Engineering Science and Technology, an International Journal 19 (2016) 1254–1265

Contents lists available at ScienceDirect

Engineering Science and Technology,

an International Journal Press: Karabuk University, Press Unit
ISSN (Printed) : 1302-0056
ISSN (Online) : 2215-0986
ISSN (E-Mail) : 1308-2043

j o u r n a l h o m e p a g e : h t t p : / / w w w. e l s e v i e r. c o m / l o c a t e / j e s t c h
H O S T E D BY

Available online at [Link]

ScienceDirect

Full Length Article

Firefly algorithm for congestion management in

deregulated environment
Sumit Verma, V. Mukherjee *
Department of Electrical Engineering, Indian School of Mines, Dhanbad, Jharkhand, India

A R T I C L E I N F O A B S T R A C T

Article history: In competitive electricity market, congestion is a serious economic and reliability concern. Congestion
Received 31 August 2015 is a common problem that an independent system operator faces in open access electricity market. This
Accepted 1 February 2016 paper presents a reliable and efficient meta-heuristic based approach to solve congestion problem. The
Available online 6 A pril 2016
proposed approach of the present work employs firefly algorithm (FFA) for alleviation of transmission
network congestion in a pool based electricity market via active power rescheduling of generators. FFA
Keywords:
is a new meta-heuristic approach based on flashing patterns and behavior of fireflies. Various impor-
Congestion management
tant security constraints such as load bus voltage and line loading have been taken into account while
Deregulation
Firefly algorithm dealing with congestion problem. The proposed methodology may help in removing the congestion of
Independent system operator line with minimum rescheduling cost. The numerical results of modified IEEE 30- and 57-bus test power
Optimal power flow systems are illustrated.
Price bids © 2016 Karabuk University. Publishing services by Elsevier B.V. This is an open access article under the CC
BY-NC-ND license ([Link] licenses/by-nc-nd/4.0/).

1. Introduction unexpected outage of generation, restriction on the construction of

new lines, unscheduled power flow in lines, tripping of transmis-
1.1. General sion lines or failures of other equipment are some of the potential
causes for congestion. In a deregulated environment, congestion is
Before restructuring of the power system, the power grids were, a primary challenge to an ISO who is responsible in managing con-
usually, used to be operated by vertically integrated utilities. These gestion in the transmission line and ensuring security as congestion
utilities had common control over both generation and transmis- may cause serious menace to stability of the power networks and
sion facilities. With unbundling: generating, transmitting and may also result in market inefficiency and electricity price hike [3].
distributing companies are working as independent entities and, Rescheduling of generator outputs, supplying reactive power support
thus, it has become a challenge for independent system operators or curtailment of transactions are, physically, the usual methods
(ISO) to operate the system in synchronism [1]. In deregulated adopted for congestion management (CM).
market, all the market players are free to interact with each other.
Buying and selling of electricity is done by the participants in such 1.2. Literature review
a way that only aims to maximize the profit, causing transmission
networks to operate beyond their operational limits. The literature survey reveals that various techniques have been
Congestion is the difference in the megawatts of the power sched- used to address the serious issues related to CM. CM in open access
uled to flow on a transmission line and the actual transfer that is electricity market has been discussed in References [4–6]. A de-
allowed on the line without violating any constraints. Congestion tailed analysis of different CM techniques, used in different electricity
occurs whenever one or more constraints are violated under which markets throughout the world, may be found in Reference [4]. A
the system operates in the normal operating condition or in any of minimum distance re-dispatch has been proposed in Reference [5]
the specified contingencies. The constraints can be either physical ignoring the economic value of the transaction adjustment. In Ref-
limits like thermal or voltage limits or specified limits to ensure erence [6], the congestion is managed by using the marginal cost
system security and reliability [2]. Increase in power demand, signals for the generators. Thukaram and Parthasarathy [7] have pro-
posed an expert system based approach for the alleviation of network
overloads using phase shifting transformers and generation re-
* Corresponding author. Tel.: +91 0326 2235644; fax: +91 0326 2296563.
scheduling. A physical flow based CM allocation mechanism for
E-mail address: vivek_agamani@[Link] (V. Mukherjee). multiple transaction networks has been discussed by Shu and Gross
Peer review under responsibility of Karabuk University. in Reference [8], which enables the independent grid operator to

[Link]
2215-0986/Ó 2016 Karabuk University. Publishing services by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license ([Link]
 S. Verma, V. Mukherjee/Engineering Science and Technology, an International Journal 19 (2016) 1254−1265 1255

acquire the congestion relief services for each transaction to remove position for itself in consideration with its current position as well
its congestion contribution at the least cost. Kumar et al. [9] have as the position of other fireflies. Hence, it moves from the local minima
proposed a zonal model based on ac load flow in which the calcu- and finds the global minima in less number of iterations. Apart from
lation of sensitivity values for all the buses in the system is required the self improving process within the current space, the FFA also
and, therefore, a large amount of computational effort is required includes the improvement among its own space from the previous
to be spent. Dutta and Singh [10] have demonstrated a technique stages. Robustness and high convergence rate make this algorithm
for the optimum selection of generators based on generator sen- most suitable to use for such kind of optimization problems [25].
sitivities to the power flow using particle swarm optimization (PSO) FFA is proposed in this paper to solve CM problem. The main mo-
algorithm with an aim to solve CM problem. A method for selec- tivation of the present work is to aid ISO to remove congestion of
tion of participating generators based on sensitivity to current flow lines in an optimal manner. In this paper FFA is applied on modi-
on congested line as well as the generation bids has been pre- fied IEEE 30-bus and 57-bus test power systems to solve congestion
sented by Talukdar et al. in Reference [11]. Conejo et al. [12] has problem under various considered contingencies.
addressed the CM problem by disregarding the off line transmis-
sion capacity limits related to stability which results in economical 1.4. Contribution
and secure operating conditions. Kumar and Chaturvedi [13] have
presented integration of fuzzy systems with genetic algorithm and The main contributions of this work are to:
PSO to solve the optimal power flow (OPF) problem for optimal
setting of control parameters. An approach for CM with flexible ac (a) project FFA as an effective optimizing tool to minimize the
transmission systems in deregulated electricity market with voltage rescheduling cost under different contingencies for the two
stability constraint, taking loadability parameter into consider- IEEE standard power systems: IEEE 30-bus system and IEEE
ation along with the line security limits using rescheduling of 57-bus system,
generators, is proposed by Kumar and Sekhar in Reference [14]. In (b) effectively remove the overload in the lines caused by various
Reference [15], an efficient PSO method has been used for real power considered contingencies with smallest shift in generation
rescheduling of generator for transmission CM in deregulated en- schedule,
vironment. The proper placement and sizing of proper flexible ac (c) minimize the total amount of rescheduling and losses for
transmission systems (FACTs) devices based on PSO in deregu- various considered cases and
lated environment has been studied in References [16,17]. The (d) demonstrate the effectiveness of the proposed FFA over the
utilization of distributed generation units for CM by improving others for this specific application.
the voltage profile using PSO has been studied in Reference [18].
The application of PSO to maximize total system social welfare in
1.5. Paper layout
a double-sided auction market by the proper allocation of FACTs
devices is proposed in Reference [19]. The application of fuzzy-
The remaining portion of this paper is structured as follows. Section
based genetic algorithm (GA) to maximize total system social welfare
2 provides the mathematical formulation of the CM problem. Section
in a double-sided auction market by the best placement and sizing
3 explains the FFA. Section 4 deals with FFA for CM problem. Simulation
of FACTs devices has been proposed in References [20,21].
results are presented and discussed in Section 5. Finally, conclusions
A methodology based on improved harmony search is pro-
are drawn and scope of future work is presented in Section 6.
posed in Reference [22] to solve transmission expansion planning
problem with adequacy-security considerations in deregulated power
system. Simulated annealing (SA) has been applied on unit com- 2. Mathematical problem formulation
mitment problem by Zhuang and Galiana [23]. Jang et al. [24] have
discussed a computationally simple random search method (RSM) The main objective of the CM is to minimize the congestion cost
that can be utilized to solve various optimization problems. while satisfying the network constraints. In the present work, the
Firefly algorithm (FFA) is a meta-heuristic approach inspired from CM problem is solved by rescheduling (increasing or decreasing)
the flashing behavior of fireflies [25] and its prevalence is increas- the active power output of generators. But change in active power
ing rapidly in almost all branches of science and technology for the output is associated with cost which, in turn, depends upon the price
purpose of optimization. FFA has been used to solve non-linear bids submitted by generating companies (GENCOs). The problem
design problem in Reference [25]. FFA has been utilized in Refer- may be stated as in Eq. (1) [15]:
ence [26] to optimize the control variables for simultaneous Minimize
optimization of real power loss and voltage stability limit of the
transmission system. The modified FFA is used in Reference [27] to
Cc = ∑ (C k ΔPGj+ + D k ΔPGj− ) $ h (1)
j εN g
design a Smith predictor controller for integration and unstable delay
processes. FFA [28] has been proposed in the current work for the where C c , C k , D k , ΔPGj+ and ΔPGj− represent the total cost incurred for
rescheduling purpose to alleviate congestion in the power networks. changing active power output ($/h), incremental price bids sub-
mitted by GENCOs ($/MWh), decremental price bids submitted by
1.3. Motivation GENCOs ($/MWh), active power increment of generator (MW) and
active power decrement of generator (MW), respectively.
Literature survey reveals that numerous techniques have been The present optimization problem is subjected to the equality
implemented by researchers in the past to solve the CM problem. and inequality constraints as stated in the next two sub-sections.
A major force behind the present work is to design a novel tech-
nique to solve the CM problem. Most traditional optimization 2.1. Equality constraints
techniques do not function admirably for the issues with nonlin-
earity and multimodality. Current pattern is to utilize nature- The equality constraints of CM represent the power flow
propelled metaheuristic algorithms to handle such difficulties, and equations as stated in Eqs. (2) to (5) [29]:
it has been demonstrated that metaheuristics are exceptionally pro-
ductive. FFA is inspired from natural behavior of fireflies. Unlike other PGk − PDk = ∑ V j V k Y kj cos (δ k − δ j − θkj ); j = 1, 2, … , N b (2)
algorithms, a firefly works individually and tries to find the best j
1256 S. Verma, V. Mukherjee/Engineering Science and Technology, an International Journal 19 (2016) 1254−1265

Q Gk − Q Dk = ∑ V j V k Y kj sin (δ k − δ j − θkj ); j = 1, 2, … , N b (3)

3.2. Light intensity and attractiveness
j
Two important things that should be considered in FFA are the
PGk = PGk
C
+ ΔPGk+ − ΔPGk− ; k = 1, 2, … , N g (4) variation of the light intensity and formulation of attractiveness. The
attractiveness of a particular firefly is determined by its bright-
PDj = PDjC; j = 1, 2, … , N d (5) ness which, in turn, is associated with the objective function value.
The attractiveness (termed as β) is relative, as it is seen and judged
where PGk and Q Gk are the generated active and reactive power at by the other fireflies and it increases as the distance between the
bus k, respectively; PDk and Q Dk are the active and reactive load two fireflies decreases. Also, light intensity decreases with the in-
power at bus k, respectively; V j and V k are voltages at bus j and k, crease in distance from the source and light is also absorbed in the
respectively; δ j and δ k are bus voltage angles of bus j and k, re- medium of its propagation. So, a degree of attractiveness is to be
spectively; θkj is admittance angle of line connected between k and set in order to vary β. The light intensity, (I (r )) , varies monotoni-
j; Nb, Ng, and Nd are number of buses, generators and loads, respec- cally and exponentially with the distance (r) between the two fireflies
tively; PGkC
and PDjC are the active power produced by generator k and it is expressed as in Eq. (11):
and active power consumed by load bus j, respectively, as ob-
I (r ) = I 0 exp ( −γ r ) (11)
tained by the market clearing value.
It is to be noted here that Eqs. (2) and (3) show active and re-
where I 0 and γ are the original light intensity and light absorption
active power balance at each node while Eqs. (4) and (5) represent
co-efficient, respectively.
final power as a function of market clearing price.
As a firefly’s attractiveness is proportional to the light intensity seen
by other fireflies, the attractiveness β can be defined as in
Eq. (12):
2.2. Inequality constraints

The inequality constraints represent the operating and physi-

cal limit of all the transmission lines, transformers and generators
and are stated in Eqs. (6) to (10) [29]: START
min
PGk ≤ PGk ≤ PGk
max
, ∀k ∈Ng (6)
Initialize Parameters
min
Q Gk ≤ Q Gk ≤ Q Gk
max
, ∀k ∈Ng (7)

(PGk − PGkmin ) = ΔPGkmin ≤ ΔPGk ≤ ΔPGkmax = (PGkmax − PGk ) (8) Initialize firefly
population
V nmin ≤ V n ≤ V nmax , ∀n ∈N l (9)
Call objective
Pij ≤ Pijmax (10)
function
where the superscripts min and max represent the minimum and
maximum values of the respected variables and Nl represents the Make a copy of
number of lines. firefly population

3. FFA Rank fireflies

according to
FFA is inspired by the flashing characteristics of fireflies to attract
attractiveness
their mating partners and is developed by Yang [25]. A brief over-
view of this algorithm is provided in the next two sub-sections.
Sort the fireflies

3.1. FFA: features

The pattern of flashes produced by bioluminescence is unique

for a particular species of fireflies. FFA, based on the nature of fire- Stopping No
flies, follows three idealized rules as mentioned below [28]. criteria
reached
(a) Each and every firefly is unisex and, hence, one firefly is at-
tracted to the other regardless of its sex.
Yes
(b) Attraction is proportional to the brightness of the fireflies. For
any two fireflies, the one having less brightness moves toward
Obtain global
the other having more brightness. The intensity of flashes is
minimum
inversely proportional to the distance between the two fire-
flies. So, as the distance increases, brightness and, hence,
attraction between the two fireflies, decreases. The brightest
END
firefly moves randomly in the population.
(c) The brightness of a firefly is determined by the objective
function value. Fig. 1. Flowchart of the FFA.
 S. Verma, V. Mukherjee/Engineering Science and Technology, an International Journal 19 (2016) 1254−1265 1257

β (r ) = β0 exp ( − γr 2 ) (12) The inequality constraints are converted to the penalty functions
and these penalty functions are added to the objective function.
where β0 is the attractiveness at r = 0. In this paper, the equality constraints are handled effectively during
The distance between any two fireflies i and j, located at positions Newton–Raphson power flow [30] and the active power inequali-
x i and x j , respectively, is the Cartesian distance given by Eq. (13): ty constraints are handled during the execution of iteration. Reactive
power inequality constraints are handled during the load flow
d solution. Other inequality constraints such as load bus voltage
∑ (x − x j ,k )
2
rij = x i − x j = i ,k (13)
and line power flow are considered as quadratic penalty func-
k =1
tions. The fitness function of CM problem may be described as in
where x i ,k and x j ,k are the components of the spatial co-ordinates Eq. (15) [15]:
x i and x j of i th and j th firefly, respectively and d is the dimen- ovl VB
Minimize Ff = C c + PF1 × ∑ (Pij − Pijmax ) + PF2 × ∑ ( ΔV j )
2 2
sion of the problem.
The movement of i th firefly, attracted to any brighter firefly j is i =1 j =1

+ PF3 × ( ΔPG )
2
given by Eq. (14). (15)

x i = x i + β0 exp ( −γ r 2
ij ) × (x j − x i ) + α × (rand − 0.5) (14) where

In Eq. (14), the first term represents the current position of i th ⎧(V jmin − V j ); if V j ≤ V jmin
ΔV j = ⎨ (16)
⎩(V j − V j ); if V j ≥ V j
max max
firefly, the second term represents the attractiveness to other brighter
fireflies and the third term represents a random walk associated with
a randomization parameter α. rand is a uniformly distributed random ⎧(PGmin − PG ); if PG ≤ PGmin
ΔPG = ⎨ (17)
number generated in the range [0,1] and the range of α is, usually, ⎩(PG − PG ); if PG ≥ PGmax
max

taken as [0,1]. The parameter γ characterizes the variation of at-

tractiveness and its value is, significantly, important as it determines Here, Ff is fitness function which is required to be minimized
the behavior and convergence of FFA and it has the range [0, ∞ ]. in order to get minimum rescheduling cost; ovl and VB represent
The operation of FFA may be summarized to the pseudo-code set of the overloaded lines and voltage violated load buses, respec-
[28], presented in Algorithm 1. The flowchart of the FFA is tively, and PFi (i = 1, 2, 3) represent penalty factors which has been
presented in Fig. 1. taken as 10,000 throughout the simulation process [15]. Moreover,

4. FFA for CM problem the second, third and fourth terms are added to the fitness func-
tion, keeping in mind the possibilities of violations.
In this work, each population has N number of design vari-
ables where N is the number of generators taking part in the CM
problem. Usually, the objective function is considered as the fitness 4.1. Computational procedure of FFA for CM
function. In this work, penalty approach [15] is adopted, which
penalizes the constraints and builds a single objective function Based on the above discussions, the procedure in applying the pro-
which, in turn, is minimized by using an optimization algorithm. posed FFA algorithm for the solution of CM problem is given below.
1258 S. Verma, V. Mukherjee/Engineering Science and Technology, an International Journal 19 (2016) 1254−1265

Step 1 Read the bus data, the line data, the price bids and the gen- load buses and six generator buses. The total active and reactive
erator information. power of load for this test system is 283.4 MW and 126.2 MVAR,
Step 2 Create contingency by either line outage or increase in load. respectively. Generation and load values (provided in the Appen-
Step 3 Run load flow while satisfying equality constraints stated dix section), are taken as the initial market clearing values for PG
in Eqs. (2) to (5). Hence, find the excess power flow and bus and PD, respectively. Contingencies like unexpected line outage and
voltage violation, if any. increase in system load are considered for the simulation purpose.
Step 4 Initial population of fireflies is generated using Eq. (6), which Two different cases of this example viz. case 1A and case 1B (Table 1)
is the amount of rescheduling required by the generators are considered for this example.
to manage congestion (randomly within the limits).
Step 5 For each generated population of fireflies, load flow is per- 5.1.1. Case 1A
formed and, hence, the fitness function is evaluated by using In this case, congestion is created by considering outage of line
Eq. (15) and the best solution is identified. During the ex- number-1 connected between bus-1 and bus-2. Due to outage of
ecution of iteration, Eqs. (9), (10), (16), and (17) are checked. line 1, congestion occurs in lines number-2 and -4, connected between
Step 6 The positions of all the fireflies are modified with refer- buses 1–7 and 7–8, respectively. OPF [29] results reveal that power
ence to their attractiveness using Eqs. (12) to (14). flows in those lines become 147.463 MW and 136.292 MW, respec-
Step 7 The fitness function, defined in Eq. (15), is evaluated with tively, against the line flow limit of 130 MW for both lines. Details
modified fireflies. Any two fireflies are randomly selected of the congested lines are presented in Table 2. Hence, the conges-
and their fitness values are compared. The firefly with better tion has to be alleviated by the optimal rescheduling of active power
fitness value is accepted while the other is rejected. generation of generators. The results, obtained by employing the
Step 8 If maximum number of iteration is reached then the program proposed FFA for the solution of CM problem for case 1A of Example
is stopped; otherwise, it goes back to Step 6. 1, are tabulated in Table 3. For comparison purpose, the results ob-
tained from RSM, SA and PSO techniques reported in Reference [15]
5. Simulation results and discussion are also included in the same table. From Table 3 it may be con-
cluded that the results obtained by proposed FFA is the best, providing
In the present work, FFA for CM is implemented using MATLAB minimum rescheduling cost compared to other methods reported
(version 7.6.0) software on an Intel Core i3 Processor based system in the literature, without overloading the other lines. The proposed
with 2.4 GHz clock speed and supported by 4 GB of RAM. To verify FFA gives the best solution as 511.8737 $/h (Table 3). The total system
the effectiveness of the proposed FFA in solving CM problem, simu- loss before CM was 16.023 MW while the same is decreased to
lations are carried out on modified IEEE 30-bus and 57-bus test 13.10 MW after CM. A comparative pictorial representation of active
systems. The bus data and line data may be found in the Appen- power rescheduling and congestion cost offered by different methods
dix section (Tables A1 and A2 for modified IEEE 30-bus test system like PSO [15], RSM [15] and SA [15] are shown in Figs. 2 and 3, in
and Tables A3 and A4 for modified IEEE 57-bus test system). The order. The convergence profile of fitness function for this test case,
price bids offered by the GENCOs to ISO for modified IEEE 30- and as yielded by the proposed FFA, is shown in Fig. 4.
IEEE 57-bus test systems are given in Tables A5 and A6, respective-
ly. Generation rescheduling cost is calculated for the simulated cases
5.1.2. Case 1B
and is compared with results reported in Reference [15].
For this case, congestion is created by considering outage of line
Details of simulated cases carried out on the two test systems
number-2 connected between bus-1 and bus-7 accompanied by
are given in Table 1. Congestion is created in lines for the simula-
tion purpose by overloading the lines. In this paper, line overloads
are created either by reducing the capacity of lines as to the com- Table 2
pared standard limits or by considering generator or line outage. Details of congested lines for modified IEEE 30-bus test system corresponding to
Case 1A.
The proposed FFA has been executed for 100 independent trial
runs, out of which the best solution set is presented here. The values Test case Congested lines Actual flow (MW) Line limit (MW)
of α and γ are taken in the range of 0 to 1, while the value of β0 is 1A 1–7 147.463 130
kept constant at 10. It has been found that population of 40 fire- 7–8 136.292 130
flies is sufficient in solving the CM problem of the present work.
The maximum number of iteration is set to 150 for all the test cases.
Table 3
The major observations of the present work are documented below.
Comparison of results obtained from different algorithms for modified IEEE 30-bus
Results of interest are bold faced in the respective tables. test system corresponding to Case 1A.

Parameters Techniques
5.1. Example 1: modified IEEE 30-bus test system
FFA [Proposed] PSO [15] RSM [15] SA [15]

The modified IEEE 30-bus test system is taken for consider- Total congestion 511.8737 538.95 716.25 719.861
cost ($/h)
ation as Example 1. It has forty-one transmission lines, twenty-four
Power flow (MW) on 129.812 129.97 129.78 129.51
previously congested
Table 1 line 1–7
Simulated cases. Power flow (MW) on 120.617 120.78 120.60 120.35
previously congested
Test system Test case Contingency considered
line 7–8
Modified IEEE 30-bus 1A Outage of line 1–2 ΔPG 1 (MW) −8.7783 −8.6123 −8.8086 −9.0763
1B Outage of line 1–7 with increase in load at ΔPG 2 (MW) +15.0008 +10.4059 +2.6473 +3.1332
all buses by 50% ΔPG 3 (MW) +0.1068 +3.0344 +2.9537 +3.2345
Modified IEEE 57-bus 2A Reduction in capacity of lines 5–6 and ΔPG 4 (MW) +0.0653 +0.0170 +3.0632 +2.9681
6–12 from 200 MW to 175 MW and from ΔPG 5 (MW) +0.1734 +0.8547 +2.9136 +2.9540
50 MW to 35 MW, respectively ΔPG 6 (MW) −0.6180 −0.0122 +2.9522 +2.4437
2B Reduction in capacity of line 2–3 from Total generation 24.7425 22.936 23.339 23.809
85 MW to 20 MW. rescheduled (MW)
 S. Verma, V. Mukherjee/Engineering Science and Technology, an International Journal 19 (2016) 1254−1265 1259

Table 4
Details of congested lines for modified IEEE 30-bus test system corresponding to
Case 1B.

Test case Congested lines Actual flow (MW) Line limit (MW)

1B 1–2 310.917 130

2–8 97.353 65
2–9 103.524 65

congestion in the transmission lines is found to be 251.794 MW. To

alleviate this overloading, the optimum rescheduling of genera-
tors are carried out by using FFA and the obtained results are
presented in Table 5. The results yielded by proposed FFA are com-
pared with the results reported in Reference [15] while adopting
PSO, RSM and SA. The cost for CM is visibly less for the proposed
FFA method than for other methods reported in Reference [15]. Also,
Fig. 2. Comparative active power rescheduling of generators for modified IEEE
the total system loss is decreased to 16.264 MW after CM, which
30-bus test system corresponding to Case 1A. was initially 37.8 MW during congestion. The up/down adjust-
ment of active power generated by the generators, as offered by the
proposed FFA method, is shown in Fig. 5. The comparative total cost
incurred while removing congestion for this case is plotted in Fig. 6.
increase of load at all the buses by 50%. This considered contin-
The convergence of fitness function, as offered by the proposed FFA
gency causes overloading of lines connected between buses 1–2,
with the number of iterations, for this test case is plotted in Fig. 7.
2–8 and 2–9 with power flow of 310.917 MW, 97.353 MW and
103.524 MW, respectively, which are beyond the limits of their
maximum power flow limits (130 MW for line 1–2 and 65 MW each
for both the lines 2–8 and 2–9). Table 4 shows the list of over- Table 5
Comparison of results obtained from different algorithms for modified IEEE 30-bus
loaded lines for this case. In this case, total power violation due to test system corresponding to Case 1B.

Parameters Techniques

FFA [Proposed] PSO [15] RSM [15] SA [15]

Total congestion cost ($/h) 5304.40 5335.5 5988.05 6068.7

Power flow (MW) on 130 129.7 129.91 129.78
previously congested
line 1–2
Power flow (MW) on 62.713 61.1 52.36 51.47
previously congested
line 2–8
Power flow (MW) on 64.979 64.67 55.43 54.04
previously congested
line 2–9
ΔPG 1 (MW) −8.5798 NR NR NR
ΔPG 2 (MW) +75.9954 NR NR NR
ΔPG 3 (MW) +0.0575 NR NR NR
ΔPG 4 (MW) +42.9944 NR NR NR
ΔPG 5 (MW) +23.8325 NR NR NR
ΔPG 6 (MW) +16.5144 NR NR NR
Total generation 167.974 168.03 164.55 164.53
rescheduled (MW)

NR means not reported in the referred literature.

Fig. 3. Congestion cost offered by different algorithms for modified IEEE 30-bus test
system corresponding to Case 1A.

Fig. 4. FFA based convergence profile of fitness function value for modified IEEE 30- Fig. 5. FFA based active power rescheduling of generators for modified IEEE 30-
bus test system corresponding to Case 1A. bus test system corresponding to Case 1B.
1260 S. Verma, V. Mukherjee/Engineering Science and Technology, an International Journal 19 (2016) 1254−1265

Table 7
Comparison of results obtained from different algorithms for modified IEEE 57-bus
test system corresponding to Case 2A.

Parameters Techniques

FFA [Proposed] PSO [15] RSM [15] SA [15]

Total congestion 6050.1 6951.9 7967.1 7114.3

cost ($/h)
Power flow (MW) on 174.318 141 148.4 146.60
previously congested
line 5–6
Power flow (MW) on 34.993 34.67 35 34.84
previously congested
line 6–12
ΔPG 1 (MW) +5.6351 +23.135 +59.268 +74.499
ΔPG 2 (MW) +2.5230 +12.447 0 0
ΔPG 3 (MW) +0.5098 +7.493 +37.452 −1.515
ΔPG 4 (MW) +0.107 −5.385 −47.391 +9.952
ΔPG 5 (MW) −39.1514 −81.216 −52.125 −85.920
Fig. 6. Congestion cost offered by different algorithms for modified IEEE 30-bus test ΔPG 6 (MW) −35.1122 0 0 0
system corresponding to Case 1B. ΔPG 7 (MW) +62.1938 +39.03 0 0
Total generation 145.227 168.70 196.23 171.87
rescheduled (MW)
5.2. Example 2: IEEE 57-bus test system

Modified IEEE 57-bus test system consists of seven generator buses,

fifty load buses and eighty transmission lines and is chosen as Example
2. The total active and reactive power loads are 1250.8 MW and
336 MVAR, respectively. The two different simulation cases consid-
ered for this example are case 2A and 2B, as presented in Table 1.

5.2.1. Case 2A
In this case, the line limits are taken as 175 MW for the line 5–6
and 35 MW for the line 6–12, instead of their original power flow
limit of 200 MW and 50 MW, respectively, to create congestion
(Table 1). The details of congested lines are provided in Table 6. Due
to this congestion, the lines 5–6 and 6–12 get overloaded and total
power violation becomes 35.322 MW. Optimum generator resched-
uling is performed using the proposed FFA to completely alleviate
this overloading of 35.322 MW. The details of the results obtained
are listed in Table 7 and these results are compared with those
yielded by PSO [15], RSM [15] and SA [15]. A comparison of the Fig. 8. Comparative active power rescheduling of generators for modified IEEE 57-
bus test system corresponding to Case 2A.
amount of active power rescheduling required for CM, as offered
by PSO, RSM and SA is presented in Fig. 8. Fig. 9 exhibits the
comparative congestion cost offered by SA, RSM, PSO and the pro-
posed FFA method. It may be noted from Table 8 and Fig. 9 that the
total cost of CM, obtained from proposed FFA method, is only 6050.1
$/h, which is the lowest among the costs obtained from the other
three methods, SA, RSM and PSO. The total system loss before CM

Fig. 7. FFA based convergence profile of fitness function value for modified IEEE 30-
bus test system corresponding to Case 1B.

Table 6
Details of congested lines for modified IEEE 57-bus test system corresponding to
Case 2A.

Test case Congested lines Actual flow (MW) Line limit (MW)

2A 5–6 195.971 175

Fig. 9. Congestion cost offered by different algorithms for modified IEEE 57-bus test
6–12 49.351 35
system corresponding to Case 2A.
 S. Verma, V. Mukherjee/Engineering Science and Technology, an International Journal 19 (2016) 1254−1265 1261

Table 8
Details of congested lines for modified IEEE 57-bus test system corresponding to
Case 2B.

Test case Congested lines Actual flow (MW) Line limit (MW)

2B 2–3 37.048 20

Fig. 11. FFA based active power rescheduling of generators for modified IEEE 57-
bus test system corresponding to Case 2B.

Fig. 10. FFA based convergence profile of fitness function value for modified IEEE
57-bus test system corresponding to Case 2A.

was 21.458 MW and it is decreased to 17.64 MW after CM while

adopting proposed FFA (Table 7). Fig. 10 portrays the convergence
profile of fitness function, as obtained by the proposed FFA.

5.2.2. Case 2B
In this case, line 2–3 is made to be overloaded by reducing its ca-
pacity to 20 MW from the original value of 85 MW. Under base load
condition, the power flow in this line is 37.048 MW and, hence, it gets
overloaded and the total power violation becomes 17.048 MW (Table 8).
To relieve this amount of power overloading, active power resched-
uling of the generators are carried out by using the proposed FFA
method. The details of the results obtained while adopting the pro-
posed FFA and the other methods reported in the literature like PSO
[15], RSM [15] and SA [15] are listed in Table 9. From Table 9, it is
Fig. 12. Comparative congestion cost for modified IEEE 57-bus test system corre-
clear that the cost incurred for CM is only 2618.1 $/h for the pro- sponding to Case 2B.
posed FFA method, which is the lowest one among all the costs,
obtained from different reported methods. The total system loss is
decreased to 21.062 MW after CM, which was 21.458 MW initially. like SA [15], RSM [15] and PSO [15] and the proposed FFA are dis-
The optimal rescheduling of active power generation required for this played in Fig. 12. The convergence of the fitness function value for
case is shown in Fig. 11. It is evident from Fig. 11 that incremental this test case, based on the proposed FFA method, is shown in Fig. 13.
change in active power generation is required for generators 1, 3 and
4, and for all the remaining generators, a decremental change is re- 6. Conclusion and scope of future work
quired. Comparative congestion cost offered by different algorithms
This paper demonstrates a novel optimization technique for so-
Table 9 lution of the CM problem in open access electricity market. FFA is,
Comparison of results obtained from different algorithms for modified IEEE 57-bus
test system corresponding to Case 2B.

Parameters Techniques

FFA [Proposed] PSO [15] RSM [15] SA [15]

Total congestion 2618.1 3117.6 3717.9 4072.9

cost ($/h)
Power flow (MW) on 19.79 19.88 20 18.43
previously congested
line 2–3
ΔPG 1 (MW) +0.3704 NR NR NR
ΔPG 2 (MW) −27.5084 NR NR NR
ΔPG 3 (MW) +31.6294 NR NR NR
ΔPG 4 (MW) +0.3308 NR NR NR
ΔPG 5 (MW) −2.2549 NR NR NR
ΔPG 6 (MW) −1.9354 NR NR NR
ΔPG 7 (MW) −0.5101 NR NR NR
Total generation 64.5393 76.314 89.320 97.887
rescheduled (MW)
Fig. 13. FFA based convergence profile of fitness function value for modified IEEE
NR means not reported in the referred literature. 57-bus test system corresponding to Case 2B.
1262 S. Verma, V. Mukherjee/Engineering Science and Technology, an International Journal 19 (2016) 1254−1265

successfully, implemented to minimize the rescheduling cost for al- previous stages. Thus, it may be concluded that FFA is a powerful
leviating congestion completely. Contingencies like line outage and and strong approach to solve optimization problems, providing most
sudden load variation are considered in this work. The proposed economical, reliable and secure operating conditions. Use of sen-
method is implemented on modified IEEE 30- and IEEE 57-bus sitivity analysis for selection of participating generators along with
systems and the results are compared with random search method, rescheduling may be the direction of future research work. FFA may
simulated annealing and PSO. It is observed that the proposed FFA be recommended as an effective optimization tool for some other
effectively relieves congestion, and rescheduling cost obtained is much power engineering optimization applications.
lower than the costs reported by the other approaches. Moreover,
total amount of rescheduling and losses are also found to be lower.
From all the considered simulated cases, it may be observed that Appendix
FFA is a potential tool to solve a non-linear, multimodal problem.
Compared to other optimization algorithms like PSO, SA and RSM, Bus data and line data for modified IEEE 30-bus system are pre-
FFA has added advantage of random reduction, lesser time to produce sented in Tables A1 and A2, respectively, while those for modified
optimum value and automatic subdivision among the fireflies. Apart IEEE 57-bus system are given in Tables A3 and A4, respectively. Price
from the self improving process within the current space, the FFA bids submitted by GENCOs for modified IEEE 30- and 57-bus systems
also includes the improvement among its own space from the are given by Tables A5 and A6, respectively.

Table A1
Bus data for modified IEEE 30-bus test system.

Bus no. Bus code Voltage (V) Angle (°) Generation Load Generation

MW MVAR MW MVAR Qmin Qmax

1 1 1.06 0.0 138.59 0.0 0.0 0.0 −30 100

2 2 1.043 0.0 57.56 50.0 21.7 12.7 −30 100
3 2 1.01 0.0 24.56 37.0 94.2 19.0 −30 100
4 2 1.01 0.0 35.0 37.3 30.0 30.0 −30 100
5 2 1.082 0.0 17.91 16.2 0.0 0.0 −30 100
6 2 1.071 0.0 16.93 10.6 0.0 0.0 −30 100
7 0 1.0 0.0 0.0 0.0 2.4 1.2 0.0 0.0
8 0 1.01 0.0 0.0 0.0 7.6 1.6 0.0 0.0
9 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
10 0 1.0 0.0 0.0 0.0 22.8 10.9 0.0 0.0
11 0 1.802 0.0 0.0 0.0 0.0 0.0 0.0 0.0
12 0 1.0 0.0 0.0 0.0 5.8 2.0 0.0 0.0
13 0 1.071 0.0 0.0 0.0 11.2 7.5 0.0 0.0
14 0 1.0 0.0 0.0 0.0 6.2 6.2 0.0 0.0
15 0 1.0 0.0 0.0 0.0 8.2 2.5 0.0 0.0
16 0 1.0 0.0 0.0 0.0 3.5 1.8 0.0 0.0
17 0 1.0 0.0 0.0 0.0 9.0 5.8 0.0 0.0
18 0 1.0 0.0 0.0 0.0 3.2 0.9 0.0 0.0
19 0 1.0 0.0 0.0 0.0 9.5 3.4 0.0 0.0
20 0 1.0 0.0 0.0 0.0 2.2 0.7 0.0 0.0
21 0 1.0 0.0 0.0 0.0 17.5 11.2 0.0 0.0
22 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
23 0 1.0 0.0 0.0 0.0 3.2 1.6 0.0 0.0
24 0 1.0 0.0 0.0 0.0 8.7 6.7 0.0 0.0
25 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
26 0 1.0 0.0 0.0 0.0 3.5 2.3 0.0 0.0
27 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
28 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
29 0 1.0 0.0 0.0 0.0 2.4 0.9 0.0 0.0
30 0 1.0 0.0 0.0 0.0 10.6 1.9 0.0 0.0

Table A2
Line data for modified IEEE 30-bus test system.

Start bus End bus R (p.u.) X (p.u.) B/2 (p.u.) Line limit (MW) Start bus End bus R (p.u.) X (p.u.) B/2 (p.u.) Line limit (MW)

1 2 0.0192 0.0575 0.0264 130 15 18 0.1073 0.2185 0.0 16

1 7 0.0452 0.1652 0.0204 130 18 19 0.0639 0.1292 0.0 16
2 8 0.0570 0.1737 0.0184 65 19 20 0.0340 0.0680 0.0 32
7 8 0.0132 0.0379 0.0042 130 12 20 0.0936 0.2090 0.0 32
2 3 0.0472 0.1983 0.0209 130 12 17 0.0324 0.0845 0.0 32
2 9 0.0581 0.1763 0.0187 65 12 21 0.0348 0.0749 0.0 32
8 9 0.0119 0.0414 0.0045 90 12 22 0.0727 0.1499 0.0 32
3 10 0.0460 0.1160 0.0102 70 21 22 0.0116 0.0236 0.0 32
9 10 0.0267 0.0820 0.0085 130 15 23 0.1000 0.2020 0.0 16
9 4 0.0120 0.0420 0.0045 32 22 24 0.1150 0.1790 0.0 16
9 11 0.0 0.2080 0.0 65 23 24 0.1320 0.2700 0.0 16
9 12 0.0 0.5560 0.0 32 24 25 0.1885 0.3292 0.0 16
11 5 0.0 0.2080 0.0 65 25 26 0.2544 0.3800 0.0 16
11 12 0.0 0.1100 0.0 65 25 27 0.1093 0.2087 0.0 16
8 13 0.0 0.2560 0.0 65 28 27 0.0 0.3960 0.0 65
13 6 0.0 0.1400 0.0 65 27 29 0.2198 0.4153 0.0 16
13 14 0.1231 0.2559 0.0 32 27 30 0.3202 0.6027 0.0 16
13 15 0.0662 0.1304 0.0 32 29 30 0.2399 0.4533 0.0 16
13 16 0.0945 0.1987 0.0 32 4 28 0.0636 0.2000 0.0214 32
14 15 0.2210 0.1997 0.0 16 9 28 0.0169 0.0599 0.065 32
16 17 0.0824 0.1923 0.0 16
 S. Verma, V. Mukherjee/Engineering Science and Technology, an International Journal 19 (2016) 1254−1265 1263

Table A3
Bus data for modified IEEE 57-bus test system.

Bus no. Bus code Voltage (V) Angle (°) Generation Load Generation

MW MVAR MW MVAR Qmin Qmax

1 1 1.04 0.0 146.39 0.0 55.0 17.0 −140 200.0

2 2 1.01 0.0 87.55 0.0 3.0 88.0 −40 50.0
3 2 0.99 0.0 41.97 0.0 41.0 21.0 −40 60.0
4 2 0.98 0.0 89.67 0.0 75.0 2.0 −30 25
5 2 1.01 0.0 461.21 0.0 150.0 22.0 −140 200
6 2 0.98 0.0 100.0 0.0 121.0 26.0 −30 9
7 2 1.02 0.0 344.95 0.0 377.0 24.0 −150 155
8 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
9 0 1.0 0.0 0.0 0.0 13.0 4.0 0.0 0.0
10 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
11 0 1.0 0.0 0.0 0.0 5.0 2.0 0.0 0.0
12 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
13 0 1.0 0.0 0.0 0.0 18.0 2.3 0.0 0.0
14 0 1.0 0.0 0.0 0.0 10.5 5.3 0.0 0.0
15 0 1.0 0.0 0.0 0.0 22.0 5.0 0.0 0.0
16 0 1.0 0.0 0.0 0.0 43.0 3.0 0.0 0.0
17 0 1.0 0.0 0.0 0.0 42.0 8.0 0.0 0.0
18 0 1.0 0.0 0.0 0.0 27.2 9.8 0.0 0.0
19 0 1.0 0.0 0.0 0.0 3.3 0.6 0.0 0.0
20 0 1.0 0.0 0.0 0.0 2.3 1.0 0.0 0.0
21 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
22 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
23 0 1.0 0.0 0.0 0.0 6.3 2.1 0.0 0.0
24 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
25 0 1.0 0.0 0.0 0.0 6.3 3.2 0.0 0.0
26 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
27 0 1.0 0.0 0.0 0.0 9.3 0.5 0.0 0.0
28 0 1.0 0.0 0.0 0.0 4.6 2.3 0.0 0.0
29 0 1.0 0.0 0.0 0.0 17.0 2.6 0.0 0.0
30 0 1.0 0.0 0.0 0.0 3.6 1.8 0.0 0.0
31 0 1.0 0.0 0.0 0.0 5.8 2.9 0.0 0.0
32 0 1.0 0.0 0.0 0.0 1.6 0.8 0.0 0.0
33 0 1.0 0.0 0.0 0.0 3.8 1.9 0.0 0.0
34 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
35 0 1.0 0.0 0.0 0.0 6.0 3.0 0.0 0.0
36 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
37 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
38 0 1.0 0.0 0.0 0.0 14.0 7.0 0.0 0.0
39 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
40 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
41 0 1.0 0.0 0.0 0.0 6.3 3.0 0.0 0.0
42 0 1.0 0.0 0.0 0.0 7.1 4.0 0.0 0.0
43 0 1.0 0.0 0.0 0.0 2.0 1.0 0.0 0.0
44 0 1.0 0.0 0.0 0.0 12.0 1.8 0.0 0.0
45 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
46 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
47 0 1.0 0.0 0.0 0.0 29.7 11.6 0.0 0.0
48 0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
49 0 1.0 0.0 0.0 0.0 18.0 8.5 0.0 0.0
50 0 1.0 0.0 0.0 0.0 21.0 10.5 0.0 0.0
51 0 1.0 0.0 0.0 0.0 18.0 5.3 0.0 0.0
52 0 1.0 0.0 0.0 0.0 4.9 2.2 0.0 0.0
53 0 1.0 0.0 0.0 0.0 20.0 10.0 0.0 0.0
54 0 1.0 0.0 0.0 0.0 4.1 1.4 0.0 0.0
55 0 1.0 0.0 0.0 0.0 6.8 3.4 0.0 0.0
56 0 1.0 0.0 0.0 0.0 7.6 2.2 0.0 0.0
57 0 1.0 0.0 0.0 0.0 6.7 2.0 0.0 0.0
1264 S. Verma, V. Mukherjee/Engineering Science and Technology, an International Journal 19 (2016) 1254−1265

Table A4
Line data for modified IEEE 57-bus test system.

Start bus End bus R (p.u) X (p.u) B/2 (p.u) Line limit (MW) Start bus End bus R (p.u) X (p.u) B/2 (p.u) Line limit (MW)

1 2 0.0083 0.0280 0.0645 150 10 29 0.0 0.0648 0.0 100

2 3 0.0298 0.0850 0.0409 85 25 30 0.1350 0.2020 0.0 100
3 8 0.0112 0.0366 0.0190 100 30 31 0.3260 0.4970 0.0 100
8 9 0.0625 0.132 0.0129 100 31 32 0.5070 0.7550 0.0 100
8 4 0.0430 0.148 0.0174 50 32 33 0.0392 0.0360 0.0 100
4 10 0.0200 0.102 0.0138 40 34 32 0.0 0.9530 0.0 100
4 5 0.0339 0.173 0.0235 100 34 35 0.0520 0.0780 0.0016 100
5 6 0.0099 0.050 0.0274 200 35 36 0.0430 0.0537 0.0008 100
6 11 0.0369 0.167 0.0220 50 36 37 0.0290 0.0366 0.0 100
6 12 0.0258 0.0848 0.0109 50 37 38 0.0300 0.1009 0.0010 100
6 7 0.0648 0.0295 0.0386 50 37 39 0.0192 0.0379 0.0 100
6 13 0.0481 0.158 0.0203 50 36 40 0.0 0.0466 0.0 100
13 14 0.0132 0.0434 0.0055 50 22 38 0.2070 0.0295 0.0 100
13 15 0.0269 0.0869 0.0115 100 12 41 0.0 0.7490 0.0 100
1 15 0.0178 0.0910 0.0494 200 41 42 0.0289 0.3520 0.0 100
1 16 0.0454 0.2060 0.0273 100 41 43 0.0 0.4120 0.0 100
1 17 0.0238 0.1080 0.0143 100 38 44 0.0 0.0585 0.0010 100
3 15 0.0162 0.0530 0.0272 100 15 45 0.0230 0.1042 0.0 100
8 18 0.0 0.5550 0.0 100 14 46 0.0182 0.0735 0.0 100
8 18 0.0 0.4300 0.0 100 46 47 0.0834 0.0680 0.0016 100
9 4 0.0302 0.0641 0.0062 100 47 48 0.0801 0.0233 0.0 100
10 5 0.0139 0.0712 0.0097 100 48 49 0.1386 0.1290 0.0024 100
11 7 0.0277 0.1262 0.0164 100 49 50 0.0 0.1280 0.0 100
12 13 0.0223 0.0732 0.0094 100 50 51 0.0 0.2200 0.0 100
7 13 0.0178 0.0580 0.0302 100 11 51 0.1442 0.0712 0.0 100
7 16 0.0180 0.0813 0.0108 100 13 49 0.0762 0.1910 0.0 100
7 17 0.0397 0.1790 0.0238 100 29 52 0.1878 0.1870 0.0 100
14 15 0.0171 0.0547 0.0074 100 52 53 0.1732 0.0984 0.0 100
18 19 0.4610 0.6850 0.0 100 53 54 0.0 0.2320 0.0 100
19 20 0.2830 0.4340 0.0 100 54 55 0.0624 0.2265 0.0 100
21 20 0.0 0.7767 0.0 100 12 43 0.0 0.1530 0.0 100
21 22 0.0736 0.1170 0.0 100 44 45 0.5530 0.1242 0.0020 100
22 23 0.0099 0.0152 0.0 100 40 56 0.2125 1.1950 0.0 100
23 24 0.1660 0.2560 0.0042 100 56 41 0.0 0.5490 0.0 100
24 25 0.0 1.1820 0.0 100 56 42 0.1740 0.3540 0.0 100
24 25 0.0 1.23 0.0 100 39 57 0.1150 1.3550 0.0 100
24 26 0.0 0.0473 0.0 100 57 56 0.0312 0.2600 0.0 100
26 27 0.1650 0.2540 0.0 100 38 49 0.0 0.1770 0.003 100
27 28 0.0618 0.0954 0.0 100 38 48 0.0 0.0482 0.0 100
28 29 0.0418 0.0587 0.0 100 6 55 0.0 0.1205 0.0 100

Table A5 [3] M. Shahidehpour, H. Yamin, Z. Li, Market Operations in Electric Power Systems,
Price bids submitted by GENCOs for modified IEEE 30-bus test system. John Wiley & Sons, Chichester, 2002.
[4] E. Bompard, P. Correia, G. Gross, M. Amelin, Congestion-management schemes:
Bus number Increment ($/MWh) Decrement ($/MWh) a comparative analysis under a unified framework, IEEE Trans. Power Syst. 18
(1) (2003) 346–352.
1 22 18
[5] F.D. Faliana, M. Ilic, A mathematical framework for the analysis and management
2 21 19
of power transactions under open access, IEEE Trans. Power Syst. 13 (2) (1998)
3 42 38
681–687.
4 43 37 [6] H. Glavitsch, F. Alvarado, Management of multiple congested conditions in
5 43 35 unbundled operation of a power system, IEEE Trans. Power Syst. 13 (3) (1998)
6 41 39 1013–1019.
[7] B.D. Thukaram, K. Parthasarathy, An expert system for alleviation of network
overloads, Electr. Power Syst. Res. 40 (1997) 143–153.
[8] T. Shu, G. Gross, A congestion-management allocation mechanism for multiple
Table A6 transaction networks, IEEE Trans. Power Syst. 17 (3) (2002) 826–833.
Price bids submitted by GENCOs for modified IEEE 57-bus test system. [9] A. Kumar, S.C. Srivastava, S.N. Singh, A zonal congestion management approach
using real and reactive power rescheduling, IEEE Trans. Power Syst. 19 (1) (2004)
Bus number Increment ($/MWh) Decrement ($/MWh) 554–562.
[10] S. Dutta, S.P. Singh, Optimal rescheduling of generator for congestion
1 44 41
management based on particle swarm optimization, IEEE Trans. Power Syst.
2 43 39
23 (4) (2008) 1560–1569.
3 42 38 [11] B.K. Talukdar, A.K. Sinha, S. Mukhopadhyay, A. Bose, A computationally simple
4 43 37 method for cost-efficient generation rescheduling and load shedding for
5 42 39 congestion management, Int. J. Electr. Power Energy Syst. 27 (5) (2005) 379–
6 44 40 388.
7 44 41 [12] A.J. Conejo, F. Milano, R. Garacia-Bertrand, Congestion management ensuring
voltage stability, IEEE Trans. Power Syst. 21 (1) (2006) 357–364.
[13] S. Kumar, D.K. Chaturvedi, Optimal power flow solution using fuzzy evolutionary
and swarm optimization, Int. J. Electr. Power Energy Syst. 47 (2013) 416–423.
[14] A. Kumar, C. Sekhar, Congestion management with FACTS devices in deregulated
References electricity markets ensuring loadability limit, Int. J. Electr. Power Energy Syst.
46 (2013) 258–273.
[15] S. Balaraman, N. Kamaraj, Transmission congestion management using particle
[1] L.L. Lai, Power System Restructuring and Deregulation, Wiley, New York, 2001. swarm optimization, J. Electr. Syst. 7 (1) (2011) 54–70.
[2] R.D. Christie, B.F. Wollenberg, I. Wangensteen, Transmission management in [16] S. Hajforoosh, S.M.H. Nabavi, M.A.S. Masoum, Coordinated aggregated-based
the deregulated environment, P. IEEE 88 (2) (2000) 170–194. particle swarm optimisation algorithm for congestion management in
 S. Verma, V. Mukherjee/Engineering Science and Technology, an International Journal 19 (2016) 1254−1265 1265

restructured power market by placement and sizing of unified power flow [22] A. Rastgou, J. Moshtagh, Improved harmony search algorithm for transmission
controller, IET Sci. Meas. Technol. 6 (4) (2012) 267–278. expansion planning with adequacy-security considerations in deregulated power
[17] S. Hajforoosh, S.M.H. Nabavi, M.A.S. Masoum, Optimal particle swarm based system, Int. J. Electr. Power Energy Syst. 60 (2014) 153–164.
placement and sizing of static synchronous series compensator to maximize [23] F. Zhuang, F.D. Galiana, Unit commitment by simulated annealing, IEEE Trans.
social welfare, J Electr. Eng. Tech. 7 (4) (2012) 501–512. Power Syst. 5 (1) (1990) 311–318.
[18] S.M.H. Nabavi, S. Hajforoosh, M.A.S. Masoum, Placement and sizing of distributed [24] J.S.R. Jang, C.T. Sun, E. Mizutani, Neuro-fuzzy and Soft Computing: a
generation units for congestion management and improvement of voltage Computational Approach to Learning and Machine Intelligence, Pearson
profile using particle swarm optimization, Innovative Smart Grid Technologies Education, Upper Saddle River, 1996.
Asia (ISGT), IEEE PES, 6 2011, 1–6. [25] X.-S. Yang, Firefly algorithm, stochastic test functions and design optimisation,
[19] S. Hajforoosh, S.M.H. Nabavi, M.A.S. Masoum, Maximizing social welfare in Int. J. Bio-Inspir. Comput. 2 (2) (2010) 78–84.
double-sided auction market by placement and sizing of TCSC using coordinated [26] P. Balachennaiah, M. Suryakalavathi, P. Nagendra, Optimizing real power loss
aggregation-based particle swarm optimization, Int. Rev. Electr. Eng. 6 (5) (2011) and voltage stability limit of a large transmission network using firefly
2557–2568. algorithm, Eng. Sci. Technol. Int. J. 19 (2) (2016) 800 – 810.
[20] S.M.H. Nabavi, A. Kazemi, M.A.S. Masoum, Social welfare maximization with [27] A. Gupta, P.K. Padhy, Modified firefly algorithm based controller design for
fuzzy based genetic algorithm by TCSC and SSSC in double-sided auction market, integrating and unstable delay processes, Eng. Sci. Technol. Int. J. 19 (1) (2016)
Sci. Iran. 19 (3) (2012) 745–758. 548–558.
[21] S.M.H. Nabavi, M.A.S. Masoum, A. Kazemi, A fuzzy-based genetic algorithm [28] X.-S. Yang, Nature-Inspired Metaheuristic Algorithms, Luniver Press, Beckington,
for social welfare maximization by placement and sizing of static 2008.
synchronous series compensator, Electr. Power Comp. Syst. 39 (13) (2011) [29] D.P. Kothari, J.S. Dhillon, Power System Optimization, PHI, New Delhi, 2011.
1329–1352. [30] H. Saadat, Power System Analysis, Tata McGraw Hill Ltd, New Delhi, 2002.