Alexandria Engineering Journal (2018) 57, 755–765

H O S T E D BY
Alexandria University

Alexandria Engineering Journal

www.elsevier.com/locate/aej
www.sciencedirect.com

ORIGINAL ARTICLE

Optimal placement of DG and DSTATCOM for

loss reduction and voltage profile improvement
Fahad Iqbal, Mohd Tauseef Khan *, Anwar Shahzad Siddiqui

Dept. of Electrical Engineering, Jamia Millia Islamia, New Delhi, India

Received 26 August 2016; revised 27 December 2016; accepted 1 March 2017

Available online 23 March 2017

KEYWORDS Abstract Due to rapid depletion of fossil fuels availability and huge environmental pollution,
Distributed generation; focus is now changing from traditional method of power generation to the non-conventional DERs
DSTATCOM; located at distribution voltage level. Technological advancement and more researches in the use of
Loss Sensitivity Factor; DERs result in the formation of microgrids and active distribution network. Small DER units when
Direct Load Flow; placed in a network near to load, makes the network stable as it can work in islanding mode in case
Optimal location of any fault outside the network. This paper proposes a scheme for reduction in losses and improve-
ment of bus voltages in the active distribution network. A DG in the form of small renewable
energy resources such as solar PV, wind, biomass, is placed optimally for reduction of losses in
the network. However, placement of DG does not result in fulfilment of reactive power needs of
the network, resulting in under voltage at several buses. This under voltage problem is solved by
the optimal placement of DSTATCOM. DLF technique is used in this paper for load flow calcu-
lations and LSF along with voltage deviation is used in objective function for optimal location
of DG on MATLAB platform. The methodology is tested on 33-bus radial distribution system.
Ó 2017 Faculty of Engineering, Alexandria University. Production and hosting by Elsevier B.V. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction topology [4,5]. The increased penetration of such type of

Renewable energy resources are gaining high attention glob-
ally due to the environmental concern, technological advance-
ment, energy crisis, power outages, technical issues and
economic incentives [1–3]. For effective utilization of renew-
able energy resources such as solar PV (photovoltaic) system,
wind turbine system, biomass etc, many researchers are focus-
ing on techniques for improvement of existing network and its
* Corresponding author.
E-mail addresses: krfahadiqbal@gmail.com (F. Iqbal), tauseefkhan_
alig@yahoo.co.in (M.T. Khan), anshsi@yahoo.co.in (A.S. Siddiqui).
Peer review under responsibility of Faculty of Engineering, Alexandria
University.
sources at different locations changes the structure of the con-
ventional system into an interconnected network of smaller
systems termed as microgrids [6] which are mostly fuelled by
the renewable energy resources [7]. Microgrids connect small
pockets of loads with the nearby available distributed genera-
tion (DG) resources and energy storage devices. Energy stor-
age devices such as batteries, ultra-capacitors, flywheels
support the operation of microgrids as these devices alleviate
the intermittent nature of renewable energy resources [8]. This
intermittent nature, due to various factors such as weather
conditions and time of use results in stability, reliability and
security issues. The importance of DGs in microgrids or an
active distribution network is well established in the rural
and remote areas where the power supply from conventional

http://dx.doi.org/10.1016/j.aej.2017.03.002
1110-0168 Ó 2017 Faculty of Engineering, Alexandria University. Production and hosting by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
756
grid is not a feasible option because of various constraints such
as unavailability of transmission infrastructure, insufficient
capacity of the generation plant and high cost of energy [9].
The operation of active distribution network should ensure
that the reliability, stability, power quality and protection
coordination of the main grid will not be affected by the addi-
tion of any energy sources [10] especially when sensitive loads
are connected with the network [11]. The extensive use of Dis-
tributed Energy Resources (DERs) in microgrids and active
distribution network helps in relieving the transmission con-
gestion, CO2 footprints, reduction in the requirement of net-
work reinforcement, delivery cost of the electricity and so on
[12]. In active distribution networks, DGs are placed much
close to the electrical loads which ultimately enhance the
power reliability and quality of the system [13,14]. There are
some economical, technical and regulatory issues which arise
due to emergence of the active distribution network; especially,
in the isolated mode of operation, frequency stabilization by
matching the load and generation, voltage stabilization by
managing the reactive power and the overall minimization of
the system losses are some of the key issues which have to be
tackled carefully [15]. In active distribution network, the dura-
tion and frequency of power outages could be reduced by
placement of DG sources nearby loads. The optimal size and
placement of DG sources help in minimization of overall losses
as well as for economical operation of the active distribution
network [16,17]. Microgrids, which are fuelled by renewable
energy resources, are becoming essential part of the active dis-
tribution network. But proper operation and design of active
distribution network is still one of the major concerns of the
utility planners. In this paper, an active distribution network
is considered for study.
Various operating conditions, such as power requirement,
voltage stability and voltage deviation are kept in mind while
finding the optimal placement of DG [18,19]. Few other condi-
tions such as geographical location and availability of primary
energy sources may effect and can change the location of DG,
but these conditions are not taken into consideration for the
sake of simplicity in this paper. DG owner will have different
aim while finding the optimal location of DG whereas the net-
work operator has different conditions for consideration in
deregulated power system because DG owner tries to reduce
its investment cost while network operator tries to reduce the
power losses [20].
In distribution network, the majority of the loads are induc-
tive in nature which results in lagging pf (power factor) of the
system [21,22]. Due to lagging pf network losses may increase,
voltage profile and the system security may deteriorate [23]. In
order to avoid these problems, custom devices such as
DSTATCOM (Distribution Static Compensator), DVR
(Dynamic Voltage Restorer), UPQC (Unified Power Quality
Conditioner), and SSTC (Solid State Transfer Switch) can be
used [24]. With the help of these devices, cost effective solution
for reactive power and unbalance loading compensation in the
distribution network can be easily achieved [25].
The present work utilizes the Loss Sensitivity Factor (LSF)
technique along with voltage deviation for optimal placement
of DG and an objective function for optimal placement of
DSTATCOM based on minimization of network losses and
voltage profile improvement [26]. For the size of DG and
DSTATCOM, factors such as cost of power losses, capital cost
F. Iqbal et al.
for installation and depreciation factors and annual charges
are being formulated in the form of an objective function.
Section 2 discusses the DSTATCOM and its modelling,
while Section 3 gives the description of system. Section 4 dis-
cusses the proposed methodology, where optimal placement
of DG using LSF technique and voltage deviation is formu-
lated in Section 4.1. Section 4.2 gives the criteria for optimal
location of DSTATCOM and Section 4.3 deals with the tech-
nique used for calculation of optimal sizes of DG and DSTAT-
COM. Section 4.4 discusses the step-by-step working
methodology in the form of flowchart. Section 5 sheds some
light on results obtained in this work while Section 6 concludes
the findings of the work.
2. Distribution static compensator
STATCOM (Static Synchronous Compensator) is a type of
shunt FACTS (Flexible AC Transmission Systems) device used
in high voltage transmission network. When it is used in low
voltage distribution network, it is known as DSTATCOM.
DSTATCOM is a shunt device that can inject and absorb
real or reactive power at the bus thereby removing the sag in
bus voltages [27]. It is a voltage source inverter having a DC
energy storage device or a DC link capacitor for providing
constant DC-link voltage [28] and it connects to the distribu-
tion network with a coupling transformer. DSTATCOM con-
trols and corrects the bus voltage and power factor because it
acts as a synchronous voltage source. Fig. 1 represents a block
diagram of DSTATCOM. In case of heavy loading conditions
or during some short circuits, DSTATCOM provides or injects
required amount of current at the point of connection so as to
lift voltage profile at the load bus where DSTATCOM is con-
nected and regulates it to desired reference value [29]. Simulta-
neously, real and reactive power can be exchanged with the
help of a DSTATCOM. The type and capacity of energy
source determine the quantity of real power exchange.
DSTATCOM improves voltage fluctuations by regularly
checking and comparing the line waveform with reference sig-
nal and based on the error it injects or absorbs the required
amount of reactive current. A coupling transformer, PWM
Figure 1 Block diagram of a DSTATCOM.
Optimal placement of DG and DSTATCOM 757

control strategy, DC-link capacitor, inverter modules and AC Equating the real and imaginary parts of Eq. (1)
filter are some of the major components of DSTATCOM [30]. v0mþ1 cos h0mþ1 ¼ Reðvm \hm Þ  Reðrm im \aÞ þ xm iDSTATCOM
Magnitude of voltage sources used in DSTATCOM will decide  
0 P
the direction and magnitude of reactive current. At the connec-  sin hmþ1 þ
tion point, if voltage magnitude is higher than the magnitude 2
 
of voltage source, DSTATCOM works as a reactor and P
 rm iDSTATCOM cos h0mþ1 þ ð2Þ
absorbs the excess reactive power. Similarly, if the voltage 2
magnitude is lower than the magnitude of voltage source,
DSTATCOM works as a variable capacitor and injects the v0mþ1 sin h0mþ1 ¼ Imðvm \hm Þ  Imðxm im \aÞ  xm iDSTATCOM
required reactive power.  
P
 cos h0mþ1 þ  rm iDSTATCOM
2
2.1. DSTATCOM modelling  
P
 sin h0mþ1 þ ð3Þ
This section represents the mathematical modelling of 2
DSTATCOM. Generally, most of the distribution systems Let us use the notations:
are of radial type in which electrical power is supplied from A ¼ Reðvm \hm Þ  Reðrm im \hm Þ
the single side. Fig. 2 shows the installation of DSTATCOM
at bus ‘m + 1’. Here ‘rm’ represents the line resistance and B ¼ Imðvm \hm Þ  Imðxm im \hm Þ
‘xm’ represents the line reactance between the buses ‘m’ and
‘m + 1’. ‘pm + jqm’ and ‘pm+1 + jqm+1’ represent the local C1 ¼ rm
loads connected at bus ‘m’ and ‘m + 1’ respectively. Similarly,
vm and vm+1 are the voltages at the buses ‘m’ and ‘m + 1’ C2 ¼ xm
respectively.
Fig. 3 represents the phasor diagram for the two bus system D ¼ v0mþ1
shown in Fig. 2. Generally, in the conventional radial system,
voltages at the buses are less than 1 pu. Voltage at bus ‘m + 1’ U ¼ iDSTATCOM
is assumed to be less than 1 pu, so DSTATCOM is placed on
this bus for improvement of voltage profile. W ¼ h0mþ1
v0mþ1 \h0mþ1 : Voltage of bus m + 1 after DSTATCOM
Eqs. (2) and (3) can be written as
consideration.
vm \hm : Voltage of bus m. D cos W ¼ A þ C1 U sin W þ C2 U cos W ð4Þ
im \a: Current flow in line after DSTATCOM
consideration. D sin W ¼ B þ C2 U sin W  C1 U cos W ð5Þ


iDSTATCOM \ h0mþ1 þ P2 : Injected current by DSTATCOM. A, B, C1 and C2 are constants.
Now, Calculating the variables U and W, where
v0mþ1 \h0mþ1 ¼ vm \hm  ðrm þ jxm Þim \a  ðrm D cos W  A
  U¼ ð6Þ
P C1 sin W þ C2 cos W
þ jxm ÞiDSTATCOM \ h0mþ1 þ ð1Þ
2 D sin W  B
U¼ ð7Þ
C2 sin W  C1 cos W
Equating Eqs. (6) and (7) by taking X = sin W, we have



K21 þ K22 X2  ð2K1 DC1 ÞX þ D2 C21  K22 ¼ 0 ð8Þ
where K1 ¼ AC2  BC1 and K2 ¼ AC1  BC2 .
The solution of Eq. (8) can be written as
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi


ffi
2K1 DC1  ð2K1 DC1 Þ2  4 K21 þ K22 D2 C21  K22
X¼
 ð9Þ
2 K21 þ K22

h0mþ1 ¼ W ¼ sin1 X
At bus m + 1, injected voltage, current and reactive power
will be the following:
v0mþ1 ¼ v0mþ1 \h0mþ1 ð10Þ
 p
iDSTATCOM ¼ iDSTATCOM \ h0mþ1 þ ð11Þ
2

jqDSTATCOM ¼ v0mþ1 iDSTACOM ð12Þ

Figure 2 Single line diagram of two buses with DSTATCOM
consideration. *denotes conjugate of complex variable.
758 F. Iqbal et al.

Figure 3 Phasor diagram of voltages and current of the system after DSTATCOM consideration.

The injected voltage at bus ‘m + 1’ is given by Eq. (10),
current through DSTATCOM by Eq. (11) and reactive power
injected by DSTATCOM can be found through Eq. (12).
3. System description
Eq. (14), the equivalent current injection matrix can be
obtained using power injection values [36]. The correlation
between branch currents and bus current injections can be
found by the application of KCL in the shown distribution sys-
tem. Branch current for the 33-bus network can be written as,

Generation, transmission and distribution are the three main B1 = i1 + i2 + . . . + i16 + i17 + . . . + i31 + i32
parts of the power system. Distribution system is responsible B6 = i6 + i7 + i8 + . . . + i16 + i17
for providing power to the end consumers. The configuration B16 = i16 + i17
of the distribution lines can be classified into three types: dou- B22 = i22 + i23 + i24
bly fed, ring and radial distribution system. Radial distribution B27 = i27 + i28 + i29 + i30 + i31 + i32
system is one of the most commonly used configurations due B32 = i32
to its rugged and economical nature [31]. Radial distribution
network suffers from high losses which happens because of In general, the correlation between branch currents and bus
voltage sag and stability limitation issue. In India, such distri- current injections is given by the following:
bution losses are in the range of 13% of the total power pro- ½B ¼ ½BIBC½i ð15Þ
duction [32,33]. These losses can be minimized by optimal
where
location of DG and DSTATCOM at weak bus in the distribu-
BIBC: Bus injections to branch currents matrix.
tion network. Generally, distribution system has high R/X
The general relationship between branch currents and bus
ratio, due to which the conventional load flow models such
voltages can be given as
as Newton–Raphson, Gauss–Seidel and Fast-Decoupled
methods cannot be applied for load flow studies [34]. These ½Dv ¼ ½BCBV½B ð16Þ
methods are unable to give the accurate values of line flows
where
and line voltages in the distribution system [35]. In the litera- BCBV: Branch Current to Bus Voltage Matrix.
ture, there are many algorithms and techniques proposed for
analysis of load flow in distribution system. In this paper, an
approach based on Direct Load Flow (DLF) analysis [34] is ½BCBV ¼ ½BIBCT Z ð17Þ
being utilized for the load flow calculations as follows:
The complex load Sm of bus m can be expressed as Z ¼ diag  ðzb Þ ð18Þ
Sm ¼ pm þ jqm ; m ¼ 1; 2; . . . n ð13Þ zb = matrix for branch impedance.
From Eqs. (15) and (16), the correlation between bus volt-
where,
ages and bus current injections can be given as
n: total number of buses,
pm: real power at mth bus, ½Dv ¼ ½BCBV½BIBC½i ð19Þ
qm: reactive power at mth bus.
½DLF ¼ ½BCBV½BIBC ð20Þ
The injection of current at mth bus can be specified as Thus,
follows:
  ½Dv ¼ ½DLF½i ð21Þ
Sm
im ¼ ð14Þ For the radial distribution system, load flow solution can be
vm
obtained by solving Eqs. (22)–(24) iteratively,
where

vm: voltage at mth bus. icm ¼ Sm =vcm ð22Þ
The two relationship matrix is developed based on the 33-
bus radial distribution network as shown in Fig. 4. Using ½Dvcþ1  ¼ ½DLF½ic  ð23Þ
Optimal placement of DG and DSTATCOM 759

Figure 4 33-bus radial distribution network.

 
½vcþ1  ¼ ½v0 ½Dvcþ1  ð24Þ p2eq: ðm þ 1Þ þ q2eq: ðm þ 1Þ xl
qline;loss ðm þ 1Þ ¼ ð26Þ
where ðv2 ðm þ 1ÞÞ
c: iteration count,
v0: initial voltage. where
peq. (m + 1): total real power supplied ahead of the node m
+ 1,
qeq. (m + 1): total reactive power supplied ahead of the
4. Proposed methodology
node m + 1,
peq. (m + 1) and qeq.(m + 1) are evaluated with the help of
This section deals with the methodology proposed for selecting BIBC as shown in below Eqs. (27) and (28).
the optimum location of DG and DSTATCOM and their
respective sizes. peq: ðm þ 1Þ ¼ BIBC  pRLPM ð27Þ

4.1. Optimal location of DG qeq: ðm þ 1Þ ¼ BIBC  qREPM ð28Þ

where:
Loss Sensitivity Factor (LSF) is one of the factors being uti- pRLPM : real power matrix of the total power system,
lized for finding the best location for DG placement in the qREPM : reactive power matrix of the total power system.
given distribution network [37,38]. LSF reduces the search
space which helps in faster calculations for the optimization Now, the Loss Sensitivity Factor can be obtained as follows
process [39]. Real and reactive power loss for lth line between
‘m’ and ‘m + 1’ buses can be written as @pline;loss 2  peq: ðm þ 1Þ  rl
¼ ð29Þ
@peq: ðv2 ðm þ 1ÞÞ
 
p2eq: ðm þ 1Þ þ q2eq: ðm þ 1Þ rl
pline;loss ðm þ 1Þ ¼ ð25Þ @pline;loss 2  qeq: ðm þ 1Þ  rl
ðv2 ðm þ 1ÞÞ ¼ ð30Þ
@qeq: ðv2 ðm þ 1ÞÞ
760 F. Iqbal et al.

Figure 5 Flowchart of the proposed methodology.

Buses having highest LSF values are best suited for optimal able limit are used for finding the optimal location of DG.
placement of DG. Here, both the real and reactive power Loss The objective function for optimal location of DG is as
Sensitivity Factors along with voltage deviation from allow- follows:
Optimal placement of DG and DSTATCOM 761

maxðF1 Þ ¼ w1  Plsf þ w2  Qlsf  w3 ploss : Net real power loss of the network before the installa-
" # tion of DSTATCOM,
Xnb n o
 2
ðvm  vmin Þ þ ðvm  vmax Þ 2
ð31Þ vmin : Minimum allowable limit of voltage,
m¼1 vmax : Maximum allowable limit of voltage,
nb: total number of buses in the system.
Here, w1, w2 and w3 are weighing factors and value of vmin is
0.95 pu and vmax is 1.05 pu.
The bus having maximum value of this objective function
4.3. Optimal size of DSTATCOM and DG
F1 will be optimal location for placement of DG.

4.2. Optimal location of DSTATCOM The optimal size is calculated by considering the variables
kVAR and kW for DSTATCOM and DG respectively. The
objective of calculating the optimal size of DSTATCOM and
Objectives behind calculating the optimal location of DSTAT-
DG was improvement in overall voltage profile of the network,
COM are the reduction in overall network losses and the overall
minimization of the network losses and energy cost.
voltage profile improvement of the distribution network. After
The size of the DSTATCOM can be calculated using Eq.
placing DSTATCOM, the voltages of all the buses should lie
(12) and is rewritten here as follows:
within the acceptable range of deviation i.e. from 0.95 pu to
1.05 pu. All the operational and system constraints must be con- jqDSTATCOM ¼ v0mþ1 iDSTACOM ð33Þ
sidered and verified for the placement of DSTATCOM. Opti-
The size of DG will be PDG when the function F3 has max-
mal placement of DSTATCOM depends on the acceptable
imum value in Eq. (34).
range of voltage deviation and on overall system losses. Hence,
the objective function (F) can be formulated as [30]: maxfF3 g ¼ cE  pPLR  8760  CDG  PDG  c ð34Þ
 0  " #
X nb n o where
ploss 2 2
minfF2 g ¼  ð0:01Þ þ ðvm  vmin Þ þ ðvm  vmax Þ cE : Cost of the energy (INR/kW h),
ploss m¼1
pPLR : Power loss reduction after the installation of DG,
ð32Þ C DG : Capital cost of the DG (per kW),
where, P DG : Total capacity of the DG (kW),
p0loss : Net real power loss of the network after the installa- c: Annual rate of depreciation & interest charges.
tion of DSTATCOM,

Real power line flows (MW) Reacve power line flows (MVAR)
4
3.5
3
2.5
2
1.5
1
0.5
0
1-2
2-3
3-4
4-5
5-6
6-7
7-8

11-12
12-13
13-14
14-15

3-23

6-26
8-9
9-10
10-11

15-16
16-17
17-18
2-19
19-20
20-21
21-22

23-24
24-25

26-27
27-28
28-29
29-30
30-31
31-32
32-33

Branches (From Bus - To Bus)

Figure 6 Real and reactive power line flows for the base case.

1
0.98
0.96
Voltages (pu)

0.94
0.92
0.9
0.88
0.86
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 27 28 29 30 31 32 33
Bus No.

Figure 7 Voltage profile without DG and DSTATCOM.

762 F. Iqbal et al.

Plsf Qlsf (34). For optimal placement and sizing of DSTATCOM, the
0.04
main criterion is voltage violation. Thus if there is voltage vio-
lation in the network, DSTATCOM will be placed according
0.03
to Eq. (32) and its size will be found using Eq. (33).
0.02
0.01 5. Simulation results and discussion
0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33
Bus No. A 12.6 kV, 33-bus radial distribution network [40] is consid-
ered to identify the best possible location and size of DG
Figure 8 Real and reactive power Loss Sensitivity Factors for and DSTATCOM. All the modelling is done in MATLAB.
33-bus radial distribution network. Fig. 6 represents the real and reactive power line flows in base
case under high loading condition (i.e., without DG and
DSTATCOM) and Fig. 7 represents the voltage profile of all
Objecve Funcon (F1) the buses in this case.
0.02 The lowest voltage is at bus 18 having a value of 0.9134.
0.015
There are total of 21 buses, whose voltages are below the
acceptable range i.e. 0.95 pu. The losses in the real and reactive
0.01 power before the installation of DG and DSTATCOM are
0.005 201.8925 kW and 134.6413 kVAR respectively.
For finding the optimal location of DG, real and reactive
0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 power Loss Sensitivity Factors along with voltage deviation
-0.005 are proposed as indicated in Eq. (31). The real and reactive
Bus No.
power Loss Sensitivity Factors for the present case are being
Figure 9 Objective function (F1) for optimal location of DG. presented in Fig. 8. According to Eq. (31), bus 29 and bus
30 are the two optimal locations for the placement of DG as
shown in Fig. 9. From Eq. (34), the optimal DG size for these
4.4. Working of proposed methodology two buses 29 and 30 comes out to be 1 MW. The total real and
reactive power losses when DG placed at bus 29 are
Fig. 5 presents the step-by-step working of the methodology 126.4621 kW and 85.6374 kVAR respectively. Similarly, the
through flowchart. The location of DG will be first finalized total real and reactive power losses when DG placed at bus
using Eq. (31) and the size of DG will be found using Eq. 30 are 113.5777 kW and 79.4291 kVAR respectively. These

Without DG and DSTATCOM With DG alone With DSTATCOM alone With DG and DSTATCOM

1.02

1
Voltage at bus (pu)

0.98

0.96

0.94

0.92

0.9
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 27 28 29 30 31 32 33
No. of Bus

Figure 10 Voltage profile comparison in four different cases.

Total Acve Total Reacve

Power Loss Power Loss
250
Losses (kW & kVAR)

200

150

100

50

0
Without DG With DG With DSTATCOM With DG
& DSTATCOM alone alone & DSTATCOM

Figure 11 Total real and reactive power losses in four different cases.
Optimal placement of DG and DSTATCOM 763

installation of DSTATCOM at optimal location i.e. bus 30,

Comparave Cost (%age)

With DG With DSTATCOM
alone
Figure 12 Cost comparison in three different cases.
results show that the placement of DG (1 MW) at bus 30 is
more effective as compared to the placement of DG at bus 29.
When DG is installed at bus 30, net reduction in the real
and reactive power losses are 88.3148 kW and 55.2122 kVAR
respectively. Initially in the base case there are 21 buses whose
voltages are not in the acceptable range of deviation but after
the placement of DG at optimal location only 13 buses left
whose voltages are below the rated limit.
Now, the next step is to find out the optimal size and loca-
tion of DSTATCOM. The best location for the placement of
DSTATCOM is calculated using Eq. (32). Objective function
has minimum value at bus 30 i.e. 0.217513 followed by bus
25 having a value of 0.236701. With the help of this result,
bus 30 is selected for the placement of DSTATCOM. From
Eq. (33), the optimal size of the DSTATCOM is 3.2 MVAR.
After the installation of DSTATCOM only, net reduction in
real and reactive power losses from the base case are
3.6425 kW and 4.0083 kVAR respectively. The net reduction
in system losses is low after the installation of DSTATCOM
only as compared to the installation of DG only. After the
all the buses of the distribution system come under the accept-
able range of voltage deviation.
From the above analysis, bus 30 is selected for the place-
ment of DG and DSTATCOM. Initially when only DSTAT-
COM is considered for the placement, then the optimal size
of DSTATCOM was 3.2 MVAR. But when both DG and
With DG & DSTATCOM are optimally placed in the network than the
DSTATCOM optimal size of the DSTATCOM is also reduced from 3.2
MVAR to 1.5 MVAR. The objective behind reducing the size
of DSTATCOM was to make system more economical while
satisfying the constraints. The constraints are bus voltages,
which should lie within the acceptable range of voltage devia-
tions i.e. from 0.95pu to 1.05 pu, and maximum capacity of the
lines. When DG and DSTATCOM both are placed simultane-
ously at bus 30, total real and reactive power losses are
86.2681 kW and 61.7757 kVAR respectively.
Fig. 10 represents the voltage profile of all the 33 busses of
the radial distribution network in four different cases i.e. base
case (without DG and DSTATCOM), with DG only, with
DSTATCOM only and when both DG and DSTATCOM
optimally placed in the network. From Fig. 10 it is clear that
only in the last two cases i.e. when only DSTATCOM is placed
and when both DG and DSTATCOM are placed, all the buses
lie within their acceptable range of voltage deviation. In the
base case, the lowest value of voltage is 0.9134 pu at bus 18.
This voltage value improves to 0.9503 pu when both the DG
and DSTATCOM are used simultaneously in the system.
Fig. 11 represents overall losses in the real and reactive
power in all the four cases. There are minimum losses in real
and reactive power when both DG and DSTATCOM are opti-
mally placed as compared to the remaining three cases.

Table 1 Comparative analysis of combinations of DG and DSTATCOM.

Description Without DG and With DG With DSTATCOM With both DG and
DSTATCOM only only DSTATCOM
Size of DG (MW) at bus 30 – 1 MW – 1 MW
Size of DSTATCOM (MVAR) at bus – – 3.2 MVAR 1.5 MVAR
30
No. of buses with under voltage 21 13 0 0
Total real power loss (kW) 201.8925 113.5777 198.25 86.2681
Total reactive power loss (kVAR) 134.6413 79.4291 130.6333 61.7757

Real power line flows (MW) Reacve power line flows (MVAR)

3
2.5
2
1.5
1
0.5
0
1-2
2-3
3-4
4-5
5-6
6-7
7-8
8-9
9-10
10-11
11-12
12-13
13-14
14-15
15-16
16-17
17-18
2-19
19-20
20-21
21-22
3-23
23-24
24-25
6-26
26-27
27-28
28-29
29-30
30-31
31-32
32-33

-0.5
-1
Branches (From Bus - To Bus)

Figure 13 Power flow in each line after placement of DG and DSTATCOM at their optimal location.
764
Fig. 12 shows the comparative cost analysis in three differ-
ent cases. The cost of the entire system will come minimum
when only DSTATCOM is optimally placed in the system,
and in this case all the voltages are within acceptable limit
but the overall real and reactive power losses are on higher side
as compared to the other two cases. The cost of the system
increases if only DG is optimally placed in the network and
the total real and reactive power losses are also further reduced
from the previous case. The cost of the system is maximum
when both DG and DSTATCOM optimally placed in the sys-
tem but this case gives the minimum losses in the total real and
reactive power as compared to the above two cases as well as
all the buses voltages are within acceptable range of voltage
deviation.
Table 1 represents the comparison of different parameters
obtained in four different cases: without DG and DSTAT-
COM, with DG only, with DSTATCOM only, and with both
DG and DSTATCOM. In this Table, size of DG and DSTAT-
COM for above mentioned cases is compared. No. of buses
with under voltage problem, total real and reactive power
losses are compared for all the four cases.
Fig. 13 represents the line flows of the network after place-
ment of DG and DSTATCOM at their respective optimal
location. It can be inferred that direction of power flows in
certain lines is now different due to placement of DG and
DSTATCOM. The change in active power flow is mainly
due to the placement of DG whereas the change in reactive
power flow is being caused by the placement of
DSTATCOM.
6. Conclusion
Microgrids and active distribution networks are extremely
important topic for research nowadays, as it provides self-
sufficiency to existing networks and also provides a scope for
construction of new grid in far areas where availability of elec-
tricity is marginal or negligible. There is a need to reduce losses
in the active distribution network as energy supplied to these
grids is very small and any losses will result in further reduc-
tion in supply. In this paper, DG and DSTATCOM are opti-
mally placed for real and reactive power loss reduction and
voltage profile improvement. It is better to place both DG
and DSTATCOM at same bus because it results in minimum
losses and minimum capacity requirement of DG and DSTAT-
COM as can be concluded from the results. It is also visible in
results that use of DSTATCOM only results in voltage profile
improvement but reduction in losses is very less. In the case of
only DG placement, losses reduced drastically but voltage pro-
file improvement is minimal. So, it is suggested on the basis of
above obtained results that improvement in the overall net-
work profile and minimization of the total real and reactive
power losses will be achieved when both DSTATCOM and
DG are placed at the same bus.
References
[1] F. Peng, H. Su, P. Li, G. Song, J. Zhao, An interactive operation
strategy for microgrid cooperated with distribution system based
on demand response, in: 5th International Conference on
Electric Utility Deregulation and Restructuring and Power
Technologies (DRPT), Changsha, 2015, pp. 156–160.
F. Iqbal et al.
[2] A.J.G. Mena, J.A.M. Garcı́a, An efficient approach for the
siting and sizing problem of distributed generation, Int. J. Electr.
Power Energy Syst. 69 (2015) 167–172.
[3] M.G. Gonzalez, A. Lopez, F. Jurado, Optimization of
distributed generation systems using a new discrete PSO and
OPF, Electric Power Syst. Res. 84 (1) (2012) 174–180.
[4] F. Shariatzadeh, N. Kumar, A.K. Srivastava, Intelligent control
algorithms for optimal reconfiguration of microgrid distribution
system, in: Industry Applications Society Annual Meeting,
IEEE, Addison, 2015, pp. 1–8.
[5] S.A. Arefifar, Y.A.R.I. Mohamed, T. El-Fouly, Optimized
multiple microgrid-based clustering of active distribution
systems considering communication and control requirements,
IEEE Trans. Ind. Electron. 62 (2) (2015) 711–723.
[6] S.A. Arefifar, M. Ordonez, Y. Mohamed, Voltage and current
controllability in multi-microgrid smart distribution systems,
IEEE Trans. Smart Grid 99 (2016) 1–10.
[7] Y. Zhang, L. Xie, Online dynamic security assessment of
microgrid interconnections in smart distribution systems, IEEE
Trans. Power Syst. 30 (6) (2015) 3246–3254.
[8] T. Tuffaha, M. AlMuhaini, Reliability assessment of a
microgrid distribution system with pv and storage, in: Smart
Electric Distribution Systems and Technologies (EDST),
International Symposium on, Vienna, 2015, pp. 195–199.
[9] Y. Ito, Y. Zhongqing, H. Akagi, DC microgrid based
distribution power generation system, Power Electronics and
Motion Control Conference, IPEMC 2004 (3) (2004) 1740–1745.
[10] K.T. Tan, P.L. So, Y.C. Chu, M.Z.Q. Chen, A flexible AC
distribution system device for a microgrid, IEEE Trans. Energy
Convers. 28 (3) (2013) 601–610.
[11] T.Y. Jyung, Y.S. Baek, Y.G. Kim, Improvement of a power
quality of microgrid system interconnected to distribution
system in emergency condition, in: Transmission &
Distribution Conference & Exposition: Asia and Pacific, Seoul,
2009, pp. 1–4.
[12] M.E. Khodayar, M. Barati, M. Shahidehpour, Integration of
high reliability distribution system in microgrid operation, IEEE
Trans. Smart Grid 3 (4) (2012) 1997–2006.
[13] G. Balakrishna, Ch. Sai Babu, Multi objective distributed
generators and D-STATCOM sitting and sizing by modified
particle swarm optimization, Int. Adv. Res. J. Sci., Eng.
Technol. 2 (2015) 12.
[14] M.M. Aman, G.B. Jasmon, A.H.A. Bakar, H. Mokhlis, A new
approach for optimum simultaneous multi-DG distributed
generation Units placement and sizing based on maximization
of system loadability using HPSO (hybrid particle swarm
optimization) algorithm, Energy 66 (1) (2014) 202–215.
[15] X. Wanwan, C. Yan, W. Bin, Microgrid influence on power
quality of distribution system and its solution, in: Power System
Technology (POWERCON), 2014 International Conference on,
Chengdu, 2014.
[16] R. Majumder, A. Ghosh, G. Ledwich, F. Zare, Power sharing
and stability enhancement of an autonomous microgrid with
inertial and non-inertial DGs with DSTATCOM, in: Power
Systems, ICPS ’09. International Conference, Kharagpur, 2009,
pp. 1–6.
[17] A. Rahim, S.R. Azmi, S.A. Sulaiman, M.H. Hussain, M.H.
Zawawi, S.N. Baizura, A study on optimization techniques for
the sizing of DG in distribution system, in: International
Conference: Electrical Energy and Industrial Electronic
Systems EEIES2009, 7–8 December 2009, Penang.
[18] B.V. Reddy, Sizing of DG units using exact loss formula to
improve efficiency of radial distribution system, Int. J. Emerging
Trends Electr. Electron. 10 (1) (2014) 13–17.
[19] A. Tahand, D. Das, Novel analytical method for the placement
and sizing of distributed generation unit on distribution
networks with and without considering P and PQV buses, Int.
J. Electr. Power Energy Syst. 78 (2016) 401–413.
Optimal placement of DG and DSTATCOM
[20] Z. Moravej, A. Akhlaghi, A novel approach based on cuckoo
search for DG allocation in distribution network, Int. J. Electr.
Power Energy Syst. 44 (1) (2013) 672–679.
[21] C. Chaithra, K. Padmaja, Optimum placement of D-
STATCOM using firefly algorithm for enhancing power
quality, Int. J. Adv. Res. Comput. Commun. Eng. 4 (2015) 6.
[22] S.R. Gampa, D. Das, Optimum placement and sizing of DGs
considering average hourly variations of load, Int. J. Electr.
Power Energy Syst. 66 (2015) 25–40.
[23] P.V.V. Rama Rao, S. Sivanaga Raju, Voltage regulator
placement in radial distribution system using plant growth
simulation algorithm, Int. J. Eng., Sci. Technol. 2 (6) (2010)
207–217.
[24] S.A. Taher, S.A. Afsari, Optimal location and sizing of
DSTATCOM in distribution systems by immune algorithm,
Int. J. Electr. Power Energy Syst. 60 (2014) 34–44.
[25] B.S. Chabok, A. Ashouri, Optimal placement of D-STATCOMs
into the radial distribution networks in the presence of
distributed generations, Am. J. Electr. Electron. Eng. 4 (2)
(2016) 40–48.
[26] P. Das, Optimal sizing and placement of distributed generation
in a radial distribution system using loss sensitivity factor and
firefly algorithm, Int. J. Sci. Res. Manage. (IJSRM) 3 (4) (2015)
2611–2618.
[27] A.S. Siddiqui, M.T. Khan, F. Iqbal, Determination of optimal
location of TCSC and STATCOM for congestion management
in deregulated power system, Int. J. Syst. Assurance Eng.
Manage. (2015), http://dx.doi.org/10.1007/s13198-014-0332-4.
[28] M.R. Balu, K.M. Rosalina, N.P. Kumar, Fuzzy logic based
optimal location and sizing of DSTATCOM in radial
distribution systems, Int. J. Adv. Technol. Eng. Sci. 02 (8)
(2014) 122–129.
[29] S. Devi, M. Geethanjali, Optimal location and sizing
determination of distributed generation and DSTATCOM
using particle swarm optimization algorithm, Int. J. Electr.
Power Energy Syst. 62 (2014) 562–570.
[30] S.M.S. Hussain, M. Subbaramiah, An analytical approach for
optimal location of DSTATCOM in radial distribution system,
765
in: Energy Efficient Technologies for Sustainability (ICEETS),
2013 International Conference, Nagercoil, 2013, pp. 1365–1369.
[31] M. Natarajan, R. Balamurugan, L. Lakshminarasimman,
Optimal placement and sizing of DGs in the distribution system
for loss minimization and voltage stability improvement using
CABC, Int. J. Electr. Eng. Inf. 7 (4) (2015) 679–690.
[32] K.R. Deva, K. Ravi, Optimal size and siting of multiple DG and
DSTATCOM in radial distribution system using bacterial
foraging optimization algorithm, Ain Shams Eng. J. (2015),
ISSN 2090-4479.
[33] R. Srinivasas Rao, S.V.L. Narasimham, M. Ramalingaraju,
Optimal capacitor placement in a radial distribution system
using plant growth simulation algorithm, Int. J. Electr. Power
Energy Syst. 33 (5) (2011) 1133–1139.
[34] J.H. Teng, A direct approach for distribution system load flow
solutions, IEEE Trans. Power Delivery 18 (3) (2003) 882–887.
[35] A. Harika, D.S.K. Kanth, P.S. Babu, Sensitivity analysis based
optimal DG allocation in radial distribution system, in:
Proceedings of 23rd IRF International Conference, 5th April
2015, Chennai, India, 2015, ISBN: 978-93-82702-90-0.
[36] S. Kansal, B.B.R. Sai, B. Tyagi, V. Kumar, Optimal placement
of distributed generation in distribution networks, Int. J. Eng.,
Sci. Technol. 3 (3) (2011) 47–55.
[37] D.R. Prabha, T. Jayabarathi, Optimal placement and sizing of
multiple distributed generating units in distribution networks by
invasive weed optimization algorithm, Ain Shams Eng. J. 7 (2)
(2016) 683–694.
[38] D.R. Prabha, T. Jayabarathi, Determining the optimal location
and sizing of distributed generation unit using plant growth
simulation algorithm in a radial distribution network, WSEAS
Trans. Syst. 13 (2014) 543–550.
[39] P. Das, Placement of distributed generation in a radial
distribution system using loss sensitivity factor and cuckoo
search algorithm, Int. J. Res. Eng. Adv. Technol. 3 (2) (2015)
30–38.
[40] K.C. Divya, P.S.N. Rao, Models for wind turbine generating
systems and their application in load flow studies, Electric Power
Syst. Res. 76 (2006) 844–856.