IEEE INDICON 2015 1570187427

1  
2  
3  
4  
Congestion Management in Deregulated Electricity
5  
6  
Market Using Distributed Generation
7  
8  
9   Md Sarwar*, Student member, IEEE, Anwar Shahzad Siddiqui
10  
Department of Electrical Engineering, F/O Engineering and Technology,
11  
Jamia Millia Islamia, New Delhi, India
12  
*[email protected]
13  
14  
15   Abstract— A congestion management approach in a deregulated distributed generations (DGs) are getting more interest in
16   electricity market using distributed generation (DG) has been restructured power system as it can be generally located in
17   proposed. Locational marginal price (LMP) has been utilized to load pockets and can be considered as negative power
18   find the optimal location of DG. Two different methods, highest demands. DGs can quickly respond to the changing conditions
19   LMP method and LMP difference method, are used to capture
of competitive electricity market and therefore has got an
20   the potential locations for DG placement. Based on the priority
of locations obtained from these two methods, the DG is placed to augmented interest for power system operation and planning
21   in deregulated environment. Other benefits of DGs include
mitigate congestion. For selection of optimal size of DG, a
22   method based on performance index for different ratings of DG easy installation and simple operation with low capital cost.
23   is proposed and its effectiveness with the two methods of locating Its strategical location and operation in system reduce losses,
24   the DG is analysed. The performance of the proposed improve voltage profile, defer system upgrades and improve
25   methodologies have been tested on IEEE 14-bus system. reliability of the system. With all these benefits DGs are being
26   extensively used for congestion management in restructured
27   Keywords— Distributed generation; deregulated electricity power system. The impact of 4istributed resources on
28   market; locational marginal price; congestion management; OPF managing congestion is discussed in [13]. In this paper, an
29   approach for the optimal placement of DGs to mitigate
I. INTRODUCTION
30   congestion based on real and reactive power flow as well as
31   In view of the fact that these days all the attempts are being voltage magnitude contribution factors is reported. Optimal
32   made to make the electricity market more competitive in order location and sizing of DG based on LMP and congestion rent
33   to meet the increasing demand of electricity around the world as well as DG benefit-to-cost ratio is discussed in [15]. But the
34   at cheaper price, thereby endangering the transmission system performance indices calculated for the optimal location and
35   security. This causes a major problem of congestion in sizing of DG do not include the generation cost of DG.
36   transmission lines of deregulated electric power systems
which would increase the electricity price and in turn will take In this paper the congestion management of transmission
37  
away all the benefits of deregulation. Transmission line lines in deregulated electricity market is done using DG. Two
38  
congestion is defined as the violation of any of the transfer different methods for optimal placement of DG in order to
39  
limits of transmission lines such as voltage limits, thermal mitigate congestion are proposed which are based on highest
40   LMP of buses and difference of LMPs between two buses. A
41   limits, stability limits etc. In competitive electricity market,
transmission line congestion occurs due to the inadequate priority list based on each of the above two methods are
42   formed to capture the potential locations for optimal DG
43   capacity of transmission line to accommodate all the power
transactions in order to meet all customers’ demands. Since placement. For the selection of optimal size of DG, a
44   methodology based on the performance index for different DG
45   congestion would affect the competitive electricity market
equilibrium, therefore its judicious management is very vital ratings is proposed. The proposed performance index
46   considers the cost of generation through DG along with
47  
which will also keep the power systems’ reliability and
security intact. A number of congestion management generation cost of the system.
48  
methodologies have been reported in literature [1][2]. Some of
49   II. PROBLEM FORMULATION
the available methods are based on generation rescheduling
50   To evaluate the nodal prices of electricity, the problem is
[3]-[6], load shedding [7], management of reactive power
51   formulated as an optimal power flow (OPF) formulation in
[8][9], zones/clustering [9][10], FACTS devices placement
52   pool based deregulated electricity market with the objective of
[11], voltage stability [12], etc. Although most of the above
53   minimization of generation cost of electricity given by
methods utilize the generation side approaches for congestion
54   alleviation, but in the new era of competitive electricity equation (1) while all other constraints are satisfied.
55   market, demand side approach are getting more attention as it
56  
Minimize C P (1)
alleviates congestion more effectively thereby improving the
57   reliability and security of the power system [13]. Therefore
60  
61   978-1-4673-6540-6/15/$31.00 ©2015 IEEE
62  
63  
64  
65  

1
where ng is the total number of generating units and Ci(PGi) is where λ and µ are Lagrangian multipliers vectors associated
the cost of electricity generation of ith generating unit given as with equality constraints and inequality constraints
quadratic cost function: respectively obtained by OPF solution. Interior point method
is used for OPF solution in MATLAB environment.
= .( ) + ! .( )+" (2)
III. OPTIMAL LOCATION OF DG
where ai, bi and ci are the cost coefficients and PGi is the
amount of electricity generation of ith unit. The method for optimal location of DG is based upon
highest LMP and difference of LMP of buses across a line in
The above objective functions are subjected to following the system. Initially, the solution space is reduced by
constraints. considering a list of bus candidate location. Since the buses
1. Power balance constraint at each node having more generation capacity than their demand have low
LMP, therefore these are not considered as potential locations
# − %
+ &% = 0; ) = 1,2, … … . , -. (3) for DG placement. Hence buses are examined for potential
location of DG according to equation (10) such that it has
0# − 0 %
+ 0&% = 0; ) = 1,2, … … . , -. (4)
either no generating unit or it has low generating capacity as
2. Generator operating limit constraint compared to its load.

P 32 ≤ P ≤ P 356 ; m = 1, 2, … , n7 (5) P H
≤ PJH ; k = 1,2, … … , nC (10)
2 2

Q 32 ≤ Q 2
≤ Q 356
2
; m = 1, 2, … , n7 (6) The two methods used for finding the optimal location for
placement of DG in order to mitigate congestion are discussed
3. Line flow constraints as follows.
356
F<= ≤ F<= ; kl = 1,2, … , n@ (7) A. Highest LMP Method
4. Bus voltage limit A simple method that can be employed to place DG
optimally for congestion alleviation is highest LMP method
356
V<3 ≤ V< ≤ V3< ; k = 1,2, … , nC (8) [15]. Since LMP gives an indication for transmission losses as
well as congestion, therefore it can be utilize for congestion
where nb is the total number of system buses, PGi and PGimax min
management. Under normal operation of a power system,
are respectively the minimum and maximum real power LMP at all buses remain almost same, but as congestion
output limits of ith generator, QGimin and QGimax are respectively occurs, the LMPs at all buses will become different. In highest
the minimum and maximum reactive power output limits of ith LMP method for optimal DG placement, buses are ranked on
generator, Fkl denotes the flow of power on transmission line- the basis of their LMPs such that the bus having highest LMP
kl connected between bus-k and bus-l due to accommodation would be ranked 1 and so on. In order to alleviate congestion,
of all contracts, Fklmax is the power flow limit of line-kl DG is placed on the top ranked bus while fulfilling equation
connected between bus-k and bus-l, nL denotes the total (10). If equation (10) is not satisfied, then the DG would be
number of lines and Vkmin and Vkmax are respectively the placed on next ranked bus and so on.
minimum and maximum voltage limits at kth bus.
B. LMP Difference Method
The optimization of the objective function incorporating all Although the highest LMP method of DG placement is
the constraints is done using Lagrangian method. The simple, but it may give rise to a situation that the congestion
Lagrangian function of the optimization problem including all increase in the network. Therefore, this method could not be
the constraints in objective function is written as: reliably used for DG placement. Another method which is
I
based on difference of LMPs of two buses across a line,
known as “ LMP difference method” can be utilized more
ℒ = C P + λGH P< − P H
+ PJH efficiently and reliably to place a DG optimally in order to
< mitigate congestion. The optimal placement of TCSC using
I M
LMP difference method for congestion management has been
+ λKH Q < − Q H
+ Q JH + μ@ (F<= − F<=
356
) discussed in [16]. Since LMP difference across a congested
< <= line will be highest as compared to other lines, therefore this
method can more effectively reflect the list of potential
locations for DG placement. Therefore, a priority list based on
+ μN P 3 −P + μO P − P 356 LMP difference across a line is formed to find the potential
location for optimal DG placement. The LMP difference
across a line is given by equation (11).
+ μN Q3 −Q + μO Q − Q356 ∆LMP<= = LMP< − LMP= ; kl = 1,2, … , n@ (11)

I I where ∆LMPkl is the LMP difference across line-kl, LMPk

and LMPl are the LMPs at bus-k and bus-l respectively.
+ μN
PH V<3 − V< + PH (V<
μO − V<356 ) (9)
< <

2
 Utilizing equation (10) as well as equation (11), the best Table I shows the priority list for optimal DG placement
possible location for DG placement in order to alleviate based on highest LMP method. Only six buses having highest
congestion is obtained. LMP before allocating DG (base case) have been listed in
Table I. Although, bus-3 has the highest LMP, but it will not
IV. OPTIMAL SIZE OF DG be selected for placement of DG as it does not satisfy equation
In competitive electricity market, the main aim of the (10). Therefore, bus-14 having next in priority is selected for
system operator is to maximize the social welfare (or DG placement. The cost of generations is 6353.65 $/h and
minimize the total generation cost when there is no demand TCR is 2119.85 $/h.
bidding) while maintaining system stability and reliability. TABLE I
But due to high cost associated with electricity production by PRIORITY TABLE FOR DG PLACEMENT BASED ON HIGHEST LMP METHOD
DG as well as its investment, there is considerable risk in their
application. Hence, not only its placement but its size is also LMP
of great concern so that its application would maximize the Priority No. Bus No.
($/MWh)
social welfare, Therefore in order to find the optimal size of 1 48.202 3
DG, a new methodology is proposed to calculate the
2 35.220 14
performance index based on generation cost, TCR as well as
generation cost of DG and is given as follows. 3 35.200 4
4 34.991 7
C UV + TCR YZ − C [\ + TCR J
PIJ = (12) 5 34.983 8
CJ
6 34.894 9
where PIDG is the performance index for DG placement, CiBC
and TCRBC are generation cost and total congestion rent Table II presents the generation cost as well as performance
respectively for base case and CiDG and TCRDG are generation index when DG of different ratings are placed at bus-14.
cost and total congestion rent respectively after DG placement
TABLE II
and CDG is the cost characteristic of DG. TCR for a network is PERFORMANCE INDEX FOR DIFFERENT RATINGS OF DG PLACED AT BUS-14
calculated as follows.
CR <= = ∆LMP<= . P<= ; kl = 1,2, … , n@ (13) DG Rating CiDG
PIDG
(MW) ($/h)
M

TCR = ∆LMP<= . P<= (14)

0.5 6333.88 0.996

<=
1.0 6316.33 1.342
1.5 6298.81 1.456
where CRkl is the congestion rent for line-kl connected
between bus-k and bus-l, and Pkl is the power flow on line-kl. 2.0 6281.32 1.512

Performance index for optimal sizing is calculated using 2.5 6263.86 1.546
equation (12) which considers the cost of generation through 3.0 6246.45 1.564
DG which is very important in order to analyse the effect on 3.5 6225.06 1.615
social welfare (or system generation cost in case of absence of
4.0 6214.71 1.576
demand side bidding). The numerator of equation shows the
revenue savings due to implementation of DG and the 4.5 6208.39 1.568
denominator shows the cost of generation through DG. 5.0 6197.10 1.560
Therefore PIDG indicates the overall savings factor due to
implementation of DG. High value of PIDG indicates gives the Table II shows that for DG rating of 3.5 MW, highest
high value of savings factor. Therefore, the size of DG for performance index is obtained. Therefore, the optimal size of
which PIDG comes out to be maximum, is the optimal size of DG selected for congestion management is 3.5 MW. Table II
DG for congestion management in competitive electricity also shows that as the rating of the DG increases the
market. generation cost of the system decreases. But since there must
be some limit to the size of DG due to high investment cost,
V. TEST CASE AND RESULTS
therefore its size is optimized using performance index.
The proposed methodologies are tested for its effectiveness
on modified IEEE 14-bus system whose network data is taken Table III presents the results for LMP difference method
from [17] while the generator data is taken from [16]. The DG which illustrate the priority list for DG placement based on
is considered to inject only real power between 1 MW to 5 LMP difference across each line. The lines having major LMP
MW in an incremental step of 0.5 MW. The cost difference have been listed in priority table, Table III, which
characteristics of DG is taken from [18]. shows that line-3 connected between bus-2 and bus-3 has the
highest LMP difference.

3
 TABLE III
PRIORITY TABLE FOR DG PLACEMENT BASED ON LMP DIFFERENCE DG a t bus-4 DG at bus-14
LMP Difference 1.8
Priority No. Line No.
($/MWh)
1.6
1 16.7 Line 3:2-3
1.4

Performance Index ( PIDG )

2 13.0 Line 6:3-4
3 12.5 Line 2:1-5 1.2
4 11.5 Line 1:1-2
1
5 3.7 Line 4:2-4
0.8
6 2.6 Line 7:4-5
0.6
Although line-3 connected between bus-2 and bus-3 has
0.4
highest LMP difference, but it is not a potential location for
DG placement as it does not satisfy equation (10). Therefore, 0.2
next line i.e line-6 connected between bus-3 and bus-4 from
the priority table is considered. Since bus-3 is a generator bus, 0
it also does not satisfy equation (10) while bus-4 is a load bus, 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
so it satisfies equation (10). Therefore, DG is optimally placed
DG rating (MW)
at bus-4. The priority Table I based on highest LMP method
shows that bus-4 is the next optimal location for DG Fig. 1 Performance Index after allocation of DG of different power ratings
placement after bus-14. But Table III shows that bus-4 is the
best location for optimal placement of DG and bus-14 does VI. CONCLUSIONS
not find place in priority table as line connecting it has very In this paper, two different methods for optimal placement
low LMP difference. Hence the LMP difference method of DG to manage congestion are proposed which are based on
captures the optimal location for DG placement more highest LMP of buses and difference of LMPs between two
efficiently than the highest LMP method. buses. Both the methods capture the potential locations for
DG placement, but the later is more effective in that. Also a
Table IV shows the results for generation cost and method based on performance index calculated considering
performance index when DG of different ratings is placed on generation cost, TCR and generation cost through DG is
bus-4. Here also, when DG is placed at bus-4, highest proposed for selection of optimal size of DG such that its high
performance index is obtained for DG rating of 3.5 MW, investment cost and production cost does not affect the
similar to when DG is placed at bus 14. This shows that benefits of deregulation. The proposed method for selection of
irrespective of the two methods for the optimal location of DG size of DG gives the optimal size irrespective of the location
placement for congestion management, the optimal size of the of placement of DG in the system. The proposed
DG is found to be same which is also illustrated in Fig. 1. methodologies are tested on modified IEEE14-bus system and
TABLE IV proved to be effective in selecting the optimal location and
PERFORMANCE INDEX FOR DIFFERENT RATINGS OF DG PLACED AT BUS-4 size of DG for congestion management.
DG Rating CiDG
(MW) ($/h)
PIDG REFERENCES
[1] R. D. Christie, B. Wollenberg, and I. Wangensteen, “Transmission
0.5 6321.55 0.238 management in the deregulated environment,” Proc. IEEE, vol. 88,
no.2, pp. 170–195, Feb. 2000.
1.0 6315.61 0.382 [2] A. Kumar, S. C. Srivastava, and S. N. Singh, “Congestion management
in competitive power market: A bibliographical survey,” Elect. Power
1.5 6297.71 0.851 Syst. Res., vol. 76, pp. 153–164, 2005.
[3] K. Talukdar, A. K. Sinha, S. Mukhopadhyay, and A. Bose, “A
2.0 6279.86 1.106 Computationally simple method for cost-efficient generation
rescheduling and load shedding for congestion management,” Int. J.
2.5 6262.06 1.238 Elect. Power Energy Syst., vol. 27, no. 5, pp. 379–388, Jun.–Jul. 2005.
[4] S. Dutta, and S.P. Singh, Optimal rescheduling of generators for
3.0 6244.3 1.314 congestion management based on particle swarm optimization, IEEE
Transactions on Power Systems, vol. 23, no. 4, pp. 1560-1569, 2008.
3.5 6226.57 1.372 [5] P. Boonyaritdachochai, C. Boonchuay, and W. Ongsakul, “optimal
congestion management in an electricity market usingparticle swarm
4.0 6214.87 1.358
optimization with time-varying acceleration coefficients”, Computers
and Mathematics with Applications,vol. 60, pp. 1068-1077, 2010.
4.5 6201.21 1.366
[6] Siddiqui A S, Sarwar M, Ahsan S, “Congestion management using
5.0 6173.58 1.366 improved inertia weight particle swarm optimization”, IEEE Power
India International Conference (PIICON), New Delhi, 2014, 1-5

4
[7] A. Shandilya, H. Gupta, and J. Sharma, “Method for generation [14] J. Liu, M.M.A Salama, and R. R. Mansour, “Identify the impact of
rescheduling and load shedding to alleviate line overloads using local distributed resources on congestion management”, IEEE Trans.
optimization”, in Proc. Inst. Elect. Eng., vol. 140, pp. 337-342, 1993. Power Deliv., vol. 20, no. 3, pp. 1998–2005, Jul. 2005.
[8] S. Hao, and A. Papalexopoulos, “Reactive power pricing and [15] M. Afkousi-Paqaleh, A. Abbaspour-Tehrani Fard, and M.
management”, IEEE Trans. Power Syst., vol. 12, pp. 95–104, 1997. Rashidinejad, “Distributed generation placement for congestion
[9] A. Kumar, S. C. Srivastava, and S. N. Singh, “A zonal congestion management considering economical and financial issues”, Elect. Eng.,
management approach using real and reactive power rescheduling,,” vol. 92, pp. 193-201, 2010.
IEEE Trans. Power Syst., vol. 19, no. 1, pp. 554–562, Feb. 2004. [16] N. Acharya, and N. Mithulananthan, “Locating series FACTS devices
[10] A. Kumar, S. C. Srivastava, and S. N. Singh, “A zonal congestion for congestion management in deregulated electricity markets”, Elect.
management approach using ac transmission congestion distribution Power Syst. Res., vol. 77, pp. 352–360, 2007.
factors,” Elect. Power Syst. Res., vol. 72, pp. 85–93, 2004. [17] Power system test case archives (2004) [Online]. Available:
[11] S. N. Singh, and A. K. David, “Optimal location of FACTS devices for http://www.ee.wasington.edu/research/pstca
congestion management”, Elect. Power Syst. Res., vol. 58, pp. 71–79, [18] K. Singh, V.K. Yadav,N.P.Padhy, and J. Sharma, “ Congestion
2007. management considering optimal placement of distributed generator in
[12] A. J. Conejo, F. Milano, and R. Bertrand, “ Congestion management deregulated power system networks”, Elec. Power Comp. Syst., vol.
ensuring voltage stability”, IEEE Trans. Power Syst., vol. 21, pp. 357– 42, no. 1, pp. 13-22, Dec. 2014.
364, 2006.
[13] F. Rahimi, and A. Ipakchi, “Demand response as a market resource
under smart grid paradigm”, IEEE Trans. Smart Grid, vol. 1, pp. 82-
88, 2010.