International Journal of ChemTech Research

CODEN (USA): IJCRGG, ISSN: 0974-4290, ISSN(Online):2455-9555

Vol.11 No.04, pp 153-157, 2018

Steady State Optimal Power Flow

*1Ravi.C.N., 2 Lakshmi.D., 3Benisha.S, 4Zahira.R.

1
Dept of EEE, Vidya Jyothi Institute of Technology, Hyderabad -500075, India
2,3
Dept of EEE, Sree Sastha Institute of Engineering & Technology,
Chennai-600123, India.
4
Dept of EEE, Tagore Engineering College, Chennai, India

Abstract : An optimal load-flow solution gives the optimal active and reactive power dispatch
for a static power system loading condition. Computationally, it is a very demanding nonlinear
programming problem, due to the large number of variables and in particular to the much
larger number of variables and in particular to the much larger number and types of limit
constraints which define the boundaries of technical feasibility This paper solves the standard
26 bus power system network.
Keywords : optimal power flow; OPF; Lamda iteration

I. Introduction
One of most important and frequently used analysis and computational procedures for the planning,
design, and operation of an electric power system is the ac power-flow program. This program constitutes a
simulation of the steady-state ac power flows and voltages in the network under study. It is used to simulate the
flows and voltages corresponding to several future load conditions for various design alternatives. If the system-
design alternative under consideration is not capable of supplying the assumed loads then the set of values for
some of the control variables such as voltage levels or power productions; capacitors are adjusted. Much of this
time-consuming trial-and error process is reduced by the optimum power flow, and provide a set of feasible
values of the control variables. Optimum power-flow solutions may be used not only for system planning but also
for system operation, which is a real-time function. In operation, it provides the most economical operating point
that meets all the flow and voltage constraints related to power-system security and quality of service. The
standard 26 bus test system which has 6 generators, 7 transformers and 9 shunt capacitor is considered in this
work.

II. Problem formulation

The steady state Optimal Power Flow problem is a minimization problem which is stated as follows. The
objective is to minimize the generation cost.

Ravi.C.N. et al /International Journal of ChemTech Research, 2018,11(04): 153-157.

DOI : http://dx.doi.org/10.20902/IJCTR.2018.110418
Ravi.C.N et al /International Journal of ChemTech Research, 2018,11(04): 153-157. 154

ng
Min Ct   i  i Pi   i Pi 2 ` (1)
i 1

Where, Ct - Total cost function, α, β, γ are cost coefficients, Pi is the ith bus real power generation

Subject To
ng ng

 Pi  PD  PL
i 1
(2) Q
i 1
i  QD  QL (3)

Vi (min)  Vi  Vi (max) for i=1 to Nbus (4) Pi (min)  Pi  Pi (max) for i=1 to ng (5)

Qi (min)  Qi  Qi (max) for i=1 to ng (6) ti (min)  ti  ti (max) for i=1 to Ntrans (7)

Where, Pi , Qi – Real and Reactive Power generation of ith bus,PD , QD – Real and Reactive Power demand in ith
bus

PL , QL – Real and Reactive Power loss ,Vi – Voltage magnitude of the ith bus ,ti – Transformer tap
position of the ith transformer, Nbus – Number of bus, ng – number of generator Ntrans – number of transformer

Lagrange Function ng ng ng
L  Ct   ( PD  PL   Pi )   i (max) ( Pi  Pi (max) )   i (min) ( Pi  Pi (min) )
(8)
i 1 i 1 i 1

Where, λ – Lagrange multiplier, μi – Khun-tucker function multiplier

The minimum of this unconstrained function is found at the point where partials of the function to its
variables are zero

L L
0 (9) 0 (10)
Pi 

L L
0 (11) 0 (12)
i (max) i (min)

µi(max) and µi(min) are zero when Pi is within its limit

From Equation – (9),

Ct C C C
  ( t  1)  0 We know that t  i since Ct = C1+ C2+ ….. Cng
Pi Pi Pi Pi

Ci P
Therefore  L  for i=1 to ng (13)
Pi Pi

Ci P Ci P 1 C Ci

  L   (1  L ) .( i )    .Li (14)
P
Pi Pi Pi Pi (1  L ) Pi Pi
Pi

1
Where, Penalty Factor, Li 
P
(1  L )
Pi
 Ravi.C.N et al /International Journal of ChemTech Research, 2018,11(04): 153-157. 155

L ng
 2 Bij Pj  B0i (15)
Pi j 1

From equation 13,

ng
 i  2 i Pi  2  Bij Pj  B0i   
j 1

i ng
1 
(  Bii ) Pi   Bij Pj  (1  B0i  i ) (16)
 j 1 2 
j i

This equation is extended to all generating plants results in following linear equation in matrix form

 1   1 
   B11 B12 . . B1ng   P1   1  B01   
 2     
 B11  B22 B23 . B2 ng   P2   1  B02  2  (17)
   .   1   
 . . .   2  . 
 . . .  .   . 
  P    
 ng
 B  Bngng    1  B0 ng  
ng ng
Bng 2 . .

ng1
    

To find the optimal dispatch for an estimated λ the simultaneous linear equation 17 is solved to find P matrix.
Then the iteration process is continued using the gradient method. To do this, from equation 16, Pi at the kth
iteration is

k (1  B0i )   i  2k  Bij Pjk

j i
Pi k  (18)
2( i  k Bii )
Substitute equation 19 in equation 10,

ng
k (1  B0i )   i  2k  Bij Pjk
i 1 2( i  k Bii )
j i
 PD  PL (19)

This can be written as f(λ) = PD + PLk (20)

Expand the equation 21 L.H.S. using Taylor‟s series about an operating point λk and neglecting the higher order
terms

f (k ) k
f (k )    PD  PL
k
(21)


P k P k
k  k  (22)
f (k ) P

 ( i ) k
 i (1  B0i )  Bii  i  2 i  Bij Pjk
ng
Pi k ng
Where  (
i 1 
) 
i 1 2( i   Bii )
k 2
j i
(23)
Ravi.C.N et al /International Journal of ChemTech Research, 2018,11(04): 153-157. 156

ng
k 1  k  k (24) P k  PD  PLk   Pi k (25)
i 1

The process is continues until ΔPk is less than a specified accuracy

III. Test System

26 bus power system network is considered. Bus 1 is taken as slack bus its voltage adjusted to 1.025
angle 0o and P1Max = 500Mw, P1Min=100 Mw, the data for the other generator buses are

Bus Voltage QMin QMax PMin PMax

No Mag. (Mvar) (Mvar) (Mw) (Mw)
(pu)
2 1.02 40 250 50 200
3 1.02 40 150 80 300
4 1.05 40 80 50 150
5 1.04 40 160 50 200
26 1.01 15 50 50 120

Transformer Data

Between Buses Tap setting (pu)

2 -3 0.96
2-13 0.96
3-13 1.017
4-8 1.05
4-12 1.05
6-19 0.95
7-19 0.95

Shunt Capacitive Data

Buses No Mvar
1 4.0
4 2.0
5 5.0
6 2.0
9 3.0
11 1.5
12 2.0
15 0.5
19 5.0

Generators operating costs in $/h

C1 = 240 + 7.0 P1 + 0.007 P12

C2 = 200 + 10 P2 + 0.0095 P22
C3 = 220 + 8.5 P3 + 0.009 P32
C4 = 200 + 11 P4 + 0.009 P42
C5=220+10.5P5+0.008P52
C26= 190 + 12 P26 +0.0075 P262
Ravi.C.N et al /International Journal of ChemTech Research, 2018,11(04): 153-157. 157

IV. Test Result

Test result for the 26 bus power system is summarized in the following table

Base Case
Optimal Power
Description Power Savings
Flow
Flow
16760.73 15447.72 1313.01
Total Generation Cost ($/h)
Total system loss (Mw) 15.53 12.807 2.723
474.1196 447.6919
173.7886 173.1938
System Power Generation 190.9515 263.4859
by the generators (Mw) 150.0000 138.8142
196.7196 165.5884
103.5772 87.0260
Incremental cost of
delivered power (λ) 13.911780 13.538113
($/MWh)

V. Conclusion

The optimum power flow has been defined and its advantages over the ordinary power flow have been
shown to be greatly reduced trial and error. It has been shown that Newton's method of power flow solution can
be extended to yield an optimal power flow solution that is feasible with respect to all relevant inequality
constraints. The main features of the method are a gradient procedure for finding the optimum and the use of
penalty functions to handle functional inequality constraints.

References
1. H. W. Dommel & W. F. Tinney, "Optimal power flow solutions", IEEE Trans. (Power App. & Syst.),
Vol. PAS-87, October, 1968 ,pp. 1866-1876.
2. Allen J. Wood and Bruce F. Wollenberg, Power Generation, Operation, and Control, John Wiley &
Sons, Inc.
3. Kundur P, „Power System Stability and Control’ , Mc Graw Hill Publication,
4. O. Alsac and B. Stott, “Optimal Load Flow with Steaty State Security”, IEEE power system engineering
conference
5. Ravi. C N, Christober Asir Rajan C Ravi, C. N. "Optimal power flow solutions using constraint Genetic
Algorithm." National Journal on Advances in Computing and Management 3.1 (2012).
6. Ghaddar, Bissan, Jakub Marecek, and Martin Mevissen. "Optimal power flow as a polynomial
optimization problem." IEEE Transactions on Power Systems 31.1 (2016): 539-546.
7. Bazrafshan, Mohammadhafez, and Nikolaos Gatsis. "Decentralized stochastic optimal power flow in
radial networks with distributed generation." IEEE Transactions on Smart Grid 8.2 (2017): 787-801.

*****