Power Trading and Congestion Management Through Real Power Rescheduling Using Unified Power Flow Controller

Indonesian Journal of Electrical Engineering and Informatics (IJEEI)
Vol. 2, No. 4, December 2014, pp. 151~160
ISSN: 2089-3272  151
Received July 27, 2014; Revised September 15, 2014; Accepted October 1, 2014
Power Trading and Congestion Management through
Real Power Rescheduling Using Unified Power Flow
Controller
S.K. Gupta, R.Bansal, Partibha Sharma, Mukesh Saini
Electrical Engineering Department, D.C.R. University of Sc. & Technology
Murthal Sonipat, India
Email: drskgupta.ee@dcrustm.org, er.richa09@gmail.com, partibhasharma66@gmail.com,
mukeshsainimvn@gmail.com
Abstract
Congestion is termed as the operating condition in which there is not enough transmission
capacity to implement all the desired transactions. This paper deals with the power trading in electricity
market to ensure regular supply at competitive rates. Bidding process of 75 Indian bus systems is
analyzed. It is shown that how can congestion cost can be addressed through active power rescheduling
with transmission line constraints using Unified power flow controller.
Keywords: GENCO; DISCO, Congestion Management, Disco Participation Matrix (DPM), UPFC, Pool
Based Transaction, Bidding, Congestion Cost, Rescheduling
1. Introduction
Open access environment may try to purchase the energy from the cheaper source for
greater profit margins, which may lead to overloading and congestion of certain corridors of the
transmission network. This may result in violation of line flow, and stability limits. Utilities
therefore need to determine adequately their available transfer capability (ATC) to ensure that
system reliability is maintained while serving a wide range of bilateral and multilateral
transactions [1]. System Operator (SO) is to manage congestion as it cause rise in electricity
price resulting in market inefficiency. In corrective action congestion management schemes, it is
crucial for SO to select the most sensitive generators to re-schedule their real and reactive
powers for congestion management [2], [3]. Whenever transmission network congestion occurs
how it segregates the wholesale electricity market and forces the market to change its price
from a common market clearing price to locational market price [4]. The voltage profile become
poor during peak loading of the network and can lead to congestion during such events [5]. . In
order to increase ATC, voltage improvement as well as minimum capital cost the deployment of
UPFC is suggested [13]. By employing a combination of capital cost indices and search for
suitable locations for UPFC a cost function is developed.
2. System under Studies
The possibility of controlling power flow in power system can improve its performance
with generation re-scheduling. The congestion is relieved by changing the line flows. In this
paper 400 kV and 200kV reduced network of one of the Electricity Boards in India which
consists of 15 generators and 97 lines, including 24 transformers is considered [3], [8]. The
single line diagram of 75-bus system is shown in Figure 1.

 ISSN: 2089-3272
IJEEI Vol. 2, No. 4, December 2014: 151 – 160
152
Figure 1. A 75 bus system under study
This system is divided into four areas to demonstrate the bidding process. Red, yellow,
blue and green color represents Control area 1, 2, 3 & 4 respectively. Detail for Control areas is
given in Table I [8].
Table I. Control Areas in 75 Bus Systems
CONTROL AREA OWNER DISCOS BUSES
AREA 1 Gencos-5,6,7 1
2
30,57,59,61,65,75
32,38,39,53,62
AREA 2 Gencos-1,2,9,12,13 3
4
16,46,50
42,47,74
AREA 3 Gencos-3,11 5
6
52,71,27,26,51,68
20,48,49,64,66,69,37
AREA 4 Gencos-4,8,10,14,15 7
8
40,56,58,60,70,72,25
28,24,34,55,63,54,73,67
Distribution companies (DISCOs) make the binary contracts with GENCOs which is
confirmed by the Power Exchange on the availability of ATC [7]. Such contract is represented
by Distribution Participation matrix (DPM). DPM for 75 bus system for a particular schedule is
shown in Table II.
Table II. Disco Participation Matrix
D1 D2 D3 D4 D5 D6 D7 D8
G6 0.1 0 0 0.05 0 0 0 0
G5 0 0.1 0 0 0 0 0 0.2
G7 0 0 0 0 0 0 0 0
G1 0 0 0.05 0 0 0.1 0 0
G2 0 0.05 0 0.1 0 0 0.15 0
G9 0.1 0 0.2 0.15 0 0 0 0
G12 0 0 0 0 0 0 0 0
G13 0 0 0 0 0 0 0 0
G3 0 0.05 0 0 0.15 0.2 0 0
G11 0 0 0 0 0 0.1 0 0
G14 0.1 0.15 0.1 0 0.25 0 0.25 0
G4 0.1 0 0.05 0.1 0 0.1 0 0.25
G8 0.1 0.15 0.1 0.1 0.1 0 0.1 0.05
G10 0 0 0 0 0 0 0 0
G15 0 0 0 0 0 0 0 0
Pool 1.1363 4.4388 2.4973 13.1255 5.4463 5.3367 5.9501 13.7502
Total 1.6363 4.9388 2.9973 13.6255 5.9463 5.8367 6.4501 14.2502

IJEEI ISSN: 2089-3272 
Power Trading and Congestion Management through Real Power … (Mukesh Saini)
153
The balance demand of DISCOs is met by Pool based transaction which is shown in the
second last raw of the Table II. Power demand by area 1 from pool (P ) is 5.5751 pu, by area 2
(P ) is 15.6228 pu, by area 3( P ) is 10.783 pu and by area 4 is P is 19.7003 pu. Total
power given by Gencos of area 1 in pool (P ) is 3.15 pu, Gencos of area 2 (P is 41.45,
Gencos of area 3 (P is 2.39 pu and Gencos of area 4 (P is 6.49 pu.
3. Bidding Process
The bidding process is for time block of 15 minutes one day ahead. Considering the
bidding from 9 am to 9.15 am on any particular day where market bidders from all areas must
submit separate bids for the area in which they have generation & loads. The bidding curves for
all areas area are shown in Figures 2, 3, 4 and 5.
Bidding Curve for area 1: It is assumed that in area 1 the Genco 6 bids for 1.05 pu
power at Rs 2100/- , Genco 5 bids for 1.5 power at Rs 1000/- and the Genco 7 bids for 0.6 pu
power at Rs 2700/-. The supply and demand curve intersects at 2700 Rs/MWh which is MCP as
shown in figure 2. MVA base is taken 100.
Figure 2. Bidding Curve for Area 1
Bidding Curve for area 2: In Area 2 the Genco 1 bids for 7.10 pu power at Rs 1200/- ,
Genco 2 bids for 2.30 pu power at Rs 2500/- , Genco 9 bids for 5.05 pu power at Rs 4600/- ,
Genco 12 bids for 18 pu power at Rs 3800/- and the Genco 13 bids for 9 pu power at Rs 5000/-.
The supply and demand curve intersects at 3800 Rs/MWh which is MCP of this area as shown
in figure 3.
Bidding Curve for area 3: In Area 3 the Genco 3 bids for 1.4 pu power at Rs 1800/-,
Genco 11 bids for 0.99 pu power at Rs 3000/. The supply and demand curve intersects at 3000
Rs/MWh which is the MCP of this area as shown in figure 4.

 ISSN: 2089-3272
154
Bidding Curve for area 4: In Area 2 the Genco 14 bids for 0.65 pu power at Rs 1000/- ,
Genco 4 bids for 0.4 pu power at Rs 2100/- , Genco 8 bids for 0.1 pu power at Rs 2800/- ,
Genco 10 bids for 0.8 pu power at Rs 3200/- and the Genco 15 bids for 4.54 pu power at Rs
3600/- . The supply and demand curve intersects at 2700 Rs/MWh which is MCP as shown in
figure 5. The interchange of active power between the Control areas is given in Table III.
Table III Interchange of power between Areas
Area Power by Gencos Pool demand Power injection to system Pool drawl from other area (s)
1 3.15pu 5.5751pu 0pu 2.4251pu
2 41.45pu 15.6228pu 25.82pu 0pu
3 2.39pu 10.783pu 0pu 8.39pu
4 6.49pu 19.7003pu 0pu 13.2103pu
The LMP for inter area transactions are obtained as shown in Figure 6. The energy and
money flow is summarized in Table IV. The load flow study [9] is performed to find out the
power flow in each transmission line to confirm the schedule of bidding. In this case it is
obtained that the line flow of line no 71 connected b/w bus no. 26 and 41 is 4.2816 pu where as
its rating is 4.15pu. Therefore this line causes congestion in the system. The congestion may be
removed by rescheduling of Generation.

IJEEI ISSN: 2089-3272 
155
Figure 6. Power Transaction from area 2 to 1, 3 and 4
4. Rescheduling of Generation
As the line no. 71, in this problem, causes congestion schedules are not confirmed by
the Power Exchange (PX) so bids are re-invited for rescheduling the generation. Gencos may
come with incremental and decremental congestion bids. The selection of sensitive generators
which may relieve the congestion by re-scheduling their generation is on the basis of their
power transmission congestion distribution factors (PTCDF) [3] can be calculated as
PTCDF =
∆
∆
Where PTCDF represents the real power flow sensitivities of line “n” with respect to real power
injection at bus ‘i’ and drawl at bus ‘j’ and termed as real power transmission congestion
distribution factor.
Objective function is chosen as minimization of the total congestion cost, CC, subjected
to various operating constraints. Mathematically, the objective function can be
Min CC=∑ c , ∆P ,
,
+ ∑ c , ∆P ,
,
The constraints are as follows:
∆Pij+ P 2 Q 2 S )2
The above equation can be written as:
∑ PTCDF ∆P + P 2 Q )2 S )2
k=1,2,. Nl
∆P ∆P ∆P i=1, 2,……. Nb
∑ ∆P ∆P 0
Where ∆P is change in the total real power transmission loss in the system. Depending upon
PTCDF some of the Gencos participates in rescheduling. Let the G-1 bids to increase its power
by a maximum of 3 pu at a bid price 4000 Rs/MWh while it offers to reduce it by -7.1 pu at a
price of 1000 Rs/MWh. The bidding prices for Gencos G-12, G-13 and G-14 are given in Table
IV.
Table IV. Re-scheduled Bids
Gencos
(Rs/MWh) (Rs/MWh)
∆Pgmin
(pu)
∆Pgmax
(pu)
G-1 4000 1000 -7.1 3
G-12 4000 2000 -10 4
G-13 5200 2000 -9 3
G-14 4000 900 - 0.65 0.5

 ISSN: 2089-3272
156
Table V. Summary of Energy and Money Flow for active Power Bidding
Area S. No. Gencos Power
with their
MCPs (Rs.)
Balance Power
from other
Areas
Total Amount
paid to Gencos
of (Rs.)
Power (pu)
received by
Discos
Fund to
Be collected from
Discos (Rs.)
Area 1 G6 1.05pu@ 2700/- 2.4251pu@5000/
-
(from area 2)
2063050 D1-
1.1363@3700.47/-
420431
G5 1.5pu@2700/- D2-
4.4388@3700.47/-
1642060
G7 0.6pu@2700/-
Area 2 G1 7.1pu@3800/- 2.4251pu@5000/
- to area 1
5936664 D3-
2.4973@3800/-
948974
G2 2.30pu@3800/- 8.393pu@5000/-
to area 3
G12 18pu@3800/- 13.2102pu@500
0/- to area 4
Area 3 G3 1.4pu@3000/- 8.393pu@5000/- 4913500 D5-
5.4463@4556.24/-
2481468
G11 0.99pu@3000/- (from area 2) D6-
5.3367@4556.24/-
2431532
Area 4 G14 0.65pu@3600/- 13.2102pu@500
0/
(from area 2)
8941550 D7-
5.9501@4538.79/-
2700625
G4 0.4pu@3600/-
G8 0.1pu@3600/-
G10 0.8pu@3600/- D8-
13.750@4538.79/-
6240925
G15 4.54pu@3600/-
Total cost 2,18,54,764 2,18,54,764
Then for this case the optimization problem is formulated as follows:
Min CC= 4000*∆P , + 4000*∆P , + 5200*∆P , + 4000 ∗ ∆P , 1000 ∗ ∆P , 2000 ∗ ∆P ,
2000 ∗ ∆P , - 900*∆P ,
This optimization problem can be formulated using the GAMS solver [10] and congestion cost
comes out to be Rs.298560/. The Energy and Money Flow for active power bidding is shown in
table V
5. Optimal Location of UPFC
The UPFC consists of a shunt (exciting) and a series (boosting) transformers [11].
Converter-1 is primarily used to provide the real power demand of converter- 2 at the common
DC link terminal from the AC power system and can also generate or absorb reactive power,
similar to the Static Compensator (STATCOM), at its AC terminal.
Converter-2 is used to generate a voltage source at the fundamental frequency with
variable amplitude and phase angle, which is added to the AC transmission line by the series
connected boosting transformer. The equivalent circuit of UPFC placed in line- k connected
between bus- i and bus- j is shown in figure 7.
Figure 7. Equivalent circuit of UPFC
Based on the basic principle of UPFC and network theory, the active and reactive
power flows in the line, from bus- i to bus- j, having UPFC can be written as

IJEEI ISSN: 2089-3272 
157
2 cos
cos sin
cos sin
(1)

2
sin
cos sin cos sin
(2)
Where = 1/ ( ) and Iq is the reactive current flowing in the shunt transformer
to improve the voltage of the shunt connected bus of UPFC.The real power and reactive power
injections at bus- i with the system loading can be written as
‫ג‬ ∑ ∈
(3)
‫ג‬ ∑ ∈
(4)
The sensitivity of system loading factor (‫,)ג‬ corresponding to the real power balance equation,
with respect to the control parameters of UPFC is defined
‫ג‬
| (5)
‫ג‬
∅ |∅ (6)
where and are the system real power loading sensitivity with respect to the series injected
voltage magnitude and the series injected phase angle of the UPFC, placed in line- k,
respectively. Using equation 1, the sensitivity factor calculated at i th bus of line- k where UPFC
is placed will be
2 cos cos sin / (7)
2 sin sin cos / (8)
Sensitivity factors for each line are calculated. From where line 26 (i=16 & j=50) found most
sensitive with value 20.124 and 26.19.
6. UPFC Model for Load Flow Studies
After selecting the location for UPFC the modeling of UPFC is important. The UPFC
circuit used to derive the steady-state model is shown [13] in Figure 8.
Figure 8. Circuit for modeling of UPFC

 ISSN: 2089-3272
158
The UPFC linearised power equations are combined with the linearised system of
equations corresponding to the rest of the network,
i.e. ∆
[∆X] is the solution vector and [ ] is the Jacobian matrix. If both nodes, i and j, are PQ-type
and the UPFC is controlling the active power, flowing from i to j, and reactive power injected at
node j, the solution vector and Jacobian matrix[12]-[13] are defined as shown in equation (11).
f X ∆P ∆P ∆Q ∆Q ∆P ∆Q ∆P ∆P (9)
∆X ∆θ ∆θ
∆V
V

∆V
V
∆θ
∆V
V
∆θ (10)
J
H
H
J
J
H
J
H H

H
H
J
J
H
J
H

N
N
L
L
N
L
H N

N
N
L
L
N
L
N

H
H
J
J
H
J
H

N
N
L
L
N
L
N

H
0
J
0
0
0
H
(11)
The series and shunt voltage parameters are adjusted by trial and error in order to achieve a
power flow solution. The rating of UPFC parameters are V =0.4882(p.u.), θ =52.76(deg),
V =0.9403 (p.u.), θ = -19.54(deg) [12].
7. Results and Discussion
The above optimization problem has been formulated using the GAMS solver [10]. The
money flow before rescheduling is shown in table V. The money flow after rescheduling through
active power bidding without UPFC is shown in figure 10
Table VI. Energy and Money flow with Active Power Rescheduling using UPFC
Area S.
No.
Gencos Power
with their MCPs
(Rs.)
Balance Power from
other Areas
Total Amount
paid to Gencos
of (Rs.)
Power (pu)
received by
Discos
Fund to Be
collected from
Discos (Rs.)
Area 1 G6 1.05pu@ 2700/- 2.4251pu@5000/-
(from area 2)
2063050 D1-1.1363@3700/- 420431
G5 1.5pu@2700/- D2-
4.4388@3700/-
1642060
G7 0.6pu@2700/-
Area 2 G1 7.1pu@3800/-
+0.8336@4000/-
2.4251pu@5000/- to
area 1
6033875 D3-
2.4973@3862/-
964513
G2 2.30pu@3800/- 8.393pu@5000/- to
area 3G9 5.05pu@3800/- D4-
13.1255@3862/-
50869362
G12 18pu@3800/-
-1.4336@2000/-
12.7102pu@5000/-
to area 4
G13 9pu@3800/-
Area 3 G3 1.4pu@3000/- 8.393pu@5000/- 4913500 D5-5.4463@4556/- 2481331
G11 0.99pu@3000/- (from area 2) D6-5.3367@4556/- 2431400
Area 4 G14 0.65pu@3600/-
+0.5@4000/-
12.7102pu@5000/
(from area 2)
8891500 D7-
5.9501@4513/-
2685280
G4 0.4pu@3600/-
G8 0.1pu@3600/-
G10 0.8pu@3600/- D8-
13.750@4513/-
6205375
G15 4.54pu@3600/-
Total
cost
2,19,01,925 2,19,01,925

IJEEI ISSN: 2089-3272 
159
The change in real power output of generators G-1, G-12, G-14 with and without UPFC
is given in Table VII. The CC after implementation of UPFC reduces to Rs. 246680/-. The total
amount paid to Gencos & funds collected from Discos (revised) after rescheduling with UPFC is
shown in Table VI.
Table VII. Change in P- Generation (pu) for the 75-bus system for Active Power Bidding
with/without UPFC
Pg -1 Pg -12 Pg -14
Without UPFC 1.0919 -1.6911 0.5
With UPFC 0.8336 -1.4336 0.5
Thus UPFC is highly effective in reducing the congestion cost. After placing UPFC the
line flows at the base and also obtained after the congestion management along with their
ratings are given in figure 9.
Figure 9. Line Flows for active power bidding with UPFC
The comparison of amount paid to Gencos & funds collected from Discos without
rescheduling and with active power rescheduling with & without UPFC is shown in figure 10.
Figure 10. Total money flow before rescheduling and With & without UPFC after rescheduling
through Active Power Bidding
0
5
10
15
20
25
1 7 131925313743495561677379859197
Line Limit
Line Flow after Rescheduling
Line Flow Before rescheduling
21854764
21901925
21922036
21820000
21840000
21860000
21880000
21900000
21920000
21940000
Before
Rescheduling
but Congested
After
Rescheduling
with UPFC
After
Rescheduling
without UPFC

 ISSN: 2089-3272
160
8. Conclusion
In this paper MCP and LMP are calculated for Pool based transaction. The Congestion
so obtained is addressed by the real power rescheduling bids of generators. A suitable objective
function is chosen for the congestion cost. Using GAMS solver the change in generations of
Gencos is calculated. The revised rates for MCP and LMP are calculated. It is obtained that the
congestion is relieved in problem under study. The UPFC is placed at an optimal location using
real power sensitivity indices. and the effect of placing UPFC at an appropriate location reduces
Congestion Cost.
NOMENCLATURE:
Nl Number of Lines in the system,
Nb Number of Buses in the system,
Pg Power generation in each area,
Pd Power demand in each area,
P Base case real power flow,
Q Reactive power flow at normal operation,
CC Congestion Cost,
Ng,up Number of participants for incremental-bid congestion,
Ng,dn Number of participants for decremental-bid congestion,
c , Incremental congestion bid of r generator, ∆P , Increase in the real power output
of                 r generator,
c , Decremental congestion bid of s generator, ∆P ,    Reduction in real power output of
s generator,
∆P , Reduction in power consumption by a tth
customer,
∆Q , Adjustment in reactive power output of vth
generator,
C , (∆Q ,     Reactive Bid Function
C , Load Curtailment Bid
∆P Changes in the total real power transmission loss,
S Line flow limit,
TCDFs Transmission Congestion Distribution Factors,
UPFC Unified Power flow Controller
References
[1] SK Gupta, Richa Bansal. “ATC in Competitive Electricity Market Using TCSC”. International Journal of
Electrical, Electronic Science and Engineering. 2014; 8(2).
[2] A Kumar, SC Shrivastava, SN Singh. "A Zonal Congestion Management Approach Using Real and
Reactive Power Rescheduling". IEEE Transaction on Power Systems. 2004; 19(01).
[3] A Kumar, SC Shrivastava and Himanshu Kumar Singh. “Sensitivity Based Approach for Transmission
Congestion Management Utilizing Bids for Generation Rescheduling and Load Curtailment”.
International Journal of Emerging Electric Power Systems. 2006; 2.
[4] NS Modi, BR Parekh. “Transmission Network Congestion in Deregulated Wholesale Electricity
Market”. IMECS. 2009; 2.
[5] A Kumar, Ram Kumar Mittapalli. “Congestion management With Generic Load Model in Hybrid
Electricity Markets with FACTS devices”. Electric Power and Energy System. 2013
[6] Kumar A and Chanana S. “Power Flow Contribution Factors based Congestion Management with
Real and Reactive Power Bids in Competitive Electricity Markets”. Electric Power and Energy System.
2008.
[7] SK Gupta, R Bansal. “TCDF Based Congestion Based Management Using TCSC”. Fifth IEEE Power
India Conference. 2012.
[8] SK Gupta. Power System Engineering. Umesh Publication First Edition. 2009: 650.
[9] Haadi Sadat. Power System Analysis. 2
nd
ed., TMH Edition. 2002: 232-239.
[10] Brooke A, Kendrick D, Meeraus A, Raman R, and Rasentha RE. A User’s Guide, GAMS Software,
GAMS Development Corporation, 1998.
[11] SN Singh and I Erlich. “Locating Unified power Flow Controller for enhancing power system
Loadability”. IEEE Trans. on Power Systems. 2011.
[12] CR Fuerte-Esquivel, E Acha and H Ambriz-Perez. “A comprehensive Newton- Raphson UPFC model
for the quadratic power flow solution of practical power networks”. IEEE trans. on Power Systems.
2000; 15(1).
[13] H Farahmand, M Rashidinejad, AA Gharaveisi and GA Shahriary. “Optimal location of UPFC for ATC
enhancement in Restructured power systems”. IEEE Trans. on Power Systems. 2007.

Power Trading and Congestion Management Through Real Power Rescheduling Using Unified Power Flow Controller

More Related Content

What's hot (16)

Similar to Power Trading and Congestion Management Through Real Power Rescheduling Using Unified Power Flow Controller (20)

More from ijeei-iaes (20)

Recently uploaded (20)

Power Trading and Congestion Management Through Real Power Rescheduling Using Unified Power Flow Controller