the business scene

Abdorreza Rabiee, Heidar Ali Shayanfar, and Nima Amjady

reactive power pricing

problems & a proposal for a competitive market

IN A DEREGULATED POWER
system, the main responsibility of an
independent system operator (ISO) is
to maintain system reliability and security by using ancillary services. To
maintain the power flow limits on transmission lines as well as voltage limits
at bus bars, sufficient reactive power is
vital for the power system. In the vertically integrated utility (VIU) structure,
generation, transmission, and distribution were owned by a single entity, and
consequently reactive power provision
and voltage support were bundled with
other services in supplying electricity to
the customers. However, in the deregulated environment of the United States,
like many other deregulated electricity
industries, reactive support and voltage
control from generation source, according to FERC order 888, is known as one
of the six ancillary services. In the new
open access environment, in pursuit
of profit, the power producer has the
incentive to sell active power as much
as possible. A generator can sell its active power if only there is enough reactive power to support it. Otherwise, the
generator is no longer able to sell active
power due to system security constraints
(e.g., the voltage stability limits). So, it
is essential to establish a mechanism for
financial compensation of the reactive
power ancillary service.
The reactive power market mechanisms are not the same in all deregulated systems. In the United Kingdom, the
National Grid Company (NGC), like
Digital Object Identifier 10.1109/MPE.2008.930483

18

ieee power & energy magazine

Reactive power is one of the least understood ancillary services in the competitive markets. Reactive power is necessary to balance the total demand with the
least amount of power flowing in the transmission lines. The line limits and the
line losses are a function of the total current, both the real and the imaginary.
Minimal power factor correction at the demand increases the reactive source
requirements on the transmission system as more distant reactive sources are
used. Reactive power is a balance between the capacitance and the inductance of
the system components as energy is alternatively stored in electric and magnetic
fields. Thus, reactive power is a localized commodity, and the goal of any system
is to reduce the size of the local community as much as possible. The goal is to enable real power flows without causing congestion and without increased losses.
The selection of reactive power sources can cause congestion and thus can be
used as a tool to increase the use of more expensive generation. Reactive power
is an integrated market with the real power market as discussed in the following
word. This work shows the dependency of real power on reactive power and
discusses alternative implementations.
Gerald B. Shebl
Associate Editor, Business Scene

the ISO, arranges the tenders of reactive support services. The generator bid
reactive power support includes capacity and utilization components. These
components are (price per MVAr and
quantity on offer) and (MVAr-h price
curve), respectively. The bidder that
is selected is paid for both the capacity and utilization components through
annual bilateral contracts with NGC.
Eliminating the power pool, the United
Kingdom instituted the NETA (New
Electricity Trading Arrangements) on
27 March 2001 to deal in the wholesale
market. NETA is based on long-time
bilateral trade, and it also includes a
short-time balancing market that settles
in bidding strategy. In the newly struc-

tured market, NETA, the reactive power market is changing, and it is expected
that the reactive power market will become an exclusively tender market and
the obligatory provision of capability of
generators will be eliminated.
In Australia, the National Electricity
Market Management Company (NEMCO) mandates that generators provide
reactive power in the power factor range
of 0.9 lagging to 0.93 leading. If a generator operates in a power factor beyond the
mandatory range, it is known as ancillary
service component, and the generator
will be compensated financially based
on the lost opportunity cost (LOC).
In the Nordic Electricity Market,
including Norway, Sweden, Finland,

1540-7977/09/$25.002009 IEEE

january/february 2009

and Denmark, due to the large distance between the generator and loads,
the reactive power is provided locally.
In fact, reactive power service in Nordic countries is on a mandatory basis.
In Sweden, for example, there is not
any organized reactive power market;
in Norway, the generators are paid for
reactive power service if only the generator operates in a power factor beyond the mandatory operating range
of 0.92 lagging to 0.98 leading.
In the New York system, the ISO
(NYISO) provides reactive power
support service at embedded costbased prices. The NYISO directs the
generators that are operating within
their capability limits to produce or
absorb reactive power to control voltages within their limits. Furthermore,
the generator will receive an LOC for
the amount of revenue it loses from
the lost generation and energy trading
to produce more reactive power.

Software

Reactive Power
Related Issues
Reactive power is one of the most important ancillary services. One of the main
reasons for recent major outages in the
power systems around the world, such
as those occurred 23 September 2003
in Sweden and Denmark, 28 September 2003 in Italy, and also the August
2003 blackout in the United States and
Canada, was reported as insufficient reactive power of the system resulting in
the voltage collapse. So reactive power
is essential for the integrity of the power
system and for maintaining the system
with an acceptable margin of security
and reliability required for trading in
the electricity market. The remainder
of this section deals with the special
characteristic of reactive power.
The amount of reactive power
losses, due to the inductive nature of
transmission lines, is significant. It can
be about ten times greater than active

Training

power losses. Even in heavy load conditions, the total amount of reactive
power losses may exceed the reactive
demand of system loads. Accordingly,
reactive power losses should be considered in the evaluation of the systems
total reactive power requirement. Reactive power cannot travel very far in
the network (due to its losses), and consequently it must be produced close to
the place it is needed.
The requirement to locally produce reactive power, on one hand, and the extreme
dependence of system stability to reactive
power, on the other hand, may encourage the reactive power producers to benefit from the situation and maximize their
own profit. In other words, market power
is another serious challenge that should
be solved in the reactive power market. In
economics, market power is the ability of
a firm to alter the market price of a good
or service. A firm with market power can
raise prices without losing all customers to

Consulting

Customers

The protection engineers productivity secret

Realistic models:
Detailed bus structures
Worlds largest library of detailed,
vendor-specific relay models
Current-limited wind generators

CAPEs dynamic
TCC display makes
it easy to see how
coordination time
intervals change
when you move
the fault around.

For serious work:

Automated relay-setting
Stepped-event simulations
Automated coordination checking,
wide area coordination reviews
Settings export to relay
vendors software
1-888-240-4044 (US only) 1-734-761-8612 www.electrocon.com/pes
20

electrocon_IEEEPE.indd 1

ieee power & energy magazine

2/25/08 10:48:33 AM

january/february 2009

competitors. When several firms control

a significant share of market sales, the resulting market structure is called an oligopoly or oligopsony. An oligopoly may
engage in collusion, either tacit or overt,
and thereby exercise market power.
Voltage and reactive power are
managed through a dynamic process.
So reactive power resources can be cat-

egorized in terms of the speed of their

response to the system reactive power
requirement. Generators, synchronous
condensers, and static var compensators (SVCs), among reactive power
compensators, adjust their reactive
power continuously in both normal and
contingent states and maintain the system integrity all the time. Therefore,

compared with other reactive power

resources such as capacitor banks and
reactors, they have a greater value in
system stability support. Capacitor
banks and reactors cannot manage the
instantaneous change of system reactive power requirement. In fact, they
can control only the system static reactive power demand. Therefore, they are
less flexible than generators for their
noncontinuous reactive compensation.
The system reactive power requirement depends on the system loading
condition. In a heavy loading condition, the generators are managed to
supply reactive power but in light load
condition the generator can absorb reactive power by operating in a leading
phase to maintain the system voltage in
a natural manner. Operating in a leading phase condition results in raising
the temperature at the armature terminal, quickening the aging of armature
insulation and decreasing the static
stability margin. Accordingly, due
to the stability constraint, the outage
probability will increase and that will
lead to a high reliability charge. Consequently, the corresponding generator
should be particularly considered in financial considerations.

Different Aspects
of Reactive Power Pricing
A proper pricing mechanism needs absolute identification of the power system
and also sources of active and reactive
power. An accurate model of active and
reactive costs results in fair pricing and,
consequently, ensures that the producers
will participate with enough incentive
as well as the consumers to utilize as
much electrical energy as required. Accordingly, this situation facilitates power
system improvement and development
in addition to optimizing social welfare.
If the reactive power price is considered
lower than its actual value, there would
be no incentive to produce reactive power and almost the whole capacity of the
generator would be specified to active
power production. On the other hand,
the consumers, because of low price, increase their reactive demand. The inadequate reactive power of the system and
22

ieee power & energy magazine

HVDC_Ad_HalfPgIsland_IEEE.indd 1

2/20/08 12:49:30 PM

january/february 2009

its demand increase would lead to unfeasibility of active power transmission,

reliability decline, and finally power
system instability (voltage collapse) risk.
Conversely, if the reactive power price
become greater than its actual value,
producers like to produce more reactive
power while the consumers, because of
high cost, decrease their demands. This
time, the surplus reactive power in the
network causes voltage stability problem.
Therfore, the pricing mechanism should
be as precise and fair as possible.
The price of reactive power is
strongly dependent on the location of
its producers. But despite that, in active
power the producer location is not as
important, i.e., a far-away producer of reactive power is not favorable
as it would be a low-cost generator.
Thus adapting a uniform auction reactive power pricing, regardless of the
producers locations, is not a priority.
However, zonal reactive power pricing
is a better solution for the problem.

The cost of producing reactive power

is much lower than that of active power
because it does not involve fuel costs.
On the other hand, the capital cost of
the reactive compensator, especially
generator and synchronous condensers, is remarkably high. Accordingly,
the pricing mechanism should include
the capital cost of the components.
Due to the high dependency of
the voltage profile on reactive power,
the system operator may be forced to
price reactive power higher than its
actual price to maintain the integrity
of the system. In other words, reactive
power pricing is a multiobjective approach that all effective factors should
be taken into account during price setting. It is worth mentioning that pricing reactive power based on the rule
of active power pricing, because of
the special characteristics of reactive
power, is not logical, although some
earlier reactive power pricing works
are similar to active power pricing.

Reactive Power
Pricing Mechanisms
Profit maximization is the main principle
used in microeconomics. In a perfectly
competitive market, the profit is maximized if the marginal revenue (MR) is
equal to marginal cost (MC). Spot pricing, the actual MC, provides the correct
economic signals and clarifies both generator and consumer states in the power
market while balancing demand and supply in the power system. The spot pricing
theory can be used for real-time reactive
power pricing by using modified optimal
power flow (OPF). In 1991, Baughman
and Siddiqi, for the first time, determined
the marginal price of active and reactive
power on an arbitrary bus of system by
means of the Lagrange multiplier. However, the application of marginal reactive
pricing may not be practical, owing to
the volatility and erratic behavior of such
prices. Moreover, MC pricing is subject
to the problem of reconciling MC prices
with the requirement to recover costs.

24$3
0/7%23934%-3)-5,!4/23
+%%0).'4(%,)'(43/.
24$34ECHNOLOGIESISTHEWORLDLEADERINREAL TIMEDIGITALSIMULATION
FORTHEPOWERINDUSTRY!SPOWERSYSTEMSCHANGERAPIDLYAROUNDTHE
WORLD SO MUST THEIR SIMULATION 3INCE WORST CASE POWER SYSTEM
CONDITIONSARERAREANDDANGEROUSTOINDUCEINTHEREALWORLDTHE
2EAL4IME$IGITAL3IMULATORTESTSTHEMWITHACCURACYINAREALISTIC
ANDSAFEENVIRONMENT
7ITHOVER24$33IMULATORSINSTALLEDWORLDWIDEINCLUDINGTHE
WORLDSLARGESTANDMOSTADVANCEDSIMULATOR24$34ECHNOLOGIES
PROVIDESUNPARALLELEDPOWERFULLEADERSHIP
0OWER3YSTEM!PPLICATIONSINCLUDE
#ONTROL3YSTEM4ESTING (6$#36#&!#43'ENERATION63#7INDx
0ROTECTIVE2ELAY4ESTING ,INE'ENERATOR4RANSFORMER"USBARx
%DUCATIONAND4RAINING %XPERIMENTATION$EMONSTRATIONS2ESEARCHx
'ENERAL0OWER3YSTEM3TUDIES 3MALLTO,ARGE3CALEWITH3TABILITY&ORMAT#ONVERSION

2%!,4)-% $)')4!,3)-5,!4)/.&/24(%0/7%2).$53429
24$34ECHNOLOGIES)NC\ )NNOVATION$RIVE\7INNIPEG-ANITOBA#ANADA24"
4EL&AX\%MAILRTDS RTDSCOM\7EBHTTPWWWRTDSCOM

24

ieee power & energy magazine

january/february 2009

In some literature, reactive power

pricing is addressed as a cost allocation problem. Reactive power tracing,
graph theory, and the modified Y-bus
method are in this category. The approach of such works is to determine
the reactive power that each generator
produces to each individual load and
the system participants charge accordingly. But due to the coupling of real
and reactive power flows in transmission networks, the calculation of a
contributing factor by these methods
is, to some extent, subjective.
The costs of reactive power consist
of explicit and implicit costs. Explicit
costs refer to those that directly must
be paid such as capital costs of reactive power participants and the reactive power production costs. The production cost of reactive power, unlike
the active power, is small, and therefore the explicit cost is almost related
to the capital cost. The implicit cost
is substantially related to opportunity
cost. If a generator is required by the
ISO to produce the amount of reactive power that is forced to decrease
its active power, due to capacity constraint, the generator will loose revenue from selling active power. Thus
the generator should be compensated
financially (LOC).
Nodal reactive power pricing is another way to design a price structure.
It is the sensitivity of the generation
production cost to the reactive power
demand and usually computed by
OPF. This production cost includes the
variable costs of generation that are
strongly related to fuel cost. The main
difference between active and reactive
power costs is that the variable cost of
reactive power production is so small
that it is often negligible. In other
words, this method represents only a
small part of actual cost of reactive
power production. It is approximately
less than 1% of the active power price
in a well-designed system. Furthermore, it is so volatile, and the capital
cost of reactive power services is not
included in this pricing mechanism.
As a result, nodal reactive power pricing is not effective enough.
january/february 2009

Due to the negligible variable cost system, for example, initially, about
of reactive power production, the 80 and 20% of the total reactive powavailability of reactive power capacity er charge was considered for capacshould be a part of the reactive power ity payment and actual reactive power
charge. In this way, the possibility of production cost, respectively. Then
gaming among generators, by creating gradually the payment for capacity
circulating reactive power flows, will was eliminated, and the generator was
compensated
only
the utilization
of
be
remarkably reduced. In the British
ESA_IEEE_HalfIsland:Layout
1 11/18/08
11:10
AMforPage
1

The Perfect Software for

Utility Systems
Arc Flash Safety
Short Circuit, Protective Device Coordination
Power Flow and more
EasyPower, the most automated, user friendly power system
software on the market, delivers a full lineup of Windows-based
tools for designing, analyzing, and monitoring utility networks.
EasyPower enables you to prevent fines and litigation by
rapidly creating and implementing a comprehensive arc flash
program. Reduce risks and improve safety, while complying
with OSHA, NFPA, CSA, NEC, and ANSI regulations. Save
valuable time and money!
EasyPowers ease of use and integration increases your
productivity 3X over all other software products, allowing you
instant problem solving and Arc Flash Hazard results for any
network system.
Download a free demo, or watch our 3 minute EasyPower video at
www.EasyPower.com.

Power made easy

intelligent | intuitive | instantaneous
power system software

Watch EasyPower in action!

Online videos and Free Demo are available at:

www.EasyPower.com

503-655-5059 x35

ieee power & energy magazine

25

reactive energy. This approach made

the generator participate in the market
arrangement with enough incentive instead of admitting the default payment
for capacity. Here some of proposed reactive power markets in the literatures
are generally studied.
Gil et al. presented a competitive
pool-based electricity market that was
suggested to be implemented in countries such as England, Wales, Argentina, and Australia. In this market, the
reactive power requirement of power
system is considered as two different
components. The first component is
related to the internal loss of the reactive control equipment for producing or absorbing reactive energy that
has a relatively smooth time variation
spot price. The second component is
related to voltage regulation services
concerning system security and usually contains a very small spot price
quantity. However, its price can reach

26

ieee power & energy magazine

a high value when the system security

is in a crucial state. Therefore, due to
dissimilar behavior of two reactive
power components from the price
variation viewpoint, two different
kinds of a reactive power market are
suggested. The first one is a reactive
energy market based on losses spot
price, and the second one is a reactive
capacity market based on a reactive
regulating capacity payment. The
reactive energy market is based on a
long-term bid provided by the generator and other control element to the
system operator. The bid format consists of the maximum and minimum
amounts of reactive power generated
by the generators or other control elements and also the losses curve. This
curve represents the internal equipment losses caused by reactive power
compensation, and it will be priced
at the marginal price of hourly active power price. The system operator

dispatches the system considering the

reactive power losses of control elements as a new term in the objective
function. Reactive power price is calculated by multiplying the amount of
reactive power by spot price obtained
from corresponding losses minimization approach. The distribution utilities or large customers would adjust
their reactive power demand regarding to the current reactive power spot
price. The information exchange between agents of reactive energy market is shown in Figure 1.
In this figure cg 1 Q g, Vg 2 , sq represent the internal loss curve and reactive spot price, respectively. The reactive capacity market would be also
based on capacity bids provided by
the dynamic compensator of reactive
power to ensure system voltage security. In this market, the selected participants are responsible for voltage
regulation and their remuneration,

january/february 2009

EPF

In region I (0 to Q base) the

due to the possibility of
high security spot price, is
provided reactive power of
based on capacity payment.
the generator is necessary for
Transmission Network
(Reactive Source Dispatch)
The remuneration of each
the generators own requireelement is determined by
ment to maintain its auxilQmax , Qmin
taking into account factors
iary equipment. Therefore
such as the amount of reacthe generated reactive power
q
cg (Qg ,Vg )
Qg
q ,Vg
tive power capacity provided
in this region is not considand type of control as well
ered an ancillary service to
as impact on the expected
be remunerated. In region II
Distribution
Generation
nonsupplied energy of the
(Q base to Q A) and (0 to Q min),
Systems
Power Plant
power system. To avoid the
because of generating or abpossibility of market power,
sorbing reactive power, lossone way is to establish both figure 1. Organization of a reactive energy market.
es of the generator increase,
of the markets base on lonand it is expected to be paid
ger contracts as compared
for its service. Thus, the EPF,
to active power contracts.
in addition to availability,
Field Current Limit
Ahmed et al. presented
will contain the cost of loss
Q, MVAr
a linear programming (LP)
term. Finally in region III
B
QB
based security constraint
(Q A to Q B), the generator is
A Armature Current Limit
QA
reactive optimal power flow
managed to reduce its active
of reactive power market arpower to generate the reQ base
rangement for the U.K. elecquired reactive power. Thus
0
tricity supply industry. This
the incurred loss of revenue
PA PRated
PB
PMax
P, MW
Q Min
market was based on comcost, and consequently, the
bined reactive power capacEPF will contain all three
Under-Excitation Limit
ity and energy payment. The
terms. Accordingly, the EPF
value of reactive power supcan be determined in any
port for both the capability figure 2. Synchronous generator capability curve.
operating condition of a synand utilization of each genchronous generator. Figure 3
erator is determined. The obillustrates the EPF generator
jective function was the minas a function of the amount
imization of capability and
of generator reactive power
Operation
Opportunity
utilization payments. The
production. Figure 4 shows
Payment
Payment
analysis showed that the rethe cost incurred by a generam3Q
active capability has a higher
tor as a function of reactive
Availability
Payment
value than the utilization of
power support.
m2
m1
reactive power to the system
For the market settlement,
a0
operator, and consequently
the participants are required to
the operator may prefer a resubmit their offer to the ISO in
QA
QB
Qmin
0
Q base
active utilization based marterms of four components, a 0,
ket to limit the exposure to
m1, m2, m3, that are the prices
market power.
for availability cost, absorbfigure 3. Structure of reactive offers from providers.
Zhong and Bhattacharya
ing reactive power price, propresented the design of a competitive need to be removed through the in- ducing reactive power price, and opmarket for reactive power ancillary vestment of reactive power devices. In portunity cost price, respectively. The
service. In this market, the generator this market, expected payment func- market can be settled by minimizing
capability curve is used to analyze re- tion (EPF) is defined for the genera- total payment, transmission losses, or
active power costs and subsequently tor. The EPF consists of three terms deviation form contracted transaction
constructed a bidding framework. The of costs, availability cost, cost of loss, or by compromised objective function
reactive power market is settled on a and opportunity cost term.
including the minimization of total
Using the reactive power capabil- payment and transmission losses and
uniform price auction, using the modified optimal power flow. It also de- ity curve of the synchronous genera- deviation from contracted transaction,
termines the strategic location of the tor shown in Figure 2, three operating simultaneously. The uniform auction
network that market power exists and regions are defined for the generator. price is considered to provide enough
28

ieee power & energy magazine

january/february 2009

FACULTY POSITION IN ELECTRIC POWER

Masdar Institute of Science and Technology,
Abu Dhabi, UAE
incentive for the producers to offer their true operating and
opportunity costs. In another work, the authors presented
reactive power market using the concept of voltage control
areas. The concept of electrical distance has been used to
identify voltage control areas within a power system. The
uniform auction price for various components of reactive
service is obtained for each control area in the same way
performed for the previous work. This pricing mechanism
results in satisfactory results. Localized reactive power
markets reduced the payment of the ISO for reactive power
procurement. Market power existence in one control area
does not affect the price of the other areas of system. The
results shows that having a localized reactive power market, compared to considering one uniform price for the
system, is likely to be a better choice. In another work the
author presents a pricing mechanism considering network
reactive power devices as a participant in the competitive
market. A two-part bidding structure for network devices,
including capacitors banks and SVCs, is proposed. The
bidding structure, as shown in Figure 5, includes the availability payment bi and operating price vi for the quantities
of reactive power supplied or absorbed from the system

Opportunity Cost + Cost of Loss

Cost
Cost of Loss

Q Base QA

Q Min

QB

vi

bi

Qmin

QC

Qmax

figure 5. Reactive power bidding structure for transmission network devices.

30

ieee power & energy magazine

About the Institute: The Masdar Institute of

Science and Technology is a new and independent
non-prot, tax-exempt research and educational
institution, founded with the assistance and advice
of the Technology and Development Program at the
Massachusetts Institute of Technology (MIT). Oering
graduate-level courses in Abu Dhabi for a highly-select
student population, The Masdar Institute is dedicated
to premier engineering research and the provision
of a denitive, research-driven education. For more
information, visit http://www.masdar.ac.ae/ and
http://web.mit.edu/mit-tdp/www/

Materials must be submitted electronically

on or before February 29, 2009 as a MS Word
attachment to:
Dr. Marwan Khraisheh, Acting Provost
Co-Chair, Search Committee for Masdar Institute of
Science and Technology
Masdar Institute of Science and Technology, Abu
Dhabi, United Arab Emirates
e-mail: mkhraisheh@mist.ac.ae

Payment

Payi

Job Responsibilities: Develop and teach graduate

courses, supervise masters and doctoral students,
develop a research program, seek external funding for
such research, and participate in the Institutes service
and outreach activities.

Application Submittal Information: The

Technology and Development Program at MIT is
assisting the Masdar Institute in the search. Initial
screening of applications will begin immediately.
Application materials should include your name,
address, telephone numbers, curriculum vitae, the
specic position you are applying for (Assistant
Professor, Associate Professor, or Professor), your
current position, a description of how your experience
matches the position requirements, and e-mail contact
information for three references.

figure 4. Reactive power production cost.

vi

The Masdar Institute of Science and Technology is

seeking applicants for full-time Professors, Associate
Professors, and Assistant Professors in the eld of
Electric Power. Applicants should have a strong record
of published research and relevant teaching experience.
Fluency in English is required. Candidates should
have a PhD. Degree in electrical engineering as well
as demonstrated research excellence in at least one of
the following electrical power areas with emphasis on
renewable energy technology applications: Electric
Power Systems; Power Electronics and Drives;
Electromagnetics; Electromechanics; and Controls.

Copy to:
Professor Fred Moavenzadeh
Co-Chair Search Committee for Masdar Institute of
Science and Technology
Technology and Development Program
Massachusetts Institute of Technology
e-mail: tdpmail@mit.edu

january/february 2009

by reactive devices. In this work, the

same as previous one, the uniform auction price is considered. The objective
function is defined as a minimization
of total payment to the generator as
well as to the network devices. Results
concluded that in an area in which reactive power sources are not balance
located, the proposed VAR auction
could help the ISO determine the most
economic schemes while minimizing
total payment. Market power problems can be easily solved as compared
to the previous work which does not
include network devices.

Final Note
Under normal operating conditions, the
generators as dynamic reactive power
sources are not allowed to operate at
their maximum capacity because a margin of reactive reserve should be held on
all generators and other dynamic compensation plants. How much reactive
power should be considered as reserve?
What is the price of reactive power reserve? And how the participants of the
market operate in a corresponding manner? Who are the participants of reactive power reserve? Can only generators
or FACTS devices such as STATCOM
and UPFC be considered as participants
of reserve market? Little attention is
paid in the literature to reactive power
reserve and its price. If this approach
is achieved and a clear reactive reserve
market established, the security margin
of the system increases and the transaction of even more active power will be
possible without considering a voltage
collapse problem.
Only synchronous generators and
condensers are recognized by the ISO
as ancillary services and compensated for reactive support service; other
reactive power compensators are not
considered ancillary services. Although power system generators are
the most important components of a
dynamic reactive power support service, due to the high capital cost of
generators as well as opportunity cost,
the reactive power of the generator will
be expensive. Therefore the reactive
power compensation of the generator
january/february 2009

is not economic. On the other hand,

other compensators of reactive power
such as capacitor banks, reactors and
SVCs have no incentive to participate
in reactive power market. It seems
that the ISO should change its policy
in the reactive power ancillary service
and identify capacitor banks, reactors,
SVCs, and FACTS devices as the other
competitors of the reactive market and
compensate them financially for their
reactive power services.
The reactive power loss of transmission lines, due to their inductive
nature, is much greater than the active
power losses of transmission lines. It
can be about ten times (or even higher)
greater than the active power losses,
and it is possible that the reactive power loss of system becomes greater than
total required reactive power of loads.
Who is responsible for this remarkable
losses cost and how should be allocated among the market participants? Is it
fair to allocate the losses to the loads?
Although the generators have only active power contracts with loads, they
use the transmission system to transfer the contracted active power to the
specified loads. In the other words,
the generators that make their profit
by selling energy cause the transmission active and reactive losses. Therefore, the transmission losses should be
shared between both generators and
loads so that neither generators nor
loads subsidize each other in any way.
In this relation, Wang and Xu studied
reactive power from a new aspect. In
their work, considering voltage stability margin, the authors determine the
minimum reactive power (Qmin) that
each generator needs to transfer its
own active power through the power
system. The Qmin is determined only
for heavy load condition of system.
However, in the light load (off peak)
condition, the performance of generators differ form that of the heavy load
conditions. In the light load condition, due to charging reactive power
of lines, the generators are supposed
to absorb reactive power. Also system
reactors should connect to absorb the
additional reactive power. The Qmin

of each generator in this case has not

been determined yet, and if it can be
determined, the market based on Qmin
of generators will be one of the most
efficient reactive power markets. It not
only determines the Qmin of generators fairly at any time but also predicts
the minimum value of reactive power
needed to operate the system with sufficient voltage security margin and
consequently ensures the ISO about
voltage stability of the power system.
The crisp treatment of network operating limit constraints in the OPF
problem usually causes too conservative solutions since those limits such
as line flow and transformer loading
limit can temporarily be relaxed. In
addition, when crisp constraints are
not satisfied, it is difficult to determine
the allowable degree of constraints
relaxation. The fuzzy treatment of
operating limit constraints provides
an opportunity for the ISO to better
trade off between system security and
social welfare in electricity market.
A fuzzy set method considering the
fuzzy nature of line flow constraints
had been applied to solve OPF earlier.
Recently, a coordinated optimal power
dispatch for bilateral contract, balancing electricity, and ancillary service
market is presented by Ongsakul and
Chayakulkheeree. Also, the genetic algorithm (GA) can be used to obtain the
best strategy in solving the OPF. Therefore, it seems that heuristic methods
as well as artificial intelligence based
techniques can be useful in the design
of a reactive power market.
All of the participants of the electricity market unanimously agree to
have a competitive reactive power market. Besides, the competitive reactive
power market ensures the ISO about
the voltage security problem. However
it is required to set up a fair and clear
reactive power market mechanism to
price reactive power properly and fairly
allocate the cost of reactive power so
that market power, voltage stability
problem, reactive power losses of system, and other reactive power related
problems can be solved while maximizing social welfare.
ieee power & energy magazine

31

For Further Reading

S. Hao and A. Papalexopoulos, Reactive power pricing and management,
IEEE Trans. Power Syst., vol. 12, no. 1,
pp. 12061215, Feb. 1997.
M.L. Baughman and S.N. Siddiqi,
Real time pricing of reactive power: Theory and study results, IEEE Trans. Power
Syst., vol. 6, no. 1, pp. 2329, Feb. 1991.
J.B. Gil, T.G.S. Romn, J.J.A. Rlo,
and P.S. Martln, Reactive power pricing: A conceptual framework for remuneration and charging procedures,
IEEE Trans. Power Syst., vol. 15, no. 2,
pp. 483489, May 2000.

J. Zhong and K. Bhattacharya,

Toward a competitive market for
reactive power, IEEE Trans. Power
Syst., vol. 17, no. 4, pp. 12061215,
Nov. 2002.
Y. Wang and W. Xu, An investigation on the reactive power support service
need of power producers, IEEE Trans.
Power Syst., vol. 19, no. 1, pp. 586593,
Feb. 2004.
W. Ongsakul and K. Chayakulkheeree,
Coordinated fuzzy constrained optimal power dispatch for bilateral contract, balancing electricity, and ancillary services markets, IEEE Trans.

Power Syst., vol. 21, no. 2, pp. 593604,

May 2006.

Biographies
Abdorreza Rabiee is pursuing the
Ph.D. degree at the Electrical Engineering Department, Iran University of
Science and Technology, Tehran.
Heidar Ali Shayanfar is a full professor with the Electrical Engineering
Department, Iran University of Science
and Technology, Tehran.
Nima Amjady is a full professor
with the Electrical Engineering Department, Semnan University, Iran.
p&e

from the editor (continued from page 6)

author concludes that WOLF pricing of unscheduled flows of
electricity tends to keep the market honest and that WOLF
prices both deliveries and receipts from the grid using the
same set of formulas.

T&D
Courses
Online!
Gonzaga Universitys
T&D Engineering Certificate
Fully Online!!

Spring 2009 Courses:

Transmission Line Design-Intro (Jan)

Transmission Line Design-Advanced (Mar)
Project Development &
Construction Methods (Mar)

Flexible schedule, 3-credit, 8-week graduate courses

Designed by industry experts; $2130 per course

Ph: 509-313-5701
Email: tandd@gonzaga.edu
www.eng.gonzaga.edu/tandd
32

ieee power & energy magazine

A Look at Pricing
Our issues The Business Scene column Reactive Power
Pricing by Abdorreza Rabiee, Heidar Ali Shayanfar, and
Nima Amjady delves into the complexities associated with
applications of reactive power that may be used as a tool to
game the system. As an accompaniment to this column, Associate Editor Gerald B. Shebl offers an introduction and
explanatory look at the relationships between real and reactive power in todays competitive market places.

A New Life for Antique Meters

Associate Editor Carl Sulzberger offers us a somewhat
different, though no less interesting, History column.
The column describes the operation of a small Pennsylvania company that specializes in the production of handcrafted electric lamps derived from antique induction
watthour meters. This column is a rather personal one for
Carl, who discovered this company several years ago, and
for Virginia Sulzberger as well, who is the proud owner of
one such meter lamp in commemoration of her retirement
from NERC several years ago. I will leave further details
to Carls erudite introduction to the column.

Answers in the Wind

As promised, we are commencing our 2009 In My View
look into the future with Associate Editor Gerald B. Shebls
Blowing in the Wind! This is Gerrys version of that title
as opposed to that of Bob Dylan. Gerry provides an in-depth
view of renewable energy sources, with cogent commentary
and insightful examples of experiences worldwide. His final
conclusion of It is interesting to live in interesting times,
as the solutions are new technologies requiring innovation
and experimentation. What an interesting time to be an engineer! adds a fitting exclamation point to a fine presentation.
Surely, this opening salvo into PESs future is worth commentary from our readers, and I look forward to printing the
p&e
dialogue that I trust will ensue.
january/february 2009