Electrical Power and Energy Systems 70 (2015) 83–90

Contents lists available at ScienceDirect

Electrical Power and Energy Systems

journal homepage: www.elsevier.com/locate/ijepes

Congestion management in power systems – A review

Anusha Pillay a, S. Prabhakar Karthikeyan a, D.P. Kothari b,⇑
a
School of Electrical Engineering, VIT University, Vellore, Tamil Nadu 632 014, India
b
Wainganga College of Engineering & Management, Wardha Road, Gumgaon, Nagpur 441108, India

a r t i c l e i n f o a b s t r a c t

Article history: In the past few decades, restructuring has overtaken all possible domains including the electric supply
Received 2 July 2014 industry. Restructuring has brought about considerable changes by the virtue of which electricity is
Received in revised form 9 November 2014 now a commodity and has converted into deregulated one. Such a competitive market has paved way
Accepted 14 January 2015
for innumerable participants. This has led to overloading and congestion of transmission lines. Moreover,
open access transmission network has ingenerated a more intensified problem of congestion. Thus,
congestion management in power systems is germane and of central importance to the power industry.
Keywords:
In this paper a review work is carried out to unite all the publications in congestion management.
Conventional methods
Optimization
Ó 2015 Published by Elsevier Ltd.
Demand response
Hybrid market
Congestion

Introduction Congestion management refers to avoiding or relieving conges-

tion. In a much broader sense, congestion management can be clas-
With the ever growing demand and recursively increasing sified under two broad paradigms. One is the cost free method and
advancements in technology, electricity market was also shifted other is the non-cost free method. The cost free measures include
from being regulated to deregulated market. In the earlier days, those which are at the disposal of the Transmission System Oper-
the electricity market was under a monopoly. One large utility ator (TSO). These employs modifying the topology of the network,
had the authority of generation, distribution as well as transmis- installment for transformer taps, operation of conventional com-
sion, usually known as vertically integrated utilities. pensation devices e.g. phase-shifters and use of Flexible AC Trans-
This called for the need of restructuring the industry. But there mission System devices. These are coined as cost free measures
were a lot of challenges in a deregulated market which includes because of nominal economical consideration. Hence, these mea-
choosing an apt auction strategy for the electricity, mitigating mar- sures would not involve the generation and the distribution com-
ket power of the participants, alleviating transmission congestion panies. The non cost free measures include generation
and related locational price spikes, maintaining system reliability, rescheduling and curtailment of load transactions.
assessing market equilibrium and market efficiency [1]. Seema and Lakshmi [2] have embodied the conventional con-
Of these challenges the recently focused one is congestion man- gestion management methods through a comprehensive review.
agement. In an open access environment, power flows from suppli- But the topics covered are confined to general nodal pricing
ers to consumers with transactions happening through method, Price Control theme, congestion management through
transmission lines. Congestion takes place when the transmission Genetic Algorithm, fuzzy logic, voltage stability, nodal and zonal
lines are not sufficient to transfer the power according to market congestions, a few points related to congestion management
desires. Thus, congestion management is a tool for efficiently mak- through FACTS devices and market based analogy.
ing use of the power available without violating the system The publications referred in this paper are from the following
constraints. journals, proceedings and books.

IEEE TRANSACTIONS ON POWER SYSTEMS.

⇑ Corresponding author at: 102, Megh, Yashodham Enclave, Prashant Nagar, Near IEEE TRANSACTIONS ON ENERGY CONVERSION.
Ajni Chowk, Wardha Road, Nagpur 440 015, India. Cell: +91 830 880 22 18, +91 IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION.
8793372375. International journal of electrical power system research (IJEPSR).
E-mail addresses: [email protected] (A. Pillay), [email protected] International journal of electrical power and energy systems
(S. Prabhakar Karthikeyan), [email protected], [email protected]
(D.P. Kothari).
(IJEPES).

http://dx.doi.org/10.1016/j.ijepes.2015.01.022
0142-0615/Ó 2015 Published by Elsevier Ltd.
84 A. Pillay et al. / Electrical Power and Energy Systems 70 (2015) 83–90

ENERGY POLICY. LMP reflects marginal cost comprising of both the generation
PROCEEDINGS OF VARIOUS IEEE CONFERENCES, REPORTS, marginal cost and transmission congestion cost. The costs are paid
REVIEWS & BOOKS. at respective nodes and the congestion charge is paid as a differ-
ence of LMP between the sink and source of a particular transac-
The review paper is organized as follows: Section ‘Conventional tion, multiplied by the energy volume of each transaction. The
methods of congestion management’ embraces upon the conven- advantage of this method is that it gives appropriate signals related
tional methods of managing congestion. Optimization techniques to setting up of new transmission lines and installation of genera-
are discussed in Section ‘Optimization techniques and expert sys- tors. But it also involves strategic gaming among the participants
tem’. The demand response management is included in Sec- who wish to increase their profits.
tion ‘Demand response management’. Section ‘Congestion
management in hybrid market’ deals with the congestion manage- Uplift cost
ment in hybrid markets. In order to get a glimpse of how helpful
these techniques actually are, real time examples of various coun- The former UK pool includes the congestion cost in the uniform
tries are discussed in Section ‘Congestion management adapted in price by addition of uplift cost.
various countries’. Section ‘Conclusions’ gives the conclusion, Uplifts refer to the cost of security and equal to the difference
review and recommendations for future work. between the total cost of supply in constrained and unconstrained
cases [10].
Conventional methods of congestion management The uplift cost includes transmission services uplift (the cost
incurred due to the physical limitations of the network), energy
Nodal pricing method uplift (the costs of demand forecast errors and generator short-
falls), reactive uplift (maintains system voltage within limits),
The nodal prices vary according to the geographic locations, unscheduled availability payments (the capacity payment paid to
hereby giving them the name Locational Marginal Prices gensets that are available but are not required to run). The compo-
(LMP). The nodal prices lead to generation of heavy surplus. nents are explained in detail in [3].
This surplus is then utilized to pay the ‘contract rights’. A con- The uplift cost can be known mathematically as follows:
tract right gives the right holder the ability to inject power at
one node and remove it at another in the transmission network PPP ¼ SMP þ Capacity Payment ðCPÞ ð1Þ
[3].
LMPs are calculated in the following four ways. First, the system PSP ¼ PPP þ Uplift ð2Þ
can be made to operate after 1 MW use and before 1 MW use. The where
difference in the two costs of operation then gives the LMP. This is
barely practical. Second, LMP based on sensitivity factors of the
marginal generators. The cost of generation at each generator can SMP it is the bid price of the marginal unit (the most
be calculated using this method. Third, LMP can be calculated as expensive generator) required to meet forecast
dual variables or Lagrange multipliers from optimal power flow demand in a market period
(OPF). Fourth, LMP can also be obtained from ‘transposed jacobian PPP pool purchase price, it is the price calculated before
matrix’ where the limiting constraints replace the rows and col- the day of trading
umns of the matrix [2]. PSP pool selling price, it is the price paid by buyers and
In [4], the author has aimed to improve the outcome of social paid to generators
welfare problem by introducing some performance indices to com- CP it is the payment for any available capacity,
pare different dispatch options thereby reducing total congestion irrespective of whether the generators generated or
cost along with. The method has been applied to three bus and not. It may rise during periods of shortages but falls
eight bus test systems. when system capacity exceeds demand
In [5], Sood et al. have proposed a generalized deregulated Uplift uplift payment covers the costs of transmission
model applied to the IEEE-30 bus test system. This model dis- (including transmission system losses) and is the
patches the pool combined with privately negotiated bilateral difference between the unconstrained schedule and
and multilateral contracts and aims to maximize the social benefit. the cost on the trading day
In [6], Acharya and Mithulananthan, have analyzed and evalu- Bid price
ated the effect of TCSC on spot price and congestion in deregulated predetermined variable costs
electricity markets. The results thus obtained help in proving that
TCSC can lead to reduction in congestion and losses.
If in an unconstrained dispatch environment, a private generator is
Jokic et al. [3] has proposed a novel control scheme by using
selected but not allowed to generate due to the system constraints,
nodal prices in electrical power systems to achieve congestion
adjustment calculations are made to provide compensation for the
management and optimal power balancing. They have also pre-
generators.
sented an explicit controller which detects all line flow constraints
in steady state and guarantees economically steady-state
Adjustmentconstrained OFF ¼ ðCapacity  GenerationÞ  ðPPP
operation.
Kang et al. implemented a novel zonal marginal pricing. The  Bid PriceÞ ð3Þ
approach is based on congestion contribution identification and
Now, in case the original dispatch violates the security con-
sequential network partition which was implemented on IEEE-39
straints, re-dispatch is needed.
nodes system [8].
Often, in the re-dispatch process, generators are paid the PPP
In [9], Murali et al. used Bat algorithm based DC optimal power
which is lower than the bid price. This requires adjustment calcu-
flow (DCOPF) to evaluate spot prices in single auction model while
lations as well.
minimizing fuel cost and improving social welfare of the system.
Bat algorithm is proved to perform better than Linear Program- Adjustmentconstrained ON ¼ ðGenerationÞ  ðBid Price  PPPÞ ð4Þ
ming (LP) and Genetic Algorithm (GA).
 A. Pillay et al. / Electrical Power and Energy Systems 70 (2015) 83–90 85

These adjustment costs are then included in generator incomes. FACTS devices can be classified into three categories: series con-
troller, shunt controller and combined series-shunt controller. The
Generator incomes ¼ ðcapacityÞ  ðPPPÞ þ Adjustment ð5Þ
series controllers like thyristor controlled series compensator
However, the generators are not charged for congestion exclu- (TCSC), static synchronous series compensator (SSSC) and thyristor
sively. Hence this method does not provide correct information controlled phase-angle regulator (TCPAR) are used in alleviating
regarding setting up of new transmission lines. line overloads and increasing transfer capability by controlling
power flow. The shunt controllers such as SVC (static var compen-
Price Area Congestion Management (PACM) sator) and STATCOM (static synchronous compensator) can be
employed to compensate voltages by injecting, directly or indi-
PACM is extensively practiced in Nordic countries as well as in rectly, reactive power at the low voltage buses. The combined ser-
India. Basically, it is encouraged in bilateral, de-centralized, day ies–shunt controllers such as UPFC (unified power flow controller)
ahead type markets. Each area consists of several generators and can be used in the system to release the power flow congestion as
loads offering different bids [11]. Initially, considering the overall well as support voltages [13].
bids and offers in the area a system price is calculated (Ps). If power In 2004, Song et al. discussed various types of FACTS devices
flow between the bidding areas exceeds the line limits, area price is and determined the location of these devices using optimization
calculated (Po). The difference between Ps and Po gives the capacity techniques considering security indices. The proposed methodol-
fee. This capacity fee is then used to decide whether or not to ogy is implemented on IEEE-57 bus system where FACTS were
install an additional unit. In a surplus area, the area price is fixed made to operate under normal conditions and line contingency sit-
in such a way that it fulfills the extra demand whose value is equal uations [14].
to the capacity in the constrained line. In a deficit area, the area In [15], Yu and Lusan, implemented a multiple time period wel-
price is adjusted so that the suppliers could supply an additional fare maximization model for installment of FACTS device in the
amount equal to the capacity of the line. deregulated power systems along with considering losses. The
results show that no FACT device can be added on a 1 h basis, suf-
ATC based congestion management ficient amount of hours has to be employed for the plan to con-
verge to an optimal placement solution.
ATC refers to the ‘‘Available Transfer Capability’’. It is the mea- The use of FACTS combined with optimization methodologies is
sure of the extra power that could be transferred on the transmis- one of the leading techniques in the present power scenario that is
sion line. capable of efficiently curtailing congestion. Effective power flow
control through FACTS applied to the transmission congestion
ATC ¼ TTC  TRM  ðETC þ CBMÞ ð6Þ
could improve the transfer capacity and also limit the voltage sta-
where bility and voltage security constraints. Reddy et al. [16] portrayed
the use of FACTS devices for eliminating transmission congestion.
TTC the maximum amount of power which can be
The authors have used Genetic Algorithm to locate FACTS devices.
transferred over the network while satisfying all
The Interline Power Flow Controllers (IPFC) are the latest type of
security constraints
FACTSs that are used to control power flow in the systems. In [17],
TRM the margin required for uncertainties in the system
the author presented an optimal power flow method incorporating
conditions
IPFC. IPFCs can be used to control power flows in multiple trans-
ETC the existing transfer commitments
mission lines.
CBM the margin reserved by load serving entities for
In [18], Karami et al. aimed to identify an optimal solution for
generation reliability requirements
the location of STATCOM and to determine the capacity of an aptly
located IPFC for managing congestion simultaneously. The pro-
Every ISO keeps a track of ATC of a particular congested line. ATC is posed method uses artificial intelligence as the optimization tech-
then placed on the website OASIS (Open Access Same-time Informa- nique to improve the voltage security and overcome the problem
tion System) controlled by ISO. ATC can be calculated using follow- of congestion.
ing methods: optimal power flow method, Continuation method, Besharat and Taher, in [19], suggested optimal location of
Real Power Transmission Congestion Factor (PTCDF) and Line Out- TCSCs based on real power performance index and reduction
age Distribution Factor (LODF). This is one of the most basic method of total system VAR power losses. In this paper two sensitivity
of relieving congestion and is in practice in many areas. based methods are used to determine the location. They are
ATC is one of the important factors that is determined in a implemented on two 5 bus power system. The results showed
transmission line. The authors in [12] have proposed a method to that sensitivity factors along with TCSC cost can be used to find
determine ATC using optimal power flow method (OPF) along with the optimal location.
ZIP model (constant power) and then modeling of FACTS devices. In [20], Gitizadeh and Kalantar analyzed the installation of
The line is then analyzed with and without FACTS and the respec- FACTS in the transmission lines. The authors have used a Sequen-
tive impacts on ATC. The result shows that ATC increases in the tial Quadratic Programming (SQP) problem for evaluation of static
presence of UPFC and Sen Transformer. The FACTS devices are dis- security margin with congestion alleviation constraint and the
cussed in detail in the following section. next stage involves a simulated annealing method for optimizing
the problem.
Flexible AC Transmission Systems (FACTS) devices Rajalakshmi et al. have described how efficiently the installa-
tion of FACTS could improve the transfer capability. The authors
There are two measures by which congestion could be man- have proposed a method to appropriately locate the FACTS devices
aged. One is the cost free measure and second is the non cost free based on the performance index and reduction of total var losses
measure. The cost free measures are more advantageous since they [21].
avoid any sort of economic discrepancies. The cost free measures In [22], Esmaili et al. proposed a multi-objective framework for
include installation of FACTS devices especially thyristor controlled congestion management wherein the authors have optimized
series compensator (TCSC) and unified power flow controller three objective functions simultaneously. The objective functions
(UPFC). include total operating cost, voltage and transient stability mar-
86 A. Pillay et al. / Electrical Power and Energy Systems 70 (2015) 83–90

gins. The proposed method efficiently locates and sizes series Benabid et al. improved the conventional PSO for optimal loca-
FACTS devices on the lines which are congested the most by prior- tion of FACTS devices like TCSC and SVC by introducing new vari-
ity listing using Locational Marginal Prices (LMP). ant specialized in multi-objective optimization problem and
FACTS devices thus play a very important role in reducing the called as non-dominated sorting PSO (NSPSO) [28]. In this tech-
congestion costs by their optimal location. Innumerable optimiza- nique, the entire population is sorted into various non-domination
tion techniques have been employed for their appropriate location fronts. The first front is completely a non-dominant set in the cur-
on the transmission line. Some of them have already been dis- rent population and the second front being dominated by the indi-
cussed while others are broadly classified in the following section. viduals in the first front only and the front goes so on. The authors
have taken IEEE 30 bus and Algerian 114 bus system for validating
their results. They maximized static voltage stability margin
Optimization techniques and expert system
(SVSM) and minimized real power losses and load voltage
deviation.
Congestion management is basically a non-linear program
Panida et al. introduced time-varying acceleration coefficients
involving a lot of variables which could be solved using optimiza-
in the conventional PSO and called as PSO-TVAC which is used
tion algorithms [23]. The most often used optimization techniques
for the optimal congestion management [29]. Their algorithm is
are categorized as Genetic Algorithm (GA), Particle Swarm Optimi-
tested on IEEE-30 bus and 118 bus systems. Results were com-
zation (PSO), Bacterial Foraging Algorithm, Expert System
pared with classical PSO and PSO with time varying inertia weight
Approach and a few other evolutionary strategies.
(PSO-TVIW). In PSO-TVIW, the main concept remains same as con-
ventional PSO except the velocity equation is modified by the con-
Genetic Algorithm striction factor and the inertia weight w decreases linearly as it
moves through iterations. PSO-TVAC is an extended version of
It has been proved Genetic Algorithm is a powerful approach to the earlier one where all coefficients are varied with iterations.
solve many non-linear programming problems especially in coun- It is an efficient optimization method. It is inspired by the
ter trading. The approach of GA with alleviating congestion using behaviors of organisms. ‘Swarm’ represents a group of particles.
the concept of chromosomes is elucidated in [24]. Every particle has two vectors associated to it: position vector
In 2006, Granelli et al. proposed a tool which provides the sys- and velocity vector. Joshi and Pandya in [30], have applied PSO to
tem operator to find the optimal configuration of a power trans- the problem of congestion management. The authors have used
mission system [25]. This tool is to suit the congestion Newton–Raphson method (NRM) to find the overloaded lines.
management problem. The authors used 33-bus CIGRE sample sys- The sensitivity factor is then found out. Depending upon the sensi-
tem and Italian based 432-bus EHV network to validate their work. tivity factors the generators on the congested lines are identified.
Firstly, they used a deterministic approach by modeling the recon- Then PSO is applied. The value of the vectors would give the values
figuration problem as a linear program with mixed (real and inte- of active power re-scheduling and amount of generator voltage re-
ger) variables. Secondly, Genetic Algorithm is used to take care of scheduling. NRM load flow is run. The values of line flows, active
the N and N  1 security constraints in a multi-objective optimiza- and reactive power rescheduling, line flows and voltage magni-
tion problem. tudes are obtained. From this the global best and personal best par-
In [26], Surender Reddy et al. have construed beautifully single ticle is found out.
objective and multi-objective optimization techniques for exem-
plary choice of location and size of FACT devices in the transmis- Bacterial Foraging Algorithm
sion line. Initially, GA is used for single objective optimization.
Later, Strength Pareto Evolutionary Algorithm (SPEA) is introduced In [31], Panigrahi and Ravikumar Pandi have proposed a new
for multi-objective optimizations i.e. considering two or more ideology based on the concept of survival of the fittest in any evo-
objective functions at the same time. In GA, initial random popula- lutionary process. The algorithm is called Bacterial foraging tech-
tion is generated then selection, crossover and mutation are per- nique. It is a stochastic optimization algorithm. The strategy
formed to select the good one of the lot. The configuration of could be explained by the behavior of Escherichia coli bacteria. It
FACTS devices is represented as three parameters: location, rating involves four stages to survive in a human intestine namely
and type. Each individual compromises of a number of strings. swarming, chemo taxis, reproduction, and elimination–dispersal.
Every individual has three values. The first indicates the location, The authors have intended to solve the problem of congestion
the second indicates the type and last value depicts the rating. management using the aforementioned methodology combined
Later, for multi-objective optimization SPEA is used. There is no with Nelder-Mead method. The authors with their calculative mea-
unique solution to a multi-objective function thus; the aim is to sures have concluded that the given algorithm is better than the
find all the tradeoffs solutions. Genetic Algorithm as well as swarm optimization.
Fuzzy adaptive bacterial foraging (FABF) approach is used for
Particle Swarm Optimization (PSO) congestion management by rescheduling the active power genera-
tion optimally [32]. Selection of generators is based on generator
Particle Swarm Optimization (PSO) is another evolutionary sensitivity over the congested line. The results are compared with
algorithm which is used for optimal location of FACTS devices. Sar- conventional bacterial foraging and PSO. IEEE 30-bus and Indian
avanan et al. explored the application of this algorithm to find the 75-bus system is taken to validate the results.
optimal location of FACTS devices like thyristor controlled series
compensators (TCSC), static var compensators (SVC) and unified Expert System Approach and evolutionary strategies
power flow controller (UPFC) [27]. Their constraints are minimum
cost of installation and to improve system loadability. IEEE 6, 30, In order to avoid multiple congested conditions, all possibilities
118 systems and a practical system from Tamil Nadu Electricity of transmission lines are to be explored. This could act as a help to
Board are used as the test system for their simulations. The authors the ISO in finding the most undisturbed state. A Congestion Man-
found that UPFC gives maximum system loadability on IEEE test agement System (CMS) makes all the searches. The search repre-
systems with high installation cost while the TCSC requires very sents a mixed integer linear programming problem. It utilizes the
less installation cost with better loadability. concept of corrective switching of generators. This change in topol-
 A. Pillay et al. / Electrical Power and Energy Systems 70 (2015) 83–90 87

ogy can create a lot of large dimension problems. These kinds of coordination of demand responses and FACTS devices with the
problems could be solved using Evolution strategies. The strategy conventional generators is the highlight of this paper [41].
is explained and exemplified in [33].  EðiÞ
Padhy in [34], proposed an hybrid model for congestion man- qðiÞ þ CRðiÞ þ penðiÞ
LðiÞ ¼ L0 ð7Þ
agement. In the first stage classical optimal algorithm is used to q0 ðiÞ
determine the curtailment strategies for specified amounts of real where
and reactive power. In the second stage fuzzy logic matrix is
applied to select optimal transaction strategy. The model is tested L0(i) customer demand before demand response program
on IEEE 30 bus system and the results prove fuzzy logic to be L(i) customer demand after demand response program
efficient. q(i) electricity price
In [35], Yesuranatam and Srilatha have proposed an Expert Sys- CR(i) paid as incentive to the customer for each unit of load
tem Approach for security and economy oriented power systems. reduction
The proposed approach alienates congestion by re-scheduling of pen(i) reduction level requested from penalty
generators by involving minimal cost in the re-scheduling process. E(i) self-elasticity of the load
The minimal cost is based upon the relative electrical distance con- q0(i) market price prior to demand response
cept (RED) and incremental fuel cost of generators. implementation
Farahmand et al. proposed hybrid mutation PSO (HMPSO) tech-
nique for estimating Available Transfer Capability (ATC) [36]. Kumar and Sekhar in their paper [42] address the congestion man-
Firstly, the authors achieved the same using GA, PSO. Secondly, agement by the same approach as [41]. The difference between
optimal installation and capacity allocation of FACTS devices using these papers is (i) the test system used (ii) congestion cost function
multi-objective optimization problem is presented. Results are val- and (iii) the congestion cases considered.
idated using IEEE 30-bus and 57-bus system. SVC and TCSC are the X
nd X
nd
devices considered in this paper. CC ¼ DCðPup
d Þ þ DCðPdown
d Þ ð8Þ
Taher and Amooshahi introduced hybrid immune algorithm d¼1 d¼1
(HIA) such as immune Genetic Algorithm (IGA) and immune parti-
cle swarm algorithm (IPSO) to locate unified power flow controller DCðPup up up
d Þ ¼ k2  DP d  bsmv a þ Rd ð9Þ
(UPFC) optimally to achieve optimal power flow [37]. The authors
have compared the results with other method like Genetic Algo- DCðPdown
d Þ ¼ k2  DPdown
d  bsmv a þ Rdown
d ð10Þ
rithm, PSO and IA and found that IPSO algorithm gives better result
in terms of minimum cost which includes total active and reactive where
production. IEEE 14-bus and IEEE 30-bus test system is used to val-
idate their results. CC congestion cost
Bhattacharyya and Gupta [38], have used fuzzy logic concept to
CðP up
d
Þ and CðPdown
d Þ linear bid functions
reduce congestion management. They aimed to find optimal loca-
DPup and P down up and down demand at bus-i for
tion of FACTS devices such as thyristor controlled series compensa- d d
congestion management bsmva base
tor (TCSC) and static var compensator (SVC) using fuzzy logic. The
MVA
proposed method is then compared with other evolutionary algo-
rithms like Genetic Algorithm (GA), Differential Evolution (DE) Rup
d
and Rdown
d
constants of bid function in $/h
and Particle Swarm Optimization (PSO). The authors have then k2 demand cost coefficient of a demand bid
concluded from the results that fuzzy logic is above all of the other function submitted to the ISO
methods.
Tarafdar Hagh et al. brought (Static Compensator) STATCOM
and Static Series Synchronous Compensator (SSSC) to predict their Congestion management in hybrid market
impact on power systems. They used an index to measure predict-
ability [39]. The authors also stress the importance of predictability Renewable energy sources are becoming more important day
concept in optimizing problems and warned the power system by day and the ways to generate electricity from these sources
operators for ignoring this issue. The results are validated using are posing a lot of challenges. Voltage instability and power flow
IEEE 14-bus and 57-bus system. control tops this list. A lot of devices and methods have been devel-
oped to overcome this including the FACTS devices as discussed in
Section ‘Conventional methods of congestion management’.
Demand response management Sood and Singh [43], have discussed the application of Locational
Marginal Pricing (LMP) to the hybrid market. The paper analyses a
It enables the customers to participate in the market power. It new optimal model for congestion management involving LMP.
has two major schemes: EDRP (Emergency Demand Response The test results show how it is useful to reducing congestion costs.
Management) and the other one re-dispatches the generations. The authors have also discussed the social benefits of promoting
Demand response is divided into two programs: Incentive renewable energy sources.
based and time based. EDRP belongs to the incentive based. Large With the advent of integration of renewable energy sources into
consumers, who want to reduce a portion of electricity, participate. the grid, challenges like relieving congestion in the hybrid market
The incentives to the consumers will be paid by the consumers. seek focus. In [44], Singh et al. proposed a methodology for sorting
The model is described in detail [40] by Parsa Moghaddam. the problem of congestion management in hydro-thermal pool
Yousefi et al. in their paper proposed an approach for conges- market. The source of electricity with the most number of uncer-
tion management by combining demand response and FACTS tainties is wind energy. Thus, integration of wind farms creates
devices. They achieved this by following two step market clearing enormous number of problems. Zhou et al., in [45], have demon-
procedure. In the first step, ISO clears for market price using social strated the concept of combined cumulants and Gram–Charlier
welfare function and in the second step, network constraints with expansion (CCGCE) and presented a research on FACTS for manag-
respect to congestion management are considered. An optimal ing congestion.
88 A. Pillay et al. / Electrical Power and Energy Systems 70 (2015) 83–90

Another device which is emerging to be of great use is Sen System containing wind farms
Transformer (ST). Kumar and Sekhar, in [46], have explained the
advantages of Sen Transformer and impact of this power flow con- Addition of wind power to the power system is highly problem-
trol device in congestion management. The paper basically dis- atic due to its limited predictability. Further, it causes the trans-
cusses how ST is capable of bidirectional flow of active and mission lines to cross their limits, leading to network congestion.
reactive power. The authors have also compared ST and FACTS A method of combined cumulants and Gram–Charlier expansion
devices, coming to the conclusion that utilization of ST would incur (CCGCE) to calculate probabilistic load flow distribution is being
low cost of power flow. Thus, STs prove to be of great help to the proposed for realistic assessment. Thus, congestion management
ISO for efficient and secure operation of hybrid electricity markets. is undertaken by installment of FACTS and CGCE method. A
Esmaili et al. [47], has proposed a two stage method involving detailed explanation of the method is presented in [45].
Benders decomposition (BD) for congestion management in hybrid In [49], Olamaei has discussed about the problems caused by
electricity market for both pool and bilateral transactions. The integration of wind farms to the grid. The first and foremost prob-
authors’ have come up with an effective convergence criteria for lem is the problem of voltage regulation. The paper describes about
BD and sub-problem type objective function This method is not the methods to overcome this problem by using FACTS devices. The
only an efficient solution but also provides transparency about authors have proven that the STATCOM and SVC can successfully
the congestion cost. control high fault current.
In [48], Kumar and Mittapalli have implemented the method of
congestion management involving rescheduling of generators with
Congestion management adapted in various countries
three bid block structure. The results show that the congestion cost
obtained with the ZIP (constant impedance, current and power)
The cases for congestion management in countries like Euro-
model is lower than those obtained with the P, Q model. The con-
pean countries and the US are exemplified in this section.
gestion cost further reduces with the FACTS devices being imple-
mented. ZIP model will be of great advantage to the ISO in
controlling real time load. Congestion management in US Electricity Network

Hydro-thermal pool market The US electricity market [50] determines two types of conges-
tions: centralized and de-centralized. The centralized approach of
The main objective is to minimize the congestion manage- congestion management is based on locational marginal pricing.
ment cost and start-up cost of hydro and thermal generators. This is exemplified through the PJM interconnection. In PJM all
The proposed ideology is based on mixed binary non-linear pro- the members collectively put their resources and are dispatched
gramming. It is a re-dispatch methodology which considers the centrally. Thus, resulting in the largest dispatch system in the
combined operation of hydro and thermal generating companies North America. The LMPs are calculated for 1750 buses in the
in a pool based energy market on hourly basis. PJM network. They are then put on the website OASIS every five
It is proposed to include the hydro-thermal constraints to the minutes for managing congestion. Thus, PJM allows the partici-
conventional congestion management problem. In addition to their pants to hedge their transactions, by following the shadow prices
bidding strategies, the hydro-thermal generator companies are on the OASIS, through FTR.
assumed to be providing the ISO with information related to their The decentralized approach involves zones and flow gates. It is
unit performance curve and amount of water allocated. assumed that there is very little congestion inside the area or the
Mathematically to minimize, zone. Thus, only the inter zone congestion has to priced and mon-
X u u X u u itored. There has been a lot of criticism against this type of CM
Congestion cost ¼ ðC j DPgj þ C dj DPdgj Þ þ ðC j DPhj technique.
j G j H FERC, NERC and regional councils play a very important role in
þ C dj DPdhj þ SC j Þ  ð;h ; V; dÞ ð11Þ managing congestion. All the organizations have worked as a unit
but the difference lies in their functions. FERC (The Federal Energy
where Regulatory Commission) focuses on affirming fair and open mar-
ket. NERC (the North American Electric Reliability Council) focus
C uj ; C dj incremental and decremental cost bids of
on technical reliability issues. Along with the organizations, ISO
generator at jth bus for rescheduling ($/MW h)
is also responsible for congestion management.
DPuhj ; DPdhj increment and decrement in real power
generation of thermal unit at jth bus
DPugj ; DPdhj increment and decrement in real power Congestion management in European Electricity Network
generation of thermal unit at jth bus
SCj a variable which is equal to start-up cost of hydro The European Commission member countries have envisioned
unit at jth bus if unit is started up for congestion to have integrated the renewable energy sources to their power
management grids by 2020 [51,52]. The addition of these unpredictable sources
;h vector of water discharge of hydro plants with would lead to new power flows and transmission feeds, thereby
individual entry as ;hj . increasing the congestion among the countries. To have learnt
d vector of bus voltage angles with individual entry from the US experience of implementing price method for conges-
as dj or di tion management, Europe tends to apprehend from following the
V vector of bus voltage magnitudes with individual same lines. Europe emphasizes on the implementation of loca-
entry as Vj or Vi tional marginal pricing.
While zonal pricing is actualized in the Nordic countries and in
the Central Western European region between Belgium, France,
The above problem is subject to few non linear constraints, linear Germany, Netherlands and Luxemburg; where the transmission
equality constraints with mixed binary variables and bound on vari- network is split into zones, it is still not widely accepted as an effi-
ables [44]. The problem is solved by using the branch and bound cient solution to avoid congestion in the European power markets.
technique. This can again be called as a shortcoming of the US Electricity Net-
 A. Pillay et al. / Electrical Power and Energy Systems 70 (2015) 83–90 89

work. It is very difficult to define zones and then formulate pricing [10] Lo KL, Yuen YS, Snider LA. Congestion management in deregulated electricity
markets. In: International conference on electric utility deregulation and
methodologies for every zone. It is indeed more tedious job to
restructuring and power technologies, 2000. Proceedings. DRPT; 2000.
define zones in a country like Europe. [11] Savagave NG, Inamdar HP. Price area congestion management in radial system
Europe witnesses bilateral transactions. Owing to the bilateral under de-regulated environment – a case study. Int J Electr Eng Technol (IJEET)
trade, locational marginal pricing appears to be the most appropri- 2013;4(1). ISSN 0976–6545 (Print), ISSN 0976–6553 (Online); January–
February.
ate method of congestion management in the EU power market. [12] Kumar Ashwani, Kumar Jitendra. Comparison of UPFC and SEN transformer for
Also, the availability of financial transmission rights enables ATC enhancement in restructured electricity markets. Electr Power Energy
smooth bilateral transactions by allowing the parties to change Syst 2012;41:96–104.
[13] Singh SN, David AK. Optimal location of FACTS devices for congestion
their bus nodes. Thus, LMP act as a supporting ideology to the management. Electr Power Syst Res 2001;58:71–9.
bilateral transactions. [14] Song Sung-Hwan, Lim Jung-Uk, Moon Seung-Il. Installation and operation of
Due to the problems caused by integration of renewable, the FACTS devices for enhancing steady-state security. Electr Power Syst Res
2004;70:7–15.
German electricity market is re-structured. There are four TSOs [15] Yu Zuwei, Lusan D. Optimal placement of FACTS devices in deregulated
(Transmission System Operators) in the German transmission net- systems considering line losses. Electr Power Syst Res 2004;26:813–9.
work for managing congestion. They have applied Nash equilib- [16] Reddy Keshi Reddy Saidi, Padhy Narayana Prasad, Patel RN. Congestion
management in deregulated power system using FACTS devices. In: IEEE
rium for evolving co-ordination between the four TSOs. Thus, the power India conference; 2006.
co-ordination lowered the overall costs of congestion. [17] Zhang J. Optimal power flow control for congestion management by interline
power flow controller (IPFC). In: Power system technology, 2006. PowerCon;
2006.
Conclusions [18] Besharat Hadi, Taher Seyed Abbas. Congestion management by determining
optimal location of TCSC in deregulated power systems. Electr Power Energy
Syst 2008;30:563–8.
An exhaustive and critical review on the topic of congestion [19] Karami A, Rashidinejad M, Gharaveisi A. Voltage security enhancement and
management has been presented in this paper. The review initially congestion management via statcom & IPFC using artificial intelligence. Iran J
Sci Technol, Trans b, Eng 31(B3):289-301.
focuses on the conventional methods of congestion management. [20] Gitizadeh M, Kalantar M. A new approach for congestion management via
Necessary discussions are made under each topic. It is also estab- optimal location of FACTS devices in deregulated power systems. Nanjing
lished that optimization tools play a very important role in reliev- China DRPT2008 6–9 April 2008.
[21] Rajalakshmi L, Suganyadevi MV, Parameswari S. Congestion management in
ing congestion. The authors have tried to include all possible deregulated power system by locating series FACTS devices. Int J Comput Appl
optimization algorithms being worked upon till now. To ease up (0975–8887) 2011;13(8).
the process many researchers are now exploring new approaches [22] Esmaili Masoud et al. Locating series FACTS devices for multi-objective
congestion management improving voltage and transient stability. Eur J Oper
like evolution strategy and pool dispatch strategy. Most of the Res 2014;236:763–73.
recent advancements in the field are discussed. Also, the tech- [23] Kothari DP, Dhillon JS. Power system optimization, 2nd ed. Prentice-Hall of
niques adopted in countries like Germany, the European countries India; 2011.
[24] Ramirez JM, Giovanni MA. Alleviating congestion of actual power system by
and the US has been portrayed in the review. All the power system
genetic algorithm. In: IEEE power engineering society meeting, June 2004, vol.
engineers should look upon the challenges faced due to congestion. 2. p. 2133–40.
It is envisaged that it will be helpful to the students, teachers, [25] Granelli G et al. Optimal network reconfiguration for congestion management
researchers and utility engineers who are teaching and working by deterministic and genetic algorithms. Electr Power Syst Res 2006;76:549–56.
[26] Surender Reddy S, Sailaja Kumari M, Sydu M. Congestion management in
in this area. In the paper, researchers can extend the work and deregulated power system by optimal choice and allocation of FACTS
come up with quicker and smarter solutions. The authors also rec- controllers using multi-objective genetic algorithm. J Electr Eng Technol
ommend that the solution to the congestion management problem 2009;4(4):467–75.
[27] Saravanan M et al. Application of particle swarm optimization technique for
could also be found out by considering variables like load and optimal location of FACTS devices considering cost of installation and system
water inflow as random variables. loadability. Electr Power Syst Res 2007;77:276–83.
[28] Benabid R et al. Optimal location and setting of SVC and TCSC devices using
non-dominated sorting particle swarm optimization. Electr Power Syst Res
Acknowledgement 2009;79:1668–77.
[29] Boonyaritdachochai Panida et al. Optimal congestion management in an
electricity market using particle swarm optimization with time-varying
The authors sincerely thank INAE (Indian National Academy of acceleration coefficients. Electr Power Syst Res 2010;60:1068–77.
Engineering) for providing funds to carry out the research reported [30] Joshi SK, Pandya KS. Active and reactive power rescheduling for congestion
management using particle swarm optimization. In: Proceedings of
in this paper and VIT University for giving the opportunity to work universities power engineering conference (AUPEC), 21st Australasian; 2011.
at such a prestigious institution. [31] Panigrahi BK, Ravikumar Pandi V. Congestion management using adaptive
bacterial foraging algorithm. Electr Power Syst Res 2009;50:1202–9.
[32] Venkaiah Ch, Vinod Kumar DM. Fuzzy adaptive bacterial foraging congestion
References management using sensitivity based optimal active power re-scheduling of
generators. Appl Soft Comput 2011;11:4921–30.
[33] Doll M, Verstege JF. An evolution strategy based approach for a congestion
[1] Prabhakar Karthikeyan S, Jacob Ragled I, Kothari DP. A review on market power
management system. In: PPT 2001 IEEE Porto power tech conference loth –
in deregulated electricity market. Electr Power Energy Syst 2013;48:139–47.
13’h September, Porto, Portugal.
[2] Saxena Abhishek, Pandey Seema N, Laxmi Srivastava. Congestion management
[34] Padhy NP. Congestion management under deregulated fuzzy environment. In:
in open access: a review. Int J Sci, Eng Technol Res (IJSETR) 2013;2(4).
IEEE international proc. on electric utility deregulation, restructuring and
[3] Yamin HY, Shahidehpour SM. Transmission congestion and voltage profile
power technologies, 5–8 April 2004, vol. 1. p. 133–9.
management coordination in competitive electricity markets. Int J Electr
[35] Yesuratnam G, Thukaram D. Congestion management in open access based on
Power Energy Syst 2003;25(10):849–61.
relative electrical distances using voltage stability criteria. Electr Power Syst
[4] Alomoush Muwaffaq I. Performance indices to measure and compare system
Res 2007;77:1608–18.
utilization and congestion severity of different dispatch scenarios. Electr Power
[36] Farahmand H et al. Hybrid mutation particle swarm optimisation method for
Syst Res 2005;74:223–30.
available transfer capability enhancement. Electr Power Energy Syst
[5] Sood Yog Raj, Padhy NP, Gupta HO. Deregulated model and locational marginal
2012;42:240–9.
pricing. Electr Power Syst Res 2007;77:574–82.
[37] Taher Seyed Abbas, Amooshahi Muhammad Karim. New approach for optimal
[6] Acharya Naresh, Mithulananthan Nadarajah. Influence of TCSC on congestion
UPFC placement using hybrid immune algorithm in electric power systems.
and spot price in electricity market with bilateral contract. Electr Power Syst
Electr Power Energy Syst 2012;43:899–909.
Res 2007;77:1010–8.
[38] Bhattacharyya Biplab, Gupta Vikash Kumar. Fuzzy based evolutionary
[8] Kang CQ et al. Zonal marginal pricing approach based on sequential network
algorithm for reactive power optimization with FACTS devices. Electr Power
partition and congestion contribution identification. Electr Power Energy Syst
Energy Syst 2014;61:39–47.
2013;51:321–8.
[39] Tarafdar Hagh M et al. Impact of SSSC and STATCOM on power system
[9] Murali M et al. Optimal spot pricing in electricity market with inelastic load
predictability. Electr Power Energy Syst 2014;56:159–67.
using constrained bat algorithm. Electr Power Energy Syst 2014;62:897–911.
90 A. Pillay et al. / Electrical Power and Energy Systems 70 (2015) 83–90

[40] Shayesteh E, Parsa Moghaddam M, Taherynejhad S, Sheikh-EL-Eslami MK. [47] Esmaili Masoud et al. Congestion management in hybrid power markets using
Congestion management using demand response programs in power market. modified benders decomposition. Appl Energy 2013;102:1004–12.
In: IEEE PES general meeting 2008, Pittsburg, USA, July 2008. [48] Kumar Ashwani, Mittapalli Ram Kumar. Congestion management with generic
[41] Yousefi A et al. Congestion management using demand response and FACTS load model in hybrid electricity markets with FACTS devices. Electr Power
devices. Electr Power Energy Syst 2012;37:78–85. Energy Syst 2014;57:49–63.
[42] Kumar Ashwani, Sekhar Charan. DSM based congestion management in pool [49] Olamaei J. Advanced control of FACTS devices for improving power quality
electricity markets with FACTS devices. Energy Procedia 2012;14:94–100. regarding to wind farms. Energy Procedia 2012;14:298–303.
[43] Sood Yog Raj, Singh Randhir. Optimal model of congestion management in [50] Kirby BJ, Van Dyke JW. Congestion management requirements, methods and
deregulated environment of power sector with promotion of renewable performance indices. Oak Ridge National Laboratory, C. Martinez, A. Rodriguez,
energy sources. Renew Energy 2010;35:1828–36. Electric Power Group, Southern California Edison, June 2002.
[44] Singh Kanwardeep, Padhy NP, Sharma J. Congestion management considering [51] Neuhoff Karsten, Hobbs Benjamin F, Newbery David. Congestion management
hydro-thermal combined operation in a pool based electricity market. Electr in European power.
Power Energy Syst 2011;33:1513–9. [52] Kunz F. Improving congestion management – how to facilitate the integration
[45] Zhou Jianhua, Yuan Yue, Yu Jiayan. Research on transmission congestion of of renewable generation in Germany. Energy J forthcoming [2012].
power system containing wind farms based on genetic algorithm. In: 4th
International conference on electric utility deregulation and restructuring and
power technologies (DRPT); 2011.
Further Reading
[46] Kumar A, Sekhar Charan. Comparison of sen transformer and UPFC for
congestion management in hybrid electricity markets. Electr Power Energy [7] Jokić A, Lazar M, van den Bosch PPJ. Real-time control of power systems using
Syst 2013;47:295–304. nodal prices. Electr Power Energy Syst 2009;31:522–30.