G4

Pline G
-0.0841 -0.1781 -0.7031 -0.090 -0.5250 -0.0941 Production of management

Pline 15 G
+0.3715 +0.0343 +0.0523 +0.3795 +0.0181 -0.3372

0 0 -1 -1 -1 0 LF_G3(MW)

0 +1 +1 0 0 +1 LF_G4(MW) 0 -36.24O9

Method PS PFC

LF_G1(MW)) 0 -23.342

LF_G2(MW) 0 1.0689 174.0983 232.6123 G1 G1 G2 G2 G3 G3 G4 G4

TABLE. 2: CASE #2 - PS METHOD GIVES WRONG ANSWER AND PFC METHOD PROVIDES ALTERNATIVES

TEST

G1MW

G2MW

G3MW

G4 MW

Pline 8 G
+0.0092 +0.0048 +0.0144 -0.0092 -0.0048 -0.0144 -0.0096 -0.0044

1 2 3 4 5 6 7 8

+1 0 0 -1 0 0 0 -1

0 +1 0 0 -1 0 +1 +1

-1 -1 -1 +1 +1 +1 0 0

0 0 +1 0 0 -1 -1 0

TABLE 3: CONTRIBUTION OF GENERATION OF 6-BUS TEST SYSTEM TO LINE FLOWS RESULTING FROM CF FORMULAS Contribution Of G-10 (MW) 90.617 127.702 131.811 70.038 32.955 53.559 49.415 20.600 Contribution Of G-20 (MW) 106.71 0 41.184 65.552 128.558 -19.321 31.099 06.723 50.409 Contribution of G-30 (MW) 14.695 -58.830 44.123 0.0104 73.524 88.230 -14.705 14.707

Lines 10-20 10-30 10-40 20-50 30-50 30-60 40-60 50-40

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

TABLE 4: CONTRIBUTION OF GENERATION OF 6-BUS TEST SYSTEM TO LINE FLOWS RESULTING FROM PFC METHOD [20] Contribution Of G-10 (MW) Contribution Of G-20 (MW)

Lines

Contribution Of G-30 (MW)

10-20 10-30 10-40 20-50 30-50 30-60 40-60 50-40

89.725 126.897 131.535 70.144 32.987 53.725 49.114 20.736

106.370 40.880 65.490 128.217 -19.051 31.258 6.657 50.315

14.346 -58.527 44.181 0.378 73.261 88.078 -1 4.643 14.811

TABLE 5: CONTRIBUTION OF GENERATION OF 6-BUS TEST SYSTEM TO LOAD CENTERS RESULTING FROM CF FORMULAS (Load centers)/ load (MW) 120)/50 (30)/100 (40)/100 (50)/200 (60)/300 Contribution of G-10 (MW) 20.579 41.188 82.396 82.393 123.574 Contribution of G-20 (MW) 14.732 29.406 58.829 58.828 88.231 Contribution of G-30 (MW) 14.684 29.416 58.828 58.827 88.232

TABLE 6: CONTRIBUTION OF GENERATION OF 6-BUS TEST SYSTEM TO LOAD CENTERS RESULTING FROM PFC METHOD [20] (Load centers)/ load (MW) Contribution of G-10 (MW) Contribution of G-20 (MW) Contribution of G-30(MW)

(20)/50 (30)/100 (40)/100 (50)/200 (60)/300

19.581 40.185 82.421 82.395 123.575

15.413 28.673 58.833 58.821 88.230

13.968 30.134 58.824 58.828 88.246

TABLE 7: VOLTAGE DIVERGENCES WHEN PFC METHOD [20] IS APPLIED IN 6-BUS TEST SYSTEM Load bus Base case Load flow solution (Voltage p. u.) bus voltage when Generation and corresponding load are removed 0.967 0.958 0.959 0.954 0.950

V ( voltage divergences) (%s) 3.3 4.2 3.4 4.2 4.4

20 30 40 50 60

1.000 1.000 0.993 0.996 0.994

TABLE 8: APPLICATION OF CF FORMULAS IN CONGESTION MANAGEMENT OF 25-BUS TEST SYSTEM (LINE-22 IS HEAVILY LOADED) Contribution of generation to flow of line-22 Method G10 G20 G70 G130 G150 G160 G180 G210 G220 G230 ( MW ) ( MW ) ( MW ) ( MW ) ( MW ) ( MW ) ( MW ) ( MW ) ( MW ) ( MW )

CF

7.643

7.940

9.323

73,799

-19.957

-21.463 -50.060 -48.928 -54.131 -285.706

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

Proposed output changes of generation in order to lower line flows in line-22 to be achieved G10 G20 G70 G130 ( MW ) ( MW ) ( MW ) ( MW ) G150 G160 G180 ( MW ) ( MW ) ( MW ) G210 G220 ( MW ) ( MW ) G230 ( MW )

Test

1 2 3 4 5

(or)+1

(or)+1

(or)+1 +1 +1 +1 (or)+1 or)-1 (or)+1 (or)-1 (or)+1 (or)+1 (or)-1 (or)-1 (or)+1 (or)-1

-1 -1 -1

(or)-1

(or)-1
line.22/

(or)-1

Results of tests 1 2 3 4 5

G
-0.29 -0.45 -0.14 -0.31 -0.16

line.21/

-0.27 -0.25 -0.11 -0.30 -0.04

TABLE 9: APPLICATION OF PFC METHOD [20] IN CONGESTION MANAGEMENT OF 25-BUS TEST SYSTEM (LINE-22 IS HEAVILY LOADED) Contribution of generation to flow of line-22 Method G10 G20 G70 G130 G150 G160 G180 G210 ( MW ) ( MW ) ( MW ) ( MW ) ( MW ) ( MW ) ( MW ) ( MW ) PFC 8.168 8.309 10.632 73.501 G220 ( MW ) G230 ( MW )

-20.312 -21.971 -51.110 -49.536 -54.203 -285.018

Proposed output changes of generation in order to lower line flows in line-22 to be achieved Test G10 ( MW ) (or)+1 G20 ( MW ) (or)+1 G70 G130 ( MW ) ( MW ) (or)+1 +1 +1 +1
G

G150 G160 ( MW ) ( MW )

G180 G210 ( MW ) ( MW )

G220 ( MW )

G230 ( MW ) -1 -1 -1

1 2 3 4 5 Results of tests 1 2 3 4 5

(or)+1 or)-1

(or)+1 (or)-1

(or)+1 (or)-1

(or)+1 (or)-1

(or)+1 (or)-1

(or)-1
line.22/

(or)-1

(or)-1
line.21/

-0.24 -0.48 -0.26 -0.37 -0.07

-0.22 -0.19 -0.23 -0.35 +0.11

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

Chaotic Particle Swarm Optimization for Reduced Order Model of Automatic Generation Control
1

Cheshta Jain, 2H. K. Verma


1, 2

SGSITS, Indore

cheshta_jain194@yahoo.co.in 2 vermaharishgs@gmail.com

Abstract This paper presents a new approach to control frequency and tie line power changes of multi area interconnected system. AGC is very important in power system to maintain system frequency and tie-line power, when system subjects to small load perturbations. Chaotic particle swarm (CPSO) is used to optimize gains of PI controller and bias frequency, to maintain system frequency and tie-line power flow at scheduled values. In this paper, model order reduction technique has been used to obtain lower order model for study of automatic generation control (AGC). Model order reduction method can preserve the identity of each generating unit. With this technique, the computational complexity and effective time has been reduced by re-sorting the lower order generating unit models. This paper also presents the selection of suitable value for governor speed regulation parameter. The proposed method shows its robustness under critical conditions when conventional optimization methods fail. Keywords chaotic particle swarm optimization, automatic generation control, reduced order of AGC.

NOMENCLATURE
F = Frequency deviation. i = Subscript referring to area (i = 1,2,). Ptie (i,j) = Change in tie line power. Pdi = Load change of ith area. Di = PDi / Fi Ri = Governor speed regulation parameter for ith area. Thi = Speed governor time constant for ith area. Tti = Speed turbine time constant for ith area. TPi = Power system time constant for ith area. KPi = Power system gain for ith area. ACEi = Area control error of ith area. Hi = Inertia constant of ith area. Ui = Control input to ith area. Bi = Frequency bias for ith area. Us = Undershoot of ACE. Mp = Overshoot of ACE. ts = Settling time ACE. tr = Rise time of ACE. ess = Steady state error of ACE. W = Inertia weight. C1, C2 =acceleration coefficient.

I. INTRODUCTION Automatic generation control is one of the most important issues in power system design. The purpose of AGC is fast minimization of area frequency deviation and

mutual tie-line power flow deviation of areas for stable operation of the system. The overall performance of AGC in any power system depends on the proper design of speed regulation parameters and gains of controller. Fixed linear feedback controller fails to provide best control performance. There are no well defined methods available to compute tie-line constant for an area with non-coherent generators and multiple tie-line. The complexity of computation of tieline increases with the increase in number of areas. This paper presents a reduced order AGC simulation technique that overcomes this problem. These reduced order systems overcome the need of identifying the non-coherent set of generators in order to control the areas. This technique is based on some assumptions. These assumptions follow the fact that, all areas are operating at same frequency and tieline flow can no longer be computed in actual system. Tieline flows are required to determine area control error [3]. The conventional controller improves steady state error (ess) but with small overshoot. PI controller has such capability to improve transient performance with minimum steady state error. The aim of this paper is to use CPSO to find optimum gains of PI controller and frequency bias with proper system parameter. CPSO is an optimization approach based on the PSO with adaptive inertia weight factor methods. It provides more precise description of natural swarm behaviour. In the view of the above, following are main objectives of the proposed paper: 1. To reduced the order of the generating unit of AGC system. 2. To optimized the gains of PI controller and frequency bias coefficient using chaotic PSO algorithm in MATLAB. 3. To examine the effect of speed regulation parameters on reduced system. The rest of the paper is organized as follow: section I presents AGC system model with proposed reduction technique. In section II chaotic particle swarm optimization is discussed in brief and an algorithm to implement CPSO based PI controller is presented in section III. Section VI shows the result with discussion and conclusion is drawn in section V.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

10

II. SYSTEM MODEL A. Conventional AGC system Automatic control system of close loop system means minimizing the area control error (ACE) to maintain system frequency and tie-line deviation are set at nominal value [15]. Block diagram of two area system is shown in Fig 1.

Where, p q n. The gain K is the same for original and approximated system. This method is based on selection of c and d in such a way that G L(s) has a response very close to that of GH(s). These coefficients are evaluated as: (4) And (5) Where, M(s) and (s) are the numerator and denominator polynomials of GH(s)/GL(s) respectively. To determine total (cp + dq) no of unknowns, requires (cp + dq) no of equations. Therefore define

Fig 1: Linear model of two area system.

The ACE of each area is linear combination of biased frequency and tie-line error.

(6) g = 0, 1, 2, ---- up to number required to solve the unknown coefficients. An analogous equation for 2g. the solution for c and d coefficient is obtained by equating M2g = 2g. In this paper two area AGC system is considered. This model has second order transfer function of generating unit of each area as:

(1) Where, Bi is frequency bias coefficient of i area. Fi is frequency deviation and Ptie is tie-line error of ith area. Based on the ACE a suitable control strategy can be taken up either continuously or discretely. A practical system consists of a number of generating units, which increases computational time and complexity. In this paper the reduction of order is used to reduce above effort.
th

(7) The value of AGC parameters (T H and TL) are given in appendix. After reduction as above method the lower order generating unit becomes: (8) III.OVERVIEW OF CHAOTIC PARTICLE SWARM OPTIMIZATON a. General PSO method Particle swarm optimization (PSO) first proposed by Kennedy and Eberhart [5]. Like evolutionary algorithms, PSO techniques conducts search using a population of particles, corresponding to individuals. In PSO, particles changes their position by flying around in a search space until computation limitations are exceed. PSO is inspired by the ability of flocks of birds to reach unknown destination. In PSO each particle is defined as moving part in hyperspace. PSO is inherently continuous and must be modified to handle design variables. The basic PSO algorithm requires three steps [5]. Step 1: Initialization PSO is initialized with the group of random particle positions (xki) and velocities (Vki) between upper and lower bound of design variable values for N particles as expressed in equation 9. X0i = Xmin + rand*(Xmax Xmin) (9) Velocity is initialized as: (10)

B. Reduced order generating unit Several methods are available for reducing the order of a system transfer function (TF) [14]. One way is to delete a certain insignificance pole of a transfer function, which has a negative real part that is much more negative than the other poles [14]. An effective approach is to match the frequency response of the reduced order transfer function with the original TF frequency response. Suppose the higher order system is described as:

(2) This system has poles in the left hand s-plane and m n. The lower order approximate transfer function is:

(3)

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

11

i = 0,1,2,.N. Step II: Update velocity Velocity of all particle updated for next (k+1) th iteration using the particles fitness values, which is function of particle positions. These fitness function value determines which particle has a global best (gbest k) value in the current swarm (kth iteration) and also determine the best position (pbesti) of each ith particles. The global best value speed up the rate of convergence. gbest value maintains only single best solution across the entire particle in the search space. If all particles are converging to this position, the further improvement of each particle stops. After finding the two best values, the particles update its velocity using following equation:

(16) B. Algorithm Step1. Choose the population size (N) and number of Step2. iteration (Nmax). Set the value of inertia weight w. Step3. Generate initial population (eq. 9) and velocities randomly as given in equation10: Step4. Set counter k=1. Step5. Run model of reduced AGC system and determine performance parameters for each particle. Step6. Calculate fitness function for each particle (eq.16). Calculate gbest and pbest position value. Using these values update velocity of each particle (eq.15). Step7. Update position of each particle (eq. 12). Step8. Calculate new fitness function for each updated particles position, if it is better than previous value of fitness function then the current value of particle position is set to pbesti. Step9. Set k = k + 1. Step10. If the last change of the best solution is greater than a pre specified number or the number of iteration reaches the maximum iteration, stop the process, otherwise go to step 4. V. RESULT AND DISCUSSION A comparative study of CPSO algorithm on reduced order and full order model is carried out in this paper. The time response plots of area control error of area 1 of the reduced order as well as full order AGC system is given in Fig2-5. These Figures shows the variation in system frequency, tie-line power flow and area control error for 1% step load perturbation on area two. From these Figures it can be seen that the response of reduced order generating unit model and actual model are approximately identical. Table I presents computational result for a input data set 1 (Tp=10, R= 8% frequency, T12=0.145). The results clearly established that reduced order model has less computational time than full order model with approximately same performance parameters. Hence, lower order model can be used to obtain desired response with saving in computation time. The gains of PI controller and frequency bias are obtained by chaotic PSO algorithm for different value of system parameters of reduced order AGC system. As Fig 6-13 shows, PI controller act too fast to the generator inputs and also exhibits fast oscillations. In all cases, an acceptable overshoot and settling time on frequency deviation signal in each area is maintained. Table-II gives transient response parameters of CPSO based PI controller for different data set. Three sets of Input data have been used in this paper, namely data set1 (Tp = 10, R = 8%, T12 = 0.145), data set2 (Tp=30, R=8% of frequency, T12=0.145) and data set3 have (Tp=30,

(11) Where w is inertia weight factor, c1 and c2 are self confidence and swarm confidence respectively. Combinations of these values usually lead to much slower convergence or sometimes non-convergence at all. Therefore proper selection of these particles is important. Step III: Update positions Now positions of particles are updated using following equation: (12) a. Chaotic PSO The parameters r1 and r2 in equation (11) are important control parameters. The use of chaotic sequence in PSO can be useful to escape from local minima than general PSO [4]. Chaotic sequence based on Henon mapping is used for random value r1as:

(14) Where, a and b are Henon map attractor. Another mapping uses the same equation for random value r2 to generate z 2(t) in range [0, 1]. Other parameters are same as in equation (10). Hence, velocity of particle is updated as:

(15) IV. IMPLEMENTATION OF CPSO-PI CONTROLEER A. Fitness function The gain of PI controller can be selected based on degree of relative stability, minimum overshoot, undershoot and settling time. To satisfy all requirements following objective function is design.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

12

R=4% of frequency, T12=0.145). Table-II and Figures 1013 shows that the large values of power system time constant (Tp=30) and low value of T12 (0.145) yield large value of undershoot, overshoot and settling time and hence high value of fitness function. These values (undershoot, overshoot, settling time for data set2) become lower as speed regulation, R decreases from 8% to 4% of frequency. The initial sharpness of the response in undershoot and overshoot lies due to choice of weighting factor (1000 and 100) in the fitness function
TABLE. I: COMPARISON OF ACTUAL AND REDUCED ORDER SYSTEM FOR DATA SET-I

Fig 3: Comparison of tie-line power for reduced order and full order AGC system

TABLE II: COMPARISON OF EFFECT ON INPUT SYSTEM PARAMETERS ON REDUCED ORDER AGC SYSTEM.

Fig 4: Comparison of area frequency 1 for reduced order and actual system.

Dataset1 (Tp=10, R=8% of freq, T12=0.145). Dataset2 (Tp=30, R=8% of freq, T12=0.145). Dataset3 (Tp=30, R=4% of freq, T12=0.145).

Fig 5: Comparison of area frequency 2 for reduced and actual system

Fig 6: Variation in area control error in dataset1 and dataset2

Fig 2: Variation in area control error in reduced order and full order system

Fig 7: Variation in area frequency 1 for dataset 1 and dataset 2 .

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

13

Step Response 0.01 f 2_dataset3 0.005 f 2_dataset2

-0.005

Amplitude

-0.01

-0.015

-0.02

-0.025

-0.03

10

15 Time (sec)

20

25

30

35

Fig 8: Variation in area frequency2 for dataset 1 and dataset 2 . Fig 12: Comparison of area frequency 2 for dataset 2 and dataset3 .
4 x 10
-3

Step Response p12_dataset1 p12_dataset2


-3

x 10

Step Response p12_dataset3 p12_dataset2

Amplitude

Amplitude
0 5 10 15 20 Time (sec) 25 30 35 40

-1

0
-2

-1

-2

10

15

20 Time (sec)

25

30

35

40

Fig 9: Comparison of tie-line power flow.

x 10

-3

Step Response ACE1_dataset3 ACE1_dataset2

Fig 13: Comparison of tie-line power flow for dataset 2 and dataset 3.

Amplitude

-1

-2

-3

-4

10

15

20 Time (sec)

25

30

35

40

Fig 10: Comparison of ACE for dataset 2 and dataset 3.

Step Response 0.01 f 1_dataset3 f 1_dataset2 0.005

VI. CONCLUSION In this paper CPSO method is used to obtain optimum gains of PI controller and frequency bias coefficient. CPSO is new variant of PSO with faster speed because of strong selection principle. In simple PSO, after certain iterations, the populations set are almost identical and no further improvement is observed. Reduced order AGC system has shown saving in computation time with identical responses. Like any other algorithms, this method also somewhat sluggish in nature but positive aspect of this method is its reliability and the number of required generation for convergence decreases with increase of population size. APPENDIX Nominal parameters of two area test system [15]: H1= H1= 5 seconds D1= D2= 8.3310-3P.U. MW/Hz R1= R1=2.4 Hz/P.U. MW Th1= Th2=80 ms Tt1= Tt2=0.3 seconds Kp1= Kp2=120HzP.U. MW Parameters for CPSO: Population size= 20 Number of iteration=100 Wmin=0.6 Wmax=1 REFERENCES
[1] Zhihua Cui, Xingjiuan Cai and Jianchao zeng, Chaotic performance dependent particle swarm optimization. Division of system simulation and computer application Taiyuan University of science and technology.

Amplitude

-0.005

-0.01

-0.015

-0.02

-0.025

10

15 Time (sec)

20

25

30

35

Fig 11: Variation in area frequency 1 for dataset 2 and dataset3.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

14

[2] [3]

[4]

[5]

[6]

[7]

[8] [9]

[10]

[11]

[12]

[13]

[14] [15]

C.F. Chen and L.S. Sheieh, A novel approaches to linear model simplification, International journal of control, 8, 561-570, 1968. K.C. Divya and P.S. Nagendra Rao, A novel AGC simulation scheme based on reduced order prime mover models, IEEE transaction on control system and applications, 1099-1103, 2003.` X. J. Cai, Z. H. Cui, J. C.Zeng and Y. Tan, Performancedependent adaptive particle swarm optimization, International Journal of Innovative Computing, Information and Control, Vol.3, No.6B,pp.1697-1706, 2007. J. Kennedy and R. Eberhart, Particle swarm optimization, in Proc. IEEE Int. Conf. Neural Networks, vol. IV, Perth, Australia, 1995, pp. 19421948. Z.-L. Gaing, A particle swarm optimization approach for optimum design of PID controller in AVR system, IEEE Trans. Energy Conversion, vol. 19, pp. 384-391, June 2004. M. Clerc, The swarm and the queen: towards a deterministic and adaptive particle swarm optimization, Proceedings of the IEEE Congress on Evolutionary Computation (CEC 1999), pp. 1951-1957, 1999. .D.Goldberg, Genetic algorithm in search optimization and machine learning, Addison-Wesley, 1989. Rania Hassan, Babak Cohanim, Oliver de Week, A comparison of Particle Swarm Optimization and the Genetic Algorithm American institute of aeronautics and astronautics. R. C. Eberhart and J. Kennedy, A new optimizer using particle swarm theory, Proceedings of the Sixth International Symposium on Micro Machine and Human Science, Nagoya, Japan, pp.39-43,1995 K.P. Wong, Z. Y. Dong, Special issue on evolutionary computation for system and control application international journal of system science, vol, 35, No. 13-14, 20 oct-15 Nov.2004, pp 729-730. R. C. Eberhart and J. Kennedy, A new optimizer using particle swarm theory, Proceedings of the Sixth International Symposium on Micro Machine and Human Science, Nagoya, Japan, pp.3943,1995. G. Yu, and R. Hwang, Optimal PID speed Control of brush less DC motors using LQR Approach, in Proc. IEEE Int. Conf. Systems, Maand Cybernetics, 2004, pp. 473-478. Richard C. Dorf and Robert H. Bishop, Modern control system, Pearson international edition 2003. O.I. Elgerd, Electric energy system theory an introduction, McGraw Hill Co., 2001.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

15

Combined Economic/Environmental Dispatch with Fuzzified Multi-Objective Particle Swarm Optimization considering Voltage Stability
1

S. Surender Reddy, 2A. R. Abhyankar, 3P. R. Bijwe


1, 2, 3

Indian Institute of Technology Delhi


2

salkuti.surenderreddy@gmail.com ,

abhyankar@ee.iitd.ac.in,

prbijwe@ee.iitd.ac.in

AbstractTraditional Economic Load Dispatch deals with minimizing generation cost while satisfying a set of equality and inequality constraints. The fossil fuel plants pollutes environment by emitting some toxic gases. Thus, conventional minimum cost operation cannot be the only basis for generation dispatch; emission minimization must also be taken care of. The objective of reactive power optimization problem can be seen as minimization of real power loss over the transmission lines. Large integrated power systems are being operated under heavily stressed conditions which imposes threat to voltage stability. The objectives economic dispatch, emission dispatch, loss minimization and voltage stability are to be met for efficient operation and control. The results of all the four objectives, considering one objective at a time are conflicting and non commensurable. Hence, an efficient control which meets all the specified objectives is required. Initially each objective is optimized individually using Particle Swarm Optimization (PSO) and then all the four objectives are optimized simultaneously using fuzzified PSO. The effectiveness of the proposed approach is tested on IEEE 30 bus system. Keywprds Economic load dispatch, emission dispatch, reactive power optimization, voltage stability, particle swarm optimization (PSO), fuzzy min-max approach.

I. INTRODUCTION The Power system should be operated in such a way that both real and reactive power are optimized simultaneously. Real power optimization problem is the traditional economic dispatch which minimizes the real power generation cost. Reactive power should be optimized to provide better voltage profile as well as to reduce total system transmission loss. Traditional Economic Dispatch [1] aims at scheduling committed generating units outputs to meet the load demand at minimum fuel cost while satisfying equality and inequality constraints. On the other hand, thermal power plants (which contribute major part of electric power generation) create environmental pollution by emitting toxic gases such as carbon dioxide (CO2), sulphur dioxide (SO2), nitrogen oxides (NOx). Initially, Combined Economic and Environmental/Emission dispatch (CEED) problem was solved by minimizing fuel cost considering emission as one of the constraints. The CEED problem is one of the fundamental issues in power system operation. The operation and planning of a power system is characterized to maintain a high degree of economy and reliability [2].

Among the options available to the power system operators to operate the generation system, the most significant is the economic dispatch. Traditionally electric power plants are operated on the basis of least fuel cost strategies and only little attention is paid to the pollution produced by these plants. The generation of electricity from the fossil fuel releases several contaminants, such as sulphur oxides (SO2), nitrogen oxides (NOx) and carbon dioxide (CO2) into the atmosphere. But, if pollution is decreased by suitably changing the generation allocation, the cost of generation increases deviating from economic dispatch. The characteristics of emissions of various pollutants are different and are usually non-linear. This increases the complexity of the CEED problem. Large integrated power systems are being operated under heavily stressed conditions which imposes threat to voltage stability. Voltage collapse occurs when a considerable part of the system attains a very low voltage profile or collapses. Hence, voltage stability of the system is also an important consideration and need to be taken care of simultaneously along with economic dispatch. The four objectives of minimization of fuel cost, minimization of emission, minimization of losses and minimization of system stability index are conflicting and non commensurable. Hence, trade off solution using fuzzy min-max approach is proposed in this paper. Assuming the decision maker (DM) has imprecise or fuzzy goals of satisfying each of the objectives, the multi-objective problem can be formulated as a fuzzy satisfaction maximization problem which is basically a minmax problem. PSO is an unconstrained optimization method. A PSO method that includes the constraints penalty factor approach is used to convert the constrained optimization to an unconstrained optimization problem [3]. In PSO, the search for an optimal solution is conducted using a population of particles, each of which represents a candidate solution to the optimization problem. Particles change their position by flying round a multidimensional space by following current optimal particles until a relatively unchanged position has been achieved or until computational limitations are exceeded. Each particle adjusts its trajectory towards its own previous best position and towards the global best position attained till then. PSO is easy to implement and provides fast convergence for many optimization problems and has

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

16

recently gained lot of attention in power system applications. In [4], it is demonstrated that PSO gets better results in a faster, cheaper way compared with other methods. In PSO, the potential solutions called particles, fly through the problem space by following the current optimum particles. Compared to GA, the advantages of PSO are that PSO is easy to implement and there are few parameters to adjust. PSO is initialized with a group of random particles (solutions) and then searches for optima by updating generations [5]. The description and applications of PSO are presented in [6, 7]. The main aim of this paper is to investigate the applicability of PSO to the various optimization problems and prove that this algorithm can be used effectively, to determine solutions to various complex problems. Fuzzified PSO approach is tested on several case studies which are extremely difficult to solve by standard techniques due to the non-convex, non-continuous and highly nonlinear solution space of the problem. The paper is organized as follows. Sections II, III, IV and V presents economic load dispatch with fuel cost minimization, emission dispatch with emission minimization, reactive power dispatch with active power loss minimization, voltage stability maximization with Lindex minimization as single objective optimization subproblems using PSO. Section VI describes multiobjective optimization problem using fuzzified PSO. Finally, brief conclusions are presented in Section VII. II. ECONOMIC DISPATCH (ED) USING PSO The ED problem is to determine the optimal combination of power outputs of all generating units to minimize the total fuel cost while satisfying the load demand and operational constraints [8]. A. Objective Function: The economic dispatch problem is a constrained optimization problem and it can be mathematically expressed as follows:

where,

(3)

(4) Bij are constants called B coefficients or loss coefficients. C. Inequality Constraint: The maximum active power generation of a source is limited by thermal consideration. Unless, we take a generator unit offline it is not desirable to reduce the real power output below a certain minimum value Pmin. Pi,min Pi Pi,max (5) Where Pi,min is minimum power output limit, Pi,max is maximum power output limit of ith generator (MW). D. Representation of individual: For an efficient evolutionary method [9], the representation of chromosome strings of the problem parameter set is important. The proposed approach uses the equal system incremental cost cost as individual (particles) of PSO [2]. Each individual within the population represents a candidate solution for solving the economic dispatch problem. The advantage of using system Lambda instead of generator units output is that, it makes the problem independent of the number of the generator units and also number of iterations for convergence decreases drastically. This is particularly attractive in largescale systems. E. Evaluation Function: We must define the evaluation function for evaluating the fitness of each individual in the population. In order to emphasize the best chromosome and speed up convergence of the iteration procedure, the evaluation value is normalized into the range between 0 and 1. The evaluation function [5] adopted is

(6) (1) where ai, bi and ci are fuel cost coefficients, subjected to number of power system network equality and inequality constraints. B. Equality Constraint: The power balance constraint is an equality constraint that reduces the power system to a basic principle of equilibrium between total system generation and total system loads. Equilibrium is met only when the total system generation equals to the total system load (PD) plus the system losses (PLoss). where, k is a scaling constant (k = 50 in this study). III. EMISSION DISPATCH (ED) USING PSO Fossil-fuel fired electric power plants use coal, gas or combinations as the primary energy resource, and produce atmospheric emissions whose nature and quantity depend on the fuel type and its quality. Coal produces particulate matter such as ash and gaseous pollutants such as carbon oxides, sulphur oxides and oxides of nitrogen. The thermal energy dissipated in cooling water raises its temperature and may be considered as a pollutant. Hydro-plants produce no such emissions. Nuclear power produces radiation emissions, which are well contained. Most of the research work summarized aims at reducing oxides of sulphur SO2 and oxides of nitrogen NOx. A major effort

(2)

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

17

has been devoted in reducing one type of pollutants or a mixture of pollutants. The approaches are designed either to reduce the total production of emissions or to reduce the concentration of pollution at ground level at certain areas which depends on both emissions and meteorological factors. A. Objective function: The objective of emission dispatch is to minimize the total environmental degradation or the total pollutant emission due to the burning of fuels for production of power to meet the load demand [10]. Dispatch the power generation to minimize emissions instead of the usual cost objective of economic dispatch is economic and easy in operation [11]. The emission function can be expressed as the sum of all types of emissions such as NOx, SO2, particulate materials and thermal radiation with suitable pricing for each pollutant emitted. In this paper only NOx emission is taken into account, since it is more harmful than other pollutants. The NOx emission can be approximated as a quadratic function of the active power output from the generating units and it is shown in Fig 1.

candidate solution for solving the emission dispatch problem. IV. REACTIVE POWER OPTIMIZATION USING PSO The purpose of Reactive Power Dispatch (RPD) is mainly to improve the voltage profile in the system and to minimize the real power transmission loss while satisfying the unit and system constraints [14]. This goal is achieved by proper adjustment of reactive power control variables like generator bus voltage magnitudes (Vgi), transformer tap settings (tk), reactive power generation of the capacitor bank (Qci) [15]. A. Objective Function: The objective of RPD is to identify the reactive power control variables, which minimizes the real power loss (Ploss) of the system. This is mathematically stated as [16],

(10) subjected to the following constraints. B. Equality Constraints: These are load flow constraints such as The power balance constraints includes real and reactive power balances.

(11)

Fig 1: NOx Emission Function

B. Objective Function: The emission dispatch problem is a constrained optimization problem and it can be mathematically expressed as [12],

(12) Where i, j are the bus indices, PGi and QGi are active and reactive power generations at bus i respectively. C. Inequality Constraints: These constraints represent the system operating constraints. Load bus voltages (Vload), reactive power generation of generator (Qgi) and line flow limit (Sl) are variables, whose limits are satisfied by adding a penalty terms in the objective function. These constraints are formulated as, (i) Voltage limits (13) (ii) Generator reactive power capability limit (14) (iii) Capacitor reactive power generation limit (15)

(7) where, E is total emission release (Kg/hr) and i, i, i are emission coefficients of the ith generating unit subject to demand constraint (8) and generating capacity limits (9).

Pi,min Pi Pi,max

(8) (9)

C. Representation of Individual: The proposed approach uses the equal system incremental emission release emission as individual (particles) of PSO [13]. Each individual within the population represents a

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

18

(iv) Transformer tap setting limit (16) (v) Transmission line flow limit (17) D. Evaluation Function: Each particle consists of voltages, taps and shunts encoded in it. The size of each particle is equal to sum of number of voltages excluding slack bus voltage, taps, and shunts. The fitness function [17] employed is

in the system. The Lindex value varies in a range between 0 (no load) and 1 (voltage collapse) [21]. Stability Index of the system is computed as

(18) V. VOLTAGE STABILITY IMPROVEMENT USING PSO Voltage stability is concerned with the ability of a power system to maintain acceptable voltages at all nodes in the system under normal condition and after being subject to a disturbance [15]. A power system is said to have a situation of voltage instability when a disturbance causes a progressive and uncontrollable decrease in voltage level. An accurate knowledge of how close the actual systems operating point is from the voltage stability limit is crucial to operators [18]. Hence, to find a voltage stability index has become an important task for many voltage stability studies. These indices provide reliable information about proximity of voltage instability in a power system. Problem Formulation: Each particle consists of voltages, taps and shunts encoded in it. The size of each particle is equal to sum of number of voltages, taps, and shunts. The fitness function [17] employed is

VI. FUZZIFIED PSO FOR MULTI-OBJECTIVE CEED The real power optimization sub-problem minimizes fuel cost by controlling controllable generator outputs while keeping the generator bus voltages unchanged. The system losses, stability index and emission computed at this power dispatch are very high compared with the results obtained when respective ones are taken as objective. Similarly the reactive power sub-problem deals with minimization of total transmission loss of the system by controlling all the reactive power sources such as taps, shunts etc. When loss minimization is taken as objective total system losses reduces but cost, emission and stability indices are high. The emission dispatch sub problem minimizes total emission output from the fossil fuel plants by controlling the generator outputs. At this power output of generator, the cost, total system losses and stability index are high [22]. In the same way Stability index sub problem minimizes the index by controlling the PV bus voltages and thus improves the system stability limit. But the cost, emission and system losses are very high. Thus, results of all the four sub problems are conflicting with one other. Hence, multiobjective PSO technique is used to find the best compromise solution. In this section results of all the four sub problems are fuzzified using fuzzy minmax approach and then PSO is used to determine the final trade off solution from all these fuzzified values. A. Problem Formulation: Each particle consists of active power generations, voltages, taps and shunts excluding slack bus voltages encoded in it. The size of each particle is equal to sum of active power generations, number of voltages excluding slack bus, taps, and shunts. Assuming the decision maker (DM) has imprecise or fuzzy goals of satisfying each of the objectives, the multiobjective problem can be formulated as a fuzzy satisfaction maximization problem which is basically a min-max problem [23]. The task here is to determine the compromise solution for all the four optimization sub problems. The goal is to minimize G(X) = compromised solution of G1(X1), G2(X2), G3(X3), G4(X4) [24], while satisfying the set of constraints Ax < B. where G1(X) is Fuel cost minimization sub problem, G2(X) is Loss minimization sub problem, G3(X) is emission minimization sub problem, and G4(X) is L-index minimization sub problem. Let F1(Xi) be fuel cost in $/hr for ith control vector, F2(Xi) be losses in p.u. for ith control vector, F3(Xi) be stability index for ith control vector, and F4(Xi) be emission release in kg/hr for ith control vector. Let the individual optimal

(19) where, index is the stability index value. Stability index of power system is computed [19] based on the solution of the power flow equations. Here, L-index is used, which is a quantitative measure for the estimation of the distance of the actual state of the system to the stability limit [20]. The L index describes the stability of the complete system and is formulated as,

where ng is number of generators, n is number of buses

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

19

control vectors for the sub problems be X1 , X2 , X3 ,X4 respectively. To find, the global optimal control vector X such that,

The imprecise or fuzzy goal of the DM for each of the objective functions is quantified by defining their corresponding membership functions i as a strictly monotonically decreasing function with respect to the objective function f. Where i=1 to 4. In case of a minimization problem, i =0 or tends to zero, if fi > fmaxi and i =1 or tends to 1, if fi < fmini . where fmaxi and f mini are the unacceptable and desirable level for respectively. In the proposed approach, we have considered a simple linear membership function for f i because none of the objectives have very strict limits. The membership function, i for ith objective function is shown in Fig 2.

2) Emission Dispatch using PSO : Table II gives all objective function values considering emission dispatch as single objective problem and the optimum value obtained is 229.145 (Kg/hr). The total system generation is 287.804 MW.

3) Reactive Power Optimization using PSO : The lower voltage magnitude limits at all buses are 0.95 p.u. and the upper limits are 1.1 for all the PV buses and 1.05 p.u. for all the PQ buses and the reference bus. The lower and upper limits of the transformer tapping are 0.9 and 1.1 p.u. respectively. Table III presents all objective function values considering loss minimization as single objective optimization problem. Table IV and V gives the positions of tap changing transformers and shunt susceptance values respectively.

VII. RESULTS AND DISCUSSION The proposed approach is tested on IEEE 30 bus system [25]. The PSO parameters used in CEED case studies are: number of particles 60, learning factors C1=2.05, C2=2.05, weight factor w=1.2, constriction factor K=0.7925. Maximum number of iterations=100. First, each objective is optimized individually at a time and then four objectives are optimized simultaneously using fuzzified PSO. 1) Economic Dispatch using PSO : Table I shows all objective function values considering fuel cost minimization as single objective optimization subproblem and the optimum value obtained is 806.498 MW. The total system generation is 293.984 MW.

4) Voltage Stability Improvement using PSO : Table VI gives all objective function values considering L-index minimization as single objective optimization problem. Table VII and VIII gives the positions of tap changing transformers and shunt susceptance values respectively.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

20

5) Fuzzified PSO for Multi-objective Optimization : From the above single objective optimization techniques it is observed that when one objective is optimized it will give optimum value for that objective but other objective function values are deviating from optimum. Hence, there is a

VIII. CONCLUSIONS In this paper, an approach to solve multi-objective optimization problem which aims at minimizing fuel cost, real power loss, emission release and improving stability index of the system has been proposed. The proposed algorithm has been tested on IEEE 30 bus system. The four objectives economic dispatch, emission dispatch, stability index and reactive power optimization are solved individually and the results from these individual optimizations are fuzzified and final best compromise solution is obtained. The multi-objective problem is handled using the fuzzy decision satisfaction maximization technique which is an efficient technique to obtain trade off solution in multi-objective problems. The proposed approach satisfactorily finds global optimal solution within a small number of iterations. REFERENCES
[1] E. H. Chowdhury, S. Rahrnan, A Review of Recent Advances in Economic Dispatch, IEEE Trans. Power Syst., vol. 5, no. 4, pp. 1248- 1259, Nov. 1990. J. Wood, B. F. Wollenberg, Power Generation, Operation and Control, John Wiley and Sona, Inc., New York, 2004. O. Alsac and B. Stott, Optimal Load Flow with Steady State Security, IEEE Trans. Power Apparatus and Syst., vol. 93, no.3, pp. 745-751, May/June 1974. M. R. AlRashidi and M. E. E. Hawary, A Survey of Particle Swarm Optimization Applications in Electric Power Systems, IEEE Trans.Evolutionary Computation., vol. 13, no. 4, pp. 913-918, Aug. 2009. J. Kennedy and R. Eberhart, Particle swarm optimization, in Proc. IEEE Int. Conf. Neural Networks (ICNN95), vol. 6, pp. 942-1948, Perth, Australia, 1995. M. A. Abido, Optimal power flow using particle swarm optimization, Electrical Power and Energy Systems, vol. 24, pp. 563571, 2002. Mozafari, T. Amraee, A. M. Ranjbar and M. Mirjafari, Particle Swarm Optimization method for Optimal Reactive Power Procurement considering Voltage Security, Scientia Iranica, vol. 14, no. 6, pp. 534545, Dec. 2007. Z. L. Gaing, Particle Swarm Optimization to Solving the Economic Dispatch Considering the Generator Constraints, IEEE Trans. Power Syst., vol. 18, no. 3, pp. 1187-1195, Aug. 2003. J. B. Park, K. S. Lee, J. R. Shin, K. Y. Lee, A Particle Swarm Optimization for Economic Dispatch with Nonsmooth Cost Functions, IEEE Trans. Power Syst., vol. 20, no. 1, pp. 34-42, Feb. 2005. Keib, H. Ma, J. L. Hart, Economic dispatch in view of the Clean Air Act of 1990, IEEE Trans. Power Syst., vol. 9, no. 2, pp. 972- 978, May 1994. J. H. Talaq, F. E. Hawary, and M. E. E. Hawary, A Summary of Environmental/Economic Dispatch algorithms, IEEE Trans. Power Syst., vol. 9, no. 3, pp. 1508-1516, Aug. 1994. S. Kumar, K. Dhanushkodi, J. J. Kumar, C. K. C. Paul, Particle Swarm Optimization Solution to Emission and Economic Dispatch Problem, IEEE TENCON, 2003. P. Dutta , A. K. Sinha, Environmental Economic Dispatch constrained by voltage stability using PSO, IEEE Electrical Engineering, IIT Kharagpur., 2006. k. Iba, Reactive power optimization by genetic algorithms, IEEE Trans. Power Syst., vol. 9, no. 2, pp. 685-692, May, 1994. Q. H Wu and J. T. Ma, Power system optimal reactive power dispatch using Evolutionary programming, IEEE Trans. Power Syst., vol.10, no.3, pp. 12431249, 1995.

[2] [3]

[4]

conflicting behavior between one objective with reference to other objective. To overcome this, multi-objective optimization using Fuzzified PSO technique is proposed in this paper. The compromise solution is given in Tables IX, X and XI. The total system generation is 288.035 MW, total load 283.4 MW.

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

21

[16] W. Zhang, Y. Liu, Reactive Power Optimization Based on PSO in a Practical Power System, IEEE Power Engineering Society General Meeting, pp. 239-243, 2004. [17] H. Yoshida, K. Kawata, Y. Fukuyama, Y. Nakanishi , A Particle Swarm Optimization For Reactive Power And Voltage Control Considering Voltage Stability, IEEE International Conference on Intelligent System Applications to Power Systems, Rio de Janeiro, April 1999. [18] P. Kundur, Power System Stability and Control, Mc GrawHill, Inc., New York, 1994. [19] Reis, F. P. M. Barbosa, A Comparison of Voltage Stability Indices, IEEE MELECON Benalmdena (Mlaga), Spain, May 2006. [20] P. Kessel and H. Glavitsch, Estimating the Voltage Stability of a Power System, IEEE Trans. Power Delivery, vol. 1, no. 3, pp. 346-355, 1986. [21] Barbier and J. P. Barret, An analysis of phenomena of voltage collapse on transmission system, Revue Gdndrale de ldlectricit6, Special CIGRE issue., pp 321, July 1980. [22] M. A. Abido, A novel multiobjective evolutionary algorithm for environmental / economic power dispatch, Electric Power Syst. Research, vol. 65, pp. 71-81, 2003. [23] S. Rajesekaran and G. A. V. Pai, Neural Networks, Fuzzy Logic, and Genetic Algorithms Synthesis and Applications, Pretice Hall Pvt Ltd. 2003. [24] Y. H. Song, G. S. Wang, P. V. Wang, A. T. Johns, Prediction And evaluation of the performance of wind-diesel systems, IEE Proceedingsof Generation, Transmission, Distribution., vol. 114, no. 4, pp. 377-382, 1997. [25] [Online].Available: http://www.ee.washington.edu/research/pstca.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

22

Congestion Management by Rescheduling of Generators based on Particle Swarm Optimization


1

Sundeep Kumar M, 2C. P. Gupta


2

1, 2
1

Indian Institute of Technology Roorkee, India


smedeti@yahoo.co.in

sundeep.2022@gmail.com,

AbstractIn a deregulated electricity market, it may always not be possible to dispatch all of the contracted power transactions due to congestion of the transmission corridors. System operators try to manage congestion, which otherwise increases the cost of the electricity and also threatens the system security and stability. In this paper, congestion management by rescheduling of generators has been proposed. For optimal selection of participating generators a technique based on generator sensitivities to the power flow of congested lines is used. Also this paper proposes an algorithm based on Particle Swarm Optimization (PSO) which minimizes the deviations of rescheduled values of generator power outputs from scheduled levels. The effectiveness of the proposed methodology has been analyzed on IEEE 30-bus and IEEE 118-bus systems. Keywords-Generator sensitivities, optimal rescheduling, particle swarm optimization, congestion management.

I. INTRODUCTION Electrical power utilities throughout the world are currently undergoing major restructuring process and are adopting the deregulated market operation. Competition has been introduced in power system around the world based on the idea that it will increase the efficiency of this industrial sector and reduce the cost of electrical energy for all consumers. This regime has resulted in the entry of a large number of new players and the proliferation in the number of transactions. Congestion occurs whenever the transmission network is unable to accommodate all the desired transactions due to the violation of one or more constraints for the resulting state under both the base case and a set of specified contingencies. The open access transmission regime results in the more intensive use of the transmission system, which, in turn, leads to more frequent congestion. Ensuring that the power system operates within its limits is vital to maintain power system security, failing which can result in widespread blackouts with potentially severe social and economic consequences. Transmission congestion management, that is to alleviate the amount of power flow in the congested transmission lines by changing the flow pattern. This process of re-dispatch can be done by modifying generation, or modifying load, or both, without compromising the system security and service quality. In the paper [3], it is proposed that prioritization of electricity transactions and related curtailment strategies in a system where pool and bilateral/multilateral dispatches coexist. In this proposed model, willingness-to-pay-toavoid-curtailment factors are introduced to deal with congestion problem and this gives insight into the nature of

competition in the emerging transmission market. In the paper proposed by Jain Fu et al [5] presented a combined framework for service identification and congestion management. The ideal objective function is to maximize the overall profit of all market participants. In both the above mentioned papers [3] and [5] congestion is relieved by load curtailment, but because of load curtailment which results in subjective factors such as importance of load, loss of good will, etc. in addition to commercial factors such as loss of revenue. In the paper by B.K.Talukdar et al [4] proposed a method for alleviation of line over load, where the cost of generation rescheduling and cost of load shedding are considered with the sensitivity of the over loaded lines to the bus injections. To avoid selection of less sensitive generators or the ones with high cost, defined functions which depend on design variables which should be chosen by the operator. However an optimal selection of the design variables is essential for regulating the number of participating generators in this work. In the paper [7] proposed by S.P.Singh et al proposed a technique for reducing the number of participating generators and optimum rescheduling of their outputs while managing congestion in a pool at minimum rescheduling cost. In this the slack bus must be first bus only for obtaining generator sensitivity values. The paper presented by S.C.Srivastava et al [2] proposed a new zonal/cluster based congestion management approach. However, in this method it is necessary to compute the sensitivity values for all the buses in the system which, given a practical power system, calls for a large amount of computational effort. The main intent of the present paper is to propose a technique for reducing the number of participating generators and optimum rescheduling of their outputs while managing congestion in a pool. In a congested power system, some generators may not be willing to participate in the congestion management process for their own economic interest or some other operational constraints. Thus, the practicalities of implementing modifications to existing schedule of generation imply that it may be advantageous to implement a good sub-optimal schedule involving only a few buses rather than a truly optimal one involving many. Hence, there is a need for methods that generate such solutions. To optimally select the generators participating in congestion management, the sensitivities of the generators are used. The second major purpose of the present paper is to explore the ability of particle swarm optimization technique in solving the congestion management problem.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

23

The congested system is modeled as an optimization problem and solved by using particle swarm optimization. II. PARTICLE SWARM OPTIMIZATION Particle swarm optimization (PSO), first introduced by Kennedy and Eberhart [8], is one of the modern heuristic algorithms. PSO is motivated by social behavior of organisms such as fish schooling and bird flocking. In PSO, potential solutions called particles fly around in a multidimensional problem space. Population of particles is called swarm. Each particle in a swarm flies in the search space towards the optimum or a quasi-optimum solution based on its own experience, experience of nearby particles, and global best position among particles in the swarm. PSO is basically developed through simulation of a flock of birds in two-dimension space. The position of each agent is represented by XY-axis position and the velocity is expressed by vx (the velocity of X-axis) and vy (the velocity of Y-axis). Modification of the agent position is realized by using the position and the velocity information. A flock of agents optimizes a certain objective function. Each agent knows its best value so far (Pbest) and its XY position. Moreover, each agent knows the best value in the group (Gbest) among Pbest, namely the best value so far of the group. The modified velocity of each agent can be calculated using the current velocity and the distance from Pbest and Gbest as shown below:

in real power flow in a transmission line k connected between bus i and j due to change in power generation by generator g can be termed as generator sensitivity to congested line (GS). Mathematically, GS for line k can be written as

Where Pij is the real power flow on congested line-k; PGg is the real power generated by the gth generator. The basic power flow equation on congested line can be written as Where Vi and i are the voltage magnitude and phase angle respectively at the ith bus ; Gij and Bij represent, respectively, the conductance and susceptance of the line connected between buses i and j; neglecting P-V coupling, (4) can be expressed as The first terms of the two products in (6) are obtained by differentiating (5) as follows

Where, vij vij+1 r1,r2 Xij Pbest Gbest C1,C2 w

The active power injected can be represented as Ps = PGs PDs (10) Where PD, is the active load at bus-s, PS can be expressed as : Current velocity of agent i at iteration k, : modified velocity of each agent i : random number between 0 and 1 : Current position of agent i at iteration k : Positional best of agent i : Global best of the group : weight coefficients for each term : weight function for velocity of agent i :

Where n is the number of buses in the system. Differentiating (12) w.r.t. s and t, the following relations can be obtained

Using the above equation, a certain velocity the gradually gets close to Pbest and Gbest can be calculated. The currents position (searching point in the solution space) can be modified by the following equation: The particles continue flying and seeking solution and hence the algorithm continues until a pre-specified number of maximum iterations are exceeded or exit criteria met. The accuracy and rate of convergence of the algorithm depends on the appropriate choice of particle size, maximum velocity of particles and inertia weight. III. FORMULATION OF GENERATOR SENSITIVITY By using generator sensitivity we select generators actually participating in congestion management. A change

Neglecting P-V coupling, the relation between incremental change in active power at system buses and the phase angles of voltages can be written in matrix form as Where

Thus

Where

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

24

To find the values of equation (6), [M] needs to be found out. However, [H] s a singular matrix of rank one deficiency. So it is not directly invertible. The slack bus is considered as the reference node. The elements of slack bus number row and column of [H] is eliminated to obtain a matrix [H-1] which can be inverted to obtain the matrix [M-1], where (.)-1 represents a matrix with its slack bus number row and column deleted. Using these relations the following equation can be obtained: The actual vector [] can be found by simply adding the element s to (19) and since the change in phase angle of slack bus is zero. Accordingly we obtain [M] by simply replacing the slack bus number row and column with zeros. Hence by knowing the values of matrix [M] we can find generator sensitivities (GS) by substituting in (6). It is to be noted that the generator sensitivity values obtained are with respect to the slack bus as the reference. So the sensitivity of the slack bus generator to any congested line is the system always zero. The system operator selects the generators having non uniform and large magnitudes of sensitivity values as the ones most sensitive to the power flow on the congested line and to participate in congestion management by rescheduling the power outputs. Based on the bids received from the participant generators, the amount of rescheduling is computed by solving the following optimization problem:

Step1: Particles are generated and initialized with values of position and velocity. Each particle has an N dimension where N denotes the number of participating generators, and the values of these N variables are the amount of rescheduling required by generators to manage congestion. Step2: Equation (23) is tested based on the system states represented by an individual particle. If that particle does not satisfy the equality constraints, it is regenerated. Step3: The binding constraints fitness values for the particles are determined using equations (21) and (22). If a particle does not satisfy the fitness requirement, it is regenerated. Step4: The optimal objective fitness values are calculated for all the particles using equation (20). Step5: Calculate current position of each particle using binding constraints fitness value and optimal objective fitness values. Current position (Sik) = optimal obj. fitness value + binding constraints fitness value

Step6: The values of position best and global best are determined. Position and velocities of particles are updated using equations (1) and (3). Step7: If the maximum number of iterations is exceeded or some pre-specified exit criteria is satisfied, the program is stopped. Else return Step 2. V. RESULTS AND DISCUSSION The congestion management by rescheduling of generators based on PSO, has been implemented using MATLAB programming language. The performance of the algorithm has been studied on IEEE 30-bus, 118-bus systems [12]. Modified IEEE 30-bus System The IEEE 30-bus system consists of six generators and 24 load buses. For the line to be congested, the line connected between buses 5 and 7 is removed. Due to the outage of line, one of the lines has been found to be congested, that is between buses 1 and 2. The amount of power flow along the congested line 1-2 is shown in Table I.
TABLE I: CONGESTED LINE DETAILS OF 30-BUS SYSTEM

Subject to

Where Pg is the real power adjustment at bus-g. Fk0 is the line flow caused by all contracts requesting the transmission service. Fkmax is the line flow limit of the line connecting bus-i and bus-j. Ng is the number of participating generators, nl is the number of transmission lines in the system. IV. SOLUTION OF RESCHEDULING VALUES USING PSO Conventional methods of solution of OPF are based on search direction determined from derivative of the function. Therefore it becomes imperative to express the problem in the form of continual differentiable function; otherwise, these methods become less efficient. In order to overcome this problem, the present paper solves the optimization problem using particle swarm optimization. PSO Algorithm for Generator Rescheduling

Congested Power Flow Line Limit Line (MW) (MW) 1-2 181.459 180 The amount of power flow above Line limit is 1.459 MW. The values of generator sensitivities computed for the congested line 1-2 are presented in Table II.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

25

TABLE II: GSS FOR IEEE 30-BUS SYSTEM (CONGESTED LINE 1-2) S. No. Bus No. GS 1 1 0.0000 2 2 -0.1872 3 5 -0.1564 4 8 -0.0261 5 11 -0.0023 6 13 0.00 12

A plot between the Generator sensitivity with respect to the serial number of generators is shown in Fig1.

Modified IEEE 118-bus System The IEEE 118-bus system consists of 54 generators and 64 load buses. There are 186 number of lines. For the line to be congested, the line connected between buses 89 and 92 is removed. Due to the outage of line, one of the lines has been found to be congested, that is between buses 85 and 89. The amount of power flow along the congested line 89-85 is shown in Table V.
TABLE V: CONGESTED LINE DETAILS OF 118-BUS SYSTEM

Fig1: Plot of Generator Sensitivities (GS) for IEEE 30-bus system

It can be seen from the Table II as well as the Fig that, generators numbered 2, 3 and 4 have the magnitudes of the sensitivity values are much larger than compared to other. Hence the generator buses selected for congestion management based on GS values are 1, 2, 5 and 8 i.e., solving the congestion management problem using PSO has been proceeded with. Generators selected for participation in congestion management are asked to reschedule their outputs optimally on the basis of their bids so that the cost of rescheduling gets minimized. The rescheduled values are obtained by solving the optimization problem discussed in Section IV based on PSO algorithm which is shown in Table III.
TABLE III: GENERATOR RESCHEDULING FOR CONGESTION MANAGEMENT IN LINE (1-2) OF 30-BUS SYSTEM

Power Flow Line Limit (MW) (MW) 89-85 87.56 80 The amount of power flow above Line limit is 7.56 MW. The values of generator sensitivities computed for the congested line 89-85 are presented in Table VI. The larger values of sensitivity are bolded in the above Table VI. A plot between the Generator sensitivity with respect to serial number of generators is shown in Fig2.

Congested Line

Fig .2: Plot of Generator sensitivities (GS) for IEEE 118-bus

S. No. 1 2 3 4

Gen. Bus No. 1 2 5 8

Gen Rescheduling (MW) -11.57 +1.71 +4.65 +5.20

The flow change along the line 1-2 after rescheduling the generator outputs at selected generators is shown in Table IV.
TABLE IV: FLOW CHANGE THROUGH CONGESTED LINE 1-2 BEFORE AND AFTER RESCHEDULING

It can be seen from the Table VI as well as the Fig that, generators numbered 38, 39, 41, 42 and 43 have the magnitudes of the sensitivity values are much larger than compared to other. Hence the generator buses selected for congestion management based on GS values are 69, 85, 87, 90, 91 and 92 i.e., solving the congestion management problem using PSO has been proceeded with. Thus, while the number of generators in the system is 54, the number of generators participating in congestion management is only 6. The rescheduled values are obtained by solving the optimization problem discussed in Section IV based on PSO algorithm. The amount of rescheduling values which are obtained at that computation which successfully relieved congestion is shown in Table VII.
TABLE VII: GENERATOR RESCHEDULING FOR CONGESTION MANAGEMENT IN LINE (89-85) OF 118-BUS SYSTEM

S. No. 30 38 39

Line connecte d between buses 1-2

Line flow (MW) (Before Rescheduling ) 181.459

Line flow (MW) (After Reschedulin g) 171.60

Line Limit (MW)

Gen. Bus No. 69 85 87 90

Generator Rescheduling (MW) +10.86 30.00 10.78 -8.00

180 41

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

26

42

91

-20.00

43 92 -23.65 The flow change along the congested line 89-85 after rescheduling the generator power outputs at selected generators is shown in Table VIII.
TABLE VIII: FLOW CHANGE THROUGH CONGESTED LINE 89-85 BEFORE AND AFTER RESCHEDULING

Line connected between buses 89-85

Line flow (MW) (Before Rescheduling) 87.56

Line flow (MW) (After Rescheduling) 72.72

Line Limit (MW) 80

VI. CONCLUSION In this paper, a simple method for congestion management using Generator Rescheduling has been proposed in which optimal number of generators is selected based on their sensitivities to the power flow of the congested line followed by corrective rescheduling. The problem of congestion is modeled as an optimization problem and solved by particle swarm optimization technique. The effectiveness of the method is tested on the IEEE 30-bus and 118-bus systems for congestion management and consistent results are found and presented. The robustness of the algorithm is demonstrated by solving different networks of various sizes and complexities with equal performance. Since the convergence of the PSO algorithm depends on the appropriate selection of particle size and inertia weight, improper choice of these parameters may lead to inferior results. However, test results reveal that the proposed implementation is effective in managing and alleviating congestion. REFERENCES

Yog Raj Sood, Narayan Prasad Padhy, and H.O.Gupta, Wheeling of Power Under Deregulated Environment of Power System-A Bibliographical Survey, IEEE Trans. on Power Sys., vol. 17, No. 3, Aug. 2002, pp. 870-878. [2] Ashwani Kumar, S.C.Srivastava, and S.N.Singh, A Zonal Congestion Management Approach using Real and Reactive Power Rescheduling, IEEE Trans. on Power Sys., vol. 19, No. 1, Feb. 2004, pp. 554-562. [3] R. S. Fang and A.K.David, Transmission Congestion Management in an Electricity Market, IEEE Trans. on Power Sys., vol. 14, No. 3, Aug. 1999, pp. 877-883. [4] B. K. Takuldar, A.K.Sinha, S.Mukhopadhyay, and A.Bose, A computationally simple method for cost-efficient generation rescheduling and load shedding for congestion management, Int.J.Elec. Power Energy Sys., vol. 27, No. 5, Jun.-Jul. 2005, pp. 379-388. [5] Jian Fu and John W.Lamont, A combined Framework for Service Identification and Congestion Management, IEEE Trans. on Power Sys., vol. 16, No. 1, Feb. 2001, pp. 56-61. [6] Loi Lei Lei, Power System Restructuring and Deregulation: Trading, Performance and Information Technology, John Wiley & Sons, ISBN 0-471-49500-X, 1st Edition, 2001. [7] Sudipta Dutta and S.P.Singh, Optimal Rescheduling of Generation for Congestion Management Based on Particle Swarm Optimization, IEEE Trans. on Power Sys., vol. 23, No. 4, Nov. 2008, pp. 1560-1569. [8] A. Kennedy and R.Eberhart, Particle Swarm Optimization, in Proc. IEEE Int. Conf. Neural Networks, vol. IV, Nov. 29-Dec. 1 1995, pp.1942-1948. [9] Shu Tao and George Gross, A congestion-Management Allocation Mechanism for Multiple Transaction Networks, IEEE Trans. on Power Sys., vol. 17, No. 3, Aug. 2002, pp. 826-833. [10] Hadi Sadat, Power System Analysis, Tata McGraw-Hill Publishing Company Limited, ISBN 0- 07-012235-0, 13th Reprint, 2008. [11] Chai Chompoo-inwai , Chitra Yingvivatanapong, Pradit Fuangfoo, and Wei-Jen Lee, Transmission Congestion Management During Transition Period of Electricity Deregulation in Thailand, IEEE Trans. on Industry Applications, vol. 43, No. 6, Nov./Dec. 2007, pp. 1483-1490. [12] L. L. Freris and A. M. Sasson, Investigation of the load flow problem, Proc . Inst. Elect. Engg., vol. 115, No. 10, pp. 14591466, 1968. [1]

TABLE VI: GSS FOR IEEE 118-BUS SYSTEM (CONGESTED LINE 89-85) S. No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Gen. Bus No. 1 4 6 8 10 12 15 18 19 24 25 26 27 31 32 34 36 40 GS +0.0001 -0.0002 +0.0001 0.0000 0.0000 -0.0001 +0.0053 -0.0014 -0.0012 +0.1048 +0.0000 +0.0000 -0.0042 +0.0036 +0.0037 -0.0000 0.0000 0.0000 S. No. 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Gen. Bus No. 42 46 49 54 55 56 59 61 62 65 66 69 70 72 73 74 76 77 GS 0.0000 0.0000 0.0000 -0.0000 -0.0000 -0.0000 +0.0003 +0.0000 +0.0000 -0.0006 -0.0000 0.0000 -0.0478 +0.2596 +0.0635 -0.3241 +0.2059 -0.1090 S. No. 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 Gen. Bus No. 80 85 87 89 90 91 92 99 100 103 104 105 107 110 111 112 113 116 GS -0.022 -0.753 -0.728 +0.198 +1.367 +1.816 +0.580 +0.013 +0.032 +0.014 +0.023 +0.018 +0.029 -0.050 -0.057 -0.059 +0.001 -0.001

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

27

Application Research Based On Sweep Frequency Response Analysis in Condition Monitoring of Power Transformers
1

Surinder Singh, 2Y.R Sood, 3Hashmat Malik


1, 2, 3

N.I.T, Hamirpur, Himachal Pradesh, India


1

surindersabb@gmail.com 2 yrsood@gmail.com

Abstract The measurement of the frequency response on power transformers is a diagnostic test technique for detecting winding displacement and deformation (among other mechanical and electrical failures), which are among the most important causes of mechanical failure in transformers. There are two different methods to carry out the measurement of frequency response: Sweep Frequency Response Analysis - SFRA and Low Voltage Impulse - LVI. SFRA presents some important advantages over LVI such as: higher signal to noise ratio, bigger repeatability and reproducibility and less requirements regarding measurement equipment. It is based on comparison between; 1) earlier measurement on same transformer, 2) measurement on sister transformers, or 3) phase to phase comparison on same transformer with higher signal to noise ratio, higher repeatability and reproducibility and less requirement regarding measuring equipment. SFRA is an electrical test that provides information relating to transformer mechanical integrity. This paper details the use of sweep frequency response as a diagnostic tool to detect winding deformation and core displacement in power transformers. Practical case studies are presented that demonstrates the effectiveness of this technique. Keywords- Sweep frequency response analysis, Condition assessment, Power Transformers, Winding Deformation

I. INTRODUCTION The high percentage of mechanical failures in power transformers caused by winding displacement and deformation makes necessary the implementation of a sensitive technique for the detection of this type of damage. The loss of mechanical integrity in the form of winding movement and core displacement in power transformers can be attributed to the large electromechanical forces due to fault currents, winding shrinkage causing the release of the clamping pressure and during transformer transportation and relocation. These winding movement and core displacement if not detected early will typically manifest into a dielectric or thermal fault [1]. This type of fault is irreversible with the only remedy been rewinding of the phase or a complete replacement of the transformer. It is therefore imperative to check the mechanical integrity of aging transformers periodically and particularly after a short circuit event to provide early warning of impending failure; hence an early warning detection technique of such a phenomena is essential. The frequency response analysis (FRA)

technique is widely used because of its high sensibility in the detection of this type of damage. It is based on the concept that windings deformation and displacement are reflected in a change of the parameters L, C, and R of the equivalent circuit of the transformer, modifying its frequency response [3]. There are two methods to obtain the frequency response: SFRA and LVI (Low voltage Impulse test). SFRA method uses a sweep generator to apply sinusoidal voltages at different frequencies to one terminal of a transformer winding. Amplitude and phase signals obtained from selected terminals of the transformer are plotted directly as a function of frequency. During a LVI test, a low voltage impulse is applied to one end of the winding and the output voltage or current of the other end of the winding are registered. The Fast Fourier Transform is applied to both measurements in order to obtain the amplitude and phase of the transfer function [4]. SFRA presents some important advantages over LVI such as: higher signal to noise ratio, higher repeatability and reproducibility, and less requirements regarding measuring equipment. An important disadvantage of the SFRA compared with LVI is the time required for the measurement. The necessary time for performing an SFRA test (typically several minutes) is related to the bandwidth and number of spot frequencies, which are not universally defined. II. POWER TRANSFORMERS AND MECHANICAL INTEGRITY Power transformers are specified to withstand the mechanical forces arising from both shipping and subsequent in-service events, such as faults and lightning. Due to transportation damage can occur if the clamping and restraints are inadequate; such damage may lead to core and winding movement. The most severe in-service forces arise from system faults, and may be either axial or radial in nature. If the forces are excessive, radial buckling or axial deformation can occur. With a core form design the principal forces are radially directed, whereas in a shell form unit they are axially directed, and this difference is likely to influence the types of damage found. Once a transformer has been damaged, even if only slightly, the ability, to withstand further incidents or short circuits is reduced. There is clearly a need to effectively identify such damage. During a field inspection, the oil has to be drained and confined space entry rules apply. So

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

28

little of the winding is visible, often little damage is seen other than displaced support blocks. Often, a complete tear down is required to identify the problem. An alternative method is to implement field-diagnostic techniques that are capable of detecting damage, such as SFRA [5]. There is a direct relationship between the geometric conFiguration of the winding and core within a transformer and the distributed network of resistances, inductances, and capacitances that make it up [6]. This RLC network can be identified by its frequency dependent transfer function. Changes in the geometric conFiguration alter the impedance network, and in turn alter the transfer function. Changes in the transfer function will reveal a wide range of failure modes. SFRA allows detection of changes in the transfer function of individual windings within transformers and consequently indicate movement or distortion in core and windings of the transformer [7]. III. DIAGNOSTIC SIGNIFICANCE OF FREQUENCY RANGES Diagnostics of frequency ranges are discussed on two levels: Per-phase Open Circuit Measurement Short Circuit Measurement A. Per-phase Open Circuit Measurement As the name implies, the per-phase measurement targets the individual phase of a given winding. At low frequencies, the influence of capacitance is negligible and the winding behaves as an inductor. Therefore, the attenuation (described by the magnitude of the transfer function) and the phase shift (described by the phase of the transfer function) of the low-frequency sinusoidal signals, passing through the winding, are determined by inductive and resistive nature of the network [8]. The inductive characteristics are determined by the magnetic circuit of the core and the resistive characteristics are dominated by the resistance of the output measuring cable. An example of transfer function magnitude and phase for a per-phase measurement is shown in Fig1 and Fig2. In Fig2 the phase angle is around 80 degrees, indicating the inductive nature of the total impedance (in the region below 1 kHz). For a three-legged core-type unit, the magnetic flux coupled with the outer phase (H1-H3 or H3-H2 in Fig1 faces a different reluctance than the flux coupled with the middle phase (H2-H1 in Fig1). Therefore, the corresponding magnitude traces, in the low frequency range, differ as well, i.e., the traces for the two outer phases correlate very closely and are shifted from the middlephase trace [4]. The presence of the residual magnetism may have an effect on relationship between the traces. This is the same phenomenon that, during exciting current and loss measurement, creates a pattern of two high similar and one lower reading under normal conditions and a slightly distorted pattern in the presence of residual magnetism.

Fig 1: Per-Phase Measurements Magnitude of the Transfer Function

Fig. 2: Per-Phase Measurements Phase of the Transfer Function

As the frequency of the input signal increases, the capacitive effects begin to dominate and the phase angle quickly becomes close to +90 degrees (in the region above 1 kHz). Now, the attenuation and the phase shift of the high-frequency sinusoidal signals, passing through the winding, are determined by inductive and capacitive nature of the network. However, in high frequency region, the inductive characteristics are determined by the leakage flux coupling and the capacitive characteristics are determined by the various capacitance elements associated with individual turns [9]. The propagation characteristic of the winding becomes complex as a result of the many resonance frequencies found in the high-frequency range. However, since the winding responses become less dependent on the magnetic circuit of the core, the traces of the three phases converge and become quite similar. As the frequency increases even further (over 100 kHz in Fig1), the sinusoidal signals travel mostly outside the winding and reflect the other elements found in the transformer, e.g., leads, support insulation, etc. The magnitude and the phase of the transfer function in that frequency region are influenced by the inductive-capacitive-resistive nature of these elements. Although most of the low-frequency magnitude responses exhibit a typical shape, there are no typical form responses in the high-frequency region. These vary greatly with design of the unit [1]. Therefore, the frequency ranges noted in description of Fig1 and Fig2, are different for different units.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

29

B. Short Circuit Measurement The aim of this measurement is to allow direct comparison between the three phases of a three phase transformer where no prior measurements exist. By making a measurement on one winding with another winding short circuited, the effect of the core at low frequencies is removed. The resulting response is that for a large inductor with no core. The responses for all three phases should be very similar at low frequencies. The theory behind the short-circuit measurement is straightforward. Any two winding transformer may be modeled at low frequencies by a simple T-model, as shown in Fig3 [10]. Normal test on HVs - LVs float

A. Tests on high voltage winding and low voltage winding with all other terminals open

Fig 5: Measurement is taken between high voltage winding and low voltage winding

B. Test on high voltage winding and neutral with low voltage winding and neutral short circuited

Fig. 3: T Model of Transformer Winding

Normal test on HVs - LVs float


Fig. 6: Measurement is taken between high voltage winding and neutral with low voltage winding and neutral short circuited

C. Test on high voltage winding and neutral with tertiary short circuited

Fig. 4: T Model with LV Short

The impedance of the windings is small, while the impedance of the core to ground is extremely high. This means that for any input signal, the response is dominated by the core [11]. By adding a short to the LV side as shown in Fig4, the effect of the core is removed and the response is dominated by the windings, which are predominantly inductors at low frequency. The response of an inductor is to have a low db response at low frequency with an inductive roll off as frequency rises. All three phases of a transformer have similar winding inductances, which mean their responses should be very similar. IV. CASE STUDIES SFRA Results: - Single-phase step-up transformer 26.67/33 MVA having current in HV winding 262.4A and current in LV winding 437.39A, make Crompton greaves. The Guaranteed Load Losses 67 KW and Guaranteed No Load Losses 13.5 KW. As there are no signature signals and no sister unit is available for the comparison of the transformer, the analysis is done with phase to phase comparison of the transformers to know the mechanical integrity of the transformer.

Fig. 7: Measurement is taken between high voltage winding and neutral with tertiary short circuited

D. Test on low voltage winding and neutral with tertiary short circuited

Fig. 8: Measurement is taken between low voltage winding and neutral with tertiary short circuited

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

30

E. Test on low voltage winding and neutral with all other terminals open

relate to some internal lead movement, but the results are acceptable. SFRA showed over all good results and the transformer is OK as far as mechanical integrity of the transformer is concerned. SFRA results indicate consistent mechanical geometry of all units, obviously core movement or winding deformation is not present. VI. CONCLUSION Sweep Frequency Response Analysis has been applied successfully as a diagnostic method for detecting mechanical deformation in power transformer. This paper discusses the case of Power transformer and various SFRA signatures obtained. It has been shown that SFRA responses of different combination of transformer winding can highlight completely the mechanical integrity of power transformer by their careful comparison. Changes in frequency response as measured by SFRA techniques may indicate a mechanical deformation inside the transformer, the cause of which then needs to be identified and investigated, so that the appropriate action can be taken before outages occurs. REFRENCES
[1] [Tony McGrail Transformer Frequency Response Analysis an Introduction, Doble Report 2005, Doble Engineering. [2] K. Fesser, J.christian, C . Neumann, U Sundemann, T. Liehfied, et.al, The Tranfer Function Method for Detection of Winding Displacement on Power Transformers Afer Transport, Short Circuit or 30 Years of Service , ClGRE Paris, paper no. 12133-04,2000 [3] E. P. Dick and C. C. Erven, Transformer diagnostic testing by frequency response analysis, IEEE Trans. Power App. Syst., vol. PAS-97, no. 6, pp. 21442153, Nov./Dec. 1978 [4] J. SecueLech , E. Mombello Approach for Determining a Reliable set of Spot Frequencies to be used during a Sweep Frequency Response Analysis (SFRA) for Power Transformer Diagnosis IEEE, 2008. [5] W. : Detecting transformer winding damage, Electrical Review, Vol 179, pp. 768, 1966 [6] Abetti P. A. and Magimiss F. J.:Natural frequencies of coils and windings determined by equivalent circuit, Trans. of AIEE, Vol. 72, pp 495-504, 1953 [7] Degeneff R. C. : A general method for determining resonances in transformer windings, IEEE Trans. on PAS, Vol. PAS-96, No. 2, pp. 423, 1977. [8] P.N. Jarman. "Transformer winding movement and fault detection" Assessment of Degradation with in Transformer Insulation Systems, lEEE Colloquium on, 6 Dec 1991. Page(s): 8/1 -8/2 . [9] Luwendran Moodley, Brian de Klerk Sweep Frequency Response Analysis as A Diagnostic tool to Detect Transformer Mechanical Integrity, eThekwini Electricity pp.1-9, 1978 [10] SFRA Basic Analysis Volume 2, Tony McGrail, August 7th 2003. [11] S. A. Ryder. "Diagnosing transformer faults using frequency response analysis". Electrical Insulation Magazine, IEEE, and Volume 19 Issue: 2, March-April 2003. Page(s): 16 -22.

Fig . 9: Measurement is taken between low voltage winding and neutral

F. Test on tertiary windings with all other terminals open

Fig. 10: Measurement is taken between low voltage winding and neutral

G. Test on tertiary windings with all other terminals open

Fig. 11: Measurement is taken between tertiary windings

V. ANALYSIS OF THE ABOVE POWER TRANSFORMER As there is no baseline and the sister unit is available, the analysis is done with phase to phase comparison. The open circuit results and the short circuit results showing variations at higher frequencies. All phases responding differently at different frequency ranges. The analysis must be compared with the base line signature signals or the sister unit of this particular power transformer. It will be very difficult to comment about the mechanical integrity of the power transformer without baseline or sister unit comparison. All graphs taken from SFRA test equipment shows little variation at high frequencies may

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

31

Congestion Management with TCSC in Competitive Electricity Markets


1

V. K. Garg, 2Ashwani Kumar


1

UIET Kurukshetra University 2 NIT Kurukshetra


1 2

vkgarg_kuk@yahoo.com ashwa_ks@yahoo.co.in

Abstract FACTS controllers such as thyristor controlled series capacitor (TCSC), by controlling the power flows in a network, can help to reduce the flows in heavily loaded lines resulting in decrease in the load curtailment. This paper proposes the effect of TCSC on load curtailment based approach for congestion management in the deregulated electricity environment. This method incorporates ac load flow equations with constraints on generation, line flows, and FACTS controller parameters. Flexible ac transmission systems (FACTS) controllers, which can minimize the load curtailment of the network by controlling the power flows, must be placed optimally due to their high cost. For optimal location, a mixed integer programming approach has been proposed in this work. The load curtailment has been determined in a pool market model. The proposed technique has been demonstrated on WSCC 9- bus test system. Keywords Optimal power flow, Mixed integer non-linear programming, FACTS controllers, TCSC, Competitive electricity market.

I. INTRODUCTION The power industry throughout world is in the process of moving from a vertically regulated industry to a deregulated structure [1-5]. Congestion occurs more frequently on many individual transmission systems due to the substantial increase in wholesale power transactions between and across regions, and the increased use of the transmission grid [6-10]. When contingency or increasing load happens in power system, some physical violation such as transmission line overload or low/high bus voltage profile may happen in the system. In these cases, we may need to shed some load to ensure that all the system security limits are satisfied. Then, the problem we will face is how to shed as little load as possible to reduce the impact on loss of economy to the system. Typically using OPF, a minimum load to be shed can be determined for congestion management. But we find that despite the fact that some branches have already reached their capacity limits when using OPF, at the same time there may exist some branches whose capacity are not fully used due to the topology and parameters of the network. If we can make better use of these branches, the load curtailment can be reduced. We also find that if we can provide voltage support at some points, the load curtailment can also be reduced. The development of Flexible AC Transmission System (FACTS) has potential contribution on load curtailment issues. Thyristor Controller Phase Angle Regulator (TCPAR) and Thyristor

Controlled Series Capacitor (TCSC) are all members of the FACTS family. Several optimal power flow (OPF) based congestion management schemes with generation redispatch and curtailment of load have been proposed in the literature [610]. Fang and David [9] proposed a transmission dispatch methodology as an extension of spot pricing theory in a pool and bilateral as well as multilateral transactions model. Prioritization of electricity transactions and willingness-to-pay for minimum curtailment strategies has been investigated as a practical alternative to deal with the congestion. An OPF model based on [9] to minimize the curtailment of the contracted powers in a power market having bilateral and multilateral as well as firm contracts has been suggested in [15]. A strategy for allocation of transmission losses among various participants and the role of FACTS controllers like SVC and TCSC on reducing the transmission congestion and curtailment of contracted power has also been suggested. Harry Singh et al. [6] proposed approaches for congestion management based on OPF, which utilizes DC load flow model to minimize the congestion cost for pool co model and bilateral model. The nodal pricing theory has been applied in the pool model whereas a method based on congestion cost allocation has been suggested for bilateral model. The role of FACTS devices for congestion management has been suggested in [11-18]. The impacts of TCSC, which changes the power flow of the transmission system as a result of the apparent change of the transmission line where TCSC has been installed has been studied in [13]. The effect of Thyristor Controlled Series Compensator (TCSC) and Static Var Compensator (SVC) to re-dispatch the generation using an OPF with the objective of minimizing the total amount of transactions being curtailed has been presented in ref. [17] for pool model while a modified with market separation constraint using both pool and bilateral model has been presented in ref. [18]. This paper has also investigated the improvement of TTC. In this paper, a strategy with minimum load curtailment has been proposed. The impact of TCSC on the load curtailment has been computed finding an optimal location of TCSC in the network for power flow control and thereby reducing the load curtailment for loss of economy due to the revenue loss after load curtailment. An optimal location of TCSC has been obtained using mixed integer non-linear programming approach. The model has been developed using GAMS and

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

32

MATLAB interfacing for solution of the complicated nonlinear programming problem [19-20]. The results of an approach have been tested on WSCC 9-bus test system. II. MODEL REPRESENTATION OF TCSC IN LOAD CURTAILMENT BASED CONGESTION MANAGEMENT An easy way to comply with the conference paper formatting requireFig1 shows the model representation of TCSC for development of load flow equations. Based on the principle of TCSC, the effect of TCSC on power system may be simulated as a controllable reactance -xc inserted in the related transmission line.
Bus-i -x c rij+jx ij jBsh Bus-j

Pg

Pli

P, Q g

Qli

(7) (8) (9)

P
Pf
Qf
Pij

At Pf , Q

At Q f

Pij , Aij 1 Pji , A ji 1


Qij , A ji Q ji , A ji
n 2Vi Gij cos2
2

1 1
nViV j {Gij cos( Bij sin(
2
ij ij

(10)

)}

cos (11)

Qij

Vi (n Bij cos
ij

Bsh ) ) Bij cos(


ij
ij

jBsh

nViV j {Gij sin(

)} cos

(12)

Pji
Fig. 1: Equivalent circuit of TCSC

V j2Gij
V j2 ( Bij

nViV j {Gij cos( Bij sin(


Bsh )
ij

)}
ij

cos
) cos

(13)

TCSC is generally installed in the substation for convenient operation and maintenance. In power flow equations, the only difference between the normal lines power flow and the flow in line with TCSC is the controllable reactance xc, where xij is replace by xij-xc. Thus, the power flow equations of TCSC line can be derived as follows:

Q ji

nViV j (Gij sin( Bij cos(


ij

(14) (15) (16)

)}

Pij , L

Pij

Pji , Qij , L

Qij

Q ji

PLT Pgmin
Pf
nb

Pij , L , QLT Pg
min Pgmax, Qg

Qij , L Qg
max Qg

(17) (18) (19) (20) (21)

Pij
Qij

Vi 2 g ij ViV j g ij cos
Vi bij
2

ij

bij sin
ij

ij

(1)

Pfmax , Pij
nb

Pfmax , Pji
0
u. * x t csc m ax

Pfmax

Bsh

ViV j g ij sin
ij

bij cos
ij

ij

(2) (3)

Pji Q ji
Where

V j2 g ij ViV j g ij cos V j2 bij


rij rij2 xij xc
2

bij sin
ij

Pg ,i
i 1,i slack

Pd ,i i 1,i slack
x t csc
m ax
m ax i

Bsh

ViV j g ij sin
, bij xij rij2 xij

bij cos
xc xc
2

ij

(4)

u. * x t csc m in
u N tm ax csc
m in
m ax i

g ij

(5)

Vi
0

Vi
i

Vi
0 li

(22) (23) (24)

In addition, we introduce integer variable uj ={0,1} that defines the presence (uj =1) or absence (uj =0) of the TCSC. III. LOAD CURTAILMENT AND OPTIMAL NUMBER AND LOCATION OF TCSC Mixed integer non-linear programming problem representing the optimal number and location of TCSC and minimizing the system load curtailment is presented. The system real load demand has been determined without and with TCSC. The problem of load curtailment has been formulated as a non-linear model, which is solved by using DICOPT solver of GAMS [19-20], is as follows:
n

Pli

When TCSC is installed in a branch, the power flow of the entire system may be changed as a result of the apparent impedance change of that transmission line. There are some modification in line flow equations and TCSC limits, while all others remain same.

Pij
Qij

n 2Vi 2 Gij
Vi (n Bij
2 2

nViV j (Gij cos


Bsh )
ij

ij

Bij sin ij )

(25) (26)

n ViV j (Gij sin

Bij cos

ij )

Min
i 1

0 li

Pli

(6)

Pji

V Gij

2 j

nViV j (Gij cos

ij

Bij sin ij )

(27)

Subjected to

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

33

Q ji

V j2 ( Bij
x t csc

Bsh )
ij

(28)
ij

n ViV j (Gij sin


0.5 x ij
Where

Bij cos

)
(29)

0.5 x ij

Equations (9) to (14) show the static power flow equations in terms of steady state bus voltages, bus angles and TCSC setting. Equations (25) to (28) are power flow equations modified with TCSC. Equations (20)-(24) are inequality constraints on voltage, angle, and load. Equations (29) is inequality constraints for setting of TCSC. IV. RESULTS AND DISCUSSIONS The proposed algorithms have been tested on WSCC-9 bus system. The WSCC 9 bus system contains 3 generators, which are distributed among 9 buses and 9 branches (line plus transformers). Load curtailments have been determined without and with TCSC. For 9-bus system, the line 4-9 has been considered as a congested line based on the flow under base case condition. The load curtailment based approach has been applied without and with TCSC to observe the impact of TCSC on the load curtailment. Load curtailment determination with TCSC The values of load curtailment without and with the presence of TCSC have been determined solving the optimization problem. The maximum and minimum limit of reactance of TCSC is taken 0.5 and 0.5 times of reactance of line in which it is connected. . It has been observed that by relaxing the bus voltage angle limit the load curtailment will be different and numbers of TCSCs will also be different. The maximum number of TCSC is taken as five above which there is no significant improvement in load curtailment. The load curtailment for the 9-bus system is found to be 0.2868 p.u. The value of load curtailment without and with TCSC for 9-bus system has been shown in Table I. It is shown in Table I that three number of TCSC are sufficient for minimizing the load curtailment. The load curtailment of the system decreases with the presence of TCSC. The system load curtailment without and with TCSC for 9-bus system is also given in the bar chart shown in Fig 2. From Table I it can be observed that net real power loss of the system is zero as in data, resistance of system is considered to be zero and the net reactive power loss increases with one TCSC and then decreases up to 2 numbers of TCSC. The variation in the reactive power loss has been shown in Fig 3. The reactive power loss is found to change due to the reactive support received from TCSC also. This changes the pattern of flows and thus the reactive power in the system.

Gij Bij

rij /{rij2 ( xij

( xij

xt csc ) 2 }
2 ij

xt csc ) /{r

( xij

xt csc ) }

rij , xij , xt csc : Resistance of line i and j, reactance of line i


and j and reactance of TCSC respectively. Where: Pli Real load demand after congestion management in pool model.

Pg , Qg , Pli0 , Qli

: Vector of real and reactive power

generation and demand respectively. P, Q : Vector of real and reactive power injections.

Pij , Qij , Pji , Q ji : Real and reactive power flow at sending


and receiving end between buses i and j respectively.

Pij , L , Qij , L , PLT , Q LT : Real and reactive power losses in


each line and total real and reactive power losses.
min max Pgmin , Pgmax, Qg , Qg : Limits on real and reactive power

generation.

u : Vector of binary variable showing the placement of


TCSC, 1 represents presence of TCSC and represents absence of TCSC.
max

xt csc , xt csc
N tm ax csc

: Vector of TCSC settings and maximum

setting that is available. : Total number of TCPAR available for use.


ij

Vi , V j ,

: Voltage magnitudes at bus i and j respectively,

bus angle difference. Equation (6) shows the objective function to be minimized, subjected to constraints given by equations (7) to (24). Equations (7) and (8) represents bus injections in terms of power flows through lines, which incorporates load demand.
No. of TCSC 0 1 2 3 Load curtailment 0.2869 0.1393 0.0869 0.0230 TCSC location (Line number) Xtcsc Setting

TABLE I : LOAD CURTAILMENT WITHOUT AND WITH TCSC (9-BUS SYSTEM) IN P.U. Total Real power loss 0.0 0.0 0.0 0.0 Total Reactive power loss -0.6039 -0.6570 -0.6448 -0.6410

8-9 5-6,8-9 4-5,6-7,8-9

0.0805 0.0504,0.0805 0.0460,0.0504,0.0805

Real power generations for the system without and with TCSC are presented in Table II for 9-bus system. As the number of TCSC increases, there is an increment in generation at certain buses and are shown highlighted in the

table. With optimal numbers of TCSCs for the minimizing load curtailment, the generators are utilized to their capacities.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

34

TABLE III: REAL POWER DEMAND WITHOUT AND WITH TCSC (9-BUS SYSTEM) IN P.U. Bus no. No TCSC 1 TCSC 2 TCSC 3 TCSC 5 0.9000 0.9000 0.9000 0.9000 7 1.0000 1.0000 1.0000 1.0000 9 1.9631 2.1107 2.1631 2.2270

Fig.. 2: Load curtailment without and with TCSC for 9-bus system

There in increase in line flow at some lines while decrease in flow at some other lines. No line reaches its maximum line flow limit. The total reactive power loss increases with presence of one TCSC and after that decreases with second and third TCSC. The highlighted values show the presence of TCSC. V.CONCLUSIONS In this work, load curtailment has been determined for a pool electricity market using ac load flow approach and the optimization problem has been solved by MINLP. The system load curtailment has been determined for WSCC 9-bus system without and with TCSCs using ac load flow approach. The line flows for optimally placed TCSCs were found to be increased compared to their base case values. The real power losses in the system increase in all the lines with tcscs. However as the number of TCSCs reaches near to its saturation value the losses in the lines has decreases slightly. The net reactive power loss in the system has been decreasing with increased number of TCSCs, but it is found that as the number of tcscs approaches to its maximum number, the reactive power loss again increases slightly. The maximum number of TCSCs are found three for 9-bus system for minimum load curtailment for the system congestion management

Fig. 3: Net system reactive power loss without and with TCSC for 9-bus system TABLE II.: REAL POWER GENERATIONS WITHOUT AND WITH TCSC (9-BUS SYSTEM) IN P.U. Bus no. No TCSC 1 TCSC 2 TCSC 3 TCSC 1 1.3831 1.5307 1.5831 1.6470 2 1.6300 1.6300 1.6300 1.6300 3 0.8500 0.8500 0.8500 0.8500

The load demand without and with TCSC is shown in Table III for 9-bus system respectively. It is found that there is variation in demand according to line reactance. The load demand at bus number 9 increases for 9-bus system. The pattern of real and reactive power flows are shown in Tables IV to Table VII.
From 1 4 6 3 6 8 2 8 4 To 4 5 5 6 7 7 8 9 9 No TCSC Pij 1.3831 0.3831 0.5169 0.8500 0.3331 0.6669 1.6300 0.9631 1.0000 Pji -1.3831 -0.3831 -0.5169 -0.8500 -0.3331 -0.6669 -1.6300 -0.9631 -1.0000 Ploss 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000

TABLE IV: REAL POWER FLOWS AT SENDING, RECEIVING END AND LOSS IN EACH LINE (9-BUS SYSTEM) IN P.U. 1 TCSC Pij 1.5307 0.5307 0.3693 0.8500 0.4807 0.5193 1.6300 1.1107 1.0000 Pji 1.5307 0.5307 0.3693 0.8500 0.4807 0.5193 1.6300 1.1107 1.0000 Ploss 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Line Rating 2.5000 2.0000 2.0000 2.0000 2.0000 2.0000 2.0000 2.0000 1.0000

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

35

TABLE V: REAL POWER FLOWS AT SENDING, RECEIVING END AND LOSS IN EACH LINE (9-BUS SYSTEM) IN P.U. From 1 4 6 3 6 8 2 8 4 To 4 5 5 6 7 7 8 9 9 2 TCSC Pij 1.5831 0.5831 0.3169 0.8500 0.5331 0.4669 1.6300 1.1631 1.0000 Pji 1.5831 0.5831 0.3169 0.8500 0.5331 0.4669 1.6300 1.1631 1.0000 Ploss 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 3 TCSC Pij 1.6470 0.6470 0.2530 0.8500 0.5970 0.4030 1.6300 1.2270 1.0000 Pji 1.6470 0.6470 0.2530 0.8500 0.5970 0.4030 1.6300 1.2270 1.0000 Ploss 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 Line Rating 2.5000 2.0000 2.0000 2.0000 2.0000 2.0000 2.0000 2.0000 1.0000

TABLE VI: REACTIVE POWER FLOWS AT SENDING, RECEIVING END AND LOSS IN EACH LINE (9-BUS SYSTEM) IN P.U. From 1 4 6 3 6 8 2 8 4 To 4 5 5 6 7 7 8 9 9 No TCSC Qij 0.2595 0.0392 -0.1700 -0.3642 -0.2497 0.3036 0.6508 0.1726 0.1062 Qji -0.1454 -0.1769 -0.1231 0.4157 0.0624 -0.4124 -0.4762 -0.3144 -0.1856 Qloss 0.1141 -0.1376 -0.2931 0.0555 -0.1873 -0.1088 0.1746 -0.1418 -0.0795 1 TCSC Qij 0.1931 0.0550 0.1990 0.3749 0.2318 0.2851 0.6747 0.2132 0.0010 Qji 0.0560 0.1811 0.1189 0.4309 0.0567 0.4067 0.4983 0.4131 0.0869 Qloss 0.1371 0.1261 0.3179 0.0560 0.1752 0.1216 0.1764 0.1999 0.0859 Line Rating 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000

TABLE VII: REACTIVE POWER FLOWS AT SENDING, RECEIVING END AND LOSS IN EACH LINE IN P.U. From 1 4 6 3 6 8 2 8 4 To 4 5 5 6 7 7 8 9 9 2 TCSC Qij 0.2067 0.0481 0.1942 0.4063 0.2697 0.3123 0.7048 0.2137 0.0118 Qji 0.0599 0.1688 0.1312 0.4639 0.0851 0.4351 0.5260 0.4029 0.0971 Qloss 0.1468 0.1207 0.3253 0.0576 0.1846 0.1229 0.1788 0.1892 0.0853 [5] 3 TCSC Qij 0.2217 0.0497 0.2172 0.4197 0.2608 0.3022 0.7070 0.2259 0.0129 Qji 0.0626 0.1839 0.1161 0.4780 0.0795 0.4295 0.5280 0.4021 0.0979 Qloss 0.1591 0.1342 0.3333 0.0583 0.1812 0.1274 0.1790 0.1762 0.0850 Line Rating 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000

REFERENCES
[1] North American Electric Reliability Council (NERC), Available Transfer Capability Definitions and Determination, NERC Report, June 1996. Marija Ilic, F. Galiana and L. Fink, Power System Restructuring, Kluwer Academic Publishers, USA, 1998. K. Bhattacharya, M.H.J. Bollen, J.E. Daalder, Operation of Restructured Power Systems, Chalmers University of Technology, Kluwer Academic Publishers, 2001. L. L. Lai, Power System Restructuring and Deregulation, Trading Performance and Information Technology, John Wiley and Sons, UK, 2001. [6]

[2] [3]

[7]

[4]

[8]

Mohammed Shahidepour and Muwaffaq Alomoush, Restructured Electrical Power Systems, Operation, Trading, and Volatility, Marcel Dekker, Inc. New York, 2001. H. Singh, S. Hao, and A. Papalexopoulos, Transmission Congestion Management in Competitive Markets, IEEE Trans. on Power Systems, vol. 13, No. 2, May 1998, pp. 672-680. F.D. Galiana, M. Ilic, A Mathematical Framework for the Analysis and Management of Power Transactions under Open Access, IEEE Trans. On Power Systems, vol. 13, No. 2, pp.681687, May 1998. D. Shirmohammadi, B. Wollenberg, A. Vojdani, P. Sandrin, M. Pereira, F. Rahimi, T. Schneider, and B. Stott, Transmission Dispatch and Congestion Management in the Emerging Energy Market Structures, IEEE Trans. on Power Systems, vol. 13, No. 4, Nov. 1998, pp. 1466-1476.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

36

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19] [20]

R.S. Fang and A.K. David, Transmission Congestion Management in an Electricity Market, IEEE Trans. on Power Systems, vol. 14, No.2, Aug. 1999, pp. 877-883. R. D. Christie, B. F. Wollenberg and I. Wangensteen, Transmission Management in the Deregulated Environment, Proceedings of The IEEE, Vol. 88, No. 2, Feb 2000, pp 170-195. S.C. Srivastava and Parveen Kumar, Optimal Power Dispatch in Deregulated Market Considering Congestion Management, International Conf. on Electric Utility Deregulation and Restructuring and Power Technologies, DRPT, city University, London, April 2000, pp. 53-59. H. Mori and Y. Goto, A parallel tabu search based method for determining optimal location of FACTS in power system, in Proc. Int. Conf. Power System-Techno., vol. 2, pp. 1077-1082, June 2000. F.D. Galiana, K.C. Almeida, M. Thoussaint, J. Griffin, B.T. Ooi, and D.T. Mcgillis, Assessment and Control of the Impact of FACTS Devices on Power System Performance, IEEE Trans. On Power System, vol. 11, pp.1931-1936, Nov.1996. K. S. Verma and S. N. Singh and H. O. Gupta, FACTS Devices Location for Enhancement of Total Transfer Capability, Proc. of IEEE PES, Winter Meeting, Ohio, USA, Jan. 28-Feb. 1, 2001, vol. 2, pp 522-527. G. Huang, P. Yan, The impacts of TCSC and SVC on power system load curtailments, in IEEE Power Engineering Society Summer Meeting, July 2001, Vancouver, Canada. S.N. Singh and A.K. David, Optimal Location of FACTS Devices for Congestion Management, International Journal of Electric Power System and Research, vol.58, pp.71-79, 2001. J.W.M. Cheng, F.D. Galiana, D.McGillis, The Application of FACTS Controllers to a Deregulated System, Conference on ACDC Power Transmission, pp.10-14, Nov. 28-30, 2001. G. M. Huang and Ping Yan, TCSC and SVC as Re-dispatch Tools for Congestion Management and TTC Improvement, IEEE PES, Winter Meeting, 2002, pp. 660-665. GAMS The Solver Manuals, GAMS Development Corporation, March 2001. MATLAB and GAMS: Interfacing Optimization and Visualization Software, Michael C. Ferris, August 10, 1999.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

37

Voltage Sag Mitigation for Load Unbalancing in a Power System Network by Placement of DSTATCOM with ANN Based Methodology
1

Brijesh Singh, 2Abdul Zeeshan


2

1, 2 1

United College of Engineering and Research, Allahabad, UP


zeeshan3970@gmail.com

brijeshsingh81@indiatimes.com,

Abstract The present paper deals with the problem of unbalanced voltages arising due to unbalanced loads in an electrical power system network. For the study, IEEE 14 bus system has been considered. Simulation is carried out in standard MATLAB environment using SIMULINK and POWER SYSTEM block-set toolboxes for both cases i.e. with and without D-STATCOM connection in the system. The optimal location for placement of D-STATCOM is found using Artificial Neural Network approach. Description of the study and some selected results of simulation are presented in this paper. The novelty of this research lies in the application of D-STATCOM for load compensation in mesh of electrical power system and the placement of DSTATCOM using ANN. Simulation results demonstrate the capability of D-STATCOM for load compensation. Keywords Power quality, D-STATCOM, Load balancing, Load compensation, ANN

I. INTRODUCTION This document is a template. Load balancing is one of the important issues for improvement of Power Quality. Most of the loads are non-linear and mostly unbalanced in the distribution system. Mainly the unbalanced load gives rise to unbalanced voltage which is objectionable for most of the industrial operations. The consumers such as industries and companies prefer electrical energy with high quality as the plants run more automated process and may experience substantial economical loss. If delivered energy to these loads have poor quality then products and equipments such as microcontrollers, computers, motor drives etc. get damaged. For these reasons, power quality problems must be addressed properly. Most of ac power system are three phase and are designed for balanced operation. In high quality power supply system, balanced phase currents and voltages must be held properly. In unbalanced operation, negative and zero sequence components arise in addition to the positive sequence components. These undesirable negative and zero sequence components cause additional losses in motors and generators, oscillating torques in ac machines, increased ripples in rectifiers, saturation of transformer, excessive neutral currents, malfunctioning of several types of equipments etc.In order to minimize these undesirable effects caused by the unbalanced loads, load compensation techniques must be applied. Load compensation is to balance the unbalance load and correct load power factor

to unity at the same time. Load compensation is very important for many applications such as compensation of single phase railway system and electric arc furnace system. Power quality is of increasing importance in worldwide distribution. The distribution system is facing severe power quality problems such as poor voltage regulation, high reactive power, harmonic current burden and load balancing [1-6]. Also, the voltage regulation is poor in the distribution system due to unplanned expansion and installation of different types of loads, most of which as already mentioned are non-linear and mostly unbalanced [7]. The improvement of power quality has been therefore one of the most important subjects in the development of distribution and low voltage system. There are several mitigation techniques for power quality problems in the distribution system and the group of devices used for the purpose are known as custom power devices [8]. The DSTATCOM is an important shunt connected custom power device which have the potential to solve numerous power quality problems [3]. D-STATCOM has effectively replaced a static VAR compensator (SVC) as the former has a number of advantages such as very fast response, low losses, capability to produce both inductive and capacitive reactive power. The D-STATCOM protects the utility transmission or distribution system from voltage sag/swell and/or flicker caused by rapidly varying reactive current demand. Thus, the D-STATCOM can be one of the viable alternatives to SVC in a distribution network. Current literature survey reveals that a no. of methods has been evolved for compensation of harmonics and unbalances. SVCs can effectively compensate flicker, unbalanced load and correct load power factor [10]. There are several literature [11-17] reported for the load compensation or load balancing using D-STATCOM. But, most of the work deals with the radial type electric power distribution system feeding unbalanced load or standalone system. A already mentioned, the present work deals with modelling and simulation of IEEE-14 bus system using D-STATCOM. The Artificial Neural Networks has been widely used successfully in power industry in fault classification, protection, fault diagnosis, relaying schemes, load forecasting , load balancing, reactive power control [ 9,18-20]. In this paper, the ANN based approach has been applied to find out the optimal location of D-

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

38

STATCOM in IEEE-14 bus system for load compensation. The ANN was trained with Levenberg Marquardth back-propagation algorithm (trainlm). II. D-STATCOM AND OPERATING PRINCIPLE The Distribution Static Synchronous Compensator (D-STATCOM) is one of the key FACTs devices. Based on voltage sourced converter, the D-STATCOM regulates system voltage by absorbing or generating reactive power. In contrast to a thyristor based Static VAR Compensator (SVC), D-STATCOM output can be controlled without depending on AC system voltage. It is similar to rotating synchronous condenser but capable of very fast reactive compensation as it has no moving part. D-STATCOM has the ability to generate or absorb reactive power by suigraph control of voltage at the inverter end with respect to the AC system voltage to which the D-STATCOM is connected in shunt as shown in Fig 1.

function f. The activation function may be a pure linear or a continuous nonlinear function like sigmoid one, hyperbolic one depending upon application. The neurons are interconnected creating different layers. The feedforward architecture which is used in the present work is shown in the Fig2 (b). This section presents the architecture of the network that is most commonly used with the back-propagation algorithmthe multilayer feedforward network; Neuron Model. An elementary neuron with R inputs is shown.

(a)

(b) Fig. 2: Artificial Neural Network architecture

Fig. 1: Basic model of D-STATCOM

During steady state operation, the D-STATCOM control system keeps the fundamental component of the VSI voltage Vc voltage in phase with the system voltage Vs. If the two voltage magnitudes are equal, the reactive power exchange is zero. If the voltage generated by the VSI is higher than the system voltage then the DSTATCOM acts as a capacitor and supplies reactive power to the system whereas if the voltage generated by the VSI is lower than the system voltage then the DSTATCOM acts as an inductor and absorbs reactive power from the system. The amount of reactive power exchange depends on the VSI voltage magnitude and on the coupling transformer leakage reactance. If the DSTATCOM has a dc source or energy storage device on its dc side, it can supply active power to the system bus and this can be achieved by adjusting the phase angle of the ac power system. III. ARTIFICIAL NEURAL NETWORK CONFIGURATION Title The Artificial Neural Network consists of a large no. of strongly connected elements called neurons. The model of artificial neurons in a schematic conFiguration is shown in below Fig2 (a). The input data p(1), p(2), .p(R) flow through the synapses weights Wij. These weights amplify or attenuate the input signals before the addition at the node represented by a circle. The summed data flows to the output through an activation

Each input is weighted with an appropriate weight w. The sum of the weighted inputs and the bias forms the input to the transfer function f. Neurons can use any differentiable transfer function f to generate their output. If the last layer of a multilayer network has sigmoid neurons, then the outputs of the network are limited to a small range. If linear output neurons are used, the network outputs can take on any value. In back-propagation it is important to calculate the derivative of any transfer function used. Once the network weights and biases are initialized, the network is ready for training. The network can be trained for function approximation (nonlinear regression), pattern association, or pattern classification. The training process requires a set of examples of proper network behaviour network inputs p and target outputs t. During training the weights and biases of the network are iteratively adjusted to minimize the network performance function. The default performance function for feed-forward networks is mean square error msethe average squared error between the networks outputs and the target outputs t. The different training algorithms for feed-forward networks use the gradient of the performance function to determine how to adjust the weights to minimize performance. The gradient is determined using a technique called backpropagation, which involves performing computations backward through the network. There are many variations of the back-propagation algorithm. The simplest implementation of back-propagation learning updates the network weights and biases in the direction in which the performance function decreases most rapidly, the negative of the gradient. Single iteration of this algorithm can be written where is a vector of current weights and biases, is

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

39

the current gradient, and is the learning rate. There are two different ways in which this gradient descent algorithm can be implemented: incremental mode and batch Mode. In incremental mode, the gradient is computed and the IV. SIMULATION PLANT MODEL Simulation model of IEEE-14 bus system which is composed of 14 buses and 20 lines is developed using MATLAB/SIMULINK. The system consists of 5 synchronous machines, three of which are synchronous compensators used only for reactive power support. There are 11 loads in the system totalling 259 MW and 81.3 MVAR. The simulation block diagram of the system is shown in Fig3. This plant model is used to find out the unbalanced voltages at different buses arising due to application of unbalanced load. Using this database of the voltage, optimal location of D-STATCOM using ANN methodology is found out. V. ANN APPROACH TO FIND OUT OPTIMAL LOCATION OF D-STATCOM In the proposed neural network architecture there are 20 hidden layers and 14 output layers. This network can be trained to give a desired pattern at the output when the corresponding input data set is applied. The training process is carried out with a large number of input and output target data. The output target data is the p.u. voltage of the IEEE-14 bus system. The training input data (p.u. voltage of different phases of different buses under unbalanced conditions) is obtained using above simulation platform model. Once the network is trained some data is used to test the network. The testing result provides information about the optimal location of most affected bus of the system. The bus corresponding to higher mse is understood to the most affected bus. The optimal location to mitigate the voltage unbalance problem is the bus having higher mse. The D-STATCOM is placed at the most affected bus and again the voltage data collection is completed under unbalanced loading condition.

weights are updated after each input is applied to the network. In batch mode, all the inputs are applied to the network before the weights are updated.

is obtained under unbalanced conditions. The test results show the effect of unbalancing .It affects the whole system voltage levels in the form of either voltage sag or swell. This gives a complete picture of voltage unbalance under unbalanced loading conditions. The voltage database has been used for training of ANN for finding the optimal location of D-STATCOM into the network. The optimal location has been decided by the mean square error Graph 2. The ANN based approach gave bus no. 5 as the optimal location for placement of D-STATCOM to compensate the voltage unbalance. After placement of D-STATCOM on bus no. 5 the voltage profile of each phase of all buses has been improved. This is very clear from the Graph 3. In order to create unbalance loading condition an additional Y- connected unbalanced load ; Phase A [P=1MW, Q=100MVAR(+ve)], Phase B [P=25KW, Q=200KVAR(+ve)], Phase C [ P=1KW, Q=0.1KVAR(+ve)] is connected at different load buses.
Voltage Profile of Different Buses During Unbalanced Loading Condition GRAPH 1

GRAPH 2 : Training performance of ANN at different buses

Fig. 3: IEEE-14 bus system (MATLAB/SIMULINK) Model

VI. SIMULATION AND TEST RESULTS For simulation studies, the voltage profiles of different buses are obtained by placing unbalanced load at different buses of the network. The p.u. voltage of different phases

VII. CONCLUSION In the present work IEEE-14 bus system is modelled and simulated using MATLAB / SIMULINK and the simulation results with and without D-STATCOM are presented. The voltage unbalance created at the buses under unbalance loading condition is compensated using D-STATCOM. For load balancing the optimal location of this device in the considered electrical network is obtained using ANN approach. The simulation results which are in

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

40

graphical form show that the D-STATCOM performance is satisfactory in load balancing for most of the buses of the considered electrical network.
GRAPH 3: Voltage Profiles of Different Buses With Placement Of Dstatcom At Bus 5

[16]

[17]

[18]

[19]

[20]

REFERENCES
[1] [2] R. Arnold, Solutions to the Power Quality Problems, Power Engineering Journal, vol. 15, No.2, pp 65-73, April 2001. A. Ghosh, and G. Ledwich,Power Quality Enhancement using Custom Power Devices, Kluwer Academic Publishers, London 2002. J. Dixon, Luis Moran, Jose Rodriguez,Reactive Power Compensation Technologies: State of art Review, Proceedings of the IEEE, vol.93, No. 12, 2005. R. C. Dugan, M. F. McGranaghan and H. W. Beaty, Electric Power System Quality, 2nd Edition, MCGrawHill, NewYork. Antonio Moreno Munoz,Power Quality: Mitigation Technologies in a distributed environment, Springer Verlag London Limited, London 2007. Ewald F. Fuchs and Mohammad A.S. Mausoum,Power Quality in Power System and Electrical Machines, Elsevier Academic Press, London, U.K. 2008. Bhim Singh, P. Jayaprakash and D. P. Kothari,Three leg VSC and a transformer based three phase four wire D-STATCOM for Distribution System, 15th National Power System Conference( NPSC) IIT Bombay, Dec. 2008. IEEE Recommended Practices and Requirements for Harmonic control in Electrical Power System, IEEE Std 519, 1992. Ismail K. Said and Marouf Pirouti , Neural Network based load balancing and reactive power control by Static VAR Compensators,International Journal of Computer and Electrical Engineering ,Vol. 1, No. 1, April 2009. San-Yi Lee, Chi-Jui Wu, Wei-Nan Chang,A Compact Control Algorithm for Reactive Power Compensation and Load Balancing with Static VAR Compensators, Electric Power System Research, Vol. 58, No. 2 , June 2001, pp 63-70. A. Ghosh and A. Joshi ,A New Approach for Load Balancing and Power Factor Correction in Power Distribution System, IEEE Transaction Power Delivery, Vol.15, No.1, January 2000, pp 417422. C. N. Bhende, Dr. M. K. Mishra, Dr. H. M. Suryawanshi,A DSTATCOM Modeling , Analysis and Performance for Unbalanced and Non-Linear loads, IE(I) Journal EL , Vol. 86, March 2006, pp 297-304. H. Nasiraghdam and A. Jalilian ,Balanced and Unbalanced Voltage Sag Mitigation using D-STA TCOM with Linear and Non-Linear loads, World Academy of Science , Engineering and Technology, 2007. Serban IOAN, Coreliu Marinescc,Unbalance Compensation in Stand Alone Microgrids, 6th International Conference on Electromechanical and Power Systems, Oct. 4-6 , 2007. Wei-Neng Chang and Kuan-Dih Yeh,Design and Implementation of D-STATCOM for Fast Load Compensation of Unbalanced

[21]

Loads, Journal of Marine Science and Technology , Vol. 17, No. 4 , pp-257-263, 2009. Zakir Husain , Ravinder Kumar Singh and Shri Niwas Tiwari,Balancing of Unbalanced Load and Power Factor Correction in Multiphase ( 4 phase ) Load Circuits using DSTATCOM, Proceedings of the World Congress on Engineering 2010, Vol. II WCE-2010, June 30-July 2, 2010, London, U.K. G. K. Bansal, Bhim Singh,Harmonic Elimination, Voltage Control, Load Balancing in an Isolated Power Generation, European Transaction on Electrical Power , Vol. 20, Issue 6, pp 771-784, Sept. 2010. Jayabarathi P. a. and Devarajan N. b.,ANN based DSPIC Controller for Reactive Power Compensation, American Journal of Applied Sciences, 4(7): pp 508-515, 2007. F. J. Alcantare, J.R. Vazquez, P. Salmeron, S. P. LItran, M.I. Arteaga Orozco,On-line Detection of Voltage Transient Disturbances using ANNs, International Conference on Renewable Energies and Power Quality (ICREPQ 09), 15th-17th April 2009, Valencia, Spain. E.A.Mohamed, N. D. Rao,Artificial Neural Network based Fault Diagnostic for Electric Power Distribution Feeders, Electric Power System Research, pp 1-10, 35, 1995. MATLAB 7 Users Guides for SIMPOWER SYSTEMS and Neural Network Tool-box.

[3]

[4] [5]

[6]

[7]

[8] [9]

[10]

[11]

[12]

[13]

[14]

[15]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

41

ATC Enhancement using Optimal Location of FACTS Controllers with Real Parameter Genetic Algorithm
1 1, 2

C.Bala Gangi Reddy, 2M. Suryakalavathi

Mahaveer Institute of Science and Technology, Hyderabad 1 dbgr.jntu@gmail.com

AbstractThe main aim of this paper is to enhance the Available Transfer Capability (ATC) from enerating/source area to sink area in deregulated power system using Continuous Power Flow (CPF) method during normal and contingency cases with optimal location and control parameter of FACTs controllers such as TCSC or SVC on IEEE 14 and IEEE 24 bus reliability test systems. Real parameter Genetic Algorithm (RGA) is used to determine location and control parameter of TCSC or SVC. ATC is dependent on many factors, such as the base case of system operation, system operation limits, network conFiguration, specification of contingencies etc. FACTs technology has a severe impact to the transmission system utilization with regards to those constraints on ATC. Hence, maximum use of existing transmission assets will be more profitable for Transmission System Operators (TSO), and customers will receive better services with reduced prices. KeywordsAvailable Transfer Capability (ATC), Continuous Power Flow (CPF), TCSC, SVC, Real parameter Genetic Algorithm (RGA), Transmission System Operator (TSO).

I. INTRODUCTION The objective of electric power industry restructuring is to promote competitive markets for electric power trading. Under new environment, the main consequence of the nondiscriminatory open-access requirement is the substantial increase in power transfers. The Available Transfer Capability (ATC) of a transmission network is the unutilized transfer capabilities of a transmission network for the transfer of power for further commercial activity, over and above already committed usage [1]. Adequate available transfer capacity (AATC) is needed to ensure all economic transactions, while sufficient ATC is needed to facilitate electricity market liquidity. It is necessary to maintain economical and secure operation over a wide range of system operating conditions and constraints. However, tight restrictions in the construction of new facilities due to the economic, environmental, and social problems, reduces the operational alternatives. It may sometimes lead to a situation that the existing transmission facilities are intensively used. On the other hand, it can be said that power suppliers will benefit from more market opportunities with reduced possibility of congestion incorporating power systems security enhancement. Maximum use of existing transmission assets will be more

profitable for transmission system owners, and customers will receive better services with reduced prices [2]. Various ATC boosting approaches have been experienced by adjusting generators terminal voltages, under load tap changers (ULTCs) and rescheduling generator outputs. Based upon the NERCs definition of ATC and its determination [1], transmission network can be restricted by thermal, voltage and stability limits. On the other hand, it is highly recognized that, with the capability of flexible power flow [3], FACTS technology has introduced a severe impact to the transmission system utilization with regards to those three constraints. From the steady state power flow viewpoint, networks do not normally share power in proportion to their ratings, where in most situations, voltage profile cannot be smooth. Therefore, ATC values are always limited by heavily loaded buses with relatively low voltage. FACTS concept makes it possible to use circuit reactance, voltage magnitude, and phase angle as controls to redistribute line flow and regulate voltage profile. Theoretically FACTS devices can offer an effective and promising alternative to conventional methods of ATC enhancement. They will provide new control facilities, both in steady state power flow control and dynamic stability control [4]. Controlling power flow in electric power systems without generation rescheduling or topological changes can improve the network performance considerably. With suitable location, the effect of a TCSC and SVC on the ATC enhancement are studied and demonstrated through case studies. It is shown that installing SVC in the proper location will improve voltage profile as well as ATC, and TCSC will improve the ATC. II. STATIC MODELING OF FACTS CONTROLLERS This section focuses on the modeling of two kinds of FACTS, namely TCSC and SVC [5]. The power flows of the line connected between bus i and bus j having series and without any FACTS controllers [6], can be written as,

where Vi, Vj , _ij are the voltage magnitudes at bus i and

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

42

bus j, and voltage angle difference between bus i and bus j, and

Similarly, the real power (Pji) and reactive power (Qji) flows from bus j to bus i in the line can be written as,

or inductive Var so as to maintain or control specific parameters of electrical power system, typically bus voltage [7, 8]. Like the TCSC, the SVC combines a series capacitor bank shunted by thyristor controlled reactor. SVC structure and its representation as a continuous variable shunt susceptance is shown in Fig 2. The SVC load flow models can be developed treating SVC susceptance as control variable. Assuming that SVC is connected at node p to maintain the bus voltage at V p, the reactive power injected by the controller is given by (8).

The linearized load flow models make use of eq. (8) to modify the corresponding jacobian elements at SVC bus [9]. The SVC load flow model can be developed treating SVC susceptance as control variable (BSV C). III. ATC COMPUTATIONAL METHODS North American Electric Reliability Council (NERC) report in 1996 [1] establishes a framework for determining ATC of the interconnected transmission networks for a commercially viable wholesale market. The report defines ATC principles under which ATC values are to be computed and it permits individual systems, power pools, regions and sub regions to develop their procedure for determining ATC in accordance with these principles. There are many methods to compute ATC. In [10], the topological information of a system is stored in a matrix form and constants for different simultaneous cases and critical contingencies have been calculated before hand and used for determination of ATC values. For very large systems, the method may be quite cumbersome. In [10], the localized linearity of the system is assumed and additional loading required to hit the different transfer limits are separately calculated and the minimum of all these is taken as the ATC. The ATC is calculated by the following methods, 1. Based on Continuation Power Flow (CPF), 2. Based on distribution factors. The method based on the CPF incorporates the limits of reactive power flows, voltage limits as well as voltage collapse and line flow limits. However, with this method the computational effort and time requirement are large. The method based on DC power transfer distribution factors utilizes the DC power flow based formulation. Computation of ATC by this method is much simple and fast [11]. However, this method gives more optimistic results. The method based on AC power transfer distribution factors (ACPTDFs) is a simple and efficient non-iterative method to calculate ATC under bilateral and multilateral contracts. The ACPTDFs have been calculated at base case Newton Raphson Load Flow (NRLF) results utilizing a sensitivity based approach. A. ATC Calculation using Continuation Power Flow (CPF) from the solved base case, power flow solutions are sought for increasing

A. Static Representation of TCSC The basic idea behind power flow control with the TCSC is to decrease or increase the overall lines effective series transmission impedance, by adding a capacitive or inductive reactance correspondingly [5]. The TCSC is modeled as variable impedance, where the equivalent reactance of the line xij is defined as:

Where, xline is the transmission line reactance [6]. The level of applied compensation of the TCSC usually varies between 20% inductive and 70% capacitive. Fig 1 shows a controllable reactance ( TCSC) placed in the jx transmission line connected between bus-i and bus-j. The real and reactive power flows from bus i to bus j and bus j to bus i in the line can be written as (1) to (4) with modified gij and bij as given below,

B. Static representation of SVC SVC is a shunt connected static Var generator or consumer whose output is adjusted to exchange capacitive

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

43

amounts of transfer in the specified direction. The quantity of the transfer is a scalar parameter, which can be varied in the model. The amount of transfer is gradually increased from the base case until a binding limit is encountered. This continuation process requires a series of power system solutions to be solved and tested for limits. The transfer capability is the change in the amount of transfer from the base case transfer at the limiting point. Continuation can be simply done as a series of load flow calculations for increasing amounts of transfers. However, when convergence could be poor, such as the case for transfers approaching voltage instability, methods that allow the transfer parameter to become a dependent variable of the model are the most successful. CPF is a method for finding the maximum value of a scalar parameter in a linear function of changes in injections at a set of buses in a power flow problem [10]. Originally introduced for determining maximum loadability, CPF is adaptable, without change in principle, for other applications, including ATC. The CPF algorithm effectively increases the controlling parameter in discrete steps and solves the resulting power flow problem at each step. The procedure is continued until a given condition or physical limit preventing further increase is reached. Because of solution difficulty and the need for the jacobian matrix at each step, the NRLF algorithm is used. CPF yields solution even at voltage collapse points. A CPF is performed by starting from an initial point and then increasing the load by a factor until some system limit is reached [12]. The loads are defined as,

_ Calculus

based method method _ Random search techniques Calculus based and enumerative methods are comfortable in their ability to deliver solutions in applications involving search spaces of limited problem domain. Both the methods are local in scope, the optima they seek are the best in a neighborhood of the current search point. But in their application to real world of search, which is fraught with discontinuities of functions and their derivatives and vast multimodal noisy search spaces, they break down on problems of even moderate size and complexity. Their inability and inefficiency to overcome the local optima and reach the global
_ Enumerate

Fig 3.: Representation of control variables in a individual Chromosome

Where PiDo, QiDo are the active and reactive power respectively of bus i in the base case. PiD , QiD are the active and reactive power of bus i increased by parameter _. For a specific source/sink transfer case calculation of the ATC may be summarized as the maximum transfer power without causing a limit violation over the base case. B. Real Parameter Genrtic Algorithm (RGA) for enhancement of ATC using FACTs devices Genetic algorithms are search algorithms based on the mechanism of natural selection and natural genetics, inspired from the biological evolution, survival of the fittest among string structures with a structured yet, randomized information exchange with in the population to form a search algorithm with some of the innovative flair of human search. In every generation a new set of strings created using bits and piece of the old, an occasional new part is tried for good measure. Being randomized, GAs exploit historical information to speculate on new search points with expecting improved performance. The current literature identifies three main types of search methods or optimization techniques. They are,

optimum make them insufficiently robust, precluding their application to complex problems as search method. On the other hand, random search algorithms managed to overcome the inherent disabilities of the calculus and enumerative methods. Yet, random schemes that searches and save the best must also be discounted because of the efficiency requirement. Random searches, in the long run can be expected to do no better than enumerative schemes. In our haste to discount strictly random search methods, we must be careful to separate them from randomized techniques. The randomized search techniques incorporated the basic advantages of random search but used it only as a tool to guide a more highly exploitative search [13]. In these methods, the search is carried out randomly and information gained from a search is used in guiding the next search. Genetic algorithm is an example of such technique, which drew inspiration from the robustness of nature. In this paper, Real Parameter Genetic Algorithm (RGA) is used. Compared with binary GA, it offers higher accuracy of the control variables. RGA is similar to normal GA except decoding the control variable information is eliminated, hence it will reduce the computation time. C. Representation of Chromosome To apply RGA, to solve a specific problem, one has to define the solution representation and the coding of control variables. The optimization problem here is to use CPF to find the Total Transfer Capability for different TCSC locations and compensations. Every individual chromosome should contain TCSC location and compensation level, as shown in Fig 3 [14]. Compared with binary GA, RGA offers higher accuracy of the control variables. As for location information, we

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

44

use a series of integrals to express different placement of TCSC/SVC. For example, if the location for TCSC/SVC has 6 choices, six integrals 1, 2, 3, 4, 5, 6 are used as candidates for this control variable [15]. In every procedure of GA (initialization, reproduction, crossover and mutation), the resulting location code will be held as one of those integers. D. Fitness Function (FF) GA is usually designed to maximize or minimize the FF, which is a measure of the quality of each candidate solution.After control variables are coded, the objective function (fitness) will be evaluated. These values are measures of quality, which is used to compare different solutions. The better solution joins the new population and the worse one is discarded. The fitness value of an individual will determine its chance to propagate its features to future generations. Here, ATC is used as the fitness function (FF) [14]. FF = maximize ATC (11) E. Blended (BLX - ) crossover Crossover is one of the main distinguishing features of Gas that make them different from other algorithms. The main objective of crossover is to reorganize the information of two different individuals and produce a new one. For two parent solutions solution in the range If ui is a random number between 0 and 1, the following (12) is an offspring [16], (assuming randomly picks a

G. Flow Chart of proposed approach In GA, individuals are simplified to a chromosome that includes control variables of the problem. The value of an individual is called fitness which is corresponding to the objective function value that should be optimized [17]. Fig 4 shows the flow chart of the proposed algorithm. As discussed earlier, RGA is selected for optimization process to improve the ATC including TCSC/SVC. IV. RESULTS AND DISCUSSION ATC is computed for a set of source/sink transfers on IEEE 14 and IEEE 24 bus reliability test systems. The ATC margin can be further increased by proper location and control parameter of FACTs controllers. The FACTs controllers selected in this paper are TCSC and SVC. RGA is used to find optimal location and control parameter of TCSC and SVC for maximization of ATC. In this paper, the total study is divided into two cases as, _ ATC calculation at base case (without line outage) _ ATC calculation at contingency case (with line outage) The ATC margin is limited by bus voltage magnitude and line flow rating. The voltage magnitude limits of all buses are

Where If is zero, this crossover creates a random solution in the range . It is reported that BLX-0.5 (with = 0:5) performs better than BLX operators with any other value [16]. F. Non-uniform mutation Mutation is used to introduce some sort of artificial diversification in the population to avoid premature convergence to local optimum. In non-uniform mutation the probability of creating a solution closer to the parent is more than the probability of creating one away from it. However, as the generations (t) proceed, this probability of creating solutions closer to the parent gets higher and higher [16]. For a given parent X = X1;X2;X3; :::XK; :::;Xl, if the gene XK is selected for mutation and the range of X K is [UK min;UK max] , then the result (13), X e K is, Xe K =

Fig. 4: Flow chart of proposed approach

set to Vmin = 0:95 p.u and Vmax = 1:1 p.u. The line ratings of IEEE 14 and IEEE 24 bus systems are given in [18] and [19] respectively. The GA parameters used in this study are, a) Population size = 60 b) Elitism probability = 0.15 c) Crossover probability = 0.9 d) Mutation probability = 0.005 e) Number of generations = 100

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

45

A. IEEE 14 bus system 1) ATC at Base case: Here, base case refers to case without any contingency. i) ATC without FACTS controllers: ATC is computed for a set of source/sink transfers using CPF. Table I shows the ATCs for IEEE 14 bus system without FACTs device.

When SVC is incorporated in the system, if we consider all buses of system, there are 14 possible locations for the SVC. The location code region are set as 14 integers as 1 to 14. The amount of compensation offered by the SVC is 0 to 0.1 p.u i.e., BSV C. After using RGA, the results obtained are shown in Table III. It shows that with the flow control function SVC increased the ATC significantly. Fig 6 shows the voltage profile for IEEE 14 bus system without and with SVC at bus 9, when ATC is computed for a transaction 1/9.

ii) ATC with the incorporation of TCSC: When TCSC is incorporated in the system, if we consider all lines of system, there are 20 possible locations for the TCSC. The location code region are set as 20 integers as 1 to 20. The amount of compensation offered by the TCSC is 20% inductive to 70% capacitive i.e., xTCSC. After usingRGA, the results obtained are shown in Table II. It shows that

Fig. 6: Bus voltage profile for without and with SVC at bus 9

Fig . 5: No. of Generations Vs Fitness

with the flow control function TCSC, the ATC has increased significantly.

Fig. 7: Bus voltage profile for without and with line outage cases

Fig 5 is the convergence characteristic of RGA and it shows a graph between generation and fitness function i.e., ATC (MW) when source/sink transfer is between bus 1 and bus 9. After 89 generations, the optimal value of TCSC location and compensation value are found. This shows the good convergence of this algorithm. iii) ATC with the incorporation of SVC:

2) ATC at contingency case: i) ATC with line outage: ATC is computed for a set of source/sink transfers using CPF, when line 16 is removed from the system that is connected between bus 13 and bus 14. Table IV shows the ATCs for IEEE 14 bus system without FACTs device, when line 16 is outage. Fig 7 shows the voltage profile for the IEEE 14 bus system with and without line outage

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

46

cases.

ii) ATC with the incorporation of TCSC: When TCSC is incorporated in the system, if we consider all lines of system, there are 19 possible locations for the TCSC. The location code region are set as 20 integers as 1 to 20 except line 16. After using RGA, the results obtained are shown in Table V. It shows that with the flow control function TCSC increased the ATC significantly even after line outage. iii) ATC with the incorporation of SVC: When SVC is incorporated in the system, if we consider all buses of system, there are 14 possible locations for the SVC. The location code region are set as 14 integers as 1 to 14.After using RGA, the results obtained are shown in Table VI.

B. IEEE 24 bus Reliability Test System i) ATC at Base case: i) ATC at base case without FACTs controllers: ATC is computed for a set of source/sink transfers using CPF. Table VII shows the ATCs for IEEE 24 bus system without FACTs device. ii) ATC with the incorporation of TCSC: When TCSC is incorporated in the system, if we consider all lines of system, there are 38 possible locations for the TCSC. The location code region are set as 38 integers as 1 to 38. After using RGA, the results obtained are shown in Table VIII. It shows that with the incorporation of TCSC, ATC has increased significantly. iii) ATC with the incorporation of SVC:

Fig. 8: Bus voltage profile for without and with SVC at bus-13

It shows that with the voltage control function SVC increased the ATC significantly during line 16 outage. Fig 8 shows the voltage profile for IEEE 14 bus system without and with SVC at bus 13, when ATC is computed for a transaction 1/13.

When SVC is incorporated in the system, if we consider all buses of system, there are 24 possible locations for the SVC. The location code region are set as 24 integers as 1 to 24. The amount of compensation offered by SVC is 0 to 0.1 p.u i.e., BSV C. After using RGA, the results obtained are shown in Table IX. It shows that with the incorporation of SVC, ATC has increased significantly.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

47

When SVC is incorporated in the system and if we consider all buses of system, there are 24 possible locations for the SVC. The location code region are set as 24 integers as 1 to 24. After using RGA, the results obtained are shown in Table XII. It shows that with the incorporation of SVC, the ATC has increased significantly even after line 8 outage.

2) ATC at contingency case: i) ATC with line outage: ATC is computed for a set of source/sink transfers using CPF, when line 8 is removed from the system that is connected between bus 4 and bus 9. Fig 9 shows a graph voltage profile for the IEEE 24 bus system with and without line outage cases. Table X shows the ATCs for IEEE 24 bus system without V. CONCLUSIONS In deregulated power systems, ATC is a critical issue either in the operating or planning because of increased area interchanges among utilities. Based on operating limitations of the transmission system and control capabilities of FACTS technology, technical feasibility of applying FACTS devices to boost ATCs are analyzed and identified. In this paper, the improvement of ATC using TCSC and SVC using CPF method is demonstrated with IEEE 14 bus and IEEE 24 reliability test systems during normal as well as contingency cases. The location and control parameter of TCSC and SVC in the system also affects the enhancement of ATC. Implementation of the proposed RGA has performed well when it is used to determine the location and compensation level of TCSC and SVC with the aim of maximizing the ATC. From the results, it is shown that installing SVC as a FACTS device will improve voltage profile as well as resulting ATC enhancement, where as TCSC can improve ATC in both thermal dominant case and voltage dominant case. It is clearly shows from the results that TCSC is more effective than SVC in improving ATC under both normal and contingency conditions. REFERENCES
[1] Transmission Transfer Capability Task Force, Available transfer capability Definitions and determination, North American Electric Reliability Council, June 1996. [2] Y. Dai, J. D. Mc Calley, V. Vittal, Simplification, expansion and enhancement of direct interior point algorithm for power system maximum loadability, Proc. 21st Int. Conf. Power Ind. Comput. Applicat., pp. 170- 179, 1999. [3] R. D. Christie, B. F. Wollenberg and I. Wangensteen, Transmission Management in the Deregulated Environment, Proceedings of the IEEE, vol.88, no.2, Feb. 2000. [4] Y. Xiao, Y. H. Song, Y. Z. Sun, Available Transfer Capability Enhancement Using FACTS Devices, Proceedings of the IEEE/PWS Summer Meeting, vol. 1, pp. 508-515, July 2000. [5] J. G. Singh, S. N. Singh and S. C. Srivastava, Placement of FACTS controllers for enhancing FACTS controllers, IEEE Power India Conference, 2006. [6] L. J. Cay, I. Erlich, Optimal choice and allocation of FACTS devices using Genetic Algorithms, IEEE PES Power Systems Conference and Exposition, pp. 201-207, 2004.

Fig 9. Bus voltage profile for without and with line outage cases

FACTs device, when line 8 is given outage.

ii) ATC with the incorporation of TCSC: When TCSC is incorporated in the system, if we consider all lines of system, there are 37 possible locations for the TCSC. The location code region are set as 20 integers as 1 to 37 except line 8. After using RGA, the results obtained are shown In Table XI. It shows that with the incorporation of TCSC, the ATC has increased significantly even after line 8 outage.

iii) ATC with the incorporation of SVC:

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

48

[7] IEEE Special Stability Controls Working Group, Static VAR Compensator Models for Power flow and Dynamic Performance simulation, IEEE Trans. on Power Syst., vol. 9, no. 1, pp. 229240, Feb. 1994. [8] H. A. Perez, E. Acha and C. R. F. Esquivel, Advanced SVC models for Newton-Raphson load flow and Newton Optimal Power flow studies, IEEE Trans. on Power Syst., vol. 15, no.1, pp.129-136, Feb. 2000. [9] L. Gyugyi, Power Electronics in Electric Utilities: Static Var Compensators, Proceedings of IEEE, pp. 483-494, vol. 76, no. 4, April 1988. [10] G. C. Ejebe, J. Tong, J. G. Waight, J. G. Frame, X. Wang, and W. F. Tinney, Available transfer capability calculations, IEEE Trans. on Power Syst., pp. 1521-1527, vol. 13, no.4, Nov. 1998. [11] M. Patel, A. A. Girgis, Review of Available Transmission Capability (ATC) Calculation Methods, Power Systems Conference, pp. 1-9, 2009. [12] V. Ajarapu, and C. Christie, The continuation power flow: a practical tool for tracing power system steady state stationary behavior due to the load and generation variations, IEEE Trans. on Power Syst., pp. 416-423, vol. 7, no. 1, Feb. 1992. [13] V. Miranda, J. V. Ranito, L. M. Proenca, Genetic Algorithms in Optimal Multistage Distribution Network Planning, IEEE Trans. on Power Syst., pp. 1927-1933, vol. 9, no. 4, Nov. 1994. [14] W. Feng, G. B. Shrestha, Allocation of TCSC Devices to Optimize Total Transmission Capacity in a Competitive Power Market, IEEE Trans. On Power Syst., pp. 587-592, Feb. 2001. [15] S. Gerbex, R. Cherkaoui, and A. J. Germond, Optimal Location of Multi-Type FACTS Devices in a Power System by Means of Genetic Algorithms, IEEE Trans. on Power Syst., vol. 16, no. 3, Aug. 2001. [16] K. Deb, Multi-objective Optimization using Evolutionary algorithms, John Wiley and Sons, Inc., New York, 2001. [17] M. Rashidinejad, H. Farahmand, M. F. Firuzabad, A. A. Gharaveisi, ATC enhancement using TCSC via artificial intelligent techniques, Electric Power Systems Research, vol. 78, pp. 11-20, 2008. [18] L. L. Freris and A. M. sasson, Investigation on the load flow problem, Proceedings of IEE, pp. 1459-1470, 1968. [19] IEEE Reliability Test System, IEEE Trans. on Power Apparatus and Syst., vol. 98, no.6, pp. 2047-2054, Nov./Dec. 1979

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

49

Unit Commitment and Economic Scheduling of 26 Thermal Units by Dynamic Programming


1
1, 2
1

Achala Jain, 2Kakoli Shaw


2

Shri Shankaracharya College of Engg and Technology, Bhilai, C.G. India


achalajai@gmail.com, kakolishaw@gmail.com

Abstract-- The decision problem of optimally scheduling the generating units in order to meet the load demand is known as unit commitment. Unit commitment under regulated conditions consists of commitment of electric power generators economically so as to meet the load demand, but in the deregulated environment there is a need to consider the decision of electric utility in a different manner. In practical situation there is a need to consider the failure and repair rates of generating units, and in order to maximize the expected profit there is a need to consider the volatility of the spot market price. This paper proposes to solve the unit commitment problem under deregulated conditions for 26 thermal units using dynamic programming on Matlab and illustrates the effectiveness of dynamic programming method for such applications. Keywordsunit commitment, deregulated system, economic scheduling, dynamic programming

Ti up

Minimum up time constrain Ramp-up constraint [MW/h] Dynamic Programming Unit Commitment Problem

up i

DP UCP

NOMENCLATURE

SU
i

t i

Start-up cost of unit i at time interval t. Combined crew start-up costs and the equipment maintenance costs of unit i [$] Cold start-up cost of unit i [$] Cooling time constant of unit i

i i

t off ,i

Continuous offline time of unit i at time interval t. Amount of power Control variable Market marginal price

Pi t

U
t

t i

Et
lt

Amount of energy

(MWh) Amount of energy sold under bilateral contract R ($/MWh) Contract price Total production cost
t i

t FTi

SD
X
t i

Shutdown cost Cumulative up time during time interval t Incremental Cost No load cost Elbow points

inci nlik eik

I. INTRODUCTION In order to run the electric power generation system economically so as to reliably meet the demand, it is thus necessary to switch the generating units on and off at appropriate times. The decision problem of optimally scheduling the operation of these machines is known as the unit commitment problem [1]. The restructuring and deregulation of electric power systems have resulted in market-based competition by creating an open market environment. A power generating utility solves the UCP to obtain an optimal production schedule of its units to meet customer demand. The unit commitment problem under deregulation is a large, non-linear, mixed integerprogramming problem [2]. Well-known mathematical programming techniques such as integer programming, branch and bound [3], Benders decomposition, simulated annealing [4] and Lagrangian relaxation method [2] have been used to solve the UCP. For small problems, they can provide the optimal solution in a reasonable amount of time. However, for large problems, the computational time required to find the optimal solution becomes prohibitive. But the DP technique [5] employs a systematic searching algorithm that tries to achieve the optimal solution without having to access all the possible combinations. The unit commitment problem can be solved using a dynamic programming algorithm. This technique can be applied because: (i) The problem satisfies the principle of optimality if all parts of an optimal solution are themselves optimal solutions to sub-problems. (ii) The number of relevant sub-problem depends on a limited number of smaller sub-problems. (iii) The number of relevant sub-problems is limited by the unit constraints. This paper proposes to solve the unit commitment problem under deregulated conditions for 26 thermal units using dynamic programming and illustrates the effectiveness of dynamic programming method for such applications.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

50

II. UNIT COMMITMENT PROBLEM UNDER DEREGULATION Unit commitment problem under deregulation [6] is particularly related with either power producer or generating company and it involves determining the optimum combination of available units (if a power producer is having M number of generating units) to serve the forecasted demand at minimum production cost and to obtain maximum profit, while observing all power system and unit constraints. The total production cost consists of fuel costs, transition costs (start-up/shut-down costs) and no-load costs (when a unit is idle or on standby) [7]. In a decentralized electricity market or in a deregulated environment, the aim is not to minimize the total production cost of a particular generating utility company or a power producer, but to maximize its total profit. This can be achieved by considering each generating unit separately and maximizing its profit independent of other units. The estimated values of the market clearing prices over the scheduling period is used as an input to the optimization of the schedule of this single unit. Once the optimal schedule is obtained, it can be used to assess which offers would be the most profitable based on a forecasted price profile. In practice, however, it may not be possible for a utility to make all the trades recommended at each trading period. III. MODELING THE OPERATION OF GENERATING
UNITS

The fuel cost,

Fi ( Pi t )

of unit i in any given time interval

t is a function of the power output, , of that unit during that time interval. The cost function is usually modeled as a second-order polynomial [1], as given by equation (2) (

Pi t

FTt,i (Pi t ) a i ( Pt t ) 2
=

bi Pi t

ci

(2)

Where For piece wise linear characteristics the cost function will be given by equation (3)
toff toff

ai , bi , and , ci

t T ,i

[(inc *
ton

k i

Pi ) SU

ton i ton

nlik ]U it

(3) Thermal units are subject to a variety of constraints. The unit constraints that must be satisfied during the maximization process are [8]: (i) Unit limits-units can only generate between given limits:

U it Pi m in

Pi t U it Pi m ax

For i=1,2,3N and

t=1,2,3 T (4) (ii) Unit minimum up time constraint: Once the unit is running, it should not be turned off immediately. ( i - i )( t=1,2,3T

Xt

T up U it
up

U it

for i=1,2,3N and (5)

For an electric utility or a power producer with M generating units, and given a certain market price profile it is required to determine the start-up/shut-down times and the power output levels of all the generating units at each time interval t over a specified scheduling period T, so that the generators total profit is maximized, subject to the unit constraints [1]. In determining an optimal commitment schedule, there are two decision variables and where, denotes the amount of power to be generated by unit i at time t and is the control variable whose value is chosen to be 1 if the generating unit i is committed at hour t and 0 otherwise [2]. All the formulae used for modeling have been taken from the Lagrangian relaxation method [2]. The total power production can be separated into two parts that is start-up cost and operating cost: (i) Start-up cost, given by equation (1) is warmthdependent, corresponding to the hot, warm or cold condition of each generating unit, as determined by the time that the unit has been off-loaded [1]. Its value depends on the shutdown time; alpha, beta and which can be obtained from unit data. (1) (ii) Each generating unit has a 'no-load' or fixed operating cost and a number of incremental operating costs, which can define a non-linear profile of operating costs [1].

Where i is the minimum up time constraint. (iii) Unit minimum down time constraint: Once the unit is decommitted (off), there is a minimum time before it can be recommitted.
i ( t=1,2,3T

Xt

Ti down U it U it
)( -

for i=1, 2, 3N and (6)

Where i is the minimum down time constraint. (iv) Unit ramp-up constraint- the amount a units generation can increase in an hour.

T down

Pi t Pi t
-

R iup

For i=1,2N and t=1,2 T

(7)

Where is the ramp-up constraint [MW/h]. (v) Unit ramp-down constraint-the amount a units generation can decrease in an hour.

up i

Pi t

Pi t
R

Ridown

For i=1,2N and t=1,2T

(8)

Where

down i

is the ramp down constraint [MW/h].

SUit

t (1 exp( X off ,i / i ))

Equations (7) and (8) apply to hours between start-up and shut down. The limit at start-up is given by: Max ( i , for i=1,2N and t=1,2T

Pi t

R up Pi m in

) (9)

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

51

The limit at shut down is given by:

Totalprofi t
t 1

Pr ofit it

Pi t

Max (

Ridown Pi m in
,

(14)

) for i=1,2N and t=1,2T

The main complication arises from the unit minimum up and down time constraints. When a unit is committed, it incurs a cost equal to its start-up cost. It then has to stay online until its minimum up time constraint has been satisfied before it can be shut down again. Similarly, once a unit is decommitted, it has to remain offline for as long as its minimum down time constraint requires before it can be recommitted [1]. Another difficulty lies in the timedependent nature of the start-up cost. Although committing a unit at a particular point in the scheduling period may not be the most profitable choice at that instant, it may still yield a better solution over the entire study period compared to the case in which the unit remained offline at the aforesaid point in time [1]. This option may have been totally lost during the optimization process though economic disqualification; i.e. if the feasible state associated with starting up the unit had been discarded because it incurred a loss during start-up. Thus the utility has to decide whether to: (i) Keep a unit committed even when the price is low during a particular period, incurring fuel cost during that period with the hope that the profit made in the following period when the price is high, together with the savings

(10) (vi) Unit status restrictions-certain units may be required to be online at certain time intervals (must run), or may become unavailable due to planned maintenance or forced outage (must not run), due to operating constraints, reliability requirements, or economic reasons. (vii) The initial conditions of the units at the start of the scheduling period must be considered. Plant crew constraints were not considered (thermal plants may have limits on the number of units that can be committed or decommitted in a given time interval due to manpower limits). Also, units may be derated (i.e. have their generating limits reduced), or required to operate at prespecified generation levels. These restrictions were also ignored. The start-up cost in any given time interval t depends on the number of hours a unit has been off prior to start-up. This can be modeled by an exponential function of the form:

SUit

t (1 exp( X off ,i / i ))

(11)

The shutdown cost. , is usually given a constant value for each unit per shutdown and in this paper is assumed to be zero. The total production cost, , for each unit at each time interval is the sum of the running cost, start-up cost and shutdown cost during that interval [1].
toff toff

SDit

t FTi

FTt ,i

[( incik
ton

Pi t ) SU iton
ton

nlik ]U it
(12)

achieved from not having to start up the unit at the start of or during the high-price period would offset the losses; or (ii) Shut down the unit during the period of low price and incur a cost when starting up the unit for the following high-price period. IV. OBJECTIVE FUNCTION The objective function is total profit, revenue minus cost over the interval (1,T) [1]. The revenue during hour t is obtained from supplying the quantity stipulated under the long-term bilateral contracts and by selling surplus energy (if any) to the power pool at the market price, t ($/MWh). The cost includes those of producing the energy, buying short falls (if needed) from the power pool, and the startup costs. Defining the amount to be sold under the bilateral contract by lt (MWh), the contract price by R

Usually the parameter is a Lagrangian multiplier [9], but in this deregulated case this parameter is considered to be the market price (whether it is observed price or a calculated one). The profit at each time interval is calculated by subtracting the total production cost during that interval from the revenue, as given by equation (13) a negative profit indicates a loss

Pr ofit

t i

( * Pi )U

t i

t T ,i

(13)

( t * Pi t )U it

= Revenue at time interval t.

As mentioned previously, the prices, t can be actual market prices or an estimate of how they would fluctuate, and are given in [$/MWh]. The total profit for unit i is then given by

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

52

($/MWh), and the amount of energy bought or sold from the market by Et, we solve the optimization problem by maximizing the expected profit over the period [1,T]. A positive value of Et indicates that Et (MWh) is bought from the power poll and a negative value indicates that E t (MWh) is sold to the pool. The objective function [1][9] is given by:
T M

2) Capacity limits:

U it Pi m in
U it

Pi t
1

U it Pi m ax
t iUP 1)

(20) (21) (22)

3) Minimum up time: (4)Minimum down time:

I (1 X it

U it

1 I ( t idown 1

X it

1)

I ( X ) =0 if X is false, Where
=1 if X is true. And

MaxE{
t 1

{l t R

Et
i 1

F }
(15)

t T ,i

t i

=1 if

t i

>0 <0
M

This is under the case that the cost function is represented by piece wise linear cost characteristics. If cost function is same as in the form of quadratic cost function then the maximum value of objective function will become
T M t

=0 if

X it
Et

Pi t
i 1

lt

E{ {l t R
t 1

Et
i 1

[CFi ( Pi t ) SU it ( X it 1 )(1 U it 1 )]U it }


(16)

After substituting in the objective function obtained from equation (15) we can rewrite equation (15) as follows.
T M M

E{
T

{
M

(l t

Pi t )

FTt ,i }

t 1 i 1 i 1 Max (23) Which after removing the constant term is equivalent to

Where the cost function [9] is given by

CFi ( Pi t ) FTt,i (Pi t ) a i ( Pt t ) 2


= =

bi Pi t

ci

(17)

E{

Pi t

FTt ,i }

but we usually consider that the cost function will be represented by piece wise linear cost characteristics because this representation will fetch us to take the values of cost function as discrete values. And hence the objective function will be given by equation (15) Since the quantity ltR is a constant, the optimization problem reduces to:
T M t t T ,i

t 1 i 1 Max (24) Subject to the operating constraints (19), (20), (21) and (22) because these constraints refer to individual units only. Equation (24) shows that the optimization problem is now separable by individual units. The optimal solution can be found by solving M-decoupled sub problems. Thus the sub problem for the i th unit [9] is

E{

Pi t

FTt ,i }

MaxE{
t 1

Et
i 1

F }
(18)
toff toff

FTt ,i
Where

[( incik
ton

Pi t ) SU iton
ton

nlik ]U it
as

given by equation (3) The time duration for operation of 26 thermal units for dynamic programming algorithm is assumed to be 10 hours. V. PROGRAMMING CONSTRAINTS Programming has been done subject to the following constraints [10] (for t=1T, i=1M). The specifications for all 26 thermal units are taken from [9].
M

t 1 Max (25) subject to the operating constraints of the i th unit. Equation (25) is similar to the sub-problem obtained in the standard version of the UCP using the Lagrangian relaxation method, except that the values of Lagrange multipliers are now replaced by the market price of electricity t and the expected value is being maximized. When the optimization sub-problem is solved for a particular unit, we assume that the market consists of N generating units (N will be much larger than M). The generating unit for which the sub-problem is solved is excluded from the market. Excluding a unit from the market does not influence the spot price because of the existence in all likelihood of a number of generating units with almost equal marginal costs, ready to produce if any of the inframarginal (or marginal) units are unavailable.

Et
1) Load:
i 1

Pi t

lt
(19)

VI. DYNAMIC PROGRAMMING ALGORITHM The unit commitment problem can be solved [5] in a bottom-up manner whereby: (i) The smallest sub-problems are solved first. This corresponds to finding the feasible states (whether 0 or 1),

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

53

the associated nominal generation or dispatch and the profit that it would entail for each unit at each time interval. (ii) These solutions are then combined to solve larger subproblems. In this case the individual profits of each feasible solution path are added together to give the total profit over the scheduling period, after which the path with the highest total profit is determined. This gives the optimal schedule for that unit for a given price profile. (iii) Finally, the individual maximum total profits are summed over all the units in the utility to give its maximum total profit for a given price profile. The dynamic programming algorithm [5] is given as follows: (i) Specify the rule that relates large problems to small problems. (ii) Store the partial feasible solutions of each sub problem. (iii) Extract the final solution for main problem considering all solutions from sub problems The flow chart [9] is as given by Fig1.

(i) The unit constraints were respected in all cases. (ii) A negative net profit was obtained whenever the unit was brought online, due to the actual profit being amortized by the start up cost. If a negative value of profit is there that means it is a loss incurred at that time due to that particular unit. If profit value is positive it is obviously a gain for the producer or a utility. So considering all the units, the best profit for the whole system is calculated. Table 1 shows the profit calculated for each unit individually in 10 hours considering the states (on/off) [9] of 26 units shown in table 2. Hence dynamic programming is useful in aiding a producer or utility to make a quick and appropriate decision to achieve optimal scheduling of thermal units for maximum profit while meeting the load demand.
TABLE 1: CALCULATED TOTAL PROFIT

UNIT 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

TOTAL PROFIT FOR (T=10) HOURS 1414.134 1396.929 1377.049 1359.646 1342.546 151.066 140.274 129.071 118.107 19122.986 19084.224 19047.897 19207.385 18663.634 18795.565 18696.600 44956.134 45002.021 44936.085 44875.456 29841.747 29614.585 29377.709 103351.435 132119.379 132066.055 776190.00($/MWh)

Fig.1: Dynamic programming flow chart

24 25 26 TOTAL PROFIT

VII. RESULTS AND DISCUSSION The dynamic programming has been executed using MATLAB. Total profit of 26 thermal units has been calculated for 10 hours. Based on the results obtained, the following general observations can be made:

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

54

REFERENCES
[1] Wood A; B. Wollenberg, Power generation operation and control Second edition, Wiley & Sons, New York 1996 [2] Baldick, R, the generalized unit commitment problem IEEE transactions on power systems, vol: 10 issue: 1, Feb. (1995) pp: 465-475. [3] Sheble, G.B; Fahd, G.N unit commitment literature synopsis, IEEE transactions on Power systems, vol: 9 issue: 1, Feb (1994) pp: 128-135. [4] Zhuang, F; Galiana, F. Unit commitment by simulated annealing, IEEE transactions on power systems, vol.5, no.1, February (1990), pp.311-317 [5] Snyder, L. Walter; David H. Powell Dynamic programming approach to unit commitment IEEE transactions on power system, May (1987), vol.PWRS-2, pp.339-347. [6] Chowdhury, N; Bhuiya, A. Counter flow in a deregulated power system network and its effect on transmission loss allocation power system research group, pp. 1047-1050.

[7] Kothari, D.P; Ahmad, A an expert system approach to unit commitment problem; TENCON 93. Proceedings Computer, communication, control and power engineering.1993 IEEE region 10 conference on issue:19-21 Oct,(1993), vol.5, pp.5-8. [8] J. Valenzuela & M. Mazumdar "Monte Carlo Computation of Power Generation Production Costs under Operating Constraints," IEEE Transactions on Power Systems, vol.16 (2001), 671-677. [9] Wang, C and Shahidehpour, S.M, Effects of ramp rate limits on unit commitment and economic dispatch IEEE transactions on power systems, vol.8, no.3, August 1993,pp.1341-1350. [10] Cohen, A.I; Brandwahjn, V; Show-kan Chang security constrained unit commitment for open markets power industry computer applications, PICA proceedings of the 21st IEEE international conference, 16-21 may (1999) pp: 39-44.

TABLE 2: STATES (ON/OFF) FOR 26 UNITS

Unit 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

t=1 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 1 1 1

2 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 1 1 1

3 1 1 1 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 0 0 0 1 1 1

4 1 1 1 1 1 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

5 1 1 1 1 1 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

6 1 1 1 1 1 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

7 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

8 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

9 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

10 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

55

Economic Emission Generation Scheduling using GA


1
1,2
1

Preetinder Kaur, 2S.K.Bath


2

GZS College of Engg. And Tech. Bathinda, Punjab, India


sjkbath@yahoo.co.in

preet.inder22@yahoo.com,

Abstract The optimal power flow has been frequently solved using classical optimization methods. The OPF has been usually considered as the minimization of an objective function representing the generation cost and/or the transmission loss. The objective of this paper is to present solution of optimization problem of large transmission systems via a simple genetic algorithm using MATLAB GA toolbox. The IEEE 30-Bus system has been studied to show the effectiveness of algorithm. The power flow is calculated using NewtonRaphon method and the results are observed and presented. Keywords Load flow, Optimization, Genetic algorithm, Fuzzy set.

I. INTRODUCTION OPF has enjoyed the renewed interest in a variety of formulations, through the use of advanced optimization techniques such as, Genetic Algorithms, Interior Point methods, simulated annealing method, decomposition and Newtons method, and their contributions have been significant. Optimization is the process of making something better. An engineer or scientist conjures up a new idea and optimization improves on that idea. Optimization consists in trying variations on an initial concept and using the information gained to improve on the idea. The optimization of a large power system problem is defined as to determine the allocation of electricity demand among the committed generating units to minimize the operating costs subjected to physical and technological constraints. In our country the existing production process is not ecologically clean. The pollution minimization problem attracted a lot of attention. The thermal power plants are the major causes of atmosphere pollution because of high concentration of pollutions they cause. All the consumers demand the electric power at the cheapest possible rate. But the minimum cost dispatch leads to very high pollution in the atmosphere. The alternative is to supply the power to consumers at minimum emission dispatch, which leads to increase in cost of production of electric power. One of the major complications introduced in above considerations are conflicting functions in that cost and emission functions are conflicting functions in that minimizing pollution maximizes cost and vice versa. So the multiple criteria must be considered simultaneously to attain more meaningful, practical and optimal schedule of operation.

The problem that handles both the objectives simultaneously is known as multi-objective optimization. Different classical techniques have been also employed to solve the OPF problem e.g. Lambda iteration method, Gradient method, Newtons method [2], linear programming method [3], Interior point method [4]. Lambda iteration method also called the equal incremental cost criterion (EICC) method has its roots in the common method of economic dispatch used since the 1930s [1][5]. However, all of these methods suffer from three main problems. Firstly, they may not be able to provide the optimal solution and usually getting stuck at a local optimum [6].Secondly, all these methods are based on the assumption of continuity and differentiability of the objective function, which is not true in a practical system. Finally, all these methods cannot be applied with discrete variables which are transformer taps. The major advantage of the GA is that it is relatively versatile for handling various qualitative constraints. It can find multiple optimal solutions in single simulation run. So they are quite suitable in solving multi-objective optimization problems. MULTIOBJECTIVE OPTIMIZATION PROBLEM FORMULATION Today the main concern is not only to achieve the minimum overall cost but is to reduce environmental pollution also because the existing energy production processes are not ecologically clean. The combustion of fuels give rise to particulate matter and gaseous pollutants. Thus reduction of this pollution is another equally important criterion for optimization in addition to cost. So consideration of both these factors in optimization process leads to multi-objective problem. Generally the objectives are in conflict with each other and all of them can not be minimized [7], only a balance can be obtained in them. One best solution can be chosen depending upon the selection criterion of decision maker. The multiple objectives taken into account for optimization are minimization of (i.) operating cost (ii.) pollutant emissions. A. Operating Cost The cost curve is approximated by a quadratic function of generators real power output, P Gi. Cost function F1 is obtained by considering the cost coefficients and active power generations. II.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

56

F1

(1)

where, ai, bi, ci are emission coefficients, and P Gi is the active power generation of ith generator. NG is the number of generators. B. Pollutant Emissions The emission curve can be directly related to the cost curve through the emission rate per Mbtu ( 1 Btu=1055.06 j) which is constant factor for a given type of fuel. Therefore, the amount of pollutant emission is given as: F2 (2)

minimized simultaneously. So the multi-objective optimization with such conflicting objective functions gives rise to a set of optimal solutions instead of one optimal solution Such optimal solutions are known as non-inferior solutions [7]. The aim of optimization is to optimize two conflicting objectives cost (FC1) and emission (FC2). To compare these two different types of objectives. We use fuzzy set theory and formulate membership function F1 and F2 to denote the achievement of objectives in different noninferior solutions. Then problem becomes: (8) Subject to: inequality and equality constraints, such as: ; i=1,2,.. NG (8) (9) where and are respectively the minimum and maximum values of real power generation allowed at generator bus i. Genetic Algorithm searches the generation schedule with in maximum and minimum limits of active power generation. MATLAB based Genetic Algorithm tool box tries to minimize the objective function. Therefore to explore minimization optimization, we redefine fuzzy function in the form of dissatisfaction in the achievement of objectives as: (10) (11) Then we use min-max technique which tries to minimize the maximum dissatisfaction between the two objectives. Therefore final fitness function becomes: (12) IV. GENETIC ALGHORITHM The genetic algorithms are part of the evolutionary algorithms family, which are computational models, inspired by Nature. Genetic algorithms are powerful stochastic search algorithms based on the mechanism of natural selection and natural genetics. GAs works with a population of binary string, searching many peaks in parallel. By employing genetic operators, they exchange information between the peaks, hence reducing the possibility of ending at a local optimum. GAs are more flexible than most search methods because they require only information concerning the quality of the solution produced by each parameter set (objective function values) and not like other optimization methods which require derivative information, or worse yet, complete knowledge of the problem structure and parameters[11]. There are some difference between GAs and traditional searching algorithms [11-13]. They are summarized as the algorithms that work with a population of strings,

The emission function F2 is formulated by considering the emission coefficients i, i, i and active power generation, PGi, at different generators. C. Active Power Transmission Loss For each generation schedule, the expected active power transmission losses are obtained from load flow by Newton Raphson method as: The complex power fed into the line connected between bus i and bus k is: = ; i=1, 2.NB; k=1, 2. NB (3)

Where is the current injected on the line between buses i and k and is given as: (4) Total transmission loss can be computed by summing all the line flows: Complex Power Loss = ) (5)

Real part of equation (5) gives the active power loss, P L. D. Membership Function Of Objectives The fuzzy sets are defined by equations called membership functions. These functions represent the degree of membership in certain fuzzy sets using values from 0 to1. The value of membership function indicates up to which degree a solution is satisfying a objective. Here (Fi) is assumed to be decreasing and continuous function defined as:[7-9]

where i=1,2.

(6)

III. MULTIOBJECTIVE PROBLEM FORMULATION Multi-objective optimization problem is formulated to schedule real power generation optimization. The objectives are the function of real power. The objectives are in conflict with each other and all of them can not be

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

57

searching many peaks in parallel, instead of a single point. GAs work directly with strings of characters representing the parameters set not the parameters themselves. GAs use probabilistic transition rules instead of deterministic rules. GAs use objective function information instead of derivatives or others auxiliary knowledge. GAs has the potential to find solutions in many different areas of the search space simultaneously [8-17]. GA uses crossover, mutation and Reproduction operators. Crossover is the primary genetic operator, which promotes the exploration of new regions in the search space. Mutation is a secondary operator and prevents the premature stopping of the algorithm in a local solution. The mutation operator is defined by a random bit value change in a chosen string with a low probability of such change. The mutation adds a random search character to the genetic algorithm, and it is necessary to avoid premature convergence after some generations. Reproduction is based on the principle of survival of the better fitness. It is an operator that obtains a fixed number of copies of previous generation. Fig1. shows a standard procedure of a Genetic algorithm.

iii.) Calculate objectives: cost and emissions using eq.1 and 2. iv.) Calculate membership functions, F1, and F2 using eq. 6. v.) Calculates the dissatisfaction D1, D2 by using eq.10 and 11. vi.) Calculates the minimization fitness function D using eq. 12. 5. GA will be run for the given number of generations to give the final best solution until otherwise terminated by some stopping criteria. VI. TEST SYSTEM The Genetic algorithm optimization method is applied to the IEEE 30 bus, 6 Generators, 41 lines, power system [18]. Real powers are taken as control variables. Different combination of real power generations are searched using Matlab GA toolbox, considering both the objectives simultaneously. The best solutions are selected using min-max technique for the dissatisfaction to be minimized. The minimum and maximum values of objectives are calculated by using lower and upper limits of generators and are shown in Tables 1 and 2 respectively. Generation coefficients and emission coefficients are given in Tables 3 and 4 respectively. Fig2 shows the IEEE 30-bus test system.
TABLE 1: MAXIMUM AND MINIMUM COST AND EMISSION VALUES.

Objective function Operating cost ($/h) Emission (Kg/h)

Fi (Min.) 239.0 142.0

Fi (Max.) 1157.0 716.0

Following is the IEEE 30 Bus, 6 Generators data [18]:


TABLE 2: REAL POWER GENERATION LIMITS

UNIT 1 Fig. 1: Standard Procedure of a Genetic Algorithm. 2 3

Pi min (MW) 50 20 15 10 10 12

Pi max (MW) 200 80 50 35 30 40

V. ALGORITHM FOR GENETIC ALGORITHM BASED OPF Stepwise algorithm is described below to achieve the solution by Genetic Algorithm. 1. Read the database for the generator data, bus data and transmission line data. 2. Assume suitably population size, maximum number of generations. 3. Input main function handle and required data 4. Run GA. It will use main function, which does the following: i.) Calculate load flow by using Newton Raphson method. ii.) Modifies slack bus power.

4 15 6

TABLE 3: EMISSION COEFFICIENTS

Unit 1 2

i (Kg/MWh) 0.0126 0.0200

i (Kg/MWh) -1.1000 -0.1000

i 22.983 22313

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

58

3 4 5 6

0.0270 0.0291 0.0290 0.0271

-0.1000 -0.0050 -0.0400 -0.0055

25.505 24.900 24.700 25.300

TABLE 4 : GENERATOR COST COEFFICIENTS

Unit 1 2 5 8 11 13

ai ($/MWh) 0.00375 0.01750 0.06250 0.00834 0.02500 0.02500

bi ($/MWh) 2.00 1.75 1.00 3.25 3.00 3.00

ci 0 0 0 0 0 0

Fitness Scaling = Rank; scales the raw scores based on the rank of each individual, rather than its score. The rank of an individual is its position in the sorted scores. Mutation = Adaptive Feasible; randomly generates directions that are adaptive with respect to the last successful or unsuccessful generation. A step length is chosen along each direction so that linear constraints and bounds are satisfied. Crossover = Two point; selects two random integers m and n between 1 and Number of variables. The algorithm selects genes numbered less than or equal to m from the first parent, selects genes numbered from m+1 to n from the second parent, and selects genes numbered greater than n from the first parent. The algorithm then concatenates these genes to form a single gene. Selection = Roulette; simulates a roulette wheel with the area of each segment proportional to its expectation. The algorithm then uses a random number to select one of the sections with a probability equal to its area. Stall Time = 200; if there is no improvement in the best fitness value for an interval of time in seconds specified by Stall time limit, the algorithm stops. Table 5 shows the conflicting nature of two objectives. When cost of operation is minimized the cost of emission is increased and vice versa.
TABLE 5: CONFLICTING VALUES OF OBJECTIVES FC1 AND FC2 No. of Gens. FC1 FC2 (minimizing FC1 ) Gen=60 445.1 261.6 (minimizing FC2 ) Gen=60 459.9 239.3

Fig. 2: IEEE 30 Bus Test System

VII. RESULTS AND DISCUSSION In this paper two objectives are considered for optimization: operating cost and pollutant emissions this multi-objective problem is solved using GA. The active power generation schedules are optimized within the lower and upper limits of the generators to optimize the conflicting objectives i.e. cost and emission which are dependent on the active power generation. The results are obtained using different options of GA such as population, mutation, crossover, selection, stall time and number of generation. These options are varied one by one and each time the results are compared and the value of each option which gives the best value is chosen. Then fixing that values of the GA options, no. of generations are varied in steps and the best optimized solution is presented. The best values of GA options chosen are: Population = 20; specify options for the population of the genetic algorithm. Population Type = Double Vector; specifies the type of the input to the fitness function.

Table 6. shows the optimized generation schedules obtained after GA run for different number of generations.
Best: 0.2293 Mean: 0.23442 0.4
Best fitness

0.38 0.36 0.34

Mean fitness

Fitness value

0.32 0.3 0.28 0.26 0.24 0.22 0.2

10

20

30 Generation

40

50

60

Fig 3: Plot of best fitness value versus the no. of generations

As shown in table 5. One of the objectives, FC1, cost is minimized and value of second, FC2 increases and in other case when FC2, emissions are minimized, value of cost increases from previous value. The no. of generations in GA are increased in the steps of 10,

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

59

starting from 10 generations to 60 generations. The observed results for the different set of optimized values of active power generation at different iterations are shown in Table 6. Table 7 shows the real power loss from the first step to last step of generations, i.e. active power loss decreased from 9.94 to 9.50 MW. Difference in value of power (P) shows the convergence of eq. (9) at each generation schedule. The values of operating cost is reduced from 489 $/h to 450.1 $/h. Also the emissions are optimized from 276.3 to 255.0 Kg/h at the last generation. The membership function f values and the minimized dissatisfaction best fitness D value for different generations are also shown in Table 7. The best fitness also varies from 0.273 to 0.230. The best fitness value is at the last generation, i.e. 0.770, it is obtained at generation 51 as no further change is seen in the fitness value as shown in Fig 3. Table 8 shows the time taken to optimize the objective function at different generations. Time increases with the no. of generations as no. of function evaluations increase with no. of generations. It is observed from the obtained results that there is not much improvement in the fitness value after 30 generations. VIII. CONCLUSIONS

REFERENCES
A.J. Wood and B. F. Wollenberg, Power Generation Operation and Control, New York, NY: John Wiley & Sons, Inc., 1996. [2]. D.I Sun, B.Ashley, B.Brewer, A.Hughes and W.F. Tinney, Optimal power flow by Newton approach, IEEE transactions on power apparatus and systems, vol. 3, pp. 103, October 1984. [3]. O.Alasc, J. Bright , M. Paris and B.Scott, Further developments in LP Based Optimal Power Flow, IEEE Transactions on Power Systems, vol 5, August 1990. [4]. Y. Wu, A. S. Debs and R. E. Marsten, Direct Nonlinear Predictor- Corrector Primal-Dual Interior Point Algorithm for Optimal Power Flows, IEEE Power Industry Computer Applications Conference, 1993. [5]. James Daniel and Weber B.S, Implementation of A NewtonBased 0ptimal Power Flow into a Power System Simulation Environment, University of Wisconsin -Platteville, 1995. [6]. Bjorndal, M.Jornsten and K.Zonal, Pricing in a Deregulated Electricity Market, The Energy Journal 22, 2001. [7]. S.K. Bath, J.S. Dhillon and D.P. Kothari, Stochastic multiobjective generation scheduling evolutionary search based GA,International conference:CERA,IIT, Roorkee,pp.66-74, Sep28-oct1,2008 [8]. S.K. Bath, J.S. Dhillon and D.P. Kothari, Fuzzy satisfying stochastic multi-objective generation scheduling by weightage pattern search methods, electric power systems research, vol.69,pp.311-320,2004 [9]. J.S.Dhillon,S.C.Parti and D.P.Kothari, Fuzzy decision making in Multi-objective long term scheduling of hydro thermal system, International journal of electrical power and energy systems, vol.24,pp19-29,2001 [10]. H John Holland, Adaptation in Natural and Artificial Systems, University of Michigan Press, Ann Arbor, 1975. [11]. Tarek Bouktir, Linda Slimani and M. Belkacemi, Genetic Algorithm for Solving the Optimal Power Flow Problem, Department of Electrical Engineering, University of Oum El Bouaghi, 04000, Algeria. [12]. L.L. Lai, J. T. Ma, R. Yokoma and M. Zhao , Improved Genetic Algorithms For Optimal Power Flow Under Both Normal and Contingent Operation States, Electrical Power & Energy System, Vol. 19, pp. 287-292, 1997. [13] B. S. Chen, Y. M. Cheng and C. H. Lee, A Genetic Approach to Mixed H2/H00 Optimal PID Control, IEEE Control system, pp. 551- 59, Oct 1995. [14] W. Stagg Glenn, H. El Abiad Ahmed, Computer Methods in Power System Analysis, Mc Graw Hill international Book Company, 1968. [15] D.E Goldberg, Sizing Populations for Serial and Parallel Genetic Algorithms, in J.D. Schaffer, editor, Proceedings of the Third International Conference on Genetic Algorithms, Morgan Kaufmann, 1989. [16] John R. Koza, Genetic Programming for Encyclopedia of Computer Science and Technology, edited by Allen Kent and G. Williams James, vol 2, Submitted August 18, 1997. [17] Koza, R. John and Andre David, Evolution of Iteration in Genetic Programming in Evolutionary Programming, V: Proceedings of the Fifth Annual Conference on Evolutionary Programming. Cambridge, MA: MIT Press, 1996. [18] http://ietd.inflibnet.ac.in/bitstream/1063/1271/18/18_appendix.pd [1].

In the present paper an attempt has been made to solve multi-objective optimization problem. Even though, excellent advancements have been made in classical methods, they suffer with disadvantages: In most cases, mathematical formulations have to be simplified to get the solutions because of the extremely limited capability to solve real-world large-scale power system problems. They are weak in handling qualitative constraints. They have poor convergence, may get stuck at local optimum, they can find only a single optimized solution in a single simulation run, they become too slow if number of variables are large and they are computationally expensive for solution of a large system. The major advantage of using the GA is that it is relatively versatile for handling various qualitative constraints. It can find multiple optimal solutions in single simulation run an consumes very less time as compared to any other methods used. So they are quite suitable in solving multi-objective optimization problems. In most cases, GA can find the global optimum solution. The added advantages of GA are adaptability to change, ability to generate good enough solutions and rapid convergence. GA is used to solve multi-objective generation scheduling to minimize total cost of power generation and pollutant emissions.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

60

TABLE 7. BEST OPERATING COST AND POLLUTANT EMISSIONS, MEMBERSHIP FUNCTIONS, DISSATISFACTION FUNCTIONS AND BEST FITNESS VALUES AT DIFFERENT GENERATIONS IN GA
No. of Gens in GA

10 Loss (MW) Real power convergence -7.11 e-015 Cost $/h Emis-sion Kg/h 276.3 F1 F2 D1 D2 f Fitness D 489.7 9.94

20 9.28

30 9.59

40 9.74

50 9.25

60 9.50

-1.73 e-014

-1.78 e-015

00

9.77 e-015

-4.44 e-015

463.9

450.0

467.2

450.6

450.1

273.0

253.1

262.6

261.7

255.0

0.727 0.766 0.273 0.234 0.727 0.273

0.775 0.772 0.245 0.228 0.775 0.245

0.770 0.807 0.229 0.194 0.770 0.229

0.752 0.789 0.249 0.210 0.752 0.249

0.769 0.792 0.231 0.209 0.769 0.231

0.770 0.803 0.230 0.197 0.770 0.230

TABLE 6: BEST GENERATION SCHEDULES AND ACTIVE POWER LOSS AT DIFFERENT GENERATIONS No. of GA Gens PG1 (MW) PG2 (MW) PG3 (MW) PG4 (MW) PG5 (MW) PG5 (MW)

10 20 30 40 50 60

130.1 70.17 50.00 110.2 110.3 191.6

169.3 55.51 50.20 49.35 48.31 48.64

47.52 49.73 43.95 49.55 49.47 49.13

45.90 31.22 35.00 34.98 35.00 34.93

32.16 29.97 30.00 29.93 29.98 29.95

25.05 39.09 40.00 39.79 39.39 39.84

TABLE 8: TIME TAKEN AT DIFFERENT GENERATIONS IN GA.


No. of Gens in GA

10

20 3 min 44 sec

30 5 min 11 sec

40 7 min 42 sec

50 8 min 43 sec

60 (51) 9 min 12 sec

Time (sec)

1 min 55 sec

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

61

Classical Approach for Optimal Allocation of Distributed Generation in a Distribution System


1

Sandeep Kaur, 2Vivek Goyal


PEC University of Technology
2

1, 2 1

Saroa_sandip@yahoo.com,

Vivek007goyal@gmail.com

Abstract recent changes in the electric utility infrastructure have created opportunities for many technological innovations, including the employment of Distributed Generation (DG) to achieve a variety of benefits. After a brief discussion of the benefits, this paper proposes a classical approach for the optimization of DG with classical method and same validated on 12 bus test system and the results are calculated by considering voltage profile improvement index and line loss reduction index. Keywords Distributed Generation, Classical Approach

NOMENCLATURE
VPII Voltage profile improvement index. LLRI Line-loss reduction index. BI Distributed generation benefit index. VPw/DG Voltage profile index of the system with DG, pu. VPwo/DG Voltage profile index of the system without DG,pu. Vi Voltage magnitude at bus i, pu. Li Load at bus i, pu. Ki Weighting factor for load bus i. LLw/DG Total line losses in the system with employment of DG, pu. LLwo/DG Total line losses in the system without DG, pu. IL,i Line current in distribution line i without DG, pu. IA,i Line current in distribution line i with employment of DG, pu. Ri Line resistance for line i, pu/km. Di Line length for distribution line i, km. BWLLR Benefit weighting factor for line-loss reduction. BWVPI Benefit weighting factor for voltage profile improvement.

As the penetration of DG in distribution system increases, it is in the best interest of all players involved to allocate DG in an optimal way such that it will reduce system losses and hence improve voltage profile. Several benefits accrue by integrating DG with utility networks. These benefits should be clearly understood, analyzed, and quantified in order to increase the potential and value of DG penetration. The benefits of DG have been evaluated and quantified in terms of capacity credit, energy value, and energy cost saving [3], [4]. In addition, quantification of voltage profile improvement, line-loss reduction, and environmental impact reduction have attracted the attention of researchers [5], [6]. The purposes of this paper are to briefly discuss the benefits of employing DG and propose a general approach and a set of indices to assess and quantify some of the technical benefits of DG in terms of voltage profile improvement, line-loss reduction, and environmental impact reduction. II. BENEFITS OF EMPLOYING DG Most of the benefits of employing DG in existing distribution networks have both economic and technical implications and they are interrelated. While all the benefits can be ultimately valuated in terms of money, some of them have a strong technical flavor than others. As such, it is proposed to classify the overall benefits as under [7], [8]. Reduced line losses Voltage profile improvement Reduced emissions of pollutants Increased overall energy efficiency Enhanced system reliability and security Improved power quality Relieved T&D congestion Deferred investments for upgrades of facilities Reduced O&M costs of some DG technologies Enhanced productivity Reduced fuel costs due to increased overall efficiency Reduced reserve requirements and the associated costs Lower operating costs due to peak shaving Increased security for critical loads In this paper, a general approach is presented to quantify the overall benefits of DG by optimal allocation of distributed generation. It is then applied to assess two major technical benefits, namely voltage profile improvement, and line-loss reduction.

I. INTRODUCTION The share of distributed generators (DGs) in power systems has been slowly increasing in the last few years. According to CIGRE report [1], the contribution of DG in Denmark and the Netherlands has reached 37% and 40%, respectively, as a result of liberalization of power market in Europe. Electric Power Research Institutes (EPRI) study forecasts that 25% of the new generation will be distributed by 2010 and a similar study by the Natural Gas Foundation believes that the share of DG in new generation will be 30% by the year 2010 [2]. The numbers may vary as different agencies define DG in different way, however, with the Kyoto protocol put in place where there will be a favorable market for DG that are coming from Green Technologies, the share of DG would increase and there is no sign that it would decrease in near future. Moreover, the policy initiatives to promote DG throughout the world also indicate that the number will grow rapidly.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

62

III. METHODOLOGY The broad benefits of introducing DG in any system are:i) Improvement of definite parameters like voltage profile, reliability and power quality. ii) Reduction in line losses and other emission related parameters. IV. EVALUATION OF BENEFITS A set of indices is proposed to quantify some of the benefits of DG. They are VPII, LLRI and BI A. VPII The proposed VPII quantifies the improvement in the VP in a simple manner with the inclusion of DG. It is defined as the ratio of the voltage profile index of the system with DG to the voltage profile index of the system without DG (base case system) and is expressed as

the system with DG to the total line losses in the system without DG and is expressed as LLRI = (3) Where LLw/DG is the total line losses in the system with the employment of DG and is given as

LLW/DG =

I2A,i Di

(4)

Where IA,i is the per-unit line current in distribution line i with the employment of DG, Ri is the line resistance for line i (pu/km), is the th distribution line length (km), and M is the number of lines in the distribution system. Similarly, LLwo/DG is given as

VPII =

(1)

LLWO/DG =

I2L,iDi

(5)

Where VPw/DG and VPwo/DG are measures of the voltage profile of the system with DG and without DG, respectively, with the same loads at the different load buses. The general expression for VP is given as

Where IL,i is the per-unit line current in distribution line i without DG. As before, the loads at the different load buses are assumed to be the same both with and without DG. C. BI The BI is a composite index proposed to quantify the overall benefits of DG. Among the several benefits offered by DG, only three major ones are considered in this paper: voltage profile improvement, line-loss reduction, and environmental impact reduction. Therefore, BI can be formulated as in (6).
BI = (BWVPI) (VPII) + (BWLLR) / (LLRI) (6)

VP =

Li ki

(2)

With

=1

Where Vi is the voltage magnitude at bus i in per-unit,Li is the load at bus i in per-unit,Ki is the weighting factor for load bus i, and N is the total number of load buses in the distribution system. As defined, the expression for VP provides an opportunity to quantify and aggregate the importance, amounts, and the voltage levels at which loads are being supplied at the various load busses in the system [11]. B. LLRI Another major potential benefit offered by DG is the reduction in electrical line losses. The loss can be significant under heavy load conditions. The utility is forced to pass the cost of electrical line losses to all customers in terms of higher energy cost. With the inclusion of DG, line loss in the distribution system can be reduced. Obviously, line-loss reductions are due to reductions in power flows resulting from the introduction of DG. However, depending on the ratings and locations of DG units, it is possible to have an increase in loss at very high (and unrealistic) penetration levels. The proposed LLRI is defined as the ratio of total line losses in

With and

0 BWVPI 1 0 BWLLR 1 BWVPI + BWLLR = 1 (7)

Where BWVPI and BWLLR are the benefit weighting factors for voltage profile improvement and line-loss reduction respectively. Once again, the choice of weighting factors comes into question. The simplest approach is to give equal weights to the two indices considered in this study. If more indices are included, they can all be given equal weights. However, if DG is introduced to mitigate a certain specific problem (such as voltage profile improvement or lowering emissions), then the corresponding index can be assigned a greater weight as compared to others. V. COMPUTATIONAL PROCEDURE Step 1: Run the base case load flow. Step 2: Increment the size of DG in small steps then calculate voltage profile improvement index and line loss reduction index for each bus by running load flow.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

63

Step 3: Select the size of DG that gives the minimum loss and voltage profile at each bus. Step 4: Compare the loss with the previous solution. If loss is less than previous solution, store this new solution and replace previous solution. Step 5: Repeat Step 2 to Step 4 for every bus. VI. TEST SYSTEM & CASE STUDY Radial distribution systems serving loads far away from transmission and sub transmission (medium voltage) lines are widely used. Any DG units introduced in such lines will typically have low ratings (a few kilowatts up to a few megawatts), most likely installed by customers in their premises. However, DG units in the tens of megawatts (up to 100 MW) will most likely find their place in mediumvoltage sub transmission systems. The test system used in this paper belongs to this category and is similar to the one illustrated in [12]. A 12-bus test system is used to evaluate the benefits of DG by employing the proposed indices and approach. All per-unit quantities used in this study are on a 400-MVA base. This system consists of three conventional generators located at buses 1, 5, and 12 with ratings of 1.0, 0.75, and 0.625 p.u, respectively. A total load of 2.013 p.u located unevenly on every bus is assumed. Resistance and reactance of all the distribution lines are assumed to be 0.000625 and 0.00375 pu/km, respectively. Fig1 shows the single line diagram of 12 bus test system.

MATLAB. From equation (6) the value of BW VPI and BWLLR is kept 0.6 and 0.4 respectively. Fig2 shows voltage in p.u at each bus for base case, which is performed by using simulation in ETAP and by Classical algorithm.

Fig. 2: Voltage at each bus during base case run.

Fig3 BI at each bus by using Classical Algorithm. TABLE I RESULTS FROM CLASSICAL ALGORITHM FOR EACH BUS WITH VARIOUS DG SIZE. Bus 2 4 No. Size of DG w.r.t Load(%) 1 2 3 4 5 6 7 8 9 10 1.0307 1.0371 1.0428 1.0476 1.0516 1.0548 1.057 1.0583 1.0587 1.0581 1.0516 1.0532 1.0533 1.0519 1.049 1.0447 1.0391 1.0322 1.024 1.0148 6 8 10

Fig. 1: Single-line diagram of the system under study.

VII. SIMULATION, CASE STUDIES AND RESULTS In the case study the simulation is run in MATLAB by using Classical algorithm. By using coding in MATLAB the results are calculated in the form of VPII, LLRI and BI for each and every bus with respect to DG size. Maximum size of DG is up to 10% of the system load and weighing factor for each bus is kept equal that is 0.083. In this case study to test the Classical algorithm, first the size of the DG is set to 10% of the to the total system load then the test system is run on ETAP as well as in

1.0365 1.0441 1.0507 1.0562 1.0606 1.0638 1.0657 1.0664 1.0658 1.0639

1.0364 1.0397 1.0415 1.0419 1.0409 1.0383 1.0344 1.0292 1.0227 1.015

1.0342 1.04 1.0439 1.046 1.0462 1.0445 1.041 1.0356 1.0285 1.0198

In the second phase, the Classical Algorithm made for this system is run in MATLAB and the results are shown by Table I and Fig3. Table I gives the value of BI only of those buses whose BI is greater than one. Fig 4 shows the curve of the BI versus DG Size for various buses.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

64

REFERENCES
[1] CIGRE. Impact of increasing contribution of dispersed generation on the power system. Working Group 37.23, 1999. [2] CIGRE. CIGRE technical brochure on modeling new forms of generation and storage, November 2000. Available from: http:// microgrids.power.ece.ntua.gr/documents/CIRE-TF-380110.pdf. [3] M. R. Milligan and M. S. Graham An enumerated probabilistic simulation technique and case study: Integrating wind power into utility production cost models, in National Renewable Energy Lab. for Wind Energy Program, 1996. [4] T. Hoff and D. S. Shugar, The value of grid-support photo voltaics in reducing distribution system losses, IEEE Trans. Energy Conversion, vol. 10, pp. 569576, Sept. 1995. [5] R. Caire, N. Retiere, N. Martino, N. Andrieu, and N. Hadjsaid, Impact assessment of LV distributed generation on MV distribution network, in IEEE Power Engineering Society Summer Meeting, Paper no. 02SM152, July 2002. [6] R. Ramakumar and P. Chiradeja, Distributed generation and renewable energy systems, in Proc. 2002 Intersociety Energy Conversion Engineering Conf., p. IECEC-20 027-1-8. [7] R. E. Brown and L. A. A. Freeman Analyzing the reliability impact on distributed generation, in IEEE Power Engineering Society Summer Meeting, vol. 2, July 2001, pp. 10131018. [8] R. E. Brown, J. Pan, X. Feng, and K. Koutlev Siting distributed generation to defer T&D expansion, Transmission and Distribution Conf. and Expo., vol. 2, pp. 622627, Oct. 2001. [9] Griffin T, Tomosovic K, Secrest D, Law A. Placement of dispersed generations systems for reduced losses, in Proceedings of the 33rd Hawaii international conference on sciences, Hawaii, 2000. [10] Bala JL, Kuntz PA, Pebles MN. Optimum capacitor allocation using a distribution-analyzer-recorder, in IEEE Trans PWRD 1997;12(1): 4649. [11] Pathomthat Chiradeja, Pathomthat Chiradeja, R. Ramakumar, An Approach to Quantify the Technical Benefits of Distributed Generation, IEEE Trans ON ENERGY CONVERSION, VOL. 19, NO. 4, DECEMBER 2004. [12] N. Hadjsaid, J. F. Canard, and F. Dumas Dispersed generation impact on distribution networks, IEEE Comput. Applicat. Power, vol. 12, pp. 2228, Apr. 1999.

Fig. 4: BI versus DG Size for some buses

VIII. DISCUSSION OF RESULTS From the Table I, Bus No.2 having maximum BI of 1.0583 when it consists the DG size of 9% of the total load. For Bus No. 4 BI is maximum when it has the DG of size 3% of the total load of the system. Bus No. 6 shows the maximum BI of 1.0664 on DG size of 8% of the total load while Bus No. 8 have maximum BI of 1.0419 when the size of the DG is 4% of the total load of the system. Bus no. 10 have Maximum BI of 1.0462 when DG of 5% of the total load is connected. From all the cases the maximum value of BI is 1.0664 for Bus number 6 at DG size of 8% of the total load which is 61.61 MW with power factor of 0.98. VIII. CONCLUSION Size and location of DG are crucial factors in the application of DG for loss minimization. This paper presents an algorithm to calculate the optimum size of DG at various buses and proposes a fast methodology to identify the best location corresponding to the optimum size for reducing total power losses in primary distribution network. The benefit of the proposed algorithm for size calculation is that a look up table can be created with only one power flow calculation and the table can be used to restrict the size of DG at different buses, with the view of minimizing total losses. However, if a DG is installed in the system, the look up table needs to be updated with new calculation. The proposed methodology for location selection correctly identifies the best location for single DG placement in order to minimize the total power losses. In practice, the choice of the best site may not be always possible due to many constraints. However, the analysis here suggests that the losses arising from different placement varies greatly and hence this factor must be taken into consideration while determining appropriate location. The paper also shows that the loss sensitivity factor approach for location selection may not lead to the best choice.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

65

High Voltage Distribution System : A Loss Reduction Technique in Power Distribution


1

N K Sharma, 2P Tiwari , 3Apoorv Sachan, 4Chand Prakash


1,2,3,4

drnikhlesh@gmail.com, 2tiwari_p@kiet.edu, 3apoorvsachan@gmail.com 4 chandprakash68@gmail.com

KIET Ghaziabad- UP, INDIA-201206

AbstractThe loads in rural area are predominantly pump sets used for lift irrigation. The loads have low power factor and low load factor. Further, load density is low due to dispersal of loads. The existing distribution system consists of three-phase 11KV/433Volts distribution transformer with lengthy Low voltage Lines. In this system, the losses are high; voltage profile and reliability are unsatisfactory. In this paper, High Voltage Distribution System (HVDS) has been introduced with small capacity distribution transformers. A simple case study has been discussed in order to prove the technical superiority of HVDS over the existing Low Voltage distribution system (LVDS). An advantage of implementing HVDS over LVDS system is discussed. Keywords- Distribution Losses, HVDS, LVDS, Rural loads.

loads and loads other than agriculture. The length of the LV lines is restricted to less than 300meters. [1] II. DISADVANTAGES OF EXISTING LV DISTRIBUTION SYSTEM The Low Voltage Distribution System (LVDS), accounts for about one third [2] of the total transmission and distribution losses in India. The main contributing factor of this loss is the present practice of using 3 Distribution Transformer of considerable capacities, which leads to the use of very long LV lines. This is based on European practice. The LV line is extended to cater a group loads and useful particularly when catering loads of high load density. Presently, the LT lines are extended irrespective of voltage drops up to full capacity of distribution transformers and sometimes even above the transformer capacity ignoring the load of lines. This leads to severe voltage drops, high line losses accompanied with low power factor, chances of unauthorized connections, etc .The practical and feasible solution is to eliminate or minimize the LV lines by changing over to HT distribution system, where the HT line is drawn as near the load as possible and small capacity 3/1 distribution transformers are installed. This is best suited to meet the scattered loads. They can be generally divided into technical losses and non-technical losses. Technical losses are caused by current and resistance (IR), hysteresis, eddy currents and dielectric losses (corona). Non-technical losses are for example due to metering errors, unmetered company or customer use and billing cycle errors [3]. Some other major disadvantages experienced with existing LVDS can be summarised as follows: A. Rampant pilferage from accessible low voltage lines causing revenue loss Field survey suggested that on an average, there existed 2 cases of unauthorized abstraction of energy for every 3 authorized agricultural connections [4]. These unauthorized connections cause not only increased losses in the system but also are one of the major factors behind the high rate of failure of the distribution transformers. A part of the it is also caused by installation of the pumps and loads of the capacity higher than the sanctioned value. B. High technical losses attributable to high LT current on the network. The copper losses are proportional to the square of the current flow in the conductors. As the voltage become low

I. INTRODUCTION High Voltage Distribution System (HVDS) project is to reconFigthe existing Low voltage (LV) network as High Voltage Distribution System, wherein the 11 kV line is taken as near to the loads as possible and the LV power supply is fed by providing an appropriate capacity transformer and minimum length of LV line with an objective to provide better quality power supply, reduction of losses and better consumer service. In rural areas, irrigation pump sets (IP sets) forms the major load and these loads are spread over a large area resulting in lengthy LV lines. In the present system large capacity transformers are provided at one point and the connections to each load is extended through LV lines. This high length of LV lines is causing low voltage condition to the majority of the consumers and high technical losses. In the HVDS project, long length LV mains are converted into 11 kV mains and thereby installing the appropriate capacity distribution transformer, the supply is provided to end consumer at usable voltage level. By converting these lines to HVDS, the current flowing through the lines shall reduce by 28[1] times and will bring down the technical losses in the LV line drastically. The prevailing low voltage in the LV line is also affecting the efficiency of the pump sets, and the failure rates of the motors are also very high. Also there is a tendency of unauthorized IP set connections to hook to the LV lines which results in over loading of the transformers and failure of the transformers. The scheme consists of converting the existing 3phase 4 wires lines to 11 kV systems using the existing supports and providing intermediate poles wherever necessary and individual transformers are provided to both agricultural

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

66

for same power current becomes high and thus the losses are increase nonlinearly in the low voltage systems. Calculations based on field data indicated that for a group of load consisting of 3 authorized and 2 unauthorized pump connections, the technical loss accounted for 8.17% of the total energy intake for 3 authorized connections [4]. These losses may drift upwards for some typical systems with C. High peak power loss of network due to unauthorized load For a given situation of 3 authorized and 2 unauthorized pumps, the peak power loss was found 40 % of the total authorized power requirement [4]. D. Unsatisfactory voltage profile at consumer nstallations. The minimum voltage at the customer premises was recorded as 370 volts against the rated voltage of 430 volts [4]. E. Unreliable supply consequent upon overloading of LT lines LT faults per annum per 100-circuit km of lines supplying agricultural pumps were as high as 1500[4]. Against this backdrop, high voltage distribution system (HVDS) was conceived as a means to curb the menace of pilferage of energy, reduce peak power loss, improve voltage profile and enhance reliability of supply arrangement. III. WHY HVDS IS NEEDED Distribution losses can be brought down considerably by suitable HT distribution System. In India, many State Electricity Boards have conducted studies and proved that distribution losses can be brought down to levels prevailing in advanced countries by adopting this system. In the existing LT distribution system 100KVA and 63KVA distribution transformers are provided and lengthy LT lines are laid till the consumer end as the consumer equipment is of 400V. In the proposed HVDS the existing LT lines are converted as 11KV lines and a small capacity transformer suitable for consumer load is provided at the nearest support of the line and LT supply is extended through aerial bunched (AB) cable. [2] Ideally, losses in an electric system should be around 3 to 16%. In developed countries, it is not greater than 10%. However, in developing countries, the active power losses percentage is around 20%; for this reason, utilities in the electric sector are currently interested in reducing it in order to be more competitive, since the electricity prices in deregulated markets are related to the system losses. [5]. Some other advantages offered by the HVDS over the existing LVDS are summarised as follows: A. Improved voltage profile to every AP Consumer Presently due to lengthy LT line and voltage drop the voltage received at the consumer end drops down. With the provision of DT's at the consumer premises (Zero LT system) there will not be any voltage drop therefore due to

improved voltage profile, efficiency of the motor will also increase. B. Theft of energy will be practically zero The unauthorized consumers at present are steeling energy by hooking or tapping LT lines feeding to various authorized consumers. After conversion of all LT lines in to HT lines hooking/tapping or theft of energy shall not be possible. C. Failure of Distribution Transformers will almost be negligible In the present system one bigger transformer is feeding to various consumers and due to theft of energy load increases which causes damage to the distribution transformer. After conversion a dedicated transformer will be provided for each agriculture consumer thus chances of damage due to unauthorized increased load shall be eliminated. D. Drastic reduction in Line losses After conversion virtually all consumers 'shall be fed through high voltage distribution system viz voltage shall increase (from 400 volt to 11000 volt) and current shall decrease accordingly. Hence there shall be considerable reduction in Line Losses (IR) [9]. E. Chances of adding illegal motors are completely eliminated As explained above after conversion chances of running illegal motors shall be eliminated as no LT line shall be available to the un-authorized consumers. Due to reduction in KVA capacity and better voltage profile, the peak power losses shall also reduce considerably. Overall system efficiency will be improved. Due to increase in voltage profile and decreased losses system efficiency will improve. More, volume of water shall be pumped out with the same capacity of the motor due to improved voltage. Due to improved voltage profile, motors will pump more water and the life of the motor running the Tube well shall increase due to less drawl of current and overheating. In the existing system, number of consumers is being fed from the same transformer which leads to blowing of the LT /HT fuse interrupting the supply to all the consumers. Thus in the HVDS system, AP consumer shall get interruption free supply. IV. RURAL ELECTRIFICATION Rural electrification is an integral component of poverty alleviation and rural growth of a nation. In India, electricity has not played effective role in the socioeconomic growth of village. Government of India has ambitious target of providing electricity to all villages by 2008 and all rural households by 2012. Steps are initiated with Rural Electric Corporation, State Electricity Boards, and Reforms in Power sector. India has one of the fastest growing economics in the world and ranked sixth place in the worldwide consumer of energy. Being the seventh largest country in the world, six thousand villages inhabit 72.2 percent of its human resource (census 2001). About,

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

67

40 percent of the total energy is in rural areas. Domestic sector constitutes major energy demand and its consumption accounts for 60 percent of energy used [6]. One of the suggested and most efficiently achieved reforms in power distribution structure in rural India is the HVDS. V. NATURE OF RURAL AGRICULTURAL LOADS IN INDIA The agricultural sector consumes about 23 percent of Indias electricity and 90 percent of its ground water. Although a number of initiatives have demonstrated that more efficient irrigation pump sets could save 30 to 70 percent of the sectors electricity use, stakeholders have been unwilling to bear the costs and risks of replacing the countrys inefficient pump sets. To address this problem, the United States Agency for International Development India (USAID) has launched a pilot project to replace about 600 irrigation pump sets in Indias Karnataka State under Water Energy Nexus (WENEXA) project. This pilot demonstrated the first market-driven public-private partnership between Bangalore Electricity Supply Company (BESCOM) and an Energy Service Company (ESCo) for agricultural demand-side management. If implemented nationwide, India could reduce its electricity supply/ demand gap from 11 to 3 percent and its annual agricultural tariff subsidy by $2400 million a year, among other socio-economic benefits [7]. Rural loads are characterized by the following featuresLoads in rural India are predominantly pump sets used for lift irrigation. These loads have low power factor, low load factor. Load density is low due to dispersal of loads. Dispersed loads require long medium voltage lines. Unreliable supply of about 6-8 hours per day and phase imbalance Average load in the villages ranges from 5 KW to 25 KW per village. Poor load factor around 0.2(average demand / maximum demand). Low load factor due to dominant domestic consumptionin particular, lighting, agricultural demand with seasonal periodicity and absence of industrial demand. Rural grids are often weak and high peak loads and relatively large inductive loads can occur. As the number of irregular, decentralized, power generators increases, so will their impact on the dynamic behaviour of the power system [8]. VI. TECHNICAL SUPERIORITY OF HVDS A Case Study: Details of 4 cases where LVDS is changed to 11 KV HVDS and the results obtained are tabulated below: Studies are conducted by erecting 11 KV CTPT set and HT Tri-vector at Distribution Transformer (DTR) location (100 KVA) existing prior to conversion. The results obtained and a comparison between the parameters before and after conversion to HVDS is given below [13]:

A. Existing system: The existing system consists of four substations viz. S/S-I, S/S-II, S/S-III and S/S-IV. The parameters of these substations such as the length of LT lines, no. of irrigation pumps under each substation, the load connected under each substation in hp, approximate no. of working days, the input units to each substation and the output of each substation in electrical units are tabulated in TABLE I. From TABLE I the loss of units in the respective substation can be easily calculated and the percentage line losses for the four substations is calculated as 18.63, 13.76, 16.82 and 16.30 for S/S-I, S/S-II, S/S-III and S/SIV respectively. B. After conversion to HVDS (with small rating 3 phase transformers): The existing system of four substations, previously working as Low Tension system, is now converted into a High Tension or High Voltage Distribution System (HVDS) by installing 15 kVA distribution transformers in place of existing 100 kVA or 63 kVA DTR. The no. of working days, input and output to each substation in electrical units is tabulated in TABLE II. The corresponding losses in each of the substations and respective percentage losses 5.47, 5.44, 5.30 and 3.77 of S/S-I, S/S-II, S/S-III and S/S-IV can be calculated .Further the percentage losses in earlier LT distribution system and current HVDS can be compared and the net reduction in percentage losses can be deduced from TABLE II as 13.16, 8.32, 11.52 and 12.50 respectively for the four substations. C. Comparison between LV Systems with HVDS The voltage improvement and loss reduction obtained in all 4 cases are as follows: S/S-I In substation S/S-I the earlier conFiguration of LT line of length 3.6 km is reconstructed into HT line of length 2.6 km and LT line of length 1 km. One DTR of rating 100 kVA is replaced by 11 no. of DTs of rating 15 kVA as tabulated in TABLE III. The tail end voltages for both conFigurations are 350 V and 420 V respectively. The percentage losses for both cases can be easily compared from TABLE III and it can be deduced that the losses are significantly reduced in the HVDS conFiguration. S/S-II In substation S/S-II the earlier conFiguration of LT line of length 3.0 km is reconstructed into HT line of length 2.04 km and LT line of length 0.96 km. One DTR of rating 100 kVA is replaced by 10 no. of DTs of rating 15 kVA as tabulated in TABLE IV. The tail end voltages for both conFigurations are 385 V and 430 V respectively. The percentage losses for both cases can be easily compared from TABLE IV and it can

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

68

be deduced that the losses are significantly reduced in the HVDS conFiguration.

REFERENCES
[1] BESCOM Bangalore: http://bangalore.justdial.com/bescombangalore-electricity-supply-co-ltdho_gandhinagar_Bangalore_hrkpoPhlq.htm [2] Amaresh, K.; Sivanagaraju, S.; Sankar, V.; , "Minimization of Losses in Radial Distribution System by using HVDS," International Conference on Power Electronics, Drives and Energy Systems, 2006. PEDES '06, 12-15 Dec. 2006, pp.1-5. [3] kert, J.; Hable, M.; Schegner, P.; , "Energy loss estimation in distribution networks for planning purposes," IEEE Bucharest PowerTech, 2009, June 28 2009-July 2 2009, pp.1-6. [4] NPCL-HVDS.pdf: http://www.powermin.gov.in/distribution/apdrpbe stprac/NPCLHVDS.pdf [5] Caicedo, N.G. et.al., "Loss reduction in distribution networks using concurrent constraint programming," Probabilistic Methods Applied to Power Systems, 2004 International Conference on , 1616 Sept. 2004, pp.295-300. [6] Kamalapur, G.D.; Udaykumar, R.Y.; , "Electrification in rural areas of India and consideration of SHS," International Conference on Industrial and Information Systems (ICIIS), July 29-Aug. 1 2010, pp.596-601, [7] Modi, S.B.; Trivedi, U.C.; Goyal, R.K.; , "Challenges & solution in determination of agriculture consumption," IPEC, 2010 Conference Proceedings , vol., no., pp.671-676, 27-29 Oct. 2010 [8] Kamalapur, G.D.; Udaykumar, R.Y.; Karajgi, S.; , "People's Participation in Rural Electrification A Successful Case," First International Conference on Emerging Trends in Engineering and Technology, 2008. ICETET '08, 16-18 July 2008, pp.945-949. [9] ESR_including_Annexures: http://www.psebindia.org/docs/esr.pdf [10] HVDS in APSPDL: http://www.apspdl.in/jsp/index.jsp. [11] Village_electrification.pdf: http://www.cea.nic.in/god/dpd/village_elec tri fication.pdf [12] High Voltage Distribution System: http://powermin.nic.in/distribution/ hvds. htm [13] High Voltage Distribution System - A Case Study: http://www.docstoc.com/docs/22995603/High-Voltage- istributionSystem---A-Case-Study.

S/S-III In substation S/S-I the earlier conFiguration of LT line of length 1.59 km is reconstructed into HT line of length 1.59 km. One DTR of rating 63 kVA is replaced by 8 no. of DTs of rating 15 kVA as tabulated in TABLE V. The tail end voltages for both conFigurations are 340 V and 420 V respectively. The percentage losses for both cases can be easily compared from TABLE V and it can be deduced that the losses are significantly reduced in the HVDS conFiguration. S/S-IV In substation S/S-IV the earlier conFiguration of LT line of length 3.3 km is reconstructed into HT line of length 2.5 km and LT line of length 0.8 km. One DTR of rating 100 kVA is replaced by 9 no. of DTs of rating 15 kVA as tabulated in TABLE VI. The tail end voltages for both conFigurations are 320 V and 430 V respectively. The percentage losses for both cases can be easily compared from TABLE VI and it can be deduced that the losses are significantly reduced in the HVDS conFiguration. VII. CONCLUSION The HVDS is technically superior to conventional LV Distribution system with regard to quality of supply, i.e., better voltage profile, reduced losses and better reliability. However, the capital cost will be marginally higher at initial stage due to increase in number of transformers and total capacity. If overall cost and benefit of the system is taken into account the HVDS may be economical/beneficial compared to LVDS.
S No. 1. 2. 3. 4. 5. 6. 7. 8. Particulars LT Lines No. of pumps Connected load No. of days Input(unit) Output (Units) Loss of units % line loss S/S-I 3.6 km. 39 179.5 hp 13 4290 3490.4 799.60 18.63%

TABLE I S/S-II 3.0 km. 24 130 hp 14 3059 2638 421 13.76% TABLE- II S No. 1. 2. 3. 4. 5. Particulars 15 kVA DTs No. of days Input(units) Output(units) Loss(unit) S/S-I S/S-II 10 13 3926 3712.2 213.8 S/S-III 8 1299.0 1229.3 69.7 S/S-IV 9 334.0 321.4 12.6 S/S-III 1.59 km. 9 72.5 hp +2.25 kW 40 17672 14700 2972 16.82% S/S-IV 3.3km. 38 130hp 19 6152 5149 1003 16.3%

11
15 5310 5019.2 290.8

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

69

6. 7. 8.

% of losses on HVDS % of losses on earlier LT distribution system Net reduction in losses

5.47% 18.63% 13.16%

5.44% 13.76% 8.32% TABLE-III

5.30% 16.82% 11.52%

3.77% 16.3% 12.5%

S No. 1. 2. 3. 4.

Particulars HT Lines(km) LT Lines(km) No. of 3-Phase distribution transformer(DT) Voltage at tail end(V)

LT System 3.6 1x100 kVA 350

HVDS 2.6 1.0 11x15 kVA 420

5.

% Line Losses TABLE IV S No. 1. 2. 3. 4. 5. Particulars HT Lines(km) LT Lines(km) No. of DTs-3 phase Voltage at tail end(V) % line losses LT System 3.0 1x100 kVA 385 13.76%

18.63%

5.47%

HVDS 2.04 0.96 10x15 kVA 430 5.44%

TABLE V S No. 1. 2. 3. 4. 5. Particulars HT Lines(km) LT Lines(km) No. of DTs-3 phase Voltage at tail end(V) % line losses LT System 1.59 1x63 kVA 340 16.82% HVDS 1.59 8x15 kVA 420 5.30%

TABLE VI S No. 1. 2. 3. 4. 5. Particulars HT Lines(km) LT Lines(km) No. of DTs-3 phase Voltage at tail end(V) % line losses LT System 3.3 1x100 kVA 320 16.30% HVDS 2.5 0.8 9x15 kVA 430 3.77%

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

70

Optimal Conductor Size Selection and Capacitor Placement in Radial Distribution System
Damanjeet Kaur
UIET, Panjab University, Chandigarh
djkb14@rediffmail.com Abstract in this paper the optimal size of conductor and capacitor in radial distribution system is discussed. In this paper, to meet the increased load demand, the size of conductor and capacitor are selected in such a manner that the total cost occurring due to cost of conductor and capacitor and cost of losses are minimized while constraints of voltage and currents are satisfied. Th0e optimization problem is solved using a population based approach Particle Swarm Optimization is implemented which gives near optimal solution. The proposed approach is demonstrated with the help of 9-bus test data. KeywordsParticle Swarm Optimization, Radial Distribution System

I. INTRODUCTION Demand of electric energy is ever increasing due to changing living standards of citizens. To meet the increased load demand, there is only solution is expansion of generation, transmission and distribution system. In earlier years power utilities were focusing on expansion of generation and transmission systems but later on it is found that distribution system is also important. In distribution system, consumers are connected to source with the help of conductors named as feeders. These conductors are lengthy and stretched at long distance. As the demand for load increases the problem of voltage drop and poor power factor becomes more severe and it becomes important to expand the system. The poor power factor which causes high current flow through the feeders and results heavy power losses. But still the expansion of distribution system can be postponed by expanding the existing network. This is generally possible by reconductoring the conductors and placing the capacitors in the system. As in distribution system, feeders share a major portion and it is economically to select conductors for feeders according to their current carrying capacity which generally decreases from source end to far end. The other option is to install capacitors at distribution system. In literature numbers of papers are available for conductor size selection [6-9] and capacitor installation separately [1-5] but to solve the two problems collectively, few authors devised techniques [10]. Authors developed methods based on heuristic, dynamic programming, local variation method, mixed integer and integer quadratic programming and evolutionary algorithms are developed to find optimal solution of the problem [1-5]. For last decades, there are many evolutionary algorithms

based techniques such as Genetic Algorithms and particle swarm optimization [11-12], have been introduced. These are populations based approaches for optimization in which different steps are applied to find optimal solution. In these approaches, a fitness function is always defined for the problem which is to be optimized. In Genetic Algorithm, initially a set of candidate solutions is generated which evolves towards a better solution through selection, crossover and mutation operations.GA has enough randomness in search space but the parameter tuning is major problem. To overcome this, PSO can be a solution as in PSO number of parameters is very less and otherwise also tuning of parameters is not required. It searches automatically for the optimum solution in the search space, and the involved search process is random. The PSO algorithm has rapidly become popular and has been applied in neural network optimization [10], data clustering [11] [12], engineering design [13], etc. Keeping advantages of PSO in mind, in this paper, Particle Swarm Optimization (PSO) technique is implemented to find optimal size of conductors and capacitors in radial distribution system. The paper is organized in different sections. Section I introduces the problem of conductor and capacitor placement in radial distribution system. In Section II the optimization problem is formulated. Section III deals about the basic steps of PSO and section IV is about algorithm of the proposed approach. Finally section V and VI are about results and discussions along with the conclusions of the problem. II. PROBLEM FORMULATION In this section, the problem of optimal conductor size selection and capacitor placement is discussed. A. Objective Function The aim is to select optimal conductor for each feeder segment and to select site, size and number of capacitors on radial distribution system in such a manner that total cost consisting of cost of conductors and capacitors and cost of power losses power losses are minimized while voltage at all buses is within the limits. Mathematically it can be defined as [10]: Min Z Kcondr Kcap C p Ploss (1)

Kcondr
b nbr

alpha ccLb At

(2)

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

71

Kcap
j n
nbr

Kc Qc j
( I b2 Rb )
b 1

(3) (4)

Ploss

After calculation of fitness, at iteration u=1, Gbest is considered equal to Pbest. Compare the fitness value of the particle p at u+1 with that of the previous best one. E. Updation of Velocity and Position Vector Update velocity and position vector of particle p using Gbest and pbest till iteration u+1 using following equations. Vel u 1 int( wVel u c1rand1 ( pbest Pos u ) p p p (7) u c2 rand 2 ( gbest Pos p )
pos u p
1

Where Kcondr, Kcap, are cost of conductor and capacitor in Rs of distribution system; At is cross-section area of t type of conductor in mm2; alpha Carrying charge rate of feeder; Constant associated with the variable installation cost of feeder in Rs./ mm2/km; Lb is the length of feeder segment b; Kc is the annual cost of capacitor in Rs/kVar; Qjc is the shunt capacitor size at bus j; n is the number of buses on which capacitors are placed; nbr is the total number of branches in the feeder; Ib and Rb are current and resistance of branch b respectively . B. Constraint (i) Voltage at all buses of the network should be within the limits as: V min V j V max (5) (ii) The current flowing through each feeder segment b for t type conductor should be less than the maximum current carrying capacity (Cap) of the t type of conductor. (6) I (b, t ) Cap(t ) III. PARTICLE SWARM OPTIMIZATION Particle Swarm Optimization is a population based evolutionary technique. It was introduced as an alternative to Genetic Algorithms. PSO evolved as a successful tool for optimization because of its simplicity and flexibility to control the balance between the global and local exploration of the search space. PSO has wide applications in continuous and discrete variables optimization. These days PSO has been extended to multiobjective variables also. The various steps involved in PSO algorithm is as given below [11-13]: C. Generation of Population Initially a population of individuals is generated. In PSO, individuals are termed as particle and group of population is known as swarm. Each particle p at kth k iteration has velocity ( Veluk ) and position ( posu ) within the search space. B. Fitness Function To optimize the problem, fitness function is defined corresponding to the objective function. For each particle p, fitness function is evaluated. C. Find Local Best and Global Best After calculation of fitness function, the best fitness function for each particle and among swarm is found out. The best fitness function of particle p is known as Pbest and that of swarm is defined as Gbest. D. At First Iteration

pos u Vel u p p

(8)

where int is the operator of rounding the variable to the nearest integer; rand1 and rand2 are random numbers generated in [0, 1]; c1 an c2 are acceleration constant. Equation (7) consists of three terms on RHS; first term is the velocity in uth iteration; second term is the cognitiononly model and the third term is the so-called social-only model, these terms are utilized to change the velocity of particle. Where w is the inertia weight factor, it provides balance between global and local explorations. w often decreases from 0.9 to 0.4 during the iterations. It is generally set using the following equation: (9) w wmax ((wmax wmin ) / kmax )* k where kmax is the maximum number of iterations and k is the current number of iteration. F. Stopping Criteria The above steps of subsection C to E is repeated till the search satisfies the termination condition. The termination condition may be maximum number of iterations or the convergence criteria set. IV. ALGORITHM OF THE PROPOSED APPROACH Generate population of particles having position and velocity vector within the given sizes of conductors and capacitors. (ii) Evaluate fitness function for the given objective function. (iii) Find Pbest and Gbest as given in subsection C. (iv) Update velocity and position vector as given in Eq. (7) and (8) respectively. (v) Repeat above steps until maximum number of iterations is achieved. (i) V. RESULTS AND DISCUSSIONS The proposed approach is tested on13.8 kV, 9-bus test data as shown in Fig 1 [9]. The busdata and conductors available in the inventory are as given in Table I and Table II respectively. The step size of capacitor is 300 kVar. The other data is as: Cost of power loss= Rs 168/kW/year Vmin=0.9; Vmax=1.0; Kc= Rs 1564/kVAr

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

72

Cc=Rs 500/mm2 alpha=0.1

TABLE III: OPTIMAL SIZES OF CONDUCTORS Segment Number L01 L02 L03 L04 L05 L06 L07 Conductor Type 5 5 5 4 5 5 3

Fig. 1: 9-bus test system TABLE I: DEMAND AT NODES IN KW

TABLE IV: CAPACITORS SELECTED FOR BUSES

Node N01 N01 N03 N04 N05 N06 N07 N08

Demand(kW) 0.0 1054.2 806.5 2632.5 609 2034.5 932.8 1731.4

Bus Number Capacitor Size (kVAR)

2 300

3 300

7 600

8 300

VI. CONCLUSIONS In this paper, the problem of conductor size selection and capacitor site and size is solved using a population based approach. In this paper, the total cost of capacitors and conductors occurring due to installation and power losses are minimized while constraints of voltage and current carrying capacity of each feeder segment are satisfied. The proposed approach is easy to implement and gives near optimal solution. REFERENCES
Fixed Cost (Rs/km) 2000 3000 4000 5000 6000 8000
H. Duran, Optimum Number, Location, and Size of Shunt Capacitors in Radial Distribution Feeders, A dynamic Programming Approach, IEEE Transactions on Power Apparatus and Systems, vol. 87, no. 9, pp. 17691774, September 1968. [2] M. Ponnavaikko and K. S. Rao, Optimal Choice of Fixed and Switched Shunt Capacitors on Radial Distributions by the Method of Local Variations, IEEE Transactions on PAS, vol.102, no 6, pp.1607-1615, June 1983. [3] M.E. Baran and F. F. Wu, Optimal Capacitors Placement on Radial Distribution Systems, IEEE Transactions on Power Delivery, vol. 4, no. 1, pp.725-734, Jan 1989. [4] M.E. Baran and F.F.Wu, Optimal Sizing of Capacitors Placed on Radial Distribution Systems, IEEE Transactions on Power Delivery, vol. 4, no. 1, pp.735-743, Jan 1989. [5] J. C. Carlisle, A. El-Keib, D. Boyd and K. Nolan, A Review of Capacitor Placement Techniques on Distribution Feeders, Proceedings of the 29th Southeastern Symposium on System Theory (SSST '97), pp. 359 365, 1997. [6] M. Ponnavaikko and K.S. Rao, An Approach to Optimal Distribution System Planning Through Conductor Gradation, IEEE Transactions on Power Apparatus and Systems, vol. 101, no. 6, pp. 1735-39, 1982. [7] Z. Wang, H. Liu, D.C.Yu , X. Wang and H. Song, A Practical Approach to The Conductor Size Selection in Planning Radial Distribution System, IEEE Transactions on Power Delivery, vol. 15, no. 1, pp , 350-354, 2000. [8] D. Kaur and Jaydev Sharma, Optimal Feeder Design in Planning Radial Distribution Systems, International Journal of Electrical Power and Energy Systems, vol 30, no. 4, pp. 261-271, 2008 [9] F. Mendoza, D. Requena, D. Navarro , L. Jose, Bemal-Agustin and A.Jose, Optimal Conductor Size Selection in Radial Power Distribution Systems Using Evolutionary Strategies, IEEE, pp1-6, 2006. [1]

TABLE II: CONDUCTORS AVAILABLE IN INVENTORY

Conduct or Type 1 2 3 4 5 6

Resistance (ohms) 0.8763 0.6960 0.5518 0.4387 0.3480 0.2765

Reactance (ohms) 0.4133 0.4133 0.4077 0.3983 0.3899 0.3610

Current (A) 180 200 230 270 300 340

Initially a population is generated as explained in Section III. After maximum number of iterations, the optimal conductor sizes options and capacitors site and sizes obtained are tabulated as shown in Table III and IV respectively. As it is clear from Table IV that reactive compensation takes place maximum at bus 7 while for other buses it is 300 kVar. The total cost involved is Rs 16560969. Hence it can be concluded that by considering reconductoring of conductors and capacitor placement power demand can be met while constraints are satisfied.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

73

[10] M. Vahid, N. Manouchehr, S. Hossein and A. Jamaleddin, Combination of Optimal Conductor Selection and Capacitor Placement in Radial Distribution Systems for Maximum Loss Reduction, 2008. [11] J. Kennedy and R. Eberhart, Particle Swarm Optimization, IEEE Proceedings on 1st International Conference on Neural Networks (Perth, Australia), pp. 1942-1948, 1995. [12] J. Kennedy and R. Eberhart, A Discrete Binary Version of the Particle Swarm Algorithm IEEE Conference on Systems, Man, and Cybernetics, Orlando, FA, pp. 4104-4109, 1997. [13] B Al-Kazemi and C. K Mohan, Multi-Phase Discrete Particle Swarm Optimization Fourth International Workshop on Frontiers in Evolutionary Algorithms, 2000.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

74

Transformer Fault Diagnosis Using Classical Methods


1

Prateek Kumar Singhal, 2Rakesh Kumar, 3R. Naresh


1,2, 3
1

NIT Hamirpur (H.P)

pks.mtech@gmail.com 2 rkyadav39@gmail.com Abstract The Power transformer is vital equipment in any electrical system, so any fault in the power transformer may lead to the interruption of the power supply and accordingly the financial losses will be high. So it is of paramount importance to detect the incipient fault of transformer as early as possible. Among the existing method for identifying the incipient fault like overheating or discharge fault inside the transformer, dissolved gas analysis (DGA) is the most popular and successful method. In this paper classical methods of DGA such as Key gas method, Rogerss ratio method and Duvals triangle method have been studied and applied on the test data of transformer as per the standard for determining the faults in transformer and hence improves the performance of transformer. Keywords Dissolved Gas Transformer, Transformer Oil. Analysis, faults, gases,

which results in generation of gases which are often combustible or harmful. The organization of paper is as follows: Section II represents the Incipient faults and fault gases. Section III represents the Methods of Incipient fault diagnosis. Section IV represents the Dissolved Gas Analysis (DGA) of power transformer. Section V represents the Condition monitoring of transformer which describes various methods for diagnosis. Section VI presents the Simulation results of IEEE standard test data and finally Section VII concludes the study and analysis of diagnosis of power transformer. II. INCIPIENT FAULTS AND FAULT GASES The operating principle of transformer is based on the slight albeit harmless deterioration of the insulation that accompanies incipient faults, in the form of arcs or sparks resulting from dielectric breakdown of weak or overstressed parts of the insulation, or hot spots due to abnormally high current densities or due to high temperature in conductors [3]. Whatever the cause, these stresses will result in the chemical breakdown of oil or cellulose molecules constituting the dielectric insulation. In condition of thermal and electrical stress condition caused by fault current in the transformers, the hydrocarbons molecules of mineral oil can decompose and form active hydrogen and hydrocarbon fragment can combine with each other to form gases like hydrogen ,methane, acetylene ,ethylene, ethane, carbon monoxide, carbon dioxide etc. So gases which are produced in the transformer are classified as follows [4]: A. Hydrocarbon and Hydrogen Hydrogen ( ) Methane ( ) Ethylene ( Acetylene ( ) Ethane ( ) B. Carbon Oxides Carbon Monoxide (CO) Carbon Dioxide ( C. Non fault gases Nitrogen ( Oxygen ( )

I. INTRODUCTION Transformers are the essential part of the electrical power system because it has the ability to change voltage and current levels, which enables the transformers to generate the electric power, to transmit and distribute electric power and utilize the power at economical and suitable levels. In electrical power system, voltage of electricity generated at the power plant will be increased to a higher level with step-up transformers [1]. A higher voltage will reduce the energy lost during the transmission process of the electricity. After electricity has been transmitted to various end points of the power grid, voltage of the electricity will be reduced to a usable level with step-down transformer for industrial customers and residential customers. Since power transformer is vital equipment in any electrical power system, so any fault in the power transformer may lead to the interruption of the power supply and accordingly the financial losses will also increase. So it is of paramount importance to detect the incipient fault of transformer as early as possible. To monitor the service ability of power transformer many devices have been evolved such as buchholz relay, differential relay, over current relay, thermal relay etc which is a part of protection in terms of determination of faults in the transformer [2]. Even in normal operation, a power transformer is subjected to internal stresses that often in time affect the performance and reliability of the transformer through the steady breakdown of its insulating materials. These material include paper and oil and after being subjected to a various of stressful condition that occur in a transformer they have been found to deteriorate

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

75

The gases produced will be evolved in oil and in the headspace as a result of various faults. The apparent difference between various types of faults is the energy dissipated per unit volume during a particular fault. The various faults can be as follows:Corona: a) Oil b) Cellulose - , CO, Corona is a low energy electrical fault. The major byproduct that contributes towards corona is hydrogen. Pyrolysis or thermal heat: a) Oil Low temperature , High temperature - , ( , ,( ) b) Cellulose Low temperature - (CO), ( High temperature - CO, ( ) Ethylene and methane are the major decomposition gases besides smaller quantities of hydrogen and ethane. The intensity of energy dissipation is more than that of corona. Arcing - , ( , ) Due to arcing, large amount of hydrogen and acetylene are produced. Methane and ethylene are the minor quantities decomposition. The intensity of dissipation is highest in arcing followed by heating and corona. It was also noticed that the amount of each individual gas is dependent on the temperature in the vicinity of the stressed point [5]. In some of the studies it has been reported that the gases are generated with an increase of temperature. Hydrogen is generated at low temperature and its amount gradually increases, while acetylene is generated at a very high temperature and also steadily increases its amount with temperature. Solid insulation degradation of transformer is conventionally diagnosed on ratio value of dissolved gases carbon monoxide (CO) and carbon dioxide (C III. METHODS OF INCIPIENT FAULT DIAGNOSIS The major concern of power transformer incipient fault is that they may decrease the electrical and mechanical integrity of the insulation system. This may progress to a point that the insulation cannot withstand transient overstress caused by fault current and electrical over voltages. Incipient fault diagnosis is therefore closely related to insulated condition assessment [6]. A. Insulation Condition Assessment Test Insulation condition assessment test refers mainly to offline routine tests, including measurement of insulation resistance(IR) , dielectric loss factor(DLF) , Interfacial polarization (IP), using anomalous IR and frequency dispersion of capacitance , turn ratio (TR) , winding resistance(WR), core ground resistance(CGR), and some excitation tests. These tests can reveal some severe problems but may not find the incipient ones [7].

These tests include measurement of the degree of polymerization (DP) and tensile strength (TS). Some relatively new methods , such as high performance liquid chromatography(HPLC) furan analysis , Interfacial polarization spectra(IPS) using return voltage(RV) measurement , and analytical chemical techniques, are also used for the same purpose. These intrusive techniques are usually not favorable because the sampling process may damage the integrity of the insulation system, but may be for very old transformers. B. Dissolved gas in Oil Analysis A more successful technique for on-line incipient fault diagnosis is dissolved gas analysis (DGA). By on-line we mean the transformer does not need to be de-energized. This type of analysis includes the conventional DGA, which is based on routine oil sampling and the modern technology of on-line gas monitors [7]. Conventional dissolved gas analysis (DGA) has been an industry standard for the detection and determination of faults in transformers for over 30 year. Developed in the late 1960s, DGA has been recognized worldwide as the main tool to prevent catastrophic failures of power transformers and has gained tremendous success compared to other techniques. The main reason for this success is that sampling and analyzing procedures are simple and inexpensive and are easy to be standardized. Many experiences have been gained from the process and several DGA standard have been set up IEC 599 (1978), IEC 60599(1999) and ANSI/IEEE Std C57.1041991(1992). Major diagnostic gases have been identified as Hydrogen, ethane, methane acetylene, ethylene, carbon dioxide and carbon monoxide [8]. Insulating material within transformers and related equipment break down to liberate gases within the unit. The distribution of these gases can be related to the type of electrical fault and the rate of gas generation can indicate the severity of the fault. The identity of the gases being generated by a particular unit can be very useful information in any preventive maintenance program. This forms the basis of dissolved gas analysis and this technique is being used quite successfully throughout the world. Since transformers of different size, Structure, manufacture, loading and maintenance history may have different gassing characteristic, they need to be considered differently in most cases. Thus, DGA is often referred as art instead of science [9]. IV. DISSOLVED GAS ANALYSIS (DGA) A. Introduction Thermal and electrical distributions in the operating transformer are two most important causes of dissolved gases in oil. The gases produced from thermal decomposition of oil and solid insulation are because of losses in conductors due to loading. Also decomposition occurs in oil and solid insulation is due to occurrence of

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

76

arc. In case of electrical disturbances the gases are formed principally by ionic bombardment. The gases are generated mainly because of cellulose and oil insulation deterioration [10]. B. Operating Procedure for DGA The following major steps are taken to analysis the DGA:Detection: - The generation of any gas above the normal level is detected first and then appropriate guidelines are utilized so that possible abnormally can be recognized at the earliest so that damage to the transformer could be minimized. Evaluation: - The impact of the abnormality or fault on the serviceability is evaluated by using a specified set of guidelines and recommendation. Action: - Lastly the action is recommended, which begins with increased surveillance and confirming or supplementary analysis and leads to either a determination of load sensitivity, reducing the load on the transformer or actually removing the unit from service. The following methods are used for detecting fault gases:Direct measurement of the amount of combustible gas in the gas space or relay(Total Combustible Gas(TCG)) Direct measurement of the amount of combustible gas dissolved in the oil (Gas in oil monitors) Chromatographic separation and analysis for the individual components in a gas mixture extracted from a sample of the transformer oil or a sample of the transformer gas space. C. Faults All Transformers generate gases to some extent at normal operating temperatures. But occasionally a gas generating does occur within an operating transformer such as a local overheating, dielectric problems, or a combination of these. In electrical equipment, these abnormalities are called faults and faults may be thermal, corona and arcing. Internal faults in oil produce the gaseous byproducts such as Hydrogen ( ), Methane ( ), Ethylene ( , Acetylene ( ), Ethane ( ). When cellulose is involved, the faults produce Hydrogen ( ), Methane ( ), Carbon Monoxide (CO), Carbon Dioxide ( . Each of these types of faults produces certain gases that are generally combustible. The total of all combustible gases may indicate the existence of anyone or a combination of thermal, electrical or corona faults. From the gases dissolved in the transformer oil following types of faults are interpreted [11]. Thermal Faults: The decomposition of oil is under the temperature ranging from 150 to 500 which produces relatively large quantities of low molecular weight gases, such as hydrogen and methane and trace quantities of the higher molecular weight gases Ethylene ( , Ethane ( ). As the fault temperature in mineral oil increases to modest

temperature, the hydrogen concentration exceed that of methane, but now the temperature are accompanied by significant quantities of higher molecular weight of gases , first ethane and then ethylene . At the upper end of the thermal fault range, increasing quantities of hydrogen and ethylene and traces of acetylene may be produced. In contrast with the thermal decomposition of oil, the thermal decomposition of cellulose and other solid insulation produces Carbon Monoxide (CO), Carbon Dioxide ( . And water vapor at temperature much lower than that for decomposition of oil and at rates exponentially proportional to the temperature. Since the paper begins to degrade at normal operating temperature in the transformer. The ratio of / CO is sometimes used as an indicator for the thermal decomposition of cellulose. This ratio is normally more than seven. As the magnitude of CO increases, the ratio / CO decrease. This may indicate an abnormality that is degrading cellulosic insulation [12]. Electrical Faults Low Intensity Discharges: Low intensity discharges such as partial discharge and very low level intermittent arcing produce mainly hydrogen, with decreasing quantities of methane and trace quantities of acetylene. As the intensity of the discharge increases, the acetylene and ethylene concentrations rise significantly. High Intensity Discharges: It is also referred as arcing. As the intensity of electrical discharge reaches arcing or continuing discharge proportion that produce temperature from , the quantity of acetylene becomes pronounced [13]. V. CONDITION MONITORING OF TRANSFORMER

A. Key Gas Method The dependence on temperature of the types of oil and cellulose decomposition gases provides the basis for the qualitative determination of fault types from the gases that are typical, or predominant, at various temperatures. These significant gases and proportions are called key gases. The principle of the Key Gas method is based on the quantity of fault gases released from the insulating oil when a fault occurs which in turn increase the temperature in the power transformer [14]-[16]. The presence of the fault gases depends on the temperature or energy that will break the link or relation of the insulating oil chemical structure. This method uses the individual gas rather than the calculation of gas ratios for detecting fault. The significant and proportion of the gases are called key gases. It gives basically four types of faults:Thermal Oil:Decomposition products include ethylene and methane, together with smaller quantities of hydrogen and ethane.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

77

Traces of acetylene may be formed due to the fault is severe and involve electrical contacts.

1 2 3 4 5

<0.1 0.1 to 3.0 <0.1 <0.1 <0.1

<0.1 0.1 to 1.0 >0.1 to <1.0 >1.0 >1.0

<1.0 >3.0 1.0 to 3.0 1.0 to 3.0 >3.0

PD ArcingHED Low temperatur e thermal Thermal <700 Thermal >700

Fig. 1: Analysis of different faults using Key Gas Method

Thermal Cellulose:Large quantities of carbon dioxide and carbon monoxide are evolved from overheated cellulose. Hydrocarbon gases such as methane; ethylene will be formed if the fault involves oil impregnated structure. In Fig1 92% carbon monoxide is shown. Electrical corona:Low energy electrical discharge produces hydrogen and methane with small quantities of ethane and ethylene. Comparable amounts of carbon monoxide and dioxide may result from discharges from cellulose. Electric Arcing:Large amount of hydrogen and acetylene are produced with minor quantities of methane and ethylene. Carbon dioxide and carbon monoxide may also be formed if the fault involves cellulose. B. Rogers Ratio Method The Rogers Ratio method use only three ratios R1, R2 and R5. This method is based on the thermal degradation principles. The validity of this method is based on correlation of the results of a much large number of failure investigation with the gas analysis for each case, but Rogers ratio method can give the ratio that do not fit into the codes. The value for the three key gas ratio corresponding are given in the table 1. These ratios according to Rogers are applicable to both gas taken from the gas space and gases extracted from the oil. The fault type has been chosen by combining some cases from the number of fault types originally suggested by Rogers. The table 1 shows the Rogerss ratio [14]-[16].
TABLE 1: DIFFERENT FAULTS IN TRANSFORMER BASED ON ROGERS RATIO

Fig. 2: Flowchart for Performing Rogers Ratio Method

C. Duvals Triangle Method M. Duval developed this method in the 1960s, to determine whether a problem exists at least one of the hydrocarbon gases or hydrogen must be at L1 level or above and the gas generation rate is at least at G2 [14][16]. The L1 level and the gas generation rate for this method are shown in table2 where G1 and G2 are two transformer ppm levels.
TABLE 2 : L1 LEVEL AND GAS GENERATION RATE FOR DIFFERENT GASES

Gas

L1 Limits 100 75 3 75 75 700 7000

CO

G1 Limits (ppm per month) 10 8 3 8 8 70 700

G2 limits (ppm per month) 50 38 3 38 38 350 3500

Case

R2 (C2H2/C 2H4) <0.1

R1 (CH4 /H2) >0.1 to <1.0

R5 (C2H4/C 2H6) <1.0

Suggested fault diagnosis Unit Normal

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

78

TABLE 4 : RESULTS FOR STANDARD DATA OF IEC 60599 AND IEC TC 10

Sample No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

Actual Fault PD PD LED LED LED LED LED HED HED HED HED HED LTF LTF LTF LTF HTF HTF HTF

Fig. 3: Duvals Triangle showing different faults

PD Partial Discharge T1 Low range thermal fault (below 300 ) T2 - Medium range thermal fault (300 - 700 ) T3 - High range thermal fault (above 700 ) D1 Low energy Electrical discharge D2 High energy electrical discharge DT Indeterminate thermal fault or electrical discharge The following points should be firstly considered while using Duvals Triangle for faults diagnosis. 1. The Duval triangle method like any other DGA diagnostic method should be applied only when there is some suspicion of a fault, based on an increase in combustible gas or some other suspicious symptom. The diagnostic method itself is not a mean of fault detection. 2. Because of the relative inaccuracy of gas in oil concentration measurement at low concentration, DGA diagnostic method including the Duval triangle should not be applied unless the gas concentration is well above the detection limits. If reasonably stable concentration of the gases present before the non set of the suspected fault, it is advisable to subtract out the background concentration, provided that the differences are large enough for interpretation [14][16]. The diagnosis should be based on recently formed gas if possible and including pre-fault gas in the diagnostic calculation can lead to misleading results [17]. VI. SIMULATION RESULTS Whenever there is any fault in the transformer there is degradation of oil and cellulose material. Due to degradation, gases are produced and these gases get evolved in the transformer oil. The transformer fault diagnosis is based on the concentration level of these gases. The main fault types that causes oil and cellulose degradation are partial discharge, low energy discharge, high energy discharge, low thermal fault and high thermal faults. For determination of these faults Key Gas Method, Rogers Ratio Method and Duvals Triangle Method are used and the test results are compared with IEEE standard results.

Key Gas Method EC EC ND ND AR ND CB ND AR ND AR ND ND LTF CB HTF HTF CB HTF

Rogers Ratio Method PD ND ND LED LED LTF LED HED HED HED HED HED ND LTF LTF HTF ND HTF ND

Duvals Triangle PD PD HED LED HED LED HED HED HED HED HED HED LTF PD LTF HTF HTF HTF HTF

TABLE 5 : TEST DATA ON POWER TRANSFORMER IN ANU SUBSTATION, HAMIRPUR, H.P, INDIA.

Sample No. 1 2 3 4 5 6 7 8

C 2091 40 1800 110 2700 1 280 105 139 5 280 20 600 20 1200 140 0 402 7700 30 705 2 5 3 2 20 907 4 690 5 800 220 18 3 30 1 50 45 225 30

CO 536 180 650 500 510 50 208 185

C 3000 2200 700 1500 1410 250 816 1300

TABLE 6 : RESULTS FOR TEST DATA GIVEN IN TABLE 5

Sample No.

Key Gas Method

1 EC, CB 2 ND 3 EC,AR,CB 4 CB 5 LTF 6 ND 7 LTF 8 ND Where, AR- Arcing CB- cellulose breakdown

Rogers Ratio Method LED HTF HTF HTF HED ND HTF HTF

Duvals Triangle PD LED LED LED HED LTF LTF HTF

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

79

EC- electrical Corona HED- high electric discharge LED- low electric discharge HTF- high thermal fault LTF- low thermal fault ND- No diagnosis PD- partial discharges DGA is done through Gas chromatography to find out the ppm level when oil is injected into it as a sample. The DGA procedure starts with extracting the gases from the oil sample. After separation, the gases will be quantified and measured. DGA is the method of gas extraction from the oil. This is also the part where it is prone to make errors. Traditionally, extraction involved some sort of high vacuum power to suck the gas from the oil. The gas is then collected and measured by using mercury pistons. Another method involves stripping the gases from the oil. This is called direct injection technique. Today, a new method called headspace method is being used by engineers mainly because it eliminates errors through automation. VII. CONCLUSION In this paper, the three classical techniques are being used for interpretation of incipient faults in power transformers by dissolved gas analysis (DGA) on standard data and the test data on Anu substation, Hamirpur, H.P, India. Out of the three Duvals triangle method is being found suitable which determines the faults related to actual faults in the power transformer. Further application of this technique as a monitor during the factory proving tests of power transformers is being developed employing very much greater detection sensitivities than used in the field. This shows great promise in that a number of faults likely to cause future trouble in service have been detected, although in each case the transformer had satisfactorily passed the routine tests. The technology presently exists and is being used to detect and determine fault gases below the part per million levels. However there is still much room for improvement in the technique, especially in developing the methods of interpreting the results and correlating them with incipient faults. REFERENCES
[1] B. Kasztenny, and M. Kezunovic, Digital Relays Improve Protection of Large Transformers, IEEE Computer Applications in Power, October 1998, pp: 39-45. P. Lui, and O.P. Malik, Improved Operation of Differential Protection of Power Transformers for Internal Faults, IEEE Trans. on Power Delivery, 7(4): 1912-1919, 1992. IEEE Guide for the Interpretation of Gases Generated in OilImmersed Transformers, IEEE Std. C 57. 104-1991. ANSI/IEEE Std C57.104-1991, "IEEE guide for the interpretation of gases generated in oil-immersed transformers," IEEE Power Engineering Society, 1992. R.R. Rogers, "IEEE and IEC codes to interpret incipient faults in transformers, using gas in oil analysis," IEEE Transaction on Er, Vol. 13, No. 5, pp. 348-54,1978.

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

IEC Publication 599, Interpretation of the analysis of gases in transformers and other oil-filled electrical equipment in service, First Edition, 1978. M. Duval and J. Dukarm, Improving the Reliability of Transformer Gas-in-Oil Diagnosis, IEEE Electrical Insulation Magazine, Vol. 21, No. 4, pp. 21-27, 2005. Interpretation of the analysis of gases in transformers and other oil filled electrical equipment in service, Book, IEC Publication 599, 1st ed., 1978.s E. Dornenburg, and W. Strittmater, Monitoring oil cooling transformers by gas analysis, Brown Boveri Review, Vol. 61, pp. 238274, 1974. M. Duval, "Dissolved gas analysis: It can save your transformer," IEEE Electrical Insulation Magazine, Vol. 5, No. 6, pp. 22-27, 1989. J.J. Kelly, "Transformer fault diagnosis by dissolved-gas analysis," IEEE Transaction On Artificial Intelligence, Vol. 16, No. 6, pp. 777-782, 1980. V.G. Arakelian, Effective diagnostics for oil filled equipment, IEEE Electrical Insulation Magazine, Vol. 18, No. 6, pp. 26-38, 2002. M. Duval, Dissolved Gas Analysis: A Powerful Maintenance tool for Transformers, Proceedings of Tech. Conference- 2004, North America, 2004. M. Wang, A.J. Vandermaar, and K.D. Srivastava, Review of Condition Assessment of Power Transformers in Service, IEEE Electrical Insulation Magazine, 2002, pp. 12-25. M. H. Wang, A Novel Extension Method for Transformer Fault Diagnosis, IEEE Transactions on Power Delivery pp. 164-169, 2003. T.K. Saha, Review of Modern Diagnostic Techniques for Assessing Insulation Condition in Aged Transformers, IEEE Transactions on Dielectrics and Electrical Insulation 10(5): pp. 903917, 2003. M. Duval, Interpretation of gas-in-oil Analysis Using New IEC Publication 60599 and IEC TC 10 Databases, IEEE Electrical Insulation Magazine, Vol. 17,No. 2, pp. 31-41,2001.

[2]

[3] [4]

[5]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

80

TABLE 3 : STANDARD DATA OF IEC 60599 AND IEC TC 10 [17] Sample No.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

37800 36036 8 305 1790 1330 57 110 3700 245 4419 10000 14 330 48 107 12705 1 3910

1740 4704 0 100 580 10 24 62 1690 120 3564 6730 14 619 610 143 23498 8 4290

8 10 101 541 619 182 30 250 3270 167 2025 10400 1 0 0 2 5188 6 1230

8 5 43 161 336 66 27 140 2810 131 2861 7330 7 2 10 222 34257 100 6040

249 554 0 33 321 20 2 90 128 18 668 345 124 58 29 34 6047 8 626

56 6 192 440 956 231 540 680 22 829 909 1980 128 51 1900 193 4004 300 1800

197 347 4067 3700 4250 1820 2518 6470 86 4250 9062 3830 2746 1 970 1330 8539 5130 11500

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

81

pH Analysis of Insulator Contaminants and Reliability Improvement of T&D Line under Chemically Polluted Environment
1

Md. Farman, 2Mohit Bansal, 3Ekram Husain, 4M.P.Sharma


Indian Institute of Technology, Roorkee, Uttarakhand, India 3 Aligarh Muslim University, Aligarh, U.P., India
1

1, 2, 4

md.farman.ee.1729@gmail.com 2 moh.bansal@gmail.com

Abstract-This paper presents a practical issue of blackout & brownouts which normally occurs in T&D lines under chemically polluted environment and adversely effects power quality of line. Different types and sources of chemical pollutants under different atmospheric conditions are discussed. Basics of flashover mechanism along with the chemical factors which regulates the magnitude of leakage current (flashover voltage) and methods of measurement of degree of contamination are discussed. Chemical analysis of various salts (soluble as well as insoluble) and ions on the flashover voltage under dry and wet condition is done to see the effectiveness of salts & ions on performance of line. Based on Hall Effect and magnetic fields, a new approach for improving grid reliability and maintenance work for T&D lines is proposed. Effect of pH on the surface conductivity is also analyzed. It is found that by increasing contamination level there is increase in surface conductivity of the insulator and reduces flashover voltage. It is also concluded that as basic or acidic nature of salt increases, in both cases flashover voltage reduces drastically. Finally methods of reducing insulator contamination level within limits are given. Keywords- Blackout, ESDD, flashover voltage, LCM, pH, power quality, soluble and insoluble metal salts and its ions.

I. INTRODUCTION In recent years, the demand of electric power has increased considerably. To satisfy this demand, electrical companies have to improve the efficiency of their transmission lines. Also, with the liberalization of electrical markets, the individual clients will have the possibility to choose the supplier companies that provide them a better quality of service. Efficiency of the system is based mainly on the continuity of the service, avoiding faults that minimize economical losses for companies and users. To maintain this continuity, one of the main problems that have been found is the effect produced by pollution on the insulators of electric lines. This pollution is one of the main causes of flashover on the insulators. The insulator begins to fail when the pollutants that exist in the air settle in the surface of the insulator and combine with the humidity of the fog, rain or dew. The mixture of pollutants, plus the humidity form a layer that can become conducting and passage of current that will facilitate the conditions of short circuit. This is due to a decrease of the resistance of the insulator surface. Unless there is a natural

cleaning or an adequate maintenance, the power quality will be affected by a possible flashover on the insulator. In other words, pollution degrades insulators and affects severely to their electric characteristics, being one of the main causes of miss operation of insulators. Therefore, the electric companies should prevent the interruptions of the service, produced by contaminated insulators. Flashover of contaminated insulator in pollution areas proved to be one of the most important factors influencing the design of EHV & UHV transmission line insulation. The mathematical analysis of the phenomenon involved is quite complicated so the majority of investigations in this field are mainly of experimental nature. Test of contaminated insulators is roughly classified into Natural and Artificial types. In natural tests insulators are exposed to climatic conditions in areas with severe pollution and are usually stressed with their maximum operating voltage. Insulator performance is judged by occurrence of flashovers and by recording leakage current. Such tests are quite useful as they correspond to actual service conditions. Their only drawback is that they are expensive and time consuming. In artificial tests insulators are dipped in contaminated solution prepared according to the contamination region and commonly occurring salts in that region and flashover tests (in dry and wet conditions) are performed in laboratory. The results obtained by this can be approximated with natural flashover of insulators. The present study has been undertaken to investigate the performance of insulators under artificially polluted conditions. The effects of different salt species present in natural contaminants, quantity of contaminants, wetting and flashover voltage of contaminated insulators has been checked. There are so many methods in determining the degree of contamination but we are mainly concerned with ESDD (Equivalent Salt Deposit Density) method in our study because it is the most reasonable and practical way of defining the degree of contamination of insulators. Based on the values of ESDD calculated in the laboratory tests the detailed chemical analysis of the contaminating salts has been done and their effects on the flashover voltage of insulators are studied.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

82

II. LITERATURE REVIEW In maintaining power quality and effective operation of power system, Transmission & Distribution system plays a vital role. Insulators are the main components T&D system which runs hundreds of KM through different atmospheric condition. Various salts and ions come in contact with these lines which adversely effects electrical characteristic and power quality of line. Flashover voltage due to presence of ions gets reduced and causes blackouts of power system. They investigated the ac flashover voltages of contaminated insulators at reduced pressure and a formulated an empirical mathematical model for height and FOV so as to see the effect of height on flashover voltage [1]. They reported the mechanism of dry band formation and temperature distribution along the surface of an energized artificially polluted insulator string [2]. Based on the experimental investigation of the performance of insulators with spiral shaped sheds under laboratory and field conditions it was shown that the flashover voltage of spiral shaped insulators is lower than that of insulators with standard sheds. The poorer performance of spiral shaped insulators is caused by the formation of non-uniform surface layer along the leakage path [3]. Study of Leakage Current Behavior on Artificially Polluted Surface of Ceramic Insulator was done [4]. A Study of Flashover Voltage of Artificially Polluted Porcelain Disc Insulator under Natural Fog Conditions for northern part of India carried out by experimental setup [5]. A mathematical relationship between the insulator resistance, leakage distance and Equivalent Salt Deposit Density (ESDD) was developed using Dimensional Analysis technique. Based on artificial contamination of NaCl and distilled water FOV and surface resistivity were measured and compared with the proposed Mathematical model [6]. The analytical expression between the creepage distance and Salt Deposit Density for the polluted insulator has been derived using Dimensional Analysis technique [9]. They developed a theoretical model to predict flashover voltage of non-ceramic insulators. The model was based on reignition and arc constants that have been derived from electric field simulations and experimental data of flashover voltage and surface resistance measurements. New and field-aged silicone rubber and ethylene propylene diene monomer rubber samples were evaluated [10]. A relationship between the minimum flashover voltage (VMF) and the insulator dry arcing distance for standard porcelain station post insulators, as typically used in Hydro-Quebec substations, was investigated under icing conditions. Their experimental results show that the VMF increases nonlinearly with an increase in insulator length. Based on these results, an improved mathematical model for predicting the minimum flashover voltage vs. length of ice-covered standard design insulators is presented [11]. Accurate measurement of low insulator contamination

levels was carried out and measurement technique was explained [12]. They presented data that aid physical interpretation of leakage currents on insulators suffering pollution under real physical conditions. The measurements were performed on the power distribution system on the island of Crete and an on-line recording system was installed on the insulators of the 150 kV networks, in order to monitor the leakage currents on the polluted insulators and investigate the various parameters of the phenomena. It is convincingly shown that in the timeframe of one period of 50 Hz, above a threshold voltage value (smaller than the peak voltage) a non-linear iv relationship is developed, leading to bi-stability and non-linear conductance phenomena. The well-established theory regarding leakage currents on high voltage insulators, as well as that concerning dielectrics in strong electric fields provide the required theoretical basis for the interpretation of the experimental results obtained [14]. They presented several measuring methods to evaluate the pollution levels on outdoor insulators are described and after comparison concluded that, Leakage Current Measurement LCM method is a reliable method for pollution level measurement for outdoor insulators and surge arresters[15]. III. CONTAMINANTS AND SOURCES
TABLE 1: LIST OF CONTAMINANTS

Geographical location

Contaminants CaCl2,Ca (NO3)2, MgCl2, NaNO3, NaCl,MgSO4,CaCO3, CaSO4.2H2O Fly-ash, industrial smokestacks, cements, fertilizers, burning of fuel MgCl2,KC1,NaCl,CaCl2, NaNO3 Salts, dew, smoke, smog NaCl,KCl,CaSO4,MgCO3, KNO3,Na2SO4,Fe2O3,CaCO3 Process industries & oil refineries products Soil dust, fertilizers, etc. Road salt, smoke Bacteria, algae, mushrooms and lichens, oxalic acid, ice

Industrial

Marine Desert

Rural areas Highways Other types

A. Basics of flashover mechanism In this section, basics of flashover mechanism are given along with electrical factors which regulates flashover voltage is also discussed: 1) Deposition of Contaminants 2) Wetting of Contaminants: Porcelain and glass are easily wet able due to their high surface energy i.e., the surface exerts sufficient energy on a drop of water to cause it to spread out into a thin film.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

83

3) Formation of dry bands: The voltage across the insulator causes leakage currents of varying magnitude and duration, resulting in heating effect. Depending on insulator shape, pollution distribution over surface and the degree of wetness the voltage distribution is differently affected. The highest heating effect is generally found in the highest stressed region, forming a dry band. 4) Occurrence of Flashover: Due to formation of dry bands the voltage distribution changes and a relatively high voltage appear across the dry bands. In certain cases the voltage gradient may exceed the critical level to initiate a surface arc which either extinguishes or evolves to a point leading to insulator flashover. B. Chemical factors which regulates the magnitude of flashover voltage. Size of atom Mobility of the hydrated ions Weight of the ions Solubility pH of the salts Hydration energy of the salts Lattice energy of the salts Structure of crystal Out of these various factors pH, solubility, mobility, size & weight of ions play an important role in deciding flashover voltage of contaminated insulators. C. Measurement methods for degree of contamination The following measurement methods are used for degree of contamination level: 1) Equivalent Salt Deposit Density(ESDD) Method: By definition, the density of equivalent salt deposit density (ESDD) equals an amount of sodium chloride which, solved in water, will change waters conductivity to the level equal to that resulting from the solution of polluted deposits gathered from insulator surface divided by the insulators surface area (mg/cm2). This method is generally used for calculating average pollution based on average density of soluble salt The equation used in ESDD calculation is [9]:
V( ESDD 0.55 1 0.02(T A 20) ) - - - - - - - - - - - - - (1)

Where, is the layer conductivity at lab temperature (in mS/cm), V is the volume of solution in conductivity measurement at laboratory temperature (in ml), T is the temperature of the solution (C), A is the area of the insulator surface (in cm2). 2) Leakage Current Method (LCM):

The general systematic structure of leakage current measurement method is illustrated in Fig1. A collarshaped ring is placed in the end of the insulator near to the earth. The leakage current sensor is placed between the insulator and the ring to create a closed-loop current. The sensor, which has a high performance speed, works based on Halls effect current transformer. The leakage current made on the insulator surface passes through the sensor and flows toward the earth. The sensor performance is based on Halls effect and magnetic fields. Sensors input impedance is of a very small value. Sensors output it directly connected to the central unit of the information recording system. This unit consists of the analog to digital (A/D) converter and a microprocessor to gather information, which records leakage current indexes in all the insulators being tested. All the adopted and saved information can be transferred through RS232 port serial or modem. For sufficient information, the sampling frequency related to the A/D converter is usually selected for 20 kHz. The central unit is also responsible for recording the phase voltage magnitude to the earth and line current for each of the phases. Since the weather conditions like humidity and temperature have an essential role in creating leakage current, it is necessary for the measurement system to be equipped with sensors of temperature and humidity and wind velocity gauges. The output of the sensors also is connected to the central unit of information recording and is saved. This sensor based leakage current scheme can be utilized to insulator condition monitoring. a message can be generated and sent to the maintenance division for insulator cleaning. 3) Non Soluble Salt Deposit Density (NSDD) Method: This method is the developed form of ESDD method, in which non-soluble pollution content in available samples is measured 4) Directional Dust Deposit Gauge (DDG) Method: This method was first chosen by research institute of Electricity Supply Commission in 1974 to examine insulator pollution. DDG includes four vertical split pipes and a pot below each pipe to gather pollution. The pipes are placed along the four geographical directions, north, south, east, and west. To facilitate international comparison of the results, the size of DDG cracks must be the same in all places to be tested DDG pollution index, according to equation (3), is the average of the four conductivities (S/cm) obtained from the four directions for a 30-day month in 500 cc normalized washing water. Pollution gathering machine in DDG method Normalized conductivity and average conductivity are worked out from the equations (2) and (3), respectively. [9]
N V 30 ) ( ) . - - - - - - - - - - - - - - - - - - - - - - - - - (2) 500 N N S E W Average condutivity - - - - - - - (3) 4 C (

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

84

High soluble salts

Low soluble salts MgSO4, Na2SO4, Na2CO3, K2SO4, CaCO3,CaSO4.2H2O, MgCO3

Where, N, S, E, w = normalized conductivity indexes in north, east, west, and south (S/cm) C = conductivity (S/cm) V = volume of distilled water (ml) N = number of days when the insulator has been under investigation

Ca(NO3)2, NaNO3, MgCl2 CaCl2, NaCl, KCl

TABLE.5: LIST OF HIGH & LOW SOLUBLE SALTS

Fig. 1: General systematic structure for LCM

IV. PH ANALYSIS FOR VARIOUS SALTS AND IONS A. Chemistry of Ions In this section, physical properties (solubility, mobility, weight) of various metal ions like chlorides, carbonates, sulphates, and nitrates are discussed which affects electrical properties of contaminants [6,7]. From table 2 & 3 we can conclude that chloride, carbonate, sulphate, and Nitrate of Na have highest solubility and mobility. From table 4 we see that Na salts has least weight among others metal salts. From table 5 we conclude that nitrates and chlorides of metal salts are highly soluble than sulphate and carbonate salts. Combining these all trends we conclude that nitrate and chlorides salts of Na and Mg can adversely affect the electrical properties of insulator surface.
TABLE 2: SOLUBILITY TREND OF VARIOUS METAL IONS

Effect of ph on flashover voltage: Fig2. shows pH scale and trend of physical properties variation along pH scale which ultimately effect flashover voltage. The term pH, indicating the hydrogen ion (positively charged hydrogen atom) concentration of a solution, is a measure of the solutions acidity. The term pH (from French pouvoir hydrogene, hydrogen power) is defined as the negative logarithm of the concentration of H+ ions: pH = -log10[H+], where [H+] = concentration of H+ ions in moles per liter. The above logarithmic formula was first introduced by S.P.L. Sorensen. pH and pOH both value of Pure water = 7.0.

Fig. 2: pH scale

E F G

TABLE 3: MOBILITY TRENDS OF VARIOUS METAL IONS

A= Acidic strength decreases B= Mobility of ions decreases C= No of (H+) ion decreases D= No of OH- ion increases E= Velocity of hydrated ion decreases F= Conductivity & leakage current increases
G= Corrosive nature

TABLE 4: WEIGHT TREND OF VARIOUS METAL IONS

B. Chemical Analysis Of Flashover Voltages For Various Salts. 1) Role of H+ and OH- ion concentration (i.e. pH) on the flashover voltage of insulators H+ ion has very small radius1.510-5 where as hydrogen atom has radius 0.7414 . Due to very high nuclear charge density it has very high polarizing power. Thus it distorts electron cloud on other atoms and ions. Its weight

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

85

is very small (atomic weight =1 amu) while weight of (OH-) ion is relatively very high (ionic weight=17amu). As the pH value increases from 0 to 14, no. of H + ion in the solution decreases and no. of hydroxyl (OH -) ions increases. As the weight and size of hydrogen ion is less, mobility of its ion is high that causes low surface resistance and large leakage current. Thus flashover voltage gets increased. When the wet contaminants are basic (pH more than 7) due to large size and weight of hydroxyl ion (OH-) mobility of ion is low that causes high surface resistance and low leakage current. Thus flash over voltage is relatively higher than acidic salts. 2) Effects of solubility of salts on flashover voltage and its comparison with experimental values [7, 8] Nitrates and chlorides salts are highly soluble and due to their high solubility in presence of moisture surface conductivity of insulator and leakage current increases and that leads to the lower flashover voltage.
Salts C (gm/l) T (C) P (cm of Hg) 74.5 74.5 74.5 74.5 74.2 74.2 74.2 74.2 74.1 74.1 74.1 74.1 73.9 73.9 73.9 73.9 73.8 73.8 73.8 73.8 73.8 73.8 73.8 73.8 74.3 74.3 74.3 74.3 73.8 73.8 73.8 73.8 73.8 73.8 Dry F.O. V (kV) 86 78 70 65 87 84 83 82 86 84 83 80 86 83 79 76 68 65 58 54 82 80 79 75 65 63 59 53 72 70 69 71 67 64 Wet F.O. V (kV) 65 61 57 52 57 54 51 49 54 52 51 48 69 67 64 60 51 50 49 47 54 52 50 49 56 54 51 49 50 46 44 50 49 47 pH

In past maintenance procedure were usually started after: (a) visual inspection of insulator surface, (b) hissing noise originating from insulator surface due to corona discharge, (c) by monitoring leakage current regularly. pH is the another method of measuring of degree of contamination, which can be effectively utilized in maintenance of insulator. In severe pollution & low rain probability zones, not only selection of good insulator but also a regular maintenance plan is required to eliminate pollutant layer settled on the insulator surface. Periodic or urgent washing of insulator is very effective for prevention flashover problem. The methods of insulator washing are roughly classified as follows: i. Hand Washing ii. Washing by using hot line washing tool iii. Washing by built in or movable hot line washing equipment In addition to washing, other techniques used for maintenance and detection are: i. Visual inspection ii. Noise detection iii. Use of Leakage Current Monitors (LCMs) EXPERIMENTAL SETUP, SUPPLY AND MEASUREMENT SYSTEM AND INSULATOR SPECIFICATION The high voltages were obtained from a testing transformer of 391V/150kV, 1-phase 30kVA rating. The Tests were carried out using the suspension insulators which are extremely used in Overhead networks in India and neighboring countries. The unit diameter is 254mm, its spacing is 146 mm, and its total leakage distance is 305mm. A. Experimental Setup VI.

CaSO4

MgSO4

MgCl2

CaCO3

NaNO3

NaCl

CaCl2 CaCO3 +NaCl +CaO CaSO4 +NaCl +CaO

25.0 37.5 62.5 75.0 25.0 37.5 62.5 75.0 25.0 37.5 62.5 75.0 25 37.5 62.5 75.0 25.0 37.5 62.5 75.0 25.0 37.5 62.5 75.0 25.0 37.5 62.5 75.0 50.0 75.0 87.5 50.0 75.0 87.5

25 25 25 25 28 28 28 28 30 30 30 30 32 32 32 32 34 34 34 34 35 35 35 35 31 31 31 31 35 35 35 35 35 35

7.8 8.3 8.5 8.7 8.3 8.4 8.5 8.6 7.6 7.7 7.8 8.0 7.4 7.4 7.5 7.5 8.8 9.0 9.1 9.1 8.4 8.5 8.5 8.5 6.3 6.4 6.5 6.5 11.9 11.9 12.0 11.8 11.9 11.9

Fig. 3: Experimental setup for flashover test

B. Observations & results: 1) Observations:


TABLE 6: OBSERVATION TABLE

V.

INSULATOR MAINTENANCE METHODS

The high voltages were obtained from a testing transformer of 391 V/150kV, 1- phase 30kVA, rating the voltages were measured using a voltmeter (accuracy 1) connected at the primary side of the transformer, that

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

86

reads the low side voltage. The corresponding high voltages were obtained from a calibration curve dawn by using the sphere sphere electrode system having diameter of 25cm (IS-1876, 1961)

Fig.7: Comparison of wet F.O.V for mixture (NaCl+CaCO3+CaO) & (CaSO4+NaCl+CaO)

Fig. 4: Flashover of insulator under test

C. Insulator Specifications
Shed diameter: 254mm Unit spacing: 146mm leakage distance: 305mm Top surface area: 691cm2 Bottom surface area: 908cm2 Total surface area: 1599cm2

Fig. 8: F.O.V (dry) vs. pH characteristics of all salts in dry condition Fig. 5: Insulator Specification Details

D. Results Graph for Flashover voltage with pH for different salt compositions in dry and wet conditions are drowned in graph from 6- 9.

Fig. 9: F.O.V (wet) vs. pH characteristics of all salts in wet condition

Fig6. Comparison of dry F.O.V for mixture (NaCl+CaCO3+CaO) & (CaSO4+NaCl+CaO)

Form the above Fig6-9 one may conclude that dry flashover voltage of contaminated insulator is not too much affected while in wet condition ions present on the surface of insulator takes active part in charge transportation and leakage current on the surface of insulator increases rapidly. Due to rise in leakage current flash over in wet condition gets reduced and at lower voltage sparking or flashing takes place. Flashover voltage

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

87

in case of contamination by single salts is much higher than in case of mixture of different ions. Insulator contaminated with chlorides and nitrates ions have lower flashover voltage than other salts. From graph 6 we can conclude that as basic or acidic nature of salt increases, in both cases flashover voltage reduces drastically. VII. CONCLUSIONS From the foregoing discussion and the study of polluted insulators, it may be concluded that the insulator flashover under contaminated conditions occurs when the contaminants on the insulator surface gets wet and the surface resistivity gets low, so the flashover voltage of polluted insulators depends on the kind and quantity of contaminants. Accordingly, both chemical analysis of the contaminants and measurement of contaminant quantity are necessary from the view point of design and maintenance of insulator under contaminated condition. There are different chemical factors which includes (Size of atom, Mobility of the hydrated ions, Weight of the ions, Solubility, pH of the salts, Hydration & Lattice energy of the salts, Structure of crystal) influences the flashover voltage. Out of these various factors pH, solubility, mobility, size & weight of ions are the main factors which play an important role in deciding flash over voltage of contaminated insulators. After going through the chemical effects of various salts of alkaline earth metals and observing their flash over voltage for dry and wet condition, we conclude that chloride and nitrates which are soluble in water affects the flash over voltage drastically. Flashover voltage due to acidic contaminants should be lower than basic contaminants. As basic or acidic nature of salt increases, in both cases flashover voltage reduces drastically. As pH increases from 0 to 7 flashover voltage should increase and as pH increases from 7 to 14 flashover voltage should slightly decrease. As reported in this work, the pH of the contaminants plays a vital role in determining the F.O.V of a contaminated insulator. Thus apart from visual inspection, conductivity measurement, ESDD calculation, leakage current monitoring, pH of the deposits on the insulator can be used as a pollution severity indicator and can be utilized for deciding maintenance scheduling of overhead line insulators. For effective condition monitoring pH monitoring or Leakage current monitoring can be used. Leakage current monitoring is relatively effective because it can be interfaced with communication system and remote controlling can be performed. REFERENCES
[1]. Sundararajan R, W. N. Robert, Effect of Altitude on the Flashover Voltage of Contaminated Insulators, IEEE Annual Report Conference on Electrical Insulation and Dielectric Phenomena, San Francisco, October 20-23, 1996. [2]. Subba B.R, Nagabhushana G.R, Study of Temperature Distribution along an Artificially Polluted Insulator String, IISc, Plasma Science & Technology, vo1.5, No2 (2003). [3]. Chrzan K.L,Kindersberger I.J, Pollution behaviour of insulators with spiral shaped sheds, IEEE trans, Germany, 1997.

[4].

[5].

[6].

[7]. [8]. [9].

[10].

[11].

[12].

[13].

[14].

[15].

Subba B.R, Nagabhushana G.R, Study of Leakage Current Behavior on Artificially Polluted Surface of Ceramic Insulator,HV Engg, Indian Institute of Science, Bangalore, 2003. Khan A.A, Masood A, Husain E,A Study of Flashover Voltage of Artificially Polluted Porcelain Disc Insulator under Natural Fog Conditions IEEE Trans , Department Of Electrical Engineering, Aligarh Muslim University, Aligarh, India 2007 Salam M A, Nadir Z, Akbar M and Islam M S, Study the Effects of Different types of Contaminants on the Insulator Resistance, ICECE 2002, 26-28, Dhaka, Bangladesh December 2002. Gupta R.K, Amit R.K, A Text Book of Inorganic Chemistry, Arihant Prakashan Meerut, 2nd Edison, 2003. LEE J. D. Concise inorganic chemistry, oxford press , Fifth edition, 2005. Salam M.A, Ahmad H, Ahmad A. S, Fuad S.A,Paul K.C, Tamsir T, Buntat Z, Piah M. A. M. , Study of Creepage Distance of the Contaminated Insulator in Correlation with Salt Deposit Density, IEEE Conference on Electrical Insulation and Dielectric Phenomena, 2000. Prediction of flashover voltage of non-ceramic insulators under contaminated conditions by S. Venkataraman and R. S. Gorur Department of Electrical Engineering Arizona State University Tempe, AZ, USA Dehkordi J.F, Zhang J, Farzaneh M, Experimental Study and Mathematical Modeling of Flashover on EHV Insulators Covered with Ice, 61st Eastern Snow Conference Portland, Maine, USA 2004. Chisholm,W. A. Accurate measurement of low insulator contamination levels, IEEE Transaction on Power Delivery, vol. 9, No. 3, July 1994. An Investigation Made About the Behaviours of Polluted Insulators Based on Artificial Contamination Tests by Mohammad Reza Shariati, Davood Mohammadi, Seyed Jamal Addin Vaseai, Mohammad Oskuoee, Mohammad Tagi Hasanzad, High Voltage Dept. Niroo Research Institute Tehran, Iran Thalassinakis E, Karagiannopoulos C.G, Measurements and interpretations concerning leakage currents on polluted high voltage insulators, Public Power Corporation, Department of Transmission and Operation, Kastorias Street, Heraklion, Crete GR-71307, Greece, Meas. Sci. Technol. 14 (2003) 421426 Nabavi S.M.H, Gholami A, Kazemi A, Masoum M.A.S, Evaluation of Leakage Current Measurement for Site Pollution Severity Assessment, Leonardo Electronic Journal of Practices and Technologies, ISSN 1583-1078, Issue 10, January-June 2007, p. 39-54.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

88

Comparative Study of Control Design Techniques for DC Motor


1

Rajat Kamboj, 2Veena Sharma


NIT Hamirpur-177005, H.P., India
1

1, 2

rajatkamboj22@gmail.com

Abstract Designing a control system is a creative

process involving a number of choices and decisions. These choices depend on the properties of the system that is to be controlled and on the requirements that are to be satisfied by the controlled system. The decisions imply compromises between conflicting requirements. This paper presents the comparative study of control system design techniques for DC Motor. Three control design methods namely feedforward control, feedback control design and linear quadratic regulator (LQR) have been applied to control the parameter of DC Motor. The main objective of these controller algorithms is to minimize the deviation of the speed of DC Motor from the reference speed. Software implementation of these controllers has been completed through MATLAB.
Keywords Mathematical model of DC motor, Feedforward control, Feedback control, LQR control.

I. INTRODUCTION DC Motors have been widely used in many industrial applications such as electric vehicles, steel rolling mills, electric cranes, and robotic manipulators due to precise, wide, simple, and continuous control characteristics. Traditionally rheostatic armature control method was widely used for the speed control of low power dc motors [1]-[2].However the controllability, cheapness, higher efficiency, and higher current carrying capabilities of static power converters brought a major change in the performance of electrical drives. The desired torque-speed characteristics could be achieved by the use of conventional proportional integral-derivative (PID) controllers [3]-[4]. As PID controllers require exact mathematical modelling, the performance of the system is Questionable, if there is parameter variation. In recent years neural network controllers (NNC) were effectively introduced to improve the performance of nonlinear systems[5]. The application of NNC is very promising in system identification and control due to learning ability, massive parallelism, fast adaptation, inherent approximation capability, and high degree of tolerance. The outer speed and inner current control loops are designed as PD or PI controllers [6]. However, the cascaded control structure assumes that the inner loop dynamics are substantially faster than the outer one. The electrical and mechanical parameters of the DC motor, i.e., resistance, inertia, back-EMF, damping are identified

from observations of the open loop response. Coulomb friction is considered as the main cause of the nonlinear motor behaviour and is adequately compensated by a feed forward control signal. The residual steady state error caused by minor nonlinearities and uncertainties in the model is compensated by an integral error feedback signal. The proposed controller is evaluated for high and low velocity reference profiles including velocity reversal to demonstrate its efficiency for high-performance servo applications. The proposed scheme attempts to bridge the current gap between the advance of control theory and the practice of DC actuator systems .This paper is structured as follows: Section II describes the Mathematical model of the DC motor derived from electromechanical relationships including friction. The DC Motor controllers are described in Section III. The details about the feedforward controller, feedback controller and LQR design of the optimal state feedback control are given in Section III. Section IV presents simulation results. Finally, section V summarizes the major conclusions of the paper .The objective of this work is to improve the system robustness against load torque fluctuation and system parameter variations. By using different control design techniques, the optimum control design analysis of D.C.Motor has been done and it has been found that the LQR design is an optimal one. II. MATHEMATICAL MODEL OF DC MOTOR The goal in the development of the mathematical model is to relate the voltage applied to the armature to the velocity of the motor[7]-[8]. Two balance equations can be developed by considering the electrical and mechanical characteristics of the system. III. ELECTRICAL CHARACTERISTICS The equivalent electrical circuit of a dc motor is illustrated in Figure1. It can be represented by a voltage source (Va) across the coil of the armature. The electrical equivalent of the armature coil can be described by an inductance (La) in series with a resistance (Ra) in series with an induced voltage (Vc) which opposes the voltage source. The induced voltage is generated by the rotation of the electrical coil through the often referred to as the back emf (electromotive force) fixed flux lines of the permanent magnets. This voltage is flux lines of the permanent magnets. This voltage is of the permanent magnets.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

89

where B is the damping coefficient associated with the mechanical rotational system of the machine. Substituting eqns. (7), (8), and (9) into eqn. (6) gives the following differential equation: (10) State Space Representation The differential equations given in eqns. (5) and (10) for the armature current and the angular velocity can be written as (11)
Fig. 1: Electrical representation of DC Motor

A differential equation for the equivalent circuit can be derived by using Kirchhoffs voltage law around the electrical loop. Kirchhoffs voltage law states that the sum of all voltages around a loop must equal zero, or (1) According to Ohm's law, the voltage across the resistor can be represented as (2) where ia is the armature current. The voltage across the inductor is proportional to the change of current through the coil with respect to time and can be written as (3) where La is the inductance of the armature coil. Finally, the back emf can be written as (4) where kv is the velocity constant. w a is the rotational velocity of the armature. substituting eqns. (2), (3), and (4) into eqn. (1) gives the following differential equation: (5) Mechanical Characteristics Performing an energy balance on the system, the sum of the torques of the motor must equal zero. Therefore (6) where Te is the electromagnetic torque, T w ' is the torque due to rotational acceleration of the rotor, T w is the torque produced from the velocity of the rotor, and T L is the torque of the mechanical load. The electromagnetic torque is proportional to the current through the armature winding and can be written as (7) where kt is the torque constant. Tw ' can be written as (8) where J is the inertia of the rotor and the equivalent mechanical load. The torque associated with the velocity is written as (9)

(12) which describe the dc motor system. Putting the differential equations into state space form gives (13)

(14) This is expressed symbolically as,

(15) (16) Transfer function block diagram A block diagram for the system can be developed from the differential equations given in eqns. (11) and (12). Taking the Laplace transform of each equation gives (17) (18) If perturbations around some steady state value are considered, the initial conditions go to zero and all the variables become some change around a reference state, and the equations can be expressed as follows

The above equations can then easily be put into block diagram form. The block diagram obtained from these equations for a Permanent magnet dc motor is shown in Figure.2

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

90

Fig. 2: Block diagram representation of eqns. (19) and (20)

by pure feed-forward without feedback: the external command or controlling signal must be available, and the effect of the output of the system on the load should be known [9]. Sometimes pure feed-forward control without feedback is called 'ballistic', because once a control signal has been sent, it cannot be further adjusted; any corrective adjustment must be by way of a new control signal. In contrast 'cruise control' adjusts the output in response to the load that it encounters, by a feedback mechanism.

The block diagram in Figure2. can be simplified by making the assumption that the load torque is constant. In the case of a sun tracking servo system, the only load torque to be concerned with is the friction in the system, which is relatively constant while the motor is moving. Since the change in TL is zero, it does not need to appear in the block diagram. Also, if one only focuses on the angular velocity as the response of interest,

Fig. 5: Feed-forward control of DC Motor

Fig. 3: Block diagram.

This block diagram is then easily reduced by block diagram algebra to an overall transfer function. (21)

2. Feedback Control Feedback describes the situation when output from (or information about the result of) an event or phenomenon in the past will influence an occurrence or occurrences of the same (i.e. same defined) event / phenomenon (or the continuation / development of the original phenomenon) in the present or future. When an event is part of a chain of cause-and-effect that forms a circuit or loop, then the event is said to "feedback" into itself [10]. Feedback is a mechanism, process or signal that is looped back to control a system within itself. Such a loop is called a feedback loop. In systems containing an input and output, feeding back part of the output so as to increase the input is positive feedback; feeding back part of the output in such a way as to partially oppose the input is negative feedback. In more general terms, a control system has input from an external signal source and output to an external load; this defines a natural sense (or direction) or path of propagation of signal; the feed-forward sense or path describes the signal propagation from input to output; feedback describes signal propagation in the reverse sense.

Fig. 4: Overall transfer function of DC Motor

IV. DC MOTOR CONTROLLER 1. feed-forward control Feed-forward is a term describing an element or pathway within a control system which passes a controlling signal from a source in the control system's external environment, often a command signal from an external operator, to a load elsewhere in its external environment. A control system which has only feed-forward behaviour responds to its control signal in a pre-defined way without responding to how the load reacts; it is in contrast with a system that also has feedback, which adjusts the output to take account of how it affects the load, and how the load itself may vary unpredictably; the load is considered to belong to the external environment of the system. Some prerequisites are needed for control scheme to be reliable

Fig. 6:.Feedback control of DC Motor

3. LQR Controller Cost function is the mathematical representation of a measure of performance, means minimum time, minimum speed, minimum energy etc. As we know that, for first order system at (22) So the control function which will minimize the cost function

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

91

(23) where, Q and R are Positive definite means both are symmetric and having positive eigen values. These assumptions imply that the cost is nonnegative, means it can be zero. There are several procedure to solve the LQR problem, one approach is to finding a controller that minimize the LQR cost function is based on finding the positive definite solution of Riccati equation [11]-[12]. (24) and the LQR has the optimal control form (25) where k(t) is a properly dimensioned matrix, given as (26) The positive definite solution of the algebraic Riccati equation results in an asymptotically stable closed loop system [13].

2. Response of D.C.Motor using Feed-forward control In feed-forward control of DC motor the feed-forward gain Kff should be set to the reciprocal of the DC gain from Va to w. Kff = 1/dc gain (dcm1) Kff=4.1000. To evaluate the feed-forward design in the face of load disturbances, the response to a step command w_ref=1 with a disturbance Td = -0.1 Nm between t=5 and t=10 seconds is simulated:
1.2 1 0.8 Setpoint tracking and disturbance rejection

Amplitude

0.6

To: w
0.4
disturbance T = -0.1Nm
d

0.2 0

-0.2 0

10 Time (sec)

15

20

Fig. 9: Feed-forward control response of DC motor

3. Response of DC Motor using Feedback control To enforce zero steady-state error, use integral control of the form C(s) = K/s where K is to be determined. To determine the gain K, we use the root locus technique applied to the open-loop.
Fig. 7: LQR control of DC Motor
15 Root Locus

10

V. SIMULATION RESULTS
5

The physical constants of motor are: R = 2.0 Ohms L = 0.5 Henrys Km = 0.1 Kb = 0.1 Kf = 0.2 J = 0.02 1. The angular velocity response of DC Motor A state-space model of the DC motor with two inputs (V a, Td) and one output (w) is constructed. The plot of the angular velocity response to a step change in voltage Va is shown in Figure8
0.25 0.2 Step Response

Imaginary Axis

-5

-10

-15 -15

-10

-5 Real Axis

Fig. 10: Root locus plot for feedback control response of D.C. Motor

From the root locus technique we find the gain K=5,


1.6 1.4 1.2 1 Setpoint tracking and disturbance rejection feedback

Amplitude

0.8 0.6 0.4 0.2

Amplitude

0.15 0.1 0.05 0 0

0 -0.2 0 5 10 Time (sec) 15 20

Fig. 11: Feedback control response of DC Motor


0.2 0.4 0.6 0.8 Time (sec) 1 1.2 1.4

Fig. 8: The plot of angular velocity response to a step change in voltage

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

92

4. Response of DC Motor using LQR control


[1]
1.4 1.2 1 Setpoint tracking and disturbance rejection feedback (LQR)

[2]

Amplitude

0.8 0.6 0.4 0.2 0 -0.2 0

[3] [4] [5]


5 10 Time (sec) 15 20

[6] [7] [8]

Fig. 12: LQR control response of DC motor

5. Comparison between the designs of feed-forward, feedback and linear quadratic regulator control response of D.C.Motor
1.6 1.4 1.2 1 Setpoint tracking and disturbance rejection feedforw ard feedback LQR

[9]

[10]

0.8

[11]
0.6 0.4 0.2 0 -0.2 0

[12]

[13]
5 10 Time (sec) 15 20

REFERRENCES T. V. Mahendiran, P. Thangam and K. Thanushkodi, Analysis on Linear and Non-Linear Drives Control of Separately Excited DC Motor Using Particle Swarm Optimization, European Journal of Scientific Research, Vol.50, No.2, pp.225-237, 2011. M. Meenakshi, Microprocessor Based Digital PID Controller for Speed Control of D.C.Motor, International Conference on Emerging Trends in Engineering and Technology, pp.960-965, 2001. M. Zhuang and D. P. Atherton, Optimum Cascade PID Controller Design for SISO System, IEE Control, Vol. 1, pp. 606-611, 1994. Tongwen Chen and B.A. Francis, Optimal Sampled-Data Control Systems, Springer, London, Vol.5, pp. 765-771, 1995. Moleykutty George, Speed Control of Separately Excited DC Motor, American Journal of Applied Sciences, Vol. 5, N0, 3, pp.227-233, 2008. M.R. Issa and E. Barbieri, Optimal PI-lead controller design, IEEE Proc. System theory, pp. 364-368, 1996. J.R.White and UMass Lowel System dynamics Mathematical Modeling of Engineering Systems, pp. 1-5, spring 1997. Samart Yachiangkam,Cherdchai Prapanavarat, Udomsak Yungyuen and Sakorn Po-ngam speed-sensorless separately excited dc motor drive with an adaptive observer, IEEE, 2004. J. Snchez, A. Visioli and S. Dormido, An event-based PI controller based on feedback and feed-forward actions, IEEE, 2009. Torsten Wik, Car-Magnus Fransson and Bengt Lennartsson, Feedforward Feedback Controller Design for Uncertain Systems, IEEE, 2003. Jian-Bo Het, Qing-Guo Wang and Tong-Heng Lee, PI/PID Controller Tuning Via LQR Approach, IEEE, 1998. M. Ruderman, J. Krettek, F. Hoffmann and T. Bertram, Optimal State Space control of DC Motor, The International Federation of Automatic Control Seoul, Korea, July 6-11, 2008. Chevrel, L. Sicot and S. Siala, Switched LQ controllers for DC motor speed and current control: a comparison with cascade control, In Proc. Power Electronics Specialists Conference PESC96 Record, pages 906912, Baveno, Italy, June 1996.

Fig. 13: Comparison between the designs of feed-forward, feedback and linear quadratic regulator control response of D.C. Motor

In the above Figthe blue line represents the response of feed-forward control, green line represents response of feedback and red line represents response of linear quadratic regulator control of DC Motor. It can be observed from the above plot that the Linear Quadratic Regulator compensator performs best at rejecting load disturbances than feedback and feed-forward control of D.C.Motor V. CONCLUSION Here in this work, three control design techniques have been applied on D.C.Motor whose performance is to be controlled. These design techniques include Feedforward, Feedback and Linear Quadratic Regulator. Using above techniques, performance of DC Motor in terms of angular speed in the presence of disturbance has been studied in this paper. From the results, it has been observed that feed-forward control handles load disturbances poorly which can be improved by using feedback control. Further it is clearly seen that Linear Quadratic Regulator compensator performs the best at rejecting load disturbances than feedback and feed-forward control.

Amplitude To: w

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

93

Mitigation of Harmonics stress in the Distribution Network of Arid Areas using Power Conditioners
1
1,2

Haroon Farooq, 2M. E. Farrag

School of Engineering & Computing, Glasgow Caledonian University, UK 2 mohamed.farrag@gcu.ac.uk


TABLE 1: IHDI LIMITS

Abstract:There are growing concerns regarding the harmonic distortion level in the power systems owing to the increase of power electronic-based equipment and loads connected to the distribution network. In order to have a sustainable future with robust distribution systems design, it is necessary to investigate the harmonic distortion caused by various nonlinear home appliances. It is also of utmost importance to come up with the novel techniques to mitigate the harmonic pollution. Air Conditioner (AC) is one rapidly increasing non-linear load, especially in warmer countries, causing harmonic distortion. In this research work, the harmonic model of an AC is implemented in Electrical Transient Analyzer Program (ETAP) for the simulation of a typical distribution system in Pakistan. The results show an alarming level of harmonic distortion. Moreover, a harmonics mitigation technique using power conditioners is introduced and implemented to significantly reduce the harmonic pollution. Keywords: harmonics mitigation, ETAP, Power Conditioners

I. INTRODUCTION Harmonic Distortion is drawing a lot of attention from utility engineers in recent years due to an ever increasing trend of non-linear electronic loads into the power systems [1]. These non-linear loads draw non-sinusoidal current waveform giving rise to harmonic distortion. Harmonic distortion is produced due to the presence of the superfluous frequency components as the integer multiple of fundamental frequency [2]. Harmonic pollution is typically computed in terms of Total Harmonic Distortion (THD). THD can be defined as The ratio of the rms of the harmonic content to the rms value of the fundamental quantity, expressed as a percentage of the fundamental [3, 4]. (%) THD = {[(h=2 M2h,rms)] / M1,rms} 100 (1) where M1,rms and Mh,rms give the rms value of fundamental and hth harmonic constituent of the relevant waveform respectively. As per IEEE standards for system voltage level up to 69 kV, the maximum allowed value for total harmonic distortion in voltage (THDv) is 5 % with each individual harmonic distortion (IHDv) not to exceed 3 %. The limits for individual harmonic distortion in current (IHDi) for a system having short circuit ratio (I sc / IL) between 20 and 50 are given in Table 1 [5].

The high level harmonic content has various devastating impacts on the distribution systems including pre-mature failure of transformers, mal-operation of circuit breakers, damage to conductors and increased losses [6, 7]. The Air conditioning load is increasing day by day [8]. Air conditioners (AC) being power electronic loads produce harmonic distortion [9]. This research work investigates the harmonic distortion produced by air conditioners followed by the successful implementation of a mitigation technique using power conditioner to reduce the harmonic pollution. After the identification of the harmonic content in the current waveform drawn by an AC, simulations of a distribution system were carried out in ETAP. Then the harmonic pollution was mitigated using power conditioners. Section 2 explains the harmonic model for an AC, section 3 presents the practical distribution system under study, section 4 gives the harmonic analysis results followed by the mitigation of harmonics and conclusions presented in section 5 and section 6 respectively. II. HARMONICS IN THE CURRENT WAVEFORM

DRAWN BY AN AC Several research works have been carried out to analyze the harmonic content in the current waveform drawn by the air conditioners [10-12]. There is a consensus with respect to the current waveform drawn by modern variable speed air conditioners as shown in Fig1.

Fig. 1: AC Current Waveform

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

94

The frequency spectrum is shown in Fig2 for the AC current, it is clear that the third harmonic is the most significant component.

from Lahore Road Grid, WAPDA Town, Gujranwala, Pakistan. The total number of homes supplied is 24, each of which is assumed to have an AC. The power required for a typical 18,000 BTU AC is 2 kW [13]. All the connected loads are assumed to have a standard power factor of 0.85 [14]. The loading details of each bus are given in Table 3.
TABLE 3: LOADING PARAMETERS

Fig. 2: Frequency Spectrum of AC Current Waveform

The total harmonic distortion in current (THDi) with 3 h 15 is equal to 46.78 %. The harmonic content is summarized in Table 2.
TABLE 2: AC CURRENT WAVEFORM HARMONIC CONTENT

The simulation was carried out for two different scenarios i.e. peak load times and off-peak load times. As per standard procedures under the electricity act of Water and Power Development Authority, the load factor for peak load times and off-peak load times is 70% and 20 % respectively [15]. 4.1 Peak load analysis: 4.1.1 Current and voltage waveform at point of common coupling: The current waveform at bus 4 which is the Point of Common Coupling (PCC) for all the consumers is shown in Fig4 along with its frequency spectrum in Fig5.

III. DISTRIBUTION SYSTEM UNDER STUDY The distribution system considered for simulation for this research project is based upon the one located in Kashmir Colony, Gujranwala, Pakistan as shown in Fig3. The simulation was conducted using ETAP, which is highly acclaimed software for the analysis of power systems.
Fig. 4: Current Waveform at PCC at Peak Load

Fig. 3: Distribution System under Study

IV. HARMONIC ANALYSIS The impacts of air conditioning load are analyzed for a 200 kVA transformer fed by an 11 kV feeder originating

Fig. 5: Frequency Spectrum of Current Waveform at PCC at Peak Load

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

95

The waveform is highly distorted with a THDi equal to 31.16 %. The IHDi for 3rd harmonic is 30 % which is exceeding the standard recommended values. While passing through system impedance, this distorted current also distorts the voltage. The voltage waveform at PCC is shown in Fig6 along with its frequency spectrum in Fig7.

Fig. 9: Frequency Spectrum of Current Waveform at Bus Bar 6 at Peak Load

The waveform is highly distorted with a THDi equal to 31.18 %. The IHDi for 3rd harmonic is 30 % which is violating the IEEE standard. The voltage waveform at bus bar 6 is shown in Fig10 along with its frequency spectrum in Fig11.
Fig. 6: Voltage Waveform at PCC at Peak Load

Fig. 10: Voltage Waveform at Bus Bar 6 at Peak Load Fig. 7: Frequency Spectrum of Voltage Waveform at PCC at Peak Load

Even though the THDv having a value of 4.31 % is marginally within acceptable limits, however IHD v for 3rd harmonic being equal to 3.66 % is beyond the standard practises. 4.1.2 Current and voltage waveform at bus bar 6: The current waveform at bus 6 which is the furthest bus bar from the transformer in terms of transmission length is shown in Fig8 along with its frequency spectrum in Fig9.
Fig. 11: Frequency Spectrum of Voltage Waveform at Bus Bar 6 at Peak Load

Fig. 8: Current Waveform at Bus Bar 6 at Peak Load

The THDv having a value of 8.96 % as well as IHD v for 3rd harmonic being equal to 8.34 % is beyond the standard practises. The reason for a high value of THDv is the distance of bus bar from the transformer, with increasing the length of the transmission line, the system impedance increases which in turn raises the distortion factor of the voltage waveform. 4.2 Off-peak load analysis: 4.2.1 Current and voltage waveform at PCC: The current waveform at PCC is shown in Fig12 along with its frequency spectrum in Fig13.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

96

Fig. 12: Current Waveform at PCC at Off-Peak Load

Fig. 15: Frequency Spectrum of Voltage Waveform at PCC at Off-Peak Load

Even though the THDv having a value of 4.53 % is still marginally within acceptable limits, but IHD v for 3rd harmonic being equal to 3.81 % is greater than the recommended value. MITIGATION OF HHRMONICS USING POWER CONDITIONER (PC) To mitigate the harmonic distortion caused by air conditioning, power conditioners can be deployed. To compensate for the harmonic current injected by a nonlinear load, power conditioner introduces the same amount of current but having harmonic content exactly opposite to that of the original non-linear load. In Simulation, the PC waveform can be generated by taking the difference of an ideal sinusoidal wave and the distorted waveform. This current will hereafter be referred as PC current. For this research work, the peak value of the distorted waveform was identified and then an ideal sine wave having the same peak value and a frequency of 50 Hz was generated. Then to obtain the filter current as shown in Fig16, the distorted waveform was subtracted from the ideal sine waveform and then the Fourier analysis was performed to obtain the harmonic content of the PC current as shown in Fig17. V.

Fig. 13:. Frequency Spectrum of Current Waveform at PCC at Off-Peak Load

The waveform is highly distorted with a THDi equal to 57.86 %. The IHDi for 3rd, 5th and 7th harmonic is equal to 55.68 %, 12.03 % and 8.80 % respectively, which is more than the standard recommended value. There is a significant increase in the THDi as compared to the one at peak-load. This is because with the decrease in linear load at off-peak times, the overall magnitude of the fundamental current flowing into the system decreases giving rise to the higher percentage contribution of the higher order harmonics. Voltage also gets distorted because of this highly distorted current waveform. The voltage waveform at PCC is shown in Fig14 along with its frequency spectrum in Fig15.

Fig. 14: Voltage Waveform at PCC at Off-Peak Load

Fig. 16: PC Current Waveform

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

97

Fig. 17: Frequency Spectrum of PC Current Waveform

The parameters for PC current waveform are summarized in Table 4.


TABLE 4:. PC CURRENT WAVEFORM HARMONIC CONTENT

Fig. 19: Frequency Spectrum of Current Waveform at PCC at Off-Peak with PC

The current waveform is almost a sine wave in the presence of active filter. The THDi drops down from 57.86 % to 2.43 %. Similarly, IHDi for 3rd, 5th and 7th harmonic has significant decline from 55.68 % to 2.34 %, from 12.03 % to 0.51 % and from 8.80 % to 0.37 % respectively. All the values are well within recommended standard values. The voltage waveform at PCC in the presence of active filter is given in Fig20 and its frequency spectrum is depicted in Fig21.

The comparison of the harmonic content of current waveform drawn by air conditioners and that of the filter makes it evident that the amplitude spectrum is almost the same. However, the harmonics are 180o out of phase of each other. 5.1 Simulation with power conditioner: To replicate the impact of active filter on the system, a load having same rating and power factor as that of an AC was inserted in each home but having a harmonic spectrum of filter current waveform. The harmonic analysis was performed to obtain the current and voltage waveforms at PCC at off-peak loading. The simulation showed a substantial decrease in the harmonic pollution. The current waveform at PCC along with its frequency spectrum in the presence of active filter is shown in Fig18 and Fig19 respectively.

Fig. 20: Voltage Waveform at PCC at Off-Peak with PC

Fig. 21: Frequency Spectrum of Voltage Waveform at PCC at Off-Peak with PC

The voltage waveform is also almost a pure sinusoidal waveform. The THDv decreases from 4.53 % to 0.32 %. Similarly, IHDv for 3rd harmonic is within the recommended standard values after falling down from 3.81 % to 0.26 %. Similar results were obtained for all bus bars.
Fig. 18: Current Waveform at PCC at Off-Peak with PC

VI. CONCLUSIONS The simulation results of a residential distribution system showed high levels of harmonic pollution in the presence

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

98

of high penetration level of air conditioning loads. Various individual harmonic components of current as well as voltage at PCC were found to be beyond the IEEE recommended values. The situation becomes more alarming with the increasing distance, in terms of transmission line length, from the transformer. Similarly, harmonic distortion became more profound considering the off-peak loads as compared to peak loads. A harmonic pollution mitigation technique using power conditioners was also successfully implemented. The harmonic distortion substantially decreased in the presence of the power conditioners. REFERENCES
[1]. M.I. Abu Bakar, Assessments for the Impact of Harmonic Current Distortion of Non Linear Load in Power System Harmonics, in Proc. IEEE/PES 2008, pp. 1 6. L. Cividino, Power Factor, Harmonic Distortion; Causes, Effects and Consideration, in Proc. INTLEC 1992, pp. 506 513. Claudio De Capua and Emilia Romeo, 2007, A Smart THD Meter Performing an Original Uncertainty Evaluation Procedure, IEEE Trans. on Instrumentation & Measurement, vol. 56, issue 4, pp. 1257- 1264. Farooq, K. Qian, M.E. Farrag, M. Allan, C. Zhou, Power quality analysis of distribution systems incorporating high penetration level of EV battery chargers, in Proc. CIRED 2011. IEEE 519-1992, Recommended Practices and Requirements for Harmonic Control in Electric Power Systems. Farooq, C. Zhou, M. Allan, M. E. Farrag, R.A. Khan and M. Junaid, Investigating the Power Quality of an Electrical Distribution System Stressed by Non-Linear Domestic Appliances, in Proc ICREPQ 2011. Rana A. Jabbar Khan and Muhammad Akmal, Mathematical Modelling of Current Harmonics Caused by Personal Computers, in International Journal of Electrical and Information Engineering 2008, Vol. 2, no. 6, pp. 336 340. Sami El-Ferik, Syed A. Hussain and Fouad M. Al-Sunni, Identification and weather sensitivity of physically based model of residential air-conditioners for direct load control: A case study, in Energy and Buildings 2006, vol. 38, issue 38, pp. 997 1005. F.A. Gorgette, J. Lachaume, W.M. Grady, Statistical summation of the harmonic currents produced by a large number of single phase variable speed air conditioners: a study of three specific designs, in IEEE Trans. on Power Delivery 2000, vol. 15, Issue 3, pp. 953 959. R.S. Wallace, The harmonic impact of variable speed air conditioners on residential power distribution, in Proc. APEC 1992, pp. 293 298. Essah, E. Feiste, R. O'Connell, R.G. Hoft, G. Brownfield, Harmonic effects of variable speed air conditioners on a singlephase lateral and on the distribution feeders in a typical residential power system. I, in Proc. APEC 1994, pp. 622 627. R.S. Thallam, W.M. Grady, M.J. Samotyj, Estimating Future Harmonic Distortion Levels In Distribution Systems Due To Single-Phase Adjustable-Speed-Drive Air Conditioners: A Case Study , in Proc. ICHPS 1992, pp. 65 69. FutureWorld Energy, Inc, Frequently Asked Questions (FAQ) / CoolSolar Series Air Conditioner, [Online], Available: http://www.futureworldenergyinc.com/pdf/CoolSolar-FAQAC.pdf (Date of access April, 2011) Linxia Guo, Yu Cheng, Lizi Zhang, Haitao Huang, Research on power factor regulating tariff standard, in Proc. CICED 2008, pp. 1 5. Electricity Act 1910, Water and Power Development Authority (WAPDA), Pakistan

[2]. [3].

[4].

[5]. [6].

[7].

[8].

[9].

[10].

[11].

[12].

[13].

[14].

[15].

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

99

Analysis of Direct Torque Controlled Techniques for Induction Motor Drives with Two Level Inverter
1
1 2

Abhishek Verma, 2Yogesh Kumar Chauhan

College of Engineering Roorkee, Roorkee, India School of Engineering Gautam Buddha University Gautam Buddha Nagar, U.P, India
1 2

abhi.ricky03@gmail.com yogeshkchauhan@gmail.com Abstract The objective of the present work is to obtain the The Conventional DTC method yields slow response reduced torque ripples, harmonics distortion. Duty Ratio during start up and change in either direction of torque and Control has been developed to improve the torque flux. In addition to this, also high torque ripples are found. performance and to obtain the voltage space vector required Several techniques have been developed to improve the to compensate the flux and torque errors. In conventional torque performance. One of them is to reduce the ripples Direct Torque Control (DTC), the selection of flux linkage and electromagnetic torque errors are made within the using Duty Ratio Control Technique [3]. In Duty Ratio respective flux and torque hysteresis bands, in order to Control Technique instead of applying a voltage vector obtain fast torque response, low inverter switching frequency for the entire switching period, it is applied for a portion and low harmonic losses. However, DTC drives utilizing of the switching period and the zero switching state is hysteresis comparators suffer from high torque ripple and applied for the rest of the period the ripples can be variable switching frequency. As in duty ratio control reduced. This is defined as duty ratio control in which technique, instead of applying a voltage vector for the entire the portion of the switching period for which a nonzero switching period, it is applied for a portion of the switching voltage vector is applied is known as the duty ratio . period and the zero switching state is applied for the rest of the period the ripples is considerably reduced. In this paper II. ASYNCHRONOUS MOTOR MODEL the simulation of different DTC schemes (Conventional DTC and Duty Ratio Control DTC) has been carried out using A. Axes Transformation MATLAB/SIMULINK and the results are compared. Keywords Asynchronous motor, Torque ripple, Direct torque control, Simulation, ABB Company.

Consider a symmetrical three-phase induction machine with stationary as bs cs axes at 2/3 angle apart, as shown in Fig 1.
bs

I. INTRODUCTION Direct torque control (DTC) drives are finding great interest, since ABB recently introduced the first industrial direct-torque-controlled induction motor drive in the mid1980s, which according to ABB can work even at zero speed. This is a very significant industrial contribution, and it has been stated by ABB that DTC is the latest AC motor control method. ABB has also introduced a DTC based medium voltage drive called the ACS1000, thus feeding pure sinusoidal voltages and currents to the motor. This is ideal for pumps and fans [1]. With the revolutionary DTC technology developed by ABB, field orientation is achieved without feedback using advanced motor theory to calculate the motor torque directly and without using pulse width modulation. The controlling variables are motor magnetizing flux and motor torque. With DTC there is no modulator and no requirement for a tachometer or position encoder to feed back the speed or position of the motor shaft. DTC uses the fastest digital signal processing hardware available. The result is a drive with a torque response that is typically 10 times faster than any AC or DC drive [2].

Vbs

Vsqs qs - axis Vcs Vas cs Vsds as

ds- axis

Fig. 1: Stationary frame a b c to d

q s axes transformation

The main goal is to transform the three-phase stationary reference frame ( as bs cs ) variables into two phase stationary reference frame ( d s q s ) variables and then transform these to synchronously rotating reference frame ( d e qe ), and vice versa. Assume that the d s q s axes are oriented at angle shown in Fig 1. The voltages Vdss and Vqss can be resolved into as bs cs components can be represented in the matrix form as:

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

100

Vas Vbs Vcs


Vqss Vdss Voss

cos cos( cos(


cos 2 sin 3 0.5

sin 120o ) sin( 120o ) sin(


cos( sin(

1 Vqss o 120 ) 1 Vdss 120o ) 1 Voss


Vas Vbs Vcs

Hysteresis-Band Controllers

VD

(1)
Flux Error

SA Voltage Vector Selector SB SC Voltage Source Inverter (VSI) Induction Motor Sk V Stator Flux and Torque Estimator

The corresponding inverse relation is:


120o ) cos( 120o ) 120o ) sin( 120o ) 0.5 0.5

+ -

T*
+ -

Torque Error

Te

(2)

where, Voss is added as the zero sequence component, which may or may not be present. The voltage has been considered as the variable. The current and flux linkage can be transformed by similar equations. It is convenient to set = 0, so that the q s -axis is aligned with the as-axis. Ignoring the zero sequence components, the transformation relations can be simplified with the help of equations (1) and (2) by assuming = 0 in these respective equations as:
Vas Vas
1 s Vqs 2 1 s Vqs 2
2 Vas 3 1

Fig. 2: Conventional DTC drive conFiguration

All of the above equations explain the modelling of induction motor in synchronous reference frame and stationary reference frame. III. CONVENTIONAL DTC In addition to vector control systems, instantaneous torque control yielding very fast torque response can be obtained by employing DTC. Drives with direct torque control DTC are finding great interest, since in the mid of 1980s, ABB introduced the first industrial direct-torquecontrolled induction motor drive. In a DTC drive, flux linkage and electromagnetic torque are controlled directly and independently by the selection of optimum inverter switching modes. The selection is made to restrict the flux linkage and electromagnetic torque errors within the respective flux and torque hysteresis bands, to obtain fast torque response, low inverter switching frequency and low harmonic losses. The required optimal switching voltage vectors can be selected by using a so called optimum switching voltage vector look-up table. This can be obtained by simple physical considerations involving the position of the stator flux linkage space vector, the available switching vectors and the required torque and flux linkage. In the Fig 2, ConFiguration of Conventional DTC drive is shown, in which the comparison between the reference and the actual value is taken and the errors are processed through hysteresis band controllers. The flux loop controller has two levels of digital output and the torque control loop has three levels of digital output. The feedback flux and torque are calculated from the induction machine terminal voltages and currents. The three phase terminals quantities are converted into two phase stationary d q components, which are used for estimating motor torque and stator linked flux. Based on the resultant flux position and the errors in flux magnitude and in torque, a three-dimensional look up table is referred to decide the inverter switching. The Stator Flux and Torque Estimator block shown in Fig 2 gives the sector number S k (in which the flux vector lies) which is fed to the Voltage Vector Selector block. C. Look Up Table Selection The required stator flux is maintained by means of choosing the most suitable VSI state. If the ohmic drop is

(3)
3 s Vds 2 3 s Vds 2
1 Vbs 3 1 Vcs 3 Vas

Vbs = Vcs =

(4) (5)

and inverse equations are expressed as:


Vqss

(6) (7)

Vdss

Vbs

1 3

Vcs

B. Modelling An induction motor is modelled using voltage and flux equations which are referred to synchronous rotating reference frame, denoted by the superscript e . Induction motor model: Voltage and flux linkage; mechanical torque equation as:
Vs e 0
e s e r

Rs ise Rr i
e r

j j(

e e

e s r

d / dt ( )
e r

e s

)
e r

(8)

d / dt (

(9) (10) (11)

Ls ise Lr ire

Lm ire Lm ise
r

Te TL
Te 3P ( 2

Jm p
e sd e isq

Bm
e sq e isd )

(12) (13)

By referring to a stationary reference frame, denoted by the superscript s , with d-axis attached on the stator winding of phase A , the mathematical equations of induction motor can be rewritten as follows.
Vs s
0
s s s r

Rs iss
Rr ir
s s s

d / dt
j
r s r s s

s s

(14)

d / dt

s r

(15) (16) (17)

Ls is Lr ir
3P 2

Lm ir
Lm ( Ls Lr

Lm is

Te

s sd

s rq

s sq

i )

s rd

(18)

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

101

neglected, then the stator voltage impresses directly the stator flux in accordance with the equation given as:
d s dt Vs

or

Vs

(19)

Decoupled control of the stator flux and torque is achieved by using radial and tangential components of the stator flux linkage space vector in its locus. These two components are directly proportional to the components of the same voltage space vector in the same directions. Fig 3 shows the possible dynamic locus of the stator flux, and its variation depending on the chosen VSI states. The possible global locus is divided into six different sectors shown by the discontinuous line. From Fig 3, the general table can be framed. It can be seen from the Table 1 that the states Vk and Vk 3 , are not considered in the torque because they can both increase (initial 30) or decrease (next 30) the torque at the same sector depending on the stator flux position.
V3 (010) 3 2 V2 (110)

Another major concern in direct torque control induction motor drive are torque and flux ripples since none of the inverter switching vectors is able to generate the exact stator voltage required to produce the desired changes in torque and flux. Possible solutions involve the use of high switching frequency or change in inverter topology, these solutions increases switching losses and increase cost. The induction motor stator flux and torque in the stationary reference frame are given by: (Vqs rs iqs ) dt (20) qs
ds

(Vds
ds qs

rs ids ) dt
qs ds

(21) (22) (23)

Te

tan
s

qs ds

(
Vdc

2 ds

2 qs

(24) (25)

V3 (FD, TI) V4 (011) 4 V5 (FD, TD) 1

V2 (FI, TI) V6 (FI, TD) V1 (100)

5 6

V5 (001)

V6 (101)

Fig. 3: Stator flux vector locus and different possible switching voltage vectors
FD: Flux Decreases, FI: Flux Increases TD: Torque decreases, TI: Torque Increases.

Fig. 4: Simulation of Conventional DTC

Finally, The DTC classical table is as follows:


TABLE I: SWITCHING VECTOR SELECTION TABLE

FI

FD

TI TE TD TI TE TD

S1 V2 V0 V6 V3 V7 V5

S2 V3 V7 V1 V4 V0 V6

S3 V4 V0 V2 V5 V7 V1

S4 V5 V7 V3 V6 V0 V2

S5 V6 V0 V4 V1 V7 V3

S6 V1 V7 V5 V2 V0 V4

Equations 20 to 25 are used in the direct self-control to estimate the stator flux and the electric torque Te in the machine. The estimated values of the stator flux and torque are compared to their command values ( ref and Tref respectively) and the inverter switching states are selected according to Table I. It can be seen from equations 20 to 25 that, the changes in torque and flux are functions of input voltage and the inverter switching time. The ripple can, therefore, be reduced either by reducing the input dc voltage amplitude or the switching time. At low input dc voltage the range of speeds in which machine can operate, become small. At higher speeds the emf generated in the machine increase to a level, where the input voltage would no longer be able to generate the required torque and flux regulation. Thus using low voltage levels is not desirable for the machine operation. Fuzzy Implementations The voltage vectors selected from the fuzzy DTC are not optimal in some region of stator flux vector positions. These disadvantages are improved by the fuzzy controller called a duty ratio controller. Since the duty ratio during each switching state is a nonlinear

FD/FI: Flux decreases/Increases. TD/TE/TI: Torque decreases/equal/Increases.

where, S k stator flux sector; stator flux modulus error after the hysteresis block; T torque error after hysteresis block. The sectors of the stator flux space vector are denoted from S1 to S 6 . Stator flux error after the hysteresis block can take three values. The zero voltage vectors V0 and V7 are selected when the torque error is within the given hysteresis limits, and must remain unchanged. IV. DUTY RATIO CONTROL STRATEGY

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

102

function of number of factors: ( Ete Tref Te ) , flux error E ref

torque error and flux e

position (), it is difficult to mathematically model its relation between the three variables. The fuzzy logic control seems to be a reasonable choice to determine the duty ratio during each switching state. Fuzzy logic converts a set of linguistic rules based on expert knowledge into an automatic control strategy. In this control the selected inverter switching state is applied for a portion of the sample period, defined as a duty ratio , and the zero switching state is applied for the rest of the period. The duty ratio is chosen to give an average voltage vector, which causes torque change with ripple reduction. Fuzzy controller includes three inputs (torque error Ete , flux error E , and the position of the stator flux linkage according to corresponding sector) and one output (duty ratio ). Fig 5 describes membership function for inputs and output. The fuzzy logic controller is a Mamdani type and contains a rule base. This base comprises of two groups of rules, each of which contains nine rules. The first group is used when the stator flux linkage modulus is smaller than its reference value (Table II), and the second group of rules is used when it is greater than its reference value (Table II). There are together 18 simple rules and only three fuzzy sets

MATLAB based display of the fuzzy inference diagram, used as a diagnostic it shows which rules are active and how individual membership function are influencing the results. The Surface Viewer is used to display dependency of one of the outputs on any one or two of \the inputs that is, it generates and plots an output surface map of the system.
Vdc FUZZY DUTY RATIO ESTIMATOR

Et

VSI

DIRECT TORQUE CONTROL

Switching Vector Stator Current

INDUCTION MOTOR Stator Voltage

Fig. 6: Block Diagram of Duty Ratio Controller DTC

Fig. 7: Fuzzy Rules for Duty Ratio Control

Fig. 5: Generation of switching pulses

D. Fuzzy Rule-Base As shown in Fig 7, the fuzzy duty ratio controller gives duty ratio () as its output. For the generation of the switching pulses, the duty ratio is multiplied with the DTC sampling period and then it is compared with a triangular wave which has also the amplitude equal to the sampling period and frequency equal to sampling frequency of DTC. Finally the comparator output gives the pulse to the inverter as shown in Fig5 in order to generate optimum voltage vector. The FIS Editor, the Membership Function Editor, the Rule Editor, and the Rule Viewer of the fuzzy controller are used here. The Membership Function Editor is used to define the shapes of all the membership functions associated with each variable. The Rule Editor is for editing the list of rules that defines the behaviour of the system. The Rule Viewer is a

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

103

GROUP 1 RULES If Ete is small and s is small then is medium. If Ete is small and s is medium then is small. If Ete is small and s is large then is small. If Ete is medium and s is small then is medium. If Ete is medium and s is medium then is medium. If Ete is medium and s is large then is medium. If Ete is large and s is small then is large. If Ete is large and s is medium then is large. IF ETE IS LARGE AND S LARGE THEN IS LARGE.
IS

GROUP 2 RULES If Ete is small and s is small then is small. If Ete is small and s is medium then is small. If Ete is small and s is large then is medium. If Ete is medium and s is small then is medium. If Ete is medium and s is medium then is medium. If Ete is medium and s is large then is large. If Ete is large and s is small then is medium. If Ete is large and s is medium then is large. IF ETE IS LARGE AND S LARGE THEN IS LARGE.
IS

Fig. 8(a):. Rotor Speed (rpm) v/s Time (second)

V. SIMULATION RESULTS The control scheme is simulated with Matlab/Simulink, for following case as: In Conventional DTC: The speed reference prescribed to (800; 1000) rpm at t = (0; 1.5) s, The load torque prescribed to (0; 12) Nm at t = (0; 1.5) s with step variation. The speed reference prescribed to (800; 1000) rpm at t = (0; 1.5) s, the load torque prescribed to (0; 10; 20) Nm at t = (0; 0.5; 1.5) s with step variation. Fig 8(a) and 8(b) shows the speed torque responses of the induction motor at different conditions i.e. the constant speed command 800 rpm is given for 0 to 1.5 seconds and the load torque is changed at 0.5 second from 0 to 10 Nm. From 1 to 1.5 seconds the load torque is maintained at 10 Nm, but the speed command is changed at 1.5 second from 800 rpm to 1000 rpm. Also at 1.5 second the load torque is changed from 10 Nm to 20 Nm. From the Fig 8(b) we can see that if the speed command is changed suddenly then the induction motor produces a large electromagnetic torque, which is undesirable. From Fig 8(d), we can see that the torque ripple in case of conventional DTC is Nm. From Fig 9(d), we can see that the torque ripple in duty ratio controlled DTC is + 3, which is very much less as compared to conventional DTC.

Fig. 8(b): Electromagnetic torque (Nm) v/s Time (second)

Fig. 8(c): Stator current (ampere) v/s Time (second)

Fig. 8(d): Torque ripples in conventional DTC

Fig. 9(a): Speed (rpm) v/s Time (second)

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

104

accuracy than the conventional strategy. Ripples of stator current in the case of Duty Ratio Controller have reduced.

Appendix
Power Rating Stator Voltage Frequency Number of Poles Stator Resistance Stator leakage Inductance Rotor Resistance Rotor leakage Inductance Mutual Inductance Inertia Friction Coefficient : : : : : : : : : : : 10 HP 460 volt 60 Hz 4 0.6837 ohm/phase 0.004152 H/phase 0.451 ohm/phase 0.004152 H/phase 0.1486 H 0.05 kg m2 0.0081412

Fig. 9(b): Torque (Nm) v/s Time (second)

REFERENCES
[1] [2] Fig. 9(c): Stator Current (Ampere) v/s Time (second) [3] [4]

[5]

[6]

[7] Fig. 9(d): Torque ripples in duty ratio control.

VI. CONCLUSIONS This Paper presents a comparative study of different direct torque control strategies of induction motor drive based on their simulation results. Two direct torque control strategies (Conventional DTC and Duty Ratio Control) are compared. A comparative analysis among the conventional direct torque control and duty ratio control, strategies are made by simulation in MATLAB/SIMULINK. Based on simulation results, it can be seen that the ripple in torque with duty ratio control is less than the one with conventional DTC. The effects of the duty ratio controller on the stator flux and current are also presented. As seen there are improvements in the waveforms of duty ratio control compared to the conventional DTC. The following conclusion can be made, Steady State torque ripple is considerably reduced. As from simulation results we can see that the torque ripple present in duty ratio controlled DTC (+3), which is much less as compared to conventional DTC ( 5) It provides fast torque response and better speed

[8]

[9]

[10]

P.C. Krause, Analysis of Electrical Machines and Drive Systems, Prentice Hall, 1985. B.K. Bose, Modern Power Electronics and AC Drives, Prentice Hall, 2002. M.H. Rashid, Power Electronics Circuits, Devices and Applications, Pearson Education, Third Edition, 2004. H.F. Abdul Wahab and H. Sanusi, Simulink model of direct torque control of induction machine, American Journal of Applied Sciences, vol. 5, no. 8, pp. 1083-1090, 2008. Pengcheng Zhu, Yong Kang and Jian Chen. 2003. Improve Direct Torque Control Performance of Induction Motor with Duty Ratio Modulation. Conf. Rec. IEEE-IEMDC03. 1: 994-998. La, K.K., Shin, M.-H., and Hyun, D.-S., Direct torque control of induction motor with reduction of torque ripple, 26th Annual Conference of the IEEE Industrial Electronics Society,Vol. 2, pp. 10821092, October 2000. L.H. Hoang, Comparison of field-oriented control and direct torque control for induction motor drives, ThirtyFourth Industrial Applications Society Annual Meeting/Industry Applications Conference, vol. 2, October 1999, pp. 1245-1252. Mir, S., Elbuluk, M. E., and Zinger, D. S., PI and fuzzy estimators for tuning the stator resistance in direct torque control of induction machines, IEEE Trans. Power. Elect. Vol. 13, No. 2, pp. 279287, May 1998. D. Casadei, G. Serra, and A. Tani, Constant frequency operation of a DTC induction motor drive for electric vehicle, Proc. ICEM 96 Conf. 3, 1996, pp.224-229. P. Satish Kumar, J. Amarnath, and S.V.L. Narasimham, An effective space-vector PWM method for multi-level inverter based on two-level inverter, IEEE International Journal, vol. 2, no.2, pp. 1793-8163, April 2010.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

105

Substation Reliability Evaluation: A Case Study


1
1

Tapsi Nagpal, 2Yaduvir Singh


1 ,2

Thapar University, Patiala


2

tapsiarora@gmail.com,

dryaduvirsingh@gmail.com

Abstract: In power system reliability evaluation, usually component failures are assumed independent and reliability indices are calculated. This paper presents a novel analytical approach for reliability evaluation by incorporating station related outages in composite or bulk system reliability analysis. Station related failures can cause multiple component outages that can propagate to other parts of the network resulting in severe damages. In order to minimize the effects of station related outages on the composite system performance, it is necessary for the designer to assess their effects. This task can be achieved by including station related outages in the composite system evaluation. The Analytical Method based on Mathematical Modeling and problem solving through mathematical formulae was used in this research to assess composite system reliability. The new method described in this paper is used to include station related outages in the reliability evaluation of the system under consideration (substation).This method is relatively simple and can be used to consider multiple component outages due to station related failures in composite system reliability evaluation. In this paper the effects of Station related outages are combined with the connected terminal failure parameters. Keywords - Reliability Evaluation, Mathematical Modeling, Outages, Composite System, Analysis

An overall power system can be divided into the three basic functional zones of generation, transmission, and distribution, and be organized into the three hierarchical levels (HL) shown in Fig1

Fig. 1: Power System Hierarchical Levels

I. INTRODUCTION Electricity is a very effective and flexible form of energy. It can be produced in a variety of ways, delivered efficiently, safely and economically and finally converted to light, heat, power and electronic or other activities. Without it, large industries equipment or small household electronics would not exist. People in modern societies have difficulty appreciating how life would be without electricity. Recent blackouts in North America and in other parts of the world have; however, focused attention on the need for a highly reliable supply of electrical energy the basic function of an electric power system is to supply its customers with electrical energy as economically as possible and with a reasonable degree of continuity and quality. In order to resolve the dilemma between the economic and reliability constraints, design, planning and operating criteria and techniques have been developed and applied in the electric power industry over many years. Most of these criteria are deterministically based and many of them are still in use today. It has, however, been recognized that power systems and their components behave stochastically. The basic weakness of the deterministic criteria is that they do not respond and reflect the probabilistic or stochastic nature of system behavior, of customer demands or of component failures. II. POWER SYSTEM HIERARCHY

Hierarchical Level I (HLI) involves only the generation facilities. Hierarchical Level II (HLII) involves both the generation and transmission facilities. Hierarchical Level III (HLIII) involves all three functional zones. III. PREDICTIVE & PAST PERFORMANCE INDICES HL-II reliability indices can be divided into the two categories of predictive indices and past performance indices. Predictive indices provide information on the future performance of the system and are important planning parameters. Past performance indices indicate the actual system performance and are related to real time operation of the system. Most utilities collect past performance indices in order to assess their composite system performance. Both analytical techniques and Monte Carlo techniques can be used to calculate predictive indices. The concepts associated with Monte Carlo simulation are still developing and more research needs to be done to understand and deal with specific system problems. System outages can occur due to conditions created by multiple component failures. These component outages can be categorized as follows 1. Independent outages, 2. Dependent outages, 3. Common cause or common mode outages and 4. Station originated outages. Data requirements for analyses depend on the kinds of outages considered in the analyses. In the real world, all types of component outages occur and therefore

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

106

consideration should be given to them if they affect the system performance. The present research takes into account only a part of a system i.e. 220 KV Grid Substation and out of Failed IV.
Name of T/Fs

Components, Reliability of transformers, transmission line and shunt capacitors has been calculated that is the Component Based approach has been assimilated in the present research.

DATA COLLECTION FOR RELIABILITY CALCULATION OF TRANSFORMERS


TABLE 4.1: PROBABILITY OF FAILURE FOR TRANSFORMERS DURING THE YEAR 2007-2008

Total Hours the year

during

Hours for which the T/F was off for schedule maintenance

Total duration of outage of T/Fs due to fault

2 220/66KV,100 MVA T/F-I 220/66KV,100 MVA T/F-II 220/132KV,90 MVA T/F-I

3 8760 hrs 8760 hrs 8760 hrs

4 100 hrs 114 hrs 93 hrs

5 59 68 0.51

TABLE 4.2: PROBABILITY OF FAILURE FOR TRANSFORMERS DURING THE YEAR 2008-2009

Name of T/Fs

Total Hours during the year(in hrs) 3 6600

2 220/66kv 100 MVA T/F no.1 ____do_____

Hours for which the T/F was off for schedule maintenance 4 45 hrs

Total duration of outage of T/Fs due to fault 5 -

6600

54 hrs

220/132kv 90 MVA T/F no.1

6600

15.15 hrs

TABLE 4.3 : PROBABILITY OF FAILURE FOR TRANSFORMERS DURING THE YEAR 2009-2010

Name of T/Fs

Total Hours during the year 3 6600 hrs

Hours for which the T/F was off for schedule maintenance 4 11.15 hrs

Total duration of outage of T/Fs due to fault 5

2 220/66kv 100 MVA T/F no.1 ____do_____ 220/132kv 90 MVA T/F no.1

6600 hrs 6600 hrs

60 hrs 15.15 Rs

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

107

V.

DATA COLLECTION FOR RELIABILITY CALCULATION OF TRANSMISSION LINES


TABLE 5.1: PROBABILITY OF FAILURE FOR TRANSMISSION LINES DURING THE YEAR 2007 2008

Name Transmission Lines(T/L)

of

Total Hours during the year

Hours for which the ckt was off for schedule maintenance 4 56.27 hrs

Total of tripping breakdown during the year

Total duration of outlay of lines due to fault

2 220kv D/C PongJalandhar ckt-I 220kv D/C PongJalandhar ckt-II 220kv D/C PongJalandhar ckt-IV

3 8784

5 1

6 35.23

8784

36.45 hrs

00.45

8784

25.04 hrs

10.38

TABLE 5.2: PROBABILITY OF FAILURE FOR TRANSMISSION LINES DURING THE YEAR 2008-2009

Name Transmission Lines(T/L)

of

Total Hours during the year

2 220kv D/C PongJalandhar ckt-I 220kv D/C PongJalandhar ckt-II 220kv D/C PongJalandhar ckt-IV
Name of Transmission Lines(T/L)

3 8784 8784 8784

Hours for which the ckt was off for schedule maintenance 4 56.27 hrs 36.45 hrs 25.04 hrs

Total of tripping breakdown during the year 5 1 2 3

Total duration of outlay of lines due to fault

6 30.00 00.40 10.38

TABLE 5.3: PROBABILITY OF FAILURE FOR TRANSMISSION LINES DURING THE YEAR 2009-2010 Total Hours during the year Hours for which the ckt was off for schedule maintenance Total of tripping breakdown during the year Total duration of outlay of lines due to fault

2 220kv D/C Pong-Jalandhar ckt-I 220kv D/C Pong-Jalandhar ckt-II 220kv D/C DasuyaJalandhar ckt-IV

3 8784 8784 8784

4 56.27 hrs 36.45 hrs 25.04 hrs

5 1 2 3

6 30.00 00.40 10.38

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

108

V.

DATA COLLECTION for RELIABILITY CALCULATION of CAPACITORS


TABLE 6.1: PROBABILITY OF FAILURE FOR CAPACITOR BANKS DURING THE YEAR 2007-2008

Name of Capacitor Bank

Total Hours during the year 3 8760 hrs

2 31.16 MVAR, 132 KV Capacitor Bank installed at Jalandhar S/Stn 19.64 MVAR, 66 KV Capacitor Bank installed at Jalandhar S/Stn

Hours for which the capacitor bank off for schedule maintenance 4 50.00 hrs

Total duration of outage of capacitor bank due to fault 5 23.00

8760 hrs

16.23 hrs

Nil

TABLE 6.2: PROBABILITY OF FAILURE FOR CAPACITOR BANKS DURING THE YEAR 2008-2009

Name of Capacitor Bank

Total Hours during the year 3 8760 hrs

Hours for which the capacitor bank off for schedule maintenance 4 16.23 hrs

Total duration of outage of capacitor bank due to fault 5 Nil

2 31.16 MVAR, 132 KV Capacitor Bank installed at Jalandhar S/Stn 19.64 MVAR, 66 KV Capacitor Bank installed at Jalandhar S/Stn

8760 hrs

33.5 hrs

10.30

TABLE 6.3: PROBABILITY OF FAILURE FOR CAPACITOR BANKS DURING THE YEAR 2009-2010

Name of Capacitor Bank

Total Hours during the year 3 8760 hrs

Hours for which the capacitor bank off for schedule maintenance 4 25.60 hrs

Total duration of outage of capacitor bank due to fault 5 22.4

2 31.16 MVAR, 132 KV Capacitor Bank installed at Jalandhar S/Stn 19.64 MVAR, 66 KV Capacitor Bank installed at Jalandhar S/Stn

8760 hrs

16.23 hrs

VII. RELIABILITY CALCULATIONS Reliability Calculations for Transformers


TABLE 7.1(A): RELIABILITY CALCULATIONS FOR TRANSFORMERS DURING YEAR 2007-2008

Availability Factor (3-4)x100/3 98.85% 98.69% 98.94%

Reliability Factor (3-4-) x 100/(3-4) 99.31% 99.21% 99.99%

TABLE 7.1(B): RELIABILITY CALCULATIONS FOR TRANSFORMERS DURING YEAR 2008-2009

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

109

Availability Factor (3-4)x100/3 99.70% 99.69% 99.81%

Reliability Factor (3-4-5) x 100/(3-4) 99.32% 99.20% 99.77%

TABLE 7.1(C): RELIABILITY CALCULATIONS FOR TRANSFORMERS DURING YEAR 2009-2010

Availability Factor (3-4)x100/3 98.75% 99.82% 99.68% Reliability Calculations for Transmission Lines

Reliability Factor (3-4-5) x 100/(3-4) 99.00% 99.77% 98.89%

TABLE 7.2(A): RELIABILITY CALCULATIONS FOR TRANSMISSION LINES DURING YEAR 2007-2008

Availability Factor (3-4)x100/3 99.35% 99.58% 99.71%

Reliability Factor (3-4-5) x 100/(3-4) 99.60% 100% 100%

TABLE 7.2(B): RELIABILITY CALCULATIONS FOR TRANSMISSION LINES DURING YEAR 2008-2009

Availability Factor (3-4)x100/3 99.19% 99.22% 98.31%

Reliability Factor (3-4-5) x 100/(3-4) 99.48% 100% 99.59%

TABLE 7.2(C): RELIABILITY CALCULATIONS FOR TRANSMISSION LINES DURING YEAR 2009-2010

Availability Factor (3-4)x100/3 99.35% 99.58% Reliability Calculations for Capacitor Banks

Reliability Factor (3-4-5) x 100/(3-4) 99.60% 100%

TABLE 7.3(A): RELIABILITY CALCULATIONS FOR CAPACITOR BANKS DURING YEAR 2007-2008

Availability Factor (3-4)x100/3 99.60% 99.91%

Reliability Factor (3-4-5) x 100/(3-4) 99.70% 100%

TABLE 7.3(B): RELIABILITY CALCULATIONS FOR CAPACITOR BANKS DURING YEAR 2008-2009

Availability Factor (3-4)x100/3 99.91% 99.60%

Reliability Factor (3-4-5) x 100/(3-4) 100% 99.80%

TABLE 7.3(C): RELIABILITY CALCULATIONS FOR CAPACITOR BANKS DURING YEAR 2009-2010

Availability Factor (3-4)x100/3 99.70% 99.81%

Reliability Factor (3-4-5) x 100/(3-4) 99.74% 100%

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

110

VIII. CONCLUSION In this research, Analytical methods, utilizing the mathematical formulations were used to assess composite system reliability. The new method described in this work is used to include station related outages in the reliability evaluation of system under consideration(Substation).This method is relatively simple and can be used to consider multiple component outages due to station related failures in composite system reliability evaluation. In this approach, the effects of station related outages are combined with the connected terminal failure parameters. IX. FUTURE SCOPE The further work incorporates the under listed areas : a) Using the Sequential Monte Carlo Simulation Technique b) Using Minimal Cut Set Method c) Using Failure Mode and Effect Analysis d) Using Capacitor Measurement Technique e) Computing Reliability Indices and Availability Indices of all the equipments installed in the substation f) Encompassing the areas of Generation, Distribution and Transmission of Electric Power REFERENCES
[1]. R. Billinton, R.N. Allan, L. Salvaderi, Applied reliability assessment in electric power systems, IEEE Press, 1991,p71-79 [2]. R. Billinton and W. Li, Reliability evaluation of electric power systems using Monte Carlo simulation method, 1st Edition Plenum Press, New York 1994,p75-80 [3]. J. Endrenyi (Chairman ), M.P. Bhavaraju, Bulk Power System Reliability Concepts And Applications IEEE PES Transactions on Power Systems , Vol. 3, No. 1 , February 1988,p95-101 [4]. R. N. Allan, R. Billinton, S.M. Shahidehpour, C. Singh, Bibliography on the application of probability methods in power system reliability evaluation: 1982-1987, IEEE Transactions on Power Systems, v 3, n 4, Nov, 1988, p 1555-1564 [5]. R. N. Allan, R. Billinton, S. H. Lee (IEEE Power System Engineering Committee, Task Force Subcommittee on Application of Probability Methods);Bibliography on the application of probability methods in power system reliability evaluation, IEEE Transactions on Power Apparatus and Systems, v PAS-103, n 2, Feb, 1984, p 275282 [6]. R. N. Allan, R. Billinton, A.M. Breipohl, C.H. Grigg, Bibliography on the application of probability methods in power system reliability evaluation. IEEE Transactions on Power Systems, v 14, n 1, Feb 1999, p 51-57 [7]. R.N. Allan, R. Billinton, A.M. Breipohl, G.H.Grigg, Bibliography on the application of probability methods in power system reliability evaluation. IEEE Transactions on Power Systems, v 9, n 1, Feb 1994, p 41-49 [8]. R. N. Allan, (IEEE, Subcommittee on the Application of Probability Methods);R. Billinton, S. H. Lee, Bibliography on the application of probability methods in power system reliability evaluation. IEEE Transactions on Power System, 1983, 83 WM 053-6,p86-93 [9]. Adolfo Crespo Marquez,Antonio Sanchez Heguedas Models For Maintenance Optimisation : a study for repairable systems and finite time periods IEEE Transactions on Reliability Engineering and System Safety,Vol-5,October 2001,p92-99

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

111

AGC using Conventional and Fuzzy Logic Controller


1 1, 2
1

Navreet Kaur, 2Rimpika


2

RIMT-IET, Mandi Gobindgarh, Punjab, India


rimpika_grover2002@yahoo.co.in

navreetwaraich@yahoo.co.in,

Abstract: Automatic generation control (AGC) plays a very important role in power system. This work presents the automatic generation control (AGC) of an interconnected two area system. For carrying out AGC for the two areas system conventional and fuzzy logic controller is used and change in generation in both the areas and change in overall frequency of the system is examined using MATLAB. Area Control Error (ACE) is reduced. The aim of the proposed controller is to restore the frequency to its nominal value in the smallest possible time whenever there is any change in the load demand etc. Keywords: Automatic Generation Control (AGC), Area Control Error(ACE)

I. INTRODUCTION When load in the system increases turbine speed drops before the governor can adjust the input. As the change in the value of speed decreases the error signal becomes smaller and the positions of governor valve get close to the required position, to maintain constant speed. However the constant speed will not be the set point and there will be an offset, to over come this problem an integrator is added, which will automatically adjust the generation to restore the frequency to its nominal value. This scheme is called automatic generation control (AGC). The role of AGC is to divide the loads among the system, station and generator to achieve maximum economy and accurate control of the scheduled interchanges of tie-line power while maintaining a reasonability uniform frequency. Modern power system network consists of a number of utilities interconnected together and power is exchanged between utilities over tie lines by which they are connected. The main role of AGC is to maintain the system frequency and tie line flow at their scheduled values during normal period. All generating units on speed governing will contribute to overall change in gyration, irrespective of the location of the load change. Restoration of the system frequency to nominal value requires supplementary control action which adjusts the load reference set point. Therefore the primary objectives of the AGC are to regulate frequency to the nominal value and to maintain the interchange power between control areas at the scheduled values by adjusting the output of selected generators. This function is commonly referred to as load frequency control. A secondary objective is to distribute the required change in generation among the units to minimize the operating costs. Generation in large interconnected power system

comprises of thermal, hydro, nuclear, and gas power generation. Nuclear units owing to their high efficiency are usually kept at base load close to their maximum output with no participation in system automatic generation control (AGC). Gas plants are used to meet peak demands only. Thus the natural choice for AGC falls on either thermal or hydro units. An interconnected power system can be considered as being divided into control areas which are connected by tie lines. In each control area, all generator sets are assumed to form a coherent group. The power system is subjected to local variations of random magnitudes and durations, Hence, it is required to control the deviations of frequency and tie-line power of each control area. A control signal made up of tie line flow deviation added to frequency deviation weighted by a bias factor would accomplish the desired objective. This control signal is known as area control error (ACE).ACE serves to indicate when total generation must be raised or lowered in a control area. In an interconnection, there are many control areas, each of which performs its AGC with the objective of maintaining the magnitude of ACE (area Control Error) sufficiently close to 0 using various criteria. In order to maintain the frequency sufficiently close to its synchronous value over the entire interconnection, the coordination of the control areas actions is required. As each control area shares in the responsibility for load frequency control, effective means are needed for monitoring and assessing each areas performance of its appropriate share in load frequency control. Fuzzy logic implements human experiences and preferences via membership functions and fuzzy rules. According to J. Nanda and A. Mangla in paper Automatic Generation Control of an Interconnected Hydro-Thermal System Using Conventional and Fuzzy Logic Controller the presence of FLC in both areas and small step perturbation in either area or in both areas simultaneously provides better dynamic response than with conventional controller. [1] Li Pingkang and Ma Yongzhen has introduced some new concepts for automatic generation control in the paper Some New Concepts in Modem Automatic Generation Control RealizationNew algorithm, such as feed forward or predictive with intelligent, will improve the control effort, and the interaction between power plant and power system. [4]Manoranjan Paridal; J. Nanda describes the application of AGC in paper Automatic Generation Control of a Hydro-Thermal System in Deregulated Environment with deregulated power system and makes a

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

112

maiden attempt to provide a new practical AGC model . [3] Barjeev Tyagi, Student Member, IEEE, and S. C. Srivastava, Senior Member, IEEE purposed in paper A Decentralized Automatic Generation Control Scheme for Competitive Electricity Markets the design of a decentralized automatic generation control(AGC) scheme for interconnected multiarea power system. [2] AGC of a two area interconnected system is carried out in this work. Change in frequency and generation in each area with load change in any of the two areas is measured. ACE is measured by using conventional method and fuzzy logic. Fuzzy logic is applied to calculate ACE (out) i.e.the control signal provided to each area so that its area control error can be minimized.. When the system frequency drifts from nominal value 50Hz, some frequency sensitive components react to this change. First, the effective load for the power system changes. This process is called load damping and modelled with damping factor D. second, if the system frequency goes beyond the governor dead band (about 35mHz); the governor will act to increase or decrease the output power of generating units. This can be done with the help of governed speed droop R, which is actually the feedback loop gain in the governor. The speed droop is defined as

w pu P

Where w = speed deviation P = output power The value of R determines the steady state speed versus load characteristics of the generating unit as shown in Fig2.1 below. The ratio of speed deviation (wr) or frequency deviation (f) to change in valve/gate position (Y) or output power (P) is equal to R. the parameter R is referred to speed regulation or droop. It can be expressed in percent as Percent R=percent speed or frequency change/percent power output change*100 = [w (nl)-w (fl)]/w0 * 100 Where w (nl) = steady state speed at no load w (fl) = steady state speed at full load w0= nominal or rated speed

generation output is relatively bigger than the loads and the vice versa. This can be understood as a strong natural inertia to stay on equilibrium. These responses are mathematically modelled as D and R respectively and are combined into one frequency response characteristic . AGC in secondary control :An automatic secondary control is conventionally referred to as Automatic Generation Control (AGC). AGC is major control function of secondary control within a utility energy control centre, whose purpose is the tracking of load variations to maintain system frequency at its nominal value, net tie line interchanges to its scheduled values, and optimal generation levels, close to its most economical points. Each control area has to continually determine and monitor system deviation in measured frequency, f and tie line flow, T, to determine the Area Control Error (ACE) as a measure for the secondary control. Area Control Error is defined by ACEi = Pij + Bi *f Where i control area for which ACE is being measured Pij power interchange in areas i and j Bi control area frequency bias coefficient f deviation in frequency ACE, as defined, represents the generation versus load mismatch for the control area. The ACE signal is used in conventional AGC which has a PI control logic. ACE serves to indicate when total generation must be raised or lowered in a control area. A general criterion can be given about which AGC is considered good.The integral of ACE over appropriate time should be small Tie-line bias control is a control philosophy developed for load frequency control in a power system. It is widely used in AGC. It has been proved efficient in maintaining interconnection reliability and its simplicity in control implementation. In Fig:2 below the abbreviations used are PD1 Incremental load change in area 1. PD2 Incremental load change in area 2 R Governor speed regulation parameter. TG Mechanical governor response time, second. T1 turbine time constant.

Fig. 2: Transfer function model of a two area interconnected system

Fig.1: Ideal steady state characteristics of governor with speed droop

These two responses, load damping and governor response; help the system interconnection to be more stable. In short the system frequency goes up when the

B1 Frequency bias constant for area 1. B2 Frequency bias constant for area 2 TP =2H/fD KP=1/D D load damping constant KI Integral gain

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

113

f1 Change in frequency for area 1 f2 Change in frequency for area 2 The power flow on tie line from area 1 to area 2 is

P 12

E1 E2 sin( XT

(1)

The tie line is represented by the synchronising torque coefficient T. synchronizing coefficient T is given by

E1 E2 cos( XT

10

20

The ACE represents the required change in area generation, and its unit is MW. The unit normally used for expressing the frequency bias factor B is mW/0.1Hz. Economic allocation of generators :An important secondary function of AGC is to allocate generation so that each power source is loaded most economically. This function reoffered to as economic dispatch control. The theory of economic dispatch is based on principle of equal incremental costs. II. FUZZY LOGIC CONTROLLER (FLC) Fuzzy Logic Controller (FLC) consists of four main parts: (i)The Fuzzification: Measure the values of input variables (ii) The Knowledge Base: The database provides necessary definitions, which are used to define linguistic control rules and fuzzy data, manipulation in an, FLC. (iii) The Decision Making Logic: It is the kernel of an FLC; it has the capability of simulating human decision making based on fuzzy concepts and of inferring fuzzy control actions employing fuzzy implication and the rules of inference in fuzzy logic. (iv) The Defuzzification: Defuzzification, which yields a non-fuzzy, control action from an inferred fuzzy control action. Algorithm for fuzzy logic application to AGC problem 1. Calculate area control error (ACE) and change of frequency (delF). 2. Convert the error and change of frequency into fuzzy variables i.e. linguistic variables such as PS,PL etc. 3. Evaluate the decision rules shown in rule base given below using the compositional rule of inference. 4. Calculate the deterministic input required to regulate the process. The control rules are formulated in linguistic terms using fuzzy sets to describe the magnitude of error, the frequency deviation and the magnitude of the appropriate control action.
TABLE :1 RULE BASE (WITH THREE MEMBERSHIP FUNCTIONS)

(2)

A positive P12 represents an increase in power transfer from area 1 to area2. For a total load change of

w1

w2

PL (3) (1/ R 1 1/ R2 ) ( D1 D2 )

consider the steady state values following an increase in area 1 load by PL1. for area 1 Pm1 - P12 - PL1 = f D1 (4) and for area 2 Pm2 + P12 = f D2 (5) The change in mechanical power depends upon regulation. Hence,

Pm1

f R1 P 12 D2 )

and

Pm 2

f R2
(7)

(6)

substitution of equation (6) in eq. (4) and eq. (5) yields

f(
and

1 R1 f(

D1 ) 1 R2

PL1 P 12 PL 2
PL
1 2
2 2

(8)

solving eq. (7) and eq. (8)

PL (1/ R 1 1/ R2 ) ( D1 D2 ) PL (1/ R2 D2 ) P12 (1/ R 1 D1 )(1/ R 2 D2 ) f

(9) (10)

PL1
1

where 1 and 2 are the composite frequency response characteristics of area1 and area2 respectively . An increase in area 1 load by PL1 results in a frequency reduction in both areas and a tie line flow of P12. A negative P12 is indicative of flow from area 2 to area1. the tie line flow deviation reflects the contribution of the regulation characteristic (1/R+D) of one area to another. Similarly for a load change in area 2 by P 12

delF/ACE PS NS ZE PS NS ZE PS NS PS NS ZE ZE ZE PS NS Membership function used is the triangular. Method used for defuzzification is centroid method.

f
1

PL
2

(11)

And

P 12

P21
1

PL 2

1 2

(12) (13) (14)


Fig. 3: Subsets defining delF

Area control error for area 2 ACE2 = P12 + B2f where B2


2

1 R2

D2

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

114

Generation capacity Frequency Load change

500 50 30MW For area 1 -0.00998 49.960 19.76 49.019 0.15 -10.88 -10.07

500 50 0MW For area2 -0.00998 49.960 9.8 40.019 0.19 -10.99 -10.07

TABLE: 6 RESULTS WITH DATA SET 2

Fig.:4: Subset defining ACE TABLE 2: RULE BASE (WITH 5 MEMBERSHIP FUNCTIONS)

delF/ACE NL NL NL NS NL ZE NS PS ZE PL ZE Where NL = negative large NS = negative small ZE = zero PS = positive small PL = positive large

NS NL NL NS PS ZE

ZE NS NS ZE PS PS

PS NS ZE PS PL PL

PL ZE ZE PS PL PL

Bias factor Frequency deviation New frequency Change in generation(MW) Change in load (MW) Area control error Net interchange power

III. RESULTS Area Control Error measured = -10.99 MW.After applying fuzzy logic (using 3 membership functions) this value is reduced to -8 MW. And applying fuzzy logic (using 5 membership functions this value is reduced to - 7.5MW.
TABLE 3: DATA SET 1

(Yellow for delF1, pink for delF2) Fig. 5: Frequency deviation step response for area 1 and area 2 with conventional controller

Regulation Hz/pu MW Damping constant pu/MW Hz Generation capacity Frequency

For area 1 2.4 0.00833 2000 50MW

For area2 2.4 0.00833 2000 50MW


(Yellow for delF1, pink for delF2) Fig. 6: Frequency deviation step response for area 1 and area 2 using fuzzy logic Controller

TABLE 4: RESULTS WITH DATA SET 1

Bias factor Frequency deviation New frequency Change in generation(MW) Change in load (MW) Area control error Net interchange power

For area 1 0.425 -0.100007 44.9997 49.019 0.8331 -50.050 -50

For area2 0.425 -5.000333 44.9997 40.019 0.8336 -50.04253 -50

TABLE: 5: DATA SET2

(yellow for Pm1, pink for Pm2, green for Ptie)

Regulation Hz/pu MW Damping constant pu/MW Hz

For area 1 0.01 0.08

For area2 0.02 1

Fig. 7: Power deviation step response for area 1 and 2, and Ptie using conventional controller

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

115

(Yellow for Pm1, pink for Pm2, green for Ptie) Fig. 8: Power deviation step response for area 1 and 2, and Ptie using fuzzy logic controller

IV.CONCLUSIONS 1. Performing the AGC of two interconnected areas change in frequency with load change in any of the two areas. How much power the two areas have to interchange to keep the frequency within permissible limits. 2. Area control is measured -10.08 and it is reduced to -8 by applying fuzzy logic to AGC problem in a two area system. 3. And frequency response for step deviation in a two area system is better by using FLC in comparison to conventional controller. Future Scope: The main objectives of AGC is to regulate frequency to specified nominal value and to maintain the interchange power between control areas. The future scope of this work is that AGC can be carried out for more than two areas. ACE can be reduced more by applying some other AI technique also. REFERENCES
[1] [2] A J Wood and B F Wollenberg. Power Generation, Operation and Control.J John Wiley & Sons, 1984 Mangla and J. Nanda , Automatic Generation Control of an Interconnected Hydro-Thermal System Using Conventional Integral and Fuzzy Logic Controller international conferencr on electrical utility, deregulation, destructuring, and power technologies,pp372-377, April 2004. Barjeev Tyagi and S. C. Srivastava A Decentralized Automatic Generation Control Scheme for Competitive Electricity Markets. IEEE transactions on power systems, vol. 21, no. 1, pp 312-320,

[3].

February 2006. [4]. J. Nanda ; Manoranjan Parida; Automatic Generation Control of a Hydro- Thermal System in Deregulated Environment pp 942-947 [5]. Li Pingkang Beijing and Ma Yongzhen Some New Concepts in Modem Automatic Generation Control Realization.pp1232-1236, 1998 [6]. P. Kundur, Power System Stability and Control. Palo Alto, CA: Mc- Graw-Hill, 1994 [7]. George Gross, Fellow, IEEE and Jeong Woo Lee Analysis of Load Frequency Control Performance Assessment Criteria. IEEE transactions on power systems, vol. 16, no. 3, pp 520-525 August 2001.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

116

Effect of Various Parameters of Gas Insulated Bus Duct on Electric Field Intensity
Nagbhushan1, J. Amarnath2, D. Subbarayudu3
PDA College of Engg, Gulbarga. Karnataka India Jawaharlal Nehru Technological University, A.P. India 3 G.Pulla Reddy Engineering College, Kurnool, A..P. INDIA
2
1

nappdaceg@yahoo.co.in

Abstract: Gas insulated substations (GIS) have received great interest in the recent years. One of the advantages of GIS over conventional substations is their compactness which makes them a favourite for service in urban residential areas. These consists conductors supported on insulator inside on enclosure filled with sulphur hexafluoride (SF6) gas. The voltage withstands capability of SF6 bus duct is strongly dependent on field perturbations such as those caused by conductor surface imperfections and by conducting particle contaminants. These acquire charge due to the electric field intensity. Nearly 50% of GIS failures are due to metallic conducting particles lifted by electric field and migrate to conductor or insulator initiating breakdown at voltages significantly below the insulation characteristics. This paper analyses the electric field intensity for different dimensional conductor and enclosures at various applied voltages in a three phase Gas Insulated Bus duct. Electric field intensity decreases with increase in diameter of enclosure. Keywords Bus duct, Image charge, Field intensity

system like breakers or disconnectors etc. Aluminium, copper and silver are the types of metal particles. GIS also suffer from certain drawbacks. One of them is outage due to conducting particles which accounts or nearly 50% of GIS failures. The particles can be lifted by the electric field and migrate to the conductor or insulators where they initiate breakdown at voltages significantly below the insulation characteristics. These particles move randomly in a horizontally mounted Gas Insulated Bus duct (GIB) system due to the electric field, and this movement plays a crucial role in determining the behaviour of GIS. Under 50 Hz AC voltages, the particle motion is complex, and under appropriate conditions, the particle may cross the gaseous gap from the low field region near the outer enclosure to the high field region near the central electrode. In order to know behaviour of particles analysis of the electric field intensity was made for different dimensional conductors, enclosures at various applied voltages. II. GIB MODELLING TECHNIQUE The Fig-1 shows the cross sectional view of typical horizontal three phase bus duct. The enclosure is filled with SF6 gas at high pressure. A particle is assumed to be at rest on the enclosure inner surface, just beneath the bus bar until a voltage sufficient enough to lift the particle and move it in the direction of field is applied. After acquiring an appropriate charge in the field, the particle lifts and begins to move in the direction of field having overcome the forces due to its own weight and drag due to the viscosity of the gas. The particle motion largely depends on applied voltage. Under AC voltages, for a wire particle of given radius, the activity increases with particle length since the particle charge-to-mass ratio at lifting increases with length.

I. INTRODUCTION The Electric power industry worldwide has some 35 years of manufacturing and field experience with Gas Insulated Substations (GIS). The advantages of GIS over conventional substations are their compactness, reliable and maintenance free operations. Basic components of GIS bay are circuit breakers, disconnectors, earthing switches, bus ducts, current and voltage transformers etc. The inner live parts of GIS are supported by insulators called spacers, which are made up of alumina filled epoxy material. The GIS enclosure forms an electrically integrated, ground enclosure for the entire substation. The enclosure is filled with sulphur hexafluoride (SF6) gas. It has excellent dielectric and heat transfer properties. Sulphur Hexafluoride (SF6) gas is the electric power industrys preferred gas or electric insulation and especially for arc quenching current interruption equipment used in transmission and distribution of electrical energy. The usefulness of SF6 gas is mainly due to its high dielectric strength, good thermal stability, economically inert, non-flammable, non-corrosive and non-toxic. Even though SF6 exhibits very high dielectric strength, the withstand voltage of SF6 within GIS is drastically reduced due to the presence of metallic particles on the inner surface of the enclosure. The origin of these particles may be from the manufacturing process, from mechanical vibrations or from moving parts of the

Fig. 1: Three Phase Bus Duct

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

117

A conducting particle motion is an electric field will be subjected to a collective influence of several forces. These forces may be divided into electrostatic force (F e), gravitational force (mg) and drag force (Fd). The motion equation is given by

C k

d2y m 2 dt

R 1
1

R 1 R2

Fe

mg

Fd

R 2

A 1 50 h P
0

Where m = mass of particle, y = displacement in vertical direction, g = gravitational constant. Here, (i) Fe
E2
2 2

E 3

K .q.E (t )
l 2 E (t 0 ) 2l ln l r
0
E2

K = Correction factor <1, and charge q is,


q
D
d

E 3

Fig. 2: Three phase GIB with image charge conductors

E (t 0 ) is ambient electric field at t=t0 Vm sin t Electric field, E (t ) R R0 y (t ) ln 0 Ri


Where R 0 are Enclosure radices and R i is conductor radius. y(t) = position of particle which is vertical distance from the surface of enclosure towards inner electrode. (ii) mg

Let the particle moves to a distance x from the inner surface of the enclosure at point P, then from Fig (2). From triangle PAB

R1

(h

x) 2

K2

2(h

x) K . cos(150 0 )

The electric field intensities for different conductors A, B, C taking into account their image charge effect, are E 1, E2 and E3 respectively.
V1 E1 1 h x 2 ln h 2h r
2

1 x

r 2 l g where,

1.68 x10 8 for Cu.

V/mm

(iii) Fd = Drag force=

0.5 y. .r[6 K d ( y) 2.6560[ g .l. y] Where y is velocity, is viscosity, r is particle radius. g is gas density, l is length of particle, K d ( y ) is drag
coefficient. Electric Field Intensity in 3-phase GIB (including image charges):

V2 E2

c os R1 2 ln

Cos R2 2h r

V/mm and

V3

Fig (2) shows three phase GIB with image charge conductors.

E3

c os R1 2 ln

Cos R2 2h r

V/mm.

The total electric field intensity E at point P is

E
E

E1

E2

E3
cos R1
2

1 1 1 V1 2h h x h x 2 ln r

cos R2

V2 V3

volts per mm, where, V1 = Vm sin t, V2 = Vm Sin ( t+120), and

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

118

V3 = Vm sin ( t+240).
III RESULTS AND DISCUSSION Simulation has been carried out for electric field intensity at different level of voltages and for different dimension of enclosure and conductor. The distance between the conductor is taken as K=215mm. Table 1 shows the variation of field intensity for different values of dimeter of enclosure and diameter of conductor at various values of maximum voltages and fixed values of x=1mm.as shown in Fig2. A conducting particle moving in an external electric field will be subjected to a collective influence of electrostatic force, Gravitational force and Drag force IV. CONCLUSION An attempt has been made to simulate the electric field intensity in a 3-phase isolated GIB on bare electrodes system considering the effect of field due to image charge.. It is seen that the field intensity is a function of dimensions of enclosure and conductor. The electric field intensity is in linear relation with diameter of conductor and it decreases with an increase in the diameter of enclosure resulting in reduction of rate of motion of particle. REFRENCES
[1] 1. G.V. Nageshkumar, J. Amarnath, B.P. Singh, K.D. Srivastava, Electric Field Effect on metallic particle contamination in a common enclosure Gas Insulated Busduct, IEEE Transactions on dielectrics and Electrical insulation, Vol. 14, No.2, April 2007. 2. M. Venugopal Rao, G.V. Nageshkumar, J. Amarnath, S. Kamakshaiah, Effect of various excitations on the performance of Gas Insulated Substations with metallic particle contamination World Journal of Modelling and Simulation, Vol.4 (2008), No.4, pp. 305-311 3. M. Ramya Priya, G.V. Nageshkumar, J. Amarnath, R. Prabhadevi, Effect of various design parameters of Gas Insulated Busduct in the performance of Gas Insulated Substations, International Conference on Control, Automation, Communication and Energy Conservation, 4th-6th June 2009. 4. K.B. Madhusahu, J. Amarnath, K.S.Lingamurthy, B.P.Singh, Simulation of Particle Movement in a single phase isolated Gas Insulated Busduct with image charge effect, 2008 International Conference on High Voltage Engineering and Applications, Chongquing, China, November 9-13, 2008. 5. M. Venugopala Rao, J. Amarnath, S. Kamakshaiah, Particle Trajectories in a three phase common enclosure Gas Insulated Busduct with Monte Carlo Technique, ARPN Journal of Engineering and Applied Sciences, Vol.3, No.6, December 2008. APPENDIX I TABLE 1: SHOWS, THE VARIATION OF FIELD INTENSITY FOR DIFFERENT VALUES OF DIAMETER OF ENCLOSURE (D) AND DIAMETER OF CONDUCTOR (D) FOR VM 300KV, 400 KV, 600 KV, 800 KV AND 1000KV .

Vm=800KV K=215mm,x=1mm K=215mm,x=1mm h=1mm,d=100mm h=125mm D(mm) E(KV/mm) d E (mm) (KV/mm) 300 7.31E-03 10 2.85E-03 400 5.19E-03 20 3.47E-03 500 4.41E-03 30 4.43E-03 600 3.98E-03 40 4.00E0-3 700 3.69E-03 50 4.86E-03 800 3.49E-03 60 5.27E-03 1000 3.21E-03 70 5.69E-03 1500 2.82E-03 80 6.10E-03 2000 2.61E-03 90 6.52E-03 Vm=300KV K=215mm,x=1mm K=215mm,x=1mm h=1mm,d=100mm h=125mm D E(KV/mm) d E(KV/mm) (mm) (mm) 300 2.74E-03 10 1.07E-03 400 1.95E-03 20 1.30E-03 500 1.65E-03 30 1.49E-03 600 1.49E-03 40 1.66E-03 700 1.39E-03 50 1.82E-03 800 1.31E-03 60 1.98E-03 1000 1.20E-03 70 2.13E-03 1500 1.06E-03 80 2.29E-03 2000 9.78E-04 90 2.45E-03 Vm=400KV K=215mm,x=1mm K=215mm,x=1mm h=1mm,d=100mm h=125mm D E(KV/mm) d E(KV/mm) (mm) (mm) 300 3.66E-03 10 1.43E-03 400 2.60E-03 20 1.99E-03 500 2.21E-03 30 1.99E-03 600 1.99E-03 40 2.21E-03 700 1.85E-03 50 2.43E-03 800 1.75E-03 60 2.64E-03 1000 1.75E-03 70 2.84E-03 1500 1.41E-03 80 3.05E-03 2000 1.41E-03 90 3.26E-03

[2]

[3]

[4]

[5]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

119

Vm=600KV K=215mm,x=1mm K=215mm,x=1mm h=1mm,d=100mm h=125mm D(mm) E(KV/mm) d(mm) E(KV/mm) 300 5.48E-03 10 2.14E-03 400 3.89E-03 20 2.61E-03 500 3.31E-03 30 2.98E-03 600 2.98E-03 40 3.32E-03 700 2.77E-03 50 3.64E-03 800 2.62E-03 60 3.95E-03 1000 2.41E-03 70 4.26E-03 1500 2.12E-03 80 4.58E-03 2000 1.96E-03 90 4.89E-03 Vm=1000KV K=215mm,x=1mm K=215mm,x=1mm h=1mm,d=100mm h=125mm D(mm) E(KV/mm) d E(KV/m (mm) m) 300 9.14E-03 10 3.57E-03 400 6.59E-03 20 4.34E-03 5.5133E500 03 30 4.96E-03 600 4.97E-03 40 5.53E-03 700 4.62E-03 50 6.07E-03 800 4.36E-03 60 6.59E-03 1000 4.01E-03 70 7.11E-03 1500 3.53E-03 80 7.63E-03 2000 3.26E-03 90 8.15E-03

Fig. 3. (a) to (h): Shows, the variation of Field intensity for different values of diameter of enclosure (D) and diameter of conductor (d)

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

120

Role of FACTS Devices in Distribution System


1

Parag Nijhawan, 2R.S. Bhatia, 3D.K. Jain


1

Thapar University, Patiala (India) National Institute of Technology, Kurukshetra (India) 3 Deenbandhu Chhotu Ram State University of Science & Technology, Muruthal (India)
2
1

parag.nijhawan@rediffmail.com 2 rsibhatia@yahoo.co.in 3 jaindk66@gmail.com

Abstract: This paper presents the need of custom power devices in a distribution network. In this paper. Converter based custom power devices are considered viz. Distribution Static Synchronous Compensator (DSTATCOM), Dynamic Voltage Restorer (DVR) and Unified Power Quality Compensator (UPQC). The complete literature review on types, modelling and applications related to these devices is compiled. Keywords: Custom power devices, distributed generation, distribution static synchronous compensator (DSTATCOM), dynamic voltage restorer (DVR), unified power quality compensator (UPQC)

collapse. These conditions lead to a loss of revenue and utility integrity. For these problems, power engineers have proposed several methods such as feeder reconFiguration, capacitor switching for VAR control, series voltage regulator and dispersed generation installation. II. FACTS DEVICES IN DISTRIBUTION SYSTEM Initially, FACTS devices which possess the latest power electronic devices and methods to control electronically the power system parameters were designed and installed on high-voltage side i.e., transmission side of the power system network. But now FACTS devices are also used in low-voltage distribution side of the power system network rated 1 through 38 kV and are also known as Custom Power Devices like Static series compensator (SSC) commercially known as Dynamic voltage restorer (DVR), High speed mechanical transfer switch (HSMTS), Advanced static VAR compensator (ASVC), Distribution STATCOM (D-STATCOM), Unified series-shunt controller (USSC) or Unified power quality conditioner (UPQC). With the increase in application of the nonlinear and electronically switched devices in distribution systems, the problems of power quality such as voltage fluctuations, flicker, harmonics and asymmetries of voltages are becoming more serious. Not only this, but lightning strikes on transmission lines, switching of capacitor banks and various network faults also result in the power quality problems such as transients, voltage sags, voltage swells and interruptions. Also, modern hightech consumer products are more sensitive to these power quality problems than before. As a result, there is a stringent need of better power quality supply. FACTS devices or Custom Power devices have all the capabilities to solve this power sector problem. According to the IEEE 1409 Custom Power Task Force, the number of custom power devices in service up to May 2000 is as follows: Transfer switches: 23 devices Static series compensators: 7 devices Static shunt compensators: 17 devices The interest in the usage of power quality devices (PQDs) arises to meet the everyday growing sensitivity of customer needs and expectations of having good quality power. Poor Power quality levels can cause costly downtimes and customer dissatisfaction. According to

I. INTRODUCTION Electricity is produced and delivered to customers through generation, and transmission and distribution systems, respectively. On the basis of reliability; generation, transmission, and distribution are referred to as functional zones of power system. Generation facilities must produce adequate power to meet customers demand. Transmission systems must transport bulk power over long distances without overloading or jeopardizing system stability. Distribution systems must deliver electric power to each customers premises from bulk power systems with an acceptable level of reliability and voltage constraint. Moreover, with the deregulation of the electric power energy market, power quality is of great concern to both electric utilities and consumers. Earlier the major focus of power system reliability evaluation has been mainly concentrated on generation and transmission as more capital cost is involved in these. But now distribution systems have begun to receive more attention for reliability assessment. The reason behind this is that the analysis of the customer failure statistics of most utilities shows that the distribution system makes the greatest individual contribution to the unavailability of supply to a customer. Also, the electrical distribution network failures account for approximately 90% of the average customer interruptions. In case of an contingency event in the distribution systems, the loads may be interrupted; lines may be overloaded or disconnected due to internal or external influences. In an extreme case, it may even lead to the entire or large part of the power system network to

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

121

Contingency Planning Research Companys annual study, downtime due to power disturbances results in major revenue losses. So, one needs to understand the steady-state and transient modelling of these custom power devices, operating conditions under which the compensated/uncompensated system is to operate and the behaviour of the system with and without custom power device(s) on occurrence of any abnormal condition in the power system network. III. LITERATURE SURVEY Reliable electric power is necessary to meet the customer loads without interruptions in supply voltage. Power quality and power quality problems [1-5] are hot topics for both suppliers and consumers with the deregulation of the electric power energy market [6-7]. The electrical distribution network failures account for approximately 90% of the average customer interruptions. The custom power devices [8-19] improve the power quality levels to meet the everyday growing sensitivity of customer needs and expectations and reduce the revenue loss. There are two categories of custom power devices Network ConFiguring type and Compensating type [20]. The former changes the conFiguration of power system network like solid state current limiter (SSCL), solid state circuit breaker (SSCB) and solid state transfer switch (SSTS). These devices use GTO or thyristors. These second category of devices are converter based. DSTATCOM, DVR and UPQC are compensating type custom power devices and are converter-based. Out of the number of power quality solutions possible, one needs to select the type and the rating of the custom power solution [21-22]. III. A. Dstatcom DSTATCOM [19], [28-70] is connected in shunt with the distribution system network as STATCOM [23-27] is used in transmission system. The structure and operating modes of DSTATCOM are shown in Figure-2. It can operate in voltage control mode [68], [43-44] or current control mode [44]. It improves voltage regulation [28], [40], [42], power factor, harmonic filtering [63] and load balancing/compensation [59-60]. It can also mitigate voltage sag/swell [57], voltage dip [64], voltage flicker [57-58] and voltage fluctuation [65] when connected to a distribution system.

Fig. 1: Structure and Operating Modes of D-STATCOM

[29] presents the modeling of DSTATCOM. [30-33], [55] discuss the analysis of the control strategies used in DSTATCOM. A comparative study of control algorithms for DSTATCOM for Load Compensation is also reported in literature [58]. At medium voltage level, 3-phase, 3wire DSTATCOMs as shown in Figure-2 are developed with voltage fed type having single stage VSI [35-36] but for higher voltage levels with higher power handling capacity, multilevel VSI [37-38], cascaded multilevel VSI [39], cascaded H-bridge [54], [59] and multipulse VSI [52], [54],[70] are used. 3-phase, 4-wire DSTATCOMs [46-48] are also developed. [46] presents the analysis, simulation and control of DSTATCOM in three-phase, four-wire isolated distribution systems. [47] presents the neural network based control of 3-phase, 4-wire DSTATCOM. [48] presents the three dimensional PWM algorithm based 3phase, 4-wire DSTATCOM. Modified double modulation signal PWM control of DSTATCOM using five level double-converter is presented in [56].

Fig. 2: Three-Phase 3-Wire D-STATCOM

DSTATCOM models for unbalanced and non-linear loads alongwith their control strategies are reporte in literature [50-52]. [66] presents the application of DSTATCOM for distributed generation.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

122

IIIB. Dynamic Voltage Restorer (Dvr) DVR [19], [28-31], [70-92] is a series connected device. The basic function of the DVR is to inject a dynamically controlled voltage (VDVR) generated by a forced commutated converter in series to the bus voltage by means of a booster transformer [76-78] as shown in Figure-3. The momentary amplitudes of the three injected phase voltages are controlled such as to eliminate any detrimental effects of a bus fault to the load voltage VL. To be able to mitigate voltage sag, DVR should be capable of generating and absorbing active and reactive power to/from the grid. To correct a given voltage dip, in addition to voltage, active and/or reactive power injections are also needed. DVR itself is capable of generating the reactive power, but the injected active power must come from the energy storage part of DVR.

Fig. 4: Single-Phase 2-Wire DVR

The conFiguration and control scheme of DVR for voltage dip mitigation are reported in literature [8490]. The circuit conFigurations of single-phase, 2-wire DVR; 3-phase, 3-wire DVR and 3-phase, 4-wire DVR are shown in Figure-4, Figure-5 and Figure-6 respectively. [88] presents the DVR for voltage sag mitigation supplying non-linear loads and [89] presents the DVR for voltage sag mitigation with zero sequence injection capability.

Fig. 3: Block Diagram of DVR

Dynamic voltage restorer installed in distribution system also enhances the power at sensitive facilities [80]. The dynamic voltage restorer (DVR) is used in power distribution networks to protect consumers from sudden sags (and swells) in grid voltage by compensating the injection voltage [74-75]. When implemented at medium voltage level, the DVR can also be used to protect a group of medium voltage or low voltage consumers. The basic functions of a controller in a DVR are: the detection of voltage sag/swell events in the system; calculation of the correcting voltage, generation of trigger pulses to the sinusoidal PWM based DC-AC inverter, correction of any anomalies in the series voltage injection and termination of the trigger pulses after the event has passed. The controller also shifts the DC-AC inverter into rectifier mode to charge the capacitors in the DC energy link in the absence of voltage sags/swells. [72-73] reported a new concept of interline dynamic voltage restoration (IDVR). In IDVR, two or more DVRs in different feeders are connected to a common dc link. One of the DVRs compensates for voltage sag, while other DVRs connected to a common dc Link replenish the dclink energy storage. A current mode control strategy is incorporated into the IDVR system.

Fig. 5: Three-Phase 3-Wire DVR

A reliability assessment algorithm for distribution systems using a Static Series Voltage Regulator (SSVR) is presented in [92]. The best location of the SSVR in distribution systems is determined based on different reliability indices. The algorithm proposed in [89] is efficient for large-scale radial distribution systems and also accommodates the effects of fault isolation and load restoration. [93] presents the positive and revolutionary impact of the existing DG in assisting the dynamic voltage restorer (DVR) in mitigating severe transient problems such as different types of voltage sags. In other words, the union of DG and DVR results in reduction in cost and rating of DVR.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

123

Fig. 6: Three-Phase 4-Wire DVR

IIIC. Unified Power Quality Compensator (Upqc) UPQC can be considered as a combination of DVR and D-STATCOM. It is a series-shunt device as shown in Figure-7. UPQC [96-112] compensates for a variety of power-quality problems in a distribution system including sag compensation, flicker reduction, unbalance voltage mitigation, and power-flow control. UPQC/USSC used in distribution systems has a similar structure to UPFC [23] used in transmission system. It can inject both series voltage and shunt current to the system. The advantage of UPQC is that it can perform both load compensation and voltage regulation at the same time.

source voltage or load current are unbalance or distort, an ac equivalent proportional-integral controller in p-q-r axis is reported in [102] which eliminates steady-state error to zero. The shunt component of a UPQC (Unified Power Quality Conditioner) uses a VSI (voltage source inverter), which is connected to the grid through a series interface inductor. This inductor reduces the switching harmonics injected by the shunt VSI into the distribution network. In case the UPQC is connected to a weak supply point, LC or/and LCL filters are used for interfacing the shunt VSI with the distribution network in order to prevent the switching frequencies generated by the shunt inverter entering into the grid. Power quality (PQ) is basically a service and many customers are ready to pay for it. In near future, distribution system operators could decide to supply their customers with different PQ levels and at different prices. A new device named as OPEN unified power quality conditioner (UPQC) can fulfill this role, is proposed in [109]. It is composed of a series main unit installed in the medium-voltage/low-voltage (LV) substation, along with several shunt units connected close to the consumers. These series and parallel units do not have a common dc link. This device can achieve general improvement in PQ, for all customers that are supplied by the mains (PQ) by using only the series unit. Additional incremental improvement in PQ (i.e., mains power interruptions), can be provided to the customers who need it (custom power) by the individual shunt units. [110] proposes a control of series compensator (SERC) of the UPQC on the basis of injecting voltage in quadrature advance to the supply current. As a result, the SERC consumes no active power at steady state. The other advantage of this scheme is that the SERC can share the lagging VAR demand of the load with the shunt compensator (SHUC). This type of UPQC is termed as UPQC-Q. IV. CONCLUSIONS The custom power devices are necessary to improve the power quality levels to meet the modern customer loads which are more sensitive to power quality problems and have complex micro-processor based controls. These devices help in reducing the large costly downtimes and the revenue loss. For this, a power engineer should know about the capital cost investment for the installation of the proposed device(s) or system(s), expected saving in the operating cost of the system, life expectancy of the proposed equipment(s), optimal placement of the device(s) [113] and energy requirement for distributed energy resources with battery energy storage for voltage support in 3-phase distribution lines [114].

Fig. 7: Block Diagram of UPQC

[99] presents the control strategy for three-phase threewire utilities. This control strategy of parallel active filter is based on combination of extended p-q theory and instantaneous symmetrical components theory while the control circuit of series active filter is based on instantaneous symmetrical components theory. [100] presents the control strategy for three-phase three-wire systems. This UPQC device combines a shunt-active filter together with a series-active filter in a back-to-back conFiguration. [101] introduces a UPQC based on improved p-q-r theory. The UPQC applying such control strategy is used for the compensation of the nonlinear and unbalance three-phase four-wire system. For UPQC, when

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

124

[1]

[2] [3]

[4]

[5] [6]

[7]

[8]

[9]

[10] [11]

[12]

[13] [14]

[15] [16]

[17]

[18]

[19]

[20]

[21]

[22]

REFERENCES Roger C. Dugan, Mark F. McGranaghan, Surya Santoso and H. Wayne Beaty, Electrical Power Systems Quality, 2nd edition, McGraw Hill, 2004. C. Sankaran, Power Quality, CRC Press, 2002. Arindam Ghosh, Gerard Ledwich, Power quality enhancement using custom power devices, Massachusetts, USA: Kluwer Academic Publishers, 2002. J. Stones and A. Collinson, "Power Quality in Modern Power Systems: Overview and Trends," Power Eng. Journal, vol.15, pp.58-64, April 2001. M. H. J. Bollen, "What is power quality?," Electric Power Systems Research, vol.66, pp.5-14, July 2003. J. Arrillaga, M. H.J. Bollen and N.R. Watson, "Power quality following deregulation," in Proc. IEEE, Vol. 88, pp.246-261, Feb.2000. R. G. Koch, P. Balgobind and E. Tshwele, "New developments in the management of power quality performance in a regulated environment," in Proc. IEEEAFRICON'02, vol.2, pp.835-840, Oct.2002. D. Daniel Sabin and Ambra Sannino, A summary of the draft IEEE P1409 custom power application guide, IEEE PES Transmission and Distribution Conference and Exposition 2003, Vol. 3, pp. 931-936, 7-12 Sept. 2003. The Electricity Technology Roadmap Initiative. A Guide for Developing the Science and Technology of the Future, EPRI, 1999. Contingency Planning Research (CPR) and Contingency Planning & Management Magazine (CPM), 2001 Cost of Downtime. Alaxander Domijan Jr., Alejandro Montenegro, Albert J.F. Keri and Kenneth E. Mattern, Custom Power Devices: An interaction study, IEEE transactions on Power Systems, Vol. 20, No. 2, pp. 1111-1118, May 2005. M. Emin Meral, Ahmet Teke, K. Cagatay Bayindir and Mehmet Tumay, Power quality improvement with an extended custom park, Electric Power Systems Research 79, pp. 1553-1560, 2009. Custom Power-State of the Art Cigre WG14-31, 2000. C. Alvarez, J. Alamar, A. Domijan Jr., A. Montenegro, and Song, "An investigation toward new technologies and issues in power quality," in Proc. 9th Int. Conf Harmon. Qual. Power, vol. 2, pp. 444-44, 2000.. N. G. Hingorani, "Introducing custom power," Proc. IEEE Spectrum, vol.32, pp.41-48, June1995. M. L. Crow, "Power quality enhancement using custom power devices", IEEE Power and Energy Magazine, vol.2, pp.50, April 2004. M. D. Stump, G.J. Keane and F.K.S. Leong, "The role of custom power products in enhancing power quality at industrial facilities," in Proc. Conf on Energy Management & Power Del., pp.507-517, March 1998. M. M. Osborne, R. H. Kitchin and H.M. Ryan, "Custom power technology in distribution systems: an overview," in Proc. IEE Symp. On Reliability, Security and Power Quality of Distribution System, pp. 10/1-l0/11, April 1995. T. Devaraju, V.C. Veera Reddy and M. Vijay Kumar, Modeling and Simulation of Custom PowerDevices to Mitigate Power Quality Problems, International Journal of Engineering Science and Technology, vol. 2(6), pp. 1880-1885, 2010. Yashpal, A. Swarup and Bhim Singh, A review of compensating type custom power devices for power quality improvement, Joint IEEE International and IEEE Power India conference on Power System technology (POWERCON 2008), pp. 1-8, 12-15 Oct. 2008. Dong-Jun Won, I1-Yop Chungm Seon-Ju Ahn and Seung-I1 Moon, A novel method to determine the kind and rating of power quality solutions, IEEE conference on Power Tech 2005, Russia, pp. 1-5, 27-30 June 2005. A. Robert, "Power Quality Monitoring at the Interface Between Transmission System and Users," Proceedings of Ninth International Conference on Harmonics and Quality of Power, vol. 2, pp. 425-430, 1-4 Oct. 2000.

[23] Yong Hua Song and Allan T. Johns, Flexible ac transmission systems, The Institution of Electrical Engineers London, 1999. [24] Yankui Zhang, Yan Zhang, Bei Wu and Jian Zhou, Power injection model of STATCOM with control and operating limit for power flow and voltage stability analysis, Electric Power Systems Research 76, pp. 1003-1010, 2006. [25] M.A. Hannan, A.Mohamed, A. Hussain and Majid Al-Dabbagh, Power quality and analysis of STATCOM using dynamic phasor modeling, Electric Power Systems Research 79, pp. 993-999, 2009. [26] Claudio A. Canizares, Massimo Pozzi, Sandro Corsi and Edvina Uzunovic, STATCOM modeling for voltage and angle stability studies, Electric Power and Energy Systems 25, pp. 431-441, 2003. [27] Feng Liu, Shenwei Mei, Qiang Lu, Yixin Ni, Felix F. Wu and Akihiko Yokoyama, The nonlinear internal control of STATCOM: Theory and Application, Electric Power and Energy Systems 25, pp. 421-430, 2003. [28] M.H.Haque, "Compensation of distribution system voltage sag by DVR and D-STATCOM," in Proc. IEEE Power Tech., vol.1, pp.5, Sept.2001. [29] P. Pohjanheimo and E. Lakervi, Steady state modeling of custom power components in power distribution networks, IEEE Power Engineering Society Winter Meeting 2000, Vol. 4, pp. 2949-2954, 23-27 Jan. 2000. [30] R.R.Errabelli, Y.Y. Kolhatkar and S.P. Das, "Experimental investigation of sliding mode control of inverter for custom power applications," in Proc. IEEE PES General Meeting, 2006, June2006. [31] A.Ghosh, A.K.Jindal and A. Joshi, "Inverter Control Using Output Feedback for Power Compensating Devices," in Proc. TENCON 2003, vol. 1, pp.48-52, Oct.2003. [32] Soo-Yong Jung, Tae-Hyun Kim, Seung-I1 Moon and ByungMoon Han, Analysis and control of DSTATCOM for a line voltage regulation, IEEE conference on Power Engineering Society Winter Meeting, Vol. 12, pp. 729-734, 27-31 Jan. 2002. [33] Rajesh Gupta, Arindam Ghosh and Avinash Joshi, Control of cascaded transformer multilevel inverter based DSTATCOM, Electric Power Systems Research 77, pp. 989-999, 2007. [34] Shiv Kumar Iyer, Arindam Ghosh and Avinash Joshi, Inverter topologies for DSTATCOM applications a simulation study, Electric Power Systems Research 75, pp. 161-170, 2005. [35] C.K.Sao, P.W. Lehn, M.R. Iravani and J.A. Martinez, "A benchmark system for digital time-domain simulation of a pulsewidth-modulated D-STATCOM," IEEE Trans. Pow.Del., vol. 17,pp.1 1 13-1 120,Oct. 2002. [36] F.Xiangyun, W. Jianze and J. Yanchao, "A Novel Control Method for DSTATCOM under Unbalanced Conditions,"in Procd Int. Conf on PowerCon 2006, pp. 1-6, Oct.2006. [37] A.Shukla, A. Ghosh and A. Joshi, "A hysteresis current controlled flying capacitor multilevel inverter based DSTATCOM,"in Procd IEEE Power Engineering Society General Meeting, vol. 1, pp.857864, 2005. [38] C.Sharmeela, G. Uma and M.R. Mohan, "Multi-level distribution STATCOM for voltage sag and swell reduction," in Proc. IEEE Power Engineering Society General Meeting, vol.2, pp. 13031307, June2005. [39] R. Gupta, A. Ghosh and A. Joshi, Switching characterization of cascaded multilevel-inverter-controlled systems, IEEE Trans. Industrial Electronics, vol. 55, pp. 1047-1058. [40] S.H. Park, Y.S. Jeon and N.H. Kwak, Analysis of voltage compensation by DSTATCOM using EMTDC program, Proceedings of the 8th IEEE International conference on Electrical Machines and Systems 2005 (ICEMS 2005), Vol. 2, pp. 14641466, 29 Sept. 2005. [41] M. Tavakoli Bina, M.D. Eskandari and M. Panahlou, Design and installation of a 250kVAR D-STATCOM for a distribution substation, Electric power Systems Research 73, pp. 383-391, 2005. [42] Sung-Min Woo, dae-wook Kang, Woo-Chol Lee ang dong-Seok Hyun, The Distribution STATCOM for reducing the effect of

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

125

[43]

[44]

[45]

[46]

[47]

[48]

[49]

[50]

[51]

[52]

[53]

[54]

[55]

[56]

[57]

[58]

[59]

[60]

[61]

voltage sag and swell, IECON01: The 27th annual conference of the IEEE Industrial Electronics Society, pp. 1132-1137, 2001. M.K. Mishra, A. Ghosh, and A. Joshi, "Operation of a DSTATCOM in voltage control mode," IEEE Trans. Pow. Del., vol. 1 8, pp.258-264, Jan.2003. G. Ledwich and A. Ghosh, "A flexible DSTATCOM operating in voltage or current control mode," Proc. IEE Generation, Transmission and Distribution, vol. 149, pp.215-224, March 2002. A.Ghosh and G. Ledwich, "Load compensating DSTATCOM in weak AC systems," IEEE Trans. Pow. Del., vol. 18, pp. 13021309, Oct.2003. B. Singh, A. Adya, A.P. Mittal and J.R.P Gupta, "Analysis, simulation and control of DSTATCOM in three-phase, four-wire isolated distribution systems," in Proc. IEEE Power India Conf, April 2006. B. Singh, J. Solanki, and V. Verma, "Neural Network Based Control of DSTATCOM with Rating Reduction for Three-Phase Four-Wire System" in Proc. Conf IEEE-PEDS'05, vol.2, pp.920925, Nov.2005. Y.Xiao-ping Yang, Z. Yan-xiao Zhang and Z. Yan-ru, "Threephase four-wire DSTATCOM based on a three-dimensional PWM algorithm," in Proc. Int. Conf Electric Utility Deregulation and Restructuring and Power Technologies, 2008, pp.2061-2066, April 2008. M.G. Molina and P.E. Mercado, Control design and simulation of DSTATCOM with energy storage for power quality improvements, IEEE/PES conference on Transmission and Distribution: Conference and Exposition 2006, Latin America, pp. 1-7, 15-18 Aug. 2006. M.K.Mishra, K. Karthikeyan and P.K. Linash, "A development and implementation of DSP based DSTATCOM to compensate unbalanced nonlinear loads," in Proc.IEEE Power India Conf 06, pp. 8, April 2006. B. Blazic and I. Papic, A new mathematical model and control of D-STATCOM for operation under unbalanced conditions, Electric Power Systems Research 72, pp. 279-287, 2004. B.Blazic, I. Papic, "Improved D-StatCom Control for Operation With Unbalanced Currents and Voltages," IEEE Trans. Power Delivery, vol.21, no. 1, pp.225-233, Jan. 2006. Walmir freitas, Andre Morelato, Wilsun Xu and Fuji Sato, Impacts of AC generators and DSTATCOM devices on the dynamic performance of distribution systems, IEEE transactions on Power Delivery, Vol. 20, No. 2, pp. 1493-1501, April 2005. K.Anuradha, B.P. Muni and A.D.Rajkumar, "Simulation of Cascaded H-Bridge Converter based DSTATCOM,"in Proc. IEEE Conf Industrial Electronics and Applications, 2006, pp. 1-5, May 2006. A.Benchaib, E. Lee-Kwet-Sun, J.-E Thierry and G.; de Preville, "Industrial D-STATCOM chain link modeling and control," in Proc. EPE'2007, pp.1-9, 2007. N. Kimura, T. Morizane, K. Taniguchi and Y. Nishida, "Modified double-modulation signal PWM control for D-STATCOM using Five-Level Double Converter," in Proc. EPE'2007, pp.1-10, Sept.2007. A.Elnady, M.M. A. Salama, "Unified approach for mitigating voltage sag and voltage flicker using the DSTATCOM", IEEE Trans. Power Delivery, vol.20, pp.992-1000, April 2005. J.Sun, D. Czarkowski and Z. Zabar, "Voltage flicker mitigation using PWM-based distribution STATCOM", in Proc. IEEE PES Summer Meeting, 2002, vol.1, pp.616-621, July 2002. Shukai Xu, Qiang Song, Yongqiang Zhu and Wenhua Liu, "Development of a D-STATCOM prototype based on cascade inverter with isolation transformer for unbalanced load compensation", in Proc. IEEE Int. Conf Industrial Technology, 2005, pp.1051-1056, Dec.2005. B. Singh and Jitendra Solanki, "A Comparative Study of Control Algorithms for DSTATCOM for Load Compensation," in Proc. IEEE Int. Conf Industrial Technology, pp.1492-1497, 2006. Chunhui Wu, Qirong Jiang and Chunpeng Zhang, "An optimization method for three-level selective harmonic eliminated pulse width modulation (SHEPWM),"in Proc. Int. Conf ICEMS 2005, vol. 2, pp.1346 - 1350, Sept.2005.

[62] L.F.Encarnacao, E.L. Emmerik and M. Aredes, "An optimized cascaded multilevel static synchronous compensator for medium voltage distribution systems," in Proc. IEEE PESC '08, pp.48054811, 2008. [63] G.Escobar, A.M. Stankovic and P.Mattavelli, "Reactive power, imbalance and harmonics compensation using d-statcom with a dissipativity-based controller," in Proc. IEEE Decision and Control, vol.4, pp.3051-3055, Dec.2000. [64] Rong Cai, M. Bongiomo, A. Sannino, "Control of D-STATCOM for Voltage Dip Mitigation," in Proc. Int. Conf Future Power Systems, 2005, pp. 1 - 6, Nov. 2005. [65] J.E Hill, "A practical example of the use of distribution static compensator (D-STATCOM) to reduce voltage fluctuations," in Proc.IEE Colloquium on Power Electronics for Renewable Energy (Digest No: 1997/1 70),pp.7/1-7/5,June 1997. [66] L.G.B. Rolim, A. Ortiz, M. Aredes, R. Pregitzer, J.G. Pinto and Joao L. Afonso, Custom power interfaces for renewable energy sources, IEEE International symposium on Industrial Electronics (ISIE 2007), pp. 2673-2678, 4-7 June 2007. [67] K.N.Choma and M. Etezadi-Amoli, "The application of a DSTATCOM to an industrial facility," in Proc. IEEE PES Winter Meeting, vol.2, pp.725-728, Jan. 2002. [68] Cristian A Sepulveda, Jose R. Espinoza, Leonardo M. Landaeta and Carlos R. Baier, Operating regions comparison of VSC-based custom power devices, 32nd IEEE annual conference on Industrial Elecetronics (IECON 2006), pp. 5344-5349, 6-10 Nov. 2006. [69] V.R. Dinavahi, M.R. Iravani and R. Bonert, Real-time digital simulation and experimental verification of a D-STATCOM interfaced with a digital controller, Electric Power and Energy Systems 26, pp. 703-713, 2004. [70] Lie Xu, Olimpo Anaya-Lara, Vassilios G. Agelidis and Enrique Acha, Development of prototype custom power devices for power quality enhancement, IEEE 9th International conference on Harmonics and Quality of Power, Vol. 3, pp. 775-783, 1-4 Oct. 2000. [71] Yun Wei Li, D. Mahinda Vilathgamuwa, Poh Chiang Loh and Frede Blaabjerg, A dual-functional medium voltage level DVR to limit downstream fault currents, IEEE transactions on Power Electronics, Vol. 22, No. 4, pp. 1330-1340, July 2007. [72] D. Mahinda Vilathgamuwa, H.M. Wijekoon and S.S. Choi, Interline dynamic voltage restorer: A novel and economical approach for multiline power quality compensation, IEEE transactions on Industry Applications, Vol. 40, No. 6, pp. 16781685, Nov./Dec. 2004. [73] H.M. Wijekoon, D. M. Vilathgamuwa and S.S. Choi, Interline dynamic voltage restorer: An economical way to improve interline power quality, IEE proceedings on Generation, Transmission and Distribution, Vol. 150, Issue 5, pp. 513-520, 15 Sept. 2003. [74] Paisan Boonchiam, Promsak Apiratikul and Nadarajah Mithulananthan, Detailed analysis of load voltage compensation for dynamic voltage restorers, IEEE (Region 10) conference on TENCON 2006, pp. 1-4, 14-17 Nov. 2006. [75] Y.D Choy, N.H. Kwak, Y.S. Jeon, D.H. Jeon, T.K. Kim and J.H. Shin, A study on the application of the DVR system using EMTDC program, IEEE T & D Conference and Exposition: Asia nd Pacific 2009, pp. 1-4, 26-30 Oct. 2009. [76] C.Huann-Keng, L. Bor-Ren and W.Kuan-Wei, "Study of dynamic voltage restorer under the abnormal voltage conditions," in Proc. Int. Conf Power Electronics and Drives Systems, vol.1, pp.308312, 2005. [77] Ambra Sannino and Jan Svensson, Application of converterbased series device for voltage sag mitigation to induction motor load, IEEE Porto Power Tech. conference (PPT 2001), Porto, Portugal, Vol. 2, 6 pp., 10-13 Sept. 2001. [78] C.S. Chang, Y.S Ho and P.C. Loh, Voltage quality enhancement with power electronics based devices, IEEE Power Engineering Society Winter Meeting, Vol. 4, pp. 2937-2942, 23-27 Jan. 2000. [79] Nikolaos Athanasiadis, Power Quality Solutions for Voltage Sags Using Dynamic Voltage Restorers, Electric Power Components and Systems, 31:159170, 2003. [80] Nicholas Abi-Samra, Dan Carnovle, Ashok Sundaram and Wade Malcolm, The role of the distribution system dynamic voltage

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

126

[81]

[82]

[83]

[84]

[85]

[86]

[87]

[88]

[89]

[90]

[91]

[92]

[93]

[94]

[95]

[96]

[97]

[98]

restorer in enhancing the power at sensitive facilities, IEEE conference WESCON/96, pp. 167-181, 22-24 Oct. 1996. C.N.M. Ho and H.S.H. Chung, "Fast Transient Control of SinglePhase Dynamic Voltage Restorer (DVR) Without External DC Source," in Proc. IEEE Power Electronics Specialists Conference, pp. 1-7, 2006. K. Piatek, "A New Approach of DVR Control with Minimised Energy Injection," in Proc. of Power Electronics and Motion Control Conference, pp.1490-1495, 2006. Stephen W. Middlekauff and E. Randolph Collins Jr., System and customer impact: Considerations for series custom power devices, IEEE transactions on Power Delivery, Vol. 13, No. 1, pp. 278-282, Jan. 1998. H.P. Tiwari and Sunil Kumar Gupta, Dynamic Voltage Restorer against Voltage Sag, International Journal of Innovation, Management and Technology, Vol. 1, No. 3, August 2010. H. Ezoji, A. Sheikholeslami, M. Tabasi and M.M. Saeednia, Simulation of Dynamic Voltage Restorer Using Hysteresis Voltage Control, European Journal of Scientific Research ISSN 1450-216X Vol.27 No.1, pp.152-166, 2009. H. Ezoji, M.Fazlali, A.Ghatresamani, M.Nopour, A Novel Adaptive Hysteresis Band Voltage Controller for Dynamic Voltage Restorer, European Journal of Scientific Research ISSN 1450-216X Vol.37 No.2, pp.240-253, 2009. Ambra Sannino and Jan Svensson, Static series compensator for voltage sag mitigation supplying non-linear loads, IEEE/PES conference Rec., USA, pp. 1147-1152, 2002. Hilmy Awad, Jan Svensson and Math H.J. Bollen, Static series compensator for voltage dips mitigation IEEE Bologna Power Tech. conference, Bologna, Italy, Vol. 3, pp. 8, 23-26 June 2003. C. Di Perna, P. Verde, A. Sannino and M.H.J. Bollen, Static series compensator for voltage dip mitigation with zero-sequence injection capability, IEEE Bologna Power Tech. conference, Bologna, Italy, Vol. 3, pp. 8, 23-26 June 2003. Ali O Al-Mathnani, Azad Mahamed and Mohd. Alauddin Mohd. Ali, Photovoltaic based dynamic voltage restorer for voltage sag mitigation, The 5th student conference on research and development (SCOReD 2007), Malaysia, pp. 1-6, 11-12 Dec. 2007. C.S. Chang and Y.S. Ho, The influence of motor loads on the dynamic voltage restoration capability of the dynamic voltage restorer, IEEE International conference on Power System Tech. 2000 (POWERCON 2000), Vol. 2, pp. 637-642, 4-7 Dec. 2000. Mehdi Hosseini, Heidar Ali Shayanfar and Mahmoud FotuhiFiruzabad, Reliability improvement of distribution systems using SSVR, ISA transactions 48, pp. 98-106, 2009. H. Funato, T. Ishikawa, and K. Kamiyama, Transient response of three phase variable inductance realized by variable active-passive reactance (VAPAR), in Proc. IEEE-APEC01, pp. 12811286, 2001. H. Funato, K. Kamiyama, and A. Kawamura, Transient performance of power circuit including virtual inductance realized by fully digital controlled variable active-passive reactance (VAPAR), in Proc. IEEEPESC00, pp. 11951200, 2000. W. EL-Khattam, A. Elnady and M. M. A. Salama, Dynamic Voltage Restorer Cost Reduction in the Distributed Generation Environment, Electric Power Components and Systems, 32:611 626, 2004. M.A. Hannan and Azad Mohamed, PSCAD/EMTDC simulation of unified series-shunt compensator for power quality improvement, IEEE transactions on power Delivery, Vol. 20, No. 2, pp. 1650-1656, April 2005. Ming Hu Heng Chen, Modeling and controlling of unified power quality compensator, Proceedings of the 5th international conference on Advances in Power System Control, Operation and Management (APSCOM 2000), Hongkong, pp. 431-435, Oct. 2000. Malabika Basu, Michael Farrell, Michael F. Conlon, Kevin Gaughan and Eugene Coyle, Optimal control strategy of UPQC for minimum operational losses, 39th International Universities Power Engineering Conference 2004, vol. 1, pp. 246-250, 2004.

[99] Haque M.T., Ise T. and Hosseini S.H., A novel control strategy for UPQC, IEEE 33rd Annual on Power Electronics specialists conference 2002, Vol.1, pp. 94-98, 2002. [100] Monteiro L.F.C., Aredes M. and Moor Neto J.A., A control strategy for unified power quality conditioner, IEEE International Symposium on Industrial Electronics 2003, Vol. 1, pp. 391-396, 2003. [101] Tan Zhili, Li Xun, Chen Jian, Kang Yong and Zhao Yang, A new control strategy of UPQC in three-phase four-wire system, IEEE Power Electronics specialist conference 2007, pp. 1060 1065, 2007. [102] Tan Zhili, Zhu Dongjiao, Chen Jian and Haobin Dong, A new control strategy for three-phase four-wire UPQC with zero steadystate error, International Conference on Electrical Machines and Systems 2008, pp. 2022 2025, 2008. [103] Kazemi A., Mokhtarpour A. and Tarafdar Haque M., A New Control Strategy for Unified Power Quality Conditioner (UPQC) In Distribution Systems, International Conference on Power System Technology 2006, pp. 1 5, 2006. [104] Axente I, Basu M and Conlon M.F., A control approach for UPQC connected to weak supply point, 42nd International Universities Power Engineering Conference 2007, pp. 619 623, 2007. [105] Jiangyuan Le, Yunxiang Xie, Zhang Zhi and Cheng Lin, A nonlinear control strategy for UPQC, International Conference on Electrical Machines and Systems 2008, pp. 2067 2070, 2008. [106] Laxmi A.J., Das G.T.R., Rao K.U., Sreekanthi K. and Rayudu K, Different control strategies for UPQC at load side, 1 st IEEE conference on Industrial Electronics and applications 2006, pp. 17, 2006. [107] Kesler Metin and Ozdemir Engin, A novel control method for unified power quality conditioner (UPQC) under non-ideal mains voltage and unbalanced load conditions, Twenty-Fifth Annual IEEE on Applied Power Electronics Conference and Exposition (APEC), pp. 374-379, 2010. [108] A. Nasiri, A. E. Amac and A. Emadi, Series-Parallel Active Filter/Uninterruptible Power Supply System, Electric Power Components and Systems, 32:11511163, 2004.

[109] Morris Brenna, Roberto Faranda, and Enrico Tironi, A New Proposal for Power Quality and Custom Power Improvement: OPEN UPQC, IEEE Transactions on Power Delivery, vol. 24, No. 4, pp. 2107-2116, October 2009. [110] M. Basu, S.P. Das and G.K. Dubey, Investigation on the performance of UPQC-Q for voltage sag mitigation and power quality improvement at a critical load point, IET Gener. Transm. Distrib., Vol. 2, No. 3, pp. 414423, 2008. [111] Khadkikar V., Chandra A., Barry A.O. and Nguyen T.D., Application of UPQC to protect a sensitive load on a polluted distributed network, IEEE Power Engineering society general meeting, 2006. [112] Yashomani Y. Kolhatkar and Shyama P. Das, Experimental Investigation of a Single-Phase UPQC With Minimum VA Loading, IEEE Transactions on Power Delivery, Vol. 22, No. 1, pp. 373-380, January 2007. [113] F. Pilo, G. Pisano and G.G. Soma, Optimal placement of custom power devices to mitigate voltage dips in distribution networks, 13th International conference on Harmonics and Quality of Power 2008 (ICHQP 2008), pp. 1-7, 28 Sept.-1 Oct. 2008. [114] M.A. Kashem and G. Ledwich, Energy requirement for distributed energy resources with battery energy storage for voltage support in 3-phase distribution lines, Electric power Systems Research 77, pp. 10-23, 2007.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

127

Indian Electricity Market: Current State and Perspectives


1

Prateek Kumar Singhal, 2Rakesh Kumar


1, 2
1

NIT Hamirpur (H.P)

pks.mtech@gmail.com 2 rkyadav39@gmail.com Abstract The Electricity in India is going through a major change with the Electricity Act-2003 coming into force on June 10, 2003. Indian Power industry is moving rapidly from regulated conventional setup to a deregulated environment. There is an urgent need to keep a track of international experiences and activities taking place in this emerging field. This paper presents the study on past, actual and the future status of the Deregulation & Restructuring activities of electricity market in India. It ensures the electricity network flexible enough to meet new and less predictable supply and demand conditions in competitive electricity markets. Keywords Deregulation, Restructuring, Indian Power System, Congestion Management, Power Market

I. INTRODUCTION Deregulation is the removal or simplification of government rules and regulations that constrain the operation of market forces [1]. Deregulation does not mean elimination of laws against fraud, but eliminating or reducing government control of how business is done, thereby moving toward a more free market. Deregulation increases the level of competitiveness which results in higher productivity, more efficiency and lower prices of electricity. Electricity industry, throughout the world, is currently under going restructuring and adopting the deregulated industry structure for better utilization of the resources and for providing choice and quality service to the consumers at economical prices [2]. The deregulated structure is envisaged to create some sort of an electricity market and introducing competition at various levels of electricity related transactions (other than transmission, which is natural monopoly). Under this structure, a competitive market of electricity is created to enable the generators to compete with each other by availing open access to the network. Over the past two decades a number of countries have restructured their electricity industry by significantly reducing the governments role in the ownership and management of domestic electricity industries. It has seen as necessary conditions for increasing the efficiency of electric energy production and distribution, offering a lower price, higher quality and secured supply [3]. The forces behind electric sector deregulation taking place worldwide are different in different countries. These deregulation processes have been developed that have defended the vertically integrated model. Asian countries including India, China, South Korea, Singapore, Japan and Thailand have also taken historic steps to restructure their

electricity industry. The vertically integrated utilities after debate and opposition by private and state monopolies that have defended the vertically integrated model. Asian countries including India, China, South Korea, Singapore, Japan and Thailand have also taken historic steps to restructure their electricity industry. Operation and control of restructured electricity market posses technical challenges far more complex than the conventional monopolistic market. The complexity arises due to involvement of several market entities, many types of contractual obligations, and separation of primary and ancillary services and varying models of market management. Some of the technical challenges include congestion management, market power, price volatility and ancillary services management [4], [5]. This paper is an attempt to study the changes going on in the Indian electricity market, its components, salient features, and the challenges that are poised for future development. The organization of paper as follows: Section II presents the overview of Indian electricity market. Section III presents the Electricity power demand in India for current year. Section IV represents the Electricity power generation in India with different fuels. Section V represents the Restructuring process of Indian Power sector. Section VI represents the Advancement in Power system problem solving techniques. Section VII represents the some suggestions for future power development in India based on the study. Section VIII represents the different websites from which the data has been taken and finally Section IX concludes the study and analysis of Electric power sector of India. II.
OVERVIEW OF INDIAN ELECTRICITY MARKET

A. Past Scenario In India, till independence, the entire power sector was under the control of private sector. After the enactment of new Electricity Act in 1948, which modified the 1910 Act, barring a few licensees in some urban areas, e.g. The Tata Power Company Ltd in Mumbai, Calcutta Electric Supply Corporation Limited (CESC) in Calcutta, Bombay Suburban Electric Supply Company (BSES) in Mumbai, Ahmadabad Electricity Company (AECO) in Ahmadabad, etc. the entire power sector is mostly owned by State Governments and is largely managed by vertically integrated electricity business through State Electricity Boards (SEBs) .During fifties and sixties, Inter State power projects were established along with the

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

128

transmission infrastructure to avail the power in an economical manner. In 1975, Government of India (GOI) entered in the field of generation and transmission through their Central sector utilities. However, distribution sector continued to remain with SEBs as a monopoly business. Over the period of operations, the sector developed techno-commercial inefficiencies. Restructuring therefore was felt necessary. Accordingly, the power generation was opened up in 1991 followed by transmission in 1998 [6], [7]. Electricity Regulatory Commissions Act was enacted in 1998 for establishing regulatory commissions in various States. Recently the Electricity Act 2003 has been notified by Government of India in June 2003. The objective of this act is to accelerate the power sector reforms. The construction and operation of generation and transmission projects in the Central sector are entrusted to Central Sector Power Corporations, viz., the National Thermal Power Corporation (NTPC), the National Hydro Electric Power Corporation (NHPC), the North Eastern Electric Power Corporation (NEEPCO) and the Power Grid Corporation of India Limited (PGCIL). The Power Grid is responsible for all the existing and future transmission projects in the Central sector and also for the formation of the National Power Grid. Two joint-venture Power Corporations namely, Nathpa Jhakri Power Corporation (NJPC) and Tehri Hydro Development Corporation (THDC) are responsible for the execution of the Nathpa Jhakri Power Project in Himachal Pradesh and Projects of the Tehri Hydro Power Complex in Uttar Pradesh respectively. Two statutory bodies, i.e., the Damodar Valley Corporation (DVC) and the Bhakra Beas Management Board (BBMB), are also under the administrative control of the Ministry of Power. Programs of rural electrification are provided financial assistance by the Rural Electrification Corporation (REC) under the Ministry of Power. The Power Finance Corporation (PFC) provides term-finance to projects in the power sector. Further, the autonomous bodies (societies), i.e., Central Power Research Institute (CPRI), the National Power Training Institute (NPTI) and the Energy Management Centre (EMC), are also under the administrative control of the Ministry of Power. B. Present Scenario Power systems in most of the states are owned, operated and managed by State Electricity Boards (SEBs), which generate, transmit and distribute power within the State territory. In addition to SEBs, few private sector utilities operate in the metropolitan cities like Bombay, Calcutta and Ahmadabad. In 1975, central government, through central public sector undertakings, also entered in the field of generation and transmission, to supplement the efforts of cash starved state electricity boards and also to take advantage of economies of scale. However, distribution continued to remain with SEBs as a monopoly business. Need for restructuring the electricity sector was felt since late 1980s, firstly, due to limited resources available with central and state governments and secondly, to improve

the technical and commercial efficiency. Consequently, power generation was opened up to private sector in 1991, followed by transmission in 1998. Electricity Regulatory Commissions Act was enacted in 1998 for establishing regulatory commissions. Today, regulatory commissions are operational at the national level and as well as in twenty-one states. As per the statute, the Central Electricity Authority (CEA) is responsible for power planning at the national level. CEA advises the Ministry of Power on the national power policy and national power planning, while the central electricity regulatory commission is looking after the regulatory issues. Day-today operation of the regional grid is carried out by Regional Load Dispatch Centers (RLDCs), which are under the operational control of CTU, i.e. PGCIL. The main function of RLDC is, to carry out the integrated operation of the power system in that region and that of Regional Electricity Board (REB), to facilitate integrated grid operation. Presently, five REBs, namely, Northern REB, Southern REB, Western REB, Eastern REB and North-Eastern REB exist to promote the integrated operation of the regional power systems. The responsibilities of REBs are to review project progress, to plan integrated operation among the utilities in the region, to co-ordinate the maintenance schedules, to determine the availability of power for interstate utilities transfer, to prescribe the generation schedule and to determine a suitable tariff for the interutility exchange of power. The names of States/Union Territories in each REBs are given in Table 3.
TABLE 1: ORGANIZATION STRUCTURE OF REGIONAL ELECTRICITY BOARDS NREB Haryana Himachal Pradesh Delhi Punjab Rajasthan Jammu & Kashmir Chandigarh Uttar Pradesh Uttaranchal WREB Gujarat Madhya Pradesh Maharashtra Goa Daman & Diu Dadar & Nagar Haveli Chhattisgarh SREB Karnataka Andhra Pradesh Kerala Tamilnadu Pondicherry
Lakshadweep

EREB Bihar Andaman & Nikobar West Bengal Sikkim Orissa Jharkhand -

NEREB Assam Arunachal Pradesh


Meghalaya

Nagaland Tripura Manipur Mizoram -

C. Current Structure Currently, State Load Dispatch Centers (SLDCs) are carrying out the optimum scheduling of the state generating units and the RLDCs are responsible for scheduling of central sector generating units only. SLDCs send the requisition to the RLDCs against their entitlements out of available power from central sector generation (CGS) and the RLDCs allocate total available power to various states in the ratio of their entitlements.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

129

Day-to-day operation of the regional grid is carried out by RLDCs, which are under the operational control of Central Transmission Utility (CTU), i.e. Power Grid Corporation of India Ltd. Apart from above, a small amount of power (about 2.5% of total generation) is being traded at wholesale level through either bilaterally or with the help of power traders. But presently, trading is mostly restricted to players such as SEBs and utilities. Power trading has generated considerable interest among power players, as it is evident from the brisk line-up for licenses at Central Electricity Regulatory Commission (CERC). Presently there are seventeen trading licenses to whom CERC has granted license for inter-state trading in electricity. These traders apply for open access on behalf of suppliers and buyers to the nodal RLDC depending upon transaction requirement. CERC has made the regulations for open access in interstate transmission and inter-state trading. The market structure of current Indian electricity industry is shown in Fig1, depicting energy flow and money flow separately.

demand in India to increase by 5.4% per year from 1997 to 2020, faster than the assumed GDP growth rate of 4.9% (IEA, 2000) [9]. The duration and number of blackouts and brownouts are beyond acceptable limits, leading to shortfalls of up to 15% of demand. Consumption is largely constrained by the supply as Fig2 shows. Fig2 represents load change in a large Indian city. As this Figshows, it seems that the seasonal variation in the load between summer and winter is limited, and changes during the day are not high either as compared to other countries. Inadequate power transmission and distribution result in shortages which in turn affect consumption patterns and induce commercial users and the most affluent domestic customers to rely on standby/in-house investments in autoproduction capacity [8].

1971, Total: 51 TWh


Industry

9%

5%
Transport
Agriculture

6%
10% 67%

3%

Commercial and Public Services Residential Other Sectors

Fig. 1: Current Indian Power Market Structure

2010, Total: 600 TWh 2% 16% 8%


Industry Transport

III. ELECTRIC POWER DEMAND IN INDIA From the time of Indias independence in 1947, the demand for electricity has grown rapidly. Final consumption of electricity has increased by an average of 7% per year since 1947. This sustained growth is the result of economic development and the increase in electrical appliances. It has been accompanied by a gradual shift from non-commercial sources of energy, such as biomass, in the household and commercial sector as well as the reduction in the use of coal for process heat in industry and kerosene for household lighting. Of total final sales of 332 TWh in 1999-2000, industry accounted for just over one-third, agriculture for 30% and the household sector for 18%. But for many years, electricity supply has fallen short of demand and the sustainability of this trend is very uncertain [7]. Though the overall demand-supply gap decreased from an estimated 8.1% in 1997-98 to 5.9% in 1998-99, it rebounded to 6.2% in 1999-2000. Peak power shortages fell from 18% in 1996-97 to 12% in 1999-2000. In spite of sustained growth, electricity consumption per capita was only 416 kWh per annum (in 1998), far below the world average of 2,252 kWh. The International Energy Agencys World Energy Outlook 2000 projects electricity

45%
Agriculture

27%

2%

Commercial and Public Services Residential Other Sectors

Fig. 2: Electricity Consumption in India

Because of unsatisfied demand, increases in electricity prices would not automatically lower consumption. A part of the population could afford a costlier electricity service, but available supply cannot satisfy their demand. A large majority of the population is rural. Officially, close to 90% of the villages are electrified. However, only half the Indian population does in fact have electricity. Since a large portion of the population lives below the poverty

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

130

line, they cannot afford electricity at current costs; this is particularly true in rural areas.
TABLE 1 HOUSEHOLD ACCESS TO ELECTRICITY IN INDIA IN 2010(%) [8]

1971, Total:61 TWh 2%

Coal Oil

Total access Rural access Urban access Rural population as % of Total Population Urban population as % of Total Population

45.7 33.1 81.5 74.0 26.0


45% 5% 1%

47%

Gas
Hydro Nuclear

2500
load(MW)

2000 1500 1 3 5 7 9 11 13 15 17 19 21 23 24 hours


Summer (22 June, 2009) Winter (6 January, 2010)
Fig. 4: Electricity Generation by fuel TABLE 2 : PROJECTED ELECTRICITY CAPACITY AND GENERATION IN INDIA

Fig. 3: Daily load Curve (New Delhi)

IV. ELECTRIC POWER SUPPLY IN INDIA India had 1,73,626.40 MW of generating capacity on 31 March 2011 which is going to be increased in future at 8% GDP growth rate. Currently an average annual rate of growth of just over 6%. Indias electricity supply is mainly based on coal burnt in boilers feeding steam turbines, an adequate technology for base-load power generation and to a lesser extent for intermediate-load generation [10]. Hydroelectricity capacity and the shares of hydro in generation have been decreasing over the years, reducing the availability of peak-load power. The Indian system is biased toward the production of base-load power, while the supply-demand gap is mainly in peakload electricity. Gas is gaining an increasing role. Nuclear accounts for a marginal share of capacity and is not expected to be a major source of power in the immediate future [11]. Indias electricity supply has long been dominated by the public sector. Public ownership and public management of the main elements of the supply industry have been the rule since independence. At that time, most existing electric utilities were integrated into 19 SEBs (see Electric Supply Act of 1948) and eight electricity departments. These boards were part of state governments [12].

Total Coal Oil Gas Nuclear Hydro Other Renewable Source: IEA

1997 GW 103 66 3 9 2 22 1

TWh 464 339 12 28 10 75 0

2020 GW 308 193 6 47 6 50 6

TWh 1484 1008 32 216 39 171 18

TABLE 2 : PEAK LOAD AND ENERGY FORECASTS IN INDIA

2001-2002
Peak Load( MW) 85,132 Generaiton (GWh)

2006-2007
Peak Load( MW) 115,70 5 Generaiton (GWh)

2011-2012
Peak Load( MW) 157,10 7 Generaiton (GWh)

529,013

719,097

975,222

India is endowed with the third-largest coal reserves in the world; hence the bulk of electric power supply is naturally based on coal. Is the time ripe for the development of a large capacity based on natural gas combined-cycle units? Should India emphasize the development of nuclear energy or other very capital-intensive technology? Should India increase harnessing its huge hydroelectric resources

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

131

in the North and East? What are the possibilities for renewable, especially wind energy? [13] The IEA projects a threefold rise in Indias generation capacity from 1997 to 2020, which corresponds to an average yearly growth rate of 5.2%. The development of coal-fired generation based mostly on domestic resources will continue, with a threefold increase to 308 GW in 2020 [14]. There will is also be a fivefold increase in the use of natural gas to 47 GW. In 1997, the load factor of the Indian power system was 51% a low rate by international standards. This rate is assumed to increase to 55% in 2020. The most salient assumption is a sharp increase in the load factor of gasfired power plants, from 36% to 53%, due mostly to the expected market deployment of natural gas combinedcycle units for base-load uses. The IEA also assumes a sizeable increase in the use of renewable energy. V. RESTRUCTURING OF INDIA The restructuring process in India has been started from year 2003, after the Electricity Act 2003 has been implemented. In that Act, PGCIL has been notified as central transmission utility at the national level and the SEBs or the main transmission companies in the restructured states are functioning as state transmission utilities at the state level. Regulatory commissions at the state and national level have been in operation. Further restructuring in the form of unbundling and corporatizing the generation, transmission and distribution is envisaged in the Electricity Act, 2003. After that Act, the competition in the Electricity market has been increased i.e., the bulk consumers and distribution companies will now be permitted to buy power directly from any generating companies [15]. The power sector in India is presently going through a process of reform and restructuring. Independent regulatory commissions are being set up, and various electricity boards are now being converted into corporate entities. The Indian power sector has been regulated and owned for many years by various government agencies. It is organized in five Regional Electricity Boards, each consisting of several state electricity boards. The Central Electricity authority is responsible for power planning at the national level. Electricity theft is quite common and is shown as transmission and distribution losses which are about 21 %. The electricity pricing has been decided by various state governments. Thus agriculture sector gets the electricity almost free whereas commercial consumer has to pay very high price, only Japan has a higher electricity price among the developing countries. Also, there is a large gap between generated capacity and demand for electricity, in spite of the fact that per capita consumption of power in India is only 400 KWh which is 1/6 th of global average. The restructuring process of Indian power sector is in progress and most of the SEBs has been privatized partially.

D. Keystones In Restructuring the Indian Electricity Industry The ABT and Electricity Act 2003 are the two major path breaking legislations that had paved the way for the restructuring in Indian Electricity industry [16]. VI. ADVANCEMENT IN POWER SYSTEM PROBLEM SOLVING TECHNIQUES In many areas the deregulated power industry behaves differently compared to classical or regulated power industry. Especially in few cases, the problem formulation and the methodology used became totally different and a major break-through in power engineering education is expected shortly. E. Case 1: Social benefit (SB) optimization In case of regulated power system, the objective function to be minimized is normally as follows:

Where is the Generation cost function for generation at bus i. Whereas the objective of the deregulated power sector market is to maximize the social benefit and decrease the cost of electricity through competition. The market environment typically consists of a pool and privately negotiated contracts. The performance of a market is measured by its social welfare, also called social benefit (SB) [17]. Mathematically, the objective is to maximize SB, which is given as follows [16]:

Subject to system constraints Where the generation cost curve bid to the pool by generator at bus i be denoted by . The worth function (which is also called benefit curve) for load that is price dependent be . It represents the price the load is willing to pay to purchase an amount of power . Similarly non-firm transaction nfbt bid some price function which they are willing to pay for transfer of power into the transmission network. Payments of this type are needed since transaction power transfers bring about incremental losses of the transmission system and may also cause congestion into the system. Whereas all firm transactions are ready to pay the system marginal price and they do not bid.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

132

F. Case 2: Price based unit commitment problem (UCP) Unit Commitment is the problem of determining the schedule of generating units within a power system subject to device and operating constraints. The generic unit commitment problem under regulated power industry can be formulated as [17, 18]: Minimize operational cost (OC)

Where - Fuel Cost - Maintenance Cost - Start-up Cost - Shut-down cost Redefining the unit commitment problem for the deregulated environment involves changing the demand constraints from an equality, to less than or equal, and changing the objective function from cost minimization, to profit (revenue-operational cost) maximization. Now the generic unit commitment problem under deregulated environment can be formulated as [18]: Maximize Profit (P)

Subject to following constraints: New Power balance

Where N - Number of units T - Number of time periods - Forecasted price for period t - Up/down status of unit i - Power generation of unit i during time period t - Forecasted demand during time period t SUGGESTIONS TO THE FUTURE DEVELOPMENT OF INDIAN ELECTRICITY MARKET Based on the above discussion, the following suggestions to the future development of the Indian electricity market have been given: A separate tariff scheme can be implemented to incorporate the charges for the reactive power. So far only the industrial sector is made to pay for the reactive power, they consume depending on their inductive load. Residential sector has so far been ignored as the serious consumers of the reactive power, although their collective reactive demand can cause a serious concern on the load side reactive power demand. Thus VII.

consumers can be billed based on the apparent power consumed by them instead of the active power. To implement the tariff on apparent power, a provision for the day and night tariffs can be introduced. Since the weather plays a major role in load forecasting, tariff structure can further be segregated with respect to the seasons. There should be a perfect completion between the generation companies and distribution companies by fare bidding option. For this, similar to many parts of the world, two models for the system operator are prevalent. The first one is the Independent system operator (ISO) model and the second one is the Transmission System Operator (TSO) model. In ISO model, transmission companies are allowed to own, manage, and control generation and distribution, and the ISO is created to facilitate open access and competitive markets. In the TSO model, operation and ownership of the grid are integrated in one single entity to provide nondiscriminatory open access to all market participants. Neutrality is an important aspect of the TSO model. Considering the current practice of the Indian electricity market, the TSO model seems to be the most suitable for future development There are also two pool models available for the structure of the restructured electricity industries in the world. The first one is the Rigid Pool model, where all the market participants have to take part in the bidding and no power transactions are allowed outside the pool. The second model is the Flexible Pool model, which is similar to the one that is followed in New York and New England where power transactions outside the pool are also permitted. Considering the fact that much of the distribution network in many Indian cities is in dire state and until the commercial losses such as theft are reduced to considerable levels, success of the Flexible Pool is highly doubtful. Thus at the present scenario, the Rigid Pool is the best alternative for power transactions at the state level. There should not be any interference of the political party. The whole command of operating generation and distribution firms should not be given to any private owned firm i.e. there should be some share of state or central government. VIII. IMPORTANT WEBSITES Finally this section contains some important websites, which provides information, data etc. related to Indian electricity market and restructuring process of India. 1. Central Electricity Authority (CEA) of India. [Online], Available: http://www.cea.nic.in 2. Central Electricity Regulatory Commission. [Online], Available: http://www.cerind.org 3. State Energy Efficiency Agency (SEEA). [Online], Available: http://www.seea.government.bg

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

133

IX. CONCLUSION In this paper the current power sector of India has been analysed & studied. Based on the analysis, it has been observed that in India, power industries are moving rapidly from regulated conventional set-up to a deregulated environment and also requires support of so many new operational methodologies including optimization of social benefit and price based unit commitment problem. So the power system engineering educators are required to upgrade the power engineering education curriculum parallel with advancement of the structure of power industry so that, the outgoing students and researchers can face the challenges with modern deregulated power industries. REFERENCES
[1] M. Shahidepour, and A. Srivastava, Restructuring choices for the Indian Power sector, IEEE Power Engineering Review, Nov. 2002. A. K. David, N. Leeprechanon, S. S. Murthy, and F. Liu, Transition to an electricity market: A model for developing countries, IEEE Transactions on Power Systems, vol.17, no.3, pp. 885-894, Aug. 2002. S.N. Singh, and S.C. Srivastava, Electric power industry restructuring in India: present scenario and future prospect, in Proceeding of IEEE International Conference on Electric Utility Deregulation, Restructuring and Power Technologies (DRPT 2004), vol.1, pp. 20-23, April 2004. F.A. Wolak, Reforming the Indian electric supply industry, Available: http://www.stanford.edu/~wolak, May 2004. S. A. Khaparde, Power sector reforms and restructuring in India, in Proceeding of Power Engineering Society General Meeting, IEEE, pp. 2329-2336, 2004. P. Baijal, Restructuring Power Sector in India, A Base Paper, Economic and Political weekly, 25 Sep, 1999. S. Ahluwalia, Power Tariff Reform in India, Economic and Political Weekly, 16 September, 2000. M. Alam, J. Sathaye, and D. Barnes, Urban household energy use in India: efficiency and policy implications, Energy Policy, 26/11, pp. 885-891, 1998. Y.R. Sood, N.P. Padhy and H.O. Gupta, Wheeling of Power under Deregulated Environment of Power System-A Bibliographical Survey, IEEE Transactions on Power Systems, vol.17, no.3, pp. 870-878, August 2002. CEA, 1997, Fourth National Power Plan 1997-2012, Central Electricity Authority, New Delhi. CEA, 1998, Fuel map of India, Central Electricity Authority, New Delhi. CEA, 1999, Perspective Transmission Plan 2011-2012, Central Electricity Authority, New Delhi. GOI, 2000, Annual report on the working of State Electricity Boards & Electric Departments, Planning Commission, New Delhi. IEA, 1999, Electricity Market Reform, an IEA handbook, International Energy Agency, Paris. Ministry of Power, Blueprint for Power Sector, Government of India, 2010. Government of India, The Electricity Act 2003, The Gazette of India, Extraordinary, 2003, Part II Section 3 Sub-Section (ii), New Delhi, Ministry of Power, June 10, 2003. Lai Loi Lai, Power System Restructuring and Deregulation, (book), Publisher: John Wiley and Sons, Ltd., New York, 2001. A. J. Wood and B. F. Wollenberg, Power Generation, Operation and Control, John Wiley and Sons, New York, 1996.

[2]

[3]

[4] [5]

[6] [7] [8]

[9]

[10] [11] [12] [13]

[14] [15] [16]

[17] [18]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

134

Availability Based Tariff Discussion and Issues


1

Shafqat Mughal, 2Yog Raj Sood


1

1, 2

NIT Hamirpur Himachal Pradesh


shafqat.mtech@gmail.com 2 yrsood@gmail.com

AbstractThe UI mechanism has dramatically reduced the problems faced by the Indian power sector in the field of grid management and discipline. It has developed investors confidence in investing in Indian power sector by creation of transparent pricing system. UI mechanism in itself supports other reform policies such as open access, demand side management, efficiency, conservation etc. Huge resources of developing India are saved by this mechanism, supporting both over injection as well as under injection depending on the conditions prevailing in the system. However in the recent days several issues regarding UI mechanism have come in to picture. The purpose of this research paper is to analyse the UI mechanism which is prevailing in India. Keywords Deregulation, Scheduling, Unscheduled Interchange, Frequency.

transparent system is required to cause a paradigm shift in the electric industry in India. II. TARIFF SCHEMES PRIOR TO ABT. At the initial stages of Bulk Power Supply (in 1980s) a single part tariff for pricing thermal power and fixed tariff for hydro power were in vogue. Although the advantage of these tariff systems being of its simplicity in implementation, it is generally felt that the single part tariff had not been conducive to motivate and encourage economic generation of power and its absorption by the Boards, as also in ensuring optimization of integrated operation of the grid system. There was a growing feeling on the part of the customer that the system of single part tariff based on a normative parameters of operation agreed to at the time of tariff fixation has resulted in substantial over recovery of fixed expenses as well as operating cost. Up to 1991, the single part tariff of Central Generating Stations was calculated such that full recovery of fixed cost was assured at a normative PLF of 62.8% (5500 hrs of generation). Generation below this target level penalized the generator on the recovery of fixed cost and conversely generation above the normative PLF level resulted in significant excess revenue. With a view to reduce or eliminate the problems caused by the original single part tariff for sale of central sector generating units to Boards and also to resolve disputes arising in the settlement of tariffs, the Department of Power constituted a Committee headed by Shri K.P. Rao, Member (E&C), CEA. The Committee, in June 1990, submitted a report entitled Report of the Committee on Fixation of Tariff for Central Sector Power Stations. This report (commonly known as K.P. Rao Tariff) which has been accepted in its entirety by the Government of India has the following salient features:i) Two part tariff consisting of fixed charge (related to investments) and variable charge (related to fuel and energy generated) was introduced. The components of annual fixed costs and variable costs and the precise manner of determining these costs were identified. Normative technical, financial and operational parameters were specified. ii) A concept of Deemed Generation was introduced which allow generators to recover its fixed charge and receive incentives, in the event of a station being available but forced to back down due to system constraint. iii) The incentive and penalty structures were rationalised. Full Capacity Charge (FCC) is continued to be payable at 5500 Hrs. (62.8%) generation. Incentives were

I. INTRODUCTION The electric power industry is experiencing a transition from power system monopolies to a deregulated market. Under this condition, utilities are forced to compete with each other [1], and for their existence in power market may cut quality and reliability cost. The primary characteristic that describes the state of power system in India is an insufficient generation to meet the rapid increase in demand. Bulging population and high growth rates have resulted in peak power deficit of 13-14% and energy shortage of 8-10% [2]. Apart from insufficient capacity additions, high transmission and distribution losses, poor interregional transmission links and fuel supply bottleneck have added to the misery of situation. During the last decade, the Government of India has been taking path-breaking initiatives on policy, regulatory and reforms front to come out of this situation [3]. The setting up of regulatory commissions in 1998, unbundling of monolithic state utilities, introduction of Availability Based Tariff in 2002 and enactment of Electricity Act in 2003 have been some of the initiatives that have gone long way in creating an environment of competitiveness and efficiency in Indian electricity sector ABT has been applied in Indian power sector since 2002 and is able to maintain the grid frequency within the specified range of 49.0-50.5 Hz without any single instance of grid failure since then [6]. Indian power system operates as a loosely connected pool regulated by Indian Electricity Grid Code (IEGC) that allows such lose control of frequency [7] However in recent years some issues have come in picture regarding UI mechanism. Thus to accelerate the process and bring about a complete transformation, a new

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

135

allowed @ 1 Paise /KWH for every 1% increase in PLF over 68.5% (6000 Hrs of generation). Disincentive is proportionate from 62.8% PLF at full fixed charge to 50% FCC at 0% PLF. The intent of original K.P. Rao tariff was to provide a mechanism to apply a two-part tariff. The report stressed that central sector generators fixed expenses were to be shared proportionately among the customers share from the generators. However, it has also specified an alternate billing methodology that the monthly fixed charge of the station could be recovered from the beneficiaries in proportion to their energy drawal. Central Government while issuing the tariff notification(s) of the individual generators in October, 1991 for the four central sector projects, viz. Singrauli, Rihand, Vindyachal and Ramgundam STPS for the sake of simplicity, has finalised the alternate billing methodology. Subsequently a number of central sector projects are established in the country, like Farakka, Kahalgaon, Talcher STPS and Rangit HEP in the Eastern Region and all the tariff notifications issued by Government of India, the alternate billing has been considered. Thus, for all practical purpose, the bulk power tariff till date remained basically more or less a single part tariff. This billing methodology has put grid operators in a piquant situation, where any particular drawee constituents violates the grid discipline in drawing more power in peak condition and inject power to the grid in the off-peak condition, instead of being penalised gets commercial advantage i.e. lesser bill from the central sector generator. III. INDIAN GRID BEFORE ABT Prior problem in the Indian power sector was not only the shortage of power but also the difficulty in performing grid operations due to acute indiscipline shown by the generators as well as the beneficiaries. The incentives to the generator were linked to actual generator and not on availability. The generators could pump as much power in the grid as they could irrespective of the frequencies and still get acknowledged for the wastage of the valuable resource. The load serving SEBs would compare the variable cost of their generator to the composite cost of the external generator causing a skewed dispatch. The regional grid operators ironically had a horrifying time trying to get generators backed down to protect the turbines of the same generator causing the situation. On the other end the stated utilities could overdraw from the grid even during deficit and still escape creating a chaos and despair all around. In the Southern Grid, there is a wide variation between the demand and supply. The distinct features of the Indian Grid before introducing ABT were: Low frequency during peak-load hours, with frequency going down to 48.0 48.5 Hz for many hours every day. High frequency during off-peak hours, with frequency going up to 50.5 51.0 Hz for many hours every day.

Rapid and wide changes in frequency, one Hz change in 5-10 minutes being a common phenomenon. Very low voltages, for substantial period. High voltages during off-peak hours, to the extent that even 400 KV trunk lines are required to be switched off to contain the over-voltages. Very frequent grid disturbances, causing tripping of generating stations, interruption of supply to large blocks of consumers, and disintegration of the Regional grids. Before introducing ABT, the two-part tariff system was prevailing, i.e. the fixed cost and the variable cost are charged from the beneficiary states/SEBs in the proportion of the actual energy drawn during a particular period. Before introducing ABT, if a beneficiary over draws a portion of the other beneficiarys share of power, there was no penalty on those who get the benefit or there was no compensation for those who suffered. IV. AVAILABILITY BASED TARIFF ABT concerns itself with the tariff structure for bulk power and is aimed at bringing about more responsibility and accountability in power generation and consumption through a scheme of incentives and disincentives. As per the notification, ABT is applicable to only central generating stations having more than one SEB/State/Union Territory as its beneficiary. Through this scheme, the CERC (Central Electricity Regulatory Commission) looks forward to improve the quality of power and curtail the following disruptive trends in power sector: i) Unacceptably rapid and high frequency deviations (from 50 Hz) causing damage and disruption to large scale industrial consumers ii) Frequent grid disturbances resulting in generators tripping, power outages and power grid disintegration This objective is to be brought about by encouraging generators to produce more during peak load hours and curtail generation adequately during off-peak hours on one hand and discouraging consumers from overdrawing on the other hand. The new tariff regime aims at inducing this discipline at the generation and consumption end through adequate monetary incentives. The most significant aspect of ABT is the splitting of the existing monolithic energy charge structure into three components viz. capacity charges (fixed), energy charges (variable) and UI (unscheduled interchange) charges. It is the last component that is expected to bring about the desired grid discipline. Splitting of the tariff into fixed and variable cost components is meant to act as an incentive for power trading which shall (ideally) conclude in a self-regulating power market regime. It is also expected to promote the concept of ELD (Economic Load Dispatch) among power generators ABT is a three part tariff comprising of:-

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

136

I. Capacity or Fixed Charges: The capacity charges are computed on the following basis and its recovery shall be related to availability: Interest of loan capital Depreciation Return of equity Operation and maintenance expenses including insurance Interest on working capital As per CERC order full capacity charge shall be recoverable at target availability (for NTPC stations this target has been fixed as 80%). At zero availability no capacity charge shall be payable. Recovery of capacity charges below the level of target availability shall be on pro rata basis. II. Energy charges: Energy (variable) charges shall cover fuel cost and shall be worked out on the basis of paise/KWH on Ex-bus energy scheduled to be sent out from the generating stations as per the following formula: Energy Charges = Rate of energy charges x scheduled generation (Ex-Bus) III. Unscheduled interchange (UI) charges: Variation in actual generation /drawal and schedule generation / drawal shall be accounted for through Unscheduled Interchange (UI) charges. UI for generating stations shall be equal to its actual generation minus its scheduled generation. UI for beneficiaries shall be equal to its actual drawal minus to its total scheduled drawal. UI shall be worked out for each 15 minutes time block. Charges for all UI transaction shall be based on average frequency at all the time block and following rate shall apply: TABLE 1: FREQUENCY UI RATES Average Frequency of time UI Rate (Paise per block kWh) 50.5 Hz and above 0.00 Below 50.5 Hz and upto 5.60 50.48 Hz 414.40 Below 49.04 Hz and upto 420.00 49.02 Hz Linear in 0.02 Hz step Below 49.02 Hz Between 50.5 Hz and 49.02 Hz The above average frequency range and UI rates are subject to change through a separate notification of CERC from time to time. Therefore, total tariff = Capacity Charge + Energy Charge (+/-) UI Charge The function of preparation of Regional Energy Accounting REA was being carried out by the Regional Electricity Boards (REBs) for the past several decades prior to April 1, 2006. After implementation of Availability Based Tariff (ABT), besides function of REA, preparation of weekly UI and Reactive Energy Accounting were also being carried out by the REBs in accordance with clause 2.3.2 of the IEGC issued in March 2002.

V. SCHEDULING AND LOAD DISPATCHING UNDER ABT In India there are five National grids which are responsible for the scheduling of generation in their respective zones. The process starts with the Central generating stations in the region declaring their expected output capability for the next day to the Regional Load Dispatch Centre (RLDC). The RLDC breaks up and tabulates these output capability declarations as per the beneficiaries' plant-wise shares and conveys their entitlements to State Load Dispatch Centres (SLDCs). The latter then carry out an exercise to see how best they can meet the load of their consumers over the day, from their own generating stations, along with their entitlement in the Central stations. They also take into account the irrigation release requirements and load curtailment etc. that they propose in their respective areas. The SLDCs then convey to the RLDC their schedule of power drawal from the Central stations (limited to their entitlement for the day). The RLDC aggregates these requisitions and determines the dispatch schedules for the Central generating stations and the drawal schedules for the beneficiaries duly incorporating any bilateral agreements and adjusting for transmission losses. These schedules are then issued by the RLDC to all concerned and become the operational as well as commercial datum. However, in case of contingencies, Central stations can prospectively revise the output capability declaration, beneficiaries can prospectively revise requisitions, and the schedules are correspondingly revised by RLDC.To be very precise Each day of 24 hours starting from 0.00 hour is being divided in to 96 time blocks of 15 minutes duration. Each generating station shall make advance declaration of the capacity available for generation in terms of MW delivery of ex-bus in each time block. The generator shall also ensure that, the capacity available during peak hours is not less than the capacity available during off-peak hours. Based on the declaration of the generating station, the RLDC shall communicate to the beneficiary SEBs, their respective share of the available capacity. The beneficiary shall prepare their schedules from the above, based on the internal generation availability, variable cost of generation of their own thermal stations, power purchase facility from other sources and the cost, system constraints and limitations etc. Based on the request from the beneficiaries as well as declaration from the generators, the RLDC shall prepare the generation schedules and drawal schedules from the CPS for each time blocks after taking into account the transmission constraints and other systems constraints if any. For transmission losses shall be apportioned among the beneficiaries in proportion of their drawls. In the case of any constraints, the RLDC have right to revise the schedule, which shall become effective from 4th time block.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

137

The generator and beneficiaries are also allowed to revise their schedule during a day, but such revisions shall be effective only from the 6th time block. VI. ISSUES CONCERNING UI In recent years a number of issues have seen to be arising in the UI mechanism .Some of these issues are as follows: i) Compatible software requirement: This is the most important issue faced by the management of power systems. Implementation of software which does all the functions of a power system is user friendly and acceptable to all the persons using it is very much important for the implementation unscheduled interchange. The software should be able to compute all the results desired, should be robust to address all regulatory issues and should be modifiable as per all the state regulatory commission requirements. ii) Avoidance of Gaming: The provision of gaming have been done away with a view to provide economic disincentive for over injection by a generating station other than hydro stations in excess of 105% of the declared capacity of the station in a time block or in excess of 101% of the average declared capacity over a day. There also are allegations made against this such as the one made by RRVPNL stating that the provisions would encourage generators iii) Trading of States surplus generation: Suppose a generating station is scheduled to give 100 MW of power to the state but the off-peak demand is only 90 MW in this case the generator has an option of either backing down on supply and save on energy charge anyhow by getting the actual payment required for scheduled generation. The other very attracting option is to trade the surplus 10 MW of power to a third party at a market determined rate without backing out the power supply. This option is most viable if the traded price is higher than the energy charge (mostly the case) which the generator would be getting because the generator is anyway paid for the capacity charge for the scheduled generation i.e. 100 MW. iv) Special meters & communication system requirement: According to ABT, UI has to be determined for each 15-minutes time block. This inherently requires metering of power supply and usage on every 15minutes time block at very interchange points. These meters are required for both inter-state and intra-state constituencies if the same time block is applied to both. As some states already had meters for 30-meters time block a question was raised that UI metering could be allowed for 30 minutes time range for intrastate UI. Corresponding to the same concern raised, a procedure was adopted as an interim arrangement for determination of UI charges for 15 minutes time block from readings given by 30 minutes recording meters. An advanced communication system is also a must requirement for the recorded datas to be transmitted to

the respective load despatch centres for timely decisions to be made. The responsibility of installation, testing and maintenance of meters and the installation of communication facility for the transmission of data rests with STU. To accelerate the process and bring about a complete transformation, a new transparent system is required to cause a paradigm shift in the electric industry in India. VII. GUIDELINES FOR INSTALLATION OF ABT
METERING EQUIPMENTS

The following guidelines are to be strictly implemented for installation of ABT compliant special energy meters (SEM). i) Metering arrangements, including installation, testing and maintenance of meters and installation of communication facility for the transmission of data shall be the responsibility of the STU on payment basis.STU shall ensure that all interface points are provided with ABT compliant meters and required communication facilities are installed before the stipulate date for implementation of ABT regime as per these Regulations. Collection and transportation of raw data to SLDC shall be the responsibility of individual licensee. Processing of the data required for accounting of energy exchanges and UI account based on average frequency of 15 minute time block shall be done by the SLDC.SLDC shall install required hardware and software for processing and analysis of date for necessary monitoring and control before the stipulated date for implementation of ABT regime as per these Regulations. Initial time synchronization of the meters and further checking and time synchronization, as and when required, shall be done by STU in coordination with the licensees and SLDC. All concerned entities (in whose premises the special energy meters are installed) shall fully cooperate with the SLDC and extend necessary assistance by taking weekly meter readings and transmitting them to the SLDC. On the basis of processed data of meters along with data relating to declared capacity and schedules etc., the SLDC shall issue the state level accounts for energy on monthly basis as well as UI charges on weekly basis. UI accounting procedures shall be governed by the orders of the Commission. In case of any dispute, the matter shall be referred to the Commission for decision, which shall be final. ii) The cost of metering shall be borne by the owners of the meters. Regarding the ownership of meters and its accessories, the entities involved shall follow Regulation-6 of CEA (Installation and operation of meters) Regulation, 2006. iii) ABT complaint meters interface meters as per CEAs (Installation and Operation of Meters) Regulations, 2006 shall be provided at the points of injection/drawl, by the respective users who shall also ensure communication of data to SLDC. The time

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

138

synchronization of the metering system shall be through Global Positioning System (GPS) with counter check from central billing station. Respective users shall bear their own expenses. iv) For the Purpose of Intra-State ABT, the existing metering CTs and PTs/CVTs installed and operated in the system shall be valid until these are replaced by those of specified accuracy and class as specified by CEA in its Regulations. v) All open access customers and transmission consumers shall provide ABT compatible. Special Energy Meters capable of time differentiated measurements (15 minutes) of active energy and voltage differentiated measurement of respective energy as may be specified by the STU or SLDC. vi) Metering and Accounting shall be as per the Tripura Electricity Grid Code as amended from time to time. vii) Where the entry and exit points are connected to the network of transmission system, the metering arrangements, including installation testing and operation and maintenance of meters and collection and transportation of data required for accounting of energy shall be organized by the STU VIII. CONCLUSION With the introduction of ABT, general improvements in grid parameters in terms of frequency and voltage have been observed. Generators and beneficiaries have become more sensitive commercially. However, some more time may be required to assess the commercial and operational achievement of ABT more conclusively. REFERENCES
[1] Newbery D., Regulation of the Electricity Sector: Comments on Some Alternative Models, presented in the International Seminar on Restructuring and Regulation of the Electric Power Sector, Buenos Aires, Argentina, 1995. http://www.powermin.nic.in The Electricity Act 2003.Constitution of India Central Electricity Regulatory commission (http ://www.cerc.gov.in) Christie Richard D., Wollenberg Bruce F. and Wangesteen Ivar, Transmission Management in the Deregulated Environment, Proceedings of IEEE, Vol.88, No.2, February 2000, pp.170-195. M. Shahidehpour, M. Alomoush, Restructured electric power systems: Operation, trading and volatility, New York: Marcel Dekker, Inc., 2001. Bhanu Bhushan, ABC of ABT_A Primer on Availability Tariff, 2005 Bhushan, A. Roy and P. Pentayya, The Indian Medicine, in Proc. IEEE PES General Meeting, June 6-10, 2004 M. Shahidehpour, H. Yamin, Z. Li, Market operations in electric power systems: Forcasting, scheduling and risk management, New York: John Wiley, 2002 Central Electricity Regulatory Commission, 2009. Unscheduled interchange charges and related matters regulation, 2009. CERC, New Delhi, India. Available online: http://www.cerc.gov.in Lai Loi Lai, Power System Restructuring and Deregulation, (book), Publisher: John Wiley and Sons, Ltd., New York, 2001. A. J. Wood and B. F. Wollenberg, Power Generation, Operation and Control, John Wiley and Sons, New York, 1996. P. Stephenson, M. Paun, Electricity market trading, Power Engineering Journal, vol.15, issue-6, pp.277-288, Dec.2001

[14]

[15]

[16]

Sood Yog Raj, Narayana Prasad Padhy and Gupta H.O.Wheeling of Power under Deregulated Environment of Power System-A Bibliographical Survey, IEEE Transactions on Power Systems, Vol.17, No.3, August2002, pp.870-878. Statement of Reason on UI Amendments dated 26.4.2010 taken from http://cercind.gov.in/Regulations/Unscheduled_SOR-UIfinal_26052010.pdf Bulk Power Tariff - with specific reference to Availability Based Tariff in the country by Shri B.K. Mishra, Member Secretary, Eastern Regional Electricity Board Kolkata http://www.sldcguj.com/compdoc/abtintro.doc

[2] [3] [4] [5]

[6]

[7] [8] [9]

[10]

[11] [12] [13]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

139

Synthesis of Renewable Clean and Green Energy: An Efficient Approach


1
1, 2

Tehseen Fatima , 2Naasir Kamaal Khan


1

Shri Jagdishprasad Jhabarmal Tibrewala University Jhunjhunu, Rajasthan, India


tehseenfatimaali@gmail.com 2 naasirk@gmail.com

Abstract Bio-fuel is one of the fastest emerging fuels in current scenario which is becoming strong competitor of present fossil fuel. Lot of research has taken out on it and many possibilities are still unexplored. Biodiesel is referred to as carbon neutral because when the oil crop grows it absorbs the same amount of CO2 as is released during fuel combustion. Here influence of various parameters such as reaction temperature, catalyst loading and molar ratio of oil to alcohol on the batch kinetics of the transesterification reaction is studied experimentally. Bio fuel is successfully synthesized from vegetable oil (triglycerides) by the process of transesterification via acid, base or enzymatic alcoholysis. Corn oil is esterified with methanol or ethanol via base catalysis to form methyl or ethyl esters known as biodiesel. As substrate corn oil was taken having 1.06.0 wt. % FFA. Catalyst effect at 1.0 1.25 wt. % was studied at a 1:8 oil/methanol molar ratio, final yield found to be 1.209%. As anticipated, the rate of reaction increases with increase in reaction temperature and catalyst loading. Summing up, the work shows much potential to synthesize the biodiesel from vegetable (corn) oil. Keywords- Biodiesel; Free fatty acid(FFA); Base catalyst; corn oil; Transesterification.

II. TRANSESTERIFICATION REACTION

I. INTRODUCTION There has been a significant increase in the production of biodiesel in the recent past. This interest has resulted in a record production of 450 million gallons biodiesel in the US in 2007 [14]. Most of the current production of biodiesel utilizes high quality soybean oil. Other sources of triglycerides (TG) such as waste cooking oil and yellow grease have also been investigated as raw material for biodiesel production. These alternative sources have been successfully converted to biodiesel with a promise of a cost effective fuel [2]. Nevertheless, processing challenges remain to be unraveled. Another low-cost source of TG may be found in distillers grains, the byproduct of dry grinded ethanol production. The annual production of distillers grains was 14.6 million metric tons in the year 2007 [17]. Distillers grains contain about 810 wt. % oil and are predominately used as an animal feed supplement. The oil content of distillers grains is beyond what is needed in the feed and if extracted is a good source of TG for biodiesel production. Transesterification of vegetable oils is the most effective process for the transformation of the triglyceride molecules into smaller, straight-chain molecules of fatty acid methyl or ethyl esters

The products of the reaction are the biodiesel (ester) and glycerol. It is an improved process for the production of fatty acid lower alkyl esters according to formula (I) R.sup.1CO--OR.sup.2 (I) in which R.sup.1CO represents a linear or branched acyl moiety having from about 6 to about 22 carbon atoms and 0 to 6 double bonds and wherein R.sup.2 is a linear or branched alkyl moiety having 1 to 6 carbon atoms, by transesterification of triglycerides or fatty acid esters using C.sub.1-C.sub.6 aliphatic alcohols, which includes the steps of (a) providing an aqueous mixture of aliphatic C.sub.1-C.sub.6 alcohol by fermenting in a first reactor a carbon source comprising carbohydrates and/or glycerol produce a fermentation both comprising aliphatic C.sub.1-C.sub.6 alcohol, (b) providing in a second reactor a triglyceride and/or an ester of a fatty acid, together with a biocatalyst capable of effecting a transesterification reaction, (c) transferring aqueous aliphatic C.sub.1-C.sub.6 alcohol mixture obtained from first reactor into second reactor to provide a two-phase system, and (d) effecting the transesterification reaction to produce fatty acid lower alkyl ester and a glycerol or alcohol containing aqueous phase. The products thus obtained are particularly useful as components of the fuel known as biodiesel [4]. In this transformation, the viscosity of vegetable oils is reduced from 10 to 15 times to about twice that of no. 2 diesel fuel. Other fuel specifications such as cetane number, energy content, and density of fatty acid esters are also similar to those of petroleum-based diesel fuel [13]. Moreover, biodiesel is essentially sulphur free and the engines fuelled by biodiesel emit significantly fewer Particulates, hydrocarbons, and less carbon monoxide than those operating on conventional diesel fuel engines. However, NO2 levels are slightly higher than those of diesel engines operating on conventional diesel fuels [18].The conventional biodiesel technology involves the use of an inorganic base or an acid catalyst at or near the

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

140

boiling temperatures of the TG/alcohol mixture. The transesterification of TG using acid catalysts has been signified by slow rate of reaction. High molar ratios of alcohol to oil and long reaction times are required for the acid-catalyzed transesterification of TG to reach high conversions. H2SO4 catalyzed transesterification of soybean oil with methanol, ethanol and 1-butanol. After 18h of incubation, the conversion of TG to esters was unsatisfactory at 1:6 and 1:20 molar ratios, while at a molar ratio of 1:30, high conversions were obtained after 69 h for all three alcohols. The effect of water on H 2SO4 catalyzed esterification of acetic acid and reported a significant decrease in the initial reaction kinetics due to catalyst impairment as water was produced from the condensation of acetic acid and methanol. The acidcatalyzed transesterification of soybean oil using 0.11.0 wt.% of sulfuric, hydrochloric, formic, acetic, and nitric acids. Sulfuric acid was reported to be the only effective acid where 99 wt.% conversions of TG was observed at 0.5 wt.% acid loading, 1:9 molar ratio of TG to alcohol, 100.C, and 8 h. The effect of the oil/alcohol molar ratio and free fatty acid (FFA) on the transesterification of soybean oil with H2SO4. They reported a strong inhibition of catalytic activity when the water concentration was increased to above 0.5 wt.%. Formation of water at or above this level was attributed to the presence of FFA at 5.0 wt.% in the initial soybean oil [11]. Most of the current production of biodiesel utilizes an alkaline catalyst which is mainly due to the fast rate of reaction and near complete conversion of reactants to biodiesel. A reaction mechanism consisting of a short initial mass transfer controlled region followed by a kinetically controlled region was proposed. The activation energies suggested the domination of Arrhenius kinetics for the forward reactions [15]. The effect of variables on base-catalyzed transesterification of soybean oil with methanol. An increase in the amount of catalyst (0.05 0.50 wt.%) and reaction time (1075 min) have little effect on the yield of fatty acid methyl esters (FAME), while increases in the rate of mixing resulted in increases in the FAME yields. In the base-catalyzed transesterification of TG, the need for a low concentration of FFA in the starting material has been stressed by researchers. An upper limit of 1% FFA for the base-catalyzed transesterification of TG. At FFA levels beyond 1%, the formation of soap prevented the separation of ester and glycerol products. Triglyceride sources containing more than 1% FFA are considered high FFA oils for basecatalyzed transesterification purposes [5]. The objective of this study was to develop a process for the conversion of distillers grains oil to biodiesel. A one stage base, a one-stage acid, and a two-stage acidbase catalyzed systems were examined. A two-stage acidbase scheme was then developed which utilized an intermediate removal of water and acid from the reaction media by a strong anionic (OH- or Cl- form) exchange resin prior to the base-catalyzed transesterification step [13]. The

procedure for the extraction of oil from distillers grains was also optimized. Samples used in the optimization experiments were model compounds made from pure corn oil and a technical grade linoleic acid (60%) with the balance consisting of oleic, linolenic, palmitic, and stearic acids. Actual corn oil extracted from condensed distillers soluble (CDS) was used to validate the results for the model substrates, Properties of biodiesel were tested according to ASTM D 6751 standard [2]. Reducing dependence on foreign energy by expanding domestic renewable fuels can have impacts for the overall U.S. economy because of energy importance in consumption, production, and trade. In the past, increasing energy independence would generally be expected to place a greater burden on the U.S. economy because of the higher domestic costs of producing alternative energy to replace relatively inexpensive foreign petroleum. However, according to the U.S. Department of Energy, petroleum prices are more likely to continue rising in the long term relative to the cost of producing domestic biofuels. Although the exact timing is uncertain, costreducing technology in bio-fuel production is expected to be a key factor in expanding production and making biofuels competitive with petroleum [6]. The Clean Cities Alternative Fuel Price Report for October 2010 is a quarterly report on the prices of alternative fuels in the U.S. and their relation to gasoline and diesel prices. This issue describes prices that were gathered from clean cities coordinators and stakeholders between October 4, 2010 and October 14, 2010, and then averaged in order to determine regional price trends by fuel and variability in fuel price within regions and among regions. The prices collected for this report represent retail, at-the-pump sales prices for each fuel, including Federal and state motor fuel taxes. The nationwide average price for regular gasoline has risen 7 cents per gallon from $2.71 per gallon to $2.78 per gallon; CNG price has risen 2 cents from $1.91 to $1.93; and ethanol (E85) has risen 14 cents from $2.30 to $2.44 per gallon. CNG is about 85 cents less than gasoline on an energy-equivalent basis, while E85 is about 67 cents more per gallon than gasoline on an energy-equivalent basis [10]. III. RESEARCH METHODOLOGY Materials : Alcohol (methyl , ethyl) Corn oi Sodium hydroxide Apparatus: Glass reactor vessel (1 L) Glass stirrer Separating funnel or gravity settler Magnetic stirrer Hot plate Thermometer Viscometer

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

141

A. Experimental setup Experimental setup consist of glass reactor vessel of volume 1L with arrangement of sample withdrawal at intermediate times. A glass stirrer for continuous stirring and baffles to avoid vortex formation. There must be arrangement for temperature and speed control. B. Transesterification reaction The required amount of oil was charged to the reactor. The required quantity of sodium hydroxide dissolves in methanol by magnetic stirrer. The solution of methanol and sodium hydroxide add to the oil in the reactor and the temperature set at 70oC. The reactor stir continuously and the methanol vapors generated were removed with the help of chilled water. The reaction can be typically carried out for 150 180 minute. After the reaction over the product mixture is unloaded from the reactor and subjected to the phase Separation in the gravity settler. Here after a settling time of 120 140 minute, ester layer is obtained at the top whereas the glycerol layer is obtained at the bottom. The glycerol discharged from the bottom whereas the ester washed with hot water. C. Alcohol Removal Once the glycerin and biodiesel phases have been separated, the excess alcohol in each phase is removed with a flash evaporation process or by distillation. In others systems, the alcohol is removed and the mixture neutralized before the glycerin and esters have been separated. In either case, the alcohol is recovered using distillation equipment and is re-used. Care must be taken to ensure no water accumulates in the recovered alcohol stream. D. Glycerin Neutralization The glycerin by-product contains unused catalyst and soaps that are neutralized with an acid and sent to storage as crude glycerin. In some cases the salt formed during this phase is recovered for use as fertilizer. In most cases the salt is left in the glycerin. Water and alcohol are removed to produce 80-88% pure glycerin that is ready to be sold as crude glycerin. In more sophisticated operations, the glycerin is distilled to 99% or higher purity and sold into the cosmetic and pharmaceutical markets. E. Methyl ester wash Once separated from the glycerin, the biodiesel is sometimes purified by washing gently with warm water to remove residual catalyst or soaps, dried, and sent to storage. In some processes this step is unnecessary. This is normally the end of the production process resulting in a clear amber-yellow liquid with a viscosity similar to petrodiesel. In some systems the biodiesel is distilled in an additional step to remove small amounts of color bodies to produce a colorless biodiesel.

F. HPLC analysis (Appendix-I) IV.RESULT Biodiesel from corn oil was synthesized. Corn oil (triglyceride) was converted into biodiesel (methyl ester) by using sodium hydroxide as a catalyst. Glycerol is also obtained after completion of reaction as by-product. Calculation: 100 gm (corn oil) + 800 gm (glycerol) =900 gm Glycerol obtained = 98 gm Methyl ester obtained = 10.881 gm % calculation for glycerol = 98 x 100 = 10.8% 900 % calculation for methyl ester (biodiesel) = 10.881 x 100 = 1.209% 900 Final yield = 1.209% IV. DISCUSSION In this work we found less yield of biodiesel. Methanol was used as reaction substrate in high quantity, but less quantity of biodiesel was obtained. During purification of biodiesel we found loss of yield. Hence, this process is not cost effective because it also needs big industrial setup. Most of reaction substrate was converted into glycerol (by-product). Glycerol is also useful product and can be used in pharmaceutical purpose, as Vaseline, as lubricant. In order to quantify these yield losses, the material balance of the process is determined by analyzing the methyl ester and glycerol phases. The material balance of the process, relative to the vegetable oil molar amount, included biodiesel molar yields and also the molar yield losses stemming from triglyceride saponification and methyl ester dissolution in the glycerol phase. All the calculated standard deviations were very low. As expected, the yield losses in the reactions with sodium or potassium methoxides were slight, 0.15 and 0.56 molar%, respectively. However, when sodium or potassium hydroxide was the catalyst, there were yield losses due to triglyceride saponification and methyl ester dissolution in the glycerol phase. As a result of the lower sodium hydroxide molecular weight in comparison with the corresponding weight of the potassium hydroxide, more soaps were produced during the saponification with sodium hydroxide, 5.65 molar%, than with potassium hydroxide, 3.46 molar%. This, in turn, involved an increase in the methyl ester proportion in the glycerol to 6.04 molar%, when sodium hydroxide was used as the catalyst. In comparison, the methyl ester dissolution in the glycerol was only 3.00 molar% after the reaction with potassium hydroxide. Nevertheless, when the transesterification was carried out with nearly equivalent amounts of potassium (1.5wt.%) and sodium hydroxide (1 wt.%), the quantity of saponified triglyceride was essentially the same: 5.65 and 5.15 molar%, respectively. However, the amount of methyl ester that dissolved in the

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

142

glycerol phase was higher after the reaction with sodium hydroxide (6.04 molar%) than after the reaction with potassium hydroxide (2.18 molar%). At the same time, after the reaction with sodium hydroxide, it was observed that the subsequent separation stage required more time because of the lower molecular weight of this catalyst. Accordingly, this effect was responsible for the higher amount of methyl ester found in the glycerol phase. V. CONCLUSION Petroleum diesel is a finite source of energy, so this work can take over the need of energy when conventional source exhausted. Have lot of advantages with some disadvantages. In todays era when the conventional petroleum source is going to be decline; in this situation we need some other renewable source of energy which can occupy the place of conventional petroleum diesel. Now a days the major problem is pollution which come from conventional fuel, this product can overcome this problem, because carbon monoxide emission is less with biodiesel. Approximate 99% particulate matter is less emitted in environment. However, catalysts are more expensive and also more difficult to manipulate since they are very hygroscopic in nature. The biodiesel yields for the sodium and potassium hydroxide were lower, respectively, because the yield losses were more substantial. Nonetheless, these yields can be higher through a modification of the value for the experimental conditions, such as, temperature and catalyst concentration. REFERENCES
Ali Y, Hanna MA, Leviticus LI (1995) Emissions and power characteristics of diesel engines on methyl soyate and diesel fuel blends. Bioresour Technol 52:185195. [2] Canakci M, Van Gerpen J (2001) Biodiesel production from oils and fats with high free fatty acid. Trans ASAE 44:14291436 [3] Canakci M, Van Gerpen J (2003) A pilot plant to produce biodiesel from high free fatty acid feed stocks. Trans ASAE 46:945954 [4] Dubreucq, Eric,(2010), Process for the production of fatty acid alkyl esters cognis IP management, US patent pub no. 2010/0234458. [5] Freedman B, Pryde EH, Mounts TL (1984) Variables affecting the yields of fatty esters from transesterified vegetable oils. J Am Oil Chem Soc 61:16381643. [6] Gahlhar M; Winston, A;somwaru,A. 10801, Effect of increased biofuels on the U.S. economy in 2022, October 2010. [7] Goff MJ, Bauer NS, Lopes S, Sutterlin WR, Suppes GJ (2004) Acid-catalyzed alcoholysis of soybean oil. J Am Oil Chem Soc 81:415420 [8] Hancsok J, Kovacs F, Krar M (2004) Production of vegetable oil fatty acid methyl esters from used frying oil by combined acidic/ alkali transesterification. Petroleum Coal 46:3644 [9] Issariyakul T, Kulkarni MG, Dalai AK, Bakhshi NN (2006) Production of biodiesel from waste fryer grease using mixed methanol/ethanol system. Fuel Process Technol 88:429436 [10] Laughlin,MD, 10795, clean cities alternative fuel price report, November 2010. [11] Liu Y, Lotero E, Goodwin JG Jr (2006) Effect of water on sulphuric acid catalyzed esterification. J Mol Catal A Chem 245:132 140 [1]

[12] Muniyappa PR, Brammer SC, Noureddini H (1996) Improved conversion of plant oils and animal fats into biodiesel and coproduct. Bioresour Technol 56:1924 [13] Mittelback M, Tritthar P (1988) Diesel fuel derived form vegetable oils. III. Emission tests using methyl esters for used frying oil. J Am Oil Chem Soc 65:11851187 [14] National Biodiesel Board, http://www.biodiesel.org/ [15] Noureddini H, Zhu D (1997) Kinetics of transesterification of soybean oil. J Am Oil Chem Soc 74:14571463 [16] Noureddini H, Santosh R. P. Bandlamudi and Emily A. Guthrie(2009), A Novel Method for the Production of Biodiesel from the Whole Stillage-Extracted Corn Oil Journal of the American Oil Chemists' Society Volume 86, Number 1, 83-91. [17] Renewable Fuels Association, http://www.ethanolrfa.org/industry/ resources/coproducts/ [18] Schumacher LG, Borgelt SC, Fosseen D, Goetz W, Hires WG (1996) Heavy-duty engine exhaust emission tests using methyl ester soybean oil/diesel fuel blends. Bioresour Technol 57:3136 [19] Wang Y, Ou S, Liu P, Xue F, Tang S (2006) Comparison of two different processes to synthesize biodiesel by waste cooking oil. J Mol Catal A Chem 252:107112 [20] Zheng S, Kates M, Dube MA, McLean DD (2006) Acid-catalyzed production of biodiesel from waste frying oil. Biomass Bioenergy 30:267272.

APPENDIX-I HPLC Analysis:

Chromatograph

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

143

Bio-diesel from Jatropha Curcas: Does Represent a Sustainable Alternative Energy Source?
Kalyan Pal
Hi-Tech Institute of Engineering & Technology, U. P. Technical University 766, Delhi-Hapur Bypass Road (NH-24), Ghaziabad-201002, India
kalyanpal@yahoo.com Abstract: There has been greater awareness on bio-diesel in developing countries in the recent times and significant activities have picked up for its production especially with a view to boost the rural economy. The properties of bio-diesel depends on the nature of the vegetable oil to be used for preparation of bio-diesel and if the developed process is scaled up to commercial levels then excellent business opportunity will be offered by the bio-diesel obtained from Jatropha oil methyl ester and it could be a major step towards the creation of an eco-friendly transportation fuel that is relatively clean on combustion and provides farmers with substantial income. A review of the following topics is proposed in order to improve the sustainability of the process: areas destined for cultivation, use of external (chemical) inputs in cultivation, processes for converting the vegetable oil to bio-diesel and above all, the location for ultimate consumption of the bio-fuel. Keywords: bio-fuel; Jatropha curcas; petroleum

I.

INTRODUCTION

The consumption of fossil fuels, above all petroleum and mineral coal, has been the foundation of the economic growth and the betterment of general levels of well-being for the so-called leading economies throughout the last century. At the same time, in the peripheral economies, like those of Latin America, a phenomenon has been presented of production of material primarily for exportation and to a lesser extent for internal transformation and consumption. There are two essential concerns which put at risk the viability of the model of growth based on export and/or excessive consumption of energy, both related with the lack of sustainability in said model: the apparent decreasing availability of fossil fuels and the notable increase in greenhouse gasses in the atmosphere caused most probably by human activity. With regard to the proven and potential reserves of petroleum in the world, undoubtedly there exists a decline in the former since the decade of 19701980 [1-3], although apparently this is not significant [4, 5]. During 2008 the international prices of petroleum rose to historically high levels, a situation utilized as an argument for the near-term emptying of the fuel reserves and for planning for the so-called energy transition, which includes, among other topics, the investigation and use of alternative sources of energy. Nevertheless, at the beginning of 2009, said prices fell substantially, which indicates that the price does not depend exclusively on the availability of petroleum, but also on multiple financial

and geopolitical interests [5]. Over the potential reserves (possible and probable), the exploration in various parts of the world [5, 6] and the application of new technologies [7-9] shows that it will be possible to extract petroleum at competitive costs for at least all of the 21st century. On the other hand, the significant emission of carbon dioxide (and other greenhouse gasses) into the atmosphere by the combustion of petroleum in various human activities and its repercussion on global climate is a fact which only in the last two decades has been taken into account [10], but which recently has become a preoccupation even for petroleum exporting countries [5]. Proposals exist to reduce the emissions of CO2, which include the design and use of more efficient machines and processes. Without considering if the foregoing may be achieved, the environmental economy postulates that technological solutions do not exist ad infinitum for the problems of the environment, and the second law of thermodynamics teaches that no completely efficient machines can exist. The two discussed preoccupations have driven the investigation of alternative sources of energy, of which there exists a large list [11], although, since the substitutes most akin to petroleum are sought after, the options most prominent in the last decade are the liquid bio-fuels, specifically ethanol [12] and bio-diesel [13-15]; these also have the advantage of being of photosynthetic origin (simple or complex sugars in the case of ethanol and greases or oils in the case of bio-diesel) and, also, renewable, as part of the carbon cycle. This present essay is focused on the bio-diesel product from the oil of the tropical plant Jatropha curcas L. (Euphorbiaceae). There are a lot of oleaginous plants with potential for producing bio-diesel, but J. curcas is frequently mentioned as the best option. Among several advantageous characteristics of this species is important to underline its adaptability to marginal soils and environments [13]. Bio-diesel is the only alternative fuel that can be used directly in any existing unmodified diesel engine. Because it has similar properties to diesel fuel, bio-diesel can be blended at any ratio with diesel fuel. In most of the developed countries, bio-diesel is produced from soybean, rapeseed, sunflower, peanut, etc., which are essentially edible in Indian context. Among the various vegetable oil sources, non-edible oils are suitable for bio-diesel production. Among the non-edible oil sources, Jatropha curcas is identified as potential bio-diesel source and

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

144

comparing with other sources, which has added advantages as rapid growth, higher seed productivity, suitable for tropical and subtropical regions of the world. In the present work the current situation of proven and potential reserves of fossil fuel, and the production and consumption model for the same are analyzed, in order to later review the sustainability of the production process which begins with the cultivation of Jatropha curcas, and culminates with the consumption of bio-diesel. II. THE SUSTAINABILITY OF BIO-DIESEL From the above, one concludes that governments and energy companies are not formulating profound changes in the model of consumption of petroleum fuels, but only technological solutions for (i) extending as much as possible the extraction of petroleum at non-prohibitive costs, (ii) minimizing the emission of CO2, and (iii) substituting for petroleum with liquid bio-fuels. Furthermore, bio-diesel cannot replace petroleum, in the first place because it cannot be produced, without causing environmental damages greater than those for which it intends to give a solution [16], on a scale similar to that of petroleum, in accordance with the present and projected demand; in the second place because from bio-diesel is not possible to extract the multitude of by-products which are produced by the petrochemical industry. One fact which exemplifies the contradiction between a preoccupation with CO2 emissions into the atmosphere and the plan to continue the model of economic development is that the leading economies, principally the United States, postulate that bio-fuels, in particular biodiesel, will permit a reduction in imports of petroleum [17], while the peripheral economies, especially those located in tropical zones, plan for the production of biodiesel for export [18]to countries like the United States and China, which are the largest emitters of CO2 into the atmosphere. Another pattern in growth is the reexportation of bio-diesel [19]; the vegetable oil is exported to one country, converted into bio-diesel, and then reexported to another country. In any case, the photosynthetic origin of bio-diesel is sufficient to justify its investigation as a renewable fuel, not only as a simple technological development, but also situating it as an element to play its part in the so-called global ecological crisis (sensu Iranzo [20]), taking into account the multitude of factors which affect the sustainability of its use. In this sense, Ovando et al. [21], put forward that the rising world interest in bio-diesel, specifically that produced from pin (J. curcas), can lead to non-sustainable practices. For this reason, they propose giving attention to at least four aspects for their importance in sustainability of the production process of pin-consumption of bio-diesel: (a) the area dedicated for cultivation, (b) the use of inputs in the cultivation, (c) the processes for converting vegetable oil to bio-diesel, and d) the final destination of the bio-fuel.

Achten et al. [22] performed a qualitative evaluation of the future sustainability of cultivating J. curcas, focusing on the environmental impacts and socioeconomic aspects; they determined that the cultivation is sustainable when practiced in marginal or degraded lands, but not when fertile areas are dedicated which could serve to cultivate foodstuffs or other crops with greater profitability. One logical conclusion from the work of Achten et al. is that if areas of forest are clear-cut to convert them into pin cultivation, the sustainability of the process is nonexistent. Nevertheless, it is possible to use areas considered as agriculturally high quality, if only one takes advantage of the periphery of the principal crops. It has been calculated, for example, that in the Mexican state of Chiapas, the cultivation of pin could occupy a maximum of 230,000 hectares on the perimeters of the cultivated areas or used for cattle-grazing [21]. It was mentioned before that bio-diesel cannot substitute for petroleum, not even for diesel, at the actual and projected levels of consumption; for example, the National Mission for Biodiesel in India has proposed the planting, by the year 2020, of an area of ten million hectares with J. curcas to produce 7.5 million tons of bio-diesel each year; this volume represents only 20% of the demand for diesel in that country [3]. To substitute bio-diesel for the actual demand for diesel in Mexico (17.4 million metric tons [23]) would require close to 23 million hectares planted with J. curcas, which is equivalent to more than 90% of the cultivable ground surface. In this country there is a recent legislation that promotes the investigation and use of bio-fuels [24]; however, the corresponding bylaw has not yet been published. To minimize the use of fossil energy and improve the energy balance it is required that the cultivation not include the use of chemical fertilizers, since these represent an elevated level of energy consumption. The data shows that 45% of commercial energy used in global agricultural production is due to the consumption of chemical fertilizers [25]; for that reason, implementing a bio-energy cultivation consuming a huge quantity of energy is, at the least, contradictory. For example, according to the same author, ammonia, the principal source of nitrogenous fertilizers, is produced from natural gas, and the petrochemical industry, which synthesizes it, consumes 1.2% of fossil fuels extracted on the global level. The post-harvest processes for the majority of agroindustrial products imply a high consumption of energy derived from the use of indispensable machines; doubtless, in this case one should pay special attention to the energy efficiency of the machines for separating the husk of the seed, the extraction of the oil and the conversion to bio-diesel. In this way, the husk can be used for fuel to heating the cauldrons if one opts for extraction of oil by extrusion by heat, or for heating the transesterification reactors for conversion to bio-diesel. Regarding this last, it is greatly relevant to mention that

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

145

although the reaction of transesterification has a high yield level (80%, that is to say that 800 mL of bio-diesel are obtained from every liter of vegetable oil); the use of methanol reduces the energy gain. Methanol, which industrially is obtained from the distillation of petroleum, requires a proportion of 200 mL for each liter of processed oil. Among the alternatives one encounters the use of ethanol, which derives from the fermentation of sugar; although it has the inconvenience that on production it is dissolved in water with a distillation yielding a maximum of 96% alcohol to 4% water. The conversion of vegetable oil to bio-diesel is favored by the absence of water, or by minimal quantities thereof. The production of completely anhydrous ethanol raises both costs and energy consumption. Other topic requiring attention is the content of toxic substances of the J. curcas seed. Several toxic molecules have been reported in the seed [26-28], but the protein curcin and the phorbol esters are the most hazardous for human and animal health. After the oil extraction, the seed cake still contains those substances, representing a potential risk for the J. curcas bio-diesel workers [29]. The potential of phorbol esters as carcinogens is known [30]. Although non-toxic genotypes of pin exist in several parts of Mexico or improved non-toxic varieties could be obtained (by conventional breeding or by transgenic methods), toxic genotypes are being used for establishing extensive plantations around the world. However, a dilemma exists: if non-toxic genotypes are used, problems with pests could be a limitation, as the plant-herbivores interaction would be substantially modified. Alternatives to use the press cake are the physical or enzymatic detoxification for using as fodder, and the composting for using in the same plantation. III. CONCLUSIONS The aspects mentioned only refer to challenges to the sustainability of production, which can be overcome by means of technological development, while the true challenge is a reflection over the viability of the model of exporting peripheral economies vs. leading economies in the consumption of fuel: namely, that given that the transport of bio-diesel by whatever means entails the consumption of energy, then in order to maximize the energy efficiency of the process, the production of biodiesel derived from pin should have a focus on local use in the rural environment, or in urban areas close to the sites for cultivation and industrialization of J. curcas, more than any export focus. The energy balance for a biofuel is a comparison of the energy contained in the fuel with the energy required to produce, process, and distribute it. It does not mean that the proposal tends toward the socalled Arcadian ecology (sensu Iranzo [20]), but this could provide rural communities with energy to stimulate processes of improvement which at present they do not possess, whether due to lack of services or energy

products or due to the high costs of these; for example, in some cases it would improve the preservation of foodstuffs at lower temperatures. Since transporting biofuels across large distances is not a sustainable result, the production for export, and later re-export, should be limited. Our criticism to current approaches for the production and consumption of J. curcas bio-diesel is not intended to disqualify the efforts of many people, but to contribute to a more sustainable productive chain. Of course, as Khan and Islam [31] explain, sustainability is not achieving only by minimizing risks and remediating problems engendered by the introduction of a given process or technology, but visualizing future potential problems, that is, having the time as main variable. REFERENCES
[1] [2] Cavallo, A.J. Predicting the Peak in World Oil Production. Natural Resources Research 2002, 11, 187-195. Bakhtiari, A.M.S. World oil production capacity model suggest output peak by 2006-07. Oil and Gas Journal 2004, Edition of April 26th, 18-20. Kumar-Prusty, B.A.; Chandra, R.; Azeez, P.A. Biodiesel: Freedom from Dependence on Fossil Fuels? Nature Precedings 2009, doi:10.1038/npre.2008.2658.1. Gmez-Cabrera, J.A. Retos y expectativas de la oferta y demanda de hidrocarburos en Mxico: Trabajo de Ingreso; Mexican Academy of Engineering: Mexico, D.F., Mexico, 2008; p. 99. World Oil Outlook 2008; Organization of the Petroleum Exporting Countries: Vienna, Austria, 2008; p. 234. Lpez, A.M. La gran tentacin: El petrleo de Mxico; Random House Mondadori (Grijalbo), Mexico, 2008; p. 207. Mendoza, A.; Miranda, F.P.; Banneman, K.; Pedroso, E.; Lpez, O.; Herrera, M. Satellite environmental monitoring of oil spills in the South Gulf of Mexico. In Proceedings of the Offshore echnology Conference; OTC: Houston, TX, USA, 2004; p. 7. Kunzig, R. The Canadian oil boom: Scraping Bottom. National Geographic Magazine 2009, 215, 38-59. Canadian Oil Sands Trust, Annual Report 2008; Canadian Oil Sands Ltd.: Calgary, Alberta, Canada, 2009; p. 88. Contribuciones del Fondo para el Medio Ambiente Mundial al Programa 21: El primer decenio; Fondo Mundial para el Medio Ambiente: Washington, DC, USA, 2000; p. 57. Douglass, S.E. Identifying the Opportunities in Alternative Energy; Wells Fargo Bank: San Francisco, CA, USA, 2005; p. 17. Pimentel, D.; Patzek, T. Ethanol production using corn, switchgrass and wood; Biodiesel production using soybean and sunflower. Natural Resources Research 2005, 14, 65-66. Gubitz, G.M.; Mittelbach, M.; Trabi, M. Exploitation of the tropical oil seed plant Jatropha curcas L. Bioresource Technol. 1999, 67, 73-82. Openshaw, K. A review of Jatropha curcas: an oil plant of unfulfilled promise. Biomass and Bioenergy 2000, 19, 1-15. Pramanik, K. Properties and use of Jatropha curcas oil and diesel fuel blends in compression ignition engine. Renewable Energy 2003, 239-248. Popluechai, S.; Breviario, D.; Sujatha, M.; Makkar, H.P.S.; Raorane, M.; Reddy, A.R.; Palchetti, E.; Gatehouse, A.M.R.; Syers, J.K.; ODonell, A.G.; Kohli, A. Narrow genetic and apparent phenetic diversity in Jatropha curcas: initial success with generating low phorbol ester interspecific hybrids. Nature Precedings 2009, doi: hdl:10101/npre.2009.2782.1. Biodiesel Handling and Use Guide, 4th ed.; National Renewable Energy Laboratory/US Department of Energy: Oak Ridge, TN, USA, 2008;

[3]

[4]

[5] [6] [7]

[8] [9] [10]

[11] [12]

[13]

[14] [15]

[16]

[17]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

146

[18] Adreani, P. Biodiesel in Argentina. In Proceedings of the National Biodiesel Conference; National Biodiesel Board: Orlando, FL, USA, 2008. [19] Zilio, J. Biodiesel economics in Brazil. In Proceedings of the Biofuels Seminar; Brazilian Vegetable Oil Industries Association: Sao Paulo, Brazil, 2005. [20] Iranzo, J.M. Inquietudes humanas, problemas cientficos y soluciones tecnolgicas: ciencia, tecnologa y poltica en [lainexistencia de la] crisis ecolgica global. Poltica y Sociedad 1993, 14-15, 99-114. [21] Ovando, I.; Adriano, L.; Salvador, M.; Ruiz, S.; Vzquez, A. Pin (Jatropha curcas): Bioenerga para el desarrollo de Chiapas. Biotecnologa Agropecuaria y Biodiversidad en Chiapas 1999, 2. [22] Achten, W.M.J.; Mathijs, E.; Verchot, L.; Singh, V.P.; Aerts, R.; Muys, B. Jatropha biodiesel fueling sustainability? Biofuels, Bioproducts and Biorefining 2007, 1, 283-291. [23] Martnez, A. El futuro de los biocombustibles biotecnolgicos. In Memorias del Foro Avances en Bioenerga; Universidad Iberoamericana: Mexico DF, Mexico, 2008. [24] Gamboa, M.C. Biocombustibles, estudio terico conceptual, iniciativas presentadas en la LX Legislatura, derecho comparado y opiniones especializadas; Centro de Documentacin, Informacin y Anlisis, H. Cmara de Diputados: Mexico, D.F, Mexico, 2009; pp. 10-20. [25] Bhattacharjee, P.K. Sustainable Fertilizer and Crop Production from Energy Security Perspective -An Overview. In Annual

[26]

[27]

[28]

[29]

[30]

[31]

AlChE (Central Florida) Clearwater Convention; AIChE: Clearwater, FL, USA, 2006; p. 28. [Martinez-Herrera, J.; Siddhuraju, P.; Francis, G.; Davila-Ortiz, G.; Becker, K. Chemical composition, toxic/antimetabolic constituents, and effects of different treatments on their levels, in four provenances of Jatropha curcas L. from Mexico. Food Chemistry 2006, 96, 80-89. Makkar, H.; Becker, K.; Schmook, B. Edible provenances of Jatropha curcas from Quintana Roo state of Mexico and effect of roasting on antinutrient and toxic factors in seeds. Plant Foods Hum. Nutr. 1998, 52, 31-36. Makkar, H.; Becker, K.; Sporer, F.; Wink, M. Studies on nutritive potential and toxic constituents of different provenances of Jatropha curcas. J. Agr. Food Chem. 1997, 45, 3152-3157. Achten, W.; Verchot, L.; Franken, Y.; Mathijs, E.; Singh, V.; Aerts, R.; Muys, B. Jatropha biodiesel production and use. Biomass and Bioenergy 2008, doi: 10.1016/j.biombioe.2008.03.003. Goel, G.; Makkar, H.; Francis, G.; Becker, K. Phorbol esters: structure, biological activity, and toxicity in animals. Int. J. Toxico. 2007, 26, 279-288. Khan, M.I.; Islam, M.R. True Sustainability in Technological Development and Natural Resource Management; Nova Science Publishers: New York, NY, USA, 2007; pp. 2-9.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

147

Simulation Model for SVPWM Inverter in Grid Connected Photovoltaic System


1 1

Guriqbal Singh, 2S.K.Bath, 3Navdeep Singh


1, 2, 3

GZSCET Bathinda, India


2

er_guriqbalsingh@yahoo.co.in,

navi_nihmal85@yahoo.com

AbstractA SVPWM inverter in Grid connected Photovoltaic system with photovoltaic array (PVA) simulation model to be used in Matlab-Simulink is developed and presented. The model is developed using basic circuit equations of the photovoltaic (PV) solar cells including the effects of solar irradiation and constant temperature 25oC. The model is tested using a directly coupled dc load as well as ac load. Test and validation studies with proper load matching circuits are simulated and results are presented here. KeywordsPhotovoltaic Array, Photovoltaic grid connected systems, Power generation, IGBT Model, Space Vector technique, Filter. I. INTRODUCTION Renewable energy sources such as wind, photovoltaic (PV) and geothermal have received much attention recently as alternative means of generating electricity. In particular, small scale PV systems are increasing in numbers due to decreasing costs, and efficiency improvements [1], which are convenient for local power generation. The generated power in PV cells can be used in a standalone system or can be fed to the AC main grid. In standalone systems, the output power of the PV system can also be stored in batteries. However, the battery systems are expensive, bulky and require high maintenance. Where utility power is also available, another solution is to feed the power into the grid, which requires a grid connected inverter . With a Grid connected inverter, excess power is bought and credited by the utility, and grid power is available at times when the local demand exceeds the PV system output. Photovoltaic (PV) systems that supply power directly to the grid are becoming more popular due to the cost reduction achieved from the lack of a battery subsystem. The use of new efficient photovoltaic solar cells (PVSCs) has emerged as an alternative measure of renewable e nergy conservation and demand-side management. Owing to their initial high costs, PVSCs have not yet been a fully attractive alternative for electricity users who are able to buy cheaper electrical energy from the utility grid. However, they have been used extensively for water pumping and air conditioning in remote and isolated areas where utility power is not available or

is too expensive to transport. Although PVSC prices have decreased considerably during the last years due to new developments in the film technology and manufacturing process.The unregulated output power of renewable energy sources should be regulated through inverters to satisfy the conditions for connection to the grid[2]. The conFiguration of a three phase grid connected PV system is illustrated in Fig 1. It consists of solar PV array, SVPWM inverter, SVPWM generator, low-pass filter and grid voltage source. In the PV array i.e solar PV modules are connected in a parallel conFiguration to match the required solar voltage and power rating. The Three phase inverter converts the DC input voltage into AC sinusoidal voltage by means of appropriate switch signals and then the output pass through an isolation step up transformer to setup the output voltage to the filter and then required voltage by the electric utility grid and load [8].

Fig. 1: Block Diagram of grid connected PV system

The PVS model proposed in this paper is a circuitry based model to be used with Simulink. The proposed model was simulated with various types of loads for performance checking. II. GENERALISED MATLAB/SIMULINK MODEL
OF SVPWM

Step by step development of a Matlab/Simulink based simulation model for implementing SVPWM is presented here. The presented simulation model is general in nature and can be conFigured very easily to simulate SVPWM in linear modulation range [12]. The simulation model with 3-Phase transformer is shown in Fig3 are described in the following sub-sections.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

148

Fig. 2: MATLAB/ SIMULINK model of SVPWM inverter In grid connected photovoltaic system

A. PV Array Block PV arrays are built up with combined series combinations of PV solar cells, which are usually represented by a simplified equivalent circuit model. In this simulation of PV system which can be connected as a DC source to the distribution power network. It means photovoltaic array[10]. The proposed model takes sunlight irradiance and cell temperature as input parameters and outputs the I-V and P-V characteristics under various conditions in following Fig3.1 and Fig3.2.

The waveform of magnitude is simply a constant line as its value remains fixed. The waveform of angle is shown in Fig 4.1 and the magnitude waveform shown in Fig 4.2 [4].

Fig. 4.1: Voltage Angle

Fig. 3.1: V-I Characteristics of PV array

Fig. 4.2: Voltage Magnitude

C. SVPWM Generator Block The block utilizes the Space Vector Pulse Width Modulation (SVPWM) technique to generate the firing pulses to the six switches of the converter. Two pattern switching can be selected. The modulation index (m) must be in the range 0 to 1.The objective of the SVPWM technique is to utilisation approximately the reference voltage vector (U*) instantaneously by combination of the switching states corresponding to the basic space vectors [7] [9] [10]. D. Three-Phase Inverter Block and Transformer Block This block is built to simulate a 3-Phase Voltage Source Inverter (VSI) assuming constant dc link voltage. The inputs to the inverter block are the switching signals and the outputs are the SVPWM voltages. The inverter model

Fig. 3.2: P-V Characteristics of PV array

B. Voltage transformation block The 3-Phase voltage is converted into two-phase voltage equivalent using Clarks transformation equations as follow: The two outputs of this block are (i) magnitude of the reference and (ii) the corresponding angle of the reference.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

149

is built using function blocks. The outputs of SVPWM inverter voltages [3][5][11]. Three phase transformers are used throughout industry to change values of three phase voltage and current[12]. Since 3-phase power is the most common way in which power is produced, transmitted and used. In this section the different types of three phase transformers connections. There are only 4 possible transformer combinations: Delta to Delta- used in industrial applications. Delta to Wye - used in most common; commercial and industrial. Wye to Delta - used high voltage transmissions. Wye to Wye - used rare, don't use causes harmonics and balancing problems.
Fig. 8: Line to line voltage at the output of the Transformer with load.

Fig. 6: SVPWM inverter Voltage waveforms

Three-phase transformers are connected in Delta or Wye conFigurations. A Wye-Delta transformer has its primary winding connected in a Wye and its secondary winding connected in a Delta. A Delta-Wye transformer has its primary winding connected in Delta and its secondary winding connected in a Wye. Here in this Simulink model Delta-Wye conFiguration is used. The outputs of SVPWM inverter with 3- phase transformer in following Figures.

E. Filter Blocks To visualize the actual output of the inverter block, transformer block filtering of the SVPWM ripple is required. The PWM voltage signal is filtered here using first order filter. This is implemented using Transfer function block[14][15]. The time constant of the firstorder filter is chosen 0.8ms. This value is chosen to ensure the cut-off frequency of the low-pass equal to 1250 Hz. So the designed filter will effectively remove the ripple of higher than 1.25 kHz. The value of the filter time constant can be changed if the desired cut-off frequency is different [6]. The filtered voltage outputs are shown in following Fig9 and Fig10.

Fig. 9: Filtered Line to line voltage at the output of the Transformer without load.

Fig.7: Line to line voltage at the output of the Transformer without load.

Fig. 10: Filtered Line to line voltage at the output of the Transformer with load.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

150

Since the main objective is the development of the PVA functional model for the Simulink environment, the other parts of the operational block diagram given in Fig 2 are not going to be explained in full detail. However, just to describe the main diagram, as it can readily be seen, the system is modeled to supply power to both dc and ac loads. The dc load is to be directly coupled while the ac load is fed through a three-phase inverter and an isolation transformer with a turn ratio 1and with appropriate filter circuit to reduce the harmonics [13]. IV. SIMULATION RESULTS The proposed PVA model is simulated using the scheme given in Fig 2. The system supplies power to a mainly resistive dc load and an RLC ac load. The system does not have any controller. The loads are just chosen to match the power generated by the PVA. Actually the voltage at dc load bus and the both voltage and frequency at ac load bus must be controlled and kept constant for the users. The control part is also done as a part of this work. However, the PV Array is included in this paper since both parts require more space. The current-voltage the I-V and P-V characteristics of PVA are given in Figs.3.1 and 3.2. Although the maximum power is above1 2 7 5 W for the current solar irradiation and the operating power is approximately 1275 W. The dc bus voltage is given in Fig4.2. The output ac voltage of the inverter is shown in Fig6.where the S V PWM mode operation of the inverter can be seen on the output voltage. The output voltage of the inverter is applied to the load over an isolation transformer with the turn ratio 1. The effects of the transformer on the voltage can be seen in Fig 7and with load in Fig 8. where the there phase line to line voltages have sinusoidal wave shapes including some harmonics. The harmonics can be eliminated or reduced by applying proper filter circuits, which are considered here in Fig.9 and Fig10.

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12] [13]

[14]

[15] V. CONCLUSION This paper introduces a simulation model for photovoltaic arrays (PVA) to be used in Matlab-Simulink GUI environment. The proposed model has a generalized structure so that it can be used as a PV power generator along with wind, fuel cells and small hydro system by establishing proper interfacing and controllers. The model is simulated connecting a three phase inverter showing that, the generated dc voltage can be converted to ac and interfaced to ac loads as well as ac utility grid system. Therefore the model proposed here can be considered as a part of distributed power generation systems.

Broeck H W, et al. Analysis and Realization of a Pulse width Modulator Based on Voltage Space Vectors. IEEE Trans. on IA , vol. 24, No. 1:142-150, 2002 Arman Roshan, A dq rotating frame controller for threee phase fullbridge inverters used in small distributed generation systems, M.Sc.thesis, Faculty of the Virginia Polytechnic Institute and State University, Jun., 2006. Dehbonei H., Borle L. and Nayar C.V., A review and a proposal for optimal harmonic mitigation in three phase pulse width modulation, Proceedings of 4th IEEE International Conference on Power Electronics and Drive Systems, 2001, Vol. 1, Oct., 2001, pp. 408 414. Peterson K. Hinga, Tokuo Ohnishi and Takayuki Suzuki, A New PWM Inverter for Photovoltaic Power Generation System. IEEE Conference on Power Electronics Specialists. Vol. 1, pp: 391-395, 1994. Martina Calais, Vassilios G. Agelidis, Multilevel inverters for Single-Phase Grid connected Photovoltaic Systems-An Overview. IEEE International Symposium on Industrial Electronics. pp: 224229, 1998. A. M. Sharaf and L. Wang: A photovoltaic powered efficient DC motor drive for pump irrigation, Proc. Canad. Solar Energy Conf., Halifax N.S., Canada, 1990. I. H. Altas and A. M. Sharaf: A Fuzzy Logic Power Tracking Controller For A Photovoltaic Energy Conversion Scheme, Electric Power Systems Research Journal, Vol.25, No.3, pp.227-238, 1992. J.H.EcksteinDetailed modeling of photovoltaic components MS thesis, Solar Energy Laboratory, University of Wisconsin, Madison, (1999) Qingrong Zeng, Liuchen Chang, Novel SVPWM Based Predictive Current Controller for Three-phase Grid Connected Inverters. Canadian Conference on Electrical and Computer Engineering. pp: 1262-1265, 2005. Juan Jose Negroni, Carlos Meza, Domingo Biel, Control of a Buck Inverter for Grid- Connected PV Systems: a Digital and Sliding Mode Control Approach. IEEE International Symposium on Industrial Electronic. Vol. 2, pp: 739-744, 2005. I. H. Altas, A. M. Sharaf, - A photovoltaic array simulation model for Matlab-Simulink GUI environment IEEE, (2007). R. Candela, V. Di Dio, E. Riva Sanseverino, P. Romano ReconFiguration techniques of partial shaded PV system for the maximization of electrical Energy production IEEE-ICCEP 2007 International Conference on Clean Electrical Power, Capri, Italia, 21-23 May 2007 Dixon, J.; del Valle, Y.; Orchard, M.; Ortuzar, M.; Moran,L.; Maffrand, C. A Full Compensating System for General Loads, Based on a Combination of Thyristor Binary Compensator and a PWM-IGBT Active Power Filter, IEEE Trans. Industrial Electronics, vol. 50, issue 5, pages 982-989. October 2003. Barrero, F.; Martinez, S.; Yeves, F.; Martinez, P.M. Active power filters for line conditioning: a critical evaluation, IEEE Trans. on Power Delivery, vol. 15, issue 1, pages 319-325. January 2000.

REFERENCES
[1] M. Depenbrock, Pulse width control of a 3-phase inverter with non sinusoidal phase voltages, in Proc. IEEE-IAS Int. Semiconductor Power Conversion Conf., Orlando, FL, 1975, pp. 389398.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

151

Design of Smart Off-Grid Energy System


1
1, 2, 3

Md. Farman , 2D. K. Khatod , 3Arun Kumar


1

Indian Institute of Technology Roorkee, Uttarakhand, India


md.farman.ee.1729@gmail.com 2 dheerfah@iitr.ernet.in 3 akumafah@iitr.ernet.in

Abstract: Smart Off-Grid Energy Systems (SOGES), which efficiently utilizes locally available renewable resources such as wind, solar, biomass, small/micro hydro, with fossil fuel powered diesel/petrol generator to provide electric power, are well suited for remote rural areas. Extension of grid to such remote areas is not economical. This paper gives various hybrid generation methods which may be used for Off-Grid electrification of remote areas. This paper also presents a generally applied software tools and their limitations. In this paper a hybrid model composed of wind, diesel and battery system is simulated based on hourly wind speed available. Analysis of per unit energy cost is performed. Forecasting of next day wind speed is done through a mixed mathematical model composed of monthly, seasonal trend and previous day wind speed using time based discrete Fourier series as well as with ARIMA. Optimal location for establishment of Hybrid system is also mathematically calculated. Load profile based on IEEE load data of that locality is also taken into account for the optimal resource utilization. Scheduling and costing of all components of proposed hybrid energy system is done using HOMER software. Features of smart off-Grid are also tabulated. Load management, power system status and spot price indicator is illustrated using A prototype model. Keywords ARIMA, Discrete Fourier series, HOMER, Hybrid System, Smart Off-Grid Energy System, Optimization, and WDB based Generation System.

resources appears to be the only solution. To energize remote rural areas, the Rural Electrification act 2003 and remote rural electrification programme of Government of India provide platform to electrify of such areas by renewable energy sources in off-grid mode. National Electricity Policy (NEP) further aims to raise the annual per capita consumption of electricity to 1000 kWh by 2012 by providing adequate reliable power, at affordable cost with access to all the citizens. Indias total installed power generation capacity is 1, 62,366 MW with 64.01% contribution from Thermal, 23.53% from Hydro, 2.7% from Nuclear and 9.84% from Renewable Energy Source(RES) [1]. Today major issue is the losses which occurs in lines, more than 32 % losses occurs in distribution lines due to inappropriate line maintenance and electricity theft. Consumers are not utilizing electrical energy smartly and efficiently due to lack of unawareness about intelligent devices and faulty energy management system which we are using today. Revenue collection by the utility is not effective because of not application of spot pricing in energy sector. Smart Off-Grid energy system comprises of three main features (1) Smart generation, (2) Smart Distribution, (3) Smart Utilization II. ENERGY SCENARIO OF INDIA Fig 1 gives the sector wise energy consumption in India which shows that maximum (49%) consumption is in industries followed by transportation (22%), household (14%) and energy required for remote area is mainly for residential and agriculture load which is around 15%. Percentage of various renewable resources worldwide is given in Fig2. Share of power generation from various resources is given in Fig3. From these Figures one may conclude that for electrification of remote area for fulfilling their energy requirement we have sufficient amount of renewable resources. Actually we are under utilizing resource. Electrification of remote area can also be done through hybrid generation where multiple energy resource is available.
Agricultur e 5% Residential 10%

I. INTRODUCTION All the countries of the world took a keen interest in the development of alternate energy resources after the OPEC made oil regulation policy and the policy was implemented that leads to oil crises in 1973. In view of exponential rise in consumption of oil, it becomes necessary to utilize energy in a smart (intelligent) manner. IREs and Hybrid energy systems are used to optimally utilize renewable energy resource. Electrical energy is the purest form of energy among all types of energy. In view of globalization, liberalization and industrialization, the demand of electric energy is increasing day by day throughout the world. In India, electricity demand is 688171 MU against availability of 620003 MU with net shortage of 9.9% during 2009-2010[1]. The peak power demand is 116281 MW while power availability during that period is only 101609 MW with net peak shortage is 12.6% [1]. Further, since more than 60-70% of rural population live in rural areas and consumes only 10-20% of the total energy. Out of total villages, about to 0.125 million villages still remain un-electrified. As the grid extension to these areas is highly capital intensive and therefore off grid electrification of these area using Renewable Energy

Others 14% Industry 49%


Transport 22%

Industry

Transport Residential Agriculture

Fig. 1: Sector wise % energy consumption in India [1, 2]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

152

Hydro 36%

Wind 4%

Biofuel 2%

Biomass Solar heat Geothermal


Solar PV

Biomass Solar 49% PV Geotherm 1% al Solar heat 2% 6%

Fig. 2: % Share of various renewable resources in world [1, 2]


RES 8%
Coal

Hydro 25% Gas 10%


Nuclear 3% Diesel 1%

Gas

Coal 53%

Diesel
Nuclear Hydro

PV array, Intelligent Charge Controller Unit (ICCU), Battery, Electronic ballast and the lighting loads such as fluorescent lamp or Compact Fluorescent Lamp (CFL). Zhiyuan et al. [7] investigated the performance characteristic for off-grid wind power system and control strategy employing fuzzy PID controller to achieve power control under different operating modes without measuring wind speed and applying mechanical sensors such as wind velocity and rotation speed sensors. The system was simulated by which are based on MATLAB/Simulink/Power system/Fuzzy Logic Toolbox.A remote wireless monitoring system based on the GPRS and the Internet was developed by Chao et al. [8]. Considering the characteristics of the off-grid wind turbine the design adopted primarily the remote communication technique to finish communication between control terminal and monitoring center. Dehbonei et al. [9] presented a novel multi-functional power processing unit capable of extracting maximum power from solar photovoltaic panels. Mizani et al. [10] demonstrated that the incorporation of optimally-rated energy storage units and renewable generators into a remote microgrid, in conjunction with an optimal dispatching strategy. Muntean et al. [11] presented architecture for a small off grid wind-power generation, control, and storage system. IV. OFF-GRID ELECTRICITY GENERATING SYSTEM Off-grid energy system is basically an electrical system which utilizes renewable energy resources in an efficient manner to electrify a cluster of villages totally or partially isolated from utility grid. Based on the energy resources, off grid energy system are of following type:- 1. Single source based a. Single renewable source based: Solar, wind, hydro, biomass etc b. Single Non-renewable source based: Diesel, gasoline, kerosene, propane etc 2. Multi source based a. Grid connected hybrid energy systems 2. Off-grid hybrid energy system b. Integrated renewable energy system In this paper we shall restrict only to WDB based hybrid energy system V. OFF-GRID WIND-DIESEL-BATTERY BASED HYBRID
ENERGY SYSTEMS

RES

Fig.3: % share of resources in power generation in India [1, 2]

III. LITERATURE REVIEW The concept of isolated/decentralized and hybrid energy system has been studied by many researchers to obtain the better renewable energy resource utilization, optimal reliability and low dependability on non-renewable energy resources. Clark et al. [3] presented scaling method for the sizing of components in distributed renewable energy systems containing multiple sources of power. The method incorporates site specific weather variability, through which it is possible to gain an understanding of numerous design trade-offs that vary depending on site specific conditions and user requirements. Gupta et al. [4] developed general methodological framework for the formulation of an action plan for the small-scale hybrid energy system for remote area. The action plan is the output of a six stage procedure: (a) Selecting cluster of villages, (b) Demand assessment, (c) Resource assessment, (d) Estimation of unit cost energy from different resources, (e) Sizing and optimization, and (f) Model formulation. A number of software tools are available for simulation, optimization and sizing of equipment in hybrid energy systems like HOMER, PVSYST, SOMES, RAPSIM, SOLSIM, INSEL, Hybrid 2, RET Screen, NREL and PV-Design Pro. A part from these, some other methodologies for optimization of HES systems are: (1)AH method [2], (2) Knowledge-based approach [2], (3) Simulation approach [2], (4) Trade-off method [2], (5) Probability method with LPSP [2], (6) Analytical method with LPSP [2], (7) Linear programming (LP) [2], (8) Dynamic programming (DP) [2], and (9) Non-linear optimization [2] Gupta et al. [5] reported the optimization of hybrid energy system model for remote hilly area in India. Soori et al. [6] presented electronic ballast design criteria for an Off-Grid Photovoltaic (PV) lighting system consisting of

Off-grid energy system covers the electrification of individual or small clusters of households completely isolated from utility grid. Using technologies based on solar, wind, biomass, and small/micro hydro and/or diesel/battery storage. Depending on whether the demand of the area is equal or more than renewable sources, the integrated renewable energy or hybrid energy system can be employed to energies such remote areas. While planning and developing off-grid energy systems, the critical issues/parameters like economic resource viability,

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

153

socio-cultural dynamics, geography, topography, agro climatic diversity, trends of development and possibility with the grid connection needs to be considered carefully. Proper and intelligent matching of resource with demand is the key for the sustainability of such off-grid energy system. Based on the availability of renewable energy sources and method of integration, the following types of hybrid energy system conFiguration may be possible: a) Wind-diesel-SPV-Battery storage b) Small/Mini/Micro-hydro Wind Diesel, c) Bio-mass SPV Diesel, d) Geothermal SPV wind Bio-mass-diesel, e) Small/Mini/Micro-hydro Bio-mass Diesel, f) Fuel cell SPV Diesel, g) SPV Small-hydro Diesel, and h) Biomass Small-hydro SPV- battery storage. In Gujarat coastal area, wind speed is sufficiently appropriate for installation of wind power station and population density of coastal area like Kandla is not very high. Besides this grid extension to this area is not too much economical. Hence we have selected wind-dieselbattery system for electrification of this area.
Power Electronics Rectifier, Controller and Charger

Start

Selection of Un-electrified Village

Is the Selected Site is Cluster of Villages?


YES YES

NO

Hourly Wind & other RES Forecasting and individual energy Assessment Sizing the individual Energy System Unit cost of energy of individual resources Problem Formulation of Hybrid Model Cost Optimization of WDB System Unit Energy cost of Hybrid energy system Operational Scheduling Strategy of WDB System Stop
NO

Load forecasting and Assessment of load profile (Minimum, Desirable & rate) Optimal Resource Selection That can meet Load Requirement

Is total Demand

Supply?

YES

Add diesel/ Conventional option to develop hybrid system

Fig. 5: Flow Chart for Development of Scheduling of WDB based Hybrid Energy system
BiDirectional Converter Mini/Micro Grid

Wind Wind Turbine Generator

Speed and power Governing System

Battery DC Bank Load

DG Set

AC LOAD

Fig. 4: Line diagram of wind-diesel-battery system

Fig. 6: Convergence of technology in smart Off-Grid power System

VI. SMART OFF GRID ENERGY SYSTEM Convergence of technology (IT + Communication + Automation & control) in off-grid power system leads to optimal digital technology i.e. smart Off-grid. Smart generation, smart distribution, and smart utilization for isolated mini/micro grid operation are collectively known as smart off grid energy system. Fig6 gives the idea of different technology applied in Smart grid and Fig7 deals with the components of Smart Off-Grid Energy system.

Fig. 7: Components of smart Off-grid Energy system

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

154

VII. OPTIMIZATION TECHNIQUES Optimization of system means to minimize the overall system cost. Optimization tools are used to select the appropriate combination of various resources so as to utilize maximum extent of renewable energy resources. A. These tools are also used to schedule the operation strategy of hybrid energy system. These tools provide detail cost components, renewable energy penetration level and hourly operation status of complete hybrid energy system.
TABLE 1: DETAILS OF VARIOUS AVAILABLE OPTIMIZATION TOOLS

Limitations: It is not publicly available. Availability: http://wind.nrel.gov/

Software HOMER

Applicability NREL's HOMER is an optimization tool used to determine energy system conFigurations given specified resources and load data. Simulation is compiled in hourly steps and is fast and comprehensive. Limitations: Publicly available and used widely. Availability:www.homerenergy.com/ The University of Massachusetts and NREL's Hybrid2 operates much like the HOMER software with a much higher degree of detail. Hybrid2 is a time series model used to predict technical and economic system performance. Limitations: The user interface is not as straightforward as HOMER. Publicly available and used widely. WINSYS allows for long term economic factors to be incorporated in system's analysis. Therefore, WINSYS focuses on full life cycle assessment. Limitations: WINSYS is not commercially available. Availability: www.winsysnetworks.net/ The IPSYS simulation package is able to predict system performance and economic analysis as well as active and reactive power flows and grid voltage levels. This level of detail is only possible with time steps down to a few seconds. The modelling of flexible supervisory system controllers is another benefit of the IPSYS software. Limitations: IPSYS is not commercially available. Availability: www.risoe.dk The package includes: VINDEC (Norway), SOMES (Netherlands), WDILOG (Denmark), RALMOD (UK), TKKMOD (Finland) and the modular model JODYMOD. Limitations: WDL Tools is not publicly available. Availability: www.weather-watch.com RETScreen is a spreadsheet based analysis and evaluation tool used to determine the economics of hybrid energy system projects. Limitations: RETScreen is publicly available Availability: http://www.retscreen.net RPM-Sim can be used to analyze the dynamic performance of wind/photovoltaic/diesel systems with and without storage.

Hybrid2

Winsys

IPSYS

VIII. CASE STUDY General: Kandla (2301N, 7012E) Gujarat under district Bhuj has been considered for hybrid energy generation system which will be composed of wind, diesel and battery energy system. Location map is given in Fig8. Especially kandla is selected because in Gujarat wind speed and solar isolation level is reasonable to install such hybrid energy system. District Bhuj is the least populated among all districts in Gujarat. Average population density of this area is less than 33 per km as per census 2001 and it is far away from national grid. Due to large distance between national grid and the selected area grid extension is not economical. Various secondary energy storage technique and there technical constraints (limitation) are also considered. Based on the operating conditions of the area and technical specifications of battery lithium-ion battery is selected. Hourly wind speed variation for various month and season are plotted in Graph.1 and 2 respectively. In study, Wind forecasting model for various season and months are created using least square method. Forecasting of next day hourly wind speed is done through a mixed mathematical model composed of monthly, seasonal trend and previous day wind speed. Next day wind hourly speed is forecasted using combined least square curve fitting and Fourier series method. Different weights are given for each parameter. HOMER 2.68 beta is used for cost analysis and operation scheduling of hourly operation of hybrid system. Monthly and seasonal model based on least square method is obtained and their trend lines are plotted in Graph 3 and 4 respectively. Finally system cost and operation schedule for each hour of year is generated.

WDL Tools

Fig. 8: Location map of Kandla Area for WDB based hybrid energy system

RET Screen

System installed capacity: Population Density (Present) =40 per km, assuming no of person per family=5 and maximum power required per family=300W. Electrified area is in the form of square or circle having area=25 km. Installed Capacity of Hybrid System= 40 300 25 1 60kW
5 1000

RPMSim

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

155

IX. WIND SPEED FORECASTING

TABLE.2: FOURIER COEFFICIENT AND VARIOUS WEIGHTS FOR BEST


PREDICTION OF NEXT DAY WIND SPEED

Fourier coefficient a0 = 9.81 a1 = 0.36 a2 = -0.02 a3 = -0.12 b1 = -0.27 b2 = -0.01


Graph.2: Hourly wind speed variation for various season

Expression

Weight age

a0 =(2/24)y a1=(2/24)ycosx a2=(2/24)ycos2x a3=(2/24)ycos3x b1=(2/24)ysinx b1=(2/24)ysin2x b3=(2/24)ysin3x

1= 0.1, 2 = 0.1, 3=0.8 1= 0.1, 2 = 0.2, 3=0.7 1= 0.2, 2 = 0.1, 3=0.7 1= 0.2, 2 = 0.2, 3=0.6 1= 0.2, 2 = 0.3, 3=0.5 1= 0.3, 2 = 0.2, 3=0.5 1= 0.3, 2 = 0.3, 3=0.4

Graph.1: Hourly wind speed variation for various month

b3 = 0.07

Graph.3: Trend line and equation for Jan by least square method

X. SOFTWARE DETAILS, MODEL COMPONENTS AND COST SUMMARY HOMER Version 2.68 Beta is an optimization tool used to determine energy system conFigurations given specified resources and load data. Simulation is compiled in hourly steps and is fast and comprehensive. In load data, here we have considered only domestic and agriculture load profile. In first case only wind power is considered, in second case, only DG is considered while in third case all the three power sources wind, diesel and battery is considered. Based on the system conFiguration and load data detailed electrical generation, generation scheduling and cost analysis is done.

Graph. 4: Trend line and equation for winter by least square method

A hybrid mathematical model composed of least square trend curves and discrete time Fourier series is adopted for the best forecasting of next day hourly wind speed data, which is given in equation (1) and (2). Y=1(0.006X-0.155X+5.533)+2(0.005X0.088X+4.985) +3(a0/2+a1cosx+a2cos2x+a3cos3x+b1sinx+b2sin2x+b3s in3x) (1) 1+2+3=1 (2) Where x is hour of the day and 1, 2, 3 are various weights given to trend curve of month, season and Fourier curve of previous day wind speed respectively. a0, a1, a2, a3, b1, b2, b3 are Fourier coefficients upto the 3rd frequency renge.

Fig. 9: Components and load in the WDB system

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

156

Fig. 10: Components wise complete cost summary

XI.

POSITIONING OF SOURCE FOR SMART OPERATION OF


DISTRIBUTION NETWORK

Various topology of distribution network has been proposed in Fig11 and to simplify the problem some assumptions are made which are tabulated in the table 5. Based on assumptions Cable length, Max Voltage drop and Total power loss are calculated using simple geometrical formula and some electrical formula like ohms law(V=IR) and Power Loss (P=I2 R). Their Assumptions and Results are tabulated in Table.5 and 6.

Fig. 11: Various Topology used in Design of Distribution network and Legends used
TABLE.5. ASSUMPTIONS USED IN POSITION IDENTIFICATION OF SOURCE FOR SMART OPERATION AND OPTIMAL UTILIZATION OF T&D NETWORK

S.NO 1 2 3 4

Assumptions Uniform load distribution in area Same resistance per unit length Half diagonal length of subblock=Unit length Unit load at each load point
Details of Controlling Actions PS Status UNIT Cost Emergency(E) 14 Saving(S) 8 Normal(N) 9 Peak1(P1) 11 Peak2(P2) 13

Preferred Topology

Voltage(V) V<4V [4V,6V) [6V,8V) [8V,10V) V>10V

LED glows Red Green White Yellow Pink

TABLE.7: CONSTRAINTS ON CONTROL VARIABLE AND THEIR RESPECTIVE ACTIONS ON STATUS, PRICE & LED LIGHT

From the Fig12 it can be seen that each feeder is sectionalized in two parts and connected through tie lines. Using two-way switch, energy supply can be fully controlled in each feeder section. System status can be easily viewed through colored LED indication available at the consumer premises. Depending on power system loading condition and system status spot prices are available in the display panel. Intelligent switches are connected to heavy power appliances which allow the consumer to use electricity only in normal or saving periods. These controlling signals can easily interfaced with the smart maters for overall tariff billing and

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

157

bidirectional communications between consumers and utility. In table.7 Constraints on control variable and their respective actions on Power system Status, unit price & LED light are tabulated. XII. CONCLUSION For successful operation of smart off-grid projects, it is essential to consider all the aspects of smart generation, distribution and utilization of generated power. Besides this energy management and public awareness about intelligent switches and devices is important for overall success of project. While selection and deciding unit capacity of wind generator it is desirable to match the local wind characteristics with the turbine characteristics because miss-match between the two characteristics turbine will not work at its rated optimal efficiency point. Another aspect in wind turbine selection is capacity calculation and operation scheduling is wind forecasting. For most accurate wind speed prediction, a mathematical model based on least square and discrete Fourier series is best fitted for the prediction of speed of that area. From the optimization results of various cases it can be concluded that generation through wind is the cheapest but reliability of supply is totally dependent on the wind speed condition in that area while generation through diesel is the costliest. Hence a proper wind-diesel mix is necessary for electrification of this area. Optimal capacity selection as well as operation scheduling strategy is the key point for success of off grid projects. Based on analysis of topology A-I, from conductor length, voltage drop and power loss point of view, topology I is best suited for SOGES. Presently consumers are not aware of power system status and are not using electricity in intelligent manner. For Smart utilization and energy management, the above prototype model can be effectively implemented in smart Off-Grid energy system. REFERENCES
[1] www.powermin.nic.in accessed on 5th Jan 2011. [2] Teri, Teri energy data directory& year book 2010. [3] Clark J.D, Stark B.H., Component Sizing for Multi-Source Renewable Energy Systems, IEEE 7th International conference on Industrial informatics, Bristol, March 2009, pp. 89-94. [4] Gupta A, Saini R. P., Sharma M. P., Hybrid Energy System for Remote Area An Action Plan for Cost Effective Power Generation, IEEE 3rd International Conference on Industrial Information Systems, Kharagpur, INDIA, Aug 2008, pp- 1-6. [5] Gupta A, Saini R P, and Sharma M P, Modeling of Hybrid Energy System for Off Grid Electrification of Clusters of Villages, IEEE International Conference on Power Electronics, Drives and Energy Systems, Roorkee India, Oct 2006, pp. 1-5. [6] Soori P K, Parthasarathy L., Okano M, and Mana A, Series Resonant-Based Ballast for Off-Grid Photovoltaic Supply Systems, IEEE International Conference on Sustainable Power Generation and Supply, Mekelle Ethiopia, Aug 2009, pp. 1-4. [7] Zhiyuan Q, Yongxin L, Haijiang L, Zhengqing H, Power Control for Off-grid Wind Power System Based on Fuzzy PID Controller, IEEE Power and Energy Engineering Conference (APPEEC), China, May 2010, pp. 1-4. [8] Chao C, Keqilao M, Muyu G, The Remote Wireless Monitoring System based on GPRS for Off-Grid Wind Turbine, IEEE International Conference on New Trends in Information and Service Science, China, March 2009, pp. 1150-1153.

Dehbonei H, Nayar C V, Borle L, A Multi-Functional Power Processing Unit for an Off-Grid PV DieselHybrid Power System, IEEE 35th Annul Power Electronics Specialists Conference, Australia, Vol. 3, Jan 2004, pp. 1969-1975. [10] Mizani S, Yazdani A, Design and Operation of a Remote Microgrid, IEEE 35th annual conference on Industrial electronics, London, Ontario, Jan 2009, pp. 4299-4304. [9]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

158

TABLE.3: UNIT AND SYSTEM COST, CAPACITY, GENERATED ENERGY

S. no

Component of system Wind generator, converter & battery DG, converter & battery

Capacity of component

Generated unit

System cost

Unit cost of energy

Wind= (625kW) Conv=60kW Battery= (820, 4V, 1900Ah) DG 1=20kW DG 2=15kW DG 3=10kW DG 4=5kW Conv=30kW Battery= (220, 4V, 1900Ah) Wind= (225 kW) DG 1=20kW DG 2=15kW DG 3=10kW DG 4=5kW Conv=30kW Battery= (220, 4V, 1900Ah)
TABLE.4: LOAD DESCRIPTION

712,880

7, 12,880$

0.046$

199,797

6, 02,822$

0.401$ 0.402$ 0.406$

Wind generator, converter, DG & battery

278,748

271,653$

0.111$ 0.112$ 0.101$ 0.115$ 0.401$ 0.402$ 0.406$

Electrical Load situation Average load Peak load

Value of load 21.78kW 59 kW

TABLE.6: PERFORMANCE COMPARISON OF VARIOUS TOPOLOGICAL NETWORKS

Topology A B C D E F G H I

Relative total Cable length 21 24 22 25 31 22 24 23 28

Max relative Voltage drop 64 44 59 64 28 186 64 59 4

Total relative Power loss 880 509 651 820 306 1770 859 712 79

Fig. 12: A prototype Model for Smart Off-Grid Energy system using spot pricing indicator

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

159

Voltage and Frequency Droop Control Stability Analysis for Parallel Inverters
1
1

Mandeep Kaur, 2Navdeep Kaur Brar


mandeep_longia25@yahoo.co.in 2 brarnav73@yahoo.com

IET, Bhaddal, Ropar(Pb.)India, 2BBSBEC, Ftg. Sahib, (Pb.)India


1

Abstract-In this paper, a new control technique for the parallel operation of inverters operating in an island grid or connected to an infinite bus. Frequency and voltage control, including mitigation of voltage harmonics are achieved without the need for any common control circuitry or communication between inverters. Each inverter supplies a current that is the result of the voltage difference between a reference ac voltage source and the grid voltage across virtual complex impedances. The reference ac voltage is synchronized with the grid, with a phase shift, depending upon the difference between rated and actual grid frequency. Normally the micro grid is intended to operate in islanded mode when the disconnection from the upstream MV network occurs. In this condition, the inverter interfaced micro sources acts as a voltage source, with the magnitude and frequency of the output voltage controlled through droops. However, under overload or under load circumstances, this droop characteristic may lead to large frequency deviation. Under this control strategy, almost all the micro sources in the micro grid can work around rated power, and when the load reduces dramatically, the distributed energy resources (like wind power and solar power) can also be taken full advantage. The analysis of the droop control stability and design procedure are provided. Keywords- Microgrids, Protection, Distributed Generation, Islanding operation, Reactive Power, Active Power.

I. INTRODUCTION The power system is in transition. Some technological advances have enabled that distributed and renewable energy sources provide a growing part of the power supply. Therefore, as a first objective, it is important to understand the current situation and future developments occurring in the electricity business and their implications for grid control. The Nature of the power network shall be different for connecting the consumer to the generation sources, which are not the bulk generator of power but generating at very small power level as compared to the conventional generation sources. This incurs a burden to the network operation and some technical II. THEORETICAL BACKGROUND This paper starts with a review of the droop control method. Although this method performs well for inductive lines, the performance in case of resistive lines is poor. In this paper, it is shown how the classical droop method can easily be adapted to account for the grid impedance, providing good performance under all conditions. A major benefit of this technique is especially prevalent in low-

voltage cable grids, generally having mainly resistive line impedance. In these limitations will appear when a great number of distributed generations are installed. One way of overcoming such problems, a micro grid system is formed to provide reliable electricity and heat delivering services by connecting distributed generations and loads together within a small area. A micro grid is usually such as battery, condenser and flywheel. Micro grids can cause several technical problems in its operation and control when operated as autonomous systems. Since the nineties, the electricity sector is subject to significant changes, the most important being the liberalization of the market, with consequences regarding the unbundling of generation, transmission, distribution and supply. Initially, power systems were small and isolated, mostly consisting of one generator and some loads, without connection to other systems. In the first half of the 20th century, however, these systems gradually adjoined, creating national grids, and bringing about benefits as economic operation and the sharing of peak load coverage and backup power. After World War II, the interconnection of grids went to a higher scale, with the interconnection of the grids, forming one large national grid with as main purpose the increase of the system reliability in case of major generator or line failures. While this trend is still ongoing, technological advances recently initiated a reverse trend, with installation of much smaller on-site distributed generators, such as Combined Heat and Power (CHP), wind or photovoltaics. In future, this trend may be accompanied by the development of local micro grids as autonomous systems, in normal conditions being connected with the backbone grid, but disconnecting in case of grid problems, to form a temporarily island grid, with the aim to increase local reliability grids, the proposed technique is appreciably more efficient in controlling frequency and voltage in proportion to the needed active and reactive power flows, as it takes the X over R ratio of the line impedance into account. Second, it is explained why the approach of controlling active and reactive current instead of active and reactive power exhibits some advantages, especially in case of a short circuit. Finally, a time-domain approach to obtain voltage and frequency droop control is described. A complex finite-output impedance voltage source is imitated, by controlling the current, flowing through a virtual impedance as a result of the voltage difference between a virtual ac source and the grid. In contrast to the classical droop control, the presented approach inherently

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

160

controls not only the fundamental voltage and frequency, but also the harmonic components of the voltage, and is therefore suited to supply highly nonlinear loads. Droop Control Through Active and Reactive Power This paper concentrates on voltage and frequency control of islanded inverter and weak LV network based microgrid The power flowing into a line at point A, as represented in Fig 1, is described .

depends predominantly on Q. In other words, the angle can be controlled by regulating P, whereas the inverter voltage U1 is controllable through Q. Control of the frequency dynamically controls the power angle and, thus, the real power flow. Thus, by adjusting P and Q independently, frequency and amplitude of the grid voltage are determined. These conclusions form the basis for the well-known frequency and voltage droop regulation through respectively active and reactive power.

Thus, the active and reactive power flowing into the line are

f0 are U0 rated frequency and grid voltage respectively, and P0 and Q0 are the (momentary) set points for active and reactive power of the inverter. The frequency and voltage droop control characteristics are shown graphically in Fig 2.

With Zej=R+jX , (2) and (3) are rewritten as

Fig. 2: (a) Frequency and (b) voltage droop characteristics.

As a first improvement to the control scheme, R is no longer neglected. This is key in low voltage cable grids, generally having a mainly resistive line impedance. Ultimately, X could be neglected instead of R. In this case, adjusting active power P influences the voltage amplitude, while adjusting reactive power Q influences the frequency. Relationships have changed radically, in such a way that the droop regulation described by (10) and (11) is no longer effective. In the general case, both X and R are to be considered. Then, the use of an orthogonal linear rotational transformation matrix T from active and reactive power P and Q to the modified active and reactive power P and Q is proposed

For overhead lines X >> R , which means that R may be neglected. If also the power angle is small, then sin = and cos =1. Equations (6) and (7) then become

Applying this transformation on (4) and (5) results in

For X>>R, a small power angle and voltage difference U1U2, (8) and (9) show that the power angle depends predominantly on P, whereas the voltage difference

For a small power angle and voltage difference U 1-U2, (13) and (14) show that the power angle depends only on P, whereas the voltage difference depends only on Q. In

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

161

other words, the angle can be controlled by regulating P, whereas the inverter voltage U1 is controllable through Q. As the grid frequency is influenced through the angle , the definition of and permits to independently influence the grid frequency and amplitude. This is illustrated graphically in Fig 3. The effect of P, Q, P, and Q on voltage and frequency is illustrated for different ratios of R/X. To derive P and Q, it suffices to know the ratio R/X. Knowledge of the absolute values of the line impedance is not needed.

From Fig 3, it can be seen that for mainly inductive lines PP and QQ, whereas for mainly resistive lines P-Q and QP. Hence, the frequency and voltage droop regulation becomes

Fig. 3: Influence of active and reactive power on voltage and frequency for different line impedance ratios: (a) R/X=0, (b) R/X=1, (c) R/X=

III IMPLEMENTATION In island grids, the use of renewable sources and other small-scale distributed generators equipped with voltage source inverters becomes more and more common. Similarly, in static Uninterruptible Power Supplies (UPS), parallel redundant inverters are used to increase the reliability. In future, a lot of distributed generators, probably with inverters as front-end, will feed into the utility networks. The low inertia and fast dynamics of voltage source inverters pose new challenges to grid stability. Stability Analysis And Dynamic Phasor Model However, this quasistatic approximation neglects the dynamics of the power network circuit elements. Whereas this approximation is, in general, acceptable for grids being supplied by slow-acting, rotating generators, the approach leads to questionable results in the presence of fast-acting, low-inertia inverters. In this paper, the dynamic phasor based modelling of the dynamics of network components is used to adequately assess the small-signal stability of grids in the presence of voltage source inverters. A systematic design procedure for a stable voltage and frequency droop controller is presented. Optimal values for the different parameters of the frequency and voltage droop controllers are selected,

according to design constraints, while ensuring stability and avoiding endless tuning of different parameters. Simulations are included to validate the results and to demonstrate the robustness of the droop controllers to grid parameter variations. A. Dynamic phasor model In literature, the concept of dynamic or time-varying phasors has been developed for a balanced, three-phase power system, enabling the inclusion of the network dynamics. The approach is used for the analysis of power system phenomena, such as sub synchronous resonance and the interaction of network and load dynamics. In this chapter, this technique is used to analyze the small-signal stability of droop control for inverters connected to a distribution grid, taking into account the distribution line phasor dynamics. Furthermore, it is shown how, based on this analysis, a stable droop controller can be designed, fulfilling some arbitrary design constraints.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

162

Thus:

(17)

The dynamic or time-varying phasor representation E(t) of e(t) is: (23) with Ud = U1dU2d and Uq = U1qU2q = U2q. Ud =U1d U2d =U1 U2 cos Uq =U1q U2q = U2 (24) Considering small disturbances of U2 and at B around the state of equilibrium (U1,U2,), the above equations can be linearized:
(18)

The dynamic phasor of the derivative of e(t) is

In the quasistatic approximation, the time-derivative on the right hand of (19) is assumed to be negligible and ignored. However, when considering systems with fastacting inverters connected to it, this time derivative becomes significant. To illustrate this, both the quasistatic and dynamic phasor transfer functions are calculated for the power flow in a line as a function of the power angle and the voltage U (Fig4). Then, making use of (19), the dynamic phasor representation is: When L >> R and both and U1U2 are small, the cross
coupling terms P/U and Q/ can be neglected, yielding two decoupled systems for P and Q:

Thus.
(27)By comparison, for the quasistatic approximation, the derivative term in (21) is neglected. The quasistatic small-signal transfer functions are:

(21)

The power flow into the line at point A, as represented in Fig2, is:

with X = L

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

163

Considering again X >> R and both and U1U2 are small, the quasistatic small signal transfer functions become

with Z 2 = R 2 + X 2. Fig4 shows the step response of the active power P as a result of a small step change of the phase angle , comparing the two small signal models with the exact (nonlinear) model. Clearly, the dynamic phasor model (dash-dotted line) resembles the exact model (solid) very well, in contrast to the quasistatic model (dashed). Only after about 15 ms, the quasistatic and dynamic phasor models yield similar results ( = 100 rad/s, U1 = 230V, Z = 0.2 , X/R = 10).

where f0 and U0 are rated frequency and voltage, respectively. P0 and Q0 are the offset values for active and reactive power. Generally, Q0 = 0 while P0 corresponds to the desired power output of a generator without droop control. kp [Hz/W] and kq [V/VAr] are the droop gain constants for frequency and voltage, respectively. They define how much active or reactive power the inverter supplies in response to frequency or voltage deviations from the rated value. For very small droop constants, the inverter operation resembles a constant voltage source. For very large droop constants, the inverter resembles a constant power source. Thus, traditional (distributed) generators without droop control could be viewed as having droop constants equal to infinity. A low-pass filter with angular cut-off frequency c = c-1 is included to filter out possible harmonics. The frequency and amplitude of the generator are:

From it can be seen that the addition of a low pass filter makes the frequency control system similar to the classical model of a rotating generator, consisting of an inertia M = c/kp and a damping torque coefficient D = 1/kp.

Fig. 4: Response of P to a =0.2 rad step change: exact model (solid), dynamic phasor model (dash dotted) and quasistatic model (dashed).

A three-phase distributed generator connected through a power line to an infinite AC power system is considered (Fig5). The AC system has a constant voltage and frequency, is balanced and contains no harmonics. The generator is assumed to be a threephase voltage-source inverter with PWM output filter. The frequency f and amplitude U of the generator output voltage are controlled by the droop characteristics as a function of active and reactive power respectively defined by (30) and represented in Fig2.

The characteristic equations for the quasistatic model are:

IV. EXPERIMENTAL RESULTS The root locus plots are shown for varying kp and kq, comparing the dynamic phasor and quasistatic small signal models. The root locus plot of the small signal model based on dynamic phasors shows that both voltage and frequency droop control systems become unstable for large gains, as some of the poles move to the right half plane. However, the quasistatic model fails to assess this instability for large gains. The time constant c = 0.1 s. All other parameters are the same as in Fig4.

Fig. 6: Root locus of dynamic phasor(equation32) for varying kp Fig. 5: Simplified scheme of a three-phase distributed generator connected to an infinite ac system.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

164

control algorithm, allowing for the design of frequency and voltage droop controllers of voltage source inverters.

REFERENCES
M. C. Chandorkar and D. M. Divan, Control of parallel connected inverters in standalone AC supply system, IEEE Trans. Ind. Appl., vol. 29, no. 1, pp. 136143, Jan./Feb. 1993. [2]. A. Tuladhar, H. Jin, T. Unger, and K. Mauch, Parallel operation of single phase inverter modules with no control interconnections, in Proc. IEEE-APEC97 Conf., Feb. 2327, 1997, vol. 1, pp. 94100. [3]. E. A. A. Coelho, P. C. Cortizo, and P. F. D. Garcia, Small-signal stability for parallel-connected inverters in stand-alone AC supply systems, IEEE Trans. Ind. Appl., vol. 38, no. 2, pp. 533542, Mar./Apr. 2002. [4]. M. C. Chandorkar, D. M. Divan, and R. Adapa, Control of parallel connected inverters in standalone AC supply systems, IEEE Trans. Ind. Appl., vol. 29, no. 1, pp. 136143, Jan./Feb. 1993. [5]. A. Engler, Regelung von Batteriestromrichtern in modularen und erweiterbaren Inselnetzen, Ph.D. dissertation, Dept. Elect. Eng., Univ. Gesamthochschule Kassel, Kassel, Germany, 2001. [6]. M. Hauck and H. Spth,Control of three phase inverter feeding an unbalanced load and operating in parallel with other power sources, in Proc.EPE-PEMC02Conf.,Sep.911,2002. [7] SJ. Chiang and J. M. Chang, Parallel control of the UPS inverters with frequency-dependent droop scheme, in Proc. IEEE PESC01 Conf., Jun. 1721, 2001, pp. 957961. [8] T. Skjellnes, A. Skjellnes, and L. E. Norum, Load sharing for parallel inverters without communication, in Proc. NORPIE02 Conf., Aug.1214,2002. [9] J.Van deKeybus, Development of a universal power measurement and control platform for low-voltage grid coupled applications in a deregulated electricity market, Ph.D. dissertation, K.U. Leuven, Leuven, Dec. 2003. [10] J. Van den Keybus, K. De Brabandere, B. Bolsens, and J. Driesen, Using a fully digital rapid prototype platform in grid-coupled power electronics applications, in Proc. IEEE COMPEL04 Conf., Aug.1518, 2004. [11] H. Tao, J. L. [15]Duarte, andM. A.M. Hendrix, Line-interactive UPS using a fuel cell as the primary source, IEEE Trans. Ind. Electron., vol. 55, no. 8, pp. 30123021, Aug. 2008. [12] R.-J. Wai, W.-H. Wang, and C.-Y. Lin, High-performance standalone photovoltaic generation system, IEEE Trans. Ind. Electron., vol. 55, no. 1, pp. 240250, Jan. 2008. [13] H. Deng, R. Oruganti, and D. Srinivasan, A simple control method for high-performance UPS inverters through output-impedance reduction, IEEE Trans. Ind. Electron., vol. 55, no. 2, pp. 888898, Feb. 2008. [14] R. H. Lasseter, "Microgrid: A Conceptual Solution," in Proc. 35th IEEE PESC Conference, Aachen, Germany, June 20-25, 2004. [15] S. Papathanassiou, D. Georgakis, N. Hatziargyriou, A. Engler and Ch.Hardt, "Operation of a prototype Micro-grid system based on microsources equipped with fast-acting power electronics interfaces," in Proc.35th IEEE PESC Conference, Aachen, Germany,June20-25,2004 [16] J. M. Guerrero, J. C. Vasquez, J. Matas, M. Castilla, and L. Garca de Vicua, Control strategy for flexible microgrid based on parallel lineinteractive UPS systems, IEEE Trans. Ind. Electron., vol. 56, no. 3, pp. 726736, Mar. 2009. [17] Xuan ZHANG, Jinjun LIU , A noval power Distribution strategy for parallel Inverters in Islanded Mode Microgrid IEEE Trans. Ind. Electron,2010 [1].

Fig. 7: Root locus of dynamic phasor(equation 32) for varying kq

From Fig 6 and Fig 7 we conclude that small signal model based on dynamic phasors shows that voltage and frequency droop control systems become unstable for large gains

Fig. 8: Root locus of quasistatic model (equation33) for varying kp

Fig. 9: Root locus of quasistatic model (equation33) for varying kq

Fig 8 and Fig 9 shows that the quasistatic model fails to access this instability for large gains. XII. CONCLUSION
This chapter started from the desire to be able to design droop controllers for distributed generation, which main purpose is to produce active power, but which may be used to improve the voltage and frequency behavior as well, be it very modestly. The classical design does not allow contributing to both voltage and frequency control in an arbitrary modest way and with predefined time constants. Moreover, it is severely limited because of stability, not clearly understood in literature, as the instability is not reflected by quasi steady-state simulations. The stability issues limit the choice of droop constants severely and hinder the development of inverters contributing in a moderate way to voltage and frequency control. It has been shown that the dynamic phasor model adequately assesses the stability of the

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

165

Energy Elasticity of India A Perspective Analysis


1

Reji B, 2Shimi S.L, 3Anu Singla


1 2, 3

1 rejib76@gmail.com lockshimi@yahoo.com 3 anu.singla@chitkara.edu.in 2

University of Kerala. India. CIET, Rajpura, Punjab, India.

Abstract :- Energy demand in India has been growing at a very rapid rate over a few decades. The trends in population growth, industrialization, Urbanization, growth in national income and energy consumption is expected to increase substantially in the coming years. This paper attempts to study the historical pattern of energy consumption, energy production and GDP growth since 1970-71 and also calculates the energy elasticity of India. Key words: - Energy, GDP, Elasticity

I.INTRODUCTION Energy is one of the basic needs of mankind. The economic development of a country significantly depends on the long-term availability of energy from sources that are affordable, accessible and environmentally friendly. The adverse effects on environment caused by the production and consumption of energy have resulted in severe environmental impacts across the globe. The supply of energy is expected to remain adequate in coming years. However, imbalance of energy consumption is prevalent around the world. Energy consumption is high in most developed countries. On the other hand, the developing countries need to consume more energy to ensure economic growth. According to estimates, energy consumption in developing countries is only one-tenth of that in the developed countries. The economic development of many countries is hindered due to energy poverty. II. ENERGY CONSUMPTION PATTERN OF INDIA India ranks fifth among the largest energy consumers of the world. According to the US Energy Information Administration data, India accounts around 4.05 percent of the worlds total energy consumption [1]. Its total primary energy demand was 595 Mtoe in 2007 and the projected demand for 2030 was 1287 Mtoe which shows this value is more than double in 2007 [2]. Its primary commercial energy consumption in 2008 stood at 19.954 quadrillion Btus [1] and involved coal, oil, gas, and electricity generated from nuclear, hydroelectric, and renewable sources. Indias commercial energy consumption is expected to more than double to 833mtoe in 2030 [2]. These Figures do not even include the energy that is consumed from traditional sources by 66 percent of Indian households[3]. Estimates of energy use from traditional

sources tend to be approximate, but Figures indicate that in 2002, 184 mtoe of energy came from such sources as fuel wood, dung, crop residue, biogas, and waste (while 354 Mtoe came from the sources mentioned above)[4]. This use is expected to grow to 215 Mtoe by 2030, though as a percentage of the total primary energy consumption, its share will drop from 34 percent to 21 percent. Per capita primary energy consumption is still fairly low in the country, i.e 17.0 million Btus (0.429toe) which is less than a third the world average.[1] Indias energy intensity, however, is still fairly high. This is particularly true of its oil intensity, which in 2004 was double the world average. The industrial sector in India is the major energy consumers accounting for about 52 percent of commercial energy consumption. Per capita energy consumption in India is one of the lowest in the world. But, energy intensity, which is energy consumption per unit of GDP, is comparable to other developed and developing countries. During 2010 Indias energy intensity (0.2 Koe/$05P) is double to U.K (0.1 Koe/$05P) , slightly higher than world average (0.19 Koe/$05P) and U.S (0.17 Koe/$05P) and lower than Asia (0.22 Koe/$05P) and China (0.28 Koe/$05P) [6] .Thus there is a huge scope for energy conservation in the country. The Energy Consumption Pattern of India during the past four decades are presented in Fig1.

Fig. 1: Energy Consumption Pattern India (1970-2010)

Source : Energy Statistics, 2011, CSO, Govt. of India. From the above graph it was found that the per capita energy consumption in india increases steadily. During 1970-71 the per capita was around 1204.39 KWHr and

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

166

increased to 4646.87 KWHr in 2009-2010. The energy consumption per capita increased around 2.85 fold (3442.48 KWHr) during the last four decades. The energy intensity was 0.1284 KWHr / Rs during 1970-71 and it increased to 0.1653 KWHr / Rs during 1985-86 . There after it decreases to 0.1224 KWHr / Rs during 2009-10. III. THE ENERGY MIX The energy mix of India is coal dominant. Around half of the primary commercial energy production depends on coal and lignite. The Fig2 and Fig 3 portraits source wise percentage production and consumption of energy in India since 1970-71.

Since 1970-71 the energy consumption pattern of India shows an inverse pattern to the production. More than half of the energy consumption is from hydro and nuclear electricity followed by coal, crude petroleum and natural gas. It was noticed that since independence the energy consumption pattern of India has changed. It has increased many folds thus in order to meet this energy demand the country has tried to harness energy from natural resources including wood, coal, natural gas, hydro, wind, solar etc. Industrialization and technological innovation has led to the over exploitation of natural resources. As people are interested in a high standard of living the energy consumption pattern has increased The per capita energy consumption in India is one of the lowest compared to world average but none the less its highly energy intensive and ranks poorly in terms of energy efficiency. Most of the energy generation is based on coal which is again a big contributor towards the green house gas emissions.
TABLE 1: ENERGY CONSUMPTION, PER CAPITA, INTENSITY INDIA (1970-2010) Year Energy Mid year GDP (Rs PerCapita Energy Consumptio population crore ) Energy Intensity n in billion '000 1999-00 Consumpt (KWH/Rs KWH price ion KWH/ ) Persons 663.99 840.53 1012.58 1477.50 1902.75 2436.77 3154.28 3909.37 4226.78 4508.26 4845.25 5462.31 551311 617248 688320 766135 852297 939540 1034931 1117734 1134023 1147677 1161495 1175480 517148 596428 695361 894041 1193650 1529453 2030710 2844942 3120029 3402716 4154973 4464081 1204.38 1361.74 1471.09 1928.51 2232.50 2593.58 3047.82 3497.59 3727.24 3928.16 4171.56 4646.88 0.1284 0.1409 0.1456 0.1653 0.1594 0.1593 0.1553 0.1374 0.1355 0.1325 0.1166 0.1224

1970-71 Fig. :2: Indias Energy Production, 1970-71 to 2009-10 1975-76 1980-81 1985-86 1990-91 1995-96 2000-01 2005-06 2006-07 2007-08 2008-09 2009-10

Source : Energy Statistics, 2011, CSO, Govt. of India. Indias energy production data shows that around 50% of the countrys energy production is dominated by coal. The second major source of energy is electricity from hydro and nuclear followed by crude oil and natural gas. During 2009-10, around 53.41 percent of total energy production is from coal, 27.11 per cent from hydro and nuclear electricity, 11 percent from natural gas and around 8.48 percent from crude petroleum.

Source : Energy Statistics, 2011, CSO, Govt. of India. IV. ELASTICITY OF ENERGY DEMAND Energy elasticity is a term used with reference to the energy intensity of Gross Domestic Production. It is the percentage in energy consumption to achieve one percent change in national GDP[7]. Unlike the other developed and developing countries, total primary commercial energy requirement in India has been falling with respect to the growth in GDP largely because higher energy prices have led to its efficient use. Estimated from the time series data of India over 1990-91 to 2003-04 comes to 0.82 which is significantly lower than 1.08 estimated for the period since 1980-81. Similarly the elasticity for per capita electricity generation is only 1.06 for the period from

Fig.:3: Indias Energy Consumption, 1970-71 to 2009-10

Source : Energy Statistics, 2011, CSO, Govt. of India.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

167

1990-91 to 2003-04 compared to 1.30 for the period since 1980-81. However, the energy elasticity of GDP growth in India may not fall as much in the future as rising income levels will foster life style changes that are more energy intense[8]. The Parikh Committee has come out with interesting projections of the energy requirements based on GDP growth rates of 7 per cent and 8 per cent at constant and falling energy elasticites. In 2005 elasticity was 0.80. The Committee considers lower elasticites of 0.75 to be attained by the decade beginning 2011-12 and 0.67 from the decade beginning 2021-22 [7]. By 2007, India's Ambassador was able to inform the United Nations Security Council that its GDP was growing by 8%, with only 3.7% growth in its total primary energy consumption, suggesting it had effectively delinked energy consumption from economic growth[7]. China has shown the opposite relationship, as, after 2000, it has consumed proportionately more energy to achieve its high double-digit growth rate. Although there are problems with the quality of the estimates of both GDP and energy consumption, by 2003-4 observers placed Chinese energy elasticity at approximately 1.5. For every one percent increase in GDP, energy demand grew by 1.5 percent. Much of this extra demand has been sourced internationally from fossil fuels, such as coal and petroleum[7].
TABLE 2: ENERGY, GDP GROWTH RATE AND ELASTICITY INDIA (19702010) Year Annual Energy Growth Rate (2)* 5.32 4.09 9.18 5.76 5.61 5.89 4.79 8.12 6.66 7.47 12.74 6.88 Annual GDP Growth Rate (3)* 3.07 3.32 5.71 6.70 5.63 6.55 8.02 9.67 9.06 6.72 7.44 6.54 Elasticity

due to high inflation in Indian economy experienced during the last few years. Measuring Elasticity using Log-liner model provides more effective measure for time series data. The equation for log-liner regression is Ln(Energy) = Constant + B ln(GDP) Here the slop coefficient B gives the elasticity of energy with respect to GDP.
TABLE:3: LOG-LINER MODEL PARAMETERS

Constant B

-5.617** 0.934**

R2 0.982 ** significance at 0.05 level Using this method the elasticity of energy was found to be 0.934 which is inelastic. V. ENERGY CHALLENGES FOR INDIA Energy is the building block of life. Now the two most important issues faced by India are the energy security and the environmental concerns. Major portion of the Indian population are still depend on biomass for their daily energy needs. With a fast GDP growth rate of the two developing countries (ie India and China) the energy demand has been increasing regularly and this has created great amount of international imbalance with regard to the energy supply and demand gap. India a fast developing country has very low per capita energy consumption but at the same time their energy intensity happens to be the highest due to inefficiency and lack in technology. This inefficiency further results in high amount of pollution and GHG emissions contributing to global warming and climate change. India today happens to be amongst the largest producers and consumers of energy in the world. The energy use here is highly intensive and happens to be amongst the highest, whereas the per capita consumption of energy is lowest compared to the OECD countries and even most of Asian countries. India is one of the fastest growing economies in the world with brilliant performance in the 10th Five year plan. The pace of growth unfortunately has not been able to match the supply and demand and as such India is still a net importer of energy. India today imports around 25% of its primary energy. Around 70% of Indias petroleum needs are met from the middle east countries. The commercial energy sector was totally regulated by the government during the pre-reform period. Since 90s private sector participated in the coal, oil, gas and electricity sectors in India due to the economic reform and liberalization. In order to meet certain socio-economic needs of the public India government provided subsidies to the energy prices which has led to distortion and inefficiency in the use of different sources of energy. The government has taken serious steps to deregulate the energy price from an Administered Price Mechanism

(1) 1970-71 to 1975-76 1975-76 to 1980-81 1980-81 to 1985-86 1985-86 to 1990-91 1990-91 to 1995-96 1995-96 to 2000-01 2000-01 to 2005-06 2005-06 to 2006-07 2006-07 to 2007-08 2007-08 to 2008-09 2008-09 to 2009-10 1970-71 to 2009-10

(4)= (2)/(3) 1.73 1.23 1.61 0.86 1.00 0.90 0.60 0.84 0.74 1.11 1.71 1.12

Calculated using data in Table 1. In the above table the elasticity is calculated from the annual growth rate of energy and GDP. It was found to be 1.73 during the period 1970-71 to 1975-76. The elasticity drop to 0.6 during the period 2000-01 to 2005-06 but beyond that it started increasing and became 1.71 during 2008-09 to 2009-10. The average elasticity during the period 1970-71 to 2009-10 was found to be 1.12 which is elastic in nature. This over all elastic nature of energy is

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

168

(APM) regime. The prices of all grades of coal and petroleum products have already been deregulated. In the electricity sector, most of the State Electricity Boards (SEBs) have started taking reform measures and regulatory commissions have been set up to determine tariffs based on economic rational. VI. CONCLUSION AND RECOMMENDATIONS The monetary inflation during the recent years and regulated price of diesel, LPG, etc. has increased energy consumption growth over national income growth since 2007-08 thus the relation between energy and GDP became elastic in nature. Necessary energy pricing mechanism can increase the efficiency in energy consumption. Thus the energy pricing can be deregulated. With deregulation the prices of oil may go up on the other hand it will increases a competition among oil producing companies in market. This will lead to better utilization of resources by technological up gradation & effective cost control. At the same time government can protect the need based population (low income group) by providing energy subsidies. In other words with deregulation there is increase in investment to fulfill the psychological need of population which is the first and fore most step of a developing country like India to strive toward a developed status. REFERENCES
[1]. http://www.eia.gov/countries/countrydata.cfm?fip s=IN [2]. International Energy Agency, World Energy Outlook, 2009 [3]. P. Khosla, Introduction, in Energy and Diplomacy, edited by I. P. Khosla (New Delhi: Konark Publishers, 2005). [4]. International Energy Agency, World Energy Outlook 2004. [5]. Expert Committee on Energy Policy, Draft Report of the Expert Committee on Integrated Energy Policy. [6]. http://en.wikipedia.org/wiki/List_of_countries _by_energy_intensity [7]. http://en.wikipedia.org/wiki/Energy_elast icity [8]. http://planningcommission.nic.in/sectors/index. php?sectors=energy

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

169

Electrical Energy Audit and Harmonic Analysis of an Industrial Unit


Pankaj Oberoi
Baddi University (HP), India
oberoi2508@gmail.com Abstract: In any industry, the three top operating expenses are often found to be, Energy, Labour and Materials. If one were to find out the potential cost savings in each of the above components, Energy would invariably emerge as a top ranker, and thus Energy Management function constitutes a strategic area for cost reduction. This paper discusses common aspects of electrical energy management in small and medium size industries. It contains the finding and the analysis of the results obtained from the electrical energy audit program which has been done in an industrial unit of Baddi (HP) named Isolloyds Engg. Tech. Ltd. The electrical energy audit was carried out under four major heads i.e. Lighting audit, HVAC (Heating Ventilation Air conditioning) audit, Power Load audit (Motors, and Melter etc.) and Harmonic analysis. Readings have been taken under these heads and analyzed to find the scope of Energy Conservation Opportunities (ECOs). Keywords: Energy Opportunities, Melter audit, Energy Conservation

A comprehensive method in checking energy usage and wastage is the "Energy Audit". II. AUDITING PROCEDURE Energy audit, similar to financial accounting process, is a process of examining energy account, checking the way energy is used and identify areas where wastage can be minimized. Energy audit cannot be successfully carried out without the commitment from the top management. Management must be firstly convinced of the necessity of implementing energy management and hence energy audit. Energy audit consists of several tasks which can be carried out depending on the type of the audit and the size and the function of the audited facility. Therefore an energy audit is not a linear process and is rather iterative. The audit describes in this paper has been carried out based on the following functional activities. Walk through survey Motor load survey used in various processes. Harmonics Analysis III. FIELD VISITS AND MEASUREMENT WORK AT ISOLLOYDS Isolloyds Engg. Tech. Ltd. is one of the oldest Indian co. in the field of fibrous high temp insulation with widest marketing and contracting base in India and Middle East. An exhaustive Electrical Energy Audit has been carried out and various data have been collected related to the audit work, for further analysis. A. Plant Electrical Energy Consumption: The factory's energy consumption has been identified in terms of the equipments and functional area wise. The results are obtained, after measurements during the factory visits. Data loggers, power analyzers, clamp meters. etc are used to measure the electrical energy consumption of the factory. The total load of the unit is approx 816.5 KW
TABLE 1: LOAD DIVISION OF ISOLLOYDS Type of Load Melter Motor Load Heating Load Plant Lighting Office Lighting Air Conditioning Water Supply Total Load Load (kW) 425 300 55 14 3 12 7.5 816.5 % Load 52.05 36.74 6.74 1.71 0.37 1.47 0.92 100 %

I. INTRODUCTION Electrical energy is the most expensive and the most important form of purchased energy. For this reason its use must be confined to a minimum for efficient operation. Because of its great flexibility, electricity offers advantages over the conventional fossil fuels and efforts to conserve electricity can result in significant cost savings. Industries use a large amount of electrical energy and that is why, it is important to ensure a loss free and energy efficient system in industries. In the developing countries like India where electrical energy resources are scarce and production of electricity is very costly, energy conservation studies are of great importance. Energy Audit is the translation of conservation ideas into realities, by blending technically feasible solutions with economic and other organizational considerations within a specified time frame. An energy audit is a study of a plant or facility to determine how and where energy is used and to identify methods for energy savings. Energy audits can mean different things to different individuals. The scope of an energy audit, the complexity of calculations, and the level of economic evaluation are all issues that may be handled differently by each individual auditor and should be defined prior to beginning of any audit activities. If we can reduce the energy usage or improve the energy efficiency in air conditioning and other mechanical and electrical installations in building, energy can be conserved and some of the resulting environmental problems, such as green house effect and ozone depletion, can be alleviated.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

170

TABLE 2: ILLUMINATION ASSESSMENT

Load Division
Melter Motor Load Heating Load Plant Lighting Office Lighting Air Conditioning Water Supply

Location Workshop(3700 m2) Ceramic Fiber Plant(1100 m2) Total

No. of Lamps 250 W /Lamp 37 20 57

Total Wattage 9250 5000 14250

Fig. 1: Load division chart of Isolloyds

Observations: 1. The major load in the plant is of Melter which consists of approx 50% of the total load 2. Motor load which consists of approx 38% of total load. 3. Total number of units consumed per month after taking average of 12 months are 3.5 lakhs to 4 lakhs/month D. Lighting Survey A walk through audit has been conducted during visits to assess the illumination requirement of the plant and scope of improvement of illumination quality and illumination level with an objective of cutting down the electrical energy consumption & cost of electricity. After survey it has been observed that Mercury vapour (MV) lamps are being used for plant lighting in the work area as well as for street lighting.

It has been observed that one MV lamp is used in 100 m2 in workshop area. Efficiency of a MV lamp = 45 lumens/watt (approx) For a 250 W lamp total lumen output = 11250 lumens That means 11250 lumens per 100 m2 area Lumens required =120 lumens/ m2 Total lumens required per 100 m2 = 12000 lumens (approx) The illumination requirement could be fulfilled by some other energy efficient option LED panel is one of such option which is more energy efficient and has longer life span. The benefits of using LEDs vary depending on the application, but typical technology features include Up to 90 percent energy-cost savings A long life of more than 100,000 hours Minimized maintenance hassles and costs Excellent cold-weather performance

TABLE 3:.COMPARISON OF PERFORMANCE BETWEEN LED PANEL & MERCURY VAPOUR LAMPS Efficiency LED For the same lumen output LED panel consumes 65 Watts of electricity The lifespan of LED bulbs is 100,000 hours plus. Due to a longer lifespan, replacement costs are decreased LEDs are monochromatic and do not need a lens to filter a certain color. MV Mercury Vapor lamps use 250 Watts of electricity. With the new options available we can be much more energy efficient The lifespan of Mercury Vapor bulbs is about 24,000 hours. That means frequent labor and material cost. Mercury Vapor bulbs produce multiple colors in the spectrum. A lens covering is needed to filter out the colors and emit monochrome light.

Life Span

Color

No. of lamps 250W/Lamp 57 57 57 57 57 57

TABLE 4: CALCULATION OF ELECTRICAL ENERGY COST FOR MERCURY VAPOR LAMPS No. of hrs of usage/day No. of hrs of Energy consumed in Cost of energy usage/year a year (KWH) (@Rs4/kWh) 24 8760 124830 499320 20 16 12 10 8 7300 5840 4380 3650 2920 104025 83220 62415 52012.5 41610 416100 332880 249660 208050 166440

Total No. of LED Panels needed to get the required lumen output Wattage of each Panel = 65 W

= 57

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

171

TABLE 5: CALCULATION OF ELECTRICAL ENERGY COST AND SAVINGS FOR LED PANEL No. of hrs usage/day No. of hrs usage/year Energy consumed in a yr (KWH) 32456 27046 21637 16228 13523 10819 Cost of Energy (@Rs4/kw h) LED Panel 129823 108186 86549 64911 54093 43274 Cost of Energy (@Rs4/k wh) MV Lanp 499320 416100 332880 249660 208050 166440 Savings/year (Rs)

24 20 16 12 10 8

8760 7300 5840 4380 3650 2920

369496 307914 246331 184748 153957 123165

E Pay Back Period


TABLE 6: CALCULATION OF PAY BACK PERIOD No. of Panels 57 57 57 57 57 57 Wattage 65 65 65 65 65 65 No. of hrs of usage/day 24 20 16 12 10 8 No. of hrs of usage/year 8760 7300 5840 4380 3650 2920 Savings (Rs./year) 369496 307914 246331 184748 153957 123165 Pay Back Period(Months) 15 19 23 31 37 46

F Motor Survey 3, 4 pole, 1470 rpm induction motors of Kirloskar make are being used for various production processes in the plant, during survey & measurement process it has been observed that some of the motors are under loaded. Two Blower motors used (60 hp) with Melter have been running at half of their rated load One needling machine (25 hp) has been running at 60% of rated capacity. One conveyor motor (20 hp) has been running at half of its rated capacity Observations
TABLE 7: RATED POWER/CURRENT VS ACTUAL POWER/CURRENT DRAWN Rated Power HP(kW) 60 (45 ) 60 (45 ) 25(18.5) 20(15) Rated Current (A) 87 87 35 30 Actual Power Drawn (kW) 24.65 27.88 11.65 7.88 Actual Current Drawn (A) 45 53 21 15.5

G. Problems with under loaded motor: i) Power Factor: An under loaded motor always run at a low lagging power factor, if motor is loaded around 50% of rated load, power factor can be as low as 0.5 lagging. That means industry is utilizing only 50% of the power from the supply mains and paying for 100% if billing is kVA based. ii) Extra expenditure on installation of power factor improvement equipment to maintain the power factor within permissible limits set by state electricity board, other wise penalty has to be paid.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

172

Fig. 2: Variation of Power factor w.r.t % of rated load TABLE 8: VARIATION OF POWER FACTOR W.R.T % OF RATED LOAD Output KW 3.7 7.5 15 18.5 45 75 HP 5 10 20 25 60 100 1/4 L 0.44 0.58 0.6 0.62 0.68 0.72 Power factor 1/2L 0.55 0.64 0.62 0.64 0.75 0.8 3/4L 0.62 0.72 0.7 0.72 0.77 0.85 FL 0.7 0.76 0.75 0.77 0.79 0.87

iii) Efficiency: Efficiency of a standard induction motor is about 87% to 90% at rated load, and it reduces drastically at half of the rated load depending upon the size of the motor and loading (about 70% to 75% for 60 hp motor), that means motor will draw more current for the same mechanical output IV. ENERGY EFFICIENT MOTORS More and more industries are switching to energyefficient motors, because energy-efficient motors lower production costs. Just a small increase in motor efficiency can significantly lower production costs. A motor can cost thousands of rupees a year to operate, so savings of even a few percent add up quickly. An energy-efficient motor generates savings whenever it is running and as long as it lasts, which may be 20 years or more. Energy-efficient motors cost more than standard motors, but purchase price pales in comparison to a motors operating costs fact, since the annual operating cost of a motor is often 5 to 10 times its initial cost, the typical 3 to 5 percent higher efficiency of an energy-efficient motor can more than offset its 15 to 20 percent higher initial cost over its life. In addition to costing less to operate, energy-efficient motors generally last longer, require less maintenance, and reduce system amperage draw. Energy-efficient motors also offer other advantages, such as longer insulation life, longer lubrication interval, greater inertial acceleration, and higher service factors. In addition, they may be better equipped to handle heavy starting loads, large sustained

loads, phase imbalances, voltage excursions, high ambient temperatures, and impaired ventilation. Energy-efficient motors also do not add as much to air-conditioning loads as standard motors, since they do not give off as much heat.

Fig. 3: Efficiency of EE Motor as per IS-12615 Eff1= 2 pole motor, Eff2= 4 pole motor

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

173

definition, designed to help users identify and compare electric motor efficiencies on an equal basis.
Losses comparison among standard & high efficiency motor(10 H.P,4 pole)

500 450 400 350 300 250 200 150 100 50 0


st t ni ca l St ra y lo ad sis re si re ne tic M ag ha

standard EE

or

or

Fig. 4: Load Vs efficiency Curve

A. Energy efficient motors performance The efficiency of a motor is the ratio of the mechanical power output to the electrical power input. This may be expressed as: Efficiency =output/input = (Input losses)/Input = output / (output + losses) Design changes, better materials, and manufacturing improvements reduce motor losses, making premium or energy-efficient motors more efficient than standard motors. Reduced losses mean that an energy-efficient motor produces a given amount of work with less energy input than a standard motor. In 1989 the National Electrical Manufacturers Association (NEMA) developed a standard definition for energy- efficient motors. The

st at

Losses

Fig. 5: Losses comparison among standard & high efficiency motors

B. Factors for Improving Efficiency Increase the amount of active material Use of double-layer windings Utilize high-performance lamination materials Optimize the air gap dimensions Improve the efficiency of fan assembly Increase the rate of heat transfer between active parts and frame Use high-efficiency bearings Optimize fabrication process.

C. Pay Back Period For the same rating, conventional motor (60 HP) consumes approx 9 KW more on 75% load.
Table 9: Comparison between EE & Standard motor (60 HP) on the basis of no. of kWh consumption per year Standardmo tor kWh 499320 416100 332880 249660 166440 Cost @Rs 4/kWh EE 1681920 1401600 1121280 840960 560640 Cost @Rs 4/kWh Standard 1997280 1664400 1331520 998640 665760 Savings(Rs /Year) 315360 262800 210240 157680 105120 Pay Back Period (Months) 3.6 4.3 5.3 7.1 10.7

No. of hours of usage/yr 8760(24hrs/day) 7300(20hrs/day) 5840(16hrs/day) 4380(12hrs/day) 2920(8hrs/day)

EE Motors kWh 420480 350400 280320 210240 140160

V.

HARMONIC ANALYSIS Baddi, Distt. Solan, Himachal Pradesh, has recently emerged as an active industrial zone of Northern India, housing industrial units of various types like textiles, pharmaceuticals, chemicals, manufacturing etc. With the increasing use of non-linear electrical loads and automation in these industrial units, poor power quality due to harmonic distortion has come up as a serious issue. To tackle the problem of increasing harmonic distortion in power distribution system, Himachal Pradesh State Electricity Board (HPSEB) has recently issued guidelines to various medium and large scale industrial units of the state to get voltage and current harmonic contents evaluated at their premises (at the Point of Common Coupling) and undertake remedial filtering solutions, if

required, for harmonic limits in excess of the limits stipulated by IEEE-519-1992 Standard. The goal of harmonic studies is to quantify the distortion in voltage and current waveforms in the power system of industrial units. The results are useful for evaluating corrective measures and troubleshooting harmonic caused problems A. Effects of Harmonics on Networks Overloading of Neutral Conductor Reduced efficiency of motors Malfunctioning of control equipment Poor power factor of the total system due to introduction of distortion power factor Overloading of power factor capacitors Increase in kVA demand of the plant due to increase in rms current drawn

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

174

M ec

ro t

B.Harmonic Measurements at Isolloyds Engg Tech. Ltd. The harmonic spectrum of L.T. currents in three-phase distribution system of the plant recorded with the help of Power and Harmonic Analyzer is indicated below

Fig 8. Harmonic spectrum of L.T current (I3) CT Ratio 1600/ 5 A

Fig 6. Harmonic spectrum of L.T current (I1) CT Ratio 1600/ 5A

Fig. 7: Harmonic spectrum of L.T current (I2) CT Ratio 1600/ 5 A TABLE 10: VALUE OF LT CURRENTS AS SHOWN BY HARMONIC ANALYZER Currents Peak Value (Amps) 2.543 4.149 4.060 rms value (Amps) 2.009 2.787 2.287 Actual rms current (Amps) 643 892 732 Total Harmonic Distortion (THD) 19.8% 5.3% 13.9 Total Demand Distortion (TDD) 11.3% 4.29% 9.15%

I1 I2 I3

C. Verification of IEEE Limit Compliance From the data provided by Electrical Section of the Plant, Per unit impedance of the t/f = 0.0514 Inductance of transmission line XL = 0.477/Km Max. Demand current IL = 1100 A C.T. Ratio = 1600/5 Short-circuit current Isc calculated at PCC = 14955.8 A Short-circuit ratio = 13.6
TABLE 11: CURRENT DISTORTION LIMIT OF IEEE-519-1992 STANDARD Short ckt ratio <20 20-50 50 -100 100 -1000 >1000 h<11 4.0 7.0 10.0 12.0 15.0 11 h<17 2.0 3.5 4.5 5.5 7.0 17 h<23 1.5 2.5 4.0 5.0 6.0 23 h<35 0.6 1.0 1.5 2.0 2.5 35 h 0.3 0.5 0.7 1.0 1.4 TDD 5.0 8.0 12.0 15.0 20.0

The industrial consumer under study falls in the category of short-circuit ratio lying in the range <20, for which the maximum allowable THD value is 4 %.(up to 11th harmonics, Ref. Table) It is clear that measured THD value at the Point of Common Coupling of Isolloyds Engg Tech Ltd. is large as compared to its allowable limit of 4 %. Further Total demand distortion (TDD) is within limits for Y phase only. Therefore, the consumer must install harmonic filter to improve Power Quality and save penalty on harmonic emission.

VI. CONCLUSION On the basis of electrical energy audit conducted the following recommendation are suggested to the consumer Mercury vapor Lamps should be replaced by LED Panels in phased manner Further electric energy cost could also be saved in office area where air conditioners are used by using false ceiling & double door system

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

175

Proper size of motor should be used, as per the rated load. If possible motor should be replaced with a proper size motor in phased manner. In due course of time, if any motor gets damaged or some new motor is to be purchased Energy Efficient Motors should be purchased. Harmonic components at PCC are greater than permissible limits Therefore, the consumer must install harmonic filter to improve Power Quality and save penalty on harmonic emission Harmonic component across individual loads is much higher where VFDs are used, which reflects at the PCC, so a more in depth analyses is required & a harmonic filter can be designed. REFERENCES
[1] H. K. Wong, C.K. Lee , Application Of Energy Audit In Buildings And A Case Study, Electrical and Mechanical Services Department, Hong Kong IEEE International Conference on Advances in Power System Control, Operation and Management, December 1993, Hong Kong. [2] Putri Zalila Yaacoh and Dr.Abdullah Asuhainu Mohd. Zin Electicai, Energy Management In Small And Medium Size Industries, Faculty of Electrical Engineering Univmsiti Tekmlogi, Malaysia. IEEE TENCON 1993,Beijing [3] IEEE Recommended Practice For Electric Power Systems In Commercial Buildings Recognized as an American National Standard (ANSI) IEEE Std 241-1990. [4] IEEE Recommended Practice for Energy Management in Industrial and Commercial Facilities Recognized as an American National Standard (ANSI) IEEE Std 739-1995 [5] Hindustan Electric Motors Catalouge Admin Off. P.O Box 18701, Vidyaailla complex, Andheri(east) Mumbai 400069 [6] Amit Tyagi, Hand Book of Energy Audit & Management TERI Press. [7] Albert Thumann, William.J.Yonger Hand Book of Energy Audits (7th Ed) 2007, Fairmont Press Inc. Lilburn. [8] Havells Catalouge, Luminaries Branch office Chandigarh. [9] Havells Catalouge, Energy Efficient 3 Induction Motors Branch office Chandigarh. [10] Cssio T. C. Andrade, Ricardo S. T,Three-Phase Induction [Motors Energy Efficiency Standards - A Case Study, Pontes Electrical Engineerigng Department Cear Federal University. [11] Catalouge, Surelink Technologies Shenzhen, China.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

176

The Diffusion of Renewable Energy Technologies in Energy Markets


Tarlochan Kaur
PEC University of Technology, Chandigarh, India.
tarlochankaur@pec.ac.in Abstract: The energy landscape is changing. Global warming, finite fossil fuel resources, cyber attack risks, power shortages, rising energy demands and new load types are just some of the drivers for the energy sector to embark on a journey to move to sustainable energies. Three major aspects integral to the transformation strategies towards sustainable energy systems are Renewable Energy, energy efficiency and less carbon intensive fossil fuel conversion. The focus in this paper is on renewable sources of energy. The global implementation of RE technology can contribute significantly to addressing these problems. A rapid expansion of energy systems based on renewable energy sources will require actions to stimulate the market in this direction. The key challenge for scaling up RETs has been the relatively high cost of deployment, resource intermittency, relatively small scale of operation and inadequate domestic research efforts in combination with major stakeholders. Barriers stand in the way of the accelerated development of renewable technologies, which can only be overcome by appropriate frameworks and policies. This paper explores the economic potentials of renewable energy, with particular focus on cost reductions and technological development, the energy markets where renewable energy might compete and make a difference, particularly in the case of developing countries and identifies the barriers that renewable energy innovations confront all along the innovation chain. It concludes with policy implications and recommendations. Key Words: Renewable Energy, Energy Markets, Innovations

A rapid expansion of energy systems based on renewable energy sources will require actions to stimulate the market in this direction. This expansion can be achieved by finding ways to drive down the relative cost of 'new' renewables in their early stages of development and commercialization, while still taking advantage of the economic efficiencies of the marketplace. Substantial cost reductions in the past few decades in combination with government policies have made a number of renewable energy technologies competitive with fossil fuel technologies in certain applications. However, making the renewables competitive will require further technology development and market deployment, as well as an increase in production capacities to massproduction levels [1]. Pricing based on the full costs of conventional energy sources (including phasing out subsidies and internalizing externalities) will make 'new' renewables more competitive. 2.1 Experience curves for renewable energy Because many renewable technologies are small in scale and modular, they are good candidates for continued cost cutting [1]. Such cost reduction can be illustrated using experience curves, which describe how cost declines with cumulative production, where cumulative production is used as an approximation for the accumulated experience in producing and employing a certain technology (see Fig 1.0).

I.INTRODUCTION Renewable energy flows are very large in comparison with humankinds use of energy. Therefore, in principle, all our energy needs, both now and into the future, can be met by energy from renewable sources. Technologies exist that convert renewable energy flows to modern energy carriers or directly into desired energy services. Technological development during the past few decades has resulted in modern renewable energy supply becoming competitive in a number of situations. Further technological development and industrial learning will continue to bring costs down. When environmental costs and security of supply considerations are included, renewable energy has even wider markets. With decisive efforts to speed up the development and dissemination of renewable energy technologies and systems, all human energy needs could be met by rerouting a small fraction of naturally occurring renewable energy flows within a country. II. ECONOMIC POTENTIALS OF RENEWABLE ENERGY

Fig. 1.0: Exemplary price experience curve: photovoltaic modules (Crystalline silicon)

Cost data are hard to find, and prices are often taken as a proxy for cost, introducing uncertainties, especially in non-competitive markets. In addition, the cost reductions illustrated by the experience curves only show the cost reduction of technologies. The cost reduction of generated heat or electricity could be larger, owing to additional sources of cost reduction such as reduced installations costs and improved availability . For some resources, such as hydro and wind, cost reductions of generated electricity

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

177

may level off when all 'good sites' are occupied. Technologies may also mature. Furthermore, the slope of experience curves may depend on the chosen timeframe and system boundaries. Experience curves have been developed for additional energy technologies [1]. III. MARKET DEVELOPMENT FOR RENEWABLE ENERGY Looking at the markets where renewable energy carriers might compete facilitates an understanding of the demand for renewable energy. The potential markets for renewable energy and the role played by the public sector to develop these markets depend on specific conditions in each country and region. Providing efficient energy-using technologies and renewable energy is a public good in many developing countries, with a wide range of benefits for sustainable development. Thus, governments must find an effective balance between liberation and directing markets towards wider social goals. It is within developing countries that much work is necessary to develop markets for renewable energy. This implies a change in focus, away from the historically dominating resources and technology assessments. A market perspective brings into question what underlies a market, such as social conditions, demand for products and services and consumer knowledge [2]. The use of renewable energy is either direct or indirect. Direct use is the immediate use of renewable energy flows to meet energy service like passive solar heating, day lighting and solar crop drying. There are often no energy markets involved here. However, policies related to areas could advance the direct use of renewable energy , for example, building codes or other instruments in the buildings area to promote passive solar heating and day lighting. Energy services cannot be measured on a Rupees per kilowatthour basis; thus, many comparisons of costs of local and integrated renewables with the costs of, for example, electricity generation by conventional plants are incorrect and misleading. Indirect use of renewable energy refers to the generation of an energy carriers that is then applied in end-use equipment to provide the desired energies. Such energy carriers include electricity, biogas, mechanical (shaft) power liquid biofuels. For some of these energy carriers there exist established markets. In other cases the use is local, such as small hydro or wind energy providing power, or stand-alone electricity use that serves niche markets, such as Solar Photovoltaics for illumination and communication uses. In industrialized countries and many developing countries, most renewable energy use takes place through markets for heat, electricity and fuels. Such markets increasingly exist in all developing countries, with some having nationwide systems for electricity, and welldeveloped fuel markets, while others rely more heavily on local markets and direct uses of renewable energy. Development of these energy markets thus relies on the use of a battery of incentives and regulations. In

developing countries it is useful to consider direct enduses and look at the opportunities for renewable energy to expand. Many of these applications encourage increased decision and participation from a variety of stakeholders, including the end users [3]. 3.1 Lighting, television, radio and telephony[4] Access to electricity opens up opportunities that are taken for granted by those who enjoy continuous access. Yet 350-400 million households in developing countries lack access to electricity. A number of options to use renewable energy for electrification exist and the markets are growing. Solar home systems usually consist of a PV solar panel, battery, charging controller and end-uses such as lighting or heating. Lanterns powered by solar energy provide lighting only. Installations may service single households or public buildings such as schools and health centres. A biogas digester can convert wastes (animal and plant) into fuels for lighting, heating, cooking and electricity generation. Digesters can be small and serve a household, or large and provide fuels for many households. Unfortunately, market development is hampered by community and political issues, as well as e technical challenges. Small-scale grids can provide electricity for communities with a high density. Traditionally, mini-grids have been powered by diesel generators or small hydro. However, solar PV, wind turbines or biomass digester, often in hybrid combinations, can replace or supplement power. 3.2 Industry, water pumping and drinking water The emerging uses of renewable energy are for agriculture, small industry, water pumping and cottage applications (sawmills and mechanical power). Furthermore, social services, such as education and health care can be supported by renewable energy. Water pumps driven by wind have historically played a role in rural areas. More recently, interest is growing in solar PVpowered water pumps along with biogas for water pumping in engines that run on diesel and biogas. Stand alone energy systems can power small industries, thereby creating local jobs opportunities. In fact, the development of mini-grids and industry goes hand in hand. As small businesses grow, the economic viability of mini-grids increases. With the availability of energy, new possibilities open up. Renewable energy can also power mechanical pumping and filtering (as well as ultraviolet disinfection) to provide clean drinking water. This is emerging as a potential major market in developing countries. 3.3 Cooking and heating water Direct combustion of biomass supports residential and commercial cooking as well as hot water in rural areas of developing countries. However, the decline in forest resources in some countries has encouraged governments to look at more efficient technologies for biomass use, as

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

178

well as solar cookers. Research and development for these technologies is still urgently needed. Markets are primarily found where resource constraints are appearing. Solar hot water heaters for residential and commercial uses are cost-effective in many regions. A large market exists for domestic solar hot water collectors worldwide [5]. 3.4 Transport Biomass-derived liquid fuels can power motor vehicles in several ways. Ethanol can power specially designed vehicles that run on pure ethanol. Ethanol is mixed, in for example Brazil and the US, with gasoline or diesel to produce gasohol for use in ordinary vehicles. Furthermore, the commercial viability of converting sugar cane to ethanol for motor vehicles has also been demonstrated. The competitiveness of ethanol and gasohol relative to conventional gasoline has continued to improve, although the global energy and automotive industries greatly affect the prospects of biomass-derived fuels in the absence of accounting for external costs and benefits. IV. RENEWABLE ENERGY INNOVATIONS Technological innovation is critical to the reshaping of energy systems in ways that encourage sustainable development [6]. However, the development and diffusion of sustainable and affordable renewable energy technologies is not occurring fast enough or widely enough [7] The challenge of stimulating novel technologies is primarily one for industrialized countries, which have the technical and economic resources for sustained research and development, and for the dissemination of renewable energy technologies. Without effective policy it is unlikely that new technologies can overcome barriers and penetrate the market to any significant extent. 4.1 Key barriers Innovations face barriers all along the innovation chain (from research and development over demonstration projects and cost buy-down to widespread diffusion). Some of these barriers reflect market imperfections; some inadequacy in the public sector domain and some differences of views about needs, corporate priorities, relevant time horizons and reasonable costs. The amount of public support needed to overcome such barriers will vary from one technology to the next depending on its maturity and market potential [8]. Direct government support is more likely to be needed for entirely new technologies than for incremental advances, where the private sector functions relatively effectively. Major criteria for deciding whether government should finance a particular field of energy research can be the contribution of that area to achieving a transition to a sustainable energy future and to strengthening the competitiveness of national industries. It is also important that the research infrastructure in the field of interest is good enough to achieve these goals. Interventions should aim at helping the most promising energy innovations

surmount bottlenecks where ever they occur in the innovation chain. Over the past couple of decades, industrialized countries have experimented with a growing number of policy instruments from target setting and procurement policies to green labelling and fiscal incentives.[9] V. CONCLUSIONS: POLICY IMPLICATIONS AND
RECOMMENDATIONS

The paper outlines the strategy and approach which may be followed for the growth and diffusion of renewable energy sector. The degree to which there will be demand for renewable energy depends on many factors. At present, only 2 per cent of the world's primary energy is new renewables. [10, 11]. One fundamental issue is that the environmental and social benefits of using renewable energy appear at the societal level, while costs have to be borne by households and investors, typically without seeing the benefits reflected in market conditions. Therefore, the demand for renewable energy is strongly linked to the market situation, and can dramatically affected by market conditions. Policies for renewable energy relate to many sectors, including land use, agriculture, buildings, transportation and urban planning. While policy and budgetary support for renewable energy have progressively increased over the years, particularly for large scale grid connected power, there continue to exist many barriers that hinder up- scaling of renewable energy deployment. And perhaps more importantly, some critical gaps remain, particularly for decentralized distribution in the areas of access to capital, technology development & adaptation, innovation induction, and strategies to up-scale deployment. Some specific areas that have to be addressed include the following: Understanding local renewable energy flows and their potential use. There is a pressing need to disseminate methodologies to estimate the local renewable energy flows and to create integrated (holistic) and sustainable solutions. Furthermore, continued research and development is required on local renewable energy technologies, such as heat pumps, building-integrated PV, passive solar and demand-side systems (integration of efficiency improvements and renewable energy). Supporting all steps in the innovation chain for renewable energy technologies and systems, including allocating a larger share of public sector funding for energy research and development to renewable energy; supporting demonstration projects (especially for modern biomass in developing countries), perhaps as publicprivate partnerships; and buying down the relative cost of 'new' renewables in their early stages of development and commercialization, while still taking advantage of the economic efficiency of the marketplace. Setting ambitious but realistic targets and timetables in combination with effective policies, such as the use of green certificates that can be traded at a national or

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

179

international market combined with agreements to reduce emissions; favourable uptake prices for renewable electricity delivered to the grid ( tax credits for investments in renewables; and concessions for the development of renewable energy resources). Methods and procedures for calculating the value of distributed generation need to be improved and disseminated, especially in situations with liberalized markets without vertical integration, where benefits may not be captured by investors in generation but by distributors. REFERENCES
Neji,L, (1997) Use of Experience Curves to Analyse the Prospects for Diffusion and Adoption of Renewable Energy Technology, Energy Policy, vol. 23, pp1099-1107 [2] Junginger, M., Faaij, A. and Turkenburg, W. (2005) 'Global Experience Curves for Wind Farms', Energy Policy, vol 33, ppl33150 [3] United Nations Development Progran: United Nations Development Programme (UNDP) (2005) Achieving the Mitten [4] Renewable Energy A Global Review of Technologies, Policies and markets, Earthscan, 2007 [5] Martinot, E., Chaurey, A., Lew, D., Moreira, J. B. and Wamukonya, N. (2002) 'Renewable Energy Markets in Developing Countries', Annual Review of Energy and the Environment, vol 27, pp309-348 [6] Johansson, T. B. and Goldemberg, J. (2002) Energy for Sustainable Development: A Policy Agenda, New York, United Nations Development Programme [7] Turkenburg, W. C. (2002) 'The Innovation Chain: Policies to Promote Ena Innovations', in Johansson, T. B. and Goldemberg, J. (eds) Energy for Sustain Development: A Policy Agenda, New York, [8] Petroncini, S. and Yemm, R.W. (2005) Introducing Wave Energy into the RenevA Energy Marketplace, www.oceanpd.com/ president's Council of Advisors on Science and Technology (PCAST) (1999) Pou Partnerships: The Federal Role in International Cooperation on Energy Innoi - Washington, DC, PCAST [9] Renewable Energy Policy Network for the 21st Century (REN21) (2005) Draft Realties 2005 [10] World Energy Assessment (WEA) (2000) Energy and the Challenge of Sustainability, >iYork, United Nations Development Programme [11] WEA (2004) Overview Update, New York, United Nations Development Programme World Energy Council (WEC) (2004) Survey of Energy Resources 2004, Oxford Elsevier [1]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

180

Restructuring and Deregulation of Indian Power Sector: Current Status, Policies and Future Perspectives
1

Naveen Kumar Sharma, 2Prashant Kumar Tiwari, 3Yog Raj Sood


1,2,3

N. I. T. Hamirpur (H.P.)-177005, India


1 2

naveen31.sharma@gmail.com prashant081.in@gmail.com 3 yrsood@gmail.com

Abstract Restructuring and deregulation of the electricity industry is a movement with the aim of achieving lower prices to customers through cost savings. The deregulation of electric power systems in many parts of the world has changed the mechanism of electricity pricing. India is in the processing on restructuring in power sector. Power planning will play an important role in the successful power reform and essential for Indian economic growth. Only a financially and commercially sound power sector can attract new investments. In Indian power sector, restructuring will substantially change power planning, especially the methodology of the power generation expansion planning. The evolution of power restructuring and deregulation policies in India needed to be considered to produce full benefits from power utility deregulation. An important possible effect of deregulation may be a reduction in maintenance and in new investments. This paper emphasizes the current strategies and policies made by the Indian government towards deregulation. Keywords Restructuring, Deregulation of power sector, Models of deregulation rated.

state-owned monopolies to finance, construct, own and operate the electricity supply network. Since the mid1990s, more than 30 countries or regions within countries have introduced policies to reform their electricity supply industries [2]. India is in the processing on restructuring in power sector. Power planning will play an important role in the successful power reform. However, restructuring will substantially change power planning, especially the methodology of the power generation expansion planning. Deregulation encouraged the growth of new independent power producers whose business requirements transformed the power plant industry. This document is a template. An electronic copy can be downloaded from the conference website. For questions on paper guidelines, please contact the conference publications committee as indicated on the conference website. Information about final paper submission is available from the conference website. II. STRUCTURE OF INDIAN POWER SECTOR The Indian power sector is presently going through a process of reform and restructuring as is the trend in many other parts of the world. Under reform, independent regulatory commissions are set up to the center as well as state wise and vertically integrated utilities are being unbundled into corporate entities. At present India is sixth largest country in the world in electricity generation, having aggregate capacity of 172 GWs out of which 63% is from thermal, 24% from hydro, 3% from nuclear and the rest about 10% is from renewable energy sources [13]. Although Over the years, Indian power sector has experienced a five-time increased in its installed capacitya jump from 30,000MW in 1981 to over 171,926.40 MW [13] by 28 February 2011 but still there is a huge gap in generation and demand in India hence need to be establish more generation plants preferably to be come from renewable sources by governmental as well as various private participation. The functions of centre electricity authority (CEA) are to advise the Ministry of Power (MoP) on national power policy, national power planning and regulatory matters on the national level where as state electricity regulatory commissions (SERCs), does the same function at state level. Indian

I.

INTRODUCTION

The power sector in India has been regulated and owned for many years by various government agencies and organizations. The role and the participation of private industry in the Indian power sector has been limited and confined to specific areas of small jurisdiction and consumer base. Power sector is one of the key sectors contributing significantly to the growth of countrys economy. Power sector needs a more useful role to be played in defining, formulating and implementing the research projects with close involvement of all utilities such that the benefit reaches the ultimate consumer [1]. During the past two decades power sector over worldwide has experienced a rapid change in generation, transmission, and distribution system. India is in the processing on restructuring in power sector. Power planning will play an important role in the successful power reform. However, restructuring will substantially change power planning, especially the methodology of the power generation expansion planning. The deregulation and restructuring of the electricity supply industry is one of the most important global energy developments of the last century. Up to the 1980s most countries relied upon

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

181

power sector is organized into five Regional Electricity Boards such as northern regional electricity board(NREB), southern regional electricity board (SREB), western regional electricity board (WREB), eastern regional electricity board (EREB) and north eastern regional electricity board (NEREB) as depicted in Fig1. As shown in Fig1, each regional electricity board covers many states electricity boards and in Table 1 region wise total installed capacity in India as on 28 February 2011. With the establishment of various inter-regional links, inter-regional power exchange has grown manifold. Growth of interregional power exchange has helped in meeting more demand in energy deficit regions besides achieving overall economy [11]. At present, the country was divided into two grids one is the Southern Grid and the other comprises of Eastern, Western Northern, and North-Eastern region grid. While India currently does not have a unified national power grid, but it is planning to hoped that India will become One Grid system by 2012. The one grid system will resolve the power woes of power deficit states by improving intra-regional transmission of power with inclusion renewable also. For renewable promotion as per the Electricity Act-2003, CEA and SERCs are required to encourage investment in renewable energy by providing suitable measures for connectivity with the grid and
S.No 1. 2. 3. 4. 5. 6. 7 REGION NREB WREB SREB EREB NEREB Islands All India COAL 23745.00 30995.50 19382.50 18235.38 60.00 0.00 92418.38 THERMAL GAS DSL 4134.76 12.99 7903.81 17.48 4690.78 939.32 190.00 17.20 787.00 142.74 0.00 70.02 17706.35 1199.75

specify a percentage of the total consumption of electricity in the area [12]. III. RESTRUCTURING OF THE INDIAN POWER SECTOR Electricity reform has been motivated by: technological developments, particularly the improved efficiency of gas turbines; the need for increased investment, especially in developing countries such like India; high electricity prices and a shift away from the view that electricity supply is a natural monopoly. Keeping in view the pros and cons of different restructuring processes in various countries, it is recognized easily that India is not yet ready for electricity restructuring. The first and major restructuring problem is the gap between demand and generation. The restructuring process as a whole is a very complex process, and steps suggested here are overlapping and interrelated. Thus, this study suggests the following steps towards the restructuring of the Indian power sector [16]:

TABLE I: REGION WISE GENERATING INSTALLED CAPACITY (MW) IN INDIA (DATA AS ON 28.02.2011)

Nuclear TOTAL 27892.75 38916.79 25012.60 18442.58 989.74 70.02 111324.48 1620.00 1840.00 1320.00 0.00 0.00 0.00 4780.00

HYDRO 13622.75 7447.50 11299.03 3882.12 1116.00 0.00 37367.40

Renewable (MNRE) 3165.55 5357.96 9341.67 359.64 223.60 6.10 18454.52

TOTAL 46301.05 53562.25 46973.30 22684.34 2329.34 76.12 171926.40

Fig. 1: Regional Electricity board of India and percentage of region wise installed capacity

Bridge the gap between power demand and electricity generation Decentralize the planning process for an easy entry of generators Increase the intrastate transmission lines Increase the tariffs incrementally

Reduce the direct government control Establish an independent regulating authority Unbundle SEBs as generation, transmission, and distribution entities Privatize and commercialize the power entities Establish a competitive power market.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

182

The power sector in India is beset with severe problems of non-payment by customers at all levels, increasing fiscal losses at REBs, large-scale thefts of power, overstaffing, under-investment in transmission and distribution, increasing power outages due to inadequate transmission, etc. A. Central Electricity Regulatory Commission Central Electricity Regulatory Commission (CERC) was set up soon after the enactment of the Electricity Regulatory Commission Act, 1998. Such commissions had already been set up in Orissa and Haryana in 1996 and 1998 respectively under state legislations. These legislations were saved by the central government at the time of the enactment of the central act. Subsequent to the enactment of the central act a large number of states have gone ahead with the setting up of regulatory commissions. With the concurrence of the central government, Andhra Pradesh has passed a separate Regulatory and Restructuring Act, in line with the Orissa and Haryana acts. There is need for other states also to pass such restructuring acts. Due to the federal nature of our Constitution, the central government had decided that whereas it would pass an Electricity Regulatory Commissions Commissions Act, it would not impose a restructuring model on any state by central legislation but would only issue guidelines and model acts for the consideration of the state governments [17]. The functions of CERC include Regulating the tariff of generating companies owned or controlled by the central government Regulating the interstate transmission of energy including the tariff of transmission utilities Promoting competition, efficiency, and economy among the power sector activities Advising the central government on the formulation of tariff policy. The central government and a few state governments have appointed regulators, and some states have unbundled the sector into transmission, generation, and distribution through state enactments. B. Electricity Act 2003 The Electricity Bill 2003, approved in Indian Parliament in May 2003, aims to enhance the scope of power sector reforms. This act consolidates all the existing laws and introduces provisions with respect to new developments in B. the sector. It focuses on creating competition, protecting consumer interests, rationalizing tariff, etc. All the necessary powers including issue of licenses are given to the regulators which are made independent entities from the government. Some of the major provisions of the Electricity Act are: Generation has been deli censed and captive generation is being freely encouraged and permitted. For hydro

projects, an approval of the State Government and clearance from the Central Electricity Authority (CEA) are needed to check the safety aspects and optimum utilization of water resources. There will be Government owned Transmission Utility at the Central as well as State level, having the responsibility of ensuring that the transmission network is developed in a planned and co-ordinate manner to meet the requirements of the sector. The load dispatch function can be integrated with or separated from the Transmission Utility and in either case it will remain under Government control. Provision for private transmission licensees has been made in this act. Open access in transmission with provision of surcharge for cross subsidy and this surcharge will be gradually phased out. Distribution licensees are free to undertake generation and generating companies are free to take up distribution licensees. For rural and remote areas stand alone systems for generation and distribution would be permitted. This provision seems to be aimed at encouraging Captive Power Plants (CPPs) and Distributed Generation (DG). For rural areas decentralized management of distribution through Panchayats, Cooperatives, etc. would be permitted. Regulatory Commissions are authorized to issue a license for power trading and they will fix up the upper limit on power trading margins. If there is directly commercial agreement between a consumer and a generating company or trader, the price of power would not be regulated and only the transmission and wheeling charges with surcharge would be regulated. State Governments can convert State Electricity Boards (SEBs) into companies or continue them as distribution licensees. An Appellate Tribunal has been created for disposal of appeals against the decision of the CERC and State Electricity Regulatory Commissions (SERC) so that there is speedy disposal of such matters. The SERC is a mandatory requirement. In the following sections the impact of this act on generation, transmission and distribution is analyzed [14, 19]. C. Concerns on Electricity Act 2003 While the Act is said to be the major thrust required to liberalize the electricity industry of the country it still has shortcomings that need to be addressed to create a transparent market free of malpractices. Until the distribution system is considerably augmented from its present state to bring down the technical losses; a consumer will not be willing to switch from its distribution licensee to other suppliers as he/she has to pay for the distribution losses with the per unit of energy

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

183

consumed. The Act leaves the question of ownership, particularly of the transmission company. If the company is state owned, then the losses of the transmission company will have to be added to other subsidies provided by the state, thus increasing the net outflow of funds from the state. As per the present structure of the Act, IPPs will not be greatly successful as the price of electricity produced by them will be very high in comparison to the one that customers are used to paying in the regulated scenario [18] One of the main impacts of the 1998 crucial deciding factor for the economic growth of the Act was the passing of Availability Based Tariff (ABT), to introduce grid discipline among the participants of the National Grid. The major change in the industry came with the started mooting on introducing these effects to the power passing of the Electricity Act 2003 that replaces the three existing legislations, namely, Indian Electricity Act 1910, the most significant and definite step taken in the Indian Electricity (Supply) Act 1948, and the Electricity Regulatory Commission Act 1998 [18]. D. Availability Based Tariff (ABT) ABT and the Electricity Act 2003 are the two major path breaking legislations that had paved the way for the restructuring in Indian electricity industry. ABT concerns itself with the tariff structure for bulk power and aims to bring about more responsibility and accountability in power generation and consumption through a scheme of incentives and disincentives. As per the notification, ABT is applicable to only central grading stations having more than one SEB/State/Union Territory as its beneficiary. Through this scheme, the Central Electricity Regulatory Commission (CERC) looks forward to improving the quality of power and curtailing the following disruptive trends in the electricity industry: Unacceptable rapid and high frequency deviations (from50 Hz) causing damage and disruptions to largescale industrial consumers. Frequent grid disturbances resulting in generators tripping, power outages, and power grid disintegration. E. Tariff Components in ABT The most significant aspect of ABT is to split the existing monolithic energy charge structure into three components, i.e., capacity charges (fixed), energy charges (variable), and UI (Unscheduled Interchange) charges. It is this last component that is expected to bring about the desired grid discipline. The splitting of the tariff into fixed and variable cost components is meant to act as an incentive for power trading which shall conclude in a self regulating electricity market regime. It is also expected to promote the concept of ELD (Economic Load Dispatch) among power generators [18]. F. Transmission and Distribution Reforms

The Electricity Laws Amendment Acts, 1998 has been passed on 10 August 1998. Under this Act, transmission has been made a separate activity which would help in inviting greater participation in investment both from public and private sector. The Central government has been emphasizing the need for unbundling and corporatizing the vertically integrated SEBs. Although distribution is a State subject, the Central government has been extending all possible help to the State governments in improving the distribution system. A Chief Ministers/Power Ministers Conference was held on 18 December 1998 to discuss and deliberate upon the critical issues pertaining to the power sector and an action Plan has been adopted. IV. POWER SYSTEM MOVING TOWARDS DEREGULATION Since the mid-1980s the electrical power supply industry around the world has experienced a period of rapid and critical changes, regarding the way electricity is generated, transmitted and distributed. The need for more efficiency in power production and delivery has led to privatization, restructuring and finally deregulation of the power sectors in several countries traditionally under control of federal and state governments. Deregulation is a relatively recent concept, whos economic, regulatory and implementation structure continues to be adopted to the specific needs of each nation.
TABLE II: CHARACTERISTICS OF MODERN AND TRADITIONAL
REGULATING APPROACH

S. No. 1. 2. 3. 4. 5. 6. 7.

Traditional approach Dominant individual role Inflexibility Traditional steady procedures Control Body role One dimensional Consequent acting Principles based on old culture

Modern approach Joint responsibility Flexibility Interactive processes Influence realization Multi dimensional Proactive approach Principles based on new culture

The need for more efficiency in power production and delivery has led to a restructuring of the power sectors in several countries traditionally under control of federal and state governments. Many countries like England, United States of America, Canada, Australia, New Zealand, Chile, Argentina, Peru, Colombia and Scandinavian are already exercising with deregulated electricity industry. Even some of the developing countries, there have been a strong drive toward deregulation and a more intense participation of privately own third party generation through a wheeling process. Though there are some pitfalls here and there, the end users of the electricity are enjoying the fruits of the deregulated electricity industry tree [3]. The main economical benefits expected from deregulation include improved quality of electricity service by allowing rates, that more closely track the true cost of service and by differentiating the product quality, for example, offering

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

184

new products with different degrees of power reliability. Table 2 shown as main Characteristics of modern and traditional regulating approach in deregulated electricity market. A. Traditional Utility Model Traditionally, power utilities have been state-owned monopolies or privately-owned monopolies, either regulated by government agencies or self-regulated without much oversight. Their traditional mission has been an engineering one: expanding supply, improving technical efficiency, and ensuring or improving reliability and access. In traditional power systems, the utilities are mostly owned by governments. Vertically integrated utilities typically own generation, transmission and distribution over a wide area, having complete control over generation, transmission and distribution over a wide area as shown in Fig2. Each utility has one or more control centres that maintain security and reliability of the specific region.

Fig. 3: Modern Deregulated power system model

Fig. 2: Traditional integrated power systems

B. Modern Deregulated Power System Model Whereas one monopoly utility traditionally performed generation, transmission and distribution functions in a vertically integrated manner, each of these functions is being parceled out to different commercial entities, some retaining a regulated monopoly status (particularly distribution utilities) and others starting to face competition (particularly generators). Unbundling can provide greater consumer incentives to self-generate using a variety of technologies, including renewable energy. If retail tariffs accurately reflect generation, transmission and distribution costs, customers may face the full costs of centralized generation and delivery, and as such may have more incentive to self-generate and thus to avoid transmission and distribution charges. On the other hand, unbundling can create transmission pricing penalties for intermittent renewable energy sources. Unbundling requires new methods and structures for transmission pricing [4]. In deregulated environment, generation companies and independent producers may be considered to wheel their electrical power for distribution companies and bulk consumers, through a third party transmission system as shown in Fig3. Electrical energy would become a product, which could be bought or sold and transported from one place to another.

C. Components of Deregulated Power System The structure components representing various segments of the electricity market are given below. Depending on the structure and regulatory framework, some of these components may be consolidated together, or may be further unbundled. In some Asian countries various regional, state, provincial or independent generators coexist. In these cases the financial and technical interrelationship are murky and are in a process of rapid evolution [5, 6]. Generation companies (Gencos): Gencos are responsible for operating and maintaining generating plants in the generation sector and in most of the cases are the owners of the plants. In some case individual generators do not market their output, but only genco market the output of all its generators. Build operate and transfer (BOT): plant or independent power producers (IPPs) BOT or IPPs can act as its own generator-serving entity and independently market its output to a trading entity or to a loadserving entity. Transmission companies (Transcos) and transmission owners (TOs): Transcos moves power in bulk quantities from where it is produced to where it is delivered. In most deregulated industry structures, the Transmission companies owns and maintain transmission lines under monopoly franchise and are called Transmission Owners (TOs), but they do not operate them. The independent system operators do that. Distribution companies (Discos) and retailers: Discos assume the same responsibility on the distribution side as in a traditional regulatory supply utility. However, a trend in deregulation is that Discos may now be restricted to maintain the distribution network and provide facilities for electricity delivery while retailers are separated from Discos and sell electric energy to end consumers. Independent system operator (ISO): The ISO is the supreme entity in the control of transmission system. The basic requirement of an ISO is disassociation from all market participants and absence from any financial interest in the generation and distribution business. Power exchanger (PX): The PX handles the electric power pool, which provides a forum to match electrical energy supply and demand based on bid prices. The time horizon of the pool market may range from half an hour to

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

185

a week or longer. The most usual is the day-ahead market to facilitate energy trading one day before each operating day. Scheduling coordinators (SCs): SCs aggregate participants in the energy trade and are free to use protocols that may differ from pool rules. In other words market participants may enter an SCs market under SCs rules through bilateral and multilateral transactions. D. Existing Models of Deregulation The worldwide current developments towards deregulation of power sector can be broadly classified in following three types of models [5, 6]. 1. Pool model In this model the entire electricity industry is separated into generation (gencos), transmission (transcos) and distribution (discos) companies. The independent system operator (ISO) and Power exchanger (PX) operates the electricity pool to perform price-based dispatch of power plants and provide a form for setting the system prices and handling electricity trades. In some cases transmission owners (TOs) are separated from the ISO to own and provide the transmission network. The England & Wales model is typical of this category. The deregulation model of Chile, Argentina and East Australia also fall in this category with some modifications. 2. Pool and bilateral trades model In this model participant may not only bid into the pool through power exchanger (PX), but also make bilateral contracts with others through scheduling coordinators (SCs). Therefore, this model provides more flexible options for transmission access. The California model is of this category. The Nordic model and the New Zeeland model almost fall into this category with some modifications. 3. Multilateral trades model This model envisages that multiple separate energy markets, dominated by multilateral and bilateral transactions, which coexist in the system and the concept of pool and PX disappear into this multi-market structure. Other models such as the New York Power Pool (NYPP) model fall somewhere in between these three models. V. INFORMATION FLOW IN A DEREGULATED POWER SECTOR Technical and market information has historically been confined in each system with minimal exchange with the other systems. However, deregulation resulted in a wide range of information flow in terms of type and quantity among the systems [9]. This information is mainly used to support the physical operation and the economic efficiency of the system as well as power quality. To maintain system reliability, system coordinators monitor inflows and outflows from the grid, as well as other

conditions, such as voltage profile, throughout the system. Operating conditions must be measured, transmitted, processed and monitored and the necessary actions must all be dispatched in real time to maintain system security and stability. In the future, real-time state information on system operations may be used to support market functions such as estimating the cost of bulk power transactions on the transmission system. Some of the technical information flows for system operations are as follows: Generation monitoring and control (standards and communications protocols are needed), Transmission system real-time monitoring and control (measurements/standards for capacity utilization and dynamical system analysis are needed), Real-time communication links to end users to support demand-side management activities (standards and communications protocols are needed). Market information primarily supports the pricing and billing activities for forecasting and dispatching generation and carrying out ancillary services [10]. It is key that information is recorded at regular intervals (for instance, hourly) and exchanged periodically (for instance, daily) with the other control centres. In addition, tariffs for energy and ancillary services must be provided to suppliers hourly (at least) in order to get their response. As the number of measurements is large, compatibility and operating costs are also important issues. In addition, these data are mostly commercial and proprietary (and valuable); thus, data security is an important issue. In the Fig4 illustrates typical system and market operations.

Fig. 4: Information and power flow in a deregulated power system

VI. PROPOSED MODEL FOR RESTRUCTURING IN INDIA In many parts of the world wherever unbundling, i.e. separation of generation, transmission and distribution has taken place, the two models are more prevalent for system operation. The first one is Independent System Operator (ISO) model and the other is Transmission System Operator (TSO) model. In ISO model, transmission companies are also permitted to own, manage and control

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

186

generation and distribution companies, an independent system operator is created to facilitate open access and competitive markets [14, 15]. In TSO model, operation of the grid and ownership of the grid are integrated in a single entity, which is responsible for development of transmission system and to provide non-discriminatory open access to all eligible market participants. Neutrality is an important aspect of the TSO to ensure an efficient market. In view of this, TSO model seems to be most suitable for future restructured electricity market in India. This is because the government owned Transmission Company is merely responsible to provide non-discriminatory open access. Some of the developed countries are also moving away from ISO model by formation of Regional Transmission Organizations (RTO), which will finally converge as a TSO model. Even though the conditions in Indian power market are not yet ripe for introducing retail competition, the necessities in a deregulated power market can be summarized below: Non-discriminatory open access to transmission network is a pre-requisite for ensuring competition in wholesale power trading. The system operation functions at the national level can be handled by central transmission utility while state transmission utilities can manage State Load Despatch Centres (SLDCs) similar to TSO concept. The regional electricity boards will have the responsibility of managing the power exchanges while the Regional Load Dispatch Centres (RLDCs) will manage the overall integrated operation of power system like outage planning, relay co-ordination, islanding schemes, etc. Some of the main characteristics of TSO model are also incorporated in the Electricity Act 2003. In a restructured power system the electricity price is no longer fixed but varies with time, locations and introduction of new distributed power production using renewable energy sources may be a big issue in this environment. VII. CONCLUSION

although at a slow pace. Several state electricity broads are being unbundled into three distinct corporations namely Generation, Transmission and distribution. The distribution system are being horizontally broken down into manageable Discos with separate accountability and privatized for better efficiency in metering, billing and revenue collection. REFERENCES
[1] R. Singh, Y. R. Sood, Transmission Tariff for Restructured Indian Power Sector with Special Consideration to Promotion of Renewable Energy Sources, IEEE TENCON Conf 2009; 1-7. J. E. Besant-Jones, B. Tenenbaum, 2001. Lessons from Californias power crisis. Finance and Development 38(3), 2428. Yog Raj Sood, Narayana Prasad Padhy, H. O. Gupta, Wheeling of Power Under Deregulated Environment of Power System A Bibliographical Survey, under publication in IEEE Trans. on power systems, paper no. 2001 TR 109.R1. Eric Martinot Grid-Based Renewable Energy in Developing Countries: Policies, Strategies, and Lessons from the GEF, World Renewable Energy Policy and Strategy Forum June 13-15, 2002, Berlin, Germany. N.P. Padhy, Y. R. Sood, Advancement in power system engineering education and research with power industry moving towards deregulation, Power Engineering Society General Meeting, 2004. IEEE, 6-10 June 2004, Vol. 1, pp. 71-76. Lai Loi Lai, Power System Restructuring and Deregulation, (book), Publisher: John Wiley and Sons, Ltd., New York, 2001. M.A. Fischetti Electric Utilities: Poised for Deregulation, IEEE Spectrum, May 1986, pp.34-43 A. J. Wood and B. F. Wollenberg, Power Generation, Operation and Control, John Wiley and Sons, New York, 1996. M. Doll and J. F. Verstege, Congestion Management in a Deregulated Environment Using Corrective Measures, In: Proceedings of the IEEE- 2001, pp. 393-398. H. Ying-Yi and H. Chuan-Yo, Locational Marginal Price Forecasting in Deregulated Electric Markets Using a Recurrent Neural Network", IEEE-2001., pp. 539-544, A. Srivastava, M. Shahidehpour Restructuring choices for the Indian Power sector IEEE Power Engineering Review, November 2002, PP-25- 29 The Electricity Act, 2003, Section 3.1 and 86.1. Ministry of Power. Available: http://www.powermin.nic.in [online]. S. A. Khaparde, Power Sector Reforms and Restructuring in India. S. S. Vindal, N. S. Saxena, and S. C. Srivastava, Industry Structure Under Deregulated Wholesale Power Markets in India, Proceedings of International Conference on Present and Future Trends in Transmission and Convergence, New Delhi, India, December 2002. A. Srivastava, M. Shahidehpour, Restructuring Choices for the Indian Power Sector, IEEE Power Engineering Review, November 2002, pp. 25-29. Pradip Baijal, Restructuring Power Sector in India, Economic and Political Weekly September 25, 1999, pp. 2795-2803. Pahwa, Zuyi Li, Electricity Industry Restructuring in India, 2006 IEEE, pp. 99- 105. Published in Gazette of India The Electricity Act, 2003. India: Universal Law Publication Company Pvt. Ltd.

[2] [3]

[4]

[5]

[6] [7] [8] [9]

[10]

[11]

[12] [13] [14] [15]

The power market deregulation, introduction of clean energy and increasing of disasters are gradually having major effects on the reliability of electric power systems. Deregulation process primarily focuses on enhancing system efficiency, improving service standards and developing competitive market. The methodology of the planning for power restructuring is very important. Electricity reform process in India is already in action

[16]

[17] [18] [19]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

187

Study on Improving the Efficiency of Electrostatic Precipitator


1
1

Vivek Goyal, 2Akanksha


PEC University of Technology
2

1, 2

Vivek007goyal@gmail.com,

Akankshasingla18@gmail.com

Abstract: The efficient operation of an Electrostatic Precipitator (ESP) in practice depends upon many variables, such as charging method, particle size, gas flow, temperature, dust resistivity, etc. Even after having efficiency upto 95% by most of the ESP, we still have problem of ash handling. In this paper, methods to improve the efficiency of ESP are discussed. Keywords: Electrostatic Precipitator (ESP), Crorna Power I. INTRODUCTION The concern with the environment is becoming more and more present in the daily life of everyone. An increasing involvement with the environmental issue is demanded from the governments, companies and, eventually, from the society as a whole. COAL-FIRED boilers in thermal power plants (TPPs) cause substantial air pollution through emission of fly ash particles. Similarly, process plants, such as cement mills and steel plants, also cause air pollution through their exhaust gases. Electrostatic precipitators (ESPs) are widely used in many industries for particle collection from flue gases. In a TPP, they serve to limit fly-ash pollution of the atmosphere, in addition to reducing erosion of the induced draft fan blades. An electrostatic precipitator (ESP) is a particle control device that uses electrical forces to move the particles out of the flowing gas stream and onto collector plates. The particles are given an electrical charge by forcing them to pass through a corona, a region in which gaseous ions flow. The electrical field that forces the charged particles to the walls comes from electrodes maintained at high voltage in the center of the flow lane. Once the particles are collected on the plates, they must be removed from the plates without re entraining them into the gas stream. This is usually accomplished by knocking them loose from the plates, allowing the collected layer of particles to slide down into a hopper from which they are evacuated. Some precipitators remove the particles by intermittent or continuous washing with water.

Fig. 1: ESP

II. COLLECTION EFFICIENCY The equations give a theoretical estimate of the overall collection efficiency of the unit operating under ideal conditions. Unfortunately, a number of operating parameters can adversely affect the collection efficiency of the precipitator. Deutsch-Anderson Equation This equation is used to determine the collection efficiency of the precipitator under ideal conditions. The simplest form of the equation is given below.

Where: = collection efficiency of the precipitator e = base of natural logarithm = 2.718 w = effective migration velocity, cm/s A = the effective collecting plate area of the precipitator, m2 Q = gas flow through the precipitator, m3/s III. FACTORS EFFECTING COLLECTION EFFICIENCY A. Operating Voltage The operating voltage should be kept as high as possible for the maximum corona leakage. The minimum penetration is obtained with some sparking, despite the fact that sparking knocks holes in the collected dust layer and retains some dust. Sparking depends on the operating voltage, electrode spacing, types of electrodes, gas composition, particle concentration, dust layer and bulk resistivity of the dust. Automatic controls lower the voltage when spark over occurs to prevent continued arcing. The operating voltage is rectified ac. Current research is aimed at a square wave voltage. The peak voltages for industrial high-voltage ESP are mainly 30 to 75 kV but may be as high as 100 kV. The field strengths will usually be 4 to 5 kV/cm.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

188

conducting electrons by current flow. The resistance of dielectric decreases with increasing temperature. The dielectric has the units of coulombs ^2 joule^-1 m^-1 and can be found by: D = Ko D. Removing Collected Material The particles collected must be removed from the collecting electrode. In low-voltage ESP applications, the build up is slow enough that the cleaning is normally done by shut down of the unit; however, for the high-voltage ESP, the dust build up is rapid and removal must be done often and usually while the unit is in service. The most widely used removal mechanism is the rapping of the electrodes. Hammers are used to tap the electrode hangers and cause the dust layer to fall off. How hard to rap and when to rap are vital questions that must be carefully answered in good design. Magnetic pulse rapping or compressed air rapping is often done at 1 to 2 raps per minute per electrode for the first sections. Vibrating the electrodes at about 50 to 100 Hz may be used for cleaning. Cleaning the electrodes , especially by rapping , releases puffs of dust. As a result, different portions of the ESP should be rapped at different times. The first section should be cleaned relatively often and the last section much less frequently. E. Particle-Migration Velocity and electrostatic force This is the speed at which a particle, once charged, migrates toward the grounded collection electrode. Variables affecting particle velocity are particle size, the strength of the electric field, and the viscosity of the gas. How readily the charged particles move to the collection electrode is denoted by the symbol, w, called the particlemigration velocity, or drift velocity. The migrationvelocity parameter represents the collectability of the particle within the confines of a specific ESP. The migration velocity is expressed as:
Fig. 3: Efficiency vs. Corona Power

Fig.2: Efficiency of ESP vs. Field Voltage

B. Corona Power The 30 to 75 kV peak voltage and the 50 to 600 mA current per section (0.1 to 1 mA/ft of wire) give corona power of 50 to 60 W/1000 cfm. Corona power is related to the penetration and efficiency by Log (q) = log (1- ) = Where q= fractional penetration, P= corona power level (W), Q= flow rate(cfm), and K= empirical constant (-0.02 to -0.2 cfm/W)

C. Bulk Resistivity and Dielectric constant The bulk resistivity and the dielectric constant for the particles to be collected will determine to a large extent the efficiency of collection. The bulk resistivity ( B, ohmcm) can be measured by placing a dust sample of known area (A, cm2) between two electrodes and measuring the voltage (V), current (I, amps), and the distance (x, cm) between the electrodes: B = B is a function of temperature, moisture, and chemical make up of the particles. It must be in the range of 10^4 to 2*10^10 ohm-cm for satisfactory ESp collection. The dielectric constant ( ) enters in to the migration velocity of particles because it affects the amount of charge that particles will take. The dielectric of a system is the resistance of the system to being broken down and

W= Where: q = particle charge(s) Ep = strength of field in which particles are collected, V/m = gas viscosity r = radius of the particle, m = 3.14 Migration velocity depends on the voltage strength of both the charging and collection fields. Therefore, the precipitator must be designed using the maximum electric field voltage for maximum collection efficiency. The migration Velocity also depends on particle size; larger particles are collected more easily than smaller ones. The particle collection inside the ESP depends on the flue gas velocity and the electrostatic force of the ESP system. The average gas velocity inside the collection chamber of an ESP varies from 0.5 m/s to 2 m/s. The flow stream with high velocity leaves the collection chamber

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

189

with poor particle collection. The particles require sufficient treatment time to get charged and collected inside an ESP. The electrostatic force, which is generated after the application of electric potential to the discharge electrodes, is strongly related to the movement of the particles. Fig3 shows how the particle collection efficiency of an ESP channel can be improved by increasing electric potential at the discharge electrodes and maintaining a reduced flow velocity at the inlet surface, which is 0.5 m/s in all four cases.

Greater radii of curvature on particles(smaller particles) Increasing dust loading up to a saturation level

Fig. 7: Efficiency of ESP vs. Drift Speed

Fig. 4: Particle Diameter vs. Collection Efficiency

Fig. 8: Efficiency vs. Drift Speed when particle diameter is varied. Fig. 5: Effective Drift Velocity vs. Resistivity

REFERENCES [1] Air pollution via. Henry C Perkins)


[2] [3] Air pollution; PART-B prevention and control.Marcel Dekker, Inc New York 1974. Nichols, G. B. 1976, September Electrostatic Precipitation Seminar presented to the U.S. Environmental Protection Agency. Research Triangle Park, NC. Pollution Control in Process Industries (SP Mahajan) Richards, J.R. 1995. Control of Particulate Emissions-Student Manual. (APTI Course 413). U.S. Environmental Protection Agency. Turner, J. H., P. A. Lawless, T. Yamamoto, D.W. Coy, G. P. Greiner, J. D. McKenna, and W.M. Vatavuk. 1992. Electrostatic precipitators. In A. J. Buonicore and W. T. Davis (Eds.), Air Pollution Engineering Manual (pp. 89-113).Air and Waste Management Association. New York:Van Nostrand Reinhold. U.S. Environmental Protection Agency. 1973. Air Pollution Engineering Manual. 2d ed. AP-40. U.S. Environmental Protection Agency. 1985. Operation and Maintenance Manual for Electrostatic Precipitators. Understanding and Controlling Air pollution (Howard E.Hesketh); Ann arbor science publishers Inc. Automatic Control and Management of Electrostatic Precipitator N. V. P. R. Durga Prasad, T. Lakshminarayana, Senior Member, IEEE, J. R. K. Narasimham, Thenmozhi M. Verman, and C. S. R. Krishnam Raju.

[4]

[5]

Fig. 6: Collection Efficiency vs. Particle Diameter

[6] [7] [8] [9]

IV. CONCLUSION Collection Efficiency can be increased by : Decreasing particle dielectric Decreasing gas velocity Increasing Electric wind Increasing collection area Increasing Electric field up to where arcing begins

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

190

Electrical Energy Conservation by Emergy Analysis


1
1
1

KamalKant Sharma, 2Gurinder Kaur Sodhi


2

M.M.UniversityMullna (Haryana), 2UGI, Lalru (Punjab)


sodhigurinder123@gmail.com

sharma_kamal@rediffmail.com,

Abstract: Energy is essential both as means of production and for its contribution to quality of life. The pace of Exploitation of energy resources has been growing over time and may result in depletion of scarce reserves.Using energy inefficiently creates waste in the entire worlds economics and has environmental impacts with local, regional and global implications Electrical energy is the most common and widely used type of Energy in the world. The subject of energy conservation is a concern for most energy users particularly industry. Energy conservation (ECON) becomes even more important for the third world, developing countries, where the rising energy costs and the use of efficient energy apparatus are of significant concern to both the industry and the efficient use of all resources is necessary both in an environmental and economic sense. The steps to create a sustainable energy system begin with the wise use of resources. It continues with the increased use of renewable resources. In recent years, integrated building design practices based on the definition of green building criteria as common standards of measurement have been promoted. In this work, an Emergy (spelled with an m) analysis has been applied to a building to account for the main energy and material inflows to the processes of building manufacturing, maintenance and use. Building materials, technologies and structural elements have been measured and compared to each other in order to evaluate their impacts and to provide a basic calculation that may be used for evaluation and selection. A comprehensive appraisal of the building industry is then expected through a series of synthetic indices. Results represent a source of information that will also be useful for future studies on the urban and regional scale. Keywords: Energy Conservation (ECON), Emergy Analysis I. INTRODUCTION

thermal power stations to develop new measures of energy conservation but also have systematic approach towards present trends of energy consumption through energy auditing and application of modern techniques and methods for minimizing energy wastage. Energy conservation is considered as a quick and economical way to solve the problem of power shortage as also means of conserving the countrys finite sources of energy. Energy conservation measures are cost effective, require relatively small investment and have short ge station as well as payback period. Electricity consumption has been growing at rate of 8 to 10 percent per year in India and the power sector accounts for more than 60 percent of the investments in the energy sector [1, 2] Emergy: (spelled with an "m") Odom H.T introduced the concept of Emergy. It is all the available energy that was used in the work of making a product and expressed in units of one type of energy embodied energy because they couldn't be generated with less". Emergy is a universal measure of real wealth of the work of nature and society made on a common basis. Emergy analysis presents an energetic basis for quantification or valuation of ecosystems goods and service. It attempts to assign the correct value to ecological and economic products and services based on a theory of energy flow in systems ecology and its relation to systems survive. Calculations of Emergy production and storage provide a basis for making choices about environment and economy following the general public policy to maximize real wealth, production and use. II EMERGY ANALYSIS 2.1 Historical backgrounds Emergy analysis is a part of a much larger theory developed by H.T. Odum about the functioning of ecological and other systems. As pointed out by Hall (1995b), the hypothesis about the role of energy in survival and evolution of systems has roots in the19th century and was first stated explicitly by the biologist Alfred Lotka (1922), who called the maximum power principle the fourth law of thermodynamics. Then, he adds that, Odum has both used and expanded the maximum power concept as a general systems hypothesis throughout his career.H.T. Odum started

Energy demand is not an exception to the economic theory of limited means and unlimited want [1]. The pace of exploitation of the energy resources has been growing over time, and has resulted in gradual depletion of scarce reserves. The critical link between energy and economy has exposed the vulnerability of nations to the volatile energy situation. The imperatives of energy shortage situation calls for energy conservation measure, which essentially means using less energy for same level of activity. While on one hand the demand for energy is increasing, on the other hand the energy resources are becoming scarce and costlier. The steady increase in the gap has not compelled technocrats and decision makers in the industry and

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

191

developing the roots of the concept of Emergy probably in the 1950s when he and E.P. Odum identified the importance of energetic to ecology that brothers subsequently realized the importance of the quality of energy and the necessity of using a common denominator for energy flows of different kinds (Hall, 1995b). From this concept, H.T. Odum extended the original concept as the Maximum Empower Principle, and developed an energy Systems language for the thermodynamics of open system. He also developed the concept that energy offered a common ground for integrating economic and ecosystems sciences.

1 sej of services are required to generate 1J of electricity(Fig 1).Solar Transformity is measured in sej/J. The Solar Transformity of a product is its Solar Emergy divided by its available energy, that is, M = t B (1) Where M is emergy, t is transformity, and B is available energy since solar energy is the baseline of all emergy Calculations, transformity of solar energy. III EMERGY ANALYSIS 3.1 Introduction About 30-40%of the total natural resources that are used in industrialized countries are exploited by the building industry. Almost 50%of this energy flow is used for weather conditioning (heating and cooling) in buildings. Almost 40% of worlds consumption of materials converts to the built environment and about 30% 0f energy use is due to housing. Since energy consumption, energy wasting, emissions and environmental impacts due to housing are expected to increase in the next few years, an accurate monitoring and management of the building industry is urgently required. Buildings could be conceived as thermodynamic engines that use energy to provide specific services, and that maintain their performance constant in time with respect to variable context conditions such as climate, temperature, humidity, sun irradiation and air motion. Buildings management therefore refers to the energy exchanges between buildings and their living context made by human beings and the surrounding environment. In particular material and energy flows to the building can be calculated in order to perform building environmental performances constant in time, with low levels of energy and material inputs.In brief eco-buildings have the following features -they make the most of energy and material inflows; -they supply a part of their energetic need through natural processes; -they use renewable and local materials; -they have minimal impact on natural cycles -they belong to their environmental context (resources, landscape, society, history) In the last few years, new sustainable building technologies have been developed and applied to buildings in order to achieve higher energetic efficiency and to reduce energy consumption and waste. An ecological assessment of buildings is expected to evaluate building technologies and materials and to define standards for making choices while taking into account the different steps of the building process from the cradle to the grave, from the extraction of raw materials to their assemblage and use and even until their disposal or recycling. Integrating accounting methods and synthetic indicators are then expected to provide general information on the environmental sustainability of buildings.

Fig. 1: Solar Transformity

2.2. Basic principle Emergy is measured in solar embodied joules, Abbreviated sej. .Emergy analysis characterizes all products and services in equivalents of solar energy, that is, how much energy would be needed to do a particular task if solar radiation were the only input. It considers the Earth to be a closed system with solar energy, deep Earth heat and tidal energy as major constant energy inputs, and that all living systems sustain one another by participating in a network of energy flow by converting lower quality energy into both higher quality energy and degraded heat energy. Since solar energy is the main energy input to the Earth, all other energies are scaled to solar equivalents to give common units. Other kinds of energy existing on the Earth can be derived from these three main sources, through energy transformations. Even the economy can be incorporated to this energy flow network as, wealth directly and indirectly comes from environmental resources measured by Emergy (Odum, 1996). 2.3 Emergy transformation Transformity: The emergy of one type required to make a unit of energy of another type. Transformity is the ratio of emergy to available energy Transformity is one example of a unit emergy value and is defined as the emergy per unit energy. For example, if 4000 solar emjoules are required to generate a joule of wood, then the solar transformity of that wood is 4000 solar Emjoules per joule (abbreviated sej/J). Transformities increase from left to right in the energy hierarchy diagrams. Solar energy is the largest but most dispersed energy input to the earth. The solar transformity of the sunlight absorbed by the earth is 1.0 b For example, since 4 coal emjoules (sej) of coal and

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

192

3.2 Indicators applied to the building industry An indicator is a tool able to give synthetic information regarding a more complex phenomenon within a wider sense.Indicators simplifies information that is often relative to multiple factors and enables investigators to communicate and compare results.The calculation of indicators follow different targets according to which of the two classes is noted: State pressure environmental indicators account for specific parameters, through conventional physical units, in order to verify their compatibility with specific environmental variables; they often evaluate much localized factors based on data collected in a specific area. Sustainability indicators provide a general evaluation based on a comprehensive balance, integrating a multiplicity of phenomena that may even be non homogeneous; they attempt to evaluate general behaviors of global sustainability with special reference to the problems of resource overexploitation and energy waste. 3.3. Emergy analysis of buildings In this section a case study is presented with emergy analysis applied to a building. This case study is applied to a contemporary building with very common characteristics, in order to provide more general information that may be applied to a widespread architecture. Since this study on a single building with common features attempts to evaluate general impacts due to the building industry and the portion of resource exploitation relative to housing, three phases have been separately: (1) Building manufacturing process (2) Building maintenance (3) Building use (1) Building manufacturing: this is the process of gathering and assembling materials to generate a built stock (the building) that persist during an indefinite lifetime as a permanent reservoir or memory of energy once spent. (2) Building maintenance: energy and material inflows are needed periodically in order to maintain the built stock (the building) constant in time; in other words, the restoring of standard technical requirements for building use resisting its physical entropic degradation. (3) Building use: a rectangle in the diagram represents interaction with users that need constant energy flows for lighting, cooling and heating; the main inputs to this phase are given by the consumption of electricity, gas and water. The emergy analysis enables us to evaluate the energy investment required for building manufacturing. The above result enable us to make a list of building materials based on their environmental cost (in terms

of sej) that depends on both their quantity and their transformity. In fact, since transformity is an indicator of energy hierarchy that accounts for all the inputs and transformation occurring in the production process, building materials have been evaluated through the emergy analysis by assessing both their environmental impact (quality) and their use in building industry (quantity). 3.4 Characteristics and criticisms of emergy analysis The most attractive characteristics of emergy analysis are: It provides a bridge that connects economic and ecological systems. Since emergy can be quantified for any system, their economic and ecological aspects can be compared on an objective basis that is independent of their monetary perception. It compensates for the inability of money to value non-market inputs in an objective manner. Therefore, emergy analysis provides an ecocentric valuation method. It is scientifically sound and shares the rigor of thermodynamic methods.Its common unit allows all resources to be compared on a fair basis. Emergy analysis recognizes the different qualities of energy or abilities to do work. For example emergy reflects the fact that electricity is energy of higher quality than solar insolation. IV. CONCLUSION Though the norms for lighting in switches have been revised, resulting into substantial savings in energy consumptions due to lighting, yet, energy efficient tubes , electronic chokes, sensors , dimmers etc. etc. have not been used. This is mainly on account of barriers it is recommended that we may identify buildings in each state for taking up energy efficient lighting installations using these energy saving devices and fixtures. This will definitely help in . increasing the awareness and energy conservation. The building industry is greatly concerned with environmental problems such as non-renewable materials and energy overexploitation. Housing involves chain processes that require inputs of materials and energy in different forms. An emergy synthesis has been applied to three phases, namely building maintenance composition and use, in order to give a measure of the environmental impact due to buildings and, more in general, to the built environment. Results show the emergy content ofa building conceived as a man-made emergy reservoir and the emergy flows for building maintenance and use. V. FUTURE SCOPE OF RESEARCH These outcomes provide a basis for future evaluations in the field of the building industry. Different building typologies, technologies and materials can be compared and contrasted with reference to different

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

193

manufacturing processes, as well as to maintenance composition and use (a materials durability, thermal efficiency and energy consumption during its lifetime). For example, different scenarios can be compared considering the emergy investment for manufacturing a special facade with an augmented thermal insulation or a passive ventilation system and its effects in terms of energy saving (reduced thermal dissipation) in the phases of building maintenance (a materials durability) and building use (energy for cooling and heating). Also, different building types can be compared through the indices of emergy per volume (referring to the building technology) and emergy per person (i.e. number of inhabitants, population density). Furthermore, these embuilding indices for different building types can be applied at the territorial level, giving a measure of the environmental impacts due to a whole urban setting. For example, emergy investments (for building manufacturing) and emergy inflows (for building maintenance and use) can be measured for a neighborhood composed of common housing and compared to a neighborhood of energy efficient ecobuildings with low environmental impacts. REFERENCES [1] Bhansali V.K, Department of electrical engineering, Jain
Narain vyas university, jodhpur, India; Energy Conservation in India-Challenges and AchievementsIEEE, Nov 1995 [2] Palanichamy c,Babu N,S Chelvan R.K, Nadarajan C Restructing the Indian powersector with energy conservations as the motive for Economic and Environmental Benefits.IEEE transaction on energy Conversion,vol 14,issue;4 pp 1589 -1596,December 1999 [3] Energy conservation/management in industryhttp://www.google.com/search? www.technopreneue.net/timeis/technology/conservation.htm+energ y+audit+industry [4] Odum H.T. & Odum E.C Environment and society in Florida; 449450 pp, 2000 [5] Odum, H.T. Environmental Accounting, Emergy and Decision Making. John Wiley, NY, 370 pp, 1996 [6] Hans-Holger Rogner; World Energy Assessment 2001, Chapter 5: Energy Resources (table 5.26) [7] Hans-Holger Rogner; World Energy Assessment 2001, Chapter 7: Renewable Energy Technologies (table 5.26) [8] M.T. Brown, H.T. Odum, S.E. Jorgensen Energy hierarchy and transformity In the universe,; Ecological Modeling 1728pp, 2004. [9] EWEA press release April, 2008 [10] Odum, Howard T. and Jan E. Arding, Emergy Analysis of Shrimp Mari culture in Ecuador, Center for Wetlands, University of Florida, Gainesville, FL 32611, 1991 [11] Odum, H.T. 1986. Emergy in ecosystems. pp. 337-369 in Environmental Monographs and Symposia, ed. by N. Polunin, John Wiley, NY. [12] Odum, H.T. Environmental Accounting, Emergy and Decision Making. John Wiley, NY, 370 pp, [13] Odum, H.T., E.C. Odum and M.T. Brown Environment and Society in Florida. Lewis Publication., Boca Raton, FL, 449 pp.1998 [14] Science man, D, G Pillet and T Murota Energy and Emergy. in Environmental Economics, Geneva, 257-276 pp,1998. [15] Science man, D., H.T. Odum, M. T. Brown. Ecological

[16]

Engineering (Elsevier) 212-218.pp 1998 Tennenbaum, S. Network energy expenditures for subsystem production. M.S. Thesis, Environmental Engineering Sciences, University of Florida, Gainesville, 132 pp.1998 [17] M.M. AboulNaga, Y.H. Elsheshtawy, Environmental sustainability assessment of buildings in hot climate

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

194

Promoting and Developments of Renewable Energy in Power System Technology and Energy Markets in India
1
1

Digambar Singhr, 2Naveen Kumar Sharma, 3Yog Raj Sood, 4R.K.Jarial


1,2,3,4

digambar.singh1986@gmail.com,2naveen31.sharma@gmail.com, 4 largi2001@gmail.com

NIT Hamirpur, India

yrsood@gmail.com,

Abstract: The renewable energy sources are regarded as related to present atmospheric movements. The Solar Electric systems, which are also called photovoltaic or PV systems are reliable and pollution-free. Solar radiation, along with secondary solar-powered resources and other such as wind and wave power, hydroelectricity and biomass, account for most of the available renewable energy on earth. India is gradually shifting focus toward its renewable energy resource, Driven by an increasing demand for electricity and widening gap between demand and supply. India has plans to generate 1,000 MW of power by 2013 and target of the solar generation electricity 20 GW in 2020. All types of energy, types of solar cell technology, renewable or non-renewable, can be traced back to either atmospheric activities in the past or to the present and future activities within the atmosphere and also describe the benefits and application of renewable energy. Keywords Renewable energy and electricity, Types of solar technologies, Solar power project, Photovoltaic application.

I. INTRODUCTION Solar electric systems, which are also called photovoltaic or PV systems, are reliable and pollutionfree. They make use of a renewable source of energy the sun and PV systems for homes and businesses, are becoming more affordable all the time. Radiant light and heat from the sun, has been harnessed by humans since ancient times using a range of ever-evolving technologies. Solar radiation, along with secondary solar-powered resources such as wind and wave power, hydroelectricity and biomass, account for most of the available renewable energy on earth. Only a minuscule fraction of the available solar energy is used [1]. India is densely populated and has high solar insolation, providing an ideal combination for solar power in India. In solar energy sector, some large projects have been proposed, and a 35,000 km area of the Thar Desert has been set aside for solar power projects, sufficient to generate 700 to 2,100 gigawatts. The average intensity of solar radiation received by India is 200 MW/km square (megawatt per kilometre square). With a geographical area of 3.287 million km square, this amounts to 657.4 million MW. However, 87.5% of the land is used for agriculture, forests, fallow lands,

etc., 6.7% for housing, industry, etc., and 5.8% is either barren, snow bound, or generally inhabitable. Thus, only 12.5% of the land area amounting to 0.413 million km square can, in theory, be used for solar energy installations. Even if 10% of this area can be used, the available solar energy would be 8 million MW, which is equivalent to 5,909 Mtoe (million tons of oil equivalents) per year. In July 2009, India unveiled a $19 billion plan, to produce 20 GW of solar power by 2020 [2].Under the plan, solar-powered equipment and applications would be mandatory in all government buildings including hospitals and hotels. On November 18, 2009, it was reported that India was ready to launch its National Solar Mission under the National Action Plan on Climate Change, with plans to generate 1,000 MW of power by 2013[2]. Technological breakthroughs for cost-effective photovoltaic technology could generate a quantum leap in the renewable energy sector since India is well endowed with solar insolation (average of 6 kWh/ sq.mt./day).India plans to announce increased subsidies for solar-power generation, as the country looks to scale up production of renewable energy and show it is committed to mitigating climate change. The Karnataka Power Corporation Limited (KPCL) has installed Indias largest solar photovoltaic power plant at Yalesandra village in Kolar district of Karnataka. Built at the cost of about $13 million, the plant makes use of modular crystalline technology to generate solar energy (Fig1).

Fig. 1: Indias largest solar photovoltaic power plant.

The numbers of solar appliances used in India are mentioned in table 1 below:

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

195

TABLE 1: TOTAL SOLAR POWER CONSUMPTION IN INDIA

Number of Solar road(Street) lights System Number of Home Light( illumination) systems Number of Solar lamp(Lanterns) Solar photovoltaic Power( energy) plants Solar water heaters System Box-shaped(type) solar cookers Solar photovoltaic pumps

55,795 342,607 560,295 Less Than< 10 MW 140 km2 of collector area 575,000 6,818

Fig. 2: Cumulative capacity additions planned in solar PV until 2012.

II. BENCHMARK CAPITAL COST NORM FOR SOLAR PV POWER PROJECTS FOR 2011-12 In Solar PV power project, the two major subsystems are Solar PV modules and balance of Systems (BOS). Solar PV modules constitute 65-70 % of the cost while BOS comprises the rest. The Solar PV modules are predominantly crystalline Silicon based. Within the modules the intermediate products at different stages of the manufacturing process are Polysilicon, Silicon Wafer, Solar Cell, and Solar PV Module. There is presently negligible domestic manufacturing capacity for Poly-silicon and Silicon Wafer in our country. The manufacturing capacity including projects under execution for Solar Cells and Modules are about 600 MW and 1000 MW respectively. The PV modules manufactured in the country meet the international standards. A. Future trends expected in global solar PV industry In 2010, increase in global PV manufacturing capacity and reduction in demand due to global recession has resulted in solar PV module price decline. According to the clean Energy Trends- 2010, the price drop, along with other internal financial and policy drivers, is causing some countries to reduce national incentive programs. Germany plans to reduce feed intariffs at the beginning of July 2010 by 11 to 16 percent, depending on the application. This move will further intensify the demand for less expensive PV in the solar market. Further, according to the Indian Semiconductor Associations report on Solar PV Industry 2010. Contemporary scenario and emerging trends: May 2010, the global capacity of solar PV modules is expected to reach 54 GW by end of year 2012 (shown in Fig2) [3].

B. Renewable Energy and Electricity There is unprecedented interest in renewable energy, particularly solar and wind energy, which provide electricity without giving rise to any carbon dioxide emission. Harnessing these for electricity depends on the cost and efficiency of the technology, which is constantly improving, thus reducing costs per peak kilowatt. Utilizing electricity from solar and wind in a grid requires some back-up generating capacity due to their intermittent nature. Policy settings to support Renewables are also generally required, and some 50 countries have these. Utilizing solar and windgenerated electricity in a stand-alone system requires corresponding battery or other storage capacity. The possibility of large-scale use of hydrogen in the future as a transport fuel increases the potential for both Renewables and base-load electricity supply. C. Condition for investing in large grid connected power plant The attractiveness of investing in large PV-grid connects power plants depends on the risk portfolio. The essential questions for investors in contradiction to home owners are: How secure is the law? What if the government changes? How safe is the legal situation of the power purchase agreement with the utility? How fast can a project be planned? How many authorizations are involved in the process for getting the permissions? In comparisons to a home owner investors are very radical with these uncertainties. As a matter of fact in most of the countries the bureaucracy and uncertainty to these questions are the lynchpins and often impedes the progression in the development of large PV-power plants. If only one of these mentioned factors is unclear, a large power plant will not be installed. There is only a YES or a NO for investors and this depends heavily on the return of investment and the risk level rather than on the nature of a particular investment. D. Solar Power Projects Map in India The map show the solar power project installation in India in which place the insolation is well in compare to other places (shows in Fig 3.).

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

196

lower contact layer is applied. From there, the surface is divided into bands and covered with a silicon layer. Lastly, the surface is re-banded and metal electrodes are applied. E. Other Solar Cells Other solar cell technologies less frequently used are the EFG (Edge Defined Film fed Growth) method and Apex solar cells from silicon, cadmium telluride solar cells and copper-indium selenide (CIS). EFG mono-crystalline solar cells are produced directly from silicon melt, eliminating sawing to wafers, which results in lower production costs and material savings due to reduced waste. IV. PHOTOVOLTAIC IS EMERGING As a major power source due to its numerous environmental and economic benefits and proven reliability. 1) The fuel is free. The sun is the only resource needed to power solar panels and the sun will keep shining until the worlds end. Also, most photovoltaic cells are made from silicon, and silicon is an abundant and nontoxic element 2) Produces no noise, harmful emissions or polluting gases. The burning of natural resources for energy can create smoke, cause acid rain, pollute water and pollute the air. Carbon dioxide CO2, a leading greenhouse gas, is also produced. Solar power uses only the power of the sun as its fuel. It creates no harmful byproduct and contributes actively to reduce the global warming. 3) PV systems are very safe and highly reliable. The estimated lifetime of a PV module is 30 years. Furthermore, the modules performance is very high providing over 80% of the initial power after 25 years which makes Photovoltaics a very reliable technology in the long term. In addition, very high quality standards are set at a European level which guarantees that consumers buy reliable products. 4) 4 PV Modules can be recycled and therefore the materials used in the production process can be reused. Recycling is not only beneficial for the environment but also for helping to reduce the energy needed to produce those materials and therefore the cost of fabrication [6]. 5) It requires low maintenance. Solar modules are almost maintenance-free and offer an easy installation. 6) It brings electricity to remote rural areas. Solar systems give an added value to rural areas House lighting, hospital refrigeration systems and water pumping are some of the many applications for off-grid systems. Telecommunication systems in remote areas are also well-known users of PV systems.

Fig. 3: Solar Power Projects installed Map in India.

III. TYPES OF SOLAR TECHNOLOGIES A. Photovoltaic Cells (PV) PV technologies make use of silicon solar cells to convert sunlight directly to electricity. PV cells can provide electric power to meet different needs - from small devices such as watches or calculators, to local electric utilities. B. Concentrating Solar Power (CSP) CSP technologies use reflectors to focus sunlight onto receivers that collect suns heat. This thermal energy can then be used to produce electricity via a steam turbine or heat engine. Previously, Floyd Associates published a research report on CSP and the major players within the sector [4]. PV Technologies: Photovoltaic modules that produce up to 100W or more of power are constructed by bringing together and connecting an array of solar cells. The properties of the module depend on the type of solar cell used. The following list represents the three main technologies currently used to create solar cells: C. Crystalline solar cells The foundation for crystalline solar cell production lies with a block of silicon ingot, which is sawn into thin silicon wafers [5]. The sawn wafers of 1mm thickness are then placed between two plates, plane and parallel to each other, which rotate in opposite directions and adjust the thickness of the wafer to 1/1000mm. From there, the wafers are cleaned, diffused at a temperature of 800 degrees Celsius, and covered with oxygen plasma. Finally, the cells are sintered at a high temperature, completing the cell production process. D. Amorphous Solar Cells Amorphous solar cell, otherwise known as thin-film solar cell, is another type of technology utilized to create PV modules. First in the solar cell production process the glass substrate is cleaned. Afterwards, a

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

197

7) It can be aesthetically integrated in buildings (BIPV). Systems can cover roofs and facades contributing to reduce the energy buildings consume. They dont produce noise and can be integrated in very aesthetic ways. European building legislations have been and are being reviewed to make renewable energies as a required energy source in public and residential buildings. This fact is accelerating the development of ecobuildings and positive energy buildings which opens up many opportunities for a better integration of PV systems in the built environment [7]. The energy pay-back time of a module is constantly decreasing. This means that the time required for a PV module to produce as much energy as it needs to be manufactured is very short, it varies between 1,5 years to 3 years. A module therefore produces 6 to 18 times more energy than is needed to manufacture it. It creates thousands of jobs. The PV sector, with an average annual growth of 40% during the past years is increasingly contributing to the creation of thousands of jobs in Europe and worldwide. It contributes to improving the security of Europes energy supply. In order to cover 100% of the electricity demand in Europe, only the 0.7% of the total land of Europe would be needed to be converted by PV modules. Therefore Photovoltaics can play an important role in improving the security of Europes energy supply. V. PHOTOVOLTAIC APPLICATIONS The Photovoltaic technology can be used in several types of applications: A. Grid-connected domestic systems This is the most popular type of solar PV system for homes and businesses in developed areas. Connection to the local electricity network allows any excess power produced to feed the electricity grid and to sell it to the utility. Electricity is then imported from the network when there is no sun. An inverter is used to convert the direct current power produced by the system to alternative current (AC) power for running normal electrical equipments (shows in Fig 4).

These systems, also grid-connected, produce a large quantity of photovoltaic electricity in a single point. The size of these plants range from several hundred kilowatts to several megawatts. Some of these applications are located on large industrial buildings such as airport terminals or railway stations. This type of large application makes use of already available space and compensates a part of the electricity produced by these energy-intensive consumers. C. Off-grid systems for rural electrification Where no mains electricity is available, the system is connected to a battery via a charge controller. An inverter can be used to provide AC power, enabling the use of normal electrical appliances. Typical off-grid applications are used to bring access to electricity to remote areas (mountain huts, developing countries). Rural electrification means either small solar home system covering basic electricity needs in a single household, or larger solar mini-grids, which provide enough power for several homes. D. Hybrid systems A solar system can be combined with another source of power - a biomass generator, a wind turbine or diesel generator - to ensure a consistent supply of electricity. A hybrid system can be grid-connected, stand-alone or Grid-support [8]. E. Consumer goods Photovoltaic cells are used in many daily electrical appliances, including watches, calculators, toys, battery chargers, professional sun roofs for automobiles. Other applications include power for services such as water sprinklers, road signs, lighting and phone boxes. F. Off-grid industrial applications Uses for solar electricity for remote applications are very frequent in the telecommunications field, especially to link remote rural areas to the rest of the country. Repeater stations for mobile telephones powered by PV or hybrid systems also have a large potential. Other applications include traffic signals, marine navigation aids, security phones, remote lighting, highway signs and waste water treatment plants. These applications are cost competitive today as they enable to bring power in areas far away from electric mains, avoiding the high cost of installing cabled networks. VI. LONG TERM POTENTIAL OF SOLAR POWER In the long term it is estimated that solar power could contribute to an increasing part of the total energy consumption. With appropriate policies both in developed and developing countries, EPIA and Greenpeace have devised in a joint scenario, that in 2030 (shows in Fig 5.), photovoltaic could produce

Fig. 4: Grid-connected domestic systems

B. Grid-connected power plants

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

198

enough energy to supply electricity to 3, 7 million people globally. The majority of them will be located in remote areas where there is no access to the electricity grid [9].

grids in the areas of Eastern Macedonia - Thrace will be completed, and thus it will be possible to implement 50% of the projects for which production authorizations have already been granted on condition that they will be implemented following the extension/enhancement of the System, of a capacity of 115MW.
TABLE 3: OPTIMISTIC SCENARIO- ESTIMATED POSSIBLE RES OUTPUT IN 2010. Source Instal led capac ity in (Sep 2009) [MW] Instal led permi t (Sep 2009) [MW] Additio nal wind farm due to already planned activity [MW] 58 Addition al RES (50-80% of installati on permit) [MW] 524 47 Estim ated total capaci ty in 2010 [MW] Esti mate d outp ut in 2010 [TW h] 3.92 0.7

Fig. 5: Global Cumulative Capacity up to 2030 - Advanced scenario.

The following table demonstrates the current capacity utilization of alternative sources of energy and their potential given in the table 2.

TABLE 2: RENEWABLE ENERGY CURRENT CAPACITY AND POTENTIAL THE END OF MARCH, 2009 S. No Alternative Energy Current capacity(MW) Potential (MW)

1 2 3 4

5 6

Wind power Bio Power Bagasse cogeneration Small Hydropower(up to 25) Energy Recovery From waste Solar photovoltaic power Biomass / cogeneration Biomass gasifier Total

10242.5 703.3 1048.73 2429.67

45195 16881 5000 15000

Wind Smallscale hydros Largescale hydros Biomas s Geother mal Solar Total

1140 180

1048 93

1722 227

3018

344

3362

4.57

41

47

23 0

64 0 133 5058

0.49 0 0.18 9.86

37 3316

83 1271

402

66+30 690

B.

92.97 2.12

2700 -

7 8

170.78 105.46 14795.53

A. Optimistic Scenario The optimistic scenario for the penetration of RES in 2010 is presented in Table 3. This predicts that the 2010 target will be realized by 80.4%, i.e. 9.86 TWh, as compared to 12.26 TWh [10]. The implementation of 50% of the investments for which installation permits have already been granted will continue without any obstacles. This assumption is realistic considering that these projects are at a mature stage, no appeals have been filed against them with the Council of State, access thereof to the grid has been guaranteed, and consequently it is possible to secure funds for these projects. The extensive works for enhancing local

Conservative scenario The three conditions of the optimistic scenario described in this work that could prevent the realization of the target. That is why a more conservative scenario [10] has also been formulated, taking into account the following: That 20% of the projects for which installation permits have already been granted (50% of PV systems and an additional 20MW from small-scale plants) will be implemented from the last months of 2009 up to the end of 2010. That there will be a delay in the completion of the works for enhancing local grids in the areas of Eastern Macedonia Thrace and the completion of the licensing process for relevant additional RES projects, and thus it will be possible to implement only 20% of the projects for which production authorizations have already been granted on condition that they will be implemented following extension/enhancement of the system. That the planned large-scale hydroelectric projects will not be implemented.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

199

TABLE 4: CONSERVATIVE SCENARIO ESTIMATED POSSIBLE RES IN 2010. Source Installe d capacity in (eally 2 009) [2009] Additio nal wind farms due to already planned activitie s[MW] -35 Additi onal RES in the rest of Greece [MW] Conser vative estima te of total capacit y in 2010 1373 199 Estim ated output in 2010 [TWh]

[9] Published in the frame of the RESTMAC project Creating Markets for Renewable Energy Sources financed by the 6th European Framework Programme for Research European Photovoltaic Industry Association (EPIA), 2009. [10] 5th Annual Report On The Penetration Level Of Renewable Energy Sources In The Year 2010.

Wind Smallscale hydros Largesacle hydros Boimass Geotherm al solar Total

1140 180

-314 -28

3.13 0.61

3018

-344

3018

4.2

41 37 4416

-397

-14

50 0 99 4739

0.38 0 0.13 8.45

-34 -390

According to the above scenario, the share of renewable energy in the gross electricity consumption will reach 69% of the target, as shown in Table 4, i.e. 13.85%. In view of the above, it is clear that additional measures and policies are required to realize the 20.1% target [10]. V. CONCUSSION This paper presents the Renewable Energy capacity in India and installation of Renewable Energy Generation. Cumulative capacity additions planned in solar PV until 2012 and Future trends expected in global solar PV industry. In this paper describe the area in which the more insolation of sun light is available in India, the solar power plant is installed project map, types of solar cell technology, the conservative scenario and the optimistic scenario for the penetration of RES in 2010. This predicts that the 2010 target will be realized by 80.4%, i.e. 9.86 TWh, as compared to 12.26 TWh and also describes the Long term potential of solar power the grid connected photovoltaic power plant and application. REFERENCES
[1] Detail of solar power in India available at: http://en.wikipedia.org/wiki/solar/power in India. [2] Future Perspectives for Renewable Energy in India available at:http://www.alternative-Energy-news.info/future-renewableenergy-india. [3] Benchmark Capital Cost Norms for Solar PV Power Projects and Solar Thermal Power Projects September 2010. [4] www.floyd-associates.com/solar2. [5] J. Zuboy, S. Sczepanski, S. Moon, D. Gwinner, and R. Nahan, DOE Solar Energy Technologies Program: Overview and Highlights. US Department Of Energy (DOE) May 2010. March 2010. [6] Information is available on website: www.pvcyle.org. [7] Information is available on website: www.pvsunrise.eu [8] More information is available on www.ruralelec.org.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

200

Important Soil Related Earthing Parameters


1

Arjunsingh Ashok Mehta , 2S. N. Singh, 3M. K. Singhal


1
2 1 sunny_arjun1@yahoo.co.in aheciitr.@gmail.com, 3ahec@iitr.ernet.in

Indian Institute of Technology Roorkee, India

Abstract: Since the earliest days of electric supply, the soil has been used as a conductor of electricity. The electrical properties of the soil are in themselves of interest and importance, particularly the specific resistance or soil resistivity. The soil resistivity is one of the particular factors in determining the resistance of any earth electrode as it determines the division of current between the mass of earth and the earthing arrangement. Factors like corrosion provide the estimate regarding the life of the earthing structure and impulse coefficient helps in the selection of protection equipments. Keywords Soil resistivity, Corrosion, Impulse coefficient.

I. INTRODUCTION The resistance of earth for a given earthing arrangement depends upon the electrical resistivity of the soil in which it is installed [1]. Therefore, it is one of the most important parameter in designing the station grounding system is the resistivity of the soil in the station area. In fact, earthing design starts from the resistivity of the local soil. II. ELECTRICAL CHARACTERISTICS OF SOIL
An easy way to comply with the conference paper formatting requirements is to use this document as a template and simply type your text into it.

material can withstand. This condition is illustrated by the presence of gaps in Fig 1. Because the substation grounding system normally is designed to comply with far more stringent criteria of step and touch voltage limits, the gradient can always be assumed to be below the critical range [2]. Step voltage is the difference in surface potential experienced by a person between his 2 feet (1 step) approximately bridging a distance of 1m without being in contact with any other grounded object, while touch voltage is the surface potential at a point where the person is standing and touching a grounded object by hand. If the actual step and touch voltage are less than the respective acceptable values i.e. Estep and Etouch given by following expressions, then the design is considered satisfactory [1]. 0.157 (1) Touch Potential Etouch (1000 1.5C s s )
tf

Step PotentialEstep

(1000 6Cs

s)

0.157 tf

(2)

Where, tf = fault duration (in s) Cs = Reduction factor s = Upper layer soil resistivity (in .m) C. Effect of Current Magnitude Soil resistivity in the vicinity of ground electrodes may be affected by current flowing from the electrodes into the surrounding soil. The thermal characteristics and the moisture content of the soil will determine if a current of a given magnitude and duration will cause significant drying and thus increase the effective soil resistivity. A conservative value of current density is not to exceed 200 A/m2 for 1 s [2]. III. SOIL RESISTIVITY VARIABLES Soil resistivity is not constant as there are number of factors affecting it. Principal factors affecting it are [3]: A. Type of Soil The soil composition can be of clay, gravel, loam, rock, sand, shale, silt, stones, etc. In many locations, soil can be quite homogenous, while other locations may be mixtures of these soil types in varying proportions, thereby having different ground resistance. D. Seasonal Conditions The effects of heat, moisture, drought and frost can introduce wide variations in normal soil resistivity. Soil resistivity usually decreases with depth, and an increase of

A. Electrical Properties of Soil The behaviour of a ground electrode buried in soil can be analysed by means of the circuit in Fig 1. As shown, most soils behave both as a conductor of resistance, r, and as a dielectric. Except for high-frequency and steep-front waves penetrating a very resistive soil material, the charging current is negligible in comparison to the leakage current, and the earth can be represented by a pure resistance [2].

Fig. 1: Electrical Model of Soil [2]

B.

Effect of Voltage Gradient The soil resistivity is not affected by a voltage gradient unless the latter exceeds a certain critical value. The value somewhat varies with the soil material, but it usually has the magnitude of several kV/cm. Once exceeded, arcs would develop at the electrode surface and progress into the earth so as to increase the effective size of the electrode, until gradients are reduced to values that the soil

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

201

only a few percent of moisture content in a normally dry soil will markedly decrease soil resistivity. The effect of seasonal variables on soil resistivity is shown in Fig 2 and Fig 3.

of various sizes are present, the spaces between the large grains may be filled with smaller ones and the resistivity will be reduced as a result. IV. USE OF SURFACE SOIL LAYER Touch and step potential are transferred in human body through the earthing resistance of the human foot [6]. A layer of crushed rock is often spread on the surface of earth above the ground grid. The area covered by this crushed rock layer is generally of significant size to validate the assumption of the feet being in contact with the material of uniform resistivity in the lateral direction. If the underlying soil has a lower resistivity than the crushed rock, only some grid current will go along the thin upper layer of the crushed rock and the surface voltage will nearly be the same as that without the rock layer. Crushed rock is used as a surface layer in substation. It has following functions [6]: 1) Increases the contact resistance between the soil and the feet of the personnel: Therefore, a surface of crushed rock around the metallic equipment and structures will provide the necessary high resistivity layer below the feet to enable a person to withstand higher voltage. However, this resistance may be considerably less than that of the crushed rock layer, how much less depends on the relative values of earth and crushed rock resistivities and the thickness of the rock layer. 2) Keeps reptile away as they find it difficult to crawl on it: India is a tropical country and has large population of reptiles. In order to minimize the hazards created by them it is necessary that the electric equipment and the structures supporting conductors are surrounded by a surface layer of crushed rock. 3) Avoids formation of oil pool: In case of oil leakage from transformer, CB etc. crushed rock layer prevents the oil pools to be formed. 4) Prevents evaporation of moisture from underlying layers [7]: The crushed rock top layer acts as a cover for lower layers and prevents it from coming in contact with atmospheric conditions thereby keeping them moistened. V. CORROSIVE PROPERTIES OF SOIL: Soil is a heterogeneous mixture of many micro and macro structures of varying physical and chemical characteristics. Corrosion of metals in soils at normal temperature is primarily electrochemical in nature and results from the operation of numerous galvanic corrosion cells [8] Corrosion, in general, can be defined as the deterioration of a substance or its properties due to the reaction with its environment. The corrosion process can be chemical, electrochemical, or physical. Corrosion should be considered when designing an earthing system. The earthing arrangement

Fig. 2: Effect of Moisture on Soil Resistivity [4]

Fig. 3: Effect of Temperature on Soil Resistivity [4]

E.

Other Factors Other soil properties conducive to low resistivity are chemical composition, soil ionization, homogeneous grain size and even grain distribution - all of which have much to do with retention of soil moisture, as well as providing good conditions for a closely packed soil in good contact with the earth rod. Some of them are explained below: 1) Effects of Dissolved salts in water: It is a well-known fact that the resistivity of water is governed by the salts dissolved in it. There is large variation in resistivity with respect to the percentage of the salts dissolved which is as shown in Fig 4.

Fig. 4: Effect of added salts on Soil Resistivity [4]

2) Effect of Size and Distribution of Grains [5]: Grain size and the presence of grains of various sizes undoubtedly play an important part in determining the resistivity. The grain size and its distribution have an effect on the manner in which the moisture is held. With large grains, moisture is probably held by surface tension at the point of contact with the grains. If, however, grains

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

202

could be affected when they are buried or submerged in a corrosive environment. The various factors which affect the underground corrosion are as follows [9]: a. Soil Resistivity It is a measure of almost all physicochemical properties of the soil. The resistivity of soil depends on natural ingredients, the amount of salts dissolved in soil, and the moisture content. The corrosivity increases with the reduction of soil resistivity. The relationship of corrosivity and soil resistivity is presented in Table I [9].
TABLE I: CORROSION RATES FOR VARIOUS SOIL RESISTIVITIES

metals and iron tends to become ennobled with aeration under some conditions. However, if there are differences in the rate of air penetration through adjacent sections of ground, well aerated areas will become the cathode and poorly rated areas become the anode of the galvanic corrosion cell. Micro structure of soil ingredients, moisture contents, capability of soil to retain moisture and depth of soil above metal, affect the transport of oxygen to metal surface. In general, transfer of oxygen is faster in loose and grainy soils as compared to dense and wetted soils. Corrosion decreases with the increase in aeration [8]. d. Moisture Moisture contents depend on season, location, soil type, particle size and ground water level. The degree of wetness contributes to the corrosion by dissolving soluble salts thereby changing the soil composition. The rate of air penetration and resistivity of soil decreases with increase in moisture content. e. Miscellaneous factors These are those that are difficult to classify because they are a combination of one or more of the above and include the effect of temperature, bacterial, or interference current effects. These factors typically contribute no more than 10% to the total corrosion rate and are usually neglected. The large variation in the chemical characteristics of soil and the resulting secondary effects afford little chance to predict precisely the rate of corrosion unless a careful and extensive study of the particular soil, where conductors are to be laid, is carried out which too may not be very dependable owing to the constant changes taking place in the soil characteristics with time [interconnection of grounding mats of diff material]. Nevertheless a fair degree of correlation has been found between corrosion and the above soil factors affecting it. Therefore, it can be written that Y = f(X1, X2, X3, X4) (3) Where, Y = Corrosion rate (mils/yr) X1 = Resistivity (.m) X2 = pH value X3 = Moisture (%) X4 = Aeration (%) Using simple multiple regression analysis, the following equation is obtained for estimation of corrosion of steel in any environment (for normal operating ranges): Y = 3.36 9.6310-5 X1 + 0.29 X2 + 0.034 X3 + 0.012 X4 (4) Equation (4) is obtained from experimental data and is limited by extreme corrosion conditions, such as, extremely high resistivity (> 10,000 ohm-cm) or extremely low aeration quantities (< 3%). Corrosion reduces by these conditions to almost zero. This equation is applicable for all steel rods up to 3-inch diameter and is applicable for the first 12 years. It has also been obtained

Soil Resistivity (.m) < 25 26 50 51 100 > 100

Typical corrosion rate (mils/year)

Severely corrosive ( > 13) Moderately corrosive (9 - 12) Mildly corrosive (4 - 9) Very mildly corrosive ( < 4)

b. pH Value pH of soil affects the corrosion process greatly. The more acidic the soil is, the higher the corrosion rate. pH value ranges generally from 5 to 10 in soil; a value of 7 indicates neutrality (lower values, acidity; and higher values, alkalinity). The presence of salts reduces the resistivity of soil and also may increase its heterogeneity. Acid salts generally increase the corrosion and alkaline salts act as corrosion inhibitors Weak oxidizing salts are extremely corrosive but strong oxidizing salts make the steel surface passive. The general relationship between the pH values and corresponding corrosion is shown in Table II [9].
TABLE II: CORROSION RATES FOR VARIOUS PH VALUES

Soil characteristics Extremely Acid Very strongly Acid Strongly Acid Medium Acid Slight Acid Neutral Mildly Alkaline Moderately Alkaline Strongly Alkaline Very strongly Alkaline

pH values Below 4.5 4.5 5.0 5.1 5.5 5.6 6.0 6.1 6.5 6.6 7.3 7.4 7.8 7.9 8.4 8.5 9.0 9.1 Higher

Corrosion Rate Highest corrosive

Least corrosive

Higher corrosive

c. Aeration It is a measure of the availability of oxygen to the metal. Aeration characteristics of a soil are dependent primarily on particle size and distribution. Some investigators consider that corrosivity of soil is directly related to its air penetration property and have postulated that poor aeration results in severe corrosion (8). This may be due to the fact that oxygen influences the static potential of

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

203

experimentally that the average corrosion rate in the following 12 years reduces to half of that value in the first 12 years and is negligible thereafter [9]. VI. IMPULSE PROPERTY OF SOIL The soil behaves like a non-linear resistance and electric breakdown may also occur. The effect of the magnitude of current flowing from the grounding arrangement is due to the fact that with increase of the current, field intensity in the soil increases. When the field intensity reaches a critical value at which the breakdown of soil occurs, branched conducting paths are developed. These conducting paths in the immediate vicinity of the electrode give the effect as if the dimensions of the electrode are increased. As a result of this, the grounding impedance of the electrode decreases. The impulse grounding impedance (Z), of a grounding system is the impedance between the point where the impulse current is injected in the grounding system and another electrode of zero grounding resistance in the ground at infinite spacing. It is defined as ratio of the peak of the voltage developed at the feeding point to the peak value of the impulse current [10]. The ratio between the impulse grounding impedance and the power frequency grounding resistance is called impulse coefficient, A. It may vary from 0.05 to 4 and depends on nature of soil, magnitude and wave shape of the impulse current, point of feeding, type of the grid and spacing between the conductors of the grid [11]. Also impulse impedance is more sensitive to the moisture than to salt content For rectangular grounding grids, if r re Where r = Radius of the circular plate having an area equal to the area of the grid (in m) re = Radius of the circular plate having an area equal to the effective area of the grid (in m) The impulse impedance of a square grid is given by Z=AR (5) Where R = Power frequency grounding resistance of the grid (in ) A = Impulse coefficient and is given by following expression.
0.333

For rectangular grids with r > re, impulse impedance is same as that for radius re. For square grids length to breadth ratio i.e. X will be 1. VII. SOIL RESISTIVITY IMPROVEMENT TECHNIQUES Several various methods had been applied to decrease the grounding resistance of the grounding system. Regular methods include enlarging the grounding grid, connecting the main grounding grid with a subsidiary external grounding grid, increasing the burial depth of the grounding grid, adding long vertical grounding electrodes, and changing the soils around the grounding grid with low resistivity materials. Sometimes these methods may be of little use. Under such conditions unique and unconventional methods to reduce earth resistance are employed a. Bentonite Although soil treatment may be applied to improve earth electrode contact resistance in special or difficult locations, but migration and leaching of applied chemicals over a period of time reduces the efficiency of the system progressively, requiring constant monitoring and replacement of the additives. For temporary electrical installations in areas of high resistivity, this may be the most economic method for obtaining satisfactory earth contact. However, if a greater degree of permanency is envisaged, earth electrodes packaged in material such as bentonite are preferable. Bentonite or similar material may be used to advantage in rocky terrain. Where holes are bored for the insertion of vertical electrodes or where strip electrodes are laid radially under shallow areas of low resistivity overlaying rock strata, bentonite packing will increase the contact efficiency with the general mass of ground [1] 5) Properties of Bentonite: The LRM, such as bentonite, is black colour natural clay containing the mineral Monmorillonite, Illite and Keolonite, which was formed by volcanic action years ago. Depending upon the proportion of bentonite to water, the bentonite suspension can be either a very liquid or viscous but pourable pasty compound. This electrolyte will not gradually leach out, as it is part of the clay itself. Provided with a sufficient amount of water, it swells up to 13 times its dry volume and will adhere to nearly any surface it touches. Due to its hygroscopic nature, it acts as a drying agent, drawing any available moisture from the surrounding environment. Bentonite needs water to obtain and maintain its beneficial characteristics [13]. The dried bentonite powder has very high electrical resistance. However, when mixed with water the bentonite suspension has a high specific conductivity. Its initial moisture content is obtained at installation when the slurry is prepared. Once installed, the bentonite relies on the presence of ground moisture to maintain its characteristics. The bentonite suspension adheres well to the earthing material and penetrates into the cavities of the soil. The bentonite suspension binds

A
Where
re

r re

2.3

(6)

K T Xc

(7)

= Soil resistivity (in .m) T = Wave front time (in -second) K = (0.05 1.45 S) for centre fed grid = (0.025 0.6 S) for corner fed grid S = Spacing between conductor (in m) X = Length to breadth ratio of the grid C = 0.029 for centre fed grid = 0.08 for corner fed grid

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

204

well over long periods of time. This has a very favourable effect on the earth resistance of earth electrodes during very dry periods. Best results are obtained by mixing bentonite with water in the ratio of 1:4 by weight [14]. It is noncorrosive, stable, and has a resistivity of 2.5 m at 300% moisture with a specific density of 1.2 g/cm3. These properties of bentonite solve the compaction and soil-rod contact problem [15]. The low soil resistivity results mainly from an electrolytic process between water, Na2O (soda), K2O (potash), CaO (lime), MgO (magnesia) and other mineral salts that ionize forming a strong electrolyte with a pH ranging from 8 to 10. Most soils have sufficient ground moisture so that drying out is not a concern. The hygroscopic nature of bentonite will take advantage of the available water to maintain it as an installed condition. If exposed to direct sunlight, it tends to seal itself off, preventing the drying process from penetrating deeper. It may not function well in a very dry environment, because it may shrink away from the electrode, increasing the electrode resistance [7]. Therefore it can be used to provide suitable earth resistance for sites in rivers, on sea-shores, in sand and in gravel soils as well as in mountainous locations. 6) Advantages of Bentonite [14] The advantages of using bentonite are as follows: The resistance values of bentonite treated electrodes varied from 70% to 20% of the reference electrodes. The variation of earth resistance is negligible in the case of bentonite treated ground electrodes whereas it is pronounced in the case of untreated reference electrodes. This means, that even during the summer months, bentonite suspension retains moisture whereas natural soil dries up. The extent of the bentonite cover around the electrode influences the efficacy of the treatment. As the cover increases, the earth resistance value decreased. A cover of 125 mm thickness around the electrode would be adequate to obtain a significant reduction in earth resistance of ground electrode was between 25 to 30 % of the untreated reference electrode. Bentonite has tremendous capacity to absorb water and retain it over long periods of time. As the resistivity of the surrounding soil drops, the bentonites benefit will diminish since the assumption that its resistivity is negligible can no longer be considered valid. b. Explosove Earthing Technique A huge deep distributed grounding system in the substation area can easily inject current into the deeper soil layers. As the area available to dissipate fault currents increases, the resistance of the grounding system decreases. Realistically, it is difficult to construct a large grounding system within the area defined. In order to decrease the grounding resistance, a special methods is proposed to decrease the grounding resistance of grounding grids in high resistivity area, it is called as

explosive grounding technique. Several vertical holes are drilled, and then appropriate explosive agents are introduced into the holes. The resulting explosions create various cracks in the soil. Finally, low resistivity materials (LRM) are injected into the holes and cracks under high pressure. As a result, a large number of cracks around the vertical conductors are filled with LRM and a large 3dimensional grounding network consisting of the ground conductors and cracks is formed as illustrated in Fig 5.

Fig. 5: Grounding System formed By Explosive Grounding Technique

The cracks around the vertical conductors filled with LRM form a complex network of low resistivity tree-like cracks linked to the substation grid is formed [15]. The basic idea of this method comes from the usual practice of building the foundations of transmission line towers in rocky regions by triggering explosions in holes and filling them with concrete. The explosion course is carefully planned, and the region close to the surface remains intact [16]. The schematic diagram of a single vertical grounding electrode formed by the explosive grounding technique is shown in Fig 6.

Fig. 6: Schematic Diagram of single vertical grounding electrode by the Explosive Grounding Technique

7) Principles of Explosion Earthing Technique[16] Contacting Deep Soil Layers with Lower Resistivity: It has been observed that there are usually low soil resistivity layers or layers saturated with underground water in the regions with high resistivity. The new method can effectively use these low resistivity layers to decrease ground resistances. Reducing Contact Resistances: The low resistivity materials which fill the holes provide a very low

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

205

contact resistance between the ground conductors and soil. Decreasing Leakage Resistance: The ground impedance of a grounding system is mainly the distributed (leakage) resistance of soil; it is mainly controlled by the leakage resistance of the adjacent soil around the grounding grid. So, reducing the resistivity of this portion soil can effectively decrease the grounding resistance. The explosive can form a huge region filled with LRM, and the equivalent diameter of this region is about 5 to 40 meters, or bigger. Therefore, expanding the number of conductive branches available for the current to dissipate in the soil can effectively decrease the ground resistance. When the new method is used, a large network of soil cracks filled with LRM is formed. This network acts like a virtual extension of the grounding system to deeper soils, resulting in lower ground resistances. Links to Intrinsic Soil Cracks: It is known that intrinsic cracks exist in rocky areas. The cracks caused by the explosion often connect with the intrinsic cracks in the rock. These intrinsic cracks are typically filled with moisture and usually extend to remote locations. The connected intrinsic cracks and explosion cracks are filled with LRM by the high pressure injection. VIII. CONCLUSION The paper discusses in brief various soil related parameters employed for designing an earthing system. Along with the explanations, an equation has been provided to roughly estimate the corrosive action for earthing conductors. Even though, impulse impedance is not a part of earthing system designing, it must be kept in limits to allow the proper operation of the protective gear. ACKNOWLEDGMENT The authors greatly acknowledge the financial support from the Ministry of Human Resource Development (MHRD), Government of India in the form of stipend for carrying out research work. REFERENCES
[1] [2] [3] [4] IS: 3043 1987, Code of practice for Earthing. pp 16. IEEE Guide for Safety in AC Substation Grounding. IEEE Std 80 2000. Publication year 2000, pp 49. Manual for Earthing, RFI Industry Pty. Ltd. IEEE Guide for Measuring Earth Resistivity, Ground Impedance and Earth Surface Potentials of a Ground System. IEEE Std 81 1983, Publication Year 1983, pp. 9. Tagg, G. F., Earth Resistance, George Newnes Limited, 1964, pp -7. Central Board of Irrigation and Power, Earthing System Parameter for HV, EHV and UHV Substation, 1985. Nair, R. G., Design of Earth Mat in EHV Stations An Optimization Technique, 53rd Research & Development Session, Central Board of Irrigation and Power, Distribution, 1986, pp. 139 -145. Manohar, V. N., Nagar R. P., Design of Steel Earthing Grids in India, IEEE Transactions on Power Apparatus and Systems, VolPAS 98, 1979, pp. 2126.

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

Sen, P. K., Malmedal K., Nelson J. P., Steel Grounding Design Guide and Application Notes, IEEE Rural Electric Power Conference, 2002. pp C2 10. Gupta, B. R., Impulse Impedance A New Design Parameter for Earthing grids, 53rd Research & Development Session, Central Board of Irrigation and Power, Distribution, 1986, pp. 207 210. Thapar, B., Dixit, S., Impulse Characteristics of Soil and Grounding Electrodes. 45th Research & Development Session, Central Board of Irrigation and Power, Vol -V Power, 1976, pp. 79 85. Parameswaran, S., Nauk, V. M., Use of Bentonite for Effective Earthing, 49th Research & Development Session, Central Board of Irrigation & Power, Vol I Power, 1981, pp. 173 - 176. Jones, W. R., Bentonite Rods Assure Ground Rods Installation in Problem Soils, IEEE Transactions on Power Apparatus and Systems, Vol PAS 99, Issue 4, 1980, pp. 1343 1346. Youping, T., Jinliang, H., Rong, Z.,Lightning Impulse Performance of Grounding Devices Covered With LowResistivity Materials, IEEE Transactions on Power Delivery, Vol 21, Issue. 3, 2006, pp. 1706 1713. Quinbo, M., Jialiang, H., Dawalibi, F. P., Ma, J., A New Method to Decrease Ground Resistance of Substation Grounding Systems In High Resistivity Regions, IEEE Transactions on Power Delivery, Vol 14, Issue 3, 1999, pp. 911 916. Rong, Z., Jinliang, H., Jun, Z., Xinfu, S., Novel Method in Decreasing Grounding Resistances of Urban Substations by Utilizing Peripheral Geographical Conditions, 37th IAS Annual Meeting. Conference Record of the Industry Applications Conference, Vol 2, 2002, pp. 11131119.

[5] [6] [7]

[8]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

206

Automatic Generation Control with Fuzzy Logic Controller in Deregulated Power System
1 1,2

Rintu Khanna, 2Vineet Vikram

PEC University of Technology, Chandigarh, India 1 rintukhanna1@rediffmail.com To create several competing electricity generation companies (horizontal unbundling) and to recognize that the power transmission system is a natural monopoly and accordingly, to make special regulatory provision in this respect. To allow consumers to exercise choice between suppliers (generating companies) while still using the existing transmission facilities. Automatic generation control for deregulated environment needs to be modified for the adaptation to the market system with several kinds of bidding strategies. In order to achieve the economic optimality, it is necessary to take into account the bidding results in the AGC scheme. The conventional AGC system adopts algorithm to obtain optimal generation allocation to generating units. However, under deregulation, AGC is allowed to control the contract power of independent power producers (IPPs) rather than output of generation units. If power pool has some generator units to directly control, the AGC should control the output of those generator units. Hence, in such a market place, the AGC simulator will need all the information required in a vertically operated utility industry plus the contract data and measurement. In a regulated market, electrical power network is vertically integrated and a single utility company monopolizes generation, transmission and distribution in a certain geographical region. Interconnection between networks and interconnection between utilities is usually voluntary to improve system reliability and performance . Tariffs however are protected and consumers are limited in choosing the suppliers of their electricity. Under deregulation, the power system structure changes in such a way that allows the evolving of industries that are more specialized for generation (Gencos), transmission (Transcos) and distribution (Discos). In addition, there will be a separate Independent Contract Administrators (ICA). It will generally constitute of the following responsibilities: A system administrator to administer the electricity energy market. System controller (scheduling co-coordinator) as the real time manager of electrical energy network operation. A transmission administrator to ensure that all have open, non-discriminatory access to the grid. The transmission administrator also contracts the services of transmission facility owner for the fulfillment of energy supply purchase contracts.

With the restructuring of power industry and adaption to open access power trading one major technical area that needs to addressed in Automatic Generation Control (AGC). The simulation study has been done for a two area non-reheat thermal turbine system in deregulated power system. A new area load frequency based on fuzzy logic has been proposed. The dynamic response of conventional controller and Hybrid controller is also presented.
Abstract:

Keywords: Automatic Generation Control, deregulation, DPM, Fuzzy logic, optimization, two area non-reheat thermal turbine system

INTRODUCTION An AGC system monitors the frequency and tie line flows, computes the net change in the area generation required. An AGC system monitors the frequency and tie line flows, computes the net change in the area generation required (generally referred to as Area Control Error, ACE) and changes the set point of the generators within the area to keep the time average of ACE to a low value. Automatic generation control [1] is defined as The regulation of power output of electric generators within a prescribed area in response to the changes in the system frequency and or tie-line loading, or the relation of these to each other, so as to maintain the scheduled system frequency and/or the established interchange with other areas within predetermined limits. Thus, aims of conventional AGC [25] are: To achieve zero static frequency error. To distribute generation among areas so that interconnected tie line flows matches the prescribed schedule. Balance total generation against total load and tie line power exchanges. To maintain each units generation at the most economic value. A worldwide trend towards restructuring and deregulation of the power industry has been developed in the last decade. With the process of deregulation, the competition has been introduced in the power industry to enhance its economical efficiency. The basic features [16] of the restructuring are To introduce competition in an otherwise monopolistic industry. In order to achieve this, to separate (vertical unbundling) the functions of power generation, transmission, distribution and electricity supply to consumers.
I.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

207

II. DEREGULATED SYSTEM Different Structures for Providing Real Power Control in a Deregulated Environment There are many possible structures [7,16] for providing real power in a deregulated environment, differentiated by who controls generators, who has the obligation to perform LFC and how free the market is. Three structures are considered progressing from no load frequency control (LFC) to a fully distributed market that covers the range of all possible LFC structures namely Free LFC, Charged LFC and Bilateral LFC. Free LFC is the present situation where a VIU (Transco) issues real power control signals to units it owns to perform LFC, with no specific cash flow tied to this function. IPPs in this environment are either captive to the VIU or contractually obligated to accept control as though they are VIU owned generators or have no obligation to perform LFC. LFC will not be a problem until proportion of VIU controlled units in generating mix decreases to the point where total controlled generator response rate drops to the near the maximum load change rate. Charged LFC is a structure in which Transco no longer owns generation, but has obligation to provide LFC. In this case, Transco must purchase real power from Gencos and resell it to Discos on a short-term real time basis (seconds). This structure can operate with a Transco monopoly on purchase and resale of LFC power with or without bilateral contracts. In Bilateral LFC, there is no central control algorithm. Each disco must purchase load matching (LM) contracts from one or more Gencos. It is possible for such bilateral LM contract to co-exist with the Transco obligation. In this case, Transco is obligated to provide LFC to Disco that have not purchased LM contracts. A key difference between the Bilateral LFC environment and the VIU environment is that the control area in the later are persistent, while the control area in the former lasts only while the contract is in operation. Communications are a significant technical issue for Bilateral LM contracts. Each contract requires a communication channel between Genco and Disco. III DISCO PARTICIPATION MATRIX The concept of a Disco participation matrix (DPM) as introduced by Donde and Pai [10] has been used which helps in visualizing and implementation of the contract between different Discos and Gencos. DPM is a matrix with the number of rows equal to the number of Gencos and the number of columns equal to number of Discos in the system. Each entry in the matrix can be thought of as a fraction of total load power contracted by Disco (column) towards a Genco (row). Thus,ijth element corresponds to fraction of total load power contracted by Disco j from a Genco i. The sum of all the entries in a column in this matrix is unity. The Disco participation matrix (DPM) is defined as

Where, cpfij, i= 1,..m j=1,n refers to contract participation factor of the ith Genco to jth Disco and m refer to the total number of Gencos and n refers to total numbers of Discos. cpfij=1 i Cpf = Fraction of load demand from Genco i by Disco j
ij

Total load demand of Disco j IV MATHEMATICAL MODEL AND BLOCK DIAGRAM The transfer function block diagram [10] of a two-area non-reheat thermal system in deregulated power system is as shown in Fig 1 and its parameters are given in Appendix 1. The block diagram has been arrived at by linearising the nonlinear model of two area thermal system around a nominal operating point. The main features of this AGC in deregulated environment as follows: 1. Computation of Set Points (CPG for Gencos) of Gencos using Discos load demand and contract participation factor. 2. Computation of area control errors. 3. Computation of local demand Pl1,loc and Pl2,loc. The other things are same as in conventional AGC system. The set point is a new input for each Gencos in deregulated power system and is arrived at as per the contract agreement between the different Gencos and Discos. Fig 3.3 illustrates the computation of contracted power demand or set points for the Gencos. It can be seen that the set point of each Genco depends on the load demand of Discos and contract participation factor of the Genco towards that particular Disco. The local loads in area I and II are denoted by Pl1,loc and Pl2,loc respectively. The local loads in the area constitute of both contracted load as well as un-contracted load. Whenever a load demands by any Disco changes, it is reflected as a local load in the area to which the Disco belongs. Coefficients that distribute ACE to several Gencos in an area are defined by area participation factor (apf). m apfj=1 j=1

Fig. 1: Transfer function block diagram of two-area non-reheat thermal system in deregulated power system

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

208

Fig. 2: Computation of set point for Gencos.

V. DYNAMIC PERFORMANCE OF TWO-AREA SYSTEM Three cases have been considered The AGC system shown in Fig 1 is characterized in the state space form as Where X, u and p are state vector, control vector and perturbation vector respectively. A,B,and are state distribution matrix,control distribution matrix perturbation distribution matrix. A. Case 1: Base Case In this case it is considered that the Gencos in each area participate equally in AGC; i.e., ACE participation factors are apf1 = 0.5, apf2 = 0.5, apf3 = 0.5 and apf4 = 0.5. It is assumed that the load change has occurred only in area I. Thus, the load is demanded only by Disco 1 and Disco 2. Value of this load demand is 0.1 pu MW for each of them. DPM becomes

Note that Disco 3 and Disco 4 do not demand power from any Gencos, and hence the corresponding participation factors (Column 3 and 4) are zero. Disco 1 and Disco 2 demand identically from their local Gencos, viz., Genco 1 and Genco 2. In the steady state, generation of a Genco must match the demand of the Discos in contract with it. This desired generation [10] of a Genco in pu MW can be expressed in terms of cpfs and the total demand of Discos as
Pmi
j

cpfij PLj
Lj

Where P is the total demand of Disco j and cpfs are given by DPM. For the case under consideration, P = 0.5 x P + 0.5 x P = 0.1 pu MW
M1 L1 L2

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

209

And similarly, P = 0.1 pu MW, P P


M2 M3 M4

= 0 pu MW, = 0 pu MW.

Dynamic response of frequency deviatioins, generated powers, and tie line powers in base case is given in Fig 3

E: Dynamic response of change in Power Generated by Genco 2

A: Dynamic response of Frequency deviation in area I in base case

F: Dynamic response of change in Power Generated by Genco 3

B:Dynamic response of Frequency deviation in area 2 in base case

G:Dynamic response of change in Power Generated by Genco 4 Fig 3: Dynamic response of frequency deviatioins, generated powers, and tie line powers in base case

C:Dynamic response of Tie-line power 1-2 flow in deregulated environment

The frequency deviation in each area goes to zero in the steady state. As only the Discos in area I have non zero load demands, the transient dip in frequency of area I is larger than that of area II. Since the off diagonal blocks of DPM are zero, i.e., there are no contract of power between a Genco in one area and a Disco in another area, the scheduled steady state power flow over the tie line is zero. The actual power on the tie line goes to zero. B. Case 2: Fully Deregulated Case Considering a case where all the Discos contract with the Gencos for power as per the following DPM:

D:Dynamic response of change in Power Generated by Genco 1

It is assumed that each DISCO demands 0.1 MW power from GENCOs as defined by cpfs in DPM and each GENCO participate in AGC as defined by following apfs: apf1 = 0.75, apf2 = 0.25, apf3 = 0.5, apf4 = 0.5.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

210

The off diagonal blocks of the DPM correspond to the contract of a DISCO in one area with a GENCO in another area. The scheduled power on the tie line in the direction from area I to area II is
2 4 4 2

Ptie1

2 , schedule i 1 j 3

cpfij PLj
i 3 j 1

cpfij PLj

Hence P

tie1-2,scheduled

= -0.05 pu MW. In the steady state,


D:Dynamic response of change in Power Generated by Genco 1

the GENCOs must generate P = 0.105 pu MW P P P


M1 M2 M3 M4

= 0.045 pu MW = 0.195 pu MW = 0.055 pu MW

Dynamic response of frequency deviations, generated powers, and tie line powers in fully deregulated environment in Fig 4
E:Dynamic response of change in Power Generated by Genco 2

A: Dynamic response of Frequency deviation in area I F:Dynamic response of change in Power Generated by Genco 3

B: Dynamic response of Frequency deviation in area II G:Dynamic response of change in Power Generated by Genco 4 Fig. 4: Dynamic response of frequency deviations, generated powers, and tie line powers in fully deregulated environment

C:Dynamic response of Tie-line power 1-2 flow in fully deregulated Environment

C. Case 3: Contract Violation It may happen that a DISCO violates a contract by demanding more power than that specified in the contract. This excess power is not contracted out to any GENCO. This uncontracted power must be supplied by the GENCOs in the same area as the DISCO. It must be reflected as the local load of the area but not as the contract demand. Consider case two again with a modification that DISCO1 demands 0.1 pu MW of excess power. The total local load in area I (P )
L1,LOC

= load of DISCO1 + load of DISCO2

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

211

= (0.1 + 0.1) +0.1 pu MW = 0.3 pu MW Similarly, the total local load in area II(P

D:Dynamic response of change in Power Generated by Genco 1


L2,LOC

= load of DISCO3 + load of DISCO4 = 0.2 pu MW (no uncontracted load) As DPM is the same as in case 2 and the excess load are taken up by GENCOs in the same area, the tie line power is the same as in case 2 in steady state. The generation of GENCOs 3 and 4 is not affected by the excess load of DISCO1. The uncontracted load of DISCO1 is reflected in the generations of GENCO1 and GENCO2. ACE participation factors decide the distribution of uncontracted load in the steady state. Thus, this excess load is taken up by the GENCOs in the same area as that of the DISCO making the uncontracted demand.

E:Dynamic response of change in Power Generated by Genco 2

F:Dynamic response of change in Power Generated by Genco 3

A: Dynamic response of Frequency deviation in area I

G:Dynamic response of change in Power Generated by Genco 4 Fig. 5: Dynamic response of frequency deviations in case of contractviolation B: Dynamic response of Frequency deviation in area II

C:Dynamic response of Tie-line power 1-2 flow

V. FUZZY LOGIC BASED AUTOMATIC GENERATION CONTROLLER The conventional integral controller (PI) has been replaced with fuzzy controller and the dynamic response of the hybrid controller for supplementary control of AGC is studied. Fuzzy logic controller is based on set of linguistic control rules and provides an algorithm, which can convert the linguistic control strategy based on expert knowledge into an automatic generation control strategy. The essential parts of fuzzy logic controller (FLC) are a set of linguistic control rules related by the dual concepts of fuzzy implication and the compositional rule of inference. In essence, then, the FLC provides an algorithm that can convert the linguistic control strategy based on expert knowledge into an automatic control strategy. There are four main blocks in a fuzzy system: the fuzzifier, the inference engine, the knowledge base (KB) and the defuzzifier. The nominal parameters are also same and given in Appendix 1. A particular Disco Participation Matrix ( DPM) has been considered.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

212

The proposed hybrid controller is as shown in Fig 6. The inputs to the proposed fuzzy logic controller (FLC) are ACE and rate change in ACE (ACE). In this study, the control rules and the membership functions are derived in such a way that it exhibits a linear PD control characteristics. Then, a hybrid supplementary controller for LFC is obtained by combining output of the fuzzy controller with proportional of the area control error. The rule base with the minimum number of rules that satisfies the above-mentioned controller characteristic is given in Table 1: NL = Negative Large, N = Negative NZ = Negative Zero Z = Zero, PL = Positive Large P = Positive PZ = Positive Zero

B:Comparison of dynamic response of frequency deviation in area 2 with integral and hybrid controller

C:Comparison of dynamic response of Tie-line power flow 1-2 with integral and hybrid controller Fig. 7: Comparison of dynamic response with conventional controller and hybrid controller Fig. 6: Hybrid Controller TABLE 1: LOOKUP TABLE FOR FUZZY LOGIC

In the proposed controller, the membership functions taken are triangle shaped and only at the boundary of the universe of discourse, trapezoidal membership functions are considered.

VI. CONCLUSION Automatic generation control (AGC) will keep on having an important role in restructured scenario of deregulated power system but with some modifications in the simulation model as contract agreement between the various DISCOS and GENCOS also become a part of it. Performance of proposed hybrid supplementary controller has been compared with conventional integral controller and observed that hybrid supplementary controller responses outperform the integral controller. The settling time and undershoot/overshoot of responses with hybrid controller are small as compared with integral controller responses. REFERENCES [1]. IEEE Recommended Definitions of Terms for Automatic Generation Control on Electric Power Systems. IEEE std94-1991(Revision of IEEE std94-1970). pp. 7-11 [2]. Louis S. VanSlyck, Nasser Jaleeli, W. Robert Kelley. A Comprehensive Shakedown of an Automatic Generation Control Process. IEEE Transactions on Power Systems, Vol. 4, No. 2, May 1989. pp. 771-781 [3]. Nasser Jaleeli, Louis S. VanSlyck, Donald N. Ewart, Lester H.Fink, Arthur G. Hoffman. Understanding Automatic Generation Control. IEEE Transactions on Power Systems, Vol. 7, No. 3, August 1992. pp. 1106-1122. [4]. Robert P. Schulte. An Automatic Generation Control Modification for Present Demands on Interconnected Power Systems. IEEE Transactions on Power Systems, Vol. 11, No. 3, August 1996. pp 1286-1294 [5]. Bjorn H. Bakken, Ove S. Grande. Automatic Generation Control in a Deregulated Power System. IEEE Transactions on Power Systems, Vol. 13, No. 4, November 1998. pp 1401-1406. [6]. M. L. Kothari, N. Sinha, Mohammad Rafi. Automatic Generation Control of an Interconnected Power System Under Deregulated Environment. pp 95-102. [7]. Feng Liu, Yonghua Song, Shengwei Mei, Qiang Lu. Optimal Load Frequency control in Deregulated Power Systems. 2002 IEEE. pp 944-948.

A: Comparison of dynamic response of frequency deviation in area 1 with integral and hybrid controller

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

213

[8]. [9].

[10].

[11].

[12].

[13].

[14].

Dr. Noureddine Bekhouche. Automatic Generation Control Before and After Deregulation. 2002 IEEE. pp 321-323. Ibraheem, Prabhat Kumar, Dwarka P. Kothari. Recent Philosophies of Automatic Generation Control Strategies in Power Systems. IEEE Transactions on Power Systems, VOL. 20, NO. 1, FEBRUARY 2005. pp 346-357. Vaibhav Donde, M. A. Pai, Ian A. Hiskens. Simulation and Optimization in an AGC System after Deregulation. IEEE Transactions on Power Systems, VOL. 16, NO. 3, AUGUST 2001. pp 481-489. Ignacio Egido, Fidel Fernndez-Bernal, Luis Rouco, Eloisa Porras, and ngel Siz-Chicharro. Modeling of Thermal Generating Units for Automatic Generation Control Purposes. IEEE Transactions on Control Systems Technology, VOL. 12, NO. 1, JANUARY 2004. pp 205-210. J. Nanda, A. Mangla. Automatic Generation Control of an Interconnected Hydro-Thermal System Using Conventional Integral and Fuzzy Logic Controller. 2004 IEEE International Conference on Electric Utility Deregulation, Restructuring and Power Technologies (DRPT2004) April 2004 Hong Kong. pp 372377 A.P. Sakis Meliopoulos, George J Cokkinides, A G Bakirtzis. Load Frequency Control Service in a Deregulated Environment. Decision Support System, Vol. 24, Issue 3-4 Jan1999. M. A. Sujan, C. Nwankpa, M. H. Gravener. Towards the Rearl
rd

APPENDIX 1
NOMINAL SYSTEM PARAMETERS FOR TWO EQUAL AREA NON-REHEAT THERMAL SYSTEM IN REGULATED POWER SYSTEM f = 60 Hz. Pr1 = Pr2 = 2000MW. 12 = 1 2 = 30 degrees. R1 = R2 = 2.4 Hz/pu MW. Ptiemax = 200 MW. a12 = -Pr1/Pr2 = -1.0
-3

D1 = D2 = 8.33 x 10 pu MW/Hz. H1 = H2 = 5secs. Tg1 = Tg2 = 0.08secs. Tt1 = Tt2 = 0.3 secs. B1 = B2 = b = 0.425 pu MW/Hz. Kp1 = Kp2 = 120 Hz/pu MW Tp1 = Tp2 = 20 secs. T12 = 0.086 pu MW/rad. C = 2*3.141*T12 = 0.545 pu MW

[15].

[16].

[17].

[18].

[19].

[20].

[21].

[22].

[23].

[24]. [25].

Time Monitoring of AGC. Proceedings of the 33 Hawaii International Conference on System Sciences-2000, pp. 1-5. George K. I. Mann, Bao-Gang Hu, Member, IEEE, and Raymond G. Gosine, Member, IEEE. Analysis of Direct Action Fuzzy PID Controller Structures. IEEE Transactions on Systems, Man, and Cybernetics Part B: Cybernetics, VOL. 29, NO. 3, JUNE 1999. pp 371-388. Richard D. Cristie and Anjan Bose. Load Frequency Control Issue in Power System Operation after Deregulation. IEEE Transactions on Power Systems, Vol. 11, No. 3, August 1996. pp 18-23. J. Kumar, K. Ng, G. Sheble. AGC Simulator for Price Based Operation: Part I. IEEE Transactions on Power Systems, Vol. 12, No. 2, May 1997. pp 527-532. J. Kumar, K. Ng, G. Sheble. AGC Simulator for Price Based Operation: Part II. IEEE Transactions on Power Systems, Vol. 12, No. 2, May 1997. pp 533-538. M. L. Kothari and J. Nanda and D.P. Kothari and D. Das. Discrete Mode Automatic Generation Control of a Two Area Reheat Thermal System with New Area Control Error. IEEE Transactions on Power Systems, Vol. 4. No. 2, May1989. pp 730738. Fernando L. Alvarado, Jienping, C.L. DeMarco and Wellington S Mota. Stability Analysis of Interconnected Power Systems Coupled With Market Dynamics. IEEE Transactions on Power Systems, Vol. 16, No. 4, November2001. pp 695-701. G.A. Chown, RC. Hartman. Design and Experience with a Fuzzy Logic Controller for Automatic Generation Control (AGC). IEEE Transactions on Power Systems, Vol. 13, No. 3, August 1998. pp 965-970. Chuen Chen Lee. Fuzzy Logic in Control Systems: Fuzzy Logic Controllers Part-I. IEEE Transactions on Systems, Man and Cybernetics, Vol. 20, No.2, March/April 1990. pp 404-418. Jawad Talaq and Fadel Al Basri. Adaptive Fuzzy Gain Scheduling For Load Frequency Control. IEEE Transactions on Power Systems, Vol. 14, No. 1, February 1999. pp 145-150. [24] I. J. Nagrath, D. P. Kothari, Power System Engineering, 13th edu, McGraw-Hill, 2003. Allen J. Wood, Bruce F. Wollenberg, Power Generation Operation, and Control, 2nd edn, John Willey & Sons, Inc., 2003.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

214

Parameter Identification of GA Based High Performance Field Oriented Controlled Induction Motor Drive
1
1,2

Dheeraj Joshi, 2P.R. Reshmi, 3RC Bansal

National Institute of Technology, Kurukshetra, Haryana, University of Queensland, Brisbane, QLD 4072 Australia 1 ee.dheeraj@gmail.com Abstract: This paper presents parameter identification of GA voltage with the produced torque or speed of the rotor. based field oriented control of induction motor drive. Field The second part of the modeling is to determine the orientation scheme is used through conventional parameters of the model. All procedures of control proportional integral (PI) controller. GA approach is used to methods need the knowledge of a system mathematical find the optimum controller parameters along with model. Any deviation between the parameters real values Autoregressive Moving Average with Extra Inputs (ARMAX) model parameters for high performance of and those belonging to any electrical apparatus induces the induction motor drive. Simulated results are shown to control performances deterioration. Hence it is indispensible to use a high performance method to identify highlight the performance of induction motor drive. KeywordsARMAX, Genetic Algorithm, induction motor, the IM parameters. It is used to identify the IM parameters field oriented control. values in industrial control. Identification of an unknown Nomenclature dynamic system can be defined as an experimental process of building a mathematical model belonging to a class of = Mechanical speed models presented which reflects precisely its behavior. Speed reference Considering the referred objective and accuracy, the Stator currents candidate model will satisfy in an equivalent manner the Stator voltages process to identify, when submitted to the same constraints leading to the definition of the identification Rotor fluxes method. In other words, identification is the art of creating a = Rotor flux magnitude mathematical description of an unknown dynamic system. Rotor flux magnitude reference The most commonly used optimization methods are based Square rotor flux magnitude on the minimization of the quadratic criteria to obtain the reference best possible set of parameter values. It will be reasonable to confirm that the best parameter values will be those = Square rotor flux magnitude which minimize the quadratic criteria J, often called as cost function and also referred as an index performance. Rotor time constant Nevertheless, it is not always possible to evaluate these p Number of pole pairs derivatives when the real measures are noisy. Rotor inductance and resistance Consequently, we use the GA method which needs only to Stator inductance and resistance compute the function and not its gradient or other M Mutual inductance auxiliary knowledge. J Moment of Inertia Many conventional identification methods were Load Torque . published such as based on the least mean square (LMS) and RLS[ ]. Nowadays, more precisely computing of real I. INTRODUCTION machine parameters is necessary for control purpose. For Induction motors (IM) have long been widely used in this reason, to efficiently identify the IM motor model, many industrial applications. IM can be considered as a when high dynamic performance in control techniques for single input, single output (SISO) system having torque- adjustable-speed drives is required. GA method is speed characteristics compatible with most mechanical preferred [2-6]. loads. This makes a IM controllable over a wide range of In this paper, two different techniques are used for speeds by using appropriate control methods. identification of IM parameters. ARMAX dynamic model Mathematical modeling is one of the most important and including the parameters minimum number are applied often difficult step towards understanding a physical under the same conditions. In the first one, we use the system. most popular RLS algorithm with its various extensions In everyday life, most of the observed phenomena have (ERLS) to identify DC motor parameters. In the second, dynamic components. The aim of IM modeling is to find we apply GA method for the estimation of these the governing differential equation that relates the applied
3

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

215

parameters. The proposed approach is employed in fitting characteristics of the block diagram of IM motor drive using data curves representing angular speed versus supplying voltage and a second order transfer function of the ARMAX model with only four parameters reducing the complexity of resolution and computing time. The conventional control methods of induction motor (IM) have various drawbacks, such as; (i) If mathematical model of the system is not accurate, it is not possible to get accurate results, (ii) Load disturbance, motor saturation and thermal variations affect motor performance adversely, (iii) Classical linear control shows good performance only at one operating speed [1]. Further, to implement conventional control, the model of the controlled system should be known. The usual method of computation of mathematical model of a system is difficult. When there are system parameter variations or environmental disturbance, the behavior of the system is not satisfactory. The field-oriented control (FOC) performance is sensitive to the deviation of motor parameters, particularly the rotor time-constant. Indirect field-oriented techniques are now widely used for the control of induction motor in high-performance applications. With this control strategy the decoupled control of induction motor is guaranteed. Sensorless vector control has received much attention in induction motor drives. The elimination of the speed sensor reduces the hardware complexity, size and cost, and increases the reliability of the drive system [2]. At present, there are two main methods of highperformance control of induction motor drives. One is the direct field-oriented control (DFOC), and the other is indirect field-oriented control (IFOC). The research for improving the IFOC due to the problem of parameter deviation, especially the rotor time-constant or rotor resistance has been developed [3]. On the other hand, the flux angle employed directly in decoupling the torque and flux components is called as the DFOC which has been investigated in many studies. Peresada et al. [4] have presented a theoretical and experimental comparison between standard and recently developed improved indirect field-oriented controllers for induction motors. It is shown that standard indirect field-oriented control algorithm provides asymptotic speed and flux regulation, while the improved one guarantees global exponential speedflux tracking under condition of constant load torque. Effect of system parameters on the performance of field oriented controlled induction motor drive are shown in [5]. Field orientation scheme is used through conventional proportional integral (PI) controller. Usually classical control is used in electrical motor drives. The classical controller designed for high performance increases the complexity of the design and hence the cost. Advanced control system based on artificial intelligence technique is called intelligent control. Genetic algorithm (GA) is a search technique used in computing to find exact or approximate solutions to

optimization and search problems. Genetic algorithms are categorized as global search heuristics. Genetic algorithms are a particular class of evolutionary algorithms (also known as evolutionary computation) that use techniques inspired by evolutionary biology such as inheritance, mutation, selection, and crossover (also called recombination). Genetic Algorithms are search algorithms based on the mechanics of natural selection and natural genetics. They combine survival of the fittest among string structures with a structured yet randomized information exchange to form a search algorithm with some of the innovative flair of human search. In every generation, a new set of artificial creatures ( strings) is created using bits and pieces of the fittest of the old; an occasional new part is tried for good measure. While randomized, genetic algorithms are no simple random walk. They efficiently exploit historical information to speculate on new search points with expected performance [10,11]. Present work shows the parameter identification of field oriented control of high performance induction motor drive using GA based PI controllers. Flux and speed error are processed using conventional PI controller. Simulated results are given to highlight the performance of induction motor drive. II. MODELLING OF FIELD ORIENTED CONTROLLED INDUCTION MOTOR DRIVE [5] The induction machine can be described by five nonlinear differential equations with four electrical variables fluxes i.e. stator currents ( ), rotor

, rotor speed (

) and stator voltages

( ). Here stator voltages are control variables. In a stator-fixed frame (, ) equation can be written as follows; =f(Y,U) = pK pK p +p ) (2) (1)

= = = = where U= Y= with =1 ( /

= ), K=M/( )

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

216

/(

)+(

=p , = . Now field oriented control (FOC) can be explained as follows; The FOC is based on the transformation of (9) in the fixed stator frame (, ) into vectors in a frame (d, q), which rotates along with the flux vector = = With =arc tan and = . It can be written as follows; (3) (4) (5) , then = (10) We obtain the following system with a simpler structure and a linear flux amplitude dynamics: = = = = = = , =0 (6) with the state feedback = Equation (1) becomes (7) = =p + When the flux amplitude (11) = is regulated to its + + Then (7) becomes

Since = from (3) and (4) ,

constant reference value , the rotor speed dynamics is also linear. If we consider the following subsystem: = = (12)

So, the flux amplitude dynamics can be controlled by via a PI controller = pK p = Using the second subsystem = = =p Let the nonlinear state feedback control be (8) = = + (14) d (13)

It is seen that the rotor speed dynamics can be also independently controlled by using a PI-type controller

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

217

+ (15)

III. OUTPUT ERROR BASED METHOD The model parameters identification procedure isillustrated by the block diagram of Fig 1 where the prediction error k between the output process y(k) and y^(k) predicted by the model, is used by a parametric adaptation algorithm( PAA) based on an optimization quadratic criteria, permitting to modify the model parameters at each instant k to minimize the output error.

Extensive RLS (ERLS): the generalized least squares (GLS), the recursive maximum likelihood (RML) and the instrumental variable (IV) methods is presented. The search of optional parameters k is made by non linear programming. It concerns to use an algorithm which from non optimal parameters k and a criterion J provides the parameters k of the model. VI. RECURSIVE LEAST SQUARES ALGORITHM (RLS) It is often desirable to dispose of methods which permit to adjust on line that digital model of a physical process or any signal for the prediction and /or the systems control. These methods are qualified as adaptive. They must be of a high performance concerning rate of convergence, stability, robustness and execution speed since they must be completed within one sample interval. The RLS method is used for adaptive identification to search in real time a digital model of any physical system of an industrial process (DC motors, synchronous and induction machines) The principle of the method consists on minimizing the quadratic criterion J corresponding to the squared error (Mean Square Error: MSE) at the present instant, between the model output and the process (for example the DC Motor) to determine the best parameters. In identification, the aim of a recursive algorithm is to find the new estimate of parameters k^ from the past one (k-1) without making all calculus. VII. THE ARMAX MODEL Considering a noisy digital model to simulate a real process very useful in industrial control with single input u(k)- single output y(k) and a delay d described by its linear digital recurrent different equation:

Fig.1: Output error based identification principle

IV. OPTIMIZATION QUADRATIC CRITERION The least mean squares method has been introduced by Karl Gauss in 1809. It has been the one of all identification methods and parameters estimation. This method is based on the minimization of a quadratic function J. It is the LMS method which permits to find the parameters optimal values ^ of linear prediction which minimizes the quantity J. The mean squared error J << MSE>> is the prediction distance based on the difference between the output of the real system and that predicted by the model in a same instant k:

(16) N: number of measurement points k: prediction error committed on the estimation. It concerns to find the parameters vector ^ minimizing this function J of the error which is a non linear function of the last ones. The prediction made using the model depends on the form of this one (AR, MA, ARMA, ARMAX, NARMAX). The model is a function of n parameters k. k varying from 1 to n. The question is then to determine the parameters k so that the criterion J will be minimal. These algorithms based on the parametric estimation are numerous, here are the most used: i) LMS. ii) RLS. iii)Recursive Prediction Error Method (RPEM) In the following sections, the equation error method using the basis RLS algorithm with its various extensions called

Application of the Z-Transform to (2) leads to the output of a system for which the model << process+ perturbation>> is ARMAX- type described by: (17) w (k) represents the effect of the noise in the output because a system is submitted to perturbation with the result that we can never perfectly calculate the output knowing the input. W (k) could be due to measurement noise of non controlled input. In addition w (k) has the same dynamic specified by the denominator A (z) of the system TF. Hence in the ARMAX model as in the ARX one, the model and the perturbation e (k) have the same dynamics represented by Fig .2

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

218

(25) VIII PROPOSED METHODOLOGY As GA is used for identification purpose, following objective function is formed; OF=perf+error; Where Perf=dev1*delta_t+dev2*delta_t dev1=abs(Psi_ref-Psi); dev2=abs(wm_ref-wm); and error=(wk-wm)^2 wk=a1*w(k-1)+a2*w(k-2)+b1*Psi(k-1)+b2*Psi(k-2) PI controller parameters are determined using GA along with IM parameters (a1,a2,b1,b2) such that peak overshoot, settling time and steady-state error in speed and flux will be minimum. PI controller send appropriate action for high performance induction motor drive. According to some stopping criterion, the tuning is complete and the IM model is identified along with controller parameters. IX RESULTS AND DISCUSSIONS This section shows simulation results of FOC IM drive. Fig1(a) and (b) shows the speed response of high performance FOC IM drive. It is seen that implementation of GA based PI controller gives very less values of settling time, rise time, peak overshoot and steady-state error under loading and unloading condition. Steady state error remains almost zero in such operations. From Fig1 (a) it is seen that constant load torque is applied at t=3 sec Tl=40 Nm, it is increased to 140Nm and 180 Nm at t=6 and t=9 sec respectively. Further Fig 1(b) shows that earlier no load is there upto 3 secs. Then Tl=140 Nm is applied at t=3 sec. Afterwards at t=6 secs, load torque is reduced as 40Nm and at 9 sec machine is unloaded.

Fig. 2:.Model structure with equation error ARMAX.

A(z) and B(z) are the polynomials of respectively degree n and m so as m<n to be realized. The (5) is given by : A(z) y(k)=Z-dB)z)u(k)+w(k) (18) Hence this ARMAX model contains AR term A(z)y(k) a control term B(z)u(k) of moving average MA. The output signal y (k) of this system is composed of a determinist part attached to the excitation input u (k) and a stochastic part attached part attached to perturbation w(k) in (19) below (GLS)

(19) Knowing that the perturbation is not reproducible (unpredictable); it cannot be described only in independent random manner specified by two parameters: Its mean and its variance 2 of the effectuated measurements. The output of a system for which the model << process+ perturbation>> ARMAX- type will be: (20) Where e(k) is often taken to be Gaussian discrete white noise ( RML0 with zero mean =0 and variance 2 =constant. Practically, these model with output error filtering observations represents the 2/3 of the cases, hence the most of the applications. Using the Z-polynomials, the equation (17) will be written as a model of the form: (21) With the z-polynomials:

(22) (17) can also be written as: (23) (24.a) (a)

Where = [a1 ..ana, b1..bnb]T: is the parameters vector.

(24.b) is the measurement vector . If the model actual parameters are known so as , We can define the equation of mean square predictor: (b)
Fig. 3: Estimated and measured speed under different operating conditions

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

219

X. CONCLUSIONS Identification of GA based field oriented control of induction motor drive has been presented in this paper. GA is used to tune the PI controller parameters for high performance of induction motor drive. Simulated results are shown to highlight the performance of induction motor drive and effectiveness of the proposed methodology. The proposed approach can be useful for control of FOCIM drive.
TABLE I: VALUES OF CONTROLLER PARAMETER

[10] D. E. Goldberg, Genetic Algorithms in Search, Optimization and


Machine Learning, Pearson Education, 2006.

[11] D. Joshi, A. K. Sharma, RC Bansal, GA-fuzzy based high


performance field oriented controlled induction motor drive, Vol.2, no.2, pp. 85-92,July-December 2010.

TABLE II: DRIVE PARAMETERS

REFERENCES [1] V. Chitra, and R. S. Prabhakar, Induction Motor Speed Control [2]
using Fuzzy Logic Controller, World Academy of Science, Engineering and Technology, vol. 23, 2006, pp-17-22. A. Mezouar, M.K. Fellah, and S. Hadjeri, Adaptive slidingmode-observer for sensorless induction motor drive using twotime-scale approach, Science Direct, Simulation Modelling Practice and Theory, vol. 16, 2008, pp.13231336. K.K. Shyu, F.J. Lin, H.J. Shieh, and B.S. Juang, Robust Variable Structure Speed Control for Induction Motor Drive, IEEE Trans. on Aerospace and Electronic Systems, vol. 35, no. 1, 1999. S. Peresada, A. Tilli, and A. Tonielli, Theoretical and Experimental Comparison of Indirect Field-Oriented Controllers for Induction Motors, IEEE Transactions on Power Electronics, vol. 18, no. 1, 2003, pp.151-163. D. Joshi, A. K. Sharma, K.S. Sandhu, I. Musirin, Effect of Drive Parameters on Field Oriented Controlled Induction Motor Drive, PEOCO 2010, 23-24 June 2010, Malaysia. V. Chitra and R. S. Prabhakar, Induction Motor Speed Control using Fuzzy Logic Controller, World Academy of Science, Engineering and Technology 23 2006,pp-17-22 R. Kumar, R.A. Gupta, and S.V. Bhangale, Indirect Vector Controlled Induction Motor Drive with Fuzzy Logic Based Intelligent Controller, IETECH Journal of Electrical Analysis, vol. 2, No. 4, 211-216, 2008 I. Birou, V. Maier, S. Pavel, and C. Rusu, Indirect Vector Control of an Induction Motor with fuzzy Logic Based Speed Controller, Advances in Electrical and Computer Engineering, vol. 10. No.1, 2010,pp.116-120 S.N. Sivanandam, S. Sumathi, and S.N. Deepa, Introduction to Fuzzy Logic Using MATLAB , Springer, 2007.

[3] [4]

[5] [6] [7]

[8]

[9]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

220

Hourly Load Forecasting of A Sub-station Feeder Using Neural Network with Back Propagation Learning
1

Manvendra Singh , 2Balwinder Singh , 3Rintu Khanna


1, 2, 3 3

PEC University Of Technology, Chandigarh rintukhanna1@rediffmail.com load with reasonable degree of accuracy.The time range, ahead of which the forecast is required, has to be viewed from the functional areas of planning, operation and management. Depending on the time range, there are three types of forecasting, e.g. Long term (a few months to a few years), medium term (one week to few months) and short term forecasting (a few minutes to a few hours). It can also accuracy. The time range, ahead of which the forecast is required, has to be viewed from the functional areas of planning, operation and management. Depending on the time range, there are three types of forecasting, e.g. Long term (a few months to a few years), medium term (one week to few months) and short term forecasting (a few minutes to a few hours). It can also be classified depending on the specific need and applicability .Long term load forecasting is mainly concern with the generation expansion planning, transmission and distribution planning, financial planning, energy exchange policy between organizations, planning for peaking capacity and maintenance of plants. Medium term forecasting refers to economic scheduling of various energy sources, inventory control of coal and liquid fuels, reservoir utilization and water management for irrigation. It also helps in maintaining security constraints and proper planning of load shedding. Short term load forecasting helps in load management with on-line dynamic voltage control, load flow studies and exchange of power as requirement for load frequency control. Selection of forecasting methods is guided by the Following factors: 1. Accuracy of forecast. This is a major criterion in selecting a model for forecasting and affects the operation cost of a system. 2. Data. Pattern, type and length of the data decide the nature of model. 3. Cost of computation. This is very important for shortterm load forecasting. 4. Ease of applicability. User must feel at ease to handle the model. 5. Interactive facility. Forecast procedure should be normally automatic and provision must be kept for intervention through external control. 6. Constant monitoring facility. for adjustments during abnormal load behaviour.

Abstract Load forecasting plays an important role in power system operation as it serves to predict the system expected power demands. It thus plays an important role in economic & financial development, expansion & planning. With Short Term Load forecasting we can make a lot of decisions like operation planning, optimum generation schedule, load scheduling etc. In this paper, the application of neural networks to design & implement Short-Term Load Forecasting (STLF) model is developed for predicting the load on a substation feeder in Agra, India. One important architecture of neural networks namely Multi-Layer Perceptron (MLP) model with back propagation learning was used to develop this forecasting model. This neural network architecture has 3 layers: an input, a hidden, and an output layer. The input to the model is the historical hourly load current data of the feeder. The number of days for which network is being trained/tested are chosen depending on the type of Forecasting model to be developed. Our forecasting model was trained and tested using three month hourly data i.e. June, August, September 2010 to predict the Month Ahead Hourly Load, Day Ahead Hourly Load & Weekdays Hourly Load on the feeder. The results show that MLP with supervised back propagation learning can be considered as a good method to model the STLF systems.

I. INTRODUCTION The increase in demand of electrical energy has drawn the attention of power system engineers towards the reliable operation of power systems. For reliable operation of integrated power supply systems, a close tracking of electrical load is required. For the economy of operations, this must be accomplished over a broad spectrum of time intervals. While for a short range of seconds or even minutes, the automatic generation control function involving economic dispatch is used to ensure the matching of the load with economic allocation among the committed generation sources, the security of supply still depends on the availability of hot and cold reserves which in turn depend on the total load demand at any time. Specially for periods of hours and day where wider variation of loads occur, meeting the demand entails the start up and shut down of the entire generating unit or interchange of the power with neighboring systems. For the preparation of the maintenance schedule of the different units and auxiliaries, it is desirable to know prior to the demand profile of important nodes of the system for wider length of time. All these necessitate an accurate forecasting of the

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

221

7. Risk due to load forecasting uncertainty: Operational risk independent of the load forecasting such as leadtime II. ARTIFICIAL NEURAL NETWORK ANNs are mathematical tools originally inspired by the way, the human brain processes information. They are composed of simple elements operating in parallel. These elements are stimulated by biological uneasy systems. As in nature, the network function is determined largely by the connections stimulated by biological uneasy systems. As in nature, the network function is determined largely by the connections between elements. You can train an ANN to perform a particular function by adjusting the values of the connections (weights) between elements. Commonly ANNs are adjusted, or trained, so that a particular input leads to specific target output. Therefore, the network is adjusted, based on a comparison of the output and the target, until the network output matches the target (Fig I)[4,5].

Fig .2: Two Layer Feed Forward Perceptron Network

Fig. 1: Artificial Neuron Structure

Typically many such input/target pairs are needed to train a network. Their basic unit is the artificial neuron. The neuron receives (numerical) information through a number of input nodes, processes it internally, and puts out a response. The processing is usually done in two stages: first, the input values are linearly combined, and then the result is used as the argument of a nonlinear activation function. The combination uses the weights attributed to each connection, and a constant bias term. The neurons are organized in a way that defines the network structure. The most concerned structure is the multilayer perceptron (MLP) type, in which the neurons are organized in layers. The neurons in each layer may share the same inputs, but are not connected to each other. If the architecture is feed-forward, the outputs of one layer are used as the inputs to the following layer. The layers between the input neurons and the output layer are called the hidden layer. Input layer Hidden layer Output Layer

Fig II shows an example of a two-layers feed-forward perceptron network with four input neuron, three neurons in the hidden layer and one neuron in the output layer. Each layer has a specified number of nodes; the interconnections are only between neurons of adjacent layers, and each neuron belonging to a layer is connected to all the neurons of adjacent layers. The parameters of this network are the weight matrix, and the bias. The estimation of the parameters is called the training of the network. The most used training algorithm in load forecasting is back-propagation one.[8]There are many variations of the back-propagation algorithm. The simplest implementation of back-propagation learning updates the network weights and biases in the direction in which the performance function decreases most rapidly, the negative of the gradient. In load forecasting applications, this basic form of multilayer feed-forward architecture is still the most popular. Nevertheless, there are a large number of other designs, which might be suitable for other applications. ANN should prove to be particularly useful when one has a large amount of data as here we have in case of Short Term Load Forecasting III. MODEL DEVELOPMENT The model development using artificial neural network consists of the following important considerations as shown in FigIII: a. Data preparation b. Selection of neural network structure c. Selection of proper training algorithm

Fig. 3: Block diagram for model development using ANN

The data preparation consists of selection of input and output variables, collection of accurate and sufficient training data, and decision of proper normalization

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

222

range for input, output data. A.Input Vector ConFiguration The input vector to the ANN is the previous days hourly load data which was collected from a 132/33 KV substation from Agra for the month of June 2010, in case of Month Ahead Hourly Load Forecasting & from the month of August & September,2010 in the case of Weekdays Hourly & Day Ahead Hourly Load Forecasting. In case of Month Ahead Hourly Load Forecasting, first twenty days hourly load data, from the month of June, 2010 was used in Training Phase & last twenty days hourly load data was used in Testing Phase. These number of input days will decide the number of input neurons in the ANN network, which is 20 in this case. In Day Ahead Hourly Load Forecasting, sample set of load data from a week of June 2010 was formed, of which, first four days data of the week was used in the Training Phase & fourth to seventh day data was used in the Testing phase, which predicts the day ahead hourly load on the feeder. The number of input neuron to the network, in this case, depending on the number of input is 4 .In the case of Weekdays Hourly Load Forecasting,, a set of past ten days hourly load data of the particular day is formed, for which load forecasting is to be done .From this sample set, first four to five days hourly data was used in Training Phase & last four to five days hourly data was used in Testing Phase. So the number of input neuron to the network, in this case are either 4 or 5. The number of hidden neuron are decided on hit & trial basis which gives best results when their number is 3. B.Data Pre-Processing The data which we have collected from the substation has wide variation s from 110A to 220A value, which needs to be normalized before being entered to the ANN for Training & Testing. So, the formula which was used to normalize the load data was Coeff= 0.8/(Xmax-Xmin) Y= 0.1 + (Xj Xmin)*Coeff ..(2)

Xmax = Maximum Value of X in the ith column of the input load data Xmin = Minimum Value of X in the ith column of the input load data Xj = Maximum Value of X in jth column C.Training Phase Back propagation algorithm with automated regularization and fixed learning rate is used in this phase. The learning rate is multiplied times the negative of the gradient to determine the changes to the weights and biases..Hence, the learning rate is kept small so that the network gats sufficient time to learn the input data. In this work the learning rate is fixed to 0.00001, which show a good performance. Similarly, other factors which affects the performance of the network are Momentum Factor, Error

Tolerance. To increase the speed of back propagation learning algorithms, momentum factor is considered. The error tolerance affects the training time and generalization capabilities of ANN. So, these factors needs to be adjusted to improve the performance of the network. The main problem in the training is called over-fitting. The error of training becomes very small, but the network will memorize the input data. The network will not be generalized for new test data, and will show a large error when it is tested. To improve the generalization of the network, an automated regularization technique is developed in this model[9].Typically, the objective of the training process is to reduce the sum of squared error which is given as: 2 Ed= ...........(3) Where tl is the target value, and al is the network output & u is the number of patterns. Sample set of load data is being formed, depending on the type of forecasting to be done (i.e. Month Ahead Hourly Forecasting, Day Ahead Hourly Forecasting or Weekdays Hourly Load Forecasting) & data is being fetched from this set for the Training Phase. The objective of training the network is to adjust the weights so that application of a set of inputs produces the desired set of outputs. In Training Phase, Back Propagation Algorithm is used to train the network with historical load data. Back Propagation algorithm is a supervised learning scheme & it incorporates an external teacher and/or global information and includes such techniques as error-correction learning, reinforcement learning and stochastic learning. The training or learning algorithm, which is commonly used for feed-forward ANN, is gradient descent back-propagation algorithm. It is a popular learning algorithm for multi-layered ANN mainly because of its computation simplicity, ease in implementation and good results generally obtained for large number of problems in many different areas of application. The gradient descent back-propagation algorithm adjusts the connection weights between different neurons in proportion to the difference between the desired and computed values of each layer in ANN structure. Addition of momentum term improves the stability of the process. The weight adjustment equation of back propagation algorithm is given by Wnew = Wold+W, Where ,

The learning rate and momentum have very significant effect on learning speed of the back propagation algorithm. Large value of results in faster convergence but subjects to network oscillations. Where a small value of stabilizes the process but results in slower convergence. Similarly for higher values of momentum

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

223

coefficient connection weights are updated in correct direction and improve the convergence. The performance of gradient descent back propagation algorithm also depends on error function used. The sumsquared error function is the most popular error function used in back propagation learning algorithm because of its computation simplicity, it is being used here for the shortterm load forecasting problem. The Training Phase can be summarized as: Step 1. Select the training pair from the training set; apply the input vector to the network input. Step 2. Calculate the output of the network. Step 3. Calculate the error between the network output and the desired output (the target vector from the training pair). Step 4. Adjust the weights of the network in a way that minimizes error. Step 5. Repeat steps 1 through 4 for each vector in the training set until the error for the entire set is acceptably low. D.Testing Phase After the Training Phase, comes the Testing Phase in which the performance of the trained network is being analyzed. In this phase, few old data which was used in the Training Phase & new data from the sample set are being used to predict the load in all the three cases i.e. Month Ahead Hourly Load Forecasting, Day Ahead Hourly Load Forecasting & Weekday Ahead Hourly Load Forecasting. The RMS Error is being calculated by comparing the hourly load data obtained after Testing Phase with the Target data, which is the desired output, in each forecasting model. IV. RESULTS The performance of the developed model for Short-Term Load Forecasting has been trained & tested with previous one month hourly load data ,in the case of Month Ahead Hourly Load Forecasting, with previous seven days hourly load data, in Day Ahead Hourly Load Forecasting & with set of load data of previous days load data for a particular weekday, in Weekday Hourly Load Forecasting. The load forecast was compared to the target/desired load data and the error is calculated, after Training & Testing Phase, in each forecasting model. The Root Mean Square (RMS), which was calculated in both Training &Testing Phase to evaluate the performance of these models, is defined as: RMS Error = sqrt((Ydesired-Ytrained)2) in training phase RMS Error = sqrt((Ydesired-Ytested)2) in testing phase ..(4) Where Ydesired is the target output in both the phases & Ytrained & Ytested is the Trained & Tested Output of the network respectively. The Training & Testing Phase results for Month Ahead Hourly Load Forecasting, Day Ahead Hourly Load

Forecasting & Weekdays Hourly Load Forecasting are as follows: (I) MONTH AHEAD HOURLY FORECAST RESULTS
Training 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 Actual Train Data

10

15

20

25

Fig. 1: Training Phase


Testing 1 Desired Output Test Data 0.9

0.8

0.7
Testing Output of Load Data

0.6

0.5

0.4

0.3

0.2

0.1

10 Hours

15

20

25

Fig. 2: Testing Phase

(II). DAY AHEAD HOURLY FORECAST RESULTS


Training 0.9 Desired Output Train Data 0.8

0.7
Training Output of Load Data

0.6

0.5

0.4

0.3

0.2

0.1

10 Hours

15

20

25

Fig. 3: Training Phase


Testing 0.9 Desired Output Test Data

0.8

0.7
Testing Output of Load Data

0.6

0.5

0.4

0.3

0.2

0.1

10 Hours

15

20

25

Fig. 4: Testing Phase

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

224

(III) WEEKDAYS HOURLY FORECAST RESULTS a.Monday


Training 0.9 Desired Output Train Data
Training Output of Load Data

c.Wednesday
Training 0.9 Desired Output Train Data 0.8

0.8

0.7

0.7
Training Output of Load Data

0.6

0.6

0.5

0.5

0.4

0.4

0.3
0.3

0.2
0.2

0.1
0.1 0 5 10 Hours 15 20 25

10 Hours

15

20

25

Fig. 5: Training Phase


Testing 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1
Testing Output of Load Data

Fig. 9: Training Phase


Testing 0.9
Desired Output Test Data

0.8

Desired Output Test Data

Testing Output of Load Data

0.7

0.6

0.5

0.4

0.3
0 5 10 Hours 15 20 25

0.2

Fig. 6: Testing Phase

0.1

b.Tuesday
Training 0.9 Desired Output Train Data 0.8

10 Hours

15

20

25

Fig. 10: Testing Phase

0.7
Training Output of Load Data

d.Thursday
Training 0.9 Desired Output Train Data 0.8

0.6

0.5

0.4
Training Output of Load Data

0.7

0.3

0.6

0.2

0.5

0.1

10 Hours

15

20

25

0.4

Fig. 7: Training Phase


Testing 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 Desired Output Test Data

0.3

0.2

0.1

10 Hours

15

20

25

Testing Output of Load Data

Fig. 11: Training Phase

10 Hours

15

20

25

Fig. 8: Testing Phase

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

225

Testing 0.9 Desired Output Test Data

Testing 0.9 Desired Output Test Data

0.8

0.8

0.7
Testing Output of Load Data

0.7
0.6
Testing Output of Load Data

0.6

0.5

0.5

0.4

0.4

0.3

0.2

0.3

0.1

0.2
0 5 10 Hours 15 20 25

0.1

Fig. 12: Testing Phase

10 Hours

15

20

25

e.Friday
Training 0.9 Desired Output Train Data
0.9

Fig. 16: Testing Phase

g.Sunday
Training Desired Output Train Data 0.8

0.8

0.7
Training Output of Load Data

0.6
Training Output of Load Data

0.7

0.5

0.6

0.4

0.5

0.3

0.4

0.2

0.3

0.1

0.2

10 Hours

15

20

25
0.1 0 5 10 Hours 15 20 25

Fig. 13: Training Phase


Testing 0.9 0.8 0.7
Testing Output of Load Data

Fig. 17: Training phase


Desired Output Test Data

Testing 0.9 Desired Output Test Data

0.6 0.5 0.4 0.3 0.2 0.1 0


Testing Output of Load Data

0.8

0.7

0.6

0.5

0.4

10 Hours

15

20

25

0.3

Fig. 14: Testing Phase

0.2

0.1

f.Saturday
Training 0.9 Desired Output Train Data 0.8

10 Hours

15

20

25

Fig. 18: Testing Phase

0.7
Training Output of Load Data

0.6

The RMS Error in Month Ahead Hourly Load Forecasting, Day Ahead Hourly Forecasting & Weekdays Hourly Load Forecasting can be tabulated as: TABLE I: MONTH AHEAD HOURLY LOAD FORECAST RESULTS RMS ERROR(in %) Training Phase Testing Phase 3.11 11.70 TABLE II: DAY AHEAD HOURLY LOAD FORECAST RESULTS
0 5 10 Hours 15 20 25

0.5

0.4

0.3

0.2

0.1

Fig. 15: Training Phase

RMS ERROR(in %) Training Phase Testing Phase 2.67 11.04

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

226

TABLE III: WEEKDAYS HOURLY LOAD FORECAST RESULTS

RMS ERROR (in %) Day of The Week Training Phase Monday 2.07 Tuesday 2.38 Wednesday 1.86 Thursday 1.89 Friday 3.20 Saturday 2.55 Sunday 2.68

Testing Phase 6.64 8.17 8.16 8.19 10.10 3.10 4.46

[14] [15]

[16]

[17]

V. CONCLUSIONS The results which have obtained for all the three model developed i.e. Month ahead hourly forecast, Day ahead hourly load forecast & Weekdays hourly load forecast confirm the applicability as well as the efficiency of neural networks in short-term load forecasting. The neural network was able to find out the nonlinear relationship that exists between the historical load data used in the training phase and on the basis of that make the prediction of load in the testing phase, in all the three model developed. REFERENCES
[1] Osofisan PB,Nawaeke CN, Application Of artificial neural Network To STLF Journal Of Engg & Applied Sciences 5(2) 7883,2010 Alfares H K & M Nazeeruddin, Electric Load Forecasting: Literarture Survey & Classification Of Methods,Int.J.Syst. Sci.,33:23-34,2002 Mohsen Hayati, Yazdan Shirvany, Artificial Neural Network Approach for Short Term Load Forecasting for Illam Region,World Academy of Science, Engineering and Technology 2007 Pradeepta Kumar Sarangi, Nanhay Singh, R. K. Chauhan and Raghuraj Singh, Short Term Load Forecasting Using Artificial Neural Network: A Comparison With Genetic Algorithm Implementation, ARPN Journal of Engineering and Applied Sciences, vol. 4, no. 9, November 2009 Nahi Kandil,Ren Wamkeue, Maarouf Saad and Semaan Georges, An Efficient Approach For Short Term Load Forecasting Using Artificial Neural Networks, International Journal of Electrical Power & Energy Systems, Volume 28, Issue 8, 2006. J. M. Zurada, Introduction To Artificial Neural Systems, West Publishing Company, Saint Paul, Minnesota, 1992. S. Haykin, Neural networks: A comprehensive foundation (A comprehensive book with an engineering perspective), Macmillan Publishing, New York, 1994. S. Haykin, Neural networks, A comprehensive foundation, Prentice-Hall, New Jersey, 1999 Hagan, M.T., H.B. Demuth, and M.H. Beale, Neural Network Design,Boston, MA: PWS Publishing, 1996. Henrique Steinherz Hippert, Carlos Eduardo Pedreira, and Reinaldo Castro Souza, Neural Networks for Short-Term Load Forecasting:A Review and Evaluation, IEEE transactions on power systems, vol. 16, no. 1, february 2001 Kwang Y. Lee, Tae-I1 Choi, Chao-Chee Ku, and June Ho Park, Neural Network Architectures for Short-Term Load Forecasting, IEEE Transactions On Power Systems Vol 9,March 1994 Pauli Murto, Neural Network Models For Short-Term Load Forecasting, Helsinki, January 5, 1998 A.D. Papalexopoulos, Application Of Neural Network Technology To Short-Term System Load Forecasting, pp. 796800, Proceedings, IEEE Second International Forum on the

[18]

[19]

[20]

Applications of Neural Networks to Power Systems, April, 1993, Yokohoma, Japan Y. Rui, A.A. El-Keib, A Review of ANN-based Short-Term Load Forecasting Models,IEEE Transactions On Power System,2001 G.A. Adepoju, Ogunjuyigbe, K.O.Alawode, Application of Neural Network to Load Forecasting in Nigerian Electrical Power System, The Pacific Journal of Science and Technology, Volume 8. Number 1. May 2007 (Spring) wenjin Dai, ping Wang, Application of Pattern Recognition and Artificial Neural Network to Load Forecasting in Electric Power System, Third International Conference on Natural Computation (ICNC 2007) Alex. D. Papalexopoulos, Shangyou Hao, Tie-Mao Peng, ShortTerm System Load Forecasting Using An Artificial Neural Work, IEEE Transactions On Power System,1993 Nahi Kandil, Rene Wamkeue, Maarouf Saad, Semaan Georges,An Efficient Approach For Shorterm Load Forecasting Using Artificial Neural Networks, IEEE ISIE 2006, July 9-12, 2006, Montreal, Quebec, Canada R. Sadownik and E.P. Barbosa, "Short-term forecasting of industrial electricity consumption in Brazil", Journal of Forecasting, Vol. 18,pp.215-224, 1999. D.W. Bunn and E.D. Farmer, "Comparative Models for Electrical Load Forecasting", John Wiley & Sons, New York, USA, 1985.

[2]

[3]

[4]

[5]

[6] [7]

[8] [9] [10]

[11]

[12] [13]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

227

Design of Functional Observer for Servomechanism System


1
1, 2
1

Sudhir, 2B.B. Sharma


NIT Hamirpur HP, India
2

sudhir.nadda87@gmail.com,

bharat.nit@gmail.com

Abstract - This paper presents a functional observer for servomechanism system, i.e., an observer which estimates directly linear functions of the state of a linear system. The order of these observers is equal to the dimension r of the vector to be estimated. Necessary and sufficient conditions for the existence and stability of these observers are given. For applying the results, linearized dynamics model of servomechanism system is used. Numerical simulations are presented in the end, to analyze the performance of functional observer. Keywords: Servomechanisms, Functional observer, existence condition.

system is described in Section IV. In Section V concluding remarks are presented. II. PROBLEM FORMULATION In servomechanism controller the most important output variable is the angular position of DC motor. Assumed that angle of DC motor is measured and torque is unmeasured output. The dynamics of the system are (1) (2) (3) (4) where V the applied voltage, T is the torque acting on the load, is the loads angular velocity, is the motor shafts angular velocity, is the loads angular position, is the motor shafts angular position, and is the torsional rigidity and motor constant, and the motor and nominal load inertia, and the motor viscous friction coefficient and load viscous friction coefficient, the gear ratio and R the armature resistance. Defining the state variables as , the linear time invariant (LTI) state space form is represented as:

I. INTRODUCTION Estimating linear functional of state vector has been the focus of research since last few decades. In some applications it is sufficient to estimate directly linear functions of the state instead of the full state vector. One well-known example for that is the state feedback design for linear systems in which it is sufficient to estimate directly the state feedback u = kx in order to stabilize a linear system [11]. The design of observers which estimate linear functions of the state, in short functional observers, were studied in [8]. Later on, a couple of other approaches have been developed to design linear functional observers, see e.g. [1], [2], [9], [11], [12]. As the entire state vector cannot be measured, in most of practical complex systems, the control law cannot be implemented. In such cases the approach which directly accounts for the non availability of the entire state vector must be devised or a suitable approximation to the state vector must be determined. In most of such situation the approach based on developing and using an approximate state vector is vastly simpler than a direct attack on the design problem. Alterative design procedures for observing a scalar linear function of the state of a multiple output system are minimal partial realization theory and decision method, respectively. In this paper, a functional observer is proposed which estimates directly linear functions of the state of a linear system i.e. servomechanism system. In [1], [2] necessary and sufficient conditions for the existence of such functional observer is summarized and presented in the subsequent section of the paper. Numerical simulations presented in the end justify the effectiveness of such observers. The remainder of the paper is organized as follows: The problem formulation is described in Section II. In Section III, the functional observer design procedure is summarized, presented in [1], [2]. The numerical example of functional observer for servomechanism

(5)

and output is (6) where and (7) The only measurement available for feedback is . The loads angular position must be set to a desired value by adjusting the applied voltage V. The elastic shaft has finite shear strength, so the torque T must stay within specified

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

228

limits. From an input/output viewpoint, the plant has a single input V, which is manipulated by the controller. It has two outputs, one measured and gives feedback to the controller and one unmeasured T. Parameters of experimental position servomechanism system are shown in table:
TABLE 4.1 PARAMETERS OF SERVOMECHANISM CONTROLLER

r where is the estimate of z, N, J, H and E are constant matrix of appropriate dimensions to be designed. The following theorem presents asymptotic functional observer design condition: Theorem 1: The state in (11) is an asymptotic estimate of z (t) for any x(0), z(0) and u(t) iff (12)

Value Symbol Value (SI units) (SI units) 1280.2 20 10 0.1 0.5 25 50 JM 20 In controller design problem, controller must set the loads angular position at a given value. The elastic shaft has a finite shear strength, so the torque T, must stay within specified limits . Also the applied voltage must stay within the range . However, here goal is to develop a functional observer for the system, which estimates the linear function of states directly. Using the values of various parameters in equation (5), following state-space model is obtained:

Symbol

(13) N is Hurwitz, i.e., has all its eigenvalues in the left-hand side of the complex plane, where P=L-EC. Proof: The observer reconstruction error is

Then, the dynamics of this estimation error are given by (14) Equation (12) is equivalent to (15) Since is of full row rank, where denotes the MoorePenrose generalized inverse of matrix L. Using the definition of matrix P, (15) leads to (16) and (17) where Theorem 2: The necessary and sufficient conditions for the existence and stability of functional observer (11) for the system (10) are: (i) There exists an rth-order observer (11) for the system (10) if and only if (18) (ii) The matrix N is Hurwitz (or the pair (G; F) is detectable) if and only if

(8) (9) Equation (8) and (9) represent the state-space model of servomechanism system. In section III, procedure for design of functional observer for the above system is discussed. III. DESIGN OF FUNCTIONAL OBSERVER FOR SERVOMECHANISM SYSTEM In this section, the procedure of designing functional observer is summarized. Consider the linear time-invariant multivariable system described by (10-a) (10-b) (10-c) n p where x and y are the state vector, the output m vector of the system, u is the input vector of the r system, and z is the vector to be estimated where r n. A, C and L are known constant matrices of appropriate dimensions. We assume, without loss of generality that rank{C} = p and rank{L} = r. Note that the dynamics of servomechanism system given in (8)-(9) belongs to (10-a) and (10-b). Our aim is to design an observer of the form (11-a) (11-b)

(19) For a more detailed discussion of Theorem 1 and Theorem 2, see [1]. The design of the matrices N, J, H, and E of the functional observer is given in [1]. First the matrices (20) and (21) are calculated whereas, e.g., where denotes the Moore Penrose generalized inverse of matrix L, i.e. . Furthermore, the matrix Q is defined as . Then the matrix N is calculated by any pole placement procedure for the pair (F, G), i.e.,

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

229

(22) where Z is the matrix obtained from the pole placement of the pair (F, G). Note that the pole placement is possible since the pair (F, G) is detectable due to the condition (ii) of Theorem 2. Using the matrix N, one obtains the matrices J and E from the equation (23) Finally, the matrix H is obtained by (24) A. Design of the functional observer:With the above results and under conditions (18) and (19), we can propose the following design method. 1) Matrices F and G can be obtained from (20), (21). 2) The matrix gain Z can be obtained by any pole placement method (since the pair (G; F) is detectable). Then we obtain N given by (22). 3) From the obtained values of Z and N, we can deduce the matrices E and J from (23). 4) Matrix H is given by (24). IV. NUMARICAL EXAMPLE State-space matrices of given servomechanism system are

(a) 1 1 0
state

(b)

state

0.5

-1 -2 -3

-0.5

10

20 30 40 t(time in seconds) (c)

50

10

20 30 40 t(time in seconds) (d)

50

2.5

6 4
state x
4

state

2 0 -2

1.5

10

20 30 40 t(time in seconds)

50

10

20 30 40 t(time in seconds)

50

Fig. 1 (a)-(d): variation of servomechanism system states with time


3.2 3 2.8 z state zhstate

2.6
z & z h state

2.4 2.2

From the results of Section III, we obtain F = [-5.8142e-013] G = [-0.0000 0.0000 0.0476 -0.0000]T Z = [0 0 62.9907 0] N = [-3.0000] E = [20.9969 -0.0156] J = [-1.0e-010 *(-0.1529 0.0001)] The estimate of z is given by the following observer:

2 1.8 1.6

10

15

20 25 30 t(time in seconds)

35

40

45

50

Fig. 2: Functional observer performance in estimating z state

Fig1 represents the behaviour of four states servomechanism system and the estimate of z state represent in the Fig2.

V. CONCLUSION In this paper a functional observer for servomechanism system has been presented, i.e. an observer which estimates directly linear functions of the state of a linear system. This method reduces the design procedure to one of full-order system. The existence and stability conditions are given and the procedure is used to develop a functional observer for servomechanism system with linear function to be estimated is taken as combination of loads angular position and . REFERENCES

[1]

M. Darouach. Existence and design of functional observers for linear systems. IEEE Transactions on Automatic Control, 45:940943, 2000.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

230

[2]

[3]

R. Engel and G. Kreisselmeier. A continous-time observer which converges in finite time. IEEE Transactions on Automatic Control, 47:12021204, 2002. Tobias Raff, Patrick Menold, Christian Ebenbauer, Frank Allgower. Finite Time Functional Observer For Linear System. 44th IEEE
Conference on Decision and Control, and the European Control Conference 2005 Seville, Spain, December 12-15, 2005.

[4] [5] [6]

[7]

[8] [9] [10]

[11]

[12]

C. T. Chen, Linear System Theory and Design. New York, NY: Holt, Rinehart and Winston, 1984. C. Rao and S. Mitra, Generalized Inverse of Matrices and Its Applications. New York, NY: Wiley, 1971. Katsuhisa Furuta and Tomomi Suzuki. Design of Functional Observer in Transfer Function Form. American Control Conference San Diego, California June 1999. Prakash K. Nakade and Girish G. Galgate. Design of Linear Functional Observer for MIMO LTI systems. International Journal of Computer Applications (0975 8887) Volume 1 No. 6 2010. D.G. Luenberger. Observers for multivariable systems. IEEE Transactions on Automatic Control, 11:190197, 1966. P. Murdoch. Observer design for a linear functional of the state vector. IEEE Transactions on Automatic Control, 18:308310, 1973. R. Engel and G. Kreisselmeier. A continous-time observer which converges in finite time. IEEE Transactions on Automatic Control, 47:12021204, 2002. T.E. Fortmann and D. Williamson. Design of low-order observers for linear feedback control laws. IEEE Transactions on Automatic Control, 17:301308, 1972. J.R. Roman and T.E. Bullock. Design of minimal order stable observers for linear functions of the state via realization theory. IEEE Transactions on Automatic Control, 20:613622, 1975.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

231

Damping of Low Frequency Oscillations using Fuzzy Logic Controlled based Unified Power Flow Controller
1

Rintu Khanna, 2Pooja Manrai, 3Rajan Manrai


1, 2, 3

PEC University of Technology Nowadays, FACTS devices can be used to control the power flow and enhance system stability. They are playing an increasing and major role in the operation and control of power systems. The UPFC (Unified Power Flow Controller) is the most versatile and powerful FACTS device [1]. The parameters in the transmission line, i.e. line impedance, terminal voltages, and voltage angle can be independently controlled by UPFC. It is used for independent control of real and reactive power in transmission lines. Moreover, the UPFC can be used for voltage support and damping of electromechanical oscillations [2~4]. Damping of electromechanical oscillations due to sudden change in input mechanical power, faults etc. is necessary for secure system operation [5]. In this paper, researches are based on single machine system with UPFC. A well-designed UPFC controller can not only increase the transmission capability but also improve the power system stability. A series of approaches have been made in developing damping control strategy for UPFC. The coordination between FACTS controllers and other power system controllers is very important and is presented in this paper by Fuzzy-coordination. The fuzzy logic controllers are rule-based controllers in which a set of rules represents a control decision mechanism to adjust the effect of certain cases coming from power system. Furthermore, fuzzy logic controllers do not require a mathematical model of the system [10]. This paper focuses on the optimization of conventional power oscillation damping (POD) controllers and fuzzy logic coordination of them. By using fuzzy- coordination controller, the coordination objectives of the FACTS devices are quite well achieved. Few researchers have presented a linearised HeffronPhillips model of a power system installed with UPFC in order to design suitable controllers for power flow, voltage and damping controls. Wang [6~8] has addressed the basic issues pertaining to the design of damping controller using UPFC. He has not presented application of fuzzy logic technique for designing UPFC based damping controllers. To present a systematic approach for designing proportional fuzzy logic based UPFC damping controller using UPFC shunt converter (exciter) phase angle ( E) To examine the relative effectiveness of UPFC damping controller ( E)for damping power system oscillations.

Abstract: The paper presents a new control method of damping low frequency power system oscillations using fuzzy logic based Unified Power Flow Controller (UPFC) Phillips-Herffron model of a single-machine power system equipped with a UPFC is used to model the system. UPFC controller based upon phase angle of shunt converter (exciter) has been designed. The effectiveness of E Proportional UPFC controller & fuzzy logic (using mamdani-type inference) based UPFC controller (PF-UPFC) has been demonstrated at variable loading conditions.. Respective models have been developed and simulated in Matlab/Simulink. The results of these studies show that the designed controller has an excellent capability in damping power system oscillations. Keywords: UPFC; Fuzzy Logic; Damping controller; Low frequency oscillations. NOTATIONS Cdc : dc link capacitance D : damping constant H : inertia constant (M = 2H) Ka : AVR gain Kdc : gain of damping controller mB : modulation index of series converter mE : modulation index of shunt converter Pe : electrical power of the generator Pm : mechanical power input to the generator T1,T2 : time constants of phase compensator Ta : time constant of AVR Tdo : d-axis open circuit time-constant of generator Vb : infinite bus voltage Vdc : voltage at dc link Vt : terminal voltage of the generator XB : reactance of boosting transformer (BT) XBv : reactance of transmission line Xd : direct axis steady-state synchronous reactance of generator XE : reactance of excitation transformer (ET) Xe : equivalent reactance of the system Xq : quadrature axis steady-state synchronous reactance of generator XtE : reactance of transformer : direct axis transient synchronous reactance of Xd generator : phase angle of series converter voltage B : phase angle of shunt converter voltage E : natural frequency of oscillation (rad/sec) n I.INTRODUCTION

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

232

To compare the performance of UPFC based damping controllers: 1. Without fuzzy logic technique 2. With proportional fuzzy logic technique To investigate the performance of this damping controller, following wide variations in loading conditions and system parameters II. SYSTEM UNDER STUDY The system used in the investigations of this paper (Fig1) consists of a synchronous generator which is connected via two transformers to an infinite bus system through a transmission line. A UPFC is installed in the midpoint of the transmission line [7]. The system dynamic model is used to represent and study the dynamic performance of the single-machine infinite-bus power system. The UPFC considered here is assumed to be based on pulse width modulation (PWM) converters. The nominal loading condition and system parameters are given in Appendix 1.

Fig. 2: A UPFC installed in a single-machine infinite-bus

The non-linear differential equations from which the well known Phillips-Heffron linear model of a singlemachine infinite-bus power system is derived are:
o o o

( Pm
o /

Pe

) / 2H

Eq
o /

( Eq

/ Eqe ) / Tdo

E qe
Fig. 1: Single Machine Infinite Bus Power System

reg ( s )(vto vt )

III. THE UNIFIED PHILLIPS-HEFFRON MODEL OF THE POWER SYSTEM INSTALLED WITH UPFC Unified power flow controller (UPFC) is a combination of static synchronous compensator (STATCOM) and a static synchronous series compensator (SSSC) which are coupled via a common dc link to allow bi-directional flow of real power between the series output terminals of the SSSC and the shunt output terminals of the STATCOM and are controlled to provide concurrent real and reactive series line compensation without an external electric energy source [9]. The UPFC, by means of angularly unconstrained series voltage injection, is able to control, concurrently or selectively, the transmission line voltage, impedance and angle or alternatively, the real and reactive power flow in the line. The UPFC may also provide independently controllable shunt reactive compensation. By controlling E, the dc voltage of dc link can be regulated. Fig2 shows a single-machine infinite-bus power system equipped with a UPFC [7]. mB, mE and B E are the amplitude modulation ratio and phase angle of the control signal of each voltage source converter respectively, which are the input control signals of the UPFC.

(1) A linearized model of the power system can be used in studying power system oscillation stability and control [6~7]. For the study of power system oscillation stability, the resistance and transient of the transformers of the UPFC can be ignored. The dynamic equations of the UPFC can be written as:
vEtd vEtq 0 XE X E iEd 0 iEq mE vdc cos( E ) 2 mE vdc sin( E ) 2 mB vdc cos( B ) 2 mB vdc sin( B ) 2
E

vBtd vBtq dvdc dt

0 XB

X B iBd 0 iBq

3mE [cos( 4Cdc

) sin(

)]

iEd iEq

3mB [cos( 4Cdc

) sin(

)]

iBd iBq

(2) By combining and linearizing Equations (2) UPFC and single machine infinite bus power system (1), the state variable equations of the power system equipped with the UPFC can be represented by equation (3). Fig3 shows the small perturbation transfer function block diagram of a machine-infinite bus system including UPFC relating the pertinent variables of electric torque, speed, angle terminal voltage, field voltage, flux linkages [7]. This linear model has been developed by linearising the nonlinear differential equations around a nominal operating point.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

233

0
o o

0 K2 M K3 Tdo / K A K6 TA
0
e

0 0 1 Tdo / 1 TA
0 Kp M Kq Tdo
b

0 K pd M K qd Tdo / K A K vd TA Vdc

Eq
o

K1 M K4 Tdo / K AKS TA
0 K Pe M K qe Tdo
/

D M 0 0

Eq

Eqe

Eqe

logic UPFC (PF-UPFC) controller structure. The membership functions of the input and output signals are shown in Fig5.The rules used in this controller are chosen as: If is P then E is P If is N then E is N

0 Kp M Kq e Tdo
/

K pb M K qb Tdo
e /

mE
E

b /

mB
B b

K A K ve TA

K A Kv TA

K A K vb TA

K A Kv TA

(3) where mE mB are the deviation of input E B control signals of the UPFC. Block diagram in Figurfe 3 can be used in small signal stability investigations of the power system. The MATLAB Simulink toolbox can be used to study the system performance under different disturbances. In Fig3

Fig. 4: Proportional Fuzzy Logic Controller (PF-UPFC)

f Kp

[ vdc K pd M
Kqd Tdo /
K A K vd TA

mE K pe M
Kqe Tdo /
K A K ve TA

mB K pb M
Kqb Tdo /
e

Kp e M
Kq Tdo
e /

Kp M
Kq Tdo

T b /
T b

Kq
Kv

K A Kv TA

K A K vb TA

K AKv TA

(4)
Fig. 5: Input & Output membership function

V. DATA & RESULTS The K - constants of the model computed for nominal operating condition of Single Machine Infinite Bus System equipped with UPFC are:
TABLE: K-CONSTANTS FOR UPFC CONTROLLER E

Fig. 3: Heffron-Phillips model of SMIB system with UPFC

IV. FUZZY LOGIC DESIGN Simple fuzzy logic controller based on mamdani type fuzzy logic controller is used in this section to damp power system oscillations in the study system. The proportional fuzzy logic controller uses angular speed as stabilizing signals. Fig4 shows the Proportional fuzzy

KConstant K1 K2 K3 K4 K5 K6 Kp e Kq e Kv e

Pe=0.8 Qe=0.167 1.1253 0.29611 0.46154 0.20755 -0.01055 0.49647 0.85772 -0.0705 0.07583

Pe=0.5 Q=0.08 1.1967 0.68794 0.46154 0.48591 -0.03328 0.48066 0.80469 -0.0705 .06364

Pe=0.2 Q=0.01 1.2612 0.99949 0.46154 0.70984 -0.06708 0.45822 0.77241 -0.0705 0.0419

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

234

Dynamic responses: Fig6 shows the comparison of dynamic responses obtained for 0.01 step change in mechanical power input Pm to SMIB. These dynamic responses are obtained using Matlab Simulink 6.0 for different values of K obtained at variable loading conditions mentioned above.

2. The performance of the UPFC damping controller E has been examined considering wide variation in the loading conditions. 3. Investigations reveal that the UPFC damping controller E provides robust performance to wide variation in loading conditions. The performance of controller improves as the loading increases. It may thus be recommended that the Fuzzy Logic controlled damping controller based on UPFC control parameters mE may be preferred damping low frequency power system oscillations. REFERENCES
A Nabavi-Niaki and M R Iravani., Steady-state and Dynamic Models of Unified Power Flow Controller (UPFC) for Power System Studies, IEEE Transactionson Power Systems, vol 11, no 4, November 1996, p 1937. [2]. K S Smith, L Ran and J Penman, Dynamic Modelling of a Unified Power Flow Controller, IEE Proceedings-C, vol 144, no 1, January 1997, p 7. [3]. T Makombe and N Jenkins. Investigation of a Unified Power Flow Controller, IEE Proceedings-C, vol 146, no 4, July 1999, p 400. [4]. Papic and P Zunko et al. Basic Control of Unified Power Flow Controller, IEEE Transaction on Power Systems, vol 12, no 4, November 1997, p 1734. [5]. Y Morioka and Y Nakach, et al. Implementation of Unified Power Flow Controller and Verification for Transmission Capability Improvement, IEEETransactions on Power Systems, vol 14, no 2, May 1999, p 575. [6]. H F Wang. Damping Function of Unified Power Flow Controller, IEE Proceedings-C, vol 146, no 1, January 1999, p 81. [7]. F. Wang, F. J. Swift, A Unified Model for the Analysis of FACTS Devices in Damping Power System Oscillations Part I: Single-machine Infinite-bus Power Systems, IEEE Transactionson Power Delivery, Vol. 12, No. 2, April, 1997, pp. 941-946. [8]. F. Wang, F. J. Swift, A Unified Model for the Analysis of FACTS Devices in Damping Power System Oscillations Part II: Multi-machine Power Systems, IEEE Transactions on Power Delivery, Vol. 13, No. 4, October, 1998, pp. 1355-1362. [9]. NarainG.Hingorani,andLaszloGyugyi,UNDERSTANDING FACTS. IEEE Press, New York, 2000. [10]. Hao Ying, Fuzzy Control and Modeling: Analytical Foundations and Applications, IEEE Press Series on Biomedical Engineering, Series Editor: Metin Akay, New York, 2000. [1]. Appendix-1 The nominal parameters and the operating condition of the system are given below. Power System data : Tdo = 5.044 s Xd = 1.0 pu Xq = 0.6 pu X = 0.3 d pu M = 2H = 8.0 MJ/MVA D = 0.0 Ka = 100 Ta = 0.01 s Operating condition : Vt = 1.0 pu Vb = 1.0 pu f = 50 Hz

Fig. 6: Deviation in angular frequency at variable loading conditions for step change in Pm

VI. CONCLUSION The significant contributions of the research work presented in this paper are as follows: 1. A systematic approach for designing proportional fuzzy logic - UPFC based damping controllers for damping power system oscillations has been presented.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

235

Small Hydro-Power-Development:An Overview and a Few Case Studies of Himalayan Region.


1
1

Sharma P.P., 2Chatterji S.


2

Directorate of Technical Education, Himachal Pradesh, Sundernagar India. 2 NITTTR, Chandigarh India.
1

ppsharma65@yahoo.com,

chatterjis@yahoo.com

Abstract --As per present power crisis small micro-hydro, non conventional plants may be planned to work during the peak demand. Integrated generation and distribution for rural area on fuel availability, small hydro capacity on run of the river, shall improve the availability and reduce energy cost. Small hydro development in India and incentives to investors for setting up small hydro-power (SHP) has been discussed. Advantages, Constraints and Measures to Reduce the Costs of SHP are also presented. This paper presents six case studies of micro hydro plants in rural area of India, Himachal Pradesh having different site conditions as on high and medium heads. The reliability of Power output from the mini/micro hydel power station is found out by river flow and rainfall estimation. Keywords -- Small hydro-power plant, regulation, automatic control, distribution networks, generators, micro-hydro power, head and discharge.

reliability, quality and reduce costs. The focus of the SHP generation is to lower the cost of equipment, increase its reliability and set up projects in areas which can provide the maximum advantages in terms of capacity utilization. II..CLASSIFICATION OF SMALL HYDRO POWER PROJECTS Hydro power projects are generally categorized in two segments: small and large hydro. Different countries are following different norms keeping the upper limit of small hydro ranging from 5 to 50 MW. In India, hydro projects up to 25 MW station capacity have been categorized as Small Hydro Power (SHP) projects. These are further classified as given in Table 1:
TABLE I: CLASSIFICATION OF SMALL HYDRO POWER PROJECTS

I. INTRODUCTION Hydropower is a renewable, non-polluting and environmentally benign source of energy. It is perhaps the oldest renewable energy technique known to the mankind for mechanical energy conversion as well as electricity generation. Hydropower represents use of water resources towards inflation free energy due to absence of fuel cost with matured technology characterized by highest prime moving efficiency and spectacular operational flexibility. Hydropower contributes around 22% of about 7,50,000 MW total electricity generated worldwide. It is the main source of power generation in many countries e.g. Norway 99%, Brazil- 86%, Switzerland 76% and Sweden 50%. Total installed power generating capacity in India is of 1,52,360 MW, which includes 39,314 MW from hydro as on 30.09.2009. Out of 39,314 MW hydro capacity, 2429 MW is being supplied from small hydro plants. It has also been emphasized that out of the total grid interactive power generation capacity that is being installed, 2% should come from small hydro-power-plants. The estimated potential of small hydro in the country is of about 15,000 MW. The Indian SHP development programme received a new dimension and tempo after the liberalization of economy and invitation to private sector for investment in power. Today the scenario is such that the SHP generation has emerged essentially as a private investment driven programme. Electricity generation from small hydro is becoming increasingly competitive with preferential tariffs and some other concessions. The challenge is to improve

Class Capacity in kW Micro Up to 100 Mini 101 to 2000 Small 2001 to 25000 In India, the Ministry of Non-Conventional Energy Sources has been assigned the business of SHP up to this capacity. III. SMALL HYDRO POWER POTENTIAL The World estimated potential of small hydro is of around 180,000 MW. India has as an estimated potential of about 15,000 MW with perennial flow rivers, streams and a large irrigation canal network with dams and barrages. Of this, 4,404 potential sites with an aggregate capacity of 10,477 MW have been identified. A comprehensive resource assessment for all the renewable energy sources including small hydro and mapping of potential sites/locations on a GIS platform is receiving utmost attention. The aim is to map renewable energy potential in the country and bring it on a GIS platform with information necessary to take investment decisions to set up projects. Models have been developed that takes into account the regional flow duration curves, geological and seismological data, vegetation cover etc. for identification of potential sites. Software packages have been developed incorporating regional hydrological models to enable users to rapidly estimate the hydro-power potential and other salient features of potential sites. IV. DEVELOPMENT OF SMALL HYDRO POWER IN INDIA India has a century old history of hydropower and the beginning was from small hydro. The first hydro power

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

236

plant was of 130 kW set up in Darjeeling during 1897 which marked the development of hydropower in the country. With the advancement of technology, and increasing requirement of electricity, the thrust of electricity generation was shifted to large size hydro and thermal power stations. However, during the last 10-15 years there is a renewed interest in the development of small hydro power projects due to its benefits particularly concerning environment and ability to produce power in remote areas. Small hydro projects are economically viable and have relatively short gestation period. The major constraints associated with large hydro projects are usually not encountered in small hydro projects. For the 7-8% growth rate that we aspire for, our energy needs will increase correspondingly. The economic growth calls for increasing use of energy and the challenge is to provide desired quality power in a sustainable manner and at reasonable cost. There is a need to tap all possible sources of energy to meet this challenge and small hydro is considered as a reliable option. The Ministrys aim is that out of the total grid interactive power generation capacity that is being installed, 2% should come from small hydro. This translates to 2000 MW capacity addition during the period 2002-2012. The Indian SHP development programme received a new dimension and tempo after the liberalization of economy and invitation to private sector for investment in power. Today the SHP programme is essentially private investment driven. Electricity generation from small hydro is becoming increasingly competitive with 3 preferential tariffs and some other concessions. The challenge is to improve reliability, quality and reduce costs. The focus of the programme is to lower the cost of equipment, increase its reliability and set up projects in areas which give the maximum advantage in terms of capacity utilisation. The installed capacity of SHP projects as on 31.12.2010 is 2850 MW. The growth of capacity addition from small hydro projects during last few years is on an average of 100 MW per year. A capacity addition of 600 MW during the period 2002-2007 is being envisaged through a mix of public and private sector projects, of this about 400 MW has already been achieved during the period 2002-2006. Further a target of adding about 1400 MW from small hydro power projects during the period 2007-2012 is fixed. Bar Chart for SHP

3000 2500 2000 1500 1000 500 0

9 08 -0 20

02 -0

03 -0

04 -0

05 -0

06 -0

07 -0

20

20

20

20

20

20

Fig. 1: Bar Chart for SHP growth in India

V. POTENTIAL OF SHP IN HIMALAYAN REGION OF INDIA Himalayan region in India stretches from Jammu & Kashmir, Himachal Pradesh, Uttaranchal to the NorthEastern States. Almost the entire potential to set up small hydro power projects is of run of river type and there are innumerous small and big streams where such projects can be set up. The mountainous topography and snow covered hills are the source of many rivers with large number of tributary systems having vast hydro electric potential. The region also has a characteristic of small settlements along the tributaries where it is very difficult to extend conventional grid. In this scenario small hydro power projects have unique relevance. Most of the Himalayan regions states are now developing their hydro potential through public and private sector participation. Till last five years or so most of the hydro power projects were set up in the public sector and State Electricity Boards were responsible for such initiatives. However, in last few years it has been realized that private sector participation is inevitable to harness this potential at a faster pace. The State of Himachal Pradesh has taken lead in this direction and over 400 projects have been offered for private sector implementation. Similarly, Uttaranchal, J&K and some North-Eastern states are now encouraging private sector participation. The Himalayan region has also utilized very small streams by setting up hundreds of micro hydel projects to provide electricity to remote villages and hamlets. A number of NGOs have been active in promoting such applications. Uttaranchal, Arunachal Pradesh and Nagaland have gained reasonably good experience in setting up community based micro hydel projects. In J&K, 1000 micro hydel projects are being set up by the Indian Army on boarder areas to electrify remote villages.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

237

20

09 -1

VI. ADVANTAGES, CONSTRAINTS AND MEASURES FOR REDUCING THE COST OF SMALL-HYDRO-POWER
STATIONS

6.1 Advantages of Small Hydro i). It is benign source of power generation, harnessing only gravitational potential of water to make it yield energy in a continuum. ii). Proven technology available in the country with very high prime moving and generation efficiencies for their development. These are least dependent on imports as indigenous technology is available. Setting up small hydro does not require any special geological contribution/ ground conditions. iii). Short gestation period. iv). SHPs are environmentally friendlier than conventional hydro plants due to: a) Non-involvement of setting up of large dams and thus not associated with problems of deforestation, submergence or rehabilitation. b) Non- polluting & environmentally benign. It is one of the least CO2 emission responsible power sources, even by considering full energy chain right from the impact of production of plant equipment etc. c) Least impact on Flora and fauna (aquatic and terrestrial) and bio-diversity. v). It serves to enhance economic development and living standards especially in remote areas with limited or no electricity at all. vi). It can be tapped wherever water flows along small streams, medium to small rivers, irrigation dam-toe/ canal drop sites etc. vii). Low investment is required which can easily be affordable by private entrepreneurs. These projects also suit to private entrepreneur due to short gestation period, quicker returns and cheapest operating costs due to low overheads. viii). It facilitates rural mass who have been able to manage to switchover from firewood for cooking to electricity and thereby checking to deforestation. ix). Small hydro is significant for off-grid, rural, remote area applications in far flung isolated communities having no chances of grid extension for years to come. Small Hydro is operationally flexible, suitable for peaking support to the local grid as well as for stand-alone applications in isolated remote areas. x). With the development of small hydro, rural communities have been able to attract new industries, mostly related to agriculture owing to their ability to draw power from SHP stations and ultimately resulting in the development in the area. xi). Small hydro does not require much expertise to build and operate. Components of small hydro projects are so simple and fairly visible at site distinctively that they can become centre of education. xii). Capital investment is less in compared to others scheme such as thermal as well as big hydro.

xiii). On the basis of project life cycle cost in real terms, inflation- free small hydro becomes several times cheaper than thermal option due to cheaper operational cost and zero cost input. xiv). Under Kyoto Protocol, the Small Hydro Project can earn extra revenue through Clean Development Mechanism (CDM). xv). Hydropower optimizes the performance of other energy technologies and supports the development of other renewable sources. 6.2 Constraints Affecting SHP Development Despite the fact that various incentives are available from IREDA, UN and MNES for development of the small hydro power schemes, the momentum for their development is less due to following reasons: Low load factor and revenues Majority of small hydro projects are located in remote places and not connected with the grid and in general stand alone power stations. Therefore, transmission of the surplus power to other places is not possible. Accordingly, they can fulfill the need of local area only. Incase, the demand is less, then the power station will continue to run at the reduced load i.e at low load factor and thereby loss in power generation which in turn result in poor revenue collection. Operation and Maintenance Cost Inaccessibility of the small hydro plants due to poor transport and communication, the adequate support of operation and maintenance for this power station is not available. Another factor due to which operation & maintenance is very high is that design of such power stations is based on inadequate hydrological and geological data which lead to frequent damages to the structures. Insufficient management for the operation Due to remoteness, the technical and management skill of local is not sufficient for operation of the power plants. Therefore, lot of training is to be provided to the managerial and operational staff. Inadequate Quality and performance The quality of small power stations is not as good as of bigger power stations as these power plants are generally designed on the basis of short term raw data. Thus the ground conditions of operation are much different from the conditions taken for design. Due to such differences the quality and performance of equipment becomes a constraint. The tariff The methodology adopted for tariff in case of small hydro is the same as in Mega Hydro, which is not reasonable. Low interest of Private Developers Private developers avoid to have a stake in small hydro because they are firstly afraid whether they will be able to generate electricity commensurate with the predicted Hydrology. And secondly, once commissioned there is no surety of a buyer at a rate that is comparable to the outcome of the investment he has put.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

238

Statutory clearances Various clearances like forest clearances, defense clearances, environmental clearances; land acquisition etc. given by different ministries takes one to two years time. A single window statutory clearance should be provided or State Govt. should get these clearances beforehand and then only entrusted these projects for execution. Transmission lines The major impediment to majority of SHP stations is nonavailability of high voltage transmission lines resulting in heavy line losses wherever the load centers were spaced far apart. Financing Small Hydro IREDA (Indian renewable Energy Development Agency), World Bank, Asian Development Bank and Japan Bank for international cooperation etc. finance small hydro projects. Govt. encourages private sectors for their participation in development of small hydro power. 6.3 Measures for Reducing Cost of SHP Units Following are some measures proposed to reduce the cost i) The number of units should be limited to 2-3 to save the cost of Civil Engineering, cost of buildings, cost of Hydro-Mechanical equipment, duplication of control system, cabling etc. ii) Instead of going for full Kaplan one can use semi Kaplan where guide vanes are fixed and variations in water is obtained from runner blade operation. So simpler version of water level controllers can be used. iii) There is ample scope for reducing Civil Engineering cost by minimizing the dimensions of Power House, control panels, panels for instrumentation and protection can be accommodated in mezzanine floor to minimize the span of Power House. Layout can be simplified and the entire generating station in semiunderground arrangement may be provided. iv) Sizing should be prudently decided with subsystems to reduce to a minimum to make the scheme economical. v) Induction Generator are economical as compared to Synchronous Generators with associated subsystems. vi) Manufacturers of Hydro-Mechanical equipments should continue standardizing the equipments. vii) Elimination of guide vanes and adoption of electronic load controller devices may reduce the cost. VII. CASE STUDIES REGARDING SHP SCOPE IN HIMACHAL
PRADESH

on Dehar Khad by Astha Projects(India) Limited . The detailed description of this small hydro power plant is given in table 2:
TABLE II: DESCRIPTION OF DEHAR KHAD SMALL HYDRO POWER PLANT

Capacity Type

Head Type of Turbine Rated output Flow Speed Type of Generator Rated Output Rated Voltage Rated Current Power factor PLC make

5MW (2*2.5MW) Grid connected (33KV grid of Himachal Pradesh State Electricity Board at Samot, District Chamba) 300 meters (net) Horizontal 2-jet Pelton 2632 KW 1 Cubic meter per second 600 RPM Hydro Turbine driven Synchronous Generator 3125 KVA 6600 Volts 273.36 Amperes 0.8 (lag) Allen Bradley

7.2 Him Kailash Small Hydro Power Plant. This power plant was commissioned in the year 2008 on Saal Khad i.e. a small rivulet by Him Kailash Hydro Power(P) Limited. It is situated at Paliur (8 KM from Chamba town) in District Chamba, Himachal Pradesh. Description of this small hydro power plant is tabulated in table 3:
TABLE III. DESCRIPTION OF HIM KAILASH SMALL HYDRO POWER PLANT

Capacity Type

Head Type of Turbine Rated output Flow Type of Generator Rated Output Rated Voltage Rated Current Power factor Type of Governor

Himachal Pradesh is one of the small and hilly states in India, which has good potential for mini/micro hydel power development. Due to its hilly terrain natural head is available with every rivulet/stream flowing in the state. The case studies of six mini/micro hydro-power-plants are appended below: 7.1 Dehar Khad Small Hydro Power Plant Dehar Khad Small Hydro Power Plant is located at village Bithal, P. O Samot, Tehsil Bhattiyat Distt. Chamba, Himachal Pradesh. It was commissioned in the year 2005

PLC make Type of Control Make

5MW (2*2.5MW) Grid connected (33KV grid of Himachal Pradesh State Electricity Board at Balloo Chamba) 160 meter Horizontal Francis Turbine 2500KW 1.94cubic meter per second Hydro Turbine driven Synchronous Generator 3125KVA 6600 Volts 273.36 Amperes 0.8 (lag) PLC Controlled Hydraulic based speed governor with pneumatic(Oxygen gas) back up pressure Hitech Guide vanes based Control Kunming Electrical Machinery Co. Ltd. China

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

239

7.3 HIMURJA Micro Hydel Power Plant at Juthed Juthed Micro Hydel Power Plant is located at Juthed (Tarwahi), 80 KM from District Headquarter Chamba near Tissa, District Chamba, H.P. It was commissioned in the year 2000 on a Tributary of Satnala rivulet by HIMURJA (Himachal Pradesh Govt. Agency). The detailed description of this small hydro power plant is presented in table 4:
TABLE IV: DESCRIPTION OF HIMURJA MICRO HYDEL POWER PLANT AT JUTHED

Table VI: Description of Ginni Global Small Hydro Power Plant

Capacity Type

100 Kilo watt 11KV Local Grid connected 55 meter Hydro Turbine driven Synchronous Generator Rated Output 150 KVA Rated Voltage 415 Volts Rated Current 208 Amperes Power factor 0.8 (lag) Speed 1000 RPM No. of Phases Three The Speed governing unit of this plant comprises of PLC Controlled, 24 Volts DC Servomotor based speed control.Ballast load used in this plant consists of 120KW rating Heaters which are immersed in continuously flowing water. Main inlet valve is on/off type operated by 0.37KW, 3-phase induction motor. 7.4 Chaba Power Plant Chaba power house is located on Shima Mandi Road near Naldehra. It is one of the oldest hydro power plant of Himachal Pradesh. This project was commissioned in 1913 by Colonel Betty, the then Chief Engineer of Punjab to meet the Power demand of Shimla, the Summer Capital of India during British rule. The detailed description of this small hydro power plant is given in table 5:
TABLE V: DESCRIPTION OF CHABA POWER PLANT

Capacity Type Head Type of Generator

Head Type of Turbine Rated output Speed Type of Generator Rated Output Rated Voltage Rated Current

5MW (2*2.5MW) Grid connected (33KV grid of Himachal Pradesh State Electricity Board at Tissa, District Chamba, H.P) 184 meter Horizontal Francis Turbine 2618 KW 1000 RPM Hydro Turbine driven Synchronous Generator 3125KVA 6600 Volts 273.36 Amperes

7.6 Aleo Small Hydro Power Project Aleo Small Hydro Power Project is a run of river scheme near Manali town in Kullu district of Himachal Pradesh. It was commissioned in the year 2005 on Allain Khad by Aleo Manali Hydropower P Ltd (AMHPL). The detailed description of this small hydro power plant is given in table 7:
TABLE VII: DESCRIPTION OF ALEO SMALL HYDRO POWER PROJECT

Capacity Type

Head Type of Turbine Rated output Flow Speed Type of Generator Rated Output Rated Voltage

Capacity Head Type of Turbine Speed Type of Generator Rated Voltage No. of Penstocks Length of Penstock Type of Excitation

1750 Kilo watt (2x500 + 3x250 = 1750 KW ) 165 meters (net) Horizontal Pelton wheel 500 RPM Hydro Turbine driven Synchronous Generator 2200 Volts 5 432 m Electro-mechanical

3MW (2*1.5MW) Grid connected (33KV grid of Himachal Pradesh State Electricity Board at Prini, District Kullu) 290.4 meters (net) Horizontal pelton turbines 1500 KW (+20% continuous overload) 1.33 cumec 750 RPM Hydro Turbine driven brushless synchronous generators 1500 KW (+20% continuous overload) each 3300 Volts REFERENCES

7.5 Ginni Global Small Hydro Power Plant It is located at Traila (85 KM from District Headquarter Chamba) Near Tissa District Chamba, Himachal Pradesh. It was commissioned in the year 2006 on a Tributary of Baira nala by Ginni Global (P) Limited. The detailed description of this small hydro power plant is tabulated in table 6:

[1] Dr. P. Saxena , Overview of Small Hydro Power Development in India , Himalayan Small Hydropower Summit, Dehradun, October, 12-13, 2006. [2] Carmen L.T. Borges , Roberto J. Pinto , Small Hydro Power Plants Energy Availability Modeling for Generation Reliability Evaluation, IEEE Transactions on Power Systems , Vol. 23, No. 3, August , 2008. [3] Raju Gupta, S.N. Singh and S.K. Singal, Automation of Small Hydropower Station, International Conference on Small Hydropower - Hydro Sri Lanka, 22-24 October, 2007. [4] Adhau Sarala P , Economic Analysis and Application of Small Micro/Hydro Power Plants , International Conference on Renewable Energies and Power Quality, Valencia, Spain ,15th to 17th April , 2009. [5] Goyal H , Bhatti T.S , Kothari D.P , Control systems for small hydro-power plants: a review, International Journal of Energy Technology and Policy 2007 - Vol. 5, No.1, pp 97 - 105.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

240

[6] www.esha.be A website on small hydro power plant by European Small Hydro-power Association ESHA [7] http//www.hpseb.com/hydro_potential.htm- A website on hydro power by Himachal Pradesh State Electricity Board. [8] Overview of Small Hydro Development Programme in Himachal Pradesh http://himurja.nic.in-A website on small hydro by HIMURJA (Himachal Govt. Agency on Small Hydro) [9] http://www.cbip.org- A website by Central Board of Irrigation and Power (Govt. of India). [10] Hydro Capacity added during 11th Plan http://www.mnes.nic.in- A website by The Ministry of New and Renewable Energy (MNRE) Government of India. [11] All India generating installed capacity http:// www.powermin.nic.inA website by The Ministry of Power Govt. of India.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

241

Selectivity Techniques of Circuit Breakers in Low Voltage System


1
1, 2

Shelly Sethi, 2K.C. Keshre


2

Tata Consulting Engineers Limited


kckeshre@tce.co.in

shellys@tce.co.in,

Abstract - No project or system can be thought of without electrical system. Therefore protection of any electrical installation is one of the important considerations while designing any system. The electrical installation should be protected against abnormal conditions such as overvoltage, over current and short circuit current conditions etc. However this paper discusses the overload and short circuit current conditions only. Fuses and circuit breakers are two different devices through which the electrical system is protected by timely isolation from live system. Nowadays, the use of fuses is not preferred in electrical system but the same are getting replaced by circuit breakers. Selectivity techniques of Circuit breakers in Low voltage system are explained in this paper. Abbreviation - MCB - Miniature Circuit Breaker, MCCB Moulded Case Circuit Breaker, ACB - Air Circuit Breaker, LV - Low voltage, kA Kilo Ampere, IEC International Electrotechnical Commission.

electricity after the problem is resolved, the switch can simply be turned back on. One advantage of Circuit breakers can simply be reset with a flip of a switch after an overload. However, the technology can be more expensive than a fusebox. Electricians are best qualified to determine whether fuses or circuit breakers are better for a particular electrical installation. When replacing a fuse, you need to take care to ensure the replacement is rated for the same amount of current. A. Types of LV Circuit breakers Air Circuit Breaker (ACB)/Moulded Case Circuit Breaker (MCCB)/Miniature Circuit Breaker (MCB) B. Classification by Category of Circuit breakers 1) Category A The limited short circuit withstand capacity is less than rated short circuit breaking capacity. The Coordination is possible partially. MCCB and MCB are the breakers which come under this category. 2) Category B The limited short circuit withstand capacity is equivalent to rated short circuit breaking capacity. The Coordination with downstream devices is possible. Type of breaker that comes under this category is ACB. In this paper, MCB and MCCB will be discussed for low voltage system. MCB: It is a simple stand smallest encased breaking device available in both domestic and industrial type. It is used for load break, isolation and protection of a sub circuit including motor sub circuits and is shown in Fig1. Its Standard Current Rating is 6125A. Its Short Circuit Rating is 6-50kA. The most essential feature of the MCB is the inverse-time tripping characteristic. This feature indicates the time required to trip the breaker in order to clear the circuit of any given level of overcurrent load. A typical inverse time tripping characteristic is depicted in Fig2 below.

I INTRODUCTION Overloads of large magnitude are dangerous, potentially destroying electrical equipment or causing a fire. Both fuses and circuit breakers will automatically isolate an incoming surge of electrical power past a certain safety limit. But while they both accomplish the same task, each uses different technology in the way that it stops the flow of electricity. Fuses are typically small objects that plug into a fusebox or other central location. Inside the fuse is a small piece of metal, across which the electricity must pass. During normal flow of electricity, the fuse permits the power to pass unobstructed. But during an unsafe overload, the small piece of metal melts due to overheating as a result of high current and then interrupts the flow of electricity. When a fuse is blown, it should be thrown away and replaced with a new fuse. One advantage of fuses is that they are cheap and can be purchased from any hardware store, but they have the drawback of needing to be replaced once they stop an overload. Fuses have another demerit that with lapse of time, their rating decreases which result into mal operation. Circuit breakers are more advanced technology to protect the equipment. Circuit breakers are switches that are tripped when the electrical flow passes a safe limit. The excess of electricity typically triggers an electromagnet, which trips the circuit breaker when an unsafe limit is reached. Once tripped, the switches simply turn off. That stops the flow of electricity, which will remain off until the switch is reset. To reset the flow of

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

242

6) Guarantee adequate support in the case of malfunction of the protection delegated to opening. In this manner, Circuit breakers must be selected in respect of time, current to operate under overcurrent and short circuit conditions. III. SELECTIVITY TECHNIQUES In the overload zone with the circuit breakers in play, time-current type selectivity is usually realised. In the short-circuit zone with the circuit breakers in play, current, time, energy & zone selectivity are used. A. Time-current selectivity Time-current selectivity makes trip selectivity by adjusting the protections so that the load-side protection, for all possible overcurrent values, trips more rapidly than the supply-side circuit-breaker as shown in Fig4. When the trip times of the two circuit-breakers are analysed, it is necessary to consider: 1) Tolerances over the thresholds and trip times. 2) Real currents circulating in the circuit-breakers.
Fig. 2: Inverse-time tripping feature of the MCB

Fig. 1: MCB

MCCB: MCCB is a circuit breaker having a supporting housing of moulded insulating material forming an integral part of the circuit breaker and is shown in Fig3. Its Current rating is 6-3200A and its Short Circuit Rating is 120kA. The MCCB works on thermalmagnetic principle.

Fig. 4: Time-current selectivity

Fig. 3: Moulded Case Circuit Breaker

II. THEORETICAL BACKGROUND Problems and requirements for the coordination of the Circuit breakers are studied in order to: 1) Guarantee safety of the installation and of people in all cases. 2) Rapidly identify and exclude just the area involved in the problem, without indiscriminate trips which reduce the availability of energy in areas not involved in the fault. 3) Reduce the effects of the fault on other integral parts of the installation. 4) Reduce the stress on components and damage to the area involved. 5) Guarantee service continuity with good quality power supply voltage.

B. Current Selectivity This type of selectivity is based on the observation that the closer the fault point is to the power supply of the installation, the higher the short-circuit current is. It is therefore possible to discriminate the zone the fault occurred in by setting the instantaneous protections to different current values. Total selectivity can normally be achieved in specific cases only where the fault current is not high and where there is a component with high impedance interposed between the two protections and therefore a great difference between the short-circuit current values as shown in Fig5.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

243

Fig. 5: Current selectivity

C. Time Selectivity This type of selectivity is an evolution of the previous one. In this type of coordination, apart from the trip threshold in terms of current, a trip time is also defined: a certain current value will make the protections trip after a defined time delay, suitable for allowing any protections placed closer to the fault to trip, excluding the area which is the seat of the fault as shown in Fig6. Generally this type of coordination: 1) Is easy to study and realize. 2) Is not very costly with regard to the protection system. 3) Allows even high selectivity limit values to be obtained. 4) Allows redundancy of the protection functions.

Fig. 7: Energy Selectivity

E. Zone Selectivity This type of coordination is an evolution of time coordination. In general, zone selectivity is made by means of dialogue between the current measuring devices which, once the setting threshold has been detected as having been exceeded, allows just the fault zone to be identified correctly and the power supply to it to be cut off as shown in Fig8. It can be realised in two ways: 1) the measuring devices send the information linked to the current setting threshold having been exceeded to a supervision system and the latter identifies which protection has to intervene;

Fig. 6: Time selectivity

Fig. 8: Zone Selectivity

However, the trip times and energy levels let through by the protections, especially by those close to the sources, are high. It is a type of selectivity which can also be made between circuit-breakers of the same size, equipped with electronic releases with delayed protection against shortcircuit. D. Energy Selectivity Coordination of energy type is a particular type of selectivity which exploits the current-limiting characteristics of moulded-case circuit-breakers. It is pointed out that a current-limiting circuit-breaker is a circuit-breaker with a sufficiently short trip time to prevent the short-circuit current from reaching the peak value which would otherwise be reached [6] as shown in Fig7.

2) When there are current values higher than their setting, each protection sends a lock signal by means of a direct connection or a bus to the higher level protection, before intervening, checks that a similar lock signal has not arrived from the load-side protection. In this way only the protection immediately to the supply side of the fault intervenes. IV SELECTIVITY WITH THE DIFFERENT TYPES OF CIRCUIT-BREAKERS A. MCB-MCB Selectivity These are circuit-breakers with a thermo magnetic release and therefore neither time selectivity let alone zone selectivity is possible. The two selectivity techniques which can be used are current selectivity and energy

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

244

selectivity. Depending on the type of MCB on the supply side, either one or the other can be realised. Only current type selectivity can be looked for between two circuit-breakers as shown in Fig9. In particular, the following prescriptions are valid: 1) In the overload zone, the load-side circuit-breaker must trip more rapidly than the supply-side circuit-breaker, taking into consideration the tolerances and the effective currents circulating in the circuit-breakers as shown in Fig10. 2) In the short-circuit zone given that the following are: a) I3MinA the lowest magnetic threshold of the supplyside circuit-breaker A. b) I3MaxB the highest magnetic threshold of the loadside circuit-breaker B. c) IkB the maximum prospective short-circuit current on the load side of B.

Fig. 11: Coordination Table

B. MCCB-MCB Selectivity The case where selectivity is between a moulded-case circuit-breaker on the supply side and a modular circuitbreaker on the load side is now analysed. In this case, different sizes of the two circuit-breakers are taken as shown in Fig12 so it is possible to obtain energy selectivity.

Fig. 9: Circuit Breaker A & B

The following relationships are verified and total selectivity is obtained. I3MinA > IkB I3MaxB < IkB Otherwise there is partial selectivity and the ultimate selectivity limit is: Is = I3MinA This is assuming that the magnetic trip thresholds of the supply-side circuit-breaker and of the load-side circuit-breaker do not create trip overlapping, taking into consideration the real currents circulating in the circuit-breakers.

Fig. 12: Circuit Breaker A & B with fault at B

Fig. 10: Energy Selectivity

The Is ultimate selectivity limit which is obtained is the one given in the coordination table as Fig11 below:

1) Overload zone In the overload zone, the load-side circuit-breaker must trip more rapidly than the supply-side circuit-breaker, taking into consideration the tolerances and the real currents circulating in the circuit breakers. 2) Short-circuit zone Supply-side circuit-breaker of thermo magnetic type. The magnetic trip threshold must be: a) higher than or equal to 10xIn when the magnetic threshold is fixed b) set to the maximum value when the magnetic threshold is adjustable c) Such so as not to create trip overlapping with the loadside circuit-breaker, taking into consideration the tolerances and the real currents circulating in the circuit-breakers. Supply-side circuit-breaker of electronic type. The instantaneous protection function I must be set to OFF i.e. I3 = OFF The I2 current threshold of function S, less any tolerance, must be adjusted so as not to create trip overlapping with the upper magnetic threshold of the loadside circuit-breaker I3MaxB, taking into consideration the real currents circulating in the circuit-breakers.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

245

With regard to the t2 trip time of function S: t2 A 100ms both with I2t=const as well with t=const as shown in Fig13.

Fig. 13: Energy Selectivity

The Is ultimate selectivity limit is obtained and is given in the Coordination Table as shown in Fig14 below:

To obtain the current type of selectivity, the following prescriptions must be respected: a) In the overload zone, the load-side circuit-breaker must trip more rapidly than the supply-side circuit-breaker, taking into consideration the tolerances and the effective currents circulating in the circuit-breakers. b) In the short-circuit zone given that the following are: I3MinA the lower magnetic threshold of the supply-side circuit-breaker A. I3MaxB the upper magnetic threshold of the load-side circuit-breaker B IkB the maximum prospective short-circuit current on the load side of B The following relationships are verified and total selectivity is obtained. I3MinA > IkB I3MaxB < IkB Otherwise there is partial selectivity and the ultimate selectivity limit is Is = I3minA. This is assuming that the magnetic trip thresholds of the supply-side circuit-breaker and of the load-side circuitbreaker do not create trip overlapping, taking into consideration the real currents circulating in the circuitbreakers as shown in Fig16.

Fig. 14: Coordination Table

C. MCCB-MCCB Selectivity The case where selectivity is between two mouldedcase circuit-breakers is now analysed. In this case, different techniques can be used to obtain selectivity between the circuit-breakers: 1) Current selectivity Current selectivity between moulded-case circuitbreakers may be necessary when there are circuit-breakers of the same size. In any case, only low selectivity values in the order of a maximum of 10 times the In rated current of the release on the supply side can be obtained as shown in Fig15.

Fig. 16: Current Selectivity

2) Time selectivity To obtain the time type of selectivity, the following prescriptions must be respected: a) In the overload zone, the load-side circuit-breaker must trip more rapidly than the supply-side circuit-breaker, taking into consideration the tolerances and the real currents circulating in the circuit breakers as shown in Fig17.

Fig. 15: Circuit Breaker A & B

Fig. 17: Circuit Breaker A & B with fault at B

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

246

b) In the short-circuit zone the I2A current threshold of function S of the supplyside circuit-breaker must be adjusted so as not to create trip overlapping with the current threshold of the protection against short-circuit (I3 or I2) of the loadside circuit-breaker, taking into consideration the tolerances and the real currents circulating in the circuit-breakers. with regard to trip time t2 of function S, the settings of the MCCBs on the supply side are indicated below according to the setting/type of MCCB on the load side: When the I2A threshold of the supply-side circuitbreaker is higher than an instantaneous protection of the load-side circuit-breaker (magnetic, I3=ON or selfprotection) the following is valid: t2A 150ms if I2t =const t2A 100ms if t =const as shown in Fig18.

[4] I.S. Msiza and M. Haffejee; "MCBs versus Fuses in Low-Voltage Applications: Critical Analysis Using the 3-AND Convergence Classifier"; Adaptation and Self-Organizing Systems; arXiv: 0709.3594, Sep. 2007. [5] Protection schemes by Reliance Energy. [6] International Electrotechnical Commission 60947 (part2): Low voltage Switchgear and Controlgear.

Fig. 18: Time Selectivity

when the I2A threshold of the supply-side circuitbreaker is only higher than the I2B threshold of the loadside circuit-breaker, by using curves with the same characteristics, the following is valid: t2A - tolerance t2B + tolerance + 50ms The ultimate selectivity limit is equal to the instantaneous trip threshold I3 of the upstream circuitbreaker minus the tolerance Is = I3minA IV CONCLUSION In this paper, selectivity techniques between circuit breakers are explained in low voltage applications so that healthy system is protected and unhealthy system is isolated from main system and service continuity with good quality power supply voltage is achieved. ACKNOWLEDGEMENTS The author would like to express a deep sense of gratitude and thanks to Tata Consulting Engineers Limited for permitting publication of the paper. REFERENCES [1] L.G. Hewitson, M. Brown and R. Balakrishnan, Practical Power System Protection, ELSEVIER,2005. [2] V Cohen, Miniature and moulded case circuit breakers: A historical review, SAIEE Historical Interest Group, Jan 1994. [3] Technical application of ABB Circuit breakers.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

247

Economy Through Energy Management


1
1
1

Gaurav Laroia, 2Navdeep Kaur Brar

grvlaroia@gmail.com,2brarnav73@yahoo.com

RIEIT, Railmajra, Nawanshahr (Pb, India), 2 BBSBEC, Fatehgarh Sahib (Pb, India)

Abstract This paper provides an overview of energy management measures that can be commonly recommended for an industrial facility. The combination of high electricity prices, limited generation capabilities and the economic situation requires case to case study to improve overall efficiency. In addition to using energy saving appliances Keywords Diversity factor, power factor capacitor banks

1 2 3 4 5 6 7 8

I. INTRODUCTION Energy management in industry in India at present is one of the most taxing problems because there is difference of about 25% in its demand and supply. Therefore the efficient use of electricity by the end users will have a multiplying effect in the saving of national economy [2, 3]. As major areas of use of electricity in modern industries are lighting, air conditioning, electrical motors etc. [1]. In order to efficiently manage electricity in the above categories, electrical energy management opportunities in these fields have been discussed in details [6, 7]. One of the most important aspects of energy management programme is electrical load analysis [5, 8]. The important parameters of electrical load analysis such as connected load, diversity factor, power factor, importance of reactive compensation using capacitor banks, capacitors for reactive compensation etc have also been taken into considerations [4]. In this paper, energy auditing of nitric acid plant is done. II. PLANT LAYOUT The total load of the plant is 1950KW. Where 1855 KW of machine load and 18 KW of light and other type of load as shown in Table 1.The segmental view of load distribution of industry is as shown in Fig 1.
TABLE 1: CATEGORY WISE LOAD OF VARIOUS PLANTS

Fig. 1: Segmental view of load distribution of industry in percentage as per Table 1

Maximum Demand of different sections is as shown in Fig 2: Sections of plant


500 400 300 200 100 0 1 2 3 4 5 6 7 8 Series1

Fig. 2: Maximum demand of different load The bar chart for different loads as for the sections (1) 295 KW, (2) 127 KW, (3) 327 KW, (4) 318 KW, (5) 427 KW , (6) 192 KW , (7) 166 KW, (8) 10 KW (varies in 24 hours).

Miscellaneous load includes light fan AC which varies time to time Maximum and minimum demand of these load is as follows

Sr. No. 1 2 3 4 5 6 7 8

Category of load Filtration plant Vaporization plant Cooling and absorption plant Bleaching plant Water treatment and workshop plant Ammonia plant Compression plant Miscellaneous load

Load in kW 312 137 351 332 429 195 175 100

Percentage value 16 7 18 17 22

3kw A.C

0.1 kw 2.16kw 0.4kw 0.5 kw

1 2 3 4 5 6

5 kw E.FAN

Fig. 3: Miscellaneous load

10 9 5

Section one to compressor remains fixed for 24 hours but the light and other load varies in 24 hours. Energy auditing

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

248

can be done by replacing florescent tubes by CFL lamps , by installing capacitor bank , by installing voltage stabilizers, by reducing load up to the required level, Halogens (spot lights) are replaced with infra read coating halogens ,Replacement of the magnetic ballast from electronic ballast etc. The energy can also be saved by installing capacitor bank near to the load. By this measure reactive current will be reduced and thus energy loss can be reduced. By installing capacitor the paper browses on installing capacitor bank on machine under three schemes as follows. III. PROBLEM FORMULATION From technical considerations it is always desirable to improve the P.F. of an installation to unity. For the given load, improvement in P.F. would reduce maximum KVA demand and hence the fixed charges. This saving however can be effected by the installation of P.F.. improvement plant, which requires annual expenditure. Consumer will therefore improve P.F. . to that value for which he is able to get maximum saving in annual expenditure. For this three different locations are analysed (scheme I, II and III) for the placement of capacitor banks and to get most economical P.F. a) Calculation: Let a consumer is supplied load P at P.F.. cos1 with maximum KVA demand S1. Let it is improved to cos2 with max KVA demand S2. If consumer is to pay maximum demand charge of Rs A/KVA per annum, then the annual savings affected by his P.F. Improvement plant = A (S1 S2) = A [P/ cos1 - P/ cos2] Reactive KVA rating of . P.F. improvement plant is given by RKVA = P (tan1 tan2) If. P.F. improvement plant costs Rs B/KVA/Annum then annual expenditure on P.F.. improvement Plant = BP (tan1 tan2) Net annual savings (S) = AP [1/ Cos1 - 1/ Cos2] - BP (tan1 tan2) For Maximum saving = ds/d2 = 0 Ds/d2 = - AP sec2 Tan2 + BPsec22 =0 Or AP sec2 Tan2 = BPsec22
A sin cos2
2 2

For Scheme I Total Cost = 1, 97, 000+35.89xTx cost per unit of energy For Scheme II Total Cost = 2, 41,500+27.97xTx Cost per unit of energy For Scheme III Total Cost=4, 35,000+16.011xTx Cost per Unit of energy i) Study of technical and economical aspects has been be done ii) Comparison has been be done between the three schemes iii) Break even analysis has to be done and optimum location has to be selected c) Comparison: Following are the comparison done of three schemes of installing capacitor bank.
TABLE II: COMPARISON OF INSTALLATION COST AND ENERGY SAVED IN
THE THREE SCHEMES

Scheme I 1) Installation Cost Distribution Losses (in kW) Energy lost in day (24 Hours in units Energy lost in a Month(in units) Cost of energy wasted in a month (@ Rs 4.50/ unit in Rs.) Cost of energy lost in one year 1,97,000/35.89 861.36 25840.8 116283.6

Scheme II 2,41,500/27.971 669.6 20088 90396

Scheme III 4,35,000/16.011 384.26 11527.8 51875.1

1395403.2

1084752

622501.2

Amount of money saved in one year if scheme II is adopted as compared to scheme I= Rs. 310651.2 Amount of money spent in excess of scheme I if scheme II is adopted = Rs. 44, 500/If scheme III is adopted, amount of money spent on installation in excess of scheme II is 1, 93,500/Amount of money saved in one year due reduction in losses = Rs. 4, 62,250.8
2000000 1900000 1800000 1700000 1600000 1500000 1400000 1300000 1200000 1100000 1000000 900000 800000 700000 600000 500000 400000 300000 200000 1 2 3 4 5 6 7 8 9 10

Series1 Series2 Series3

B cos2
B A
2

sin

Time in hrs x 1000


Fig. 4: Break Even Analysis

sin

B A

Annual expenditur e per KVA of phase Advancing plant Annual fixed charges per KVA of max Demand Total cost= Installation cost + Cost of energy lost =

b) Methodology: The following methodology has been adopted: For this different three locations for the capacitor bank have been selected. (i) At LT panel (ii) at distribution board (iii) at the load

Installation cost + Power lost x Operating hours x Cost of one Unit of Energy . Where T represents total operating hours of plant Total costs (Installation + Cost of energy) for different working hours for the three schemes have tabulated as below:

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

249

TABLE III: TOTAL COST FOR DIFFERENT WORKING HOURS FOR THREE
SCHEMES

Working Hours

Scheme I 1,97,000+35.89x TxCost of electricity per unit 358505 520010 681515 843020 1004525 1166030

1000 2000 3000 4000 5000 6000

Scheme II 2,41,500+2 7.97xTx cost of electricity per unit 367365 493230 600150 725700 851250 976800

Scheme III 4,35,000+16.011xT x cost of electricity per unit

example of a research paper example of a research paper example of a research paper example of a research paper example of a research paper example of a research paper

in [3] in [4] in [5] in [6] in [7] in [8]

507045 579090 651135 723180 795225 867270

IV. RESULTS AND DISCUSSIONS Scheme1 Cost of installing capacitor bank of 790 KVAR at LT panel = Rs. 1, 97,000/Scheme 2 Total cost of installing capacitor banks in scheme II= Rs. 2, 41,500/ Scheme 3 Total cost in scheme III=Rs. 4, 35, 000/ Amount of money saved in one year if scheme II is adopted as compared to scheme I= Rs. 3, 10,651.2/Amount of money spent in excess of scheme I if scheme II is adopted = Rs. 44, 500/If scheme III is adopted, amount of money spent on installation in excess of scheme II is 1, 93,500/Amount of money saved in one year due to reduction in losses = Rs. 4, 62, 250.80/Amount of money saved in one year if scheme II is adopted as compared to scheme I= Rs. 3, 10,651.2 Therefore the additional expenses occurred in construction of central panels for connecting capacitor banks at the load points are recoverable in a period, which is less than one year. Break even Analysis when the scheme II and III are compared with scheme I, shows that cost of scheme II equalized with cost of scheme I after 2000 working hours, after which the cost in scheme II is less, which shows, that if working life of the plant is more than 2000 hours, the scheme II will be beneficial. The cost of scheme III equalizes with the cost of scheme I between 3000 to 4000 working hour after which scheme III costs less, therefore if the operating life of the plant is more than 4000 working hours, scheme III will be beneficial over scheme I. When scheme II is compared with scheme III, It is found that the costs of the two equalizes at between 4000 to 5000 hour of operation after which the operating costs of scheme III will be less than that of scheme II. Therefore scheme III must be followed for the plant if its working life is more than 5000 hours. References example of a research paper in [1] example of a research paper in [2]

V CONCLUSIONS Energy saving by connecting Power factor improvement equipment near to load instead of connecting at LT Panel In scheme I distribution losses are 35.89 kW In scheme II distribution losses are 27.971 kW In scheme III distribution losses are 16.011 kW So if one will adapt the method used in Scheme II instead of method used in Scheme I, one can save 7.919 kW. And if one will adapt the method used in Scheme III, instead of method used in Scheme I, one can save 19.879 kW. As power run on 24 hour basis, total unit of energy saved in one year are 19.879 x 24 x 365 = 174131.28 KWH If cost of energy is Rs. 4.5 per unit, it will save Rs. 783463.50 annually. Total Industrial load in India at present is 80,000 MW which will rise to 100,000 MW by the end of this century. If 2% of energy is saved by following the Principles discussed 2% saving in energy would mean saving of 2000 MW of Power, the cost of which comes out to be Rs. 90,00,000.00 Lacs. So it is concluded that in addition to using energy saving appliances if capacitor bank has to be installed at proper location then optimum economy energy efficiency can be achieved. REFERENCES
[1]. H.K. Wong C.K, Application of Energy Audit in Buildings, IEEE International Conference on Advances in Power System Control, Operation and Management, pp. 977-981, 1993. [2]. N. Vasudevan, Energy efficiency energy consideration in the Indian foundry industry, pp. 2263-2267, 1996. [3]. Susila M Tennakoon W. W. L Keerthipala, W B Lawrance, Solar Energy for Development of a Cost-Effective Building Energy System, pp. 55-60, 2000. [4]. S.S. Murthy, C.S. Jha , S.R. Lakshmi, A.K. Rai, D.A. Desai , Energy conservation aspects of induction motors using design and power controllers, pp. 677-682, 2000. [5]. Mohd Amran, Mohd Radzi, Software Development for Energy [6]. Auditing Practice, pp. 420-424, Student Conference on Research and Development, 2003. [7]. M. Al-Mujahid, M.A. El-Kady King Saud University, Advancedenergy auditing and conservation in industrial facilities, pp.656-662, 2003. [8]. M. A. Rosen, The role of energy efficiency in sustainable development, pp. 140-149, 2004. [9]. A. Prudenzi, M. Di Lillo, A. Silvestri, M.C. Falvo, A software tool for energy audit activities in buildings, pp. 452-457, 2008.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

250

Parameter Variation in Separately Excited DC Motor


1

Rakesh Sing Lodhi, 2H.K.Verma, 3D.P.Kothari


1, 2, 3

SGSITS Indore, India

Abstract:-This Paper proposes the MATLAB simulation of Separately Excited DC Motor. The novelty of this paper lies in the modeling & transfer function model of Separately Excited DC Motor, PID Controller & Chopper. The separately excited DC motor has been modeled using MATLAB 7.0 toolbox. It has been found the simulation results of Separately Excited DC Motor in different parameters & different simulation techniques. Keywords: Separately Excited DC Motor, MATLAB 7.0 toolbox, Chopper, PID Controller.

I.

INTRODUCTION III. MODELING OF DC MOTOR The state space model of separately excited dc motor is as follows:

Computer modeling and simulation tools have been extensively used to support and enhance electric machinery courses.MATLAB with its toolboxes such as Simulink [1] and Sim Power Systems [2].Traditionally rheostatic armature control method was widely used for the speed control of low power DC motors. The desired torque-speed characteristics could be achieved by the use of conventional proportional-integral-derivative (PID) controllers. The difference between the desired current, and the current being injected is used to control the switching of the chopper circuit [3].Choppers are used for the control of DC motors because of number of advantages, such as high efficiency, flexibility in control, light weight, quick response, and Regeneration down to very low speeds. Chopper controlled DC drives have also applications in servos in battery operated vehicles such as forklift trucks, trolleys, and so on. The separately excited DC motor is also employed in traction. Regenerative braking of a SEDM is fairly simple and can be carried out down to very low speeds. A chopper can be operated at high frequency [4]. II. SPEED CONTROL SYSTEM In speed control system it has six blocks DC motor, chopper, current controller, speed controller current transducer & Speed transducer as shown in Figure:1.Speed control of Separately excited DC motor by using chopper.PID controller is used in place of speed controller. Chopper is used switched to force the current down. On the other hand when the error reaches the lower hysteresis limit a positive pulse is produced to increases the current. Speed Transducer is used to sense the speed & Current transducer is used to sense the current.

X
Y
x1
x2

A x B u
C x D u
K La
B J
X1 X2

Ra
K

x La 1

x2

1 La
1 J 1
La 0

u (1)

x1 J Ra K
La K J La B J

x2

TL (2)
0 u 1 J

(s) V (s) (Ra a

K SLa )(JS B)S KK b S

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

251

IV. SPEED CONTROLLER PID Controller is used as speed controller. The characteristic of PID control action are no oscillations, improves the transient response, and improves the steady state response. The mathematical formulation is:

C(t)

KC E(t)

KC t TI 0

E(t).dt KC TD

dE dt

kept constant [3].The speed control circuit of an armature controlled SEDM using chopper circuit The desired torque speed characteristics could be achieved by the use of conventional proportional integral derivative (PID) controller [4].The block diagram of speed control of SEDM using chopper as shown in Fig5.

By Taking Laplace Transform;

C(s)
C (s) E (s)

K C E(s)
G(s)

K C E(s) TI s
C T s T s I I K

sK T E(s) C D
s 2T T D I 1

Where: G(s) is Gain of PID Controller TI = Integral control TD = Derivative Control V. CHOPPER Choppers are used for the control of DC motors because of number of advantages, such as high efficiency, flexibility in control, light weight, quick response, and Regeneration down to very low speeds. Chopper controlled DC drives have also applications in servos in battery operated vehicles such as forklift trucks, trolleys, and so on. The separately excited DC motor is also employed in traction. Regenerative braking of a SEDM is fairly simple and can be carried out down to very low speeds. A chopper can be operated at high frequency. The chopper is modeled as a first order lag with a gain of Kr. The time delay corresponds to the statistical average conduction time, which can vary from zero to T.The Transfer Function is then:

VII. SIMULATION AND RESULTS Simulation of Separately excited dc motor in deferent parameters & different simulation technique.

Gr ( s ) 1

Kr sT 2

Kr
1

1 sT 2

Figure 4: Transfer Function of Chopper

VI. SPEED CONTROL OF SEDM BY PARAMETER VARIATION The motor must be separately excited to use armature voltage control. When the armature voltage is increased, the no load speed of the motor increases while the slope of the torque speed curve remains unchanged since the flux is

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

252

VIII. CONCLUSION The speed takes a certain time to initially reach the level of the final value (the rise time). The speed overshoots the level of the final value. The speed oscillates about the level of the final value. The chopper-fed de series motor is essentially a sampled data system. It can be approximated by an equivalent continuous model for predicting transient response, involving variations of average speed and current with time, due to change in either duty ratio or load torque T L. REFERENCES
[1]. SIMULINK, Model-based and system-based design, Using Simulink, Math Works Inc., Natick, MA, 2000. [2]. SimPowerSystems for use with Simulink, users guide,MathWorks Inc., Natick, MA, 2002. [3]. Handbook of power electronics second edition Tata Mc-Graw-Hill by MD Singh, K B Khanchandani page 888. [4]. Brijesh Singh Intelligent PI Controller for Speed Control of D.C. Motor International Journal of Electronic Engineering Research ISSN 0975 - 6450 Volume 2 Number 1 (2010) pp. 87100 Research India Publications. [5]. S.J.Chapman, Electric machinery fundamentals, 3rd Edition WCB/McGraw-Hill, New York1995. [6]. Moleykutty George Speed Control of Separately Excited DC Motor American Journal of Applied Sciences 5 (3): 227-233, 2008 ISSN 1546-9239 2008 Science Publications. [7]. R.KRISHNAN, Handbook of Electric Motor Drives page (133134) PEARSON EDUCATION. [8]. Wester, G. W. & Middle brook, R. D. Low- Frequency Characterization of Switched dcde Converters, IEEE Transactions on Aerospace and Electronic Systems, vol. AES- 9, n.3, pp.376-385, 1973. [9]. Rakesh Singh Lodhi Analysis & Design of PID Controller for Chopper-fed Separately Excited DC Motor national conference of ECOMM-11.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

253

Power Enhancement of Transmission Line


1
1,2
1

S. S .Matharu, 2Vikramjit Singh Sanghotra


2

DAV Institute of Engineering & Technology, Nawanshahr


labhsanghotra@yahoo.co.in

ssmatharu2002@yahoo.com,

Abstract -- With ever-growing need of electric power in day to day life, the power generation /transmission is faced with limited resources & its high cost. To overcome the problem of transmission of electric power, HVDC lines are the only alternative than the traditional AC lines, because of having advantages of reduced voltage drops over the long distances & having lesser operating cost. On the existing AC lines with having maximum current limits, the most effective method power enhancement is simultaneous transmission of ac & dc, without doing any modification in structure as well as of insulators , however the effects of faults occurring on AC lines toward HVDC lines circuit & vice versa ,the effect of radio interference, audible noise & corona is required to be considered. The paper present the various methods of power enhancement , their comparison & signifies the best method of all. Keywords - HVDC, EHVAC.

line into HVDC pole on one circuit for bulk transmission & other circuit remains as AC in order to tap the small loads in between, are different alternatives . The paper discusses the various factors encountered & possible remedial action for implementation of simultaneous transmission of AC & DC. II. CHOICE OF SELECTION HVDC ON ALTERNATE CIRCUIT

I. INTRODUCTION Because of industrialization of developing countries & more & more increasing usability of domestic electric appliances it is mandatory to increase the generation of electric power. Developing countries like India, china are fast approaching to enter in the devolve nations. Normally the generation of electrical power is economical at the points where raw material producing this is cheaper, also the power resources are not evenly distributed through out the world as hydel projects can be installed at places where sufficient quantity of water & head is available , thermal plants can be installed near coal mines /Gas fields/Oil wells etc.& these places are always away from the load centers, thereby transmission of electric power from generating plants to load centers is the need of the day. While planning, initially most power transmission lines are designed considering the future demand of load & initially these lines are loaded with low capacity. As the demand of the power increases the utilization of reserve capacity of lines regularly utilized & the time comes where whole optimum capacity of the lines are depleted. Initially as the load is less the lines are uncompensated but these are required to be series compensated using shunt reactors at the suitable intervals, as the load on the lines continue to be increasing as time lapse, Shunt capacitors & FACT devices pushes more power by way of voltage control in the transmission lines. By modification in replacement of Conductors & improving the cooling system of thyristor valves, converter transformer, increment in the capacity of transmission line is an alternative. Hybrid transmission system using AC superimposed on DC, Transmission of AC/DC on same tower by converting one circuit of double circuit of AC

Fig. 1: Economic Comparison of EHVAC-HVDC

From the graph above EHVAC is economical below the distance of 600 km & HVDC is cheaper for longer distances. III. DIFFERENT METHODS FOR UPGRADING OF POWER IN TRANSMISSION SYSTEM From one AC system to an other AC System, more power can be transformed by following methods: By installing new transmission line along with the existing line. Power upgrading by SIL using series compensation [1]. Power upgrading by combining AC-DC transmission.[7] By converting one circuit of double circuit AC into an DC circuit (Hybrid Transmission).[6] IV. POWER UPGRADING BY SIL USING SERIES COMPENSATION The surge impedance loading or the natural loading does not impose any restriction on the length of the transmission line but it restricts the amount of power for the certain value of voltage. Keeping in view Stability considerations, If resistance and shunt leakage of a transmission line is neglected the power that can be transmitted within stability limit is P=

V S Vr sin , V S = Sending end Voltage X V r = Receiving and voltage


X = Series reactance in ohm/plan = Load angle - angle between Vs and V r

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

254

When VS P=

Vr
=

V
Kv sin X
0
2

V 2 Sin X

for 3 phase where KV =

line to line voltage. If = 90 maximum power can be transmitted. But is not allowed to vary more than 20 0 0 to 30 . Otherwise, the machines will loose synchronism under transient conditions. Consider a case, when x= 0.3475 ohm per km.

V. POWER UPGRADING BY COMBINING AC-DC TRANSMISSION In this method for the upgrading of the power in the existing transmission line is to modulate the DC power on the AC power has devolved by Mr H Rahman & B . H . Khan of Aligarh Muslim University. [7]

300 S = length in km of line. We have 2 2 2 kv kv kv sin x0.5 1.43 MW P= x 0.34755 S


If the line in to operate at natural load then PSIL =

KV ZS
2

Equating the two

1.43 kv S

kv ZS

Fig. 2: Composite AC-DC System

or S= 1.43 Z S km.

This means that the length of the line that can operate at natural loading without loosing stability is 1.43 Z S (For the assumed value of x, for any other value of X these is another fixed value of S). This imposes a limit of on the length of transmission line in an A.C. system. In EHV line exceeding 200 km additional equipment like series capacitance and shunt reactor is to be installed. Series compensation is an important method in improving the performance of EHV lines. It increases the power handling capacity and reduces voltages regulation. Power transfer capacity P =

VSVr Sin X

Is the load angle or torque angle (angle between VS and Vr) If a series compensator in installed in the line having X C as the capacities reactance, the net reactance is ( X L X C ) then power transfer capacity of such a line is given by P
1

VSVr Sin XL XC
XL XL Xc XL XL I

then ratio

P P

The basic scheme for simultaneous AC-DC power flow through the double circuit transmission line is to modulate the DC power on an AC power. As shown in the diagram above The DC power is obtained through line commutated 12 pulse converter controlled rectifier & injected to the neutral point of zigzag connected secondary winding of the transformer & reconverted to again AC power by 12 pulse inverter circuit. Connected to neutral circuit of zigzag secondary winding of receiving end transformer. The double circuit of AC transmission line carries both 3 phase AC & DC power. Each conductor of each line carries 1/3rd of the DC current along with the AC current. The three conductors of the 2nd line provide the return path for the DC current. Zigzag winding of both the transformers are connected to avoid the saturation of the core of the transformer because the two flux are produced by the DC current (Id/3) through each winding in the each limb of core of zigzag transformer are equal in magnitude & opposite in direction so the net DC flux at any instant of time becomes zero in each limb of the core, result in the DC saturation of the core is avoided. A high value of reactance Xd is helpful in reducing harmonics in the DC current. The circuit was modeled by the authors in PSCAD/EMTDC. Table below shows the comparison of computed/simulated results that almost power is doubled.
TABLE I. COMPUTED/SIMULATED RESULTS Power Angle in (degree) AC power (MW) AC Current(KA ) DC Current (KA) DC 30 45 60 75 80

1 Xc X2 1 k

whereK

Xc XL

K=

XC in known as degree of series compensation. XL


0.4

290/295 .416/.41 5 5.25/5.2 4 1685/17

410/411 .612/.61 1 5.07/5.1 14. 1624/16

502/495 .805/.79 6 4.82/4.9 1 1545/15

560/541 .98/.96

571/548 1.03/1.02

Obviously Xc < xL

XC The optimum compensation in obtained with XL


to 0.5 i.e. the power upgraded is almost doubled. [1]

4.52/4.6 3 1419/14

4.41/4.52 1413/146

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

255

Power(MW) P Total Pr. (AC+DC) (MW)

15 1971/19 88

57 2034/20 51

85 2048/20 62

98 2010/20 19

7 1985/199 5

VI. HYBRID EHVAC/HVDC SYSTEM On the existing EHVAC transmission line, one of the circuit of 230KV AC is converted into HVDC line which runs parallel on the same tower conFiguration. The input tapped from the Bus Bar is given to converter transformer & controlled rectifier
Fig. 4: Proposed hybrid transmission Line

VII. COMPARISION OF VARIOUS METHODS OF POWER ENHANCEMENT


Comparison Of Power Flow
Sl No. SIL Conversi on from AC to DC 180% Parallel Operatio n of AC & DC 147% Superim posing AC on DC 3 to 3.5 times the original power

01

1.5 to 2 times

Fig. 3: Hybrid EHVAC/HVDC System

Circuit which may be formed from power electronic devices such as IGBT (Insulated gate Bipolar transistor) having suitable capacity to handle the required amount of power. The DC output is controlled with the help of rectifier firing control circuit, which is connected with the sensing element of power. At the other end of the transmission line, an inverter circuit is provided which converts DC supply into AC supply. Here also, a firing control circuit is associated along with the inverter to control required amount of power. This AC is given to converter transformer for suitable voltage level adjustment with the next power system. The smoothing reactors & filters are provided in the transmission line to tackle with the problems created from different types of harmonics & noises. The power flow in AC line = ( Va x Vb x Sin ) / X Where Va = Voltage of system A Vb = Voltage of system B = Angle between voltage & current X = reactance of the line. Power Flow in DC line Pd = ( Ud1 Ud2 ) x Ud2 / R Where Ud1 = Dc Voltage level at the rectifier end Ud2 = DC voltage level at inverter side end R = Resistance of the line The followings are the various considerations required to be taken into account while considering the parallel operation of EHVAC & HVDC circuits. a) Interaction between the two lines .i.e EHVAC & HVDC b) Corona radio interference & audible noise c) Effect on sensation level. d) Faults in converter transformer & its protection circuits.

VIII.CONCLUSION Keeping in view the national integration, economic feasibility of the system, Bulk transmission of power between the national grid, low losses & better system stability, there exist the feasibility of simultaneous transmission of ac &dc. REFERENCES
[1]. EHVAC & HVDC Transmission & distribution Engineering by S. Rao . Khanna Publisher. P 928. [2]. B.A.Clarmont , G.B.Johnson , L.E.Zaffanella, S.Zelindger . Effect Of HVAC-HVDC line Separation in a hybrid corridor Ieee Transaction on power delivery, vol. 4 April 1989. [3]. D Woodford. Secondary Arc Effect in AC/DC Hybrid transmission . Vol.8 pp 704 711, 1993 [4]. R. Verdi in , A M Gole , E kuffel , Ndiseko , B . Bisewski . Induced Over voltage On An AC-DC Transmission System. IEEE Transaction on power delivery,vol . 10 pp 1514-1524 , 1995 . [5]. M.I. Khan, R.C. Agrawal Conversion of AC Line into HVDC . IEEE PES 2005 Conference & Exposition in Africa, 11-15 July 2005. [6]. D. A . Halamay, K . M . Sexby & J.L. Bala. Feasibility study of high voltage dc & ac multi circuit hybrid transmission line. Proceeding of 37th annual North American Power Symp , pp 310316 Oct 23-25 2005 [7]. Rahman & B. H. Khan Power upgrading of transmission line by combining AC-DC Transmission, IEEE Transaction on power system Vol. 22 Feb. 2007 . [8]. Study of dynamic performance of AC & DC hybrid System in china southern power grid Kaijjain OU IEEE PES Power Africa 16-20n July.2007

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

256

Wind Power Generation for Rural India


1Sanjay

Jaiswal, 2Kumkum verma, 3Dheeraj Jain, 4Neeru Devi


4

1, 2, 3, 4
1

to.sanjay1985@gmail.com,2kumkum.verma1983@gmail.com,3dheeraj.suryamtech@gmail.com

Institute of Technology & Management, Bhilwara, Rajasthan

1, 2, 3, 4 neeru130@gmail.com
and development of alternative energy sources. Many such energy sources, such as wind energy and photovoltaic are now well developed, cost effective and are being widely used, while others, such as fuel cells are in their advanced developmental stage. As mentioned, these energy sources are preferred for being environmental friendly. The integration of these energy sources to form a hybrid system is an excellent option for distributed energy production. However, there are some major concerns which obstruct the free and full-fledge development of alternate energy sources. One of the major issues is cost, but it has been observed that the cost of renewable energy shows a down hill trend with the increase in its demand and production. Also rapid advances in the field of power electronics has enabled the cost reduction of renewable energy systems and have also ensured better reliability of such systems. Among the renewable energy sources, photovoltaic cell and wind turbine systems make use of advanced power electronics technologies. The problem of energy storage is also of major concern for renewable energy. Since most of these energy sources are discontinuous in nature, efficient storage devices need to be designed to store such forms of energy. The focus in this dissertation will be on small size variable speed wind energy, photovoltaic, and fuel cell as a hybrid system. II. THE WIND ENERGY Generation of electricity from modern wind turbines is now an established technology, although many developments are yet to come. Worldwide there are more than 20,000 turbines; with the most cost effective for grid integration starting at approximately 400kW capacity and 40m in rotor diameter [1]. A typical capacity factor on a good site (wind speed average > 6m/s) is 25 to 30%. Such turbine can be expected to supply 20 to 40% of its rated annual energy into a local grid, annually [2]. Maximized electricity generation by wind turbines is an interesting topic in electrical engineering and many types of variable speed generating systems have been researched to achieve this goal. Use of a variable speed generating system in wind power applications can increase the captured wind energy by 10-15% annually. This can yield a significant revenue increase over a 20 or 30 years life of operation [6, 7]. III. BASIC OF WIND TURBINE The schematic representation of a variable speed concept is shown in Fig3.1. The variable speed turbine can generate electricity from winds with speeds ranging from 119 to 65 miles per hour. In conjunction with the variable

Abstract- In this paper focuses on studying the operating control of direct-drive grid-connected wind-power generation system of variable speed PMSWG. After wind turbine simulation is analyzed first, the closed-loop current control method of DC motor based on BUCK speed regulating circuit is put forward. Non-controllable rectifier and grid-connected PWM inverter is adopted as the AC gridconnected circuit of PMSWG. Using DC voltage as gridconnected parameter, the method which observes speed from DC voltage without speed sensor is proposed. The novel maximal wind-energy tracing strategy of direct control of grid-connected power is proposed The simplest control method for maximum power tracking by employing a step-up dc-dc boost converter in a variable speed wind turbine system, using permanent magnet machine as its generator, is introduced. The Output voltage of generator is connected to a fixed dc-link voltage through a three-phase diode rectifier and the dc-dc boost converter. So maximum power-tracking algorithm calculates the reference speed, corresponds to maximum output power of the turbine, as the control signal for the dc-dc converter. The dc-dc converter uses this speed command to control the output power of the generator, by controlling the output voltage of the diode rectifier and input current of the boost converter, such that the speed of generator tracks the command speed. A current regulated pulse width modulation voltage source inverter maintains the output voltage of the dc-dc converter at a fixed value by balancing the dc-link input and output power. Keywords-SWG, non-controllable rectifier, PWM inverter, grid-connected power direct-control, maximal wind energy tracking.

I. INTRODUCTION The energy sources, such as oil, coal, natural gas, and nuclear are finite and generate pollution, use of renewable energy sources, for instance, solar, wave, biomass, wind, monohydric, and tidal power, as a major form of clean technology could be the right solution to solve energy crisis in the recent century. The main advantage of renewable energy over fossil fuels and nuclear power is the absence of harmful emissions, including carbon, sulphur, nitrogen oxides, and radioactive products. In this way renewable energy sources do not have the high external cost and social issues of the alternates. Moreover, supply and consumption of energy based on conventional fossil fuel is considered as a significant factor of global warming and environmental deterioration. The utilization of natural energy is recognized as a new energy source which will eventually replace conventional energy sources [2,]. The ever increasing demand for conventional energy sources has driven society towards the need for research

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

257

speed feature, this wide operating envelope increases the turbines energy capture by 10 to 15 percent or more over a comparably sized constant speed turbine, as shown in Fig3.2.

Fig. 3.1: Variable speed concept

Fig. 3.2: Increased energy capture using variable speed turbine.

The variable speed turbines rotor can turn faster as wind speed increases, storing some of the winds energy as kinetic energy, which generates additional electricity when released. Fig3.3 shows a typical small size variable speed wind turbine which is using a permanent magnet machine as its generator.

V. POWER CALCULATION AT VARIOUS STAGES The power output of a wind generator is proportional to the area swept by the rotor - i.e. double the swept area and the power output will also double. The Power of Wind Since air is made up of air molecules and each molecule have some mass, so kinetic energy is generated due a moving particle. Therefore kinetic energy is defined as Kinetic Energy = 0.5 x Mass x Velocity2 Air has a known density (around 1.23 kg/m3 at sea level), so mass of wind striking wind turbine per second is given by the following equation Mass/sec (kg/s) = Velocity (m/s) x Area (m2) x Density (kg/m3) Therefore, Power = 0.5 x Swept Area x Air Density x Velocity3 P = 0.5 x x A x Cp x V where: P = Power in watts = Air density (about 1.225 kg/m at sea level, less higher up) A = Rotor striking area, exposed to the wind (m) Cp = Coefficient of performance V = Wind speed (m/s) Now, the tip speed ratio of wind turbine is given by dividing the speed of the tips of the turbine blades by the speed of the wind i.e; Tip speed ratio (Tsr) = Tip Speed of blade at 60 rpm (1rps) (2) x (1) x (radius) where radius = 0.5m, therefore Tip speed of blade = 3.14 m/s If wind speed = 9 m/s, then Tsr = 3.14/9 = 0.35

Fig. 5.1: Tip speed v/s Cp Fig. 3.3: Small size variable speed wind turbine using PMG.

IV. CLASSIFICATION OF WIND TURBINE It is two types: Horizontal axis wind turbine Vertical axis wind turbine 1. Vertical-axis wind turbines [VAWT] in which the axis of rotation is vertical with respect to the ground (and roughly perpendicular to the wind stream), 2. Horizontal-axis wind turbines [HAWT] in which the axis of rotation is horizontal with respect to the ground (and roughly parallel to the wind stream.).

From the above graph of coefficient of performance (Cp) v/s Tip Speed Ratio we get Cp corresponding to 0.35 as 0.11. From power equation P = 0.5 x x A x Cp x V3 A = 0.5 x 1m In above relation, 0.5 is horizontal length of striking and 1is the height of blade and V is 9m/s So, Mechanical power P = 0.5 x 1.225 x (0.5 x1) x 0.11 x 9 P = 25 W Assume 40% efficiency to convert mechanical power in to electrical power.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

258

So, Electrical power = 0.4 mechanical power = 0.4 25 = 10 W. E.M.F (VOLTAGE) PRODUCED E= E= Induced E.M.F in volt. = flux in Wb. Z = no. of conductor in turns. N = Blade speed in r.p.m P = no. of pole. a = no. of parallel path.
Fig:. 5.3 Generated E.m.f v/s Blade speed

Where,

E= = 12 V
TABLE 1: CALCULATION FOR ELECTRICAL POWER OUTPUT AT VARIOUS
WIND VELOCITIES

Wind speed Km/h 0 5 10 15 20 25 30 35 40 45 50 55

Wind speed m/s 0 1.385 2.770 4.155 5.540 6.925 8.310 9.695 11.080 12.465 13.850 15.230

Calculated mechanical power in watt 0 0.089 0.715 2.410 5.720 11.180 19.320 30.690 45.810 65.230 89.470 119.09

Electrical power =0.4 mech.power Watt 0 0.035 0.286 0.966 2.290 4.470 7.730 12.270 18.320 26.090 35.790 47.630

VI. CONCLUSION The implementation of vertical axis wind turbine on road dividers would be a great asset to the ministry of Non-conventional energy Resources as it would reduce the burden on the consumption of conventional energy sources. They can be installed on any highway with the width being the only constraint. These turbines are simple in construction and require less investment. Since, turbine is in small size, it can harness a limited amount of wind. Therefore they can be used for low power application such as for street lighting on any busy road. Moreover it can also be use to light up the advertisement hoardings. Other application could be in diversions on highways and traffic lights. Further more, these turbines can find application in lighting up the toll plazas on highways. Since the battery is portable we can use it in some other location for any low voltage purpose. Thus there is balance between the cost and the power available. Future prospect, the addition of speed governing system and control circuit may make the model much acceptable. The emerging trends in the technology have shown a way to the use of non-conventional energy sources so efficiently and a little effort at the side may find an effective solution for the boom of the electrical energy by the society. REFERENCES
[1]. P. Klein, J. Schmid, Eurowin- the European Wind Turbine Data Base, Annual reports from 1990 and 1991, DG XII. Brussels, 1992. J. Twidell, Renewable Energy: Implementation and Benefits, Advances in Power System control, Operation and Management, 1993, pp. 418-424, vol.1. Z. Chen and E. Spooner, Wind Turbine Power Converters: a Comparative Study, in Proc. 1998 IEE Power Electronics and Variable Speed Drives, Seventh International Conference, pp. 471 476. Z. Chen and E. Spooner, A Modular Prmanent Magnet Generator for Variable Speed Wind Turbines, in Proc.1995 IEE Electrical Machines and Drives, Seventh International Conference, pp. 453 457. E. Muljadi, S. Drouilhet, R. Holz, and V. Gevorgian, Analysis of Permanent Magnet Generator for Wind Power Battery Charging, in Proc. 1996 IEEE Industry Applications Conference, pp. 541 548. Y. Tang and L. Xu, Flexible Active and Reactive Power Control Strategy for a Variable Speed Constant Frequency Generating z system, in Proc. 1993 IEEE Power electronics Specialists Conference, pp. 568 573.

: Fig. 5.2 Wind speed v/s power curve TABLE II: GENERATED E.M.F AT VARIOUS BLADE SPEED

[2].

[3]. Blade speed (r .p.m) 5 10 15 20 25 30 35 40 45 50 55 60 65 E.m.f ( volts) 1 2 3 4 5 6 7 8 9 10 11 12 13

[4].

[5].

[6].

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

259

S. Song, S. Kang, and N. Hahm, Implementation and Control of Grid connected AC-DC-AC Power Converter for Variable Speed Wind Energy Conversion 141System, in Proc. 2003 IEEE Applied Power electronics Conference and Exposition, pp. 54158, vol.1. [8]. F. Martinez Rodrigo, J. M. Ruiz Gonzalez, J. A. Dominguez Vazquez and L. C Herrero de Lucas, Sensor less control of a squirrel cage induction generator to track the peak power in a wind turbine, in Proc. 2002 IEEE Industrial Electronics Society Conference, pp.169 -174 vol. 1. [9]. R. Datta and V. T. Rang Nathan, A Method of Tracking the Peak Power Points for a Variable Speed Wind Energy Conversion System, IEEE Trans. On Energy Conversion, Mar. 2003, pp. 163 168, vol. 18. [10]. Y. Higuchi, N. Yamamura, M. Ishida, and T. Hori, An Improvement of Performance for Small-Scaled Wind Power Generating System with Permanent Magnet Type Synchronous Generator, in Proc. 2000 IEEE Industrial Electronics Society Conference, pp.1037-1043, vol. 2. [7].

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

APPENDIX SPECIFICATIONS Number of poles =8 Rated power =80W Rated speed =60 Rpm Rated voltage = 12 V Air gap length, mm =0.6 No of coils =6 maximum current capacity =10 A No of turns/coil =800 Resistance/coil =17 Total no of magnets =32 Magnet dimensions, =50 x 50 x 10 mm Flux density =1.5 Wb/m2 flux = 3.75 mWb Total height of structure =1600mm Diameter of rotor =300mm

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

260

Optimal Placement of UPFC in Transmission Line in Deregulated Environment


1
1, 2
1

Raminder Kaur, 2Tarlochan Kaur


2

PEC University of Technology, Chandigarh, India.


tarlochan.kaur@gmail.com

raminder84@gmail.com,

Abstract - In this paper a new approach to find the optimal location and size of UPFC in a deregulated power system has been proposed. In a deregulated market, transmission system being an independent entity, catering to a number of independent generating units. Placement of UPFC at ideal position help make the system more efficient and reliable. In this paper, an elicit non dominating genetic algorithm has been proposed for finding proper placing and sizing of UPFC, while maintaining the constraints. Keywords: Optimal location, deregulated power system, elicit non dominating genetic algorithm

I. INTRODUCTION The economic, social and legislative development, particularly the world wide movement towards deregulation, demand new concepts and technologies to best exploit the existing power system resources without compromising system security. Without any doubt the implementation of above objective requires effective power flow control methodologies. FACTS technology provides unique control flexibility and versatility [1-3]. In the present paper UPFC has been taken as FACTS device because UPFC, being most versatile and flexible, is an effective means for regulating voltage profile and power flow in modern power systems. Optimizing its location become a concern when coming to the practical implementation stage. A mathematical model has been developed to work out a pragmatic approach based on Multi-objective optimization using Evolutionary algorithm to determine the optimal location of UPFCs to be installed [4,5]. Investment considerations of UPFC and real power loss of the network concerned are taken as objective functions. The method takes into account the steady state security constraints. II. FORMULATION OF COST OF UPFC A UPFC should be designed with its shunt inverter rating capacity large enough to supply the reactive current Iq for upholding the bus voltage and active current I T, for satisfying requirement of series voltage source Vsi [6,7]. For the series inverter, its rating is defined by the thermal limit of the transmission line. Hence, the cost of the UPFC can be regarded as proportional to its rating which is related to the rated voltage and maximum current of the associated transmission line. From the mathematical modeling equations the apparent power of the shunt inverter Se can be obtained; Se = Vi (IT + jIq)* = ViIT + jViIq = Re(VsiIs*) + ViIq 1

Since (phase angle difference between both ends of transmission line) can vary from zero to , the real power passing through the series inverter is at the maximum value, regardless of the phase angle of Is where Is is the line current passing through secondary of series transformer. Thus, capacity of shunt component of the UPFC, can be obtained: = 2 By the same method, the capacity of the series component of the UPFC is = = Is max VsI max 3 From the equations 2 and 3, the parameter VSI max is seen to be mutually independent to Iqmax as they are not related; These equations also show that the MVA rating of the shunt inverter is large than that of the series inverter, I qmax represents the shunt reactive compensation function of the UPFC and VSI max together with represents the respective shift of voltage and phase angle function of the UPFC [8,9]. Hence, these values can be used to describe not only the capacity but also the function of the UPFC. Thereby, the installation cost of the UPFC is defined as: CUPFC =(F1+ l Vsl max + 1 Iql max) 4 Where F1 = Installation cost (constant value) 1 Vsl max = Cost relating to the capacity of series inverter 1 Iql max = Cost relating to the additional shunt reactive compensation capacity in the shunt inverter. 1 = represents the per unit voltage cost, determined by the maximum current of the transmission line (Rs x kA/MVA) 1= per unit current cost, determined by the rated voltage of the transmission line, (Rs y kV/MVA) III. MATHEMATICAL FORMULATION Objective Function Two objectives are being considered in a multiobjectie optimization problem. If ni be the total number of lines in system and nu be the number of UPFC placed for minimization of loss and mathematical problem is given as follows: Minimize 5 And PL the loss in one line, say 1 (between nodes i and j), is given by formula PL = gl [Vi2 + Vj2 - 2ViVj cos( i - j )] 6 So the second objective is Minimize 7

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

261

Operating Constraints Equality constraints : Power flow equations corresponding to both real and reactive power balance equations are the equality constraints which can be written, for all the buses except buses i and j in which UPFC is connected, as

9 For buses I and j, the equality constraints can be written as

10

11

12

13 Where Pi = Real power injection at bus - i Qi = Reactive power injection at bus - i Pgi = Real power generation at bus - i Qgi = Reactive Power generation at bus - I Pdi = Real Power load at bus - I Qdi = Reactive power load at bus = i Vi = Voltage magnitude at bus - i i = Load angle at bus - i Pig, Pig = Real power injection at bus i and j due to FACTS devices Qig, Qig = Reactive power injection at bus i and j due to FACTS devices Yij = gij + bij = i-jth element of Y - bus Matrix. Nq = Number of reactive power sources in the system. Inequality Constraints: (i) Reactive Power Generator Limit: Let Q gimin and Qgimax are the minimum and maximum reactive power generation limit of the reactive source generators (Nq) respectively. Qgimin < Qgi < Qgimax i = 1,2,..Nq 14 (ii) Voltage Limit: This includes the upper (Vimax) and lower (Vimin) limits on the bus voltage magnitude. min Vi < Vi < Vimax i = 1,2,..Nb 15

(iii) Phase Angle Limits: The phase angle at each bus should be between lower min and upper imax i limits. imin < i < imax i = 1,2,..Nb 16 These limits may vary depending upon the problem under consideration. Imposing phase angle limits at load buses is another way of limiting the power flow in the transmission lines and as, for generator buses this is done for stability reasons. (iv) Line Flow Limit: These constraints represent the maximum power flow in a transmission line and are usually based on thermal and dynamic stability considerations. Let Plimax be the maximum active poor flow in line i respectively. The line flow limit can be written as Plimax > Plii = 1,2,..Ni 17 (v) UPFC control parameter limits: The voltage magnitude (VT) and phase angle ( T) of series voltage of UPFC must lie within limit. Mathematically, it can be written as, 0.5 > Vslmax > 0 Iq max > 0 < Vsl < Vslmax - Iq max < Iq < Iq max 0 < sl < 2 Mathematically equations (7 17) can be written as Min F1 (x,u,p) Min F2 (x,u,p) Subject to H(x,u,p) = 0 (equality constraints) g(x,u,p) > 0 (inequality constraints) Where Control variables = x = Vsim, Iqmax, Vsi, sl & Iq State variables = u = Pgj, Qgi, Pig, Qig, Qjg Fixed parameters = P = Pdi, Qdi A proposed algorithm for optimal placing and sizing of UPFC Genetic algorithm is a stochastic search algorithm [1012]. In a genetic algorithm, the decision variables are coded into a gene string according to the coding design. Each gene represents a solution to the problem. The genes are evaluated through a predefined evaluation function, an selected to generate new genes through gene operation for a new iteration. The iteration stops when the specified criteria are satisfied. At the same time, the optimal solution set is obtained [13-15]. An efficient algorithm is devised, which combines the power searching mechanism of genetic algorithm with the mathematical foundations for global optimization. The proposed algorithm is :Step 1 : Combine parent and offspring population and cerate Ri = PiUQ1. Perform a non-dominated sorting to Ri and identity different fronts Fi, i=1,2.etc. Step 2 : Set new population Pi+1 = 0. Set a counter i=1 Until + < N perform Pt+1 = Pt+1 UPi and i=i+1

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

262

Step 3 : Perform the crowding sort procedure and include the most widely spread (N) solutions by using the crowding distance values in the sorted Fi to Pt+1. Step 4 : Create offspring population Qi+1 from Pi+1 by using the crowded tournament selection, crossover and mutation. Simulation Result This section presents the computation results on IEEE 14 bus 5 generator system as shown in Fig1.

8 9 10 11 12 13 14 2 3 4

07.6 47.8 09.0 03.5 06.1 13.5 14.9 21.7 11.2 94.2

1.6 -3.9 5.8 1.8 1.6 5.8 5.0 12.7 7.5 19.0

TABLE IV: TRANSFORMER DATA


Line No. From bus To bus Series impedance Resistance Reactance Taps Rating (p.u.)

1 2 3 Fig. 1: IEEE 14 bus system TABLE L: GENERATOR POWER DATA


Line

8 9 9

3 6 7

0.00001 0.00001 0.00001

0.25202 0.20912 0.55618

0.962 0.978 0.969

0.50 0.50 0.50

TABLE V : LINE DATA


From bus To bus Series impedance (p.u.) Resistance Reactance Half line charging Rating (p.u.) No.

Bus No

Real Power Max Min

Reactive Power Max Min

Specified voltage (pu) 1.060 1.045 1.070 1.010 1.090

4 5 6 7 8 9

1 2 4 9 1 2 6 2 6 7 3 3 3 7 10 12 13

8 8 9 8 2 4 5 9 7 10 11 12 13 14 11 13 14

0.05403 0.05695 0.06701 0.01335 0.01938 0.04699 0.00001 0.05811 0.00001 0.03181 0.09498 0.12291 0.06615 0.12711 0.08205 0.22092 0.17093

0.22304 0.17388 0.17103 0.04211 0.05917 0.19797 0.17615 0.17632 0.11001 0.08450 0.19890 0.25581 0.13027 0.27038 0.19207 0.19988 0.34802

0.02460 0.01700 0.01730 0.00640 0.02640 0.02190 0.0000 0.01870 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000

1.00 0.50 0.50 1.00 1.00 1.00 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50

(MW) (MW ) (MVAR) (MVAR) 1 2 3 4 5 250 200 200 250 000 50 200 000 000 000 100 50 24 40 24 -45.0 -40.0 -06.0 0 -06.0

10 11 12 13

TABLE II : REACTOR / CAPACITOR DATA

14 15 16 17 18 19

Bus No.

Impedance Resistance Reactance -5.263

0.000
TABLE III : LOAD BUS DATA

20

Bus no. Real (MW)

Load Reactive (MVAR) VI. RESULTS Table 5 shows the optimal placement and size of the UPSC to be placed in the bus system in different lines following all the constraints of line flows, bus voltage level, reactive power generation limits etc. as shown in the Tables 1-4. This leads to improvement in power quality,

6 7

000.0 029.5

000.0 16.6

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

263

increased secure loading with higher stability limits for the economic operation of deregulated power systems. VII. CONCLUSIONS The proposed algorithm combines the powerful searching mechanism of genetic algorithm with the mathematical foundations of global optimization. A multi objective formulation for placing and sizing of UPFC has been presented keeping in view the operation constraints such as maximum line flaws, bus voltage levels, reactive power generation limits of generations. It is seen that with the proper placing and size of UPFC, the transmission efficiency and reliability is increased. REFERENCES
[1]. Preecha P. and Srivastava S.C., "Optimal Reactive Power Dispatch Considering FACTS Devices," EPRI, vol. 46, issue 13, pp. 251257, sept. 1998. [2]. W.L. Fang and H.W. Ngan, "Optimizing Location of Unified Power Flow Controllers using the method of Augmented Lagrange multipliers," Proc. IEE, Part-C, Vol. 146, No. 5, September, 1999, pp. 428-434. [3]. N.G. Hingorani, "Role of FACTS in a Deregulated Market," IEEE Power Engineering Society, Summer Meeting, Vol.3, 2000, pp. 1463-1467. [4]. A.Edris,, "FACTS Technology Development: An update," IEEE Power Engineering Review, March 2000, pp 4- 11. [5]. A.J.F. Ken, X. Lambard, A. EDris, A.S. Kehraban, A. Elate, "UPFC: Modeling and Analysis," IEEE Trans on Power Delivery, Vol. 14, No.2, April 1999 pp.648- 654. April 2000, pp 53-59. [6]. S.N. Singh and A.K. David, "A New Approach for placement of FACTS devices in Open Power markets," IEEE Power Engineering Review, September 2001, pp. 58-60. [7]. S.N. Singh and A.K. David, "Optimal Location of FACTS Devices for Congestion Management", Electric Power Systems Research, Vol. 58, No.2, July 2001, pp. 71 79. [8]. T.T. Lie and W. Deng, "Optimal Flexible AC Transmission Systems (FACTS) devices Allocation," International Journal of Electrical Power and Energy systems. Vol. 19. No.2, pp.125-134. [9]. K.S. Verma, S.N. Singh and H.O. Gupta, "Optimal Location of UPFC for Congestion Management," Electric Power Systems Research, Vol. 58, No.2, July 2001, pp.89-96. [10]. F.A. Rahimi and A. Vojdani, Meet the Emerging Transmission Market Segments, IEEE Computer Applications in Power, Vol. 12, No. 1, pp. 26-32, Jan. 1999. [11]. H. Rudnick, R. Varela, and W. Hogan, Evaluation of Alternatives for Power System Coordination and Pooling in a Competitive Environment, IEEE Trans. Power Systems, Vol. 12, No. 2, pp. 605 613, May 1997. [12]. J.A momoh, R.J Koessler, M.S Bond, B.Stott, D.Sun, A.Papalexopoulos, P. Ristanovic, Challenges to optimal power flow, IEEE Power Engineering Society Vol 12,No.1, pp.445-455 February 1997. [13]. G. Huang, Q. Zhao, A Multi-Objective Optimization Solution for Deregulated Power Markets, IEEE SM99, Canada. [14]. R Ouiddir, M Rahli & L Abdelhakem-Koridak, Economic Dispatch using a Genetic Algorithm: Application to Western Algerias Electrical Power Network Power Systems Optimization Laboratory, Faculty of Electrical Engineering University of Science and Technology of Oran: Journal of information science and engineering, Vol21, pp. 659-668 (2005). [15]. K.S.Pandya, S.K.Joshi. A Survey of optimal power flow methods, Journal of Theoretical and Applied Information Technology, pp.450-458 (2005-2008).

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

264

Cross Border Power Trading Model for South Asian Competitive Electricity Market
1

Sumit Saroha, 2Rohit Verma, 3Rajnesh Pawar


1,3
1

DCRUST, (Murthal) Sonepat, Haryana, India National Power Training Institute, Faridabad , Haryana, India
saroha_sumit0178@yahoo.com

Abstract South Asian countries continue to be characterised by low per capita consumption of energy, poor quality of energy infrastructure, skewed distribution and inaccessible and costly energy availability. Economic gains based on regional cooperation in the energy sector have become a firmly established practice across the world. Although a number of market models are prevalent in the International arena, the same could not be directly adopted for South Asian cross border power market. Rather, a careful study of the existing models is to be carried out to evolve a model suitable to Indian conditions, which is easy to adopt, sustain and to take care of existing participants. Complexity of electricity market in cross border international model scenario has been analyzed in this paper & a model for South Asian cross border power trading has been proposed. Demand and supply side bidding scenarios with case studies are presented in the paper. This paper is focused on integrating the energy Market in South Asia, examining the opportunities, benefits, and challenges for transparent price discovery of power in South Asia. KeywordsPower Tariff, Electricity Market, Power Pool, Market Clearing Price.

I. INTRODUCTION Electricity Industry, throughout the world is currently undergoing restructuring. Countries are adopting the deregulated industry structure for better utilization of the resources and for providing choice & quality service to the consumers at economical prices [1] resulting in transparent price discovery. The bordering countries of India have abundant energy resources and unequal geographical distribution of energy resources for generating electric power. In these countries, various resources of power generation are unevenly distributed and power demand pattern of household and industrial consumers is also different [2]. South Asian Association for Regional Cooperation (SAARC) in which Bangladesh, Bhutan Nepal, India, Pakistan, Maldives and Sri Lanka are member countries, had agreed for a regional free trade agreement. This concept was further sponsored by USAID (United State Agency for International Development) for the South Asian regional energy transmission partnership. Many other countries like China, South Africa, Nord Pool, USA, Canada, countries in West Africa etc. are also working on cross border power trading [1], [3], [9], [13]-[16]. Lee and Lee [3] had developed a Cross-border Power Trading System for West African Power Pool. Bajpai and

S.N. Singh [1] developed an Electric Power Trading Model for Indian Electricity Market but none of the researchers have developed a cross border power trading model for India. In cross border power markets there are inter relationships between mechanisms for congestion management, transit and trading. Each Transmission System Operators (TSO) maintains power system security within its own area, but differs in its impact on the economics of the energy market [5]-[6]. A concept of Independent System Operator (ISO) is introduced in this paper for coordination between all participating countries TSOs. In this paper design concept of cross border power trading model in competitive power market has been introduced by taking consideration of major issues faced by present power traders in Indian electricity market. This paper focuses on trading arrangements, operation of power exchange and effectiveness of proposed pricing mechanism which is tested on linear demand and supply side bidding scenarios of market model. Suitable mathematical models are developed for calculations of market clearing price (MCP) and matrix laboratory version 7.5 software is used for the MCP simulations. This paper could be guide line for the policy makers, power systems designers and market operators to promote the cross border power trade in South Asian countries with system reliability and security. II. REQUIREMENT ANALYSIS SACPT (South Asian Cross border Power Trading) model is created to coordinate multi-national power system operations into a unified regional electricity market. The main objective of SACPT is to match the demand and supply of South Asian countries, with quality and reliability of power supply, through the development and realization of key infrastructure that would permit accessibility to economic energy resources, for all member countries of South Asian region. Presently India trades electric power from Bhutan through bilateral contract. To facilitate international power exchange in South Asian region countries are planning high capacity transmission lines. A. SACPT PX Objective Your The SACPT Power Exchange will promote operational coordination and information sharing regarding power generation and demand forecasting

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

265

between SACPT region members. Its proposed objectives are: To make Indian cross border power trading model for competitive electricity market To analyze and propose the pricing mechanism for the competitive electricity market To provide guideline for the transmission system design To provide transparent price discovery of electricity B. Present and future scenario Nepal, Pakistan, India and Bhutan have a potential of hydro energy resource of approximately 83000 MW, 21000 MW, 84000 MW and 21000 MW respectively [2]. Besides this Myanmar and Bangladesh have significant unexploited gas reserves. India currently trades power from Bhutan through bilateral contracts. The power generating capacity of Bhutan is relatively small and most of the generation is exported to India. C. Assumptions For the operational coordination and information sharing between SACPT member countries, it is essential to have real time monitoring and data exchanging facilities in the SACPT power exchange. SACPT-PX and independent system operator (ISO) are the source of the operational data. ISO will coordinate between the TSOs of the respective countries. The assumptions in the SACPT- PX evolution estimation are mainly based on the availability of data via the data link and the introduction of the regional electricity market in South Asian region. Therefore, it is necessary that all the TSOs should have the capability to provide the data to PX via the data communication. It is assumed that there would be a stepby-step evolution from the current situation to a final competitive regional electricity market. After completion of the energy management system (EMS) upgrade or renovation projects in ISO, it is assumed that the current offline data is going to be managed in the EMS or, at least, managed in the electronic form which is transferable via the data link. In the electricity market environment, the power trading and settlement is basically dependant on its market design and market rule. It is assumed that the participating entities in this International power trade have non discriminatory open access. III. DESIGN CONCEPTS Electricity Act 2003 of India was enacted in June 2003 which made open access provisions. Cross-border electricity trading with India is limited. Electricity trade in India is restricted to wheeling and selling power to states within India. Moreover, cross-border power trading is done only through Power Trading Corporation of India (PTC), through bulk power transmission services agreements with Power Grid Corporation of India Ltd. (PGCIL) [15]. The Electricity Act 2003 has opened up energy trading within India by authorizing the Central Electricity

Regulatory Commission (CERC) and State Electricity Regulatory Commission (SERCs) to grant licenses valid for 25 years to any person to undertake trading in electricity as an electricity trader. Also, in a related order by CERC, PGCIL as the central transmission utility (CTU) was ordered to provide open access in transmission and distribution. The National Load Dispatch Center (NLDC) has also been conceived to take care of interregional and cross boarder exchanges [12]. NLDC will act as TSO for India and similar TSOs will be there in the participating countries. ISO will coordinate between TSOs. A number of power traders from the member countries will be participating in this cross-border power trading arrangement and majority of power will be traded on SACPT PX which will give the benchmark for power pricing [1], [3]. Fig 1 shows the SACPT - PX model:

Fig. 1: South Asian Cross Border Power Trading Model

D. Open Access to Transmission Services A robust transmission network is essential for power market operations such that market reforms should start with transmission and not generation. To meet the needs of cross-border power trade, as is the case with all new power markets, the adequacy and open access of transmission facilities is critical for market efficiency. By ensuring open access, a competitive environment is created that allows generators and traders to choose their customers, and vice versa. To realize the potential of regional power trade, India should extend the principles of open access and licensing of power traders to cross-border trading, and each trading country in the region should develop and enact similar open access rules consistent with their regulatory regimes that draw upon Indias experience in restructuring policy guidelines regarding network planning, investment, and access issues and pricing issues. A common regulatory framework involving trading countries in the region needs to be formulated for determining the structure and level for network charges and common operational rules of transmission system operators, such as requisite metering and communication

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

266

arrangements, revised technical codes for interconnection, and grid codes for power scheduling and dispatch [15]. E. Common Power Market Design A common power market design comprises a core and common set of philosophies translated to standard rules, procedures and products designed to create a seamless transmission system across multiple wholesale and retail markets and different regulatory jurisdictions across the countries. This design must facilitate implementation and must promote reliability, enhance efficiency and offer non-discriminatory access, whilst making access more transparent by offering common standards and validation. For cross-border power trading to be competitive and to ensure the harmonious operation of cross-border power markets, a common market design must be developed and applied on a regional basis, a common set of rules and procedures for governing wholesale energy markets must first be developed that facilitates implementation in multiple markets across different countries, ultimately leading to seamless competitive power markets across the region [15]. In this paper it has been assumed that all participating companies are on similar regulatory platform. F. Power Exchange A power exchange as a supplementary market for day ahead balancing is proposed for promotion of inter regional trading and maximum utilization of available resources. The proposed PX concept is cross border competitive electricity market where all eligible participants will submit portfolio based bids for each hour on Day-Ahead (DA) basis. The PX characterized by hourly cleared physical delivery market should have uniform price double auction system with option of linear bidding. The existing PPAs should be honoured, means PX should be only a supplementary market to match unallocated surplus with likely requirements through DA auction based market. Fig1 illustrates a simplified, theoretical model of the proposed cross border power market having different interactions between the relevant actors that participate in electricity market. The financial flows that result from electricity trade is referred as commercial transactions and the physical electricity flows referred as physical transactions. All the market participants will electronically submit their bids and offers to the PX and PX will clear the market by determining Market Clearing Price (MCP) from intersection of aggregated supply and demand curves. If this market settlement causes congestion in transmission networks then PX splits the market and determines area wise MCP with transmission constraints settlement. The PX will get the available transmission capacity (ATC) information from ISO & TSOs on DA basis for congestion management and prepares the final DA schedule for the transactions to be cleared. Then PX submits this schedule to ISO for real-time implementation. Next day 00h00 to

24h00 electricity will be delivered and financial clearing can be done once a week. Role and Responsibilities: In decentralized trading arrangement, PX has to be a separate entity from system operator, as an independent market operator. But PX should have co-ordination with ISO for ATC information and ISO has coordination with TSOs for the dispatch of DA schedules and imbalance settlement through balancing power market. In the proposed DA auction based market, PX should have time specific flow of activities which is shown in Fig2 below [1, 9]:

Fig. 2: Flow of Activity in PX

G. System Operator Model Independent system operator (ISO) model is better option for restructured electricity market in India. The unbundling of transmission and system operation is essentially required to address the possibility of the transmission utility exerting market power through the system operator. Common management of two functions could potentially result in conflict of interest. Sometimes there is conflict of interest between NLDC and CTU on the issue of maintenance scheduling of transmission lines. Similar instances may happen in other participating countries. While option exists under EA 2003, to organizationally separate the transmission and system operation function, this can be deferred for the short term. In effect the system operation function should operate as a separate division within transmission utility till these are organizationally separated. It is preferable that power market model should be established first before separation of transmission and system operation is undertaken [1]. H. Congestion Management When the schedule of power flow in a particular transmission corridor discovered in power exchange is more than the transfer capacity of that corridor, then

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

267

congestion is said to occur. The entire power scheduled cannot flow in the line, as it would endanger the entire transmission system. When congestion occurs, area prices across the system are different. Then the higher cost generators within the same area have to run, because a contingency or transmission line makes the lowest cost generators in others areas unable to supply load [6]. During congestion Transmission System Operators (TSO) operating in connected local areas can coordinate cross border trade by re dispatching or counter trading and are responsible for the security of local networks [5]. IV. SIMULATION OF MCP & MCV The MCP (Market Clearing Price) is the lowest price obtained at the point of intersection of aggregated supply and demand curves and volume of power at the point of intersection is called MCV (Market Clearing Volume). At this price both suppliers of generation and customers are satisfied and would provide enough electricity from accepted sales bids to satisfy all the accepted purchase bids. At MCP, total sales bids in their merit order would be equal to the total purchase bids down to that price in their merit order. That price would be the MCP. Two types of markets exist based on the bidding mechanism. If bidding is done only by the suppliers, it is termed as a single-sided bidding, whereas, if both suppliers and customers are allowed to bid into the market, it is known as a double-sided bidding mechanism. The bidders can be allowed to bid their outputs or demands in the linear form [4], as shown in Fig3

If there are Ns no. of suppliers who bid into market, the combined supply curve will be given by the following equation.

For the fixed demand D, the MCP will be obtained by following equation.

In this, it is assumed that bidders have enough capacity. Capacity of any individual supplier can be included accordingly. If the ith generator has the minimum (qi min) and maximum (qi max) power output limits & u is unit function, the supply curve defined in equation (1) will become If the capacity limit both minimum generation (Qmin) and maximum generation (Qmax) is specified then combined supply curve would be

Where functions u(Qi-Qmin) = 1, when QQmin = 0, when Q<Qmin And u(Q-Qmax) = 1, when QQmax = 0, when Q<Qmax B. Double Side Bidding In the single side bidding the demand bid is inelastic but in the double bid market, elasticity of demand curve has been considered. Both supply side and demand side bidding are taken into account for determination of market clearing price (MCP). Demand curve for any individual in linear bid mode can be expressed as: Di(p) = mdi is the slope of the demand curve and p i0 is the intercept of price axis as shown in Fig 3. If there are Nd no. of demand bidders who take part into bidding then the combined demand curve will be D(p) = The market clearing price (MCP) can be obtained by solving the following equation.

Fig. 3: Linear Demand and Supply curve

A. Single Side Bidding The bidding is done only by the suppliers and the demand is fixed then this is called single side bidding. In a linear bid model, a supply curve that is a function of the market price (MCP) of any bidder i can be expressed as: Where, msi is the slope of supply curve as shown in Fig3, Qi(p) is quantity of the ith supply curve, p is the price in cents/KWh.

V. STUDY CASES In this different bidding scenarios of cross border power in the electricity market are studied. In this study, uniform pricing approach is considered. For the simplicity only one supply bidder is taken from each country other than the power taken from India during the congestion. The output part of the Indian power is 20 KW.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

268

TABLE I: LINEAR BID DATA

Msi, Qgmax Qgmin cents/KWh2 Bidder 1 (Bhutan) 0.10 70 10 Bidder 2 (Nepal) 0.15 50 10 Bidder3 (Bangladesh) 0.20 60 10 Bidder 4 (Myanmar) 0.25 50 10 Bidder 5 (Sri Lanka) 0.30 65 10 Bidder 6 (Pakistan) 0.32 40 10 The Table I shows the linear supply bid data of each country of Indian cross border power trading region. In the study following bidding strategies have been considered: Case-1: Linear supply bid with fixed demand (i.e., single sided bid market) Case-2: Linear supply bid with linear demand bid (i.e., double sided bid market) CASE-1 In this case, a constant demand of 150 KW is considered. Analysis is performed for the following cases: (a) First, the power generated from India is not available. Demand is met only by the other six cross border locations. MCP is obtained from the intersection of the cumulative supply curve of bidders 1, 2, 3, 4, 5, 6 respectively and fixed demand line of 150 KW, as shown in Fig4 based on their high capital cost & Maintenance cost.

Supply Bidders

Qi = (i) Impact of bidding of Indian Generator: The congestion into the system is uncertain so output variation of Indian power as well as MCP with the different bidding rates are shown in Fig 5.

Fig. 5: MCP at Diff. Bid Rate

There are two situations shown in the Fig5 (i) Restricted Indian Generator: If the Indian gen. bid at zero, it will be completely dispatched and MCP will be 4.6693 cents/KWh. (ii) Unrestricted Indian Power: The MCP with restricted Indian power (i.e., 20 KW) is 4.0467 cents /KWh which is not same as MCP at the unrestricted Indian power. Maximum and minimum MCP can be found out by Equation (5) with respect to bidding rate (ms) and then equating to zero. For various bidding rate (ms) varying from 0 to 10, payments and output are shown in the Table II for the following two options: Option 1: with fixed Indian Power of 20 KW, MCP=4.0467cents/KWh when ms <1 Option 2: When ms between 1 and 10, MCP calculated without Indian Power and output of bidders 1, 2, 3, 4, 5, 6 as adjusted by Eq. (9) are shown in Table II.
TABLE II: OUTPUT AND PAYMENT AT DIFFERENT CASES Bidders Output Payment Output Payment Payment KW At At 4.669 KW At At 4.046 At 4.669 4.669 4.046
Bidder-1 Bidder-2 Bidder-3 Bidder-4 Bidder-5 Bidder-6 Indian Generator Total ms<1 46.69 31.12 23.34 18.67 15.56 14.59 0 150 218.02 145.34 109.01 87.208 72.673 68.131 0 700.38 1<ms<10 40.4669 26.9780 20.2335 16.1868 13.4890 12.6459 20 150 163.75 109.17 81.87 65.50 54.58 51.17 80.934 607.00 188.93 125.96 94.47 75.576 62.980 59.043 93.38 700.38

Fig. 4: Linear Supply Curve

As from Fig4 the MCP= 4.6693 Power supplied by each generator to meet the demand of 150 KW is shown in Table II. (b) Now consider the power generated from India is available upto 20 KW and partly demand is met by the other six cross border locations. Therefore, Indian generation only reduces the bidders (i.e., 1, 2, 3, 4, 5, 6) total dispatch from 150 KW to 130 KW. The corresponding MCP is now 4.0467 cents /KWh (as per Fig 5). Since MCP is reduced, it may not be possible to recover the cost of Indian generator hence output of bidders would be reduced by

CASE- 2 In the competitive electricity market, instead of taking the constant demand, the demand side participants are also allowed to bid into the market with their demand elasticity. Let three different demand bidders are considered in this paper for simplicity and their bid data are given in Table III.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

269

TABLE III : LINEAR DEMAND BID DATA

Mdi(Cents/KWh2 Poi(Cents/KWh) ) Bidder 1 0.16 14 Bidder 2 0.18 16 Bidder 3 0.20 18 The MCP, which is the intersection of cumulative supply and total demand curves as shown in Fig 6, is found to be 5.4442 cents/KWh. The corresponding individual supply and demand curve is shown in Fig 6. Consumers

It can be noted that bidding rate less than 1.0 will have no impact on the MCP with restricted Indian power of 20 KW (considered maximum output). However, with higher bidding rate, the MCP increases but the power dispatched from the Indian power is less. The payment at various MCP and corresponding output is shown into Table IV. VI. RESULT & DISCUSSION Case 1: There are six bidders one bidder from each country, these bidders meet a fixed demand of 150 KW and the MCP will be 4.6693 cents as shown in Fig 5. Table II shows the bidding quantity of each supply bidder and the total amount of money at MCP 4.6693 cents is 700.38 cents. When 20 KW power taken from a local Indian generator then the MCP will be reduced 4.046 cents as shown in Fig 5 and local area price will be 4.046. The total quantity supplied by each supply bidder is also reduced and the total amount of money at MCP of 4.046 cents is 607.004 cents. If we consider the supply quantity at Unrestricted Indian generator power to meet the demand of 150 KW and MCP of Restricted Indian generator power then total amount of money paid to supply bidders is 700.38 cents as shown in Table II. There is a possibility of monopoly into the single sided market bidding. Case 2: there are six bidders one bidder from each country; these supply bidders meet the variable demand of three demand bidders. Table IV shows the bidding quantity of each supply bidder and total demand of 174.8957 KW bidders in this case MCP is 5.4442 cents as shown in Fig7; the total amount of money at MCP 5.4442 cents is 952.1712 cents. When 20 KW power taken from a local Indian generator then the MCP will be reduced 5.0355 cents as shown in Fig 7 and local area price will be 5.0355. The total quantity supplied by each supply bidder is also reduced but overall demand & supply is increase as shown in Table IV is 181.7648 KW and the total amount of money at MCP of 5.0355 cents is 915.2748 cents. If we consider the supply quantity at Unrestricted Indian generator power to meet the demand of 181.7648 KW and MCP of Restricted Indian generator power then total amount of money paid to supply bidders is 989.5637 cents as shown in Table IV. There is less chance of monopoly in double side bidding as compare to single side bidding. So this will result in transparent price discovery. The power taken from Indian generator will also reduce the possibility of congestion into SACPT region. VII. CONCLUSION The growth of power demand in India is very high. So the proposed SACPT model will try to meet the Indian power demand by promoting competition into the Indian cross border region and result in better demand side management in the region. Besides this it will help the poor developing countries in its neighbourhood to better utilize its resources and earn foreign exchange. Power Exchange as a market operator will operate the proposed Day Ahead auction market with uniform price

Fig. 6: Linear Supply and Demand Curve

Demand of three bidders:- Bidder 1: 53.47 KW, Bidder 2: 58.64KW and Bidder 3: 62.77KW. If Indian generator do not participating into bidding then the MCP will be the intersection of a new supply curve parallel below (20 KW in x-axis) to the cumulative supply curve of six supply bidders. It is similar to the intersection of a parallel line below the demand curve to the total supply curve without the Indian power. The MCP is found to be 5.0355 Cents/KWh. Mathematically; it can also be obtained as:

Due to the elastiity of demand bidders, there is reduction in the spot price which causes the increased in the consumption of demand bidders and it is Bidder 1: 56.0282, Bidder2: 60.9140, Bidder3: 64.8226. Total demand is 181.7648. The effect on the variation of bidding of Indian generator on Indian output and the MCP is shown in Fig 7

Fig. 7: MCP at Diff. Bid Rate

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

270

clearing in a price area. PX will provide equal and sametime information to all the market participants through online bulletin board. Power exchange, as institution, is able to pool risks and can create liquidity (through physical and financial products), thereby limiting the adverse exposure to traders within reasonable limits. In addition, such institution facilitates efficient clearing and settlement in cooperation with the system operator. This paper also presents the various market strategies in cross border competitive power market. A proper use of such model can avoid the congestion in competitive power market by which efficiency and reliability of the market can be increased. The power taken from the cross border can play a vital role in mitigation of market power, ancillary services but their cost must be recovered for successful promotion of cross border power trading energy. This paper could be guide line for the policy makers, power systems designers and market operators to promote the cross border power market with system reliability and security. REFERENCES
[1] P. Bajpai and S. N. Singh, An Electric Power Trading Model for Indian Electricity Market Power Engineering Society General Meeting, IEEE 2006. [2] Potential and Prospects for Regional Energy Trade in the South Asia Region, report 334/08, Energy Sector Management Assistance Program/South Asia Regional Cooperation Program, South Asia Regional Cooperation Program, March 2008. [3] Jinsu Lee, Hyosang Lee, and Sunghak Kim, Designing the Crossborder Power Trading System for West African Power Pool Transmission & Distribution conference and Exposition IEEE T&D Asia, 2009. [4] S. N. Singh and I. Erlich, Strategies for Wind Power Trading in Competitive Electricity Markets IEEE Transactions on Energy Conversion, Vol. 23, No. 1, pp. 249-256, March 2008. [5] Lucia Parisio and Bruno Bosco, Electricity prices and cross-border trade: Volume and strategy effects Science Direct Energy Economics, vol. 30, issue 4, pp. 1760-1775, 2008.

[6] I. Androcec, I. Wangensteen, S. Krajcar, Impact of Cross-Border Electricity Trading on Market Participants Power Engineering conference Lisbon, Portugal, March 18-20,IEEE 2009. [7] M. Shahidepour and A.Srivastava, Restructuring choices for the Indian Power sector IEEE Power Engineering Review, Nov 2002. [8] Ning Lu, Joe H. Chow, and Alan A. Desrochers, Pumped-Storage Hydro-Turbine Bidding Strategies in a Competitive Electricity Market IEEE Transactions on Power Systems, Vol. 19, No. 2, May 2004. [9]. Nils Flatab, Gerard Doorman, Ove S. Grande, Hans Randen, and Ivar Wangensteen, Experience With the Nord Pool Design and Implementation IEEE Transaction on Power Systems, Vol. 18, No. 2,pp 541-547, MAY 2003. [10] S.N. Singh, S.C. Srivastava, Electric power industry restructuring in India: present scenario and future prospect, in Proc. of IEEE International Conference on Electric Utility Deregulation, Restructuring and Power Technologies (DRPT 2004), vol.1, pp.2023, April 2004. [11] Walter reinisch and Tetuo Tezuka, Market Power and Trading Strategies on the Electricity Market: A Market Design View IEEE Transactions on Power System, Vol. 21, No.3, August 2006. [12] Subrata Mukhopadhyay, and Sudhindra K Dube, Status of Power Exchange in India: Trading, Scheduling, and Real Time Operation of Regional Grids Power Engineering Society General Meeting, IEEE 2005. [13] USAID SARI/Energy Program, Indias Electricity Act, 2003 Implications for Regional Electricity Trade Volume I, September 2004. [14] USAID SARI/Energy Program, The Four Borders Project: Reliability Improvement and Power Transfer in South Asia, Nov. 2001. [15] USAID SARI/Energy Program, Regional Energy and Trade Laws in South Asia Volume I September 2004. [16] Dr. Mahendra P Lama, Dr. Mohan Man Sainju and Dr. QK Ahmad, Reforms and Power Sector in South Asia: Scope and Challenges for Cross Border Trade, October 2002.

TABLE IV: PAYMENTS AT VARIOUS MCP AND CORRESPONDING OUTPUT

Bidders

Bidder-1 Bidder-2 Bidder-3 Bidder-4 Bidder-5 Bidder-6 Indian Generator Total Demand Side Bidder 1 Bidder 2 Bidder 3 Total

Generator Output, Payment in KW At MCP 5.44 Cents At MCP 5.4442 54.4422 296.3987 36.2984 197.5971 27.2211 148.1979 21.7769 118.5583 18.1474 98.7986 17.0132 92.7287 0 0 174.8957 53.4736 58.6432 62.7789 174.8957 952.171 291.1223 319.266 341.782 952.171

Generator Output, KW At MCP5.03 50.3548 33.5699 25.1774 20.1419 16.7849 15.7359 20 181.7648 56.0282 60.9140 64.8226 181.7648

Payment At MCP 5.0355 253.5607 169.0405 126.7804 101.4243 84.5202 79.2371 100.71 915.274 282.129 306.731 326.413 915.273

Option 2 Payment at 5.4442 274.1416 182.7612 137.0708 109.6565 91.3803 85.6693 108.884 989.563 305.028 331.627 352.907 989.563

MCP

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

271

Power System Transient Stability Improvement with Superconducting Magnetic Energy Storage
1

Rintu Khanna, 2Gurnam Singh, 3T.K.Nagsarkar


PEC University of Technology, Chandigarh, India 1 rintukhanna1@rediffmail.com superconducting coil active power absorbed /released by superconducting coil reactive power absorbed /released by superconducting coil electric MVA power absorbed /released by superconducting coil terminal voltage of synchronous machine magnitude of driving point admittance of the synchronous machine in p.u., admittance angle associated with Y11 in radians, magnitude of transfer admittance between the synchronous machine and infinite bus in p.u. admittance angle associated with Y12 in radians is the total computation time gain coefficient for rotor angle as control signal gain coefficient for rotor relative speed as control signal gain coefficient for change in terminal voltage of synchronous machine as control signal

1, 2,3

Abstract: This paper presents improvement of transient stability of power systems with superconducting magnetic energy storage (SMES). SMES has been connected at the generator terminal for single machine connected to infinite bus (SMIB) . A comparison of performance of proportional type of SMES (P-SMES) and that of an optimally switched conventional dynamic brake suggested by Minisey et al is presented. Transient stability investigations were carried out on SMIB with SMES for diverse loading conditions and different fault clearing times. The results established superior performance of SMES in damping transient swings over optimally switched conventional dynamic brake. Keywords: power system stabilization, optimally switched conventional dynamic brake, SMES, transient stability, rotor angular velocity control, AC-DC conversion, firing angle control, power control, power system, SMIB, critical fault clearing time

Psmes Qsmes Ssmes Vt


Y11

11

Y12

12

NOMENCLATURE Vdo Vd E the ideal no-load maximum DC voltage of the 6-pulse bridges average dc voltage impressed on the superconducting coil by the converter the voltage on the secondary side of the transformer or the bus voltage at which the SMES is connected firing delay angle of converter 1 firing delay angle of converter 2 estimated active power estimated reactive power may be equal to , p, acc change in voltage electrical power output of generator in p.u. at the instant of fault clearance turbine power in p.u. phase angle between E' and infinite bus voltage VB in electrical radians rotor angular velocity in electrical radians per second the steady state angular velocity in electrical radians per second 60 Hz , the nominal frequency per unit inertia constant in seconds admittance of superconducting coil susceptance of superconducting coil conductance of superconducting coil unidirectional dc current flowing in the

Tc kd
k kv

1 2

Pd Qd c v Pc PT
'

o fo fo
H Ysmes Bsmes Gsmes Id

I. INTRODUCTION The power systems are continuously growing in size and complexity in order to supply the increasing power demand. Accordingly transient stability of power system has become an important problem. In order to improve transient stability various measures such as discrete control of excitation systems, ultra-fast fault clearing, high speed re-closing, dynamic braking, Static Var Compensators (SVC), etc., are used. Since the successful commissioning test of BPA 30 MJ unit of superconducting storage (SMES) system [1], SMES has received much attention in power system application. SMES is a device consists of a superconducting coil, in which ac power is converted into dc by ac-dc converter and stored in superconducting coil in the form of magnetic energy. Its storage efficiency is as high as 90%. As SMES can be charged or discharged very rapidly through the semiconductor ac-dc converter system, it can be applied for both active and reactive power compensation for power system stabilization [2-3]. This paper is concerned with control method of a SMES for effective use of active and

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

272

reactive power modulation for power system stabilization during large disturbances. II SYSTEM MODELING WITH SMES FOR POWER SYSTEM STABILIZATION The inductor-converter unit of SMES consists of a dc superconducting inductor coil, a 12-pulse cascaded bridge type AC/DC converter and a star-delta/delta-delta step down transformer.

terminal voltage v, of the synchronous machine (synchronous machine to which SMES is connected) . The block diagram of control scheme is given in Fig 2. Thus the governing equations are: Pd = f1 ( c ) (5) Qd= f 2 ( v ) 2 S*= Pd2 Qd
1= 2=

(6) (7) (8)

f 3 (Pd, Qd)

(9) f 4 (Pd,Qd ) * The value of S gives the estimated electrical power of SMES when S* < Smax, then Psmes = Pd and Qsmes = Qd, whereas for S* > Smax , where Smax is the maximum MVA available in SMES.Pd = Pdmax, and the magnitude 2 of Qdmax can be calculated as Qdmax = Smax Pd2 .

Fig. 1: Conceptual Scheme of SMES Unit

Control of the converter firing angle enables the dc voltage Vd appearing across the inductor to be continuously varied between a wide range of positive and negative values. Such type of controller can be designed using Gate turn off thyristors (GTO). The converter dc output current Id being unidirectional, the control for the direction of power flow in the superconducting coil is achieved by continuously regulating the firing angle a. SMES unit can absorb and release electric power at high speed by the controlled operation of the AC-DC converter. The average dc voltage Vd of the DC side of the 12-pulse converter is expressed by Vd = Vdo ( cos 1 + cos 2) for, 0 < n < 360 0, n = 1,2 (1) where, Vdo= 3 2 E (2) Psmes=VdId=VdoId(cos 1+cos 2) Qsme=VdId(sin +sin 2) A
Control Scheme

Fig. 2: Control Scheme of SMES as Controller

(3) (4)

The control of DC power flow of converter is basically a function of power system variables. DC power flow can be controlled by firing angle of converters. The DC power of the converter can be varied with respect to any of the variables that reflect the state of the AC power system. During a disturbance, large or small, the variables affected are electrical power output, rotor acceleration /deceleration, system frequency (rotor velocity), rotor angle, voltage etc. The signals that are considered as input signals for the control circuit are (a) deviation in rotor angular velocity from the synchronous angular velocity, , (b) change in electrical active power output, P, of synchronous machine, (c) rotor acceleration, acc. (d) deviation in the

Controlling 1 and 2 the active and reactive power injected/absorbed by the SMES can simultaneously be controlled. Therefore, when the SMES is installed at the generator terminal of the power system, it can be utilized to damp out the rotor oscillations in a power system caused by large disturbances. B Governing Equations of Single Machine Connected to Infinite Bus Provided with SMES Equipped with Proportional Controller The following governing equations have been used for transient stability investigation of single machine connected to infinite bus (SMIB) equipped with proportional controller. A general schematic diagram of a simple power system consisting of a synchronous machine connected to infinite bus through a transformer and a double circuit transmission line is shown in Fig3. While carrying out the investigations the assumption are: (i) Turbine power is kept constant. (ii) A synchronous machine for transient stability studies is represented by constant transient internal emf E' behind direct-axis transient reactance x'd, which is a function of air-gap flux linkages.

Fig. 3: Schematic Representation of a Simple Power System

with SMES

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

273

The swing equation of a synchronous machine during fault condition may be represented in state space form as: d ' (10)
dt d dt
o

fo (P H t

Pe )

(11)

The swing equation of a synchronous machine during post- fault condition may be represented in state space form as: d ' (12)
dt d dt
Pe E
o

fo ( Pt H
2

Pe

Psmes )
11 )

(13)
12

Y11 cos(

E VB Y12 cos(

(14)

To insert for proportional type of controller with speed signal its input Pd and Qd are given by: Pd=k * (15) Qd=kv* v (16) (17) Psmes Qsmes
Ysmes Gsmes jBsmes Vt
2

this in mind the performance of the P-SMES is compared with that of an optimally switched conventionally dynamic brake suggested by Minisey et al [4]. As such the transient stability computations have been carried out with reference to the system given by Minisey [4]. It may be added that the system considered for investigations to find the effectiveness of a SMES in damping out transient swings is by Minisey [4] itself. Conventional dynamic brake is assumed to be connected to the terminals of the synchronous machine soon after the clearance of the fault and is disconnected after an optimum time interval determined in ref. [4]. For a fault clearance time of TFC = 0.2 sec, the optimal inclusion time of the dynamic brake with R = 5.5 per unit as reported by Minisey et.al [4] is 0.3 sec (or in other words the time of disconnection of the dynamic brake is TBR = 0.5 sec, the time measured from the time of incidence of fault). The corresponding curve for optimally switched dynamic brake are shown by curveb in Fig5 for TFC = 0.2 secs. and for TFC = 0.21667 secs in Fig6. C Comparison of System Performance with Optimally Switched Dynamic Brake and SMES Fig5 shows the comparison of swing curves in three cases namely without any stabilizing device, with optimally switched dynamic brake and with P-SMES. In Fig5 curve a corresponds to the case when no dynamic brake is applied to the system, curve-b corresponds to the case when conventional dynamic brake is applied for an optimal time. Curve-c shows swing curve with PSMES in operation with k= 0.07 and kv = 0.5. It may be seen from the curves of Fig5 and Fig6 that P-SMES offers a much better control in damping rotor oscillations caused by the fault compared to an optimally switched dynamic brake.

Vt

This relationship is used to obtain v signal at every step of numerical integration system differential equations in the mathematical model. The Eq. (10), Eq. (11) and Eq. (12), Eq. (13) may be solved using numerical methods such as modified Euler method ,Runge Kutta etc. III A SYSTEM CONSIDERED FOR TRANSIENT STABILITY INVESTIGATIONS

SMIB Schematic diagram of the sample system considered for stability studies with SMES connected to the generator terminal is shown in Fig4.

Fig . 4: Schematic Representation of the System Considered forTransient

Stability Investigations Infinite bus voltage is chosen as the reference phasor for transient stability investigations. With reference to Fig4 the system data in p.u. is given in Appendix I System Performance with Optimally Switched Dynamic Brake and SMES While attempting comparison with a conventional brake it is thought fit to compare it with the best designed conventional dynamic brake available. With B

Fig. 5: Performance comparison of Optimally Switched Dynamic Brake and SMES for at TFC = 0.20 secs

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

274

Fig. 6: Performance comparison of optimally switched dynamic brake and SMES for TFC = 0.21667 secs

Effect of Fault Clearing Time on the SMES Rating It has been found that the maximum active power requirement of SMES depends on the fault clearing time. With increase in fault clearing time it increases and is shown in Table 1 and the trend is plotted in Fig7. Performance of SMES is very satisfactory under different operating conditions (for light load as well as heavy load conditions), in case where the fault clearing time is more than the critical fault clearing time under the representative operating conditions considered. It has been found that the requirement of SMES rating increases with increase in TFC accordingly. If a fast acting circuit breaker is used to clear the fault in the system, the requirement of SMES rating for the maintenance of transient stability increases.
H Table I: Variation of the Maximum Active Power Requirement of SMES VARIATION OF THE MAXIMUM ACTIVE POWER REQUIREMENT OF SMES TFC Psmes(max) 0.0500 3cycles .0867 0.0834 5 cycles .1993 0.1334 8cycles .3084 0.16667 10 cycles .3800 0.20000 12cycles .4158 0.21667 13cycles .4522 0.23334 14cycles .4909 0.25001 15cycles .5253 0.26668 16cycles .5631

The practical implementation of the control signal was of the main deciding factor in favor of as the promising signal for the investigations. Proportional-SMES has been tried for the transient stability investigations. It was observed that transients damp out in 3.0 seconds. In order to investigate the effectiveness of the SMES in damping the oscillations the investigations have been carried out for two cases namely when the fault is cleared at the instant of critical fault clearing time and also for a case when the fault is cleared at a time beyond the critical fault clearing time. To find out the effectiveness of SMES in improving transient stability of a power system, its performance is compared to a conventional dynamic brake presently in use. It is observed that the performance of the system considered with SMES is better than that with optimally switched dynamic brake under similar conditions of operation. The power requirement of SMES was obtained as the fault clearing time was varied. It has been found that the requirement of SMES rating increases with increase in TFC accordingly if a fast circuit breaker is used to clear the fault in the system, the requirement of SMES rating for the maintenance of transient stability decreases. APPENDIX I With reference to Fig 3 the system data in p.u. is as follows [4]: Data for The SMIB System [4 ]
Xd GL+jBL Xq S X'd
Vt

1.0 0.18 + j 0.067 0.6 0.753 + j0.03 0.27 1.05 0.13 1.075 0.15 0.7488 9 secs 0.2 secs 4 secs

Xt
E

RL XL Tdo TFC H

ACKNOWLEDGMENT Facilities are provided by the PEC University of Technology (formerly Punjab Engineering College), Chandigarh for completing the research work. REFERENCES
Fig. 7: Variation in Psmes vs variation in fault clearing time [1]. H.J. Boenig, J.F.Hauer, H.J .Boenig and J.F Hauer, Commissioning Test of the Bonneville Power Administration on 30 MJ Superconducting Magnetic Energy Storage Unit, IEEE Transactions on Power Apparatus and Systems, Vol. 2, PAS 104, pp. 302- 309, Feb. 1985. T. Ise and Y. Murakami, Simultaneous Active and Reactive Power Control of Superconducting Magnetic Energy Storage Using GTO Converter, IEEE Trans. on Power Delivery, Vol. PWRD-I, No.1, pp. 143-150, January 1986.

IV. CONCLUSION In this paper the transient stability investigations have been carried out for single machine connected to infinite bus system where SMES devices are used to control stability. The system variable namely was selected.

[2].

[3].

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

275

[4].

[5].

[6]. [7]. [8]

Rabbani, J. B.X. Devotta, S. Elangovan, Application ofSimultaneous Active and Reactive Power Modulation of SMES Unit under Unequal Mode for Power System Stabilization, IEEE Transaction on Power Systems, Vol. 14, No.2, May 1999. S.M. Minisey and E.V. Bohn, Optimum Network Switching inPower Systems, IEEE Transaction on Power Apparatus and Systems, Vol. PAS- 90,pp. 2118-2123, Sept.1971. M.A. Pai, Computer Techniques in Power System Analysis, (Book), Tata Mc -Graw-Hill Publishing Company Ltd., New Delhi, 1979. R. Khanna, "Some Aspects of transient stability Improvement usingSuperconducting Magnetic Energy Storage," Ph.D. dissertation, Deptt. Elect. Engg.., Panjab University, Punjab Engineering College, Chandigarh, 2010.

Member of IEEE Power Electronics Society (USA). He is also an active Member of Indian Microelectronics Society. His current areas of interests include Power Electronics and Drives, Power System Stability and Control, Renewable Energy Resources and Smart Grid Technologies. Tusar Kanti Nagsarkar (M83,SM85) received B.Sc. degree and B.E. degree in Electrical Engineering from the Nagpur University, Nagpur, M.Tech degree from the Indian Institution of Technology,Kharagpur and Ph.D degree from Panjab University, Chandigarh in 1961,1965,1967 and 1981 respectively. He worked in Punjab Engineering college, Chandigarh during 1967-2001 from where he retired as Professor and Head, Department of Electrical Engineering. Dr. Nagsarkar is a fellow of Institution of Engineers , India.

VII

Biographies

Rintu khanna (M99) was born in India, on Nov. 11, 1963. She did her B.E., M.E. degree from the Punjab Engineering College, Panjab University, Chandigarh and doctrate from PEC University of Technology (formerly Punjab Engineering College) , Panjab University ,Chandigarh Her employment experience includes teaching in Punjab Engineering College. Chandigarh. She has guided a number of M Tech dissertations and published research papers in national/international journals as well as conferences .Her special fields of interest are power system, Soft Computing, applications of power electronics, Renewable Energy Resources and Smart Grid Technologies. She is member of IEEE (power system) and Life member of the Institution of Engineers (India) also. Gurnam Singh (M1988) was born in Zira, Punjab India, on December 1, 1948. He graduated from the Punjab Engineering College Chandigarh now known as PEC University of Technology. He did his Masters degree in 1973 at Panjab University, Chandigarh (India). He is Professor of Electrical Engineering and currently posted as Deputy Director at PEC University of Technology, Chandigarh, India. He opted to join the position of Assistant Professor of Electrical Engineering at Punjab Engineering College, Chandigarh way back in 1972. He earned a Doctorate Degree in the Faculty of Engineering and Technology from Panjab University, Chandigarh with a specialization in the area of excitation control of Power Systems. He was selected as a Professor of Electrical Engineering through Direct Selection on All India Basis conducted by Union Public Service Commission of India. During the past 30 years, he has helped to establish teaching and research laboratories at Punjab Engineering College for Microprocessor Applications, Real Time Systems and Advanced Power System Simulation. His research efforts are focused on ITIntensive Computer Applications to Stability, Power Electronics, Microcomputers, Microelectronics and Systems Simulation. He has published numerous papers in technical journals and presented many more at workshops & conferences both national as well as international. He has the distinction of developing IT-Intensive Curriculum for B.E (Electrical) Degree Course at Punjab Engineering College and has taught most of these subjects as a pioneer in this area. Such subjects include Microprocessor and its Applications, Digital Design, Analog & Mixed Signal Design, Power Electronics Control of AC & DC Motors, Data Structures Using C, Analog and Digital Microelectronics, etc. The Chandigarh Administration honored him by awarding him the Certificate of Commendation on an Independence Day in recognition of outstanding services rendered in the field of Dedication of Duty. He has guided a number of PhD and M Tech dissertations and published research papers in national/international journals as well as conferences. Dr. Gurnam Singh is the Mentor of IEEE Student Branch at Punjab Engineering College and was the Counselor of this Branch for the last so many years. He is a Fellow of Institution of Electronics and Telecommunication Engineers, Member of IEEE Circuits and Systems Society (USA), IEEE Signal Processing Society (USA) and Permanent

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

276

Deregulated Energy Market and Congestion Management: A Review


1
1, 2
1

Manvir Kaur, 2Sumit Verma

Thapar University Patiala (India) operation, paper elucidates the concepts of network congestion in deregulated environment. Network congestion divides wholesale market to different zones, different techniques are explained to deal with network congestion. II. DEREGULATED ELECTRICITY MARKET Deregulation in power industry is a restructuring of the rules and economic incentives that government set up to control and drive the electric power industry. The electric power industry has over the years been dominated by large utilities that had an overall authority over all activities in generation, transmission and distribution of power within its domain of operation. Such utilities have often been referred to as vertically integrated utilities. One of the principal characteristics of a competitive structure is the identification and separation of the various tasks which are normally carried out within the traditional organization so that these tasks can be open to competition whenever practical and profitable. This process is called unbundling. An unbundled structure contrasts with the socalled vertically integrated utility of today where all the tasks are coordinated jointly under one umbrella with one common goal, that is, to minimize the total costs of operating the utility. One of the first steps in the restructuring process of the power industry has been the separation of transmission and generation activities [2]. The subsequent step was to introduce competition in generation activities, either through the creation of power pools, provision for direct bilateral transactions or bidding in the spot markets. An important point to note is that the restructuring process was however not uniform in all countries. While in many instances, it started with the breaking up of a large integrated utility, in certain other instances. III NETWORK CONGESTION When the producers and consumers of the electric energy desire to produce and consume in amounts that would cause the transmission system to operate at or beyond one or more transfer limit, the system is said to be congested. Congestion is a consequence of network constraints characterizing a finite network capacity that precludes the simultaneous delivery of power from an associated set of power transactions. Line outages or higher load demands are the causes of congestion in the transmission network. When a generator bids other than its incremental costs, in an effort to exploit imperfections in

manvir.kaur@gmail.com, 2sumitverma1990@yahoo.com

Abstract Power companies world-wide have been restructuring their electric power systems from a vertically integrated entity to a deregulated, open-market environment. Transmission network is vital entity in open access deregulated wholesale electricity market. Whenever transmission network congestion occurs in electricity market, it divides the market in different zones and the trading price of electricity will no longer remains the same for the whole system. In this paper, after presenting a conceptual understanding of deregulated electricity market, how network congestion effect the energy market and forces the market to change its price from a common market clearing price to locational market and a comprehensive review of utility practices as well as research methods in the area of transmission congestion management in deregulated electricity markets is presented. Keywords: Power System Deregulation, Network Congestion, electricity pricing, FACTS.

I. INTRODUCTION The success of privatization of most of the industries led people to think for the deregulation of electric power system. This yields to restructuring of currently vertically integrated utility (VIU) to the main three utilities, namely generation company (GENCO), transmission company (TRANSCO) and distribution company (DISCO). The success in the energy privatization in the countries like UK, USA, Norway and Australia has encouraged many more countries to privatize their electricity industry. India has also participated in the process and most of the states of India have restructured their electricity boards. Ever since the restructuring has taken place, the electric power industry has seen tremendous changes in its operation and governance. The sole purpose of introduction of deregulation and electricity market is to create a healthy competition among the participant of the market and to make electricity market more efficient, liquid and complete [1]. The development of electricity market also aims for the maximum participation from the electric utilities to provide transparent and non-discriminatory platform for energy producers. The efficiency of market decreases in the event of transmission line congestion. The congestion results in price change and reduces the market efficiency. Congestion can be managed by different approaches. One of the approach is real and reactive power rescheduling [2]. As it proceeds, the paper explains the structure of electricity market and its major components. After giving the background of electricity market architecture and

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

277

the market to increase profits, its behaviour is called strategic bidding. If the generator can successfully increase its profit by strategic bidding or by any means other than lowering its costs, it is said to have market power. The obvious example of market power is a nonregulated monopoly with a zero elasticity demand, where the generator can ask whatever the pr ice i t wants for electric energy. Market power resul t s in market inefficiency.

[3],[4].Three broad methods of congestion management are currently in use around the world. One method is basically a centralized optimization exercise; either explicitly with some form of generation re-dispatch with security and transmission constraints, or implicitly, direct control of system operators control congestion [5]. A second method is based on the use of price signals derived from ex ante market resolution to deter congestion by allowing congestion to constrain scheduled generator output prior to real time operation. Inevitably some congestion may still arise and must be corrected in real time by centralized control [6]. A third method seeks to control congestion by allowing or disallowing bilateral transmission, agreements between a producer and a consumer, based on the effect of the transaction on the transmission system [7]. Several methods have been reported that address the congestion management problem in deregulated electricity markets. These can be classified into broad groups, as : A) Security-Constrained Generation Re-dispatch One of the most common approaches to alleviate congestion in the network is to re-schedule of the generation by the optimal power flow model with transmission constraints as well as bus-voltage constraints to maintain the system security. An unified framework for the representation of market dispatch and re-dispatch problems that the independent grid operator must solve in congestion management in various jurisdictions was developed in [8]. This framework is used to compare the performance of different congestion management approaches that exist today in various markets in the world. An optimization procedure for re-dispatch of generation is proposed in [9 in order to alleviate transmission congestion on the network. Consequently, a new approach to allocate the cost of congestion and losses to the nodes of the transmission network based on nodes responsibility is proposed. While a minimum-distance generation re-dispatch was suggested in [10], where the economic value of the transaction adjustment is disregarded. While solving the congestion problem with generation re-dispatch, Huang and Yan [11] investigates the impacts of thyristor controlled series capacitor (TCSC) and static VAr compensator (SVC) with the objective of minimizing the total amount of transactions being curtailed on this re-dispatch method. This paper suggested that the improvement of total transfer capability (TTC) by using TCSC and SVC with the consideration of transaction patterns would reduce the possibility of congestion occurrence.

Fig. 1 (a, b)

There are many possible causes of market power. One of the main reasons is congestion. Consider an example shown in Figure(1). Let each zone have a 100MW constant load. Zone A has 200 MW generator with an incremental cost of Rs.100/MWh. Zone B has a 200 MW generator with an incremental cost of Rs.200/MWh. Assume both generators bid their incremental cost. If there is no transfer limit between zones, all 200 MW of load will be brought from generator A at Rs.100/MWh as shown in Fig(1a). If there is a 50 MW transfer limit, then 150 MW will be brought from A at Rs.100/MWh and the remaining 50 MW must be brought from generator B at Rs.200/MWh, as shown in Fig(1b). The total cost of 200 MW in un-congested market is Rs.20000/h and that in the congested market is Rs.25000/h. Congestion has created a market inefficiency of 25% of the optimal cost, even without strategic behavior of the generator. Congestion has also created unlimited market power for generator at B. IV CONGESTION MANAGEMENT TECHNIQUES The main objectives of congestion management would now be i) to develop MW schedules which minimizes the system cost variation of the initial market clearing while fulfilling system security criteria; ii) to provide appropriate economic signals that are consistent with the MW schedules; iii) to facilitate the management of transmission congestion risks. Congestion management is also done with respect to maximizing the overall satisfaction degree of all participants in the market

B) Congestion Pricing and Market-Based Methods Price signals were used for the generators to manage congestion and the solution under rational behaviour assumption is identical to an OPF solution [12]. A similar approach was suggested for the pool model in [13], where

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

278

the cost of congestion was bundled with the marginal cost at each bus. A bilateral model was also investigated, and a congestion cost minimization approach was proposed. A framework for real-time congestion management under a marker structure similar to the newly proposed UK trading arrangement is presented in [14], in which not only resources in balancing market but also some bilateral contracts can be dispatched if necessary. The linearized model of a modified optimal power flow (OPF) is proposed to implement such a framework. A new framework was presented in [15] to manage dynamic congestion. The main feature of the proposed model is that system stability is incorporated into the congestion management, and the concept of market-based congestion management is extended into the dynamic scenario. Under the proposed framework, the ISO can eliminate the dynamic congestion with available resources in the real-time dynamic congestion management market. The total dynamic congestion management cost will be minimized, and system security as well as the scheduled transactions is maintained. C) Network Sensitivity Factors Methods Other method for congestion management proved to be efficient is the use of network sensitivity factors, which is the relationship between the change in power injection and the change in power flow in the network, has been demonstrated in [16]-[22]. In a deregulated electricity market, it may always not be possible to dispatch all of the contracted power transactions due to congestion of the transmission corridors. System operators try to manage congestion, which otherwise increases the cost of the electricity and also threatens the system security and stability. In this paper, a new zonal/cluster-based congestion management approach has been proposed. The zones have been determined based on lines real and reactive power flow sensitivity indexes also called as real and reactive transmission congestion distribution factors. The generators in the most sensitive zones, with strongest and non-uniform distribution of sensitivity indexes, are identified for rescheduling their real power output for congestion management. In addition, the impact of optimal rescheduling of reactive power output by generators and capacitors in the most sensitive zones has also been studied [16]. D) Application of FACTS Devices An alternative to building new transmission lines to solve the frequently occurred congestion problems is to use Flexible AC Transmission System (FACTS) [23]-[27]. FACTS devices such as thyristor controlled series compensators and thyristor controlled phase angle regulators, by controlling the power flows in the network, can help to reduce the flows in heavily loaded lines resulting in an increased loadability of the network and reduced cost of production [24]. Congestion management using FACTS devices requires a twostep approach. First,

the optimal location of these devices in the network must be ascertained and then, the settings of their control parameters optimised. The development of simple and efficient models for optimal location of FACTS devices that can be used for congestion management by controlling their parameters optimally. E) Cluster-Based Management Approach The congestion management zones have been determined based on lines real and reactive power flow sensitivity indexes also called as real and reactive transmission congestion distribution factors. The generators in the most sensitive zones, with strongest and non-uniform distribution of sensitivity indexes, are identified for rescheduling their real power output for congestion management [16]. A new method has been proposed to calculate and settle zonal congestion cost in deregulated power markets, which are a mix of the pool model and the bilateral model. Based on the different settlement methods deployed, deregulated power markets can be divided into pay-as-marginal-price market, pay-asmarket-clearing-price market, and pay-as-bid market. All the three kinds of markets are considered in the proposed new method by solving the same optimization problem of zonal congestion management F) Demand-Response for Congestion Management As we have observed, congestion relief is normally carried out through generators in the short-term by redispatching available generation to avoid congestion and other associated problems in various contingency situations. This form of congestion service provision is referred to as preventive management [28]. If however, the cost of such preventive management is too high, then it would be more cost-effective to invest in transmission system reinforcement which can be referred to as longterm congestion management. A somewhat in-between alternative to the above two is to create provisions for load interruption in a judicious manner that could aid in transmission congestion relief, which can be referred to as corrective management [29]. In the context of deregulated markets, introducing the provision that allows customers to offer their interruptible load for competitive procurement by the ISO is a topical issue. Participation of the customers in this provision for congestion relief could significantly increase the number of service providers, and hence locations, available to the ISO. G) Financial Transmission Rights (FTRs) An FTR is a financial risk-management instrument. FTRs are used to hedge the costs associated with transmission congestion. It represents a specified MW amount between (usually) two points in the power transmission network. It is valid over a defined period of time, typically a month, season or year, and often only for peak or off-peak hours. Whenever there is transmission congestion in the FTRs defined direction, the FTR will earn congestion revenue for its holder from the ISO. The

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

279

FTRs primary purpose is to offset a transmission users congestion charges, which are typically quite volatile [30]. V. CONCLUSION This paper presents a systematic and comprehensive review of the current research trends in the issues related to transmission congestion management. Depending on the structure and objectives of the electricity market, different congestion management methods are put into practice. Effective congestion management will help mitigate the effects of market power in electricity markets. Attempt has been made to cover most major transmission congestion management techniques in deregulated electricity markets. REFERENCES
[1]. [2]. [3]. A. Pahwa, Student Member, Zuyi Li, Member,IEEE , ElectricityIndustry Restructuring in India,Power symposium 2006, Loi Lai Lei Power System Restructuring and Deregulation, Edited), John Wiley & Sons Limited, 2001 Gomes, M.H.; Saraiva, J.T., Congestion management by maximizing the overall satisfaction degree of all participants in the market, Proc. IEEE Power Tech, 1, Porto, 2001, 6 pp. R.S. Fang, A.K. David, Transmission congestion management in an electricity market, IEEE Transactions on Power Systems, 14(3), 1999, 877-883. K. Bhattacharya, M.H.J. Bollen, & J.E. Daalder, Chapter 4, Operation of Restructured Power Systems (Kluwer Academic Publishers, 2001). S. Oren, Market-based congestion management, Presented at the PUCT Public Workshop on Congestion Management, Texas, 2003. R.D. Christie., B.F. Wollenberg, & I. Wangensteen, Transmission management in the deregulated environment, Proceedings of the IEEE, 88(2), 2000. E. Bompard, P. Correaia, & G. Gross G., M. Amelin, Congestionmanagement schemes: A comparative analysis under a unified framework, IEEE Transactions on Power Systems, 18 (1), 2003, 346-352. N.S. Rau, Transmission loss and congestion cost allocation An approach based on responsibility, IEEE Transactions on Power Systems, 18 (1), 2003, 346-352. F.F. Falania, M. Ilic, A mathematical framework for the analysis and management of power transactions under open access, IEEE Transactions on Power Systems, 13(2), 1998, 681-687. G.M. Huang, P. Yan, TCSC and SVC as re-dispatch tools for congestion management and TTC improvement, Proc. IEEE Power Engineering Society Winter Meeting, 1, 2002, 660 665. H. Glavisch and F. Alvarado, Management of multiple congested conditions in unbundled operation of a power system, IEEE Transactions on Power Systems, 13(3), 1998, 1013-1019. H. Sing, S. Hao, & A. Papalexopoulos, Transmission congestion management in competitive electricity markets, IEEE Transactions on Power Systems, 13(2), 1998, 672-680. X. Wang, Y.H. Song, Advanced real-time congestion management through both pool balancing market and bilateral market, IEEE Power Engineering Review, February 2000. J. Ma, Y.H. Song; Q. Lu, S. Mei, Market-based dynamic congestion management, IEEEPower Engineering Review, 22(5), 2002, 54 56. A. Kumar, S.C. Srivastava, S.N. Singh, A zonal congestion management approach using real and reactive power rescheduling, IEEE Transactions on Power Systems, 19(1), 2004, 554 562. J.G. Vlachogiannis, Accurate model for contribution of generation to transmission system effect on charges and congestion management, IEE Proceedings on Generation, Transmission and Distribution, 147(6), 2000, 342 348.

[4].

[5].

[6].

[7].

[8].

[9].

[18]. J. Yang, M.D. Anderson, Tracing the flow of power in transmission networks for use-of-transmission-system charges and congestion management, Proc. IEEE Power Engineering Society Meeting, 1, 1999, 399 405. [19]. M. Liu, G. Gross, Role of distribution factors in congestion revenue rights application, IEEE Transactions on Power Systems, 19 (2), 2004, 802-808. [20]. R. Baldick, Variation of Distribution Factors with Loading, Harvard Electricity Policy Group, Harvard University, available online at http://www.ksg.harvard.edu/hepg/ [21]. ] Kopcak, I.; da Silva, L.C.P.; da Costa, V.F.; Naturesa, J.S.; Transmission systems congestion management by using modal participation factors Power Tech Conference Proceedings, 2003 IEEE Bologna , Volume: 2 , June 23-26, 2003 Pages:6 11 [22]. G.C.; Erlich, I.; Congestion analysis and participants' behaviour in a pool market Stamtsis Generation, Transmission and Distribution, IEE Proceedings- , Volume: 151 , Issue: 1 , 14 Jan. 2004 Pages:127 131 [23]. G.M. Huang, P. Yan, Establishing pricing schemes for FACTS devices in congestion management, Proc. IEEE Power Engineering Society General Meeting, 2, 2003, 1030-1030. [24]. S.N. Singh, A.K. David, Congestion management by optimising FACTS device location, Proc. International Conference on Electric Utility Deregulation and Restructuring and Power Technologies, 2000, 23 28. [25]. J. Brosda, E. Handschin, Congestion management methods with a special consideration of FACTS-devices, Proc. IEEE Power Tech, 1, 6 pp. [26]. S. Bruno, M. La Scala, Unified power flow controllers for security-constrained transmission management, IEEE Transactions on Power Systems, 19(1), 2004, 418 426. [27]. X. Wang; Y.H. Song, Q. Lu, Y.Z. Sun, Series FACTS devices in financial transmission rights auction for congestion management, IEEE Power Engineering Review, 21(11), 2001, 41 44. [28]. E. Lobato, L. Rouco, T. Gmez, F. M. Echavarren, M. I. Navarrete, R. Casanova, & G. Lpez, Preventive analysis and solution of overloads in the Spanish electricity market, Electric Power Systems Research, 68(3), 2004, 185-192. [29]. M. Doll, J.F. Verstege, Congestion management in a deregulated environment using corrective measures, Proc. IEEE Power Engineering Society Winter Meeting, 2, 2001, 393 398. [30]. D.S. Kirschen and G. Strabac, Chapter 6 in Fundamentals of power system economics: Transmission networks and electricity markets (John Wiley & Sons Ltd, 2004)

[10].

[11].

[12].

[13].

[14].

[15].

[16].

[17].

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

280

Implementation of PID based Intelligent Fuzzy Controller for Temperature Control in Process Industries
1

Rakesh Kumar, 2Mandeep Sharma


1, 2
1

Bhsbiet Leheragaga Poly. Wing subsequently reported for the four-rule controllers, and linear-like fuzzy controllers. Palm has analytically demonstrated the equivalence between the fuzzy controller and sliding-mode controllers . It is possible to build a fuzzy controller which provides better performance than a conventional PID controller, but a fundamental question remainsIs a fuzzy PID controller guaranteed to outperform a conventional PID controller for any type of process! It has been reported that a specific fuzzy controller is not necessarily better than a conventional PI controller. This raises another question Under what conditions can a fuzzy PID system provide better performance than a linear PID controller! To date, no satisfactory formal techniques have been developed to solve these problems. Moreover, in a study of optimal design for fuzzy controllers, two relationships must be established: 1) design parameters and control nonlinearity, and 2) control nonlinearity and process performance. This Paper is an attempt to undertake the development of a new analytical approach to the optimal design of fuzzy controllers. We propose a new methodology for the optimal design of fuzzy PID controllers. In the proposed work the main objective of the investigator is to compare the performances of conventional PID controllers and the intelligent fuzzy logic controller. For this comparison, two parameters needs to be evaluated i.e. Overshoot and settling time. This paper suggests a fuzzy logic based controller which acts with the help of artificial intelligence techniques. There are many artificial intelligence techniques and fuzzy logic is one of them. Fuzzy logic can handle imprecise data and can effectively used in controller design. Table I Shows design parameters of a fuzzy PID Controller and table II shows comparison of PID and a fuzzy PID Controller showing membership functions and relation matrices.

raj5sept@rediffmail.com

AbstractThis paper presents a systematic approach for the design and implementation of temperature controller using PID based intelligent fuzzy-controller for chemical plants and investigates the robustness of the designed control law. The proposed approach employs PID based intelligent fuzzycontroller for determination of the optimal results than PID controller parameters for a previously identified chemical plant. Results indicate that the proposed algorithm significantly improves the performance of the chemical plant. It is anticipated that designing of PID based fuzzy controller using proposed intelligent techniques would dramatically improves the speed of response of the system.Rise time and settling time would be reduced in magnitude in the intelligent scheme as compared with conventional PID controller. Keywords: Chemical plants, Steam temperature control, Industrial system, Multiobjective control; Optimal-tuning; PID control Fuzzy logic control, genetic algorithms, nonlinear control, optimal control, PID control

I. INTRODUCTION Fuzzy Logic Control (FLC) technique has found many successful industrial applications and demonstrated significant performance improvements. However, fuzzy controller design remains a fuzzy process due to the fact that there is insufficient analytical design technique in contrast with the well-developed linear control theories. Although the functions of fuzzy systems have the advantage of being relatively easy to understand, the systems become complex or nontransparent due to many design parameters involved. Considering fuzzy PID-like controllers, we summarize the design parameters within two groups: structural parameters and tuning parameters. While the structural parameters are determined during offline design, tuning parameters can be calculated during online adjustments of the controller to enhance the process performance, as well as to accommodate the adaptive capability to system uncertainty and process disturbance. Some parameters can be called either structural or tuning parameters depending on their usage. A wide variety of fuzzy PID-like controllers have been developed. In most cases, fuzzy controller design is accomplished by trial-and-error methods using computer simulations. Significant studies based on the closed-form analysis of fuzzy PID-like controllers started with the work of Ying, Siler, and Buckley , where they have used a simple four-rule controller similar to that of Murakami and Maeda ,More analytical work in this regard was

Fig. 1: Proposed methodology with respect to the data or parameter flows for optimal design of fuzzy PID controllers.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

281

form a control curve, is closely analogous to a linear PID controller. III. SIMULATION RESULTS Oil Tank Temperature Controller This temperature controller is used to control the temperature of raw oil in oil tank. In this the set temperature is 255oC and PID temperature controller reaches set temperature in five hours and ten minutes. Fuzzy model was developed using error, change in error and fuzzy output to improve the settling time.
TABLE 4.1: FUZZY SYSTEM, FOR OIL TANK TEMPERATURE CONTROLLER

Fig. 2: Structure of the proposed INTELLIGENT fuzzy PID controller.

II. NEW METHODOLOGY A new methodology is proposed for the analytical design of a fuzzy PID controller. Fig 1 shows the proposed methodology with respect to the data or parameter flow in off-line design. In Step 1, the structure of a fuzzy PID controller is designed and the structural parameters are set for the preliminary design. The tuning parameters are identified in Step 2, while in Step 3 an analytical fuzzy calculation is performed, which produces a closed-form relationship between the design parameters and control action for the fuzzy inference. In Step 4, numerical simulation (or control theory) is used to obtain the control performance data. In Step 5, genetic-based optimizations are carried out to produce optimal design parameters. This also provides useful information for the redesign of the original system. Finally, if necessary, redesign is undertaken using the designer's expertise for further improvement to the control system. Note that the theoretical study in Step 3 makes the fuzzy controller transparent. This step is important since it will establish a close link between fuzzy control design technique and classical/modern control theory. Simplicity is a key principle of this design methodology. The reason is obvious if we see that fuzzy logic controllers are systems which simulate human control exercise. For many everyday control tasks, people initially try to apply simple rules. Three rules used in this work are very common in a feedback set-point control problem. If a satisfactory control process can be achieved by applying simple rules, the use of complex rules, which is often associated with a higher cost of computation, becomes unnecessary. Simplicity is the best and direct way to maintain a clearly physical insight into the control laws. It also makes highdimensional fuzzy systems tractable for using simple mathematical expressions for describing functionality between design parameters and nonlinearity. The simplicity of a system can be assessed by examining the structure of the information flow and the total number of design parameters. It is preferable for a system to include high modularity, parallelism, concurrency, and normality for implementation in both software and hardware. To simulate the simplest control exercise of human beings, we develop a one-input/three-rule fuzzy PID controller associated with at most six design parameters. This system, having a simple structure to

(a) Membership functions of Error input.


Membership function for Error Initial Peak value value Very Small (VS) -2 20 Linguistic variable Small (S) Medium (M) High (H) Very High (VH) 20 60 100 160 50 100 140 190 Final value 50 90 140 190 230

(b) Membership functions of Change in Error input.


Membership function for Change in Error Initial Peak Linguistic variable value value Very Small (VS) -10 -6 Small (SM) -6 0 Medium (MD) 0 8 Large (L) 9 17 Membership function for Fuzzy Output Initial Peak Linguistic variable value value Very Small (VS) 28 39 Small (S) 38 70 Medium (M) 70 110 Large (L) 120 165 Very Large (VL) 180 220 Final value -2 6 15 25

(c) Membership functions of Fuzzy output.


Final value 50 90 140 210 255

To develop Fuzzy controller, firstly error signal(e) is calculated by subtracting output of PID temperature controller from set temperature then change in error(e) was calculated by subtracting previous error from current error. Considering error and change in error as input and fuzzified output as output function membership functions are created for each input and output. Membership functions for these quantities are defined as in above Table 4.1. The membership functions are shown in schematic form in Fig 4.1.

(a) Membership functions of Error input.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

282

(b) Membership functions of Change in Error input.

Red graph shows the fuzzy output of fuzzy model of oil tank temperature controller, black line represent the output of PID temperature controller and blue line represent the set temperature enter in PID temperature controller. Fuzzy output has some oscillations in rising time and a steady state error of 35oC. To improve this fuzzy response the membership functions of all the input and output are increased. Membership functions of error and change in error inputs have been increased to six and that of fuzzy output has been increased to seven, as shown in Table 4.3
TABLE 4.3: IMPROVED FUZZY SYSTEM, FOR OIL TANK TEMPERATURE
CONTROLLER

(c) Membership functions of Fuzzy output.


Fig. 4.1: Fuzzy system, for oil tank temperature controller

A rule base was developed for the fuzzy model using simple IF-THEN rules, as follows: (i) IF Error is Very Small (VS) and Change in Error is also Very Small (VS) THEN Fuzzy Output is Very Large (VL). (ii) IF Error is Very Small (VS) and Change in Error is Small (SM) THEN Fuzzy Output is Very Large (VL). (iii) IF Error is Very Small (VS) and Change in Error is Medium (MD) THEN Fuzzy Output is Very Large (VL). (iv) IF Error is Small (S) and Change in Error is also Small (SM) THEN Fuzzy Output is Large (L). (v) IF Error is Small (S) and Change in Error is Medium (MD) THEN Fuzzy Output is Very Large (VL). This rule base is summarized as in Table 4.2
TABLE 4.2: RULE BASE Fuzzy output (Fz) Error(e) VS S M H VH VS VL S VS Change in Error (e) SM VL L L M VS MD VL VL L M S L L S

(a) Membership functions of Error input. Membership function for Error Linguistic Initial Peak Final variable value value value Very Small (VS) -2 20 40 Small (S) 20 50 80 Medium (M) 60 90 120 High (H) 90 120 150 Very High (VH) 130 165 200 Extremely High 170 220 230 (EH) (b) Membership functions of Change in Error input.
Membership function for Change in Error Initial Peak value Final value value Negative (N) -10 -6 -2 Negative Small (NS) -6 -1 4 Small (SM) 0 4 8 Medium (M) 5 9 13 Large (L) 9 14 19 Very Large (VL) 14 19 25 Linguistic variable

(c) Membership functions of Fuzzy output.


Membership function for Fuzzy Output Initial Peak Linguistic variable value value Very Small (VS) 28 39 Small (S) Medium (M) Medium Large (ML) Large (L) Very Large (VL) Extremely Large (EL) 38 70 110 160 200 240 70 110 150 190 225 250 Final value 50 90 140 190 220 250 260

On the basis of this rule base a fuzzified output is calculated. This Fuzzy model is simulated in MATLAB fuzzy logic toolbox GUI, and results are obtained. Then results are plotted along with the actual temperature and set temperature obtained from the process, are plotted in Fig4.2.

The rule base is also revised as shown in Table 4.4. By using this rule base, oscillations in fuzzy response decreases and steady state error was also reduced than the last fuzzy model.

TABLE 4.4 : IMPROVED RULE BASE Fig4.2. Response curve of PID temperature controller Vs Fuzzy temperature controller in oil tank temperature controller Fuzzy output (Fz) Change in Error (e)

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

283

Error (e) VS S M H VH EH

N EL S VS

NS EL VL L ML M VS

SM EL VL L L M S

M EL VL L L ML M

L ML M

VL M

Here the steady state error and settling time both have been improved. Steady state error is decreased to zero and settling time is reduced by 2 hours and 10 minutes. For Further improvements, Fuzzy output and rule base have been revised. Revised Fuzzy output and rule base are shown in Table 4.6.
TABLE 4.6: IMPROVED FUZZY SYSTEM, FOR OIL TANK TEMPERATURE
CONTROLLER

The MATLAB simulation results are plotted along with the actual temperature and set temperature obtained from the process, are plotted in Fig4.3.

(a) Membership functions of Fuzzy output.


Membership function for Fuzzy Output Initial Linguistic variable Peak value value Very Small (VS) 28 50 Small (S) Medium (M) Medium Large (ML) Large (L) Very Large (VL) Extremely Large (EL) 50 100 140 180 200 250 85 130 165 200 220 255 Final value 80 120 165 200 220 255 260

Fig. 4.3: Improved Response curve of PID temperature controller Vs Fuzzy temperature controller in oil tank temperature controller

Here the steady state error is decreased to 5oC, and settling time also improved. Further improvements in Fuzzy output and rule base have been made. To achieve this requirement, the range of last two membership function of fuzzy output has been changed. The new range is VL 200-220-255 and EL 250-255-260 and rule base is shown in Table 4.5.
TABLE 4.5: IMPROVED RULE BASE Fuzzy output (Fz) Error (e) VS S M H VH EH Change in Error (e) N EL VS NS EL EL VL ML M VS SM EL EL VL L M S M EL EL VL L ML M L ML M VL M

(a) Improved Rule base


Fuzzy output (Fz) Error (e) VS S M H VH EH N EL VS NS EL EL VL L ML S Change in Error (e) SM EL EL VL VL ML VS M EL EL VL VL L M L L M VL -

The MATLAB simulation results are plotted along with the actual temperature and set temperature obtained from the process, are plotted in Fig4.5.

The MATLAB simulation results are plotted along with the actual temperature and set temperature obtained from the process, are plotted in Fig4.4.

Fig. 4.5. Improved response curve of PID temperature controller Vs Fuzzy temperature controller in oil tank temperature controller

Fig. 4.4. Improved response curve of PID temperature controller Vs Fuzzy temperature controller in oil tank temperature controller

Here fuzzy output response contains lesser oscillations. Settling time also reduce by 45 minutes from last fuzzy model. Total time reduced is 2 hours and 55 minutes. To

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

284

further improve fuzzy response, fuzzy system is revised. Membership functions of error input and fuzzy output are increased. Membership functions of error input have been increased to seven and that of fuzzy output has been increased to nine. Improved error input, fuzzy output and rule base are shown in Table 4.7
TABLE 4.7 IMPROVED FUZZY SYSTEM, FOR OIL TANK TEMPERATURE
CONTROLLER

(a) Membership functions of Error input. Membership function for Error Linguistic Initial Peak Final variable value value value Very Small (VS) -2 20 40 Small (S) 20 50 80 Medium Low 55 85 110 (ML) Medium High 90 115 140 (MH) High (H) 120 150 180 Very High (VH) 150 185 210 Extremely High 200 215 230 (EH) (b) Membership functions of Fuzzy output. Membership function for Fuzzy Output Initial Peak Final Linguistic variable value value value Very Small (VS) 28 50 69 Low Small (LS) 52 70 90 High Small (HS) 72 95 118 Medium Low 95 120 140 (ML) Medium High 120 145 175 (MH) Medium Large 145 170 200 (ML) Large (L) 180 200 230 Very Large (VL) 210 230 255 Extremely Large 250 255 260 (EL) (c) Improved Rule base Fuzzy Change in Error (e) output (Fz) Error (e) N NS SM M L VL VS EL EL EL EL S EL EL EL EL ML EL EL EL EL MH VL VL H ML ML L L L VH HS HS LM HM HM EH VS VS LS LS The MATLAB simulation results are plotted along with the actual temperature and set temperature obtained from the process, are plotted in Fig4.6.

Fig4.6. Improved response curve of PID temperature controller Vs Fuzzy temperature controller in oil tank temperature controller

Here fuzzy output response contains lesser oscillations than the last fuzzy output response. Settling time is also reduced by 20 minutes, i.e. settling time is 2 hours and 15 minutes. But one spick is produced during the rise time that can be removed by changing the rule base as in Table 4.8
TABLE 4.8 IMPROVED RULE BASE

Fuzzy output Change in Error (e) (Fz) Error (e) N NS SM M L VL VS EL EL EL EL S EL EL EL EL ML EL EL EL EL MH VL EL H ML ML L L L VH HM HM HM HM HM EH VS VS LS VS The MATLAB simulation results are plotted along with the actual temperature and set temperature obtained from the process, are plotted in Fig4.7.

Fig. 4.7: Improved response curve of PID temperature controller Vs Fuzzy temperature controller in oil tank temperature controller

Here settling time reduced by 10 minutes from last fuzzy model. Finally fuzzy model give fuzzy output response with lesser oscillations. This fuzzy model reduces the settling time by 3 hours and 15 minutes. Comparing first fuzzy model in Fig 4.2 and last fuzzy model in Fig 4.7 developed for oil tank temperature controller, an analysis is made that by increasing the number of membership functions from 5 to 7 membership

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

285

functions for error input, for change in error input from 4 membership functions to 6 membership functions and that of fuzzy output from 5 membership functions to 9 membership functions, a response curve has been obtained that has a settling time of 1hour 55 minutes, and oscillations in response curve are all most removed. This fuzzy model reduces the settling time by 3 hours and 15 minutes. IV. CONCLUSIONS Aiming at characteristic of agro plants and control requirement, a Fuzzy-PID hybrid controller with advantages of both fuzzy controller and PID controller integrated is presented in this paper. The available field application shows Fuzzy-PID hybrid controller can not only restrain the large fluctuation to temperature effectively, but also has excellent static performance. Fuzzy-PID hybrid controller has decisive effect on keeping stable temperature of agro and provides powerful support for smooth production process. Owing to improving production and super quality product by application of Fuzzy-PID hybrid controller, considerable economy benefit is brought to the enterprise. REFERENCES
[1]. Erdal Kayacan and Okyay kaynak, An Adaptive Grey Fuzzy PID Controller With Variable Prediction Horizon, Tokyo, Japan, pp 760-765, September 20-24, 2006. [2]. B.G. Hu, G.K.I Mann and R.G Gosine, New methodology for analytical and optimal design of fuzzy PID controllers, IEEE Transactions on Fuzzy Systems, Vol. 7, no. 5, pp. 521-539, 1999. [3]. Awang N.I. Wardana, PID-Fuzzy Controller for Grate Cooler in Cement Plant, IEEE Transactions on Fuzzy System, Vol. 32, no.7, pp.1345-1351,2005. [4]. Han-Xiong Li,Lei Zhang, Kai-Yuan Cai, And Guanrong Chen, An Improved Robust Fuzzy-PID Controller With Optimal Fuzzy Reasoning, IEEE Transactions on Systems, Vol. 35, no. 6, 12831292, December 2005. [5]. Is in Erenoglu, Ibrahim Eksin, Engin Yesil and Mujde Guzelkaya, An intelligent hybrid fuzzy PID controller, Proceedings of 20 th European Conference on Modeling and Simulation, 2006. [6]. Leehter Yao and Chin-Chin Lin, Design of Gain Scheduled Fuzzy PID Controller, World Academy of Science, Engineering and Technology, pp.152-1561, 2005. [7]. Zhen-Yu Zhao, Masayoshi Tomizuka, Satoru Isaka, Fuzzy gain scheduling of PID controllers, IEEE Transactions on Systems, man and cybernetics, Vol. 23, no. 5, September/October 1993, pp. 1392-1398. [8]. Nagaraj, S. Subha, B. Rampriya, Tuning Algorithms for PID Controller Using Soft Computing Techniques, IJCSNS International Journal of Computer Science and Network Security, VOL.8 No.4, April,2008, pp. 278-281. [9]. Zeyad Assi Obaid, Nasri Sulaiman and M.N. Hamidon, Developed Method of FPGA-based Fuzzy Logic Controller Design with the Aid of Conventional PID Algorithm, Australian Journal of Basic and Applied Sciences, vol 3, no 3 pp. 2724-2740, 2009. [10]. Zeyad Assi Obaid, Nasri Sulaiman, M. H. Marhaban and M. N. Hamidon, Analysis and Performance Evaluation of PD-like Fuzzy Logic Controller Design Based on Matlab and FPGA, IAENG International Journal of Computer Science, pp 37-38. 2001. [11]. Nasri Sulaiman, Zeyad Assi Obaid, M. H. Marhaban and M. N. Hamidon, FPGA-Based Fuzzy Logic: Design and Applications a Review, IACSIT International Journal of Engineering and Technology Vol.1, No.5, December, 2009, pp. 491-503. [12]. Chao-Ying Liu, Xue-Ling Song, Zhe-Ying Song, Ying-Cai Sheen, Control System Design of Heat Exchange Station based on fuzzy

[13].

[14].

[15].

[16].

[17].

[18].

[19].

[20].

technology, Proceedings of Fifth International Conference on Machine Learning And Cybernetics, Dalian, pp. 380-384, 13-16 August, 2006. Seema Chopra, R. Mitra, Vijay Kumar, Auto tuning of fuzzy PI type controller using fuzzy logic, in the Proceedings of IJCC, Vol. 6, no.1, pp. 12-18, March 2008. Seema Chopra, R. Mitra, Vijay Kumar, Neural network tuned fuzzy controller for MIMO system, Proceedings of IJCSSE, pp. 78-85, 2:1, 2007. Manish Kumar, and Devendra P Garg, Intelligent Learning Of Fuzzy Logic Controllers Via Neural Network And Genetic Algorithm, Proceedings of 2004 JUSFA 2004 Japan USA Symposium on Flexible Automation Denver, Colorado, pp. 1-8, July 19-21, 2004. Herrera, F., Lozano, M., Verdegay, J. L., Tuning Fuzzy Logic Controllers by Genetic Algorithms International Journal on Approximate Reasoning 1995, 12, pp. 299-315. Lee, M.A., Takagi, Integrating Design Stages of Fuzzy Systems Using Genetic Algorithms, Proc. 2nd IEEE International Conference on Fuzzy Systems, San Francisco, pp. 612-617, 1993. Belarbi K; Titel F, Genetic algorithm for the design of a class of fuzzy controllers: An alternative approach, IEEE Transactions on Fuzzy Systems, , Vol 8, no 4, pp. 398-405, 2000. Park, Y. J., Cho, H.S., Cha, D.H., Genetic Algorithm-Based Optimization of Fuzzy Logic Controller Using Characteristic Parameters, Proceedings of IEEE International Conference on Evolutionary Computation, pp 831-836. Cheong, F., Lai, R., Constraining the Optimization of a Fuzzy Logic Controller Using an Enhanced Genetic Algorithm, IEEE Transactions on Systems, Man and Cybernetics-Part B: Cybernetics, Vol 30, no.1, Feb 2000.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

286

User Controlled Switch Mode Power Supply


1
1, 2, 3, 4
1

Sarpreet Kaur, 2Sakshi Dhingra, 3Bhavesh Singhal, 4Vibhor Gupta


2

University Institute of Engineering & Technology, Panjab University, Chandigarh


sakshi_dhingra18@yahoo.com, 4 vibhoruiet@gmail.com
3

sarpreetdua@yahoo.co.in,

bhaveshsinghal@hotmail.com,

Abstract Switch Mode Power Supplies (SMPS) are efficient devices used to provide energy alternative solutions when the main source is not available. Since the electrical energy is saved in the form of DC in batteries, the SMPS provides a link for power source in required standards. In this manuscript, system has been designed and simulated to develop a user controlled switch mode power supply in order to maintain portability standards for SMPS at global level of load usage. The features proposed through this article includes, incorporation of voltage and frequency control, logic control for breaker switching, and load changing options. Each control system is realized by a user input option or Embedded matlab function block. These are realized in hardware, by using a microcontroller, in which inputs are fed or programming is embedded. These systems are modeled in MATLAB SIMULINK 7.10, and results are shown in order to explain the working of proposed SMPS. Keywords Switch Mode Power Supplies (SMPS)

given as per the user requirement. But for AC loads, this DC voltage from the battery is inverted using an Inverter Circuit, converted into a pulsating AC wave. A square to sine wave filter then provides complete sinusoidal AC as per the user voltage requirement. In this mode the battery discharging phenomenon takes place. II. MODE 1- ACTIVE MODE The basic blocks in the active mode are: (a) 220Volts AC mains supply, (b) Step down transformer, (c) Rectifier, (d) DC load, (e) AC load, (f) DC Battery (charging phenomenon), (g) user input for voltage, as shown in the Fig1 block diagram. This section has an AC main section and the Rectifier section. A three winding transformer is connected to this supply. After transformer action first secondary winding gives 220 volts and is used to feed the AC load. Second secondary winding voltage is stepped down to 24 Volts AC. This voltage is rectified using a diode rectifier to feed the DC load. The DC from the rectifier is not pure therefore a capacitor is provided for the filtering process.

I. INTRODUCTION A Switched-Mode Power Supply is an electronic power supply that incorporates a switching regulator in order to be highly efficient in the conversion of electrical power. Like other types of power supplies, an SMPS transfers power from a source like the electrical power grid to a load while converting voltage and current characteristics. This article includes the modelling of a User Controlled Switched Mode Power Supply using Simulink MATLAB. This SMPS circuit is versatile. It provides the following features: 1. It provides voltage to AC load as per user requirement (0-220 Volts) with user defined frequency 2. It rectifies the AC voltage to multi output DC voltage for DC load, as per user requirement (0-24 Volts) 3. A battery is provided which can feed both AC (0-220 Volts) as well as DC load (0-24 Volts) when mains is switch off. The block diagram of the SMPS is given in the Fig1. The SMPS works in 2 modes: 1. Active Mode 2. Battery Mode Active Mode: In Active Mode the SMPS is connected to the main supply of 220 Volts. AC load can be directly fed from this supply according to the user voltage. For DC load, this voltage is stepped down to 24 Volts AC, and then rectified using a Diode Rectifier to DC providing DC voltage as per user requirement. This DC voltage is also the source for charging a DC battery which comes into action when mains supply is switched off. In this mode battery charging phenomenon takes place. Battery Mode: In Battery Mode, the main supply is off. For DC loads, the voltage from the battery is directly

Fig. 1: The Block Diagram of the User Controlled Switched Mode Power Supply

A) Rectifier: For the rectification process, a full wave bridge diode rectifier is used. The four diodes are arranged in series pairs with only two diodes conducting current during each half cycle. During the positive half cycle of the supply, two diodes in a half cycle conducts in series while the other two are reverse biased and the current flows through the load. During the negative half cycle of the supply, next two set conduct in series, but other two switch "OFF" as they are now reversing biased. The current flowing through the load is the same direction as before. The standard bridge rectifier is modeled, using a capacitor to

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

287

filter ripples.[5] Rectifier simulations are depicted in Fig2. AC supply (220 V), step down AC voltage (24 V) and the rectified DC is shown in the graph.
300 200 100
voltage

A) Battery Discharging: In the absence of mains supply, when the battery is fully charged, it then discharges to feed the load. The Battery Characteristics at Nominal voltage of 24V at rated capacity of 2Ah using Simulink MATLAB are given in Fig4[3].
Nominal Current Discharge Characteristic at 0.2C (0.5A) 28 26 24 22 20 18 0 1 Discharge curve Nominal area Exponential area

-100 -200 -300


220V mains 24V AC (step down sec.) Rectif ied 24V DC (f iltered)

Voltage

2 3 Ampere-hour (Ah) E0 = 24.538, R = 0.096, K = 0.2697, A = 1.6416, B = 360 28 26

Voltage

0.2

0.5

2 time

24 22 20 18 0 1 2 Ampere-hour (Ah) 3

1.5 A 3A

Fig. 2: Simulations of the Rectifier Mode

B) Battery Charging: A DC battery is provided in SMPS so that, when mains supply is switched off, the battery is then used to feed the load. In Active mode, when supply is there, battery charging phenomenon takes place. The typical charge characteristics of Lead acid, Lithium Ion, NiMH and NiCD batteries are given in the Fig3[1].

Fig. 4: Discharging Characteristics of battery with following parameters


[2]

E0 = Constant voltage (V) EBatt = Nonlinear voltage (V) Exp(s) = Exponential zone dynamics (V) Sel(s) = Represents the battery mode. Sel(s) = 0 during battery discharge, Sel(s) = 1 during battery charging. K = Polarization constant (Ah-1) or Polarization resistance (Ohms) i* = Low frequency current dynamics (A) i = Battery current (A) it = Extracted capacity (Ah) Q = Maximum battery capacity (Ah) A = Exponential voltage (V) B = Exponential capacity (Ah)

Fig.3: Typical Charge characteristics of Lead acid, Lithium Ion, NiMH and NiCD batteries[2]

III. MODE 2- BATTERY MODE The battery mode is the offline mode. This section comes into picture when the mains supply is off. The battery discharges and feeds the load. This mode consists of following blocks (a) Battery Discharging Mode (b) Inverter Mode (c) AC load (d) DC load

B) Inverter Mode: The battery can provide only the DC load; therefore to feed the AC load, an Inverter is used which converts the DC voltage (24 Volts) from the battery to AC voltage (24 AC, which is converted to 220 Volts using a step up Transformer). In this SMPS Full wave inverter has been used. This inverter gives square wave of 24 V as the output. Hence a square to sine wave converter (LC filter) is applied for pure sinusoidal output[4]. The pure sine wave obtained is at high voltage (36.7443 V) hence we use an attenuator to bring it back to 24 V. This voltage is either stepped up to 220 V or user input voltage using a digital potentiometer explained later. The frequency of the output wave depends upon the firing pulses given to the thyristors. The time period of the pulses shall change the firing of the thyristors. The user will enter the desired frequency, given as input to the microcontroller. The frequency shall be converted to corresponding time period. Using pulse width modulation, the logic output from controller will fire the thyristors. For very high load options, GTO (Gate turn off thyristor) should be preferred, as it is designed to provide self commutation on reverse voltage. Using L-C filter the user defined frequency AC pulse shall be converted to sinusoidal waveform. However, LC filters works for high frequency

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

288

only. Since the set of standard frequencies are high, so LC is preferred. RC circuit may also be used for low frequencies. The Simulink Model for inverter is given in Fig5 and its simulations are given in the Fig6.

Fig. 7: For user voltage as 16 V

Fig. 5: Inverter Circuit with input user voltage block


40 30 20

Voltage

10 0

-10 -20 -30 -40 0


LC filtered sine wave inverted square wave 24V attenuated sine wave

0.5

1 Time

1.5

Fig. 6: Inverter Simulations

User Input voltage: This switched mode power supply has advantage over other supplies that it also provides an option for user input voltage. For this unit weve used an Embedded MATLAB function integrated with the inverter circuit. In hardware implementation we can use a microcontroller. The Embedded function acts a digital potentiometer and its programming is done as: Function [v_user1, original_sine] = user_voltage(v,v_user) amp=36.7443; temp=v/amp; original_sine=temp*24; v_user1=temp*v_user; Following graph is the simulation for v_user=16 volts
40

IV. SWITCHING TESTS The basis of the switch mode power supply is the capability of the system to feed the load by switching the operation to the DC battery source. The main switching phenomenon comes into the picture when the supply of the mains is cut off and the system is loaded to the charging battery. The battery discharges along the inverter and feeds the AC load. This phenomenon or the simulation are shown in Fig8, the devices used for switching is the ideal switch, which on giving HIGH pulse switches ON and on giving LOW pulse switches OFF. To maintain the pulse cycle TIMER block is used. The system would work as follows: 1. AC Mains OFF then Main AC load switch OFF and Battery charging switch OFF. 2. AC Mains OFF then secondary AC load switch ON, DC load switch ON and Battery discharge ON. The timings of the switching may be changed by having more timer block synchronized with each other. As for now, only one timer is used. The conditions 1 and 2 are opposite to each other. To implement this logic for switches, an embedded MATLAB function may be used. The programming is given below. Function [y, z] = fcn(t) t1=abs (t-1); y = t; z= t1; Switching phenomenon is shown in Fig 10 for following modes Active Mode: (i) Switching 1: Mains (user dependent switch) and AC mains load directly synchronized (ii) Switching2: Battery Charging works after switch 3 (iii) Switching3: DC load after rectification works after switch 1 Inverter Mode: (i) Switching 4: Switch1 (OFF) + Switch 2 (OFF) + Switch3 (OFF) then works after some time Battery Discharging (ii) Switching 5: DC load + to inverter (iii) Switching 6: AC load on inverter mode After Switch 5
1 0.5 0 1 0.5 0 1 0.5 0 0 0.2 0.4 0.6 Switch 3 & 4 0.8 2.0 0 0.2 0.4 0.6 Switch 2 0.8 2.0 0 0.2 0.4 0.6 Switch 1 0.8 1.0

20

1 0.5 Switch 5 0 0.2 0.4 0.6 0.8 2.0

Voltage

0 1 0.5

Switch 3 0 0.2 0.4 0.6 0.8 2.0

-20

0
LC filtered sine wave 24V sine wave user voltage=16V

1 0.5 Switch 6 0 0.2 0.4 0.6 0.8 2.0

-40

0.2

0.4

0.6

0.8 Time

1.0

1.2

1.4

1.8

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

289

Fig. 8: Switching Phenomenon

The debouncing of the switches used in the circuit causes transients which are of less magnitude and less time but they introduce harmonics. Therefore the switch maybe opened after delay. Such that when two circuits are opened at same time, they are not isolated. Delay can be given using a microcontroller using its clock cycle. In hardware the delay is given using optocoupler or relay. Here ideal switch is used therefore no delay is given. If given then we can use Discrete Fourier transform or scalar delay (off on delay) blocks available in new version of MATLAB 7.10. V. RESULTS AND CONCLUSION The simulations of the various systems have been shown and explain in the respective sections. The main idea to develop the SMPS with user control was to introduce flexibility in operation as well as portability. As per various outputs depicted, the proposed design is able to show the desired expectations. Conclusion can be made that user controlled SMPS can be made depending upon the working voltage and the set of desired frequencies, in which system can work. All the operation is user defined, in continuous state of operation.
REFERENCES [1] Tremblay, O.; Dessaint, L.-A.; Dekkiche, A.-I., A Generic Battery Model for the Dynamic Simulation of Hybrid Electric Vehicles Vehicle Power and Propulsion Conference, 2007. VPPC 2007. IEEE 9-12 Sept. 2007, pp. 284-289. MATLAB SIMULINK v7.10 - Simpower system toolbox (functions, model blocks and help). C. M. Shepherd, "Design of Primary and Secondary Cells - Part 2. An equation describing battery discharge" Journal of Electrochemical Society, Volume 112, Jul. 1965, pp. 657-664 P.S. Bhimbra, Introduction to Power Electronics, Chapter 8Inverters P.S. Bhimbra, Introduction to Power Electronics, Chapter 3Diode Circuits and Rectifiers

[2] [3]

[4] [5]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

290

Automatic Reclosing - Transmission Lines Applications and Its Considerations


1
1

Gagandeep Kaur Gill, 2Nishu Gupta


GNDEC, Ludhiana, Punjab, India
2

1,2

gagangill_87@yahoo.co.in,

er.nishugupta@gmail.com

Abstract- Autoreclosing success rates vary from one company to another, it is clear that the majority of faults can be successfully cleared by the proper use of tripping and autoreclosing. This de-energizes the line long enough for the fault source to pass and the fault arc to de-energize, then automatically recloses the line to restore service. Thus, autoreclosing can significantly reduce the outage time due to faults and provide a higher level of service continuity to the customer. Furthermore, successful high-speed reclosing on transmission circuits can be a major factor when attempting to maintain system stability during fault clearing, as we will see later in this paper. The remainder of this paper covers the use of autoreclosing, primarily on transmission lines, and the issues to consider for proper application of autoreclosingThe purpose of this paper is to collect the various topics of protection that are associated with reclosing and present them here for use in applying autoreclosing to transmission circuits. KeywordsPower Quality, Power System Monitoring, Voltage sag, Harmonics, Under voltage, Overvoltage, Transients, Voltage Flicker, Magnetic synthesizer, Super Conducting Storage Devices.

Permanent faults are those that will not clear upon tripping and reclosing. An example of a permanent fault on an overhead line is a broken wire causing a phase to open, or a broken pole causing the phases to short together. Faults on underground cables should be considered permanent. Cable faults should be cleared without autoreclosing and the damaged cable repaired before service is restored. There may be exceptions to this, as in the case of circuits composed of both underground cables and overhead lines, as we will explore later. For those faults that are permanent, autoreclosing will reclose the circuit into a fault that has not been cleared, which may have adverse affects on system stability (particularly at transmission levels). II. HISTORY Automatic reclosing was first applied in the early 1900.s on radial feeders protected by instantaneous relays and fuses. These schemes reclosed the circuit two or three times prior to lockout, with a 73% to 88% success rate on the first reclose actions, and covered both radial and looped circuits, predominantly at distribution voltages, but also including 154kV. High-speed reclosing (HSR) was first used by American Electric Power System (then known as American Gas & Electric) in 1935 as a means to defer construction of redundant transmission lines. System continuity was maintained on these radial lines by rapidly reclosing a single line rather than providing a second, redundant path for power to flow. Modern systems with single radial lines to transmit power from one point to another are commonplace. It is more common to have a network with parallel transmission lines. HSR is used more for maintaining system stability and synchronism than for point-to-point continuity. The development of high-speed breakers for transmission lines by the late 1930's led to the application of high-speed reclosing (HSR) on these lines, resulting in improved system stability. Probability studies of the insulator flashover were initiated to determine minimum reclosing times that still permitted enough time for arc de-ionization. Early applications of HSR on multi-terminal lines tripped all terminals and then reclosed the circuit breaker at highspeed at one terminal. If this high-speed reclosure was successful, the remaining terminals were reclosed with time delay to complete the through circuit. The preceding historical information touches on a number or reasons for using autoreclosing on both distribution and transmission systems. Following is a summary of reasons for using

H.B.1 INTRODUCTION Various studies have shown that anywhere from 70%, to as high as 90%, of faults on most overhead lines are transient. A transient fault, such as an insulator flashover, is a fault which is cleared by the immediate tripping of one or more circuit breakers to isolate the fault, and which does not recur when the line is re energized. Faults tend to be fewer transients (near the 80% range) at lower, distribution voltages and more transient (near the 90% range) at higher, subtransmission and transmission voltages. Lightning is the most common cause of transient faults, partially resulting from insulator flashover from the high transient voltages induced by the lightning. Other possible causes are swinging wires and temporary contact with foreign objects. Thus, transient faults can be cleared by momentarily de-energizing the line, in order to allow the fault to clear. Auto reclosing can then restore service to the line. The remaining 10 - 30% of faults are semi permanent or permanent in nature. A small branch falling onto the line can cause a semi-permanent fault. In this case, however, an immediate de-energizing of the line and subsequent auto reclosing does not clear the fault. Instead, a coordinated time-delayed trip would allow the branch to be burned away without damage to the system. Semipermanent faults of this type are likely to be most prevalent in highly wooded areas and can be substantially controlled by aggressive line clearance programs.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

291

autoreclosing. This list may not be complete, and each engineer needs to consider any additional reasoning when applying autoreclosing in any given application. 1 Minimizing the interruption of the supply to the customer. 2 Maintenance of system stability and synchronism (high-speed tripping/autoreclosing on transmission lines). 3 Restoration of system capacity and reliability with minimum outage and least expenditure of manpower. 4 Restoration of critical system interconnections. 5 Restoration of service to critical loads. 6 Higher probability of some recovery from multiple contingency outages. 7 Reduction of fault duration, resulting in less fault damage and fewer permanent faults. 8 The use of high-speed tripping and autoreclosing schemes in fuse saving schemes to prevent permanent outages for transient faults beyond tap fuses. 9 The use of delayed tripping and autoreclosing schemes in fuse blowing schemes to allow time delayed tripping to clear semi-permanent faults. 10 Ability to run substations unattended, resulting in saved wages. 11 Relief for system operators in restoration during system outages. III. FUNDAMENTALS The application of autoreclosing requires the evaluation of many factors. These factors may vary considerably depending upon the system conFiguration, the system components, and the reclosing philosophy utilized by the protection engineer or company. The following factors are of fundamental concern: 1. The benefits and possible problems associated with reclosing 2. The choice of dead time 3. The choice of reset time 4. The decision to use single- or multiple-shot reclosing The factors noted above are fundamental when evaluating autoreclosing applications. Decisions when choosing these in autoreclosing applications are influenced by the type of protection and switchgear used the nature of the system, and the possibility of stability problems, and the effects on various consumer loads. Distribution networks and transmission systems present some similar and some different problems in respect to the application of autoreclosing. Arcing time (of a mechanical switching device) The interval of time between the instant of the first initiation of the arc and the instant of final arc extinction in all poles. Breaker reclosing time The elapsed time between the energizing of the breaker trip coil and the closing of the breaker contacts to re establish the circuit by the breaker primary contacts on the

reclose stroke. (i.e., breaker operating time plus breaker dead time). Closing time (of a mechanical switching device) The interval of time between the initiation of the closing operation and the instant when metallic continuity is established in all poles. Dead time (of a circuit breaker on a reclosing operation) The interval between interruption in all poles on the opening stroke and re establishment of the circuit on the reclosing stroke. The dead time of an arcing fault on a reclosing operation is not necessarily the same as the dead time of the circuit breakers involved, since the dead time of the fault is the interval during which the faulted line is de-energized from all terminals. De-ionizing time The time following the extinction of an overhead line fault arc necessary to ensure dispersion of ionized air so that the arc will not re-strike when the line is re-energized. Delayed autoreclosing The autoreclosing of a circuit breaker after a time delay that is intentionally longer than for high-speed auto reclosing. High-speed autoreclosing The autoreclosing of a circuit breaker after a necessary time delay (typically less than one second) to permit fault arc de-ionization with due regard to coordination with all relay protective systems. This type of autoreclosing is generally not supervised by voltage magnitude or phase angle. Operating time (circuit breaker) The time from the energizing of the trip coil until the fault arc is extinguished. Operating time (protection) The time from the inception of the fault to the closing of the tripping contacts. Where a separate auxiliary tripping relay is employed, its operating time is included. Reset time (of an automatic circuit recloser or automatic line sectionalizer) The time required, after one or more counting operations, for the counting mechanism to return to the starting position. System disturbance time The time between the inception of the fault and the circuit breaker contacts making on successful reclosing. Single-shot reclosing An operation sequence providing only one reclosing operation, lockout of the circuit occurring on subsequent tripping.
IV. AUTOMATIC RECLOSING ON TRANSMISSION AND SUBTRANSMISSION SYSTEMS

The voltage classes considered as transmission and sub transmission levels are: Subtransmission 34.5 kV - 138 kV Transmission 115 kV and higher with transmission divided into: High Voltage (HV) 115 - 230 kV

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

292

Extra High Voltage (EHV) 345 - 765 kV Ultrahigh Voltage (UHV) 1000 kV and higher The voltage values indicated represent the nominal and typical rms system voltages (line to-line) . These classes are general and may vary from one system to another, as well as overlap from one class to another. 4.1 Transmission and Subtransmission Overview The sub transmission and transmission systems are generally accepted as those circuits with voltages of 34.5 kV and higher, although the lines between distribution and sub transmission/transmission are not always clear. Transmission and sub transmission lines are more likely to be looped interconnected systems, meaning that the line has a positive-sequence source at two or more ends. Fault current to line faults is supplied by the source terminals, and all source terminals must be tripped to clear both phase and ground faults. 4.2 Application of Auto reclosing on Transmission Systems A primary concern in the application of autoreclosing, especially on EHV-rated lines and higher, is the maintenance of system stability and synchronism. This is normally done through the application of high-speed tripping and autoreclosing. The problems involved with maintaining stability on these lines when autoreclosing during a fault on the line depend on the characteristics of the system - whether it is loosely connected, for example, with two power systems connected by a single tie line, or, conversely, highly interconnected, in which case maintaining synchronism during autoreclosing is much easier. The intent of autoreclosing on transmission and sub transmission systems, other than the maintenance of stability, is to return the system to its normal conFiguration with minimum outage of the line with the least expenditure of manpower. System restoration becomes increasingly important when applied to lines that interconnect systems and are critical for reliable power exchange between the systems. Individual utility policy and system requirements dictate the complexity and variety of automatic reclosing schemes in service today. V. HIGH-SPEED AUTORECLOSING ON TRANSMISSION AND
SUBTRANSMISSION SYSTEMS

developed from empirical data gathered from laboratory tests and field experience. 3. The protection characteristics 4. The circuit breaker characteristics and limitations. 5. Choice of reclose reset time 6. Number of reclose attempts Protection Characteristics On transmission lines where stability is a concern, simultaneous tripping of both circuit breakers ensures the quickest arc de-ionization of the fault. Any time during which one circuit breaker is open in advance of the other represents an effective reduction in the dead time, and may jeopardize the chances of a successful reclosure Simultaneous tripping, in conjunction with high-speed reclosing, keeps the system disturbance time to a minimum. Simultaneous tripping can be accomplished by use of high-speed distance relays. When distance relays are used and the fault is near one end of the line, special measures need to be taken to ensure simultaneous tripping of each end Where distance relaying of this type is not feasible, some form of pilot relaying can be used. While it is important to autoreclose on lines using distance or pilot relaying, it is desirable that reclosing should be blocked and the breaker should remain open for out-of-step conditions. An out-of-step condition is generally an indication that the power swing is too great to maintain synchronism of the two separated systems; therefore, autoreclosing will not be effective and should be blocked. If single-pole switching is used, there must be a method for the relays to properly detect the faulted phase and trip the proper poles accordingly. Circuit Breaker Characteristics In order to interrupt faults that are permanent, circuit breakers used with high-speed reclosing must have an interrupting duty capable of interrupting faults twice or more in rapid succession. This requires evaluation, and possible de rating, of the breaker from the standard duty of two operations 15 seconds apart. Circuit breakers used for high-speed reclosing are fitted with operating mechanisms and control circuits that will automatically reclose at high speed and, if necessary, trip a second time. These breakers are often designed with special mechanisms that give higher speeds than are attainable with standard closing solenoids. Special control circuits on breakers are used for high-speed autoreclosing. After the breaker has been tripped by the protective relays, the trip coil is deenergized and the closing coil is energized well before the end of the opening stroke, thus reversing the motion of the piston and breaker contacts Number of Reclose Attempts High-speed reclosing on transmission and sub transmission systems where stability is a concern is invariably single shot. Repeated attempts of reclosing with high fault levels would have serious effects on system stability. Furthermore, the incidence of semi-permanent

High-speed auto reclosing, used in conjunction with high-speed clearing of faults, is used on transmission and sub transmission for improving stability. Factors to consider when using high-speed auto reclosing include: 1. The maximum time available for opening and reclosing the system without loss of synchronism (maximum dead time). This time is a function of the system conFiguration and the transmitted power. 2. The time required for de-ionization of the arc path so that the arc will not restrike when the breaker is reclosed. This time can be estimated by the use of a formula

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

293

faults that could be cleared by repeated reclosures is to be less likely than on distribution systems. VI. DELAYED AUTORECLOSING ON TRANSMISSION AND SUBTRANSMISSION SYSTEMS On highly interconnected transmission and sub transmission systems, where the loss of a single line is unlikely to cause loss of synchronism between the two ends, delayed autoreclosing can be used. In this case, the dead time is allowed to be long enough for any power swings on the system to settle down before reclosing. Thus, the problems of fault arc de-ionization times and circuit breaker operating characteristics are eliminated. Where delayed autoreclosing is used on transmission systems, it is usual practice to use a synchronism check relay in the reclosing scheme. Even though the tripping of the line is unlikely to cause a loss of synchronism, there may be a voltage and/or phase difference developed between the two ends of the tripped line, which might cause problems if reclosed out of phase. Synchronism check relays generally check for phase angle, voltage and frequency difference when employed in autoreclosing schemes. On a line of this type, it is common practice to reclose the breaker at one end first, a process known as .dead line charging.. Reclosing on the other end is then under the control of the synchronism check relay for .live line reclosing VII. THREE PHASE VERSUS SINGLE PHASE AUTORECLOSING If single-phase autoreclosing is used on a transmission line for, for example, a single-line to- ground fault, tripping only the faulted phase will allow an interchange of synchronizing Power that would otherwise be unavailable through the use of three phase autoreclosing. In some installations, all three poles operate on any fault other than a single-line-to-ground fault. In other cases, selective-pole tripping is used not only for single-line faults, but also for line-to-line and two-line-to ground faults. When two conductors are open, some power is carried on the remaining conductor with a ground return. The stability limit of the line can be raised above the limit obtainable with three-pole tripping and reclosing at the same speed. Alternatively, the same stability limit can be achieved with slower autoreclosing. Single pole switching also has the advantage of reducing mechanical shock to generators compared to three phase reclosing. A disadvantage of single pole switching is that each pole in the breaker must have its own tripping and closing mechanism, and the relay scheme must be able to properly select and trip the faulted phase or phases. VIII. AUTOMATIC RECLOSING APPLICATION CONSIDERATIONS 1) Effects of Auto reclosing on Breaker Interrupting Ratings

The design of the power circuit breaker has evolved over the years and has undergone many improvements in design, mechanism speed, and operating reliability. These improvements, as well as protective relay development and scheme sophistication, have led to higher speed, higher interrupting ratings, and longer duty ratings for the applications employed today. When developing a reclosing philosophy, the limitations of the breaker to interrupt the fault must be considered. Circuit breaker interrupting capabilities are defined based on the Rated Standard Operating Duty (Standard Duty Cycle). The standard duty cycle, as defined by IEEE C37.04-1979, is 2 operations with a time interval of 15 seconds between operations (CO + 15 sec. + CO). This means that the breaker can operate twice to interrupt its symmetrical interrupting capability current with 15 seconds of dead time between operations. The speed of the breaker when interrupting the fault is also important, especially where stability is critical. Therefore, breakers also have rated interrupting times. The rated interrupting time of a circuit breaker is the time between the trip circuit energization and power arc interruption on an opening operation, and is used to classify breakers at different speeds. 2) Lines with Generators On single-tie circuits with dispersed generation, reclosing on the circuit must be delayed long enough for the dispersed generation to be isolated from the utility. If this does not happen, the generator may be damaged due to the utility source closing into the generator out of synchronism. As an additional safety factor, where there is customer generation, voltage supervision is often applied to the autoreclosing scheme. In this case, autoreclosing is delayed until a dead line is sensed (also known as live line blocking, or LLB), thus preventing reclosing into the dispersed generation. If the dispersed generator has the capacity to maintain the connected load, it may be used to do so in the event that the utility tie is lost. In this case, the dispersed generation needs to have the ability for dead line closing. In addition, before the utility tie is re established, this generation must be isolated from the utility to prevent the utility from damaging the generator when reenergizing. This can be done either locally or remotely. The generation can also be tied back to the utility system using synchronism check. If the generation capacity is insufficient to supply the connected load, it should be removed from the system upon a trip of the utility supply and prior to the utility reclosing 3) Lines with Motors Switching operations on motor loads, both induction and synchronous, can produce high transient torques on the motor and, thereby, cause damage to or destruction of the motor. One example may be in an industrial plant with critical induction motor loads that have a primary bus to supply the motors and an auxiliary bus with a separate supply. If the voltage on the main bus is lost, a fast bus transfer is made to the auxiliary in order to maintain the

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

294

critical motor load. Large torques can result. If this rapid transfer is mandatory, there are safe limits that need to be considered for reconnection of motors. If rapid transfer is not critical, the best policy is to delay the re-energization of the induction motors until the motor voltage has dropped to a safe level. Levels of 33% or less are in common use. 4) Line with Cables Faults on lines that are underground cable tend to be permanent in nature. Thus, reclosing on completely underground lines is not generally used, as doing so is likely to aggravate the damage. Circuits comprised of both underground cable and overhead lines could have autoreclosing depending on the utility practice. In this case, the number of reclosing operations may be reduced to a single shot. The basis for determining whether autoreclosing should be used is usually based on the possibility of the fault occurring on the overhead portion of the line and, thus, being able to reclose successfully. In some instances, where a small portion of the circuit from the substation is cable and a larger portion beyond this is overhead, an autoreclosing scheme that blocks reclosing for close-in faults (e.g., on the cable) may be used. Another approach is to install separate relaying on the cable portion to block reclosing for a fault on the cable such as current differential or pilot wire relaying. This may be cost prohibitive as there would also be the need for some form of communication channel connecting each end of the cable protection, freestanding current transformers on the line at the cable/overhead line transition, etc. 5) Lines with Automatic Sectionalizing A sectionalizer is a circuit-isolating mechanism that is not rated to interrupt fault current. It will typically open while de-energized after counting a number of fault current pulses or on loss of potential. It will be closed either manually or, after a time delay, on restoration of potential. Sectionalizing applications on transmission lines are similar to those on distribution lines. The line usually feeds tapped loads. By using sectionalizing schemes, a permanent fault can be isolated, and maintain power to these loads. Successful sectionalizing requires that both reclosing and reset times of the reclosing relays associated with the line breakers and the line sectionalizing equipment coordinate. IX. CIRCUIT BREAKER MODEL The circuit-breaker model is shown in Fig1. Current icb through the circuit-breaker flows between I/O pins cb+ and cb-, passing through the voltage source Vsense, voltage-controlled voltage source E(arc) and voltagecontrolled switch cbmod1. Vsense acts as an ideal current meter. To model the thermal characteristic of the circuit breaker, the current icb measured by Vsense is passed to the current-controlled current source G(i*i), which outputs a current equal to icb raised to the power n, whenever icb exceeds the rated current ir of the circuit breaker

Fig. 1: Circuit Breaker Model

Model parameter values are given in Table 1, based on typical measured characteristics of a 125A circuit breaker.
TABLE

1: 125A CIRCUIT-BREAKER MODEL PARAMETERS

Va ir n Ccb im1

initial arc voltage rated operate current current-time thermal curve fit in t value @ i=10ir

13V
125A

3.5 6.9 E10 An.sec 2,000 A

current level to start magnetic operation current level to end 2,500 A im2 magnetic operation DC resistance with 0.8 m Rd contacts closed Measured voltage and current waveforms of a 125A circuit-breaker operating in both a DC high-current test facility and a distributed power system rack were used to validate the model when operating in the magnetic region. Operation in the thermal region is not shown, as this mode is typically of secondary importance when investigating over-current protection in telecommunications power systems. Figures 2 and 3 show simulated waveforms of the circuit-breaker model operating in circuits with equivalent characteristics to the test circuits used to obtain the measured waveforms Simulated results show quite good agreement with measured results considering the complex physical arcing process that takes place during circuit-breaker operation. The major area of discrepancy is the overvoltage transient generated as the arc extinguishes. It should be noted that minor waveform variations have been observed with repeated tests under the same test conditions.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

295

Fig. 2: Simulated current and voltage waveforms for a 125A circuitbreaker operating in 54VDC circuit with 5.2kA prospective current and 0.25ms prospective time constant.

Fig. 5: Simulated current and voltage waveforms for a 125A circuitbreaker when protecting a battery string within one of Telstra's distributed power supplies.

Fig 3: Simulated current and voltage waveforms for a 125A circuitbreaker operating in 54VDC circuit with 12kA prospective current and 1ms prospective time constant.

The circuit-breaker model has assisted Telstra's power system designers to analyse the operation of a 125A circuit-breaker operating in a Telstra distributed power system battery rack. The measured current waveform of a circuit-breaker interrupting a short circuit from the 48V battery string negative (active) output terminal to the rack frame is shown in Fig4. No voltage waveforms were taken in this test. Fig5 shows simulated waveforms of the circuit-breaker model operating in a circuit with equivalent characteristics to that used to obtain the measured waveform shown in Fig4. Again, quite good agreement is obtained between the simulated and measured current waveform

X. CONCLUSION A number of the applications considered in this paper includes a requirement for some form of recloser supervision and/or reclose blocking for various situations. Below is a summary of some of these conditions that uses reclose supervision and blocking. Consider blocking auto reclosing for the following conditions: Receipt of transfer trip Manual trip Breaker failure Hot line maintenance Three phase faults Faults on buses - bus differential relay operation Faults on transformers - transformer differential relay operation Faults on underground cables Out-of-step condition Under frequency / under voltage load shedding trips High impedance fault detection on distribution lines High current, close-in faults Line side voltage supervision Auto reclosing will be blocked for sensed voltage on the line. Live line blocking is generally used where large motors or generators are connected to the line. This blocking prevents damage to the motor or generator from being energized out of phase with the system. To ensure complete line side voltage supervision, all three phases should be monitored. ACKNOWLEDGEMENT The authors are highly grateful to M.S. Saini, Principal, Guru Nanak Dev Engineering College, Ludhiana, for providing this opportunity to carry out the paper work at topic titled as AUTOMATIC RECLOSING TRANSMISSION LINES APPLICATIONS AND ITS CONSIDERATIONS .We would like to express a sense of gratitude and thanks profusely to Prof. Arvind Dhingra, Department of Electrical Engineering, GNDEC, Ludhiana. The constant guidance and encouragement has been of great help in carrying out the work. We also want to thank the staff of Electrical Engineering Department, GNDEC, Ludhiana, and all our

Fig. 4: Measured current waveform for a 125A circuit-breaker when protecting a battery string within one of Telstra's distributed power supplies. 667A/div and 0.5ms/div

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

296

friends and family members who directly or indirectly helps us during the course of work. Discussions with them have been incalculably beneficial. Authors owe a great debt to Prof. Jaswinder Singh, Head of Department of Electrical Engineering, GNDEC, Ludhiana for providing this opportunity and for their intellectual support throughout the course of work. The references that have been consulted in bringing out this report are also gratefully acknowledged. REFERENCES
[1] IEEE Power Systems Relaying Committee; Automatic Reclosing of Transmission Lines; IEEE Transactions, Vol. PAS-103, Feb. 1984, no. 2, pages 234 - 245 [2] Protection Relay Application Guide; GEC Measurements, 1975 [3] IEEE Standard Definitions for Power Switchgear; IEEE Std. C37.100-1992 [4] IEEE Application Guide for AC High-Voltage Breakers Rated on a Symmetrical Current Basis; ANSI/IEEE C37.010-1979 [5] IEEE Power Systems Relaying Committee; Guide for Automatic Reclosing for Line Circuit Breakers for AC Distribution and Transmission Lines; Draft document, 1998 [6] [ Kimbark, Edward Wilson, ScD; Power System Stability; John Wiley & Sons, Inc., N.Y., London [7] M.C. Jackson, et al.; Turbine Generator Shaft Torque and Fatigue: Part I Simulation Methods and Fatigue Analysis; IEEE Transactions, Vol. PAS-98, 1979, pages 2299- 2307, Part I [8] M.C. Jackson, et al.; Turbine Generator Shaft Torque and Fatigue: Part II - Impact of System Disturbances and High-speed Reclosing; IEEE Transactions, Vol. PAS-98, 1979, pages 23082313, Part II [9] NPCC; Guide for the Application of Autoreclosing to the Bulk Power System; NPCC, 1979 [10] Blackburn, J. L., et al; Applied Protective Relaying, Westinghouse Electric Corporation, 1982 [11] Elmore, Walter A., et al; Protective Relaying Theory and Applications, Marcel Dekker, Inc., 1994 [12] Blackburn, J. L.; Protective Relaying Principles and Applications - Second Edition; Marcel Dekker, Inc., New York-Basel-Hong Kong, 1998 [13] IEEE Guides and Standards for Protective Relaying Systems; IEEE, Inc., NY, Spring 1991 Edition. [14] Anderson, P.M. and Fouad, A.A.; Power System Control and Stability - Volume 1;The Iowa State University Press, Ames, Iowa, 1977

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

297

Technical Problems Associated with Economic Operation of Wind Farm Integrated into Power System
1
1

Arrik Khanna, 2Anitpal Singh


1,2

Chitkara University, Punjab


2

arrik1433@gmail.com,

er_apsbhamra@yahoo.com

Abstract: Wind farms interconnected to power system bring new challenges to power system economic operation. Wind power is going through a very rapid development. The integration of wind power in the power system is now an issue in order to optimize the utilization of the resource and to continue the high rate of installation of wind generating capacity, which is necessary so as to achieve the goals of sustainability and security of supply. This paper presents the main technical challenges that are associated with the integration of wind power into power systems. These challenges include effects of wind power on the power system, the power system operating cost, power quality, power imbalances, power system dynamics, and impacts on transmission planning. The main conclusion is that wind powers impacts on system operating costs are small at low wind penetrations and at higher wind penetrations. Keywords: Wind power; Power generation; Power systems; Grid integration

I. INTRODUCTION The progress in wind energy technology has been extraordinary. The cost of electricity from wind power has fallen to about one seventh of the cost in the early 1980s [1]. This, added to worldwide growth in energy demand, environmental concerns, the rising cost of fossil fuel generation in many developing parts of the world, and a good wind resource base make wind a competitive option. In the United States, there has been substantial recent growth in wind energy generating capacity, with growth averaging 24% annually during the past 5 years [2]. In India as on 31 Dec 2010 the installed capacity of wind power in India was 13065.37 MW. But as on March 2011 the installed capacity of wind power in India has raised to 14157 MW. A subsequent rise of of 8.35 % in three month In order to facilitate the implementation of wind power projects, various computer algorithms have been developed for the choice and design of wind power and other renewable energy sources [3-6]. For Example Offshore Code Comparison Collaboration Continuation (OC4), HOMER [7] and RET Screen [8]. HOMER allows the designer to compare many different design options based on their technical and economic merits. It also assists in understanding and quantifying the effects of uncertainty or changes in the inputs. RET Screen is dedicated to the preparation of pre-feasibility studies for the evaluation of energy production, life-cycle costs and greenhouse gas emissions reduction for various renewable energy technologies including wind power Grid

integration concerns have come to the fore in recent years, as wind power penetration levels have increased in a number of countries as an issue that may impede the widespread deployment of wind power systems. Three of the strongest challenges to wind powers future prospects are the problems of wind intermittency, grid reliability and voltage stability. The conventional management of transmission and distribution operation is challenged by electricity market restructuring, security of supply concerns and the integration of newer generation technologies such as wind power. Transmission availability can be a barrier to wind power development. Favourable wind locations are often in areas distant from existing transmission. Building new transmission lines can be difficult due to planning barriers, land use rights and costs. Internationally accepted standards for power performance, safety, noise and other environmentalrelated conditions are needed to reduce market barriers, as well as administrative and installation costs. This paper presents the main technical challenges that are associated with the integration of wind power into power systems. These challenges include effects of wind power on the power system, the power system operating cost, power quality, power imbalances, power system dynamics, and impacts on transmission planning. II. EFFECTS OF WIND POWER ON THE POWER SYSTEM Regarding the power system, the drawbacks of wind power are that wind power production is variable, difficult to predict and cannot be taken as given. However, integration of variable sources is much less complicated if they are connected to large power systems, which can take advantage of the natural diversity of variable sources. A large geographical spreading of wind power will reduce variability, increase predictability and decrease occasions of near-zero or peak output. The power system has flexible mechanisms to follow the varying load that cannot always be accurately predicted. As no production unit is 100% reliable, part of the production can come from variable sources, with a similar risk level for the power system. The impacts of wind power in the electricity system depend to a large extent on the: Level of wind power penetration; Grid size; and Generation mix of electricity in the

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

298

system Wind energy penetration at low to moderate levels is a matter of cost, as demonstrated by various national and regional integration studies. And the integration costs related to the impacts listed above are fairly modest. For low penetration levels of wind power in a system, system operation will hardly be affected. The established control methods and system reserves available for dealing with variable demand and supply are more than adequate for dealing with the additional variability at wind energy penetration levels of up to around 20 per cent, depending on the nature of a specific system. For higher penetration levels, some changes to systems and their method of operation may be required to accommodate the further integration of wind energy. Short and long term impacts The impacts of wind power on the power system can be categorised into short -and long-term effects. The shortterm effects are caused by system balancing at the operational timescale (minutes to hours), whereas the long-term effects are related to the contribution wind power can provide to the system adequacy (its capability to reliably meet peak load situations). Impacts in the system: Local and system wide Locally, wind power plants interact with the grid voltage, just like any other power station. In this context, steady state voltage deviations, power quality and voltage control at or near wind farm sites must all be taken into consideration. Wind power can provide voltage control and active power (frequency) control. Wind power plants can also reduce transmission and distribution losses when applied as embedded generation. On the system-wide scale, there are other aspects to consider. Wind power plants affect voltage levels and power flows in the networks. These effects can be beneficial to the system, especially when wind power plants are located near load centres, and certainly at low penetration levels. For example, wind power plants can support the voltage in the system during fault (low voltage) situations. Also, wind plants that have a reactive power control system installed at the end of long radial lines benefit the system, since they support the voltage in (normally) low voltage quality parts of the grid. Wind power may need additional upgrades in transmission and distribution grid infrastructure, as is the case when any power plant is connected to a grid. In order to connect remote high resource sites, such as offshore or very large Wind plants in remote areas, to the load centres, new lines need to be constructed (just as new build pipelines had to be built for oil and gas). In order to maximise the smoothing effects of geographically distributed wind, and to increase the level of firm power, additional cross-border transmission is necessary to reduce the challenges of managing a

system with high levels of wind power. Wind power requires measures for regulating control, just as any other generation technology and, depending on the penetration level and local network characteristics, it affects the efficiency of other generators in the system (and vice versa). In the absence of sufficient intelligent and well managed power exchange between regions or countries, a combination of (non-manageable) system demands and production may result in situations where wind generation has to be constrained. Finally wind power plays a role in maintaining system stability and contributes to the system adequacy and security of supply. III. WIND POWER IMPACTS ON ELECTRIC POWER SYSTEM
OPERATING COSTS

Wind power plants generate electricity when the wind is blowing, and the plant output depends on the wind speed. Wind speeds cannot be predicted with high accuracy over daily periods, and the wind often fluctuates from minute to minute and hour to hour. Consequently, electric utility system planners and operators are concerned that variations in wind plant output may increase the operating costs of the system. This concern arises because the system must maintain balance between the aggregate demand for electric power and the total power generated by all power plants feeding the system. This is a highly sophisticated task that utility operators and automatic controls perform routinely, based on well-known operating characteristics for conventional power plants, sophisti-cated decision-support algorithms and systems, and a great deal of experience accumulated over many years. In general, the costs associated with maintaining this balance are referred to as ancillary-services costs. System operators are concerned that variations in wind plant output will force the conventional power plants to provide compensating variations to maintain system balance, thus causing the conventional power plants to deviate from operating points that are chosen to minimize the total cost of operating the entire system. The operators concerns are compounded by the fact that conventional power plants are generally under their control and thus are dispatch able, whereas wind plants are controlled instead by nature.

Fig. 1.1: Impacts of Wind Power

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

299

A graphical overview of the various impacts of wind power in the power system is given in Fig1.1 It shows the local and system wide impacts, as well as the short- and long-term impacts, for the various affected aspects of the power system, which include grid infrastructure, system reserves and system adequacy. Although these are valid concerns, it is important to understand that the key issue is not whether a system with a significant amount of wind capacity can be operated reliably, but rather to what extent the system operating costs are increased by the variability of the wind. Recently several investigations of wind power impact on electric power system operating costs have been conducted. These studies addressed utility systems with different generating resource mixes and employed different analytical approaches. The common element of the studies is that they seek to determine the cost of ancillary services necessary to accommodate a wind plant on a utility system. There are typically three time scales of interest, which correspond to the operation of the utility system and the structure of the competitive electricity markets: 1. Unit-commitment horizon of 1 day to 1 week with 1 h time increments, 2. Load-following horizons of 1 h with 510 min increments (intra-hour), 3. Regulation horizon of 1 min1 h with 15 s increments. IV. POWER QUALITY The location and intermittent nature of wind turbine machines can cause power quality problems such as voltage dips, frequency variations, and low power factor. Wind turbines, especially inductive machines, tend to absorb reactive power from the system and produce a low power factor. If wind turbines absorb too much reactive power, the system can become unstable. Poor power quality can cause the end users equipment to operate inefficiently, i.e. lights to flicker, or the utility system becomes unstable and disrupts power to the customer. Power quality problems caused by wind power are best solved at the point of interconnection of the wind generator to the utility grid [18]. New state-of-the-art wind generators utilize power electronics and variable-pitch turbines that allow the wind turbine to produce energy at various wind speeds [19]. The same power electronics regulate theturbines output voltage while keeping the power factor close to unity. The power electronics in the turbine control the voltage and phase angles of the rotors currents to control the output voltage and power factor. In the past, utilities relied on switched capacitors to keep the voltage steady and maintain power factor. However, switching capacitors could only provide fixed amounts of voltampere reactive (VAR) and tended to put stress on the wind turbines gearbox and increase its cost of maintenance. The power electronics in new wind turbines provide the appropriate amount of VARs without stressing the wind turbine gearboxes. In addition to adding power

electronics inside the wind turbines, some utilities are looking to DC and voltage source converter (VSC) to connect the wind turbines to their grid to stabilize the voltage and frequency. V. POWER IMBALANCES The power fluctuation at each wind turbine is affected by the type of turbine, the control algorithm, the wind speed fluctuation, and the tower shadow effect. The power measurement from a single wind turbine usually shows a large fluctuation of output power. Because many turbines are connected in a wind farm, the power fluctuation from one turbine may cancel that of another, which effectively rectifies the power fluctuation of the overall wind farm. As wind energy technologies progress, wind turbines become larger. Manufacturers are currently producing multimega watt wind turbines. Thus, fewer turbines are needed to deliver the same power and the power fluctuation of an individual wind turbine will have a greater impact on the power network. The impact on a weak grid will be even greater. We used a simulation program to investigate the impact of the turbine distribution on a large wind farm. Many researchers have investigated various aspects of electrical power systems on a wind farm. Wind farms with variable speed or fixed speed wind turbines were investigated under varying conditions. The voltage fluctuations as a function of X/R ratio, the reactive power fluctuations on voltage variation, the harmonics components at the point of common coupling (PCC), and the flicker emission out of a wind farm were presented in reference [5] and reference [6]. The flicker emission from a wind farm is reduced as the grid stiffness (Sk/Sn ratio) increases [7]. Also, the flicker emission is affected dramatically by the turbulence intensity. The flicker emission at 16% turbulence intensity is twice as high as a turbulence intensity of 8% [8]. VI EFFECTS OF WIND POWER ON POWER SYSTEM
DYNAMICS

Power system dynamics investigates how a power system responds to disturbances that change the systems operating point. Examples of such disturbances are frequency changes because a generator trips or a load is switched in or disconnected; voltage drops due to a fault; changes in prime mover mechanical power or exciter voltage, and so on. The power system is considered stable if the system reaches a new steady state and all generators and loads that were connected to the system before the disturbance are still connected. The original power system is considered unstable if, in the new steady state, loads or generators are disconnected. The dynamics of a power system are governed mainly by the generators. Thus, if conventional power generation with synchronous generators is on a large scale replaced with wind turbines that use either asynchronous squirrel cage induction generators or variable-speed generation systems with power electronics, the dynamics of the

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

300

power system will at some point be affected, and perhaps its stability, too. Although wind turbines indeed affect the transient and small signal dynamics of a power system, power system dynamics and stability are not a principal obstacle to increasing the penetration of wind power. By taking adequate measures, the stability of a power system can be maintained while increasing the wind power penetration. In the case of constant-speed wind turbines, measures must be taken to prevent voltage and rotor speed instability in order to maintain transient stability. This can be done by equipping them with pitch controllers in order to reduce the amount of over speeding that occurs during a fault; by combining them with a source of reactive power to supply the large amount of reactive power consumed by a squirrel cage generator after a fault, such as a static condenser or static VAR compensator; or by changing the mechanical and/or electrical parameters of the turbine [22]. In the case of variable-speed wind turbines, the sensitivity of the power electronic converter to over currents will have to be counteracted in another way than is presently done, namely, by switching off the turbine. However, the literature seems to indicate that there may indeed be other options. The small signal stability shows that increased levels of wind power penetration do not seem to require any additional measures in order to maintain small signal stability. The generator type that is most likely to engage into power system oscillations is the synchronous generator, and that generator type is not used in the wind turbine types that are presently on the market. Therefore, replacing synchronous generation seems to have either a negligible or a favorable impact on power system oscillations. However, the impact of wind power on the small signal stability of power systems is a rather recent research subject. Hence, results are still very limited. Therefore, the conclusions presented here should be considered preliminary and be used prudently. VII. TRANSMISSION PLANNING Wind turbines are of course mainly installed in places with good wind resources. This can lead to high concentrations of wind power in some regions. This means that large amounts of wind power will be generated in regions where the local consumption is low and the excess has to be exported. In order to do that it is necessary to have a strong transmission grid designed for the transmission of wind power. This basically means that it has the right capacity of power transport from the windy regions to the load centers. This of course is a set of problems that has been known in the past when the transmission grid capacity was designed to fit the needs of power transmission from large central (coal, natural gas or nuclear) power plants. The main problem is the rapid implementation of wind power, which means that the reinforcement of the grid has not been planned and that the procedure for determination of the transmission

capacity is different due to the stochastic nature of wind energy. When wind power is going to be exploited at high levels it means that large amounts of power have to be transmitted over long distances and that in the regions with a high wind power production there will be very little if any conventional production. This also has an impact on how inter-area connections are handled with respect to controllability, security of supply of the areas, etc. [23]. In most areas of the world with deregulated power markets, existing transmission planning practices do not look ahead towards expanding the transmission grid to serve high wind resources [24]. Individual wind projects cannot afford to pay for major expansion on their own, so the wind resources may remain stranded and underdeveloped. FERC introduced economic transmission planning, which is planning to build transmission to reduce congestion and not just for reliability purposes [24]. Such economic transmission planning represents an opportunity to access remote wind resources and for this reason the wind industry is keenly interested in it. Yet economic transmission planning faces at least three challenges: 1. Economic transmission planning is viewed separately from transmission planning for reliability, yet the two may be intertwined, i.e., certain reliability fixes may be necessary in order for an economic transmission addition to move forward. 2. These economic transmission studies may not result in any action; market participants are asked, rather than required, as is the case with reliability studies, to contribute financially to support any identified transmission upgrades or expansion. 3. Related to the previous point, economic transmission planning studies are time, labor, and cost intensive, and efforts to keep them going may fail without some sign of success. VIII. WIND FORECASTING Wind forecasting is increasingly showing value for the improved scheduling of wind energy and such forecasts can have substantial value even if they are not perfectly accurate [25,26]. Using physics-based forecasting models, real-time wind and energy data from the wind plants, and computational learning systems such as artificial neural networks [27] or support vector machines, it is possible to provide forecasts of wind energy delivery that are significantly better than simplistic forecasts based on climatology (historic values) or persistence (assuming that what is currently happening will continue without change). The accuracy of forecast results depends on the specific challenges of the wind plant location, the surrounding terrain, and the local climatology. Because wind plants are internationally located at sites that tend to amplify wind effects when compared with the surrounding space, it is often worthwhile to run customized fine-scale forecasting models and local wind flow models to simulate the

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

301

enhanced local effects. These results may be combined with other regional forecasts, using computational learning systems to detect complex relationships and optimize the wind energy forecast. Typically, the operator will want to receive forecasts that run hour-by-hour for the next few days to reduce the unit commitment costs. While results are site-specific, a reasonable expectation at most locations is that the accuracy of the next day hour-by-hour power forecasts using current state-of-the-art methods will have a mean absolute error (MAE) of perhaps 1015% of the rated (nameplate) capacity of the wind park. Notably, the accuracy of energy forecasts (for example, total wind energy for tomorrow) can be significantly better than the accuracy of power forecasts. The exact timing of the passage of weather fronts can be difficult to predict, and because the winds can change with the passage of such fronts, errors in timing can result in large power errors. These errors tend to balance out over longer periods of time (several hours or a full day, for example) so the accuracy of energy forecasting can be quite good. This may be useful on systems with substantial gas or hydro generation that have flexibility to reschedule energy delivery over the day. In addition, forecasting errors are significantly reduced when aggregated on a system-wide basis. Due to the smoothing effects of geographic dispersion, system-wide forecasting errors for multiple dispersed wind plants may be reduced by perhaps 3050% when compared with the errors of individual wind plants IX. WIND POWER CURTAILMENT In most European countries, renewable energy sources have priority access to the grid. In economic terms, wind power should be on top in merit order due to its marginal price of almost zero. The constraining of wind generation at certain times could reduce the overall system integration costs [28]. Curtailment of wind power production could arise because of network limitations or system aspects. Examples of wind power curtailment because of network limitations are: 1. Local limitations could give rise to a need to constrain the output of a group of wind generators. The connection of generators is normally planned so that constraints are not required under the conditions covered by the planning criteria. However, conditions not covered by the criteria may arise from time to time. 2. The time frame from application to connection of a wind farm can be down to one year, whereas the time frame from planning to the commissioning of new transmission lines might be a decade. Examples of wind power curtailment because of system aspects are: At times of low system demand, a minimum level of conventional generation plant must be kept connected to ensure system stability and control, and to ensure that the

demand can be met on a day-ahead basis. This is especially a problem in systems dominated by nuclear power due to the high minimum load capability. Due to a risk of a reduction in wind generation output at the same time as rising demand, wind output may have to be constrained to ensure that the load gradient that must be met by the remaining power plants is within their dynamic capacities. In case the non dispatch able generation exceeds demand plus interconnection capacity, the generation surplus must be removed. 3. Modern wind turbines are capable of going from partial load to full power within approximately 10 s, given the necessary wind conditions. This makes wind power a valuable asset for fast regulation. The risk of curtailment is seen as a barrier to securing financing for future investment in wind generation, so it is important that the level of curtailment is estimated and that the rules and commercial arrangements are in place. X. CONCLUSIONS The main technical challenges that are associated with the integration of wind power into power systems were presented in this paper. These challenges include effects of wind power on the power system, the power system operating cost, power quality, power imbalances, power system dynamics, and future of wind energy in India. Work conducted to date shows that wind powers impacts on system operating costs are small at low wind penetrations (about 5% or less). In most cases, these incremental costs would detract from the value of wind energy on current wholesale markets by 10% or less. At higher wind penetrations, the impact will be higher, although current results suggest the impact remains moderate with penetrations approaching 20%. What remains is to integrate these techniques into the day-to-day operation of power systems. On any given power system, costs arising from wind variability are a strong function of the characteristics of the system, such as generation mix and fuel costs, and will increase with increasing wind penetration, assuming the non-wind characteristics of the system remain unchanged REFERENCES
[1] Ancona DF, Goldman PR, Thresher RW. Wind program technological developments in the United States.Renew Energy 1997; 10:2538. [2] Bird L, Bolinger M, Gagliano T, Wiser R, Brown M, Parsons B. Policies and market factors driving wind power development in the United States. Energy Pol 2005;33:1397407. [3] Jager-Waldau A, Ossenbrink H. Progress of electricity from biomass, wind and photovoltaics in theEuropean Union. Renew Sustain Energy Rev 2004;8:15782. [4] Georgilakis PS. State-of-the-art of decision support systems for the choice of renewable energy sources for energy supply in isolated regions. Int Distrib Energy Resour 2006;2:12950. [5] Stampolidis VL, Katsigiannis YA, Georgilakis PS. A methodology for the economic evaluation of photovoltaic systems. Oper Res An Int J 2006;6:3754. [6] Katsigiannis YA, Georgilakis PS. Reliability and economic evaluation of small autonomous power systems containing only renewable energy sources. In: Proceedings of the international

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

302

[7]

[8]

[9] [10]

[11]

[12]

[13]

[14] [15] [16] [17]

[18]

conference on electrical machines, Chania, Greece, September 2006. Lambert T, Gilman P, Lilienthal P. Micropower system modeling with HOMER. In: Farret FA, Simoes MG,editors. Integration of alternative sources of energy. New York: Wiley; 2006 pp. 379 418. RETScreen International Renewable Energy Decision Support Centre.wind energy project analysis.Minister of Natural Resources Canada, 20012002 [Chapter 2]. Available at /http://www.retscreen.netS,accessed October 26, 2005. Ackermann T. Wind power in power systems. Chichester: Wiley; 2005. Smith JC, DeMeo EA, Parsons B, Milligan M. Wind power impacts on electric power system operating costs: summary and perspective on work to date. Presented at the American wind energy association global wind power conference, 2004, Illinois, Chicago. Brooks D, Lo E, Zavadil R, Santoso S, Smith J. Characterizing the impact of significant wind generation facilities on bulk power system operations planning. UWIG; 2003. Dragoon K, Milligan M, Assessing wind integration costs with dispatch models: a case study. Presented at WINDPower2003, 2003, TX, Austin. Hirst E. Integrating wind energy with the BPA power system: preliminary study. Prepared for BPA PowerBusiness Line; September, 2002. Hirst E. Integrating wind output with bulk power operations and wholesale electricity markets. Wind Energy 2002;5:1936. Electrotek Concepts Inc. Systems operations impacts of wind generation integration study, 2003. Prepared for We Energies. GRE Wind Integration study. Presented at UWIG technical workshop. WA, Seattle, 2003. Kirby B, Milligan M, Hawkins D, Makarov Y, Jackson K, Shui H. California renewable portfolio standard renewable generation integration cost analysis. Phase 1: one year analysis of existing resourcesResults and recommendations. December 2003. Kennedy BW. Integrating wind power: transmission and operational impacts. Refocus 2004:367.

[19]

[20] [21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

Kanellos FD, Hatziargyriou ND. The effect of variable-speed wind turbines on the operation of weak distribution networks. IEEE Trans Energy Conv 2002;17:5438. Mosey Ed. Wind farms get a boost from BPA. Portland, Oregon: BPA press release; 2002. Nielsen LH, Morthorst PE, Skytte K, Jensen PH, Jrgensen, Eriksen PB, et al. Wind power in a liberalizedNorth European electricity exchange. In: Proceedings of the European wind energy conference and exhibition, Nice, France; 1999. Akhamatov V, Knudsen H, Nielsen AH, Pedersen JK, Poulsen NJ. Modelling and transient stability of largewind farms. Int J Electr Power Energy Systems 2003;25:12344. Bindner H, Lundsager P. Integration of wind power in the power system. Proceedings of the 28th annual conference of the IEEE industrial electronics society, November 2002. Pwiko R, Osborn D, Gramlich R, Jordan G, Hawkins D, Porter K. Wind energy delivery issues: transmission planning and competitive electricity market operation. IEEE Power Energy 2005;3:4756. Ahlstrom M, Jones L, Zavadil R, Grant W. The future of wind forecasting and utility operations: planning for improved system operations. IEEE Power Energy 2005;3:5764. Tsikalakis A, Katsigiannis Y, Peorgilakis P, Hatziargyriou N Determining and exploiting the distributionfunction of wind power forecasting error for the economic operation of autonomous power systems.Proceedings of the IEEE society general meeting, Montreal, Quebec, Canada, June 1822, 2006. Katsigiannis YA, Tsikalakis AG, Georgilakis PS, Hatziargyriou ND. Improved wind power forecasting using a combined neurofuzzy and artificial neural network. Lecture Notes Artif Intell 2006;3955:10515. Eriksen PB, Ackermann T, Abildgaa . System operation with high windpenetration. IEEE Power Energy 2005;3:6574.

Effects Short-term effects Production efficiency thermal and hydro Transmission distribution efficiency Regulating reserves Discarded wind) energy Long term effects

Effect or Area Time scale impacted element

Wind power contribution

Voltage management Seconds/ minutes Wind farms can provide (dynamic) voltage support Local/regional (design dependent) of System and Sysyem or local 1-24 hours 1-24 hours Impact depends on how the system is operated and on the use of short-term forecasting Depending on penetration level, wind farms may create additional investment costs or benefits. Spatially distributed wind energy can reduce network losses.

System System System (generation transmission System)

Several minutes Wind power can partially contribute to primary and to hours secondary control Hours reliability Years and adequacy Wind power may exceed the amount the system can absorb at very high penetrations Wind power can contribute (capacity credit) to power system adequacy

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

303

Trends and its Predictive Control in Power Electronics & Drives


1

Mamatha Sandhu, 2Shimi Sudhaletha


1, 2

Chitkara University, Punjab Campus

Abstract: Today, electric drives, power electronic systems and their controls are getting more and more complex, and their use is widely increasing in all sectors such as power systems, traction, hybrid vehicles, industrial and home electronics, automotive, naval and aerospace systems, etc. Advances in Microprocessors, Microcomputers, Microcontrollers such as DSP, FPGA, dSPACE etc. and power semiconductor devices have made tremendous impact on electric motor drives. The purpose of this paper is to acquaint the reader, the trends in Induction Motor Drives in todays real world and also the predictive control in power electronics and drives. Keywords: drives, predictive control, power electronics

I. INTRODUCTION Power electronics is about control of power electronically. Range of power can extend from some fraction of watt to several mega watts (MW). 3-Phase motor drives dominate the industry in most applications from low to high power. The use of power converters has become very popular in the recent decades for a wide range of applications, including drives, energy conversion, traction, and distributed generation. The control of power converters has been extensively studied, and new control schemes are presented every year. Several control schemes have been proposed for the control of power converters and drives. This paper presents the trends in the induction motor drives for the present industries and also a survey of the most important types of predictive control applied in power electronics and drives. A classification of them is proposed, and each type of predictive control is explained in the following sections, including some application examples. Hysteresis-based predictive control, trajectorybased predictive control, deadbeat control, model predictive control (MPC), MPC with finite control set, and finally, some conclusions are presented. New Trends of Induction Motor Drives: 3- Phase induction motors are used widely to serve general purpose applications, both adjustable speed and servo drives. 3Phase synchronous motor is found in special applications, mostly as servo drives. Some very large power adjustable speed drives also prefer synchronous motors because of the possibility of using low cost load-commutatedinverters (LCI) built from thyristors. The induction motor is without doubt the most used electrical motor and a great energy consumer. Three-phase induction motors consume 60% of industrial electricity and it takes considerable efforts to improve their efficiency. The vast majority of induction motor drives are

used for heating, ventilation and air conditioning (HVAC). These applications require only low dynamic performance and in most cases only voltage source inverter is inserted between grid and induction motor as cheapest solution. The classical way to control these devices is constant V/f ratio and in a conventional setting, the field excitation is kept constant at rated value throughout its load range. It is known that efficiency improvement of IMD can be implemented via motor flux level and this method has been proven to be particularly effective at light loads and in a steady state of drive. Recent trends in variable speed drive technology for induction motor drives include the development of high torque control at low speed, highly responsive sensor less vector control, low rotational fluctuation and minimum time acceleration/deceleration technology, In addition adaptive drive technology that includes the mechanical system and high efficiency sensor less synchronous motor drives are also being developed. Thus, the three technologies that are concerned are low rotational fluctuation control, minimum time acceleration/deceleration technology for induction motor drive, and sensor less for synchronous motor drive. II. PREDICTIVE CONTROL METHODS Predictive control is a very wide class of controllers that have found rather recent application in power converters. The classification proposed in this paper for different predictive control methods is shown in Fig 1. The main characteristic of predictive control is the use of the model of the system for the prediction of the future behavior of the controlled variables. This information is used by the controller in order to obtain the optimal actuation, according to a predefined optimization criterion. The optimization criterion in the hysteresis-based predictive control is to keep the controlled variable within the boundaries of a hysteresis area, while in the trajectory based; the variables are forced to follow a predefined trajectory. In deadbeat control, the optimal actuation is the one that makes the error equal to zero in the next sampling instant. A more flexible criterion is used in MPC, expressed as a cost function to be minimized. The difference between these groups of controllers is that deadbeat control and MPC with continuous control set need a modulator, in order to generate the required voltage. This will result in having a fixed switching frequency. The other controllers directly generate the switching signals for the converter, do not need a modulator, and present a variable switching frequency.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

304

One advantage of predictive control is that concepts are very simple and intuitive. Depending on the type of predictive control, implementation can also be simple, as with deadbeat control and finite control set MPC (FSMPC) (particularly for a two-level converter with horizon N = 1). However, in general, some implementations of MPC can be more complex. Variations of the basic deadbeat control, in order to make it more robust, can also become very complex and difficult to understand. Using predictive control, it is possible to avoid the cascaded structure, which is typically used in a linear control scheme, obtaining very fast transient responses. An example of this is the speed control using trajectory-based predictive control.
Predictive Control

a hysteresis area or space. The simplest form of this principle is the so-called bangbang controller. Although bangbang controllers usually are not considered as predictive controllers in literature, they clearly show the characteristics of a typical predictive controller. An improved form of a bangbang controller is the predictive current controller proposed by Holtz and Stadtfeld [1]. The block diagram of the hysteresis-based predictive control is shown in Fig 2. Using predictive current control, the switching instants are determined by suitable error boundaries.

Deadbeat Control
--Needs Modulator --- Fixed Switching frequency ---Low computatio ns

Hysteresis Based

Trajectory based
--No Modulator ---Variable Switching frequency

--No Modulator
--- Variable Switching frequency ---Simple concepts

---No cascaded structure

Fig. 2: Hysteresis- based predictive control.

---no constraints include


Model

Predictive Control (MPC) Control With finite Continuous Control Set


--Needs Modulator -Fixed Switching --No Modulator. --Variable Switching frequency. --Online optimization.

frequency ---no constraints include

--Low complexity

Fig. 1: Block Diagram of Predictive control methods

Nonlinearities of the system can be included in the model, avoiding the need of linearzing the model for a given operating point and improving the operation of the system for all conditions. It is also possible to include restrictions to some variables when designing the controller. These advantages can be very easily implemented in some control schemes as MPC, but it is very difficult in schemes as deadbeat control. A more detailed description of each type of predictive control is shown in the next sections. III. HYSTERESIS-BASED PREDICTIVE CONTROL Hysteresis-based predictive control strategies try to keep the controlled system variables between the boundaries of

IV. TRAJECTORY-BASED PREDICTIVE CONTROL The principle of trajectory-based predictive control strategies is to force the systems variables onto precalculated trajectories. Control algorithms according to this strategy are direct self control or direct mean torque control. Some additional methods like sliding mode control [2] or direct torque control [3] are a combination of hysteresis- and trajectory-based strategies, whereas direct speed control (DSPC) can be identified as a trajectory-based control system, although it also has a few hysteresis-based aspects. DSPC will be further explained as an example of trajectory-based predictive controllers. Unlike cascade controllers, predictive control algorithms offer the possibility to directly control the desired system values. Most predictive control methods published so far only deal with stator currents, torque, or flux (linear) directly; the drive speed is controlled by a superimposed control loop. DSPC, shown in Fig 3, in contrast, has no control loop of this type; the switching events in the inverter are calculated in a way where speed is directly controlled in a time-optimal manner. The switching states of the inverter are classified as torque increases, slowly torque decreasing, or rapidly torque decreases. For small time intervals, the inertia of the system and the derivatives of machine and load torques are assumed as constant values.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

305

Fig. 3: DSPC

V. DEADBEAT-BASED PREDICTIVE CONTROL A well-known type of predictive controller is the deadbeat controller. This approach uses the model of the system to calculate, once every sampling period, the required reference voltage in order to reach the reference value in the next sampling instant. Then, this voltage is applied using a modulator. It has been applied for current control in three-phase inverters [4][5], rectifiers [6], [7], active filters [8], [9], power factor correctors [10], power factor pre-regulators [11], uninterruptible power supplies [12], dcdc converters [13], and torque control of induction machines [14]. While this method has been used when a fast dynamic response is required, being deadbeatbased, it is often fragile. Indeed, errors in the parameter values of the model, unmodeled delays and other errors in the model often deteriorate system performance and may even give rise to instability. Another disadvantage of these deadbeat control schemes is that non-linearity and constraints of the system variables are difficult to incorporate.

A. Deadbeat Current Control A typical deadbeat current control scheme is shown in Fig 4. It can be noted that, compared to a classic current control scheme, the PI controller has been replaced by the deadbeat controller. The reference voltage is applied using a modulator. The load model for a generic RLE load is described by the following space vector equation v = Ri + L di + e -------(1) dt where v is the voltage space vector, i is the current space vector, and e is the EMF voltage space vector. The following discrete-time equation can be obtained from (1) for a sampling time Ts: 1 i(k + 1) x i(k) = v(k) e(k) --------- (2) where = eTsR/L and " = 1/R(1 eTsR/L). Based on the discrete-time model (2), the reference voltage vector is obtained as V*(k) = 1 [i*(k + 1) xi(k)] + e(k). -------- (3) Reference voltage v* is applied in the converter using a modulator. The basic operating principle of deadbeat current control is shown in Fig 5. Here, the load current i at time k is different to the reference current i*. This error is used for calculation of the reference voltage v*, which is applied to the load at time k. ideally, at time k + 1, the load current will be equal to the reference current. When implemented in a real system, several problems may appear and deteriorate the performance of a deadbeat controller. One of them is the delay introduced by calculation time and modulation. This problem has been solved in [15], by considering this delay in the model. Another important issue is the sensitivity to plant uncertainties and errors in the model parameter values. This problem has been studied, and several solutions have been proposed, including the use of an adaptive selftuning scheme [16], a predictive internal model [17], and neural networks [18]. VI. MPC (MODEL PREDICTIVE CONTROL) MPC, also referred to as receding horizon control, is the only one among the so-called advanced control techniques (usually understood as techniques more advanced than a standard PID control) which has been extremely successful in practical applications in recent decades, exerting a great influence on research and development directions of industrial control systems. Applications and theoretical results abound; see, e.g., the books [19][21] and survey papers [22][23]. An attractive feature of MPC is that it can handle general constrained nonlinear systems with multiple inputs and outputs in a unified and clear manner. In this section, we will focus on MPC formulations where a continuous control set is considered and controller outputs are first passed through modulators, which then provide the switch positions. A survey on FSMPC formulations which control the states of the

Fig. 4: Deadbeat Current Control

Fig. 5: Illustration of Deadbeat Current Control Operations

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

306

converter switches directly, i.e., without any intermediate modulators, are discussed. Explicit MPC for Drive Control The advantages of replacing the PI current controller in the field-oriented drive control scheme in Fig 6 by Explicit MPC. The dynamics of an induction machine are governed by nonlinear differential equations. To obtain a model suitable for Explicit MPC, in [24], the timecontinuous machine model was discretized, and the nonlinear cross-coupling between the stator current components isd and isq was neglected.

minimizes the given cost function. In addition to this, if the horizon length is set to N = 1, the calculation of the optimal actuation is very simple and easy to implement experimentally. A. System Model When modeling a converter, the basic element is the power switch, which can be an insulated-gate bipolar transistor, a thyristor, a gate turn-off thyristor, or others. The simplest model of these power switches considers an ideal switch with only two states: on and off. Therefore, the total number of switching states of a power converter is equal to the number of the different combinations of the two switching states of each switch. However, some combinations are not possible, for example, those combinations that short-circuit the dc link. Let us consider the example of an H-bridge single-phase inverter. It has four switches; therefore, the total number of combinations is 24 = 16. However, two switches in the same leg of the inverter cannot be on at the same time; therefore, it is usual to drive them as complementary switches. This way, the number of possible states is reduced, and each leg has only two states. Therefore, the number of possible states is 22 = 4. As a general rule, the number of possible switching states can be calculated as xy, where x is the number of possible states of each leg of the converter and y is the number of phases (or legs) of the converter. This way, a three-phase two-level converter has 23 = 8 possible switching states, a three-phase three-level converter has 33 = 27 switching states, and a five-phase two level converter has 25 = 32 switching states. Some examples of different converter topologies and their corresponding number of possible switching states are shown in Table I. For some other converter topologies, the way of calculating the possible switching states may be different.
TABLE I : POSSIBLE SWITCHING STATES FOR DIFFERENT CONVERTER
TOPOLOGIES

Fig. 6: Typical Field Oriented Control of an Induction machine [24].

Comments The results of [24] are certainly promising. However, we feel that MPC has significantly more to offer than replacing individual modules within a cascaded control structure. Indeed, the main advantage of MPC, with respect to other control approaches, lies in the possibility to control nonlinear and constrained systems. In addition, various objectives, such as reference tracking and disturbance compensation, can be incorporated. Thus, it is worth studying the feasibility of developing MPC architectures which govern the entire drive control architecture without using any additional PI loops or PWM modules. For that purpose, nonlinearities need to be taken into account, and the finite set nature of switching states should be respected. In the following section, we will revise some MPC formulations where the switches are controlled directly, without using PWM modules. VII. FS-MPC Considering the discrete nature of power converters, it is possible to simplify the optimization problem of MPC by avoiding the use of modulators. Taking into account the finite set of possible switching states of the power converter, which depends on the possible combinations of the on/off switching states of the power switches, the optimization problem is reduced to the evaluation of all possible states and the selection of the one which

Converter 3-phase inverter 3-phase NPC inverter 5 phase inverter Matrix converter Direct converter Flying capacitor

Switching States 8 27 32 27 9 8

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

307

2) Torque and Flux Control For the same three-phase inverter, considering an induction motor as a load, the torque and flux of the machine can be directly controlled, as shown in [30]. 3) Power Control For a three-phase PWM rectifier, the same possible switching states and voltage vectors shown in Table II are valid. Here, the inductive filter model is considered for the prediction of the input current and power of the converter. As well as a predictive current control can be used in the rectifier, another approach for the control of this converter considers the direct control of the active and reactive powers .where the reactive power reference is zero and the active power reference is obtained from the dc link voltage control loop. 4) Control of NPC Converter In a three-level three-phase inverter, the number of switching states is 27, generating 19 different voltage vectors. Here, it is possible to take advantage of the redundancy of switching states by considering additional terms in the cost function. As proposed in [31], it is possible to control the load currents while balancing the capacitor voltages and reducing the average switching frequency. 5) Control of Direct Converter The direct converter has six bidirectional switches which allow nine possible switching states. The control strategies presented in [27] consider the control of the capacitor voltages of the LC filter at the input of the converter in a cascaded control structure, in order to obtain unity power factor and controlled amplitude of the output voltage. 6) Control of Flying Capacitor Converter: The application of FS-MPC to a single-phase four-level flying capacitor converter is presented in [29]. This converter has three pairs of switches allowing eight possible switching states that generate four different voltage levels. The output current and capacitor voltages are controlled considering also the reduction of the switching frequency and control of the spectrum of the output current. VIII. CONCLUSION This paper has reviewed the latest trends used in Induction motor drives applications in the real world. It has explained the different types of predictive control methods used in power electronics and drives. The predictive control methods are divided into following groups: deadbeat control, hysteresis-based control, trajectory-based control, and continuous MPC and FSMPC. The basic principles and the latest developments of these methods have been systematically described, and application examples have been indicated. It presents several advantages that make it suitable for the control of power converters and drives. Predictive control has been

Fig. 7: FS- MPC TABLE II V V0=0 V1=2/3 Vdc V2=Vdc/3 +j sqrt(3)/3 Vdc V3=-Vdc/3 +j sqrt(3)/3 Vdc V4 = -2/3 Vdc V5= -Vdc/3 - j sqrt(3)/3 Vdc V6= Vdc/3 +j sqrt(3)/3 Vdc V7 =0

Sa 0 1 1 0 0 0 1 1

Sb 0 0 1 1 1 0 0 1

Sc 0 0 0 0 1 1 1 1

Possible Switching states and voltage vectors for 3 phase Inverter B. Horizon Length When a horizon length N is used, the number of possible input sequences, considering the possible switching states of the converter, can be quite large. Then, the idea of predicting the behavior of the system for all possible switching state sequences becomes difficult to apply in a real system. A simple solution is the use of N = 1, reducing the number of calculations to the number of possible switching states of the converter. The block diagram of an FS-MPC considering a prediction horizon N = 1 is shown in Fig7. Here, the state variables of the system x(k) are measured (or estimated) and used as initial condition for the predictions. Some Application Examples: 1) Current Control Considering a three-phase two-level inverter, the eight possible switching states and voltage vectors are summarized in Table II. Here, variables Sa, Sb, and Sc represent the switching states of the a, b, and c legs of the inverter. For the case of the three-phase PWM rectifier, the possible switching states and voltage vectors generated by the inverter are the same as the ones shown in Table II for an inverter. As well as the current can be controlled, it is also possible to control the active and reactive powers [25]. For more complex converter topologies, the control strategy is the same but with a different set of possible switching states. This is the case of the matrix converter, as presented in [26], the direct converter [27][28], and the flying capacitor converter [29].

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

308

applied to a very wide range of systems, and it is open for new applications and converter topologies. However, the best suited type of predictive control will depend on the application and requirements of the system. As a conclusion of survey, it is the belief that predictive control strategies will continue to play a strategic role in the development of modern high-performance power electronics and drive systems and will offer a new interesting perspective for future research in this area. REFERENCES
[1] J. Holtz and S. Stadtfeld, A predictive controller for the stator current vector of AC machines fed from a switched voltage source, in Proc.IPEC, Tokyo, Japan, 1983, pp. 16651675. S. V. Emeljanov, Automatic Control Systems With Variable Structure (Automatische Regelsysteme mit Veranderlicher Struktur). Munich, Germany: R. Oldenbourg-Verlag, 1969. Takahashi and T. Noguchi, A new quick response and high efficiency control strategy of an induction motor, in Conf. Rec. IEEE IAS Annu.Meeting, 1985, pp. 16651675. O. Kukrer, Discrete-time current control of voltage-fed threephase PWM inverters, IEEE Trans. Ind. Electron., vol. 11, no. 2, pp. 260269, Mar. 1996. Q. Zeng and L. Chang, An advanced SVPWM-based predictive current controller for three-phase inverters in distributed generation systems, IEEE Trans. Ind. Electron., vol. 55, no. 3, pp. 12351246, Mar. 2008. L. Malesani, P. Mattavelli, and S. Buso, Robust dead-beat current control forPWMrectifier and active filters, IEEE Trans. Ind. Appl., vol. 35, no. 3, pp. 613620, May/Jun. 1999. Y. Nishida, O. Miyashita, T. Haneyoshi, H. Tomita, and A. Maeda, A predictive instantaneous-current PWM controlled rectifier with AC-side harmonic current reduction, IEEE Trans. Ind. Electron., vol. 44, no. 3, pp. 337343, Jun. 1997. S.-G. Jeong and M.-H. Woo, DSP-based active power filter with predictive current control, IEEE Trans. Ind. Electron., vol. 44, no. 3, pp. 329336, Jun. 1997. J. Mossoba and P. W. Lehn, A controller architecture for high bandwidth active power filters, IEEE Trans. Power Electron., vol. 18, no. 1, pt. 2, pp. 317325, Jan. 2003. W. Zhang, G. Feng, and Y.-F. Liu, Analysis and implementation of a new PFC digital control method, in Proc. Conf. Rec. PESC. Acapulco, Mexico, 2003. CD-ROM. [27] P. Mattavelli, G. Spiazzi, and P. Tenti, Predictive digital control of power factor preregulators, in Proc. Conf. Rec. PESC, Acapulco, Mexico, 2003. CD-ROM. P. Mattavelli, G. Spiazzi, and P. Tenti, Predictive digital control of power factor reregulators with input voltage estimation using disturbance observers, IEEE Trans. Power Electron., vol. 20, no. 1, pp. 140147, Jan. 2005. P. Mattavelli, An improved deadbeat control for UPS using disturbance observers, IEEE Trans. Ind. Electron., vol. 52, no. 1, pp. 206212, Feb. 2005. S. Saggini,W. Stefanutti, E. Tedeschi, and P. Mattavelli, Digital deadbeat control tuning for dcdc converters using error correlation, IEEE Trans. Power Electron., vol. 22, no. 4, pp. 15661570, Jul. 2007. P. Correa, M. Pacas, and J. Rodriguez, Predictive torque control for inverter-fed induction machines, IEEE Trans. Ind. Electron., vol. 54, no. 2, pp. 10731079, Apr. 2007. Abu-Rub, J. Guzinski, Z. Krzeminski, and H. A. Toliyat, Predictive current control of voltage source inverters, IEEE Trans. Ind. Electron., vol. 51, no. 3, pp. 585593, Jun. 2004. Y. A. R. I. Mohamed and E. F. El-Saadany, An improved deadbeat current control scheme with a novel adaptive self-tuning load model for a threephase PWM voltage-source inverter, IEEE Trans. Ind. Electron., vol. 54, no. 2, pp. 747759, Apr. 2007. Y. A. R. I. Mohamed and E. F. El-Saadany, Robust high bandwidth discrete-time predictive current control with predictive

internal modelA unified approach for voltage-source PWM converters, IEEE Trans.Power Electron., vol. 23, no. 1, pp. 126 136, Jan. 2008. [18] Y. A. R. I. Mohamed and E. F. El-Saadany, Adaptive discretetime gridvoltage sensorless interfacing scheme for grid-connected DG-inverters based on neural-network identification and deadbeat current regulation, IEEE Trans. Power Electron., vol. 23, no. 1, pp. 308321, Jan. 2008. [19] F. Camacho and C. Bordons, Model Predictive Control. New York: Springer-Verlag, 1999. [20] J. M. Maciejowski, Predictive Control With Constraints. Englewood Cliffs, NJ: Prentice-Hall, 2002. [21] C. Goodwin, M. M. Sern, and J. A. De Don, Constrained Control & EstimationAn Optimization Perspective. London, U.K.: Springer- Verlag, 2005. [22] S. J. Qin and T. A. Badgwell, A survey of industrial model predictive control technology, Control Eng. Pract., vol. 11, no. 7, pp. 733764, Jul. 2003. [23] Q. Mayne, J. B. Rawlings, C. V. Rao, and P. O. M. Scokaert, Constrained model predictive control: Optimality and stability, Automatica, vol. 36, no. 6, pp. 789814, Jun. 2000. [24] Linder and R. Kennel, Model predictive control for electrical drives, in Proc. IEEE PESC, Recife, Brazil, 2005, pp. 1793 1799. [25] P. Antoniewicz and M. P. Kazmierkowski, Predictive direct power control of three-phase boost rectifier, Bull. Pol. Acad. Sci., vol. 54, no. 3, pp. 287292, 2006. [26] S. Muller, U. Ammann, and S. Rees, New time-discrete modulation scheme for matrix converters, IEEE Trans. Ind. Electron., vol. 52, no. 6, pp. 16071615, Dec. 2005. [27] M. Catucci, J. Clare, and P. Wheeler, Predictive control strategy for ZCS single stage resonant converter, in Proc. 32nd Annu. Conf. IEEE IECON, Nov. 2006, pp. 29052910. [28] Q. S. Dang, P. Wheeler, and J. Clare, A control analysis and implementation of high voltage, high frequency direct power converter, in Proc. 32nd Annu. Conf. IEEE IECON, Nov. 2006, pp. 20962102. [29] I. Silva, B. P. McGrath, D. E. Quevedo, and G. C. Goodwin, Predictive control of a flying capacitor converter, in Proc. Amer. Control Conf., New York, Jul. 2007, pp. 37633768. [30] Rodrguez, J. Pontt, C. Silva, P. Corts, S. Rees, and U. Ammann, Predictive direct torque control of an induction machine, in Proc. 11th Int. EPE-PEMC, Riga, Latvia, Sep. 24, 2004. [31] R. Vargas, P. Cortes, U. Ammann, J. Rodriguez, and J. Pontt, Predictivecontrol of a three-phase neutral-point-clamped inverter, IEEE Trans. Ind. Electron., vol. 54, no. 5, pp. 2697 2705, Oct. 2007. [32] http://www.drivesys.com/asdis.html [33] http://www.engineeringexchange.com/profiles/blogs/the-types-ofac-drives [34] http://tungvp.wordpress.com/2010/03/27/motor-control-tutorials/ [35] IEEE Journal, Electrical India, Jan, April 2011.

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

309

Energy Efficiency Improvement through Innovation and Technology


1
1

Pushpanjali Chauhan, 2Shikha Gupta , 3Medha Walia, 4Mandip Singh


1, 2

Chitkara University, Punjab, India , 3 ,4SBBSIET , Punjab, India pushpanjali.chauhan@chitkara.edu.in, 2shikha.gupta@chitkara.edu.in, 3 medhawalia@gmail.com, 4mandipmaini408@gmail.com introduction of modern technologies and improved management practices. Taking advantage of all the opportunities that are available for each sector has the cumulative potential for cost-effective efficiency improvements of up to 50 percent, which is the projected increase in demand in the next quarter century. Today 33 percent of all energy consumption is in electrical energy, and is expected to grow to 67 percent by 2040 [Fig2.4]. Power electronics is enabling technology to efficiently distribute and use electric power, and to generate electrical energy from renewable sources (wind, solar ,etc.). Power electronic converters can be found wherever there is a need to modify the electrical energy form (i.e. modify its voltage, current or frequency). Many market segments, such as domestic and office appliances, HVAC, consumer digital, communications, factory automation and drives, traction, automotive and renewable energy, can potentially benefit from the application of power electronics technology. Advanced power electronics could, for example ,realize savings of more than 50 percent in energy losses in converting from mains or battery voltages to those used in electronic equipment. II ENERGY USAGE AND FUTURE DEMAND Energy demand is growing in parallel with population growth and economic expansion. According to the International energy Outlook 2007*, projections for global energy growth include: Market energy consumption is projected to increase by 57 percent in the next 25 years. Total energy demand in the developing (non-OECD) countries increases by 95 percent, compared with an increase of 24 percent in the developed (OECD) countries [Fig2.1].

Abstract The aim of this paper is the presentation of a participatory and interactive approach for foresight of saving energy through carbon capture ,renewable energy sources also new power electronics technologies are part of the conservation of energy. They will yield improved efficiency by significantly reducing wasted energy throughout the entire energy ecosystem, from generation through transmission to consumption (EndUse). Through this approach, a number of innovative energy technologies are assessed in the Greek energy sector, taking into account economical, environmental, social and technological aspects, and engaging experts and stakeholders from different companies, associations and institutions, in a methodological and consistent way. The results of its application point out the significance of collective interactive approaches, engaging all interested stakeholders, for supporting the discussion of the technological priorities and for guiding the future funding and Research and Development (R&D) efforts. Keywords Carbon Technologies Capture, New Power Electronics

I. INTRODUCTION The ability to maintain existing levels of energy services for the next 25 to 50 years is seriously challenged. It is put in jeopardy by the global forces of increasing economic activity in developing countries, geopolitical tensions in the key energy producing countries, and the threat to the environment caused by the rise in emissions and global temperatures. In the next quarter century, global energy demand and corresponding CO2 emissions are projected to increase by more than 50 percent. This increase in demand is expected to be met through traditional energy sources, primarily fossil fuels (including natural gas). Liquid fuels will remain the most important fuel for transportation; most of the increase in oil supply will be met by a small number of OPEC producers, and the consuming countries will be vulnerable to supply disruptions and price shocks. Consequently, sustainability and security policies require a concerted effort from governments and industry across the globe. The solution to short-term growth in energy consumption can be delivered, not by creating more, but by wasting less. Today the global efficiency rate for converting primary energy into useful (or transmittable) energy at power plants is about 35 percent. That efficiency rating could be increased to 45 percent by replacing existing conversion and transmission technologies with stateof-the-art semiconductor solutions (Section 1.3). In transmission and distribution, technical losses are up to 10 percent, another area which can be improved by the

Fig. 2.1: World Market Energy Consuption, 2004-2030

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

310

During the same period, energy-related carbon dioxide emissions are projected to increase by 59 percent, from 27 billion metric tons to 43 billion metric tons [Fig2.2].

services in residential, industrial, transportation, public, and commercial sectors. During the next 20 years, the amount of primary energy required for a given level of energy services could be cost-effectively reduced by 25 to 35 percent in industrialized countries. Reductions of more than 40 percent are cost-effectively achievable in transitional economies within the next two decades. And in most developing countrieswhich tend to have high economic growth and old capital and vehicle stocksthe costeffective improvement potential ranges from 30 to more than 45 percent, relative to energy efficiencies achieved with existing capital stock. Although not exactly to scale, the green line in Fig2.3 shows the relative efficiencies achievable along the energy value-chain.

Fig. 2.2: World Energy-Related Carbon Dioxide Emissions by Region, 20032030

Emissions from developing countries are projected to exceed those from developed countries by 57 percent, the twin forces of urbanization and industrialization will accelerate this trend. Much of the growth in energy demand is coming from developing economies in Asia, including China and India. Energy demand in this region is projected to more than double over the next 25 years and account for more than 65 percent of the increase in energy use for the developing regions as a whole. A. ENERGY EFFICIENCY Energy efficiency refers to the ratio between energy outputs (services such as electricity, heat and mobility) and inputs (primary energy). Energy efficiency has increased considerably since the 1970s. This improvement resulted from a response to energy price increases and supply, as well as independent technology improvements and government policies. By IEA estimates, without the energy savings achieved since 1973, energy demand in 1998 would have been 50 percent higher. However, the rate of energy savings has slowed significantly after 1990, when rapid demand growth and lower energy prices reduced focus on energy efficiency. In the last two years, energy efficiency has returned to prominence due to the dramatic increase. in energy prices, threats to fuel supplies and a general consensus on the harmful impact of emissions on physical health and global climate change.Today, the global efficiency of converting primary energy to useful energy is about one-third, according to the World Energy Council (WEC), a 94-member country organization that provides research and education on all aspects of energy consumption. In other words, two-thirds of primary energy is dissipated in the conversion process, mostly as lowtemperature heat. Further significant losses occur when the useful energy is transmitted and delivered as energy

Fig. 2.3: Role of Electrical Energy along the Energy Chain

Fig. 2.4: 2004 Final Energy Demand in the world by the most important sectors and fuel source

III. ROLE OF POWER ELECTRONICS


IN ENERGY EFFICIENCY

Power electronics is the technology associated with the efficient conversion, control and conditioning of electric power by static means from its available input form into the desired electrical output form. Power electronic converters can be found wherever there is a need to modify the electrical energy form (i.e. modify its voltage, current or frequency). Therefore, the power handling capability of converters ranges from several milliwatts (as in a mobile phone) to hundreds of megawatts (as in a HVDC transmission system). A power electronic converter is built around one (or more) device(s) operating in switching mode (either on or off). With such a structure, the energy is transferred from the input of the

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

311

converter to its output by bursts. The power conversion systems can be classified according to the type of the input and output power:AC to DC (rectification): Used every time an electronic device is connected to a main supply (computer, television, etc.). DC to AC (inversion): Used in a wide range of applications, from small-switched power supplies for a computer, to adjustable speed motor control for HEV, to large electric utility applications to transport bulk power. DC to DC (conversion): Used in most mobile devices (mobile phone, PDA, etc.) to maintain the voltage at a fixed value. AC to AC (conversion): Used to change either the voltage level or the frequency (international power adapters, light dimmer, etc.). What power electronics can do? - Photo voltaic converters, converters and generators for wind turbines and biogas turbines - Smart drives, energy recovery components and systems - Hybrid and electric traction (high voltage) - Integration of control ICs and power switches - ReconFigurable power solutions - Power device technology for on-chip power processing, fast, and high-temperature conversion - Further research will reduce the weight and volume of components and systems while increasing power density Today 33 percent of all energy consumption is in electrical energy, and is expected to grow to 67 percent by 2040 [Fig3.1]. Power electronics is the enabling technology to fficiently distribute and use electric power, as well as to generate electrical energy from renewable sources (wind,solar etc).

Over half of all electricity is consumed by electric motors. In industry, electric drives account for nearly 60 percent of industrial electricity consumption (Fraunhofer ISI, Motor Challenge, 2005). These are the motors in elevators, refrigerators, air conditioners, washing machines and factories. The vast majority of these motors do not have electronic controls. Electric motion (motion excluding transportation) accounts for 80Q British thermal units. Simple electric motors (without intelligent control electronics) account for about 70Q Btu. These simple electric motors are either fully on or fully off, which is like driving with the gas pedal pushed all the way which is like driving with the gas pedal pushed all the way to the floor then taking it off, over and over again. Not only is this not a good way to drive, it also turns out to be less efficient. By converting all such simple electric motors to variable speed, it is possible to cut power consumption by almost half [Fig3.2] or approximately 40Q Btu, which amounts to 10 percent of the total energy consumption .An example is the typical air conditioner [Fig3.3], which uses a bimetallic switch to turn the motor on when the temperature gets too hot and turn the motor off when the temperature gets too cold. This method of control typically wastes half of the energy that could be saved if a modern, variable-speed electric motor were used. For a typical 20 cubic foot unit, this cuts ahousehold electrical bill by about US$70 per year

Fig. 3.2: Electric Drives with Variable speed controls Fig. 3.1: Electrical Energy Saving Potential through power electronics

Many market segments such as domestic and office appliances, HVAC, consumer digital, communications, factory automation and drives, traction, automotive and renewable energy can potentially benefit from the application of power electronics technology [Fig3.1]. Advanced power electronics could, for example, realize savings of more than 50 percent in energy losses in converting from mains B. MOTOR CONTROL: ENERGY SAVINGS WITH VARIABLE SPEED DRIVES (VSD)

Fig. 3.3: A/C Controls with and without Smat Electronics

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

312

This savings represents approximately US$900 million of energy savings from just one motor application. Smart IC control electronics, with IGBTs transistors and some simplifications to the motion control circuitry are dramatically reducing system costs and cause an ever-faster adoption of energy-efficient electronic motor controllers. While the evidence is compelling to the expert, the consumer cost value placed on energy savings is not as compelling. To be adopted, variable-speed electric motors have to offer a significant and tangible benefit. C. VSDS IN TRANSPORTATION: HYBRID ELECTRIC VEHICLES The transportation sector, using about 27 percent of the enduse energy, relies heavily upon petroleum-based liquid fuels. This is likely to increase as the global number of vehicles is expected to grow to 1.2 billion vehicles by 2020from the current 750 million vehiclesas the developing economies add more road transport. In a typical road transport vehicle, only 25 percent of the energy contained in one unit of oil when burned is used for mobility and accessories; the rest is wasted in the form of exhaust heat (40 percent), cooling (30 percent), friction and radiation (5 percent). While in motion, energy is also wasted in braking, idling, and in stop-and-go traffic. Hybrid Electric Vehicles work on the principle of recapturing this wasted energy with an electric motor (acting as a generator) and storing the energy in the battery for later use. Along with saving energy, HEVs have other advantages over conventional vehicles. However, the benefits provided by HEVs are heavily dependent on the type of the HEV system [Fig3.4]. The larger the size of the electric motor and the energy storage system, the higher the functionality and fuel efficiency (FE) benefit. One process that is common across all the HEV systems is the conversion, storage and later usage of energy. Power Electronics is the enabling technology that makes this process possible. A High Voltage (HV) DC line runs from the battery to the sub-systems and the components of the HEV system [Fig3.5]. An inverter will convert direct current (DC) from the cars batteries to alternating current (AC) to drive the electric (traction) motor that provides power to the wheels. The inverter also converts AC to DC when it takes power from the generator to recharge the batteries.

Fig 3.5: HEV DC Bus

D. WIND ENERGY One of the best sources of renewable energy is wind energy. Wind energy has the biggest share in the renewable energy sector. Over the past 20 years, grid connected wind capacity has more than doubled and the cost of power generated from wind energy based systems has reduced to one-sixth of the corresponding value in the early 1980s. Early versions of wind turbine generators consisted of fixed-speed wind turbines with conventional induction generators. This class of machines was rugged but was limited to operation in a narrow wind-speed range. In addition, the conventional induction generator, which was connected directly to the electrical grid, required that reactive power support be provided locally to achieve the desired voltage level. Advances in power electronics have revolutionized wind turbine technology and led to the development of the doubly fed induction generator (DFIG) (Fig 3.6).

Fig. 3.6: Schematic drawing of a doubly fed induction generator wind turbine

The stator of the DFIG is connected directly to the grid, and the rotor winding is connected via slip rings to a converter. The highly efficient, variable speed DFIG is designed to get maximum energy from the wind, and puts out electricity at a constant frequency, no matter what the wind speed

Fig. 3.4: HEV Systems Functions and FE Impact

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

313

Fig. 3.7: depicts this variable speed wind energy conversion system in a block diagram.

Fig. 3.10: Z-source based variable speed wind energy conversion systemversion system

It shows a doubly-fed-induction generator where the rotor circuit is controlled by the power converter system via the slip rings, and the tator circuit is connected to the grid. This method is dvantageous as the power converter has to handle a fraction ~ 25%50% of the total power of the system. The power converter system employs a rotor side ac-dc converter, a dc link capacitor, and a dc ac inverter connected to the grid as shown in Fig 3.8. The power converter enables vector control of the field which facilitates active/reactive power control.
Electrica

A Z-source inverter based conversion system is a more novel conversion system making the role of power electronics more clear. Fig 3.10 illustrates its topology. The following lists some merits of this topology. Buck-boost ability. Inherent short circuit protection due to Z-source conFiguration. Improved EMI as dead bands are not required Most modern wind farms have DFIGs and are available in ratings that range from 1.5 megawatts (MW) to 4.5 MW. Newer generations of wind generators, which have permanent magnet synchronous generators and fully rated converters, have a range of control over both real power and reactive power for varying wind speeds. E. ENERGY EFFICIENCY IN LIGHTING APPLICATIONS Electric lighting accounts for approximately 15 percent of the worlds total electrical energy consumed. A large part of this is attributed to the roughly 11 billion incandescent lamps in the world today .Incandescent bulbs work by using electricity to heat up the wire causing it to glow and produce light. Unfortunately the heat comprises about 90 95 percent of the electricity and only 5 10 percent is emitted as useful light. A much more efficient way to create light is to use fluorescent lamps. Here, smart electronics (electronic ballasts*) convert and then regulate the flow of electricity through the glass tube. Fluorescent lamps equipped with smart IC technology use approximately one-fourth of the energy of incandescent bulbs; a 23-watt fluorescent lamp is comparable to the light from a 100-watt incandescent bulb Another technology that has been appearing in more new lighting products is the LED. By directly converting electricity to light with bright LEDs, the efficiency improves to between 28 60 percent vs. 5 10 percent for incandescent bulbs [Fig3.11]. New technologies like silicon carbide light emitting diodes (LEDs) are lighter, more flexible, and offer even more energy savings. These devices also last longer than incandescent bulbs by a factor of 25 or more..With the growing non linear loads, it is becoming important to monitor the power quality. Only few of the methods have been discussed here. Other methods are also available. Depending upon the need and utility any one method may be suitable for a particular type of situation.

l
generatorFig. 3.8: Fully variable wind energy conversion system

On the other hand, in Fig 3.9 the wind turbine can be completely decoupled from the grid, enabling a wider range of optimal operation.
13 Fig 13

rangFig. 3.9: Fully variable wind energy conversion system.e, variable wind energy conversion imited range, variable wind energy c

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

314

and reduce the standby power that provides instant-on capability.

Fig. 3.11: Comparison of Light Intensities

F. ENERGY EFFICIENCY IN INFORMATION TECHNOLO GY The broadband revolution is a major computer-based transformation, on par with the birth of the personal computer and the rise of the Internet. Broadband refers to a communication channel that has high bandwidth, or equivalently, can transfer a lot of information very quickly.There are only 700 million personal computers in the world, and there are 1.6 billion TV sets in the world [Fig3.12].

Fig 3.13: Stand by Power Efficiency using Power Electronics

IV.TECHNOLOGY AND INNOVATION As discussed the role of power electronics in the energy efficiency spectrum of applications. In this section, we discuss the key technologies involved in power electronics pplications, Infineons position in supplying these technologies, and the future trends in power electronics technology. V. POWER SEMICONDUCTOR TECHNOLOGY Production of the analog chips used in power electronics is a specialized process that is very different from the more widely known digital chips, such as computer memory and logic ICs. Analog circuits process continuous signals (these have many different voltage levels, while digital chips process signals that have discrete levels, two voltage levels that represents either on or off).For smart power and analog the key competencies required are: Analog (linear) design: This kind of design is dependent on a deep understanding of the relationship between the integrated circuit design elements and the goals of the device. Materials science: The electrical properties of the integrated circuit are defined by the physics of the materials used, which often are not the materials used in the standard or digital CMOS-integrated circuit products. Examples of power semiconductors include products such as the Coo lMOS MOSFET transistor technology. In this product family, Infineon design innovation circumvents whatengineers had previously called the silicon limit, a characteristic of the material that limited performance ofMOSFET transistors. A second example is power diodes based on silicon carbide (SiC). SiC was previously used to make integrated circuits, but was abandoned for easier-toprocess pure silicon. Advances in technology and design, many led by Infineon, have made SiC very useful in power supplies. These power supplies are used in computers,

Fig. 3.12: Forecast for CRT and Flat Panel Displays

As usage of on-demand content grows, the shift to broadbanddigital media will lead to a large migration to digital TV, which in turn will lead to much greater energy demand. A large screen HDTV set uses approximately 2.5 times more power than an analog TV made in 2000 (500 W/year vs. 200 W/ year). In addition, more Internet bandwidth directly increases the sales of servers (a multi-processor computer with large disk space), routers, and switches (these last two electronic devices are used to move information around on the Internet). Also, when these devices are off they are still on; they require what is called standby power. Standby power alone accounts for about 10 percent (or 50W 100W for a typical home) of the wasted power in power supplies (a power supply is the unit in these devices that converts the utility electricity into a form that is useful for the electronic circuits) Power electronics enable the manufacturers of the power supplies that drive consumer electronics and computer servers to increase the operating efficiency of these supplies by more than 95 percent [Fig3.13]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

315

communication equipment, TVs, audio equipment, and other electronic devices. Infineon innovations in the field are driven by commitment of resources, including large R&D expertise and investment in its own manufacturing (unlike standard digital chips, tight control over processes is a key competitive element). V. CONCLUSION With the growing demand for energy, it is becoming important to save the energy through new power electronics technology. Only few of the methods have been discussed here. Other methods are also available. Depending upon the need and utility any one method may be suitable for a particular type of situation. REFRENCES
[1] International Energy Outlook 2007, (published in May 2007 by the U.S. Energy Information Administration (EIA), http://www.eia.doe.gov/oiaf/ieo/index.html). Comparison of Energy Systems Using Life Cycle Assessment, 2004 ISBN: 0 946121 16 8 Energy Efficiency: A Worldwide Review, 2004 ISBN: 0 946121 17 6 Performance of Generating Plant: New Realities, New Needs, 2004 ISBN: 0 946121 19 2 Energy End Use Technologies for the 21stCentury, 2004 ISBN: 0 946121 15 X EC (1999). VSDs for Electric Motor Systems. SAVE study UU (2001). Icarus-4. Universiteit Utrecht. The Netherlands. Kumana,(2000).The Pinch Analysis, http://www. oilandgasnewsworldwide.com/bkArticlesF.asp?Article=17610&Sectio n=2463&IssueID=382 Worrell, E. and C. Galitsky (2005). Energy Efficiency Improvement and Cost Saving Opportunities For Petroleum Refineries. Lawrence Berkeley National Laboratory. Joosen S, K. Blok (2001). Economic Evaluation of Sectoral Emission Reduction Objectives for Climate Change. Economic Evaluation of Carbon Dioxide Emission Reduction in the Household and Services Sectors in the EU. Ecofys Energy and Environment, AEA Technology Environment and National Technical University of Athens. Harmelink, M., K. Blok, M. Chang, W. Graus and S. Joosen (2005). Opportunities for acceleration of energy savings in the Netherlands. Ecofys, Utrecht, The Netherlands. Commissioned by Dutch Ministry of Economics affair Harmelink, M., M. Voogt, W. Graus (2003). Low Carbon Electricity Systems. Methodology and results for the EU. Ecofys, Utrecht, the Netherlands. Prepared for WWF. Munkejord, M. (2006). Saving the answer. Energy saving is without doubt the quickest, most effective and most cost-effective manner for reducing greenhouse gas emissions. Montel Powernews No1 2006. Lebot B, A Meier, A Angelade (2000). Global Implication of Stand-by Power Use. IEA, France, LBNL, USA, Ademe, France (published in proceedings of ACEEE Summer Study on Energy Efficiency in Buildings) Cicero (1997). Towards a sustainable energy system a long term transition strategy. Report 1997: 6. University of Oslo, Norway. Kaan, H.F. and B.J. de Boer (2006). Passive houses: achievable concepts for low CO2 housing. ECN, Petten, The Netherlands. Keulenaar et al. (2004). Energy efficiency motor systems. European Copper Institute. Global Warming and Appliances: Increasing Consumer Participation in Reducing Greenhouse Gases, a report released by the Pew Center on Global Climate Change.

[2] [3] [4] [5] [6] [7] [8]

[9]

[10]

[11]

[12]

[13]

[14]

[15] [16] [17]

[18]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

316

Sustainability and CDM Project in the Indian Power Transmission Sector by Replacement of SF6 Gas in Circuit Breaker
naveen_yadav35@yahoo.com1, rohit12_verma@yahoo.com2, dahiyasurender73@gmail.com3

Abstract - This paper gives a case of large Sub-Station Clean


Development Mechanism (CDM), a small-scale CDM project to analyze its influence on Indias Sustainable Development. SF6 gas is being extensively used in circuit breakers. SF6 gas has the highest Global warming potential which is 23900 times of Carbon Dioxide. Extensive Survey of Sub-Stations using SF6 circuit breaker has been done in a sample state of India. Results indicate that there is high leakage of SF6 gas in atmosphere from Circuit Breakers. Quantification of leakage of SF6 gas has been done and CDM project for replacement of SF6 gas from the Sub-Station has been made for sustainable development. Baseline methodology has been proposed for SF6 gas replacement in Circuit Breaker. This paper may become a helpful tool for the policy makers and power system designers.

amount of SF6 gas is being released which cause a lot of environmental effects. Therefore, it is one of the strategic options for optimizing for promoting sustainable development in the power System through the Replacement of Circuit Breaker or replacement of SF6 gas with other kind of gas which has lower Green house effect. For the sample survey done in Haryana, SubStations of 132 KV and above capacity has been taken, as lower capacity substations are rarely using this type of circuit breakers. II. CDM PROJECT IN SUB-STATION Although CDM was designed to encourage any types of GHG reduction project development, it has a strict procedure as well as monitoring to accurately measure the reduction of Green House Gas. India has large number of Sub-Stations using SF6 gas, so there will be great potential for the development of such CDM projects by quantifying the SF6 gas that will be released during operation, maintenance, commissioning and salvage of SF6 circuit breakers. This project is a small-scale project which is referred to in paragraph 6(c) of decision 17/CP.7. These types have since been amended by 1/CMP.2, paragraph 28. A. Operation of Circuit Breaker: A circuit breaker is a switching device which can open or close a circuit in a small fraction of second. This is achieved due to its separable contacts. The closing and opening of the circuit allows to establish or to interrupt the circulation of current through the circuit under usual or unusual working conditions, such as short circuits. The interruption process of the current in a circuit breaker, begins when the movable contacts start to separate. As a consequence, the contacts area is reduced and the current density gets larger, until the energy causes the metal to begin vaporizing and an arc appears. In spite of the existence of a physical separation of the switching contacts, the established arc makes possible that the current continues flowing. The interruption of the circulating current will be achieved when the interrupting medium gets to turn the carrying arc plasma into an isolating medium. ([3]-[6]) In SF6 circuit breakers, the current continues flowing after contact separation through arc plasma of ionized SF6.

Keywords Sub-Station, Circuit Breaker, SF6 gas, CDM, Sustainable Development, Baseline I. INTRODUCTION Clean development mechanism is a market-based flexible mechanism, as defined in Article 12 of the Kyoto Protocol and the core is that developed countries and developing countries are permitted to cooperate to implement Green House Gas (GHG) emission reduction [1]. India is an important player in the carbon credit market, through the implementation of the Clean Development Mechanism. India would be responsible for 6% of the global emissions of gases that cause the greenhouse effect. This paper gives an analysis of CDM projects influence on Indias Sustainability in the case of large Sub-Stations. Extensive Survey of Sub-Station using SF6 circuit breaker has been done in a sample state of India namely Haryana. Seventeen sub-stations of 132 KV and above capacity have been surveyed and the amount of leakage of SF6 gas has been quantified. SF6 gas has the highest Global Warming potential which is 23900 times of Carbon Dioxide. Besides this a SF6 circuit breaker has a life of twenty years [2]. At the end of the life of a circuit breaker, on an average 25 kg of gas is released in the atmosphere which is called salvage gas leakage in this case. This salvage gas leakage has been uniformly distributed over the life of the circuit breaker. So on annuity bases salvage gas leakage has been taken as 1.25 kg per year. Besides this annual gas leakage from the circuit breaker has been found by the survey. Large

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

317

The circuit breaker is designed to direct a constant gas flow to the arc that extracts heat from the arc and so allows achieving its extinction at current zero. The gas flow de-ionizes the contact gap and reestablishes the required dielectric strength to prevent an arc re-strike [7]. B. Assumption: Total SF6 losses include leakage in service plus losses during salvage of the circuit breaker. SF6 losses during maintenance are not taken into account. Total SF6 losses are considered to be equivalent to 2% vol. /year as the worst case. They were considered to be equivalent to 0.5% vol. /year as the best case possible [8]. However serving in Indian conditions i.e. extreme hot and extreme cold, its total leakage is taken as 2% and this hypothesis has been tested by our survey report. It is assumed that leakage may be higher which is checked by our survey report. It has been assumed that each SF6 circuit breaker will have 25 kg SF6 gas and its life will be 20 years. So each circuit breaker on Salvage will release 25 kg gas after 20 years operation. So an annuity bases the salvage leakage of SF6 has been taken as 1.25 kg per circuit breaker per year. SF6 gas is assumed to be used in sub-stations of 132 kv and above capacity. Cost of 1 CER will be 19 [19] It has been assumed that SF6 gas will be replaced by the mixture of SF6/N2. As an immediate replacement of pure SF6, the large amount of available physical and laboratory data suggest that a 40% SF6 - 60% N2 mixture may exhibit dielectric characteristics suitable for use as insulation in high voltage equipment [9]. N 2 may have overall very less GWP as compare to SF6. So its GWP has been neglected or taken as zero. Only big size i.e. 132 kv & above sub-stations will be considered for the survey, as 11/33 kv sub-stations, if using SF6 circuit breaker, are having only 1-5 kg of SF6 gas. Generally they are sealed-for-life. So leakage of SF6 from this category of circuit breaker is very small, and hence neglected. C. Green House Gas Effect Greenhouse gases are atmospheric gases which absorb a portion of the infrared radiation emitted by the earth and return it to earth by emitting it back. Potent green house gases have strong infrared absorption in the wavelength range from 7 m to 13 m. Sulphur hexafluoride is an efficient absorber of infrared radiation, particularly at wavelengths near 10.5 m [10]. Because of the many and increasing commercial uses of SF6, there has been an increased demand for it. The estimated world production of SF6 has increased steadily since the 1970s to approximately 7000 metric tons per year in 1993 [1113, 14, 15]. Additionally, unlike most other naturally occurring greenhouse gases (e. g., CO2, CH4), SF6 is largely immune to chemical and photolytic degradation,

therefore its contribution to global warming is expected to be cumulative and virtually permanent [9]. D. Process for the CDM project The Project owner gives the project idea note and makes the Project Design Document (PDD) for Host Country Approval given by Designated National Authority. CDM Executive Board approves the baseline Methodology and sends the PDD for validation to the operational entity. After validation, registration is done by CDM Executive Board. The Project Owner monitors the emission reduction and verification is done by operational entity. Finally Certified emission reduction (CER) certificates are issued by CDM Executive Board. A project can continue to earn CERs for a maximum of either 10 years (with no change of The baseline) or 7 years with at most two renewals (i.e. up to 21 years) [16]. III. BASELINE METHODOLOGY A. Project boundary The physical boundary is the electrical grid or subset of electrical grid where the project activity of recycling and leak reduction program is implemented. The greenhouse gas included is SF6, which is commonly used as an insulator in electrical transmission and distribution grids. Any part of the Grid where SF6 leak reduction was being implemented prior to the start of project activity shall be excluded from the project boundary B. Procedure for the selection of the most plausible baseline scenario The methodology covers following categories of SF6 emissions reductions from the equipments within the project boundary: Released SF6 encapsulated in existing equipment during repairs; Released SF6 encapsulated in existing equipment during decommissioning; and Reduction in leaks by repairing the equipments. The baseline scenario shall be determined by analyzing the following potential alternatives Implementing the project activity without CDM; and Continuation of the present practice, which shall be described in the CDM-PDD. Step 1: Assessment of National Policy/regulations on SF6 List national or regional policies/regulation that either require reduction of SF6 emissions from the power sector or prescribe maintenance standards that affects SF6 release to atmosphere. If such policies exist, assess the enforcement of the policies. If above-mentioned policies/regulation exists and is enforced, then the project activity implemented without CDM is the baseline scenario.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

318

Step 2: Assess if implementation of SF6 recycling by any utility or by the utility in any part of its electrical grid is being undertaken. Identify and list the level and extent of SF6 recycling being undertaken with the region where the project activity is implemented. If some utilities do under taken SF6 recycling, are there factors that prevent the implementation of the same activity within the project boundary of the project activity. If not then the project activity implemented without CDM is the baseline scenario. Documented evidence for factors preventing implementation shall be reported in the CDM-PDD and validated by the DOE [17]. This methodology is applicable only if the baseline scenario is continuation of the present practice. C. Additionality: Additionality shall be demonstrated using the latest version of the Tool for the demonstration and assessment of additionality. In addition, it must be shown that no sectoral or regional/national-level policies exist that require the recycling or leak management of SF6 in electric utility infrastructure. The barriers listed below should be evaluated as part of the application of the latest version of the tool for the demonstration and assessment of additionality: 1. Investment barriers, other than the economic/financial barriers: Real and/or perceived risks associated with the technology or process is too high to attract investment; Funding is not available for innovative projects. 2. Technological barriers: Skilled and/or properly trained labour to operate and maintain the technology is not available, leading to equipment disrepair and malfunctioning. 3. Barriers due to prevailing practice: Developers lack familiarity with state-of-the-art technologies and are reluctant to use them; The project is the first of a kind. Management lacks experience using state-of-the-art technologies, so that the project receives low priority by management. Perceived technical and financial risks to enterprises (fears that a new technology may not work, could interrupt production, take time to perfect, or will not actually result in financial savings). Real and perceived insignificance of many investments for example, if energy efficiency (or SF6) projects are relatively small and the value of the savings achieved typically is only a small percentage of enterprise operating costs. These identified barriers are to be considered only if they would prevent potential project proponents from

carrying out the proposed project activity were it not registered as a CDM activity. IV. MATERIAL AND METHODS Thorough analyses of the Survey of sub-stations, SF6 gas emission from circuit breaker is known for Haryana state. The Survey list of the Sub-Station in this paper is arranged in Table 2. V. CALCULATION A. For Complete One Sub-Station Step 1: Baseline emissions BEy = (TGx + SVx) x GWPSF6 1000 Where: BEy = Baseline emissions during the year y (tCO2/yr) TGx = Total gas (SF6) emitted from one sub-station (kg SF6) SVx = Salvage Value of each sub-station (kg SF6) GWPSF6 = Global warming potential of SF6 (tCO2e/tSF6) = (6.676 + 1.25) x 23900 1000 BEy = 189.44 tCO2/yr Step 2: Project Emissions As we replaced pure SF6 with mixer of gas having composition 40% SF6 and 60% N2 PEy = [40 % of leaked SF6 gas i.e. (TGx + SVx)] x GWPSF6 + 1000 [60 % of leaked N2 gas i.e. (TGx + SVx)] x GWPN2 1000 = (3.17) x 23900 + (4.75) x 0000 1000 1000 PEy = 75.76 tCO2/yr Where: PEy = Project emissions during the year y (tCO2/yr) GWPSF6 = Global warming potential of SF6 (tCO2e/tSF6) GWPN2 = Global warming potential of N2 (tCO2e/tSF6) Step 3: Emission reductions Emission reductions are calculated as follows: ERy = BEy PEy Where: ERy = Emission reductions during the year y (tCO2/yr) BEy = Baseline emissions during the year y (tCO2/yr) PEy = Project emissions during the year y (tCO2/yr) = 189.44 - 75.76

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

319

ERy = 113.68 tCO2/yr Step 4: Calculation of the cost of emission reduction ERy = 113.68 tCO2/yr CER will be equal to tones of CO2 released in a year [18] So in this case CER will be =113.68 Cost of 1 CER will be 19 [19]. So total earning from CER=19 x 113.68 = 2159.92 per year per Sub- Station B. For One Circuit-Breaker Step 1: Baseline emissions BEy = (TGx + SVx) x GWPSF6 1000 Where: BEy = Baseline emissions during the year y (tCO2/yr) TGx = Total SF6 emitted from one sub-station (kg SF6) SVx = Salvage Value of each sub-station (kg SF6) = (0.485 + 1.25) x 23900 1000 = 41.46 tCO2/yr. Step 2: Project Emissions As we replaced pure SF6 with mixer of gas having composition 40% SF6 and 60% N2 PEy = [40 % of leaked SF6 gas i.e. (TGx + SVx)] x GWPSF6 + 1000 [60 % of leaked N2 gas i.e. (TGx + SVx)] x GWPN2 1000 = (0.694) x 23900 + (1.04) x 0000 1000 1000 = 16.58 tCO2/yr. Where: PEy = Project emissions during the year y (tCO2/yr) GWPSF6 = Global warming potential of SF6 (tCO2e/tSF6) GWPN2 = Global warming potential of N2 (tCO2e/tSF6) Step 3: Emission reductions Emission reductions are calculated as follows: ERy = BEy PEy Where: ERy = Emission reductions during the year y (tCO2/yr) BEy = Baseline emissions during the year y (tCO2/yr) PEy = Project emissions during the year y (tCO2/yr) = 41.46 - 16.58 = 24.88 tCO2/yr Step 4: Calculation of the cost of emission reduction ERy = 24.88 tCO2/yr CER will be equal to tones of CO2 released in a year [18] So in this case CER will be =24.88 Cost of 1 CER will be 19 [19]. So total earning from CER=19 x 24.88 = 472.72 per year per circuit breaker

Replacement cost with a mixture of SF6/N2 (40:60) will approximately be zero or negative (60% of SF 6 from breaker will be taken out and stored in containers) and will involve only man-power cost which is already employed in the existing sub-station and treated as a fixed cost in operation and maintenance. VI. RECOMMENDATION Sulfur hexafluoride is a superior dielectric gas for nearly all high voltage applications. It is easy to use, exhibits exceptional insulation and arc-interruption properties, and has proven its performance by many years of use and investigation. It is clearly superior in performance to the air and oil insulated equipment which was used prior to the development of SF6-insulated equipment. However, the extremely high global warming potential of SF6 mandates that users actively pursue means to minimize releases into the environment, one of which is the use of other gases or gas mixtures in place of SF6. An evaluation of the results of the last two decades, and a detailed analysis of the data presented in this report, indicate that no replacement gas is immediately available for use as an SF6-substitute (drop-in gas) in existing electric utility equipment. For gas insulated transmission lines and gas insulated transformers, the limitation is primarily due to the need for re-certification and possible re-rating of equipment that is already in use. For gas insulated circuit breakers there are still significant questions concerning the performance of gases other than pure SF6. However, various gas mixtures show considerable promise for use in new equipment, particularly if the equipment is designed specifically for use with a gas mixture: Mixtures of nearly equal amounts of SF6 and N2 exhibit dielectric properties that suggest that they could be used as a universal application gas for both electrical insulation and arc/current interruption purposes. In this connection, standard procedures for mixture handling, use, and recovery would need to be further developed. Mixtures of low concentrations (<15%) of SF6 in N2 show excellent potential for use in gas insulated transmission lines, although further work on their performance in practical systems is necessary. Pure high pressure nitrogen may be suitable for some electrical insulation applications. Consideration of the use of such environmentally friendly gases where SF6 is not absolutely required should be investigated and promoted. A mixture of SF6 and helium has shown promise when used in gas insulated circuit breakers, and should be investigated further. Finally, it is clear that a significant amount of research must be performed for any new gas or gas mixture to be used in electrical equipment. Such a program necessarily would require the systematic study of potential replacements,

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

320

including their physical, chemical, and performance properties. A concerted effort in this area by equipment manufacturers, utilities, government labs, universities, and gas manufacturing companies would be beneficial [9]. VII. CONCLUSIONS Since our hypothesis is 2% i.e. 0.5 kg/year leakage of SF6 gas and by survey reported this hypothesis is tested and it has been found to 1.94% i.e.0.485 kg/year. The replacement of SF6 gas by SF6/N2 40%/60% composition results in lesser GWP of the leaked gases. Average of SF6 gas filled per circuit breaker per year is found to be 0.489 kg by the survey report. Number of carbon credits earned will be 113.68 CER per sub-station per year. If this project of SF6 gas replacement by SF6/N2 mixture is approved as CDM project then earnind per sub-station per year will be 2159.92 This survey has been undertaken in the state of Haryana which has large number of 132 kv & above sub-stations using SF6 circuit breaker and the result can be extrapolated for country as whole. Technology Status/Application Electric power systems: At least one utility is known to use SF6/N2 and SF6/CF4 gas mixtures for circuit breakers used in cold weather, at transmission and sub-transmission voltage levels (i.e., 500 kV and below) [20]. REFERENCES [1] Yang Wei-hua Xu Tao Zhao Sheng-nan & Jiang Dong The Contribution of Clean Development Mechanism to CO2 Emission Reduction: a Case Study Power and Energy Engineering Conference, 27-31 March 2009. APPEEC 2009 Vol. 1. [2] Maren Istad and Magne Runde:- Thirty-Six Years of Service Experience with a National Population of Gas-Insulated Substations IEEE Transactions on Power Delivery, Vol. 25, No. 4, pp.2448-54 October 2010. [3] R.D.Garzon, High Voltage Circuit Breaker. Marcel Dekker, Inc.2002. [4] C.L. Wadhwa, Electrical Power System. John Wiley & Sons, Inc. 1991

[5] T.E. Browne, Circuit Interruption. Theory and Techniques, Marcel Dekker, Inc.1984 [6] C.H. Flurscheim, Power Circuit Breaker Theory and Design.Peter Peregrinus Ltd, 1982 [7] A.Iturregi, E.Tirres,I.Zamora,O.Abarrategui,High voltage circuit breaker: SF6 vs. Vacuum International Conference on Renewable Energies & Power Quality (ICREPQ09), Spain 15th to 17th April, 2009 [8] Jean-Luc Bessede, Werner Krondorfer Impact of High-Voltage SF6 Circuit Breakers on Global One Warming - Relative Contribution of SF6 Losses, 1st International Conference on SF6 and the Environment Agenda AEG Place, Charleroi PA, USA, November 2, 2000 pp. 1-6 [9] L. G. Christophorou, J. K. Olthoff,D. S. Green Gases for Electrical Insulation and Arc Interruption: Possible Present and Future Alternatives to Pure SF6 NIST Technical Note 1425, pp. 1-43, November 1997. [10] [10] D. Edelson and K. B. McAffe, Notes on the Infrared Spectrum of SF6, J. Chem. Phys., Vol. 19, pp.1311 12, 1951. [11] ] G. Mauthe, L. Niemeyer, B. M. Pryor, R. Probst, H.Brutigam, P. A. O'Connell, K. Pettersson, H. D. Morrison, J. Poblotzki, and D. Koenig, Task Force 01 of Working Group 23.10, SF6 and the Global Atmosphere, Electra No. 164, pp. 121131, February,1996. [12] A. Diessner, Capiel draft paper on The Influence of Switchgear on the Greenhouse Effect, Private Communication, 1996. [13] G. Mauthe, K. Pettersson, P. Probst, J. Poblotzki, D.Kning, L. Niemeyer, and B. M. Pryor, Members of WG23.10 Task Force 01. Draft document on SF6 and the Global Atmosphere, March 1995. [14] E. Cook, Lifetime Commitments: Why Climate Policy-Makers Can't Afford to Overlook Fully Fluorinated Compounds, World Resources Institute,Washington, D.C., February, 1995. [15] L. G. Christophorou and R. J. Van Brunt, SF6 Insulation: Possible Greenhouse Problems and Solutions, NISTIR 5685, July, 1995. [16] Dr. Rohit Verma, Deputy. Director N.P.T.I CDM-Hydro Potentials Uttaranchal in Himachal Pradesh [17] CDM Executive Board Approved baseline and in Monitoring methodology AM0035, SF6 Emission Reductions Electrical Grids. AM0035 / Version 01Sectoral Scope 11 and 01, 29 September 2006, pp.1-19 [18] Julien Chevallier1 Carbon Price Drivers:An Updated Literature Review Universities Paris Dauphine, France 01Sectoral Scope 11 and 01, 29 September 2006 General Conference pp. 1-2. [19] Huang Yingli, Chen Weiqing, MA Fangyuan Equilibrium Analysis on Price Paradox of Chinese CERs Market: based on the Imperfect Competition 2nd Conference on Environmental Science and Information Application Technology, IEEE 2010. [20] U.S. Climate Change Technology Program, Technology Options for the Near and Long Term, November 2003.

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

321

Source Baseline SF6 emission from utility equipment (circuit breakers)

Gas SF6

Inclu-ded? Yes

Justification / Explanation The project activity is prevention of SF6 release into atmosphere.

CO2 CH4 SF6 emission from utility equipment (circuit breakers) Project Activity SF6 CH4 CH4

No No Yes No No The project is prevention of SF6 release into atmosphere.

TABLE I SUMMARY OF GASES AND SOURCES INCLUDED/EXCLUDED IN THE PROJECT BOUNDARY


TABLE II: SURVEY LIST SUBSTATION

Sub-Station BadshaPur Narnaul Ch. Dadri Karnal Karnal Mohindergarh Palla Rewari Nising Murthal Kurukshtra Sect56,Gurgoan Sect52,Gurgoan Daulta Baad IMT,Manesar Sonepat BBMB Dadri

Size 220 KV 220 KV 132 KV 132 KV 220 KV 220 KV 220 KV 220 KV 220 KV 132 KV 220 KV 220 KV 220 KV 220 KV 220 KV 220 KV 220 KV

No. of C.B 21 11 10 4 16 18 12 14 23 5 10 15 19 22 10 10 14

Gas Filled in a C.B kg/year 2-2.5 3-4 1 2-3 4-5 2-3 2-3 4 4 2 2 0 0 4 0 5 0

Gas Filled in Sub-Station kg/year 20 7 2.5 4 7 8 9 12 12 5 7 0 0 12 0 8 0

Total no. of circuit breaker=234 Average of SF6 gas filled per sub-station per year=

Average of gas filled per circuit breaker per year=0.489 kg 6.676 kg

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

322

Micro Grid: The Concept


Mamta Chamoli
Manav Rachna International University, Faridabad Haryana
Mamtas4chamoli2000@yahoo.co.in Abstract: The emerging potential of distributed generation is to take a system approach which view generation and associated load as a subsystem or Micro Grid. Here in it the modeling of Solid Oxide fuel Cell, Micro turbine connection to electrical power system and the effects of the various parameters have been analyzed. The Matlab/Simulink software is used here. Keywords: Distributed Generation, Micro Grids, Solid Oxide Fuel Cell

I. INTRODUCTION The actual Electrical Power Systems generation is based on large scale Central generation (hydro power plants, fossil fuelled power plants and/or nuclear power plants), being the energy transmitted over long distances up to the final consumers. Consequently, the electrical energy chain in each country is largely dependent on imported fossil and/or nuclear fuels. The increasing nations dependence on imported fossil fuels, the international political instability that have been affecting the primary energy resources prices and the security of supply, together with environmental concerns and climate change issues are compromising the current energy paradigm. In fact, energy is the basis of the economic development and these shortcomings can significantly affect the economic growth and the well being of millions of citizens worldwide. Therefore, issues related with security of supply of fossil and nuclear fuels, sustainable economic development and climate changes are the main drivers for the definition of ambitious energy policies. In particular, by focusing on these issues, they defined a set of very ambitious targets: Reducing Greenhouse Gases (GHG) emissions at least 20% by 2020; Improve energy efficiency by 20% by 2020; Raising the share of renewable energy to 20% by 2020. The development of Distributed Generation (DG) and renewable electricity generation technologies are essential means in order to achieve the proposed goals. DG and renewable energy sources (RES) cover a wide range of technologies (wind generators, photovoltaic panels, fuel cells and micro turbines just to mention a few examples) that are suitable for supplying power at customers sites. However, in order to achieve such ambitious goals, the electric networks must be prepared for the large scale deployment and integration of these emerging technologies. The exploitation of DG sources can Deferral of investments on transmission and distribution systems; Reduction of losses in the distribution system; Provision of network support services or ancillary services.

A. Micro Grid Architecture The first major effort devoted to MG was initiated with the Fifth Framework Program (1998-2002), which funded the R&D project entitled MICROGRIDS Large Scale Integration of Micro-Generation to Low Voltage Grids]. Within this work, an MG can be defined as a LV network [1]. (e.g., a small urban area, a shopping centre or an industrial park) plus its loads and several small modular generation systems connected to it, providing both power and heat to local loads. An MG may also include storage devices (such as batteries, flywheels or super capacitors) and network control and management systems. This MG example includes: B. Several feeders supplying electrical loads. Micro-Generation systems based on renewable energy sources such as Photovoltaic (PV) or micro wind generators and fuel-based MS in CHP applications. Storage devices A hierarchical-type management and control scheme supported by a communication infrastructure, in order to ensure all the elements of MG are aggregated in a single Cell that is interfaced to the electrical power system in a similar way as is done in the case of ordinary Consumers or DG sources.

Fig .1: .MG Architecture

II. FUEL CELLS SYSTEMS The basic element of a fuel cell is a unit cell, as shown in Fig2. These basic elements convert the chemical energy contained in fuels directly into electric energy. Each basic fuel cell unit consists of a cathode (positively charged electrode), an anode (negatively charged electrode) and an electrolyte layer. The anode provides an interface between the fuel and the electrolyte, catalyzes the fuel reaction, and provides a path through which free electrons are conduct to the load via the external circuit. The overall fuel cell reactions occur in two steps: the oxidation reaction at the anode and the reduction reaction at the cathode. The oxidation reaction is the dissociation of hydrogen atoms into protons and electrons. The reduction reaction occurs

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

323

when the oxygen atoms dissociate and bond with the protons coming through the membrane and the electrons from the external circuit, forming water.

Fig. 2: .Schematic Diagram of fuel Cell plant

A. Solid Oxide fuel Cell A Solid Oxide fuel Cell (SOFC) was chosen to be included in the MG system, since it is particular interesting for stationary power generation applications due to the following advantages: The fuel processor requires a simple partial oxidation reforming process, eliminating the need of an external reformer. SOFC has relatively low requirements for the fuel reformation process. It can use carbon monoxide directly as a fuel, which do not require a very sophisticated reformer. As it operates at extremely high temperatures, it can tolerate relatively impure fuels. This SOFC dynamic model is also adopted in this work and is it based on the following assumptions: The gases are ideal. The stack is fed with hydrogen and air. If natural gas instead of hydrogen is used as fuel, the dynamics of the fuel processor must be included in the model. The channels that transport gases along the electrodes have a fixed volume, but their lengths are small. Thus, it is possible to consider one single pressure value in their interior. The exhaust of each channel is via a single orifice. The ratio of pressures between the interior and exterior of the channel is large enough to consider that the orifice is choked. The temperature is stable at all time. B. Micro turbines The development of micro turbines technology in the recent year is the result of intensive research and development in the field of small stationary and automotive gas turbines, reliable auxiliary system used on board commercial aircraft and turbochargers, which started its development in the 1950s. C. Simulation of the Micro turbine Connected to Three Phase Electrical Power System The circuit represents the DC to AC conversion unit of a 480 V, 50 kW Micro turbines connected to a 600 V power grid. The Turbine-Generator-Rectifier group is modeled as a simple DC current source injecting the requested DC

power into the DC bus. The inverter uses a PWM, twolevel IGBT converter (using the Universal Bridge block). The control system uses two regulators: an inner current loop controlling the current at bus B2 and an outer DC voltage regulator controlling the DC bus voltage. LC filters are used to reduce harmonic voltages produced at the 480 V bus B2 (harmonic frequencies around multiples of 15 kHz). In a two-level voltage-sourced converter (VSC) using ideal switches, the two pulses sent to the upper and lower IGBT of each arm could be complementary. However, in practical VSCs the turn-off of semiconductor switches is delayed because of the storage effect. Therefore, a time delay of a few microseconds (storage time + safety margin) is required to allow complete extinction of the IGBT which is switched off before switching on the other IGBT. Otherwise, a short-circuit could result on the DC bus. The dead time is modeled by introducing a "Discrete On/Off Delay" block (found in the Extras/Discrete Control Blocks" library) at the pulse input of the converter block. The delay specified in that block is applied on the rising edges of pulses. ("Type" parameter set to "On Delay") and the dead time is specified in the "Time Delay" parameter. In order to get an acceptable accuracy with a 15 kHz switching frequency, a sample time Ts_Power= 0.5 microseconds is used to discretize the circuit. The discrete control system uses a much larger sample time (Ts_Control=50 microseconds)

Fig:.2.2: Simulation of the Micro turbine connected to three phase electrical Power system 1.

Simulation with zero dead time Start the simulation and observe voltages, current, power and control signals on Scope1. The simulation has been automatically initialized and starts in steady state. The quadrature-axis current Iq_ref is set to zero inside the current controller in order to generate power at unity power factor. Therefore, voltage and current at bus B2 (trace 1) are in phase. The DC voltage (trace 4) is regulated at 900 V. On trace 3, you can observe the power demand (set at nominal power of 50 kW) and the measured output power at bus B2 (49 kW, because of 1 kW losses in converter and filters). The Total Harmonic Distortion (THD) is displayed above the spectrum (THD=0.74%). THD is also measured during simulation by the "Discrete THD" block (trace 5 of Scope1).

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

324

2. Impact of dead time on harmonic distortion Now open the "Discrete On/Off Delay" block menu and specify a Time delay=1e-6 s. This will introduce a dead time of 1 microsecond in the firing pulses. Repeat simulation and FFT analysis for dead times up to 3 microseconds. You will note an increase of THD when the dead time is increased. Results from frequency analysis are summarized in the table shown on the model for dead times comprised between zero and 3 microseconds.. The THD increase is caused mainly by the introduction of low frequency harmonics (mainly 5th, 7th and 11th). For a 3 us dead time the THD has increases from 0.74 % to 1.75 %. The table2.1 also indicates the modulation index increases when the dead time increases, indicating a less efficient use of the DC voltage. With a 900 V DC voltage, if dead time would be increased above 3 us, the modulation index would be higher than 1and additional distortion would be introduced because of over modulation.

[5]

Electric Power Systems Research, vol. 62, no.1, pp. 1-11, May 2002. J. Padulls, G. W. Ault, and J. R. McDonald, "An integrated SOFC Plant dynamic model for power systems simulation." Journal of Power Sources, vol. 86, pp.495-500, March 2000.

Fig. 2.4: Harmonic Distortion of B2 as a function of dead time TABLE 2.1: NOMINAL OUTPUT POWER 50 KW P.F =1.0: VDC =900V Dead Time Index (us) THD H5 H7 H11 15*KHZ +/K*60H z Modul ation

0 1 2 3

0.74 0.85 1.27 1.75

.02 .29 .72 .71

.03 .22 .50 .97

.02 .09 .34 70

.44 .44 .45 .47

0.87 0.91 0.97 0.94

III. CONCLUSION As the total harmonic distortion increase the modulation index goes on increasing for the micro turbine system connected to the power system.. REFERENCES
[1] Fang.Gao A Control Strategy for a Distributed Generation unit in Grid connected and autonomous modes of operation. IEEE Transactions on Power Systems.vol.23 no.2, pp-850-859 April 2008. J.A Pecas Lopes C.L Moreira, and AG Madureia Defining Control Strategies for Micro Grids Islanded Operation. IEEE Transcations on Power Systems.vol.21 no.2, pp-916-920 May 2006. F.Katirae Micro Grid Autonomous Operation during and subsequent to islanding process. IEEE Transcations on Power systems. vol.20.no.1, pp-248-257 January 2008. Y. Zhu and K. Tomsovic, "Development of models for analyzing the load-following performance of Micro turbines and fuel cells."

[2]

[3]

[4]

International Conference on Deregulated Environment and Energy Markets, July22-23, 2011

325

Das könnte Ihnen auch gefallen