Chapter 4

Solution methods of Economic

Dispatch and Optimal Power Flow
 Contents
• Economic Dispatch Problem
• Solution Methods of Economic Dispatch
• Base Point and Participation Factors
• Solution of Optimal Power Flow
• Linear Sensitivity Analysis
• Security Constrained Optimal Power Flow
 1. Economic Dispatch Problem
• Problem statement
– Classical ED
• Determine the power that must be generated by given
n generators so that the total cost of generation is
minimum subject to the constraints of power demand
 1.1 input-output characteristics
• To determine the total cost of generating a
given power, the input output characteristics
has to be known
• Thermal
 For hydroelectric
• Input output is linear
– F=aP+b
 Example: determining equation of input-output

• Determine the coefficients a,b, c for the

generator

Using simulated annealing

0.0017
0.0673
0.0207
2. Solution Methods of Economic Dispatch

• Equal incremental cost method

• Problem statement
• Minimize

• Subject to
 Algorithm of equal incremental cost
• Neglect the min max inequality and determine
the loading based on equal incremental rate
• Check the power output for each
• If any limit is violated, use

• Handle the violated units as neglected

• Recompute the load dispatch with
Example: suppose the output characteristics of two units is given
by the following equation. The constraints are also given

• Solution: the equal incremental cost method

gives us

• Which violates the constraint, rearranging

 Example 2- equal incremental cost

• For the following two generators, determine

the equal incremental cost operating point
when delivering 500MW

• solution
 Lagrangian Method

• An easy and analytical method is to use the

method of Lagrangian
• Procedure
– Form the Lagrangian equation by taking a
Lagrangian variable

– set the partial derivatives to zero

 Necessary condition
• Incremental cost has to be equal to the
Lagrangian variable
 Security constrained ED
• Economic dispatch has to consider two other
constraints
– Transmission line carrying capacity
– Loss on TL
– Such ED is complex and difficult to solve
• Problem statement
 Solution methods
• Change to linear programming problem by
Taylor series approximation
• Use piecewise linear approximation
• Quadratic programming method
• Gradient descent method
• GA
4. Solution of Optimal Power Flow
• Power flow problem
– Is network solution that shows current, voltage,
real and reactive power flows at every bus in a
system
– It is determining the electrical response of the
transmission system for a given load and
generated powers
– Result is used to decide line and transformer
loading and acceptability of bus voltages
 Conditions considered
• Various system loading
• When equipments fail or with outage
• Addition of new generators
• Addition of new TL
• Interconnection with other system
• Load growth studies
 Problem formulation
• Consider the following 3 bus system

• Problem is to determine bus voltages, branch

currents and real and reactive power flows
through the various branches
 Problem formulation cont…

• The branch currents and bus voltages are

related by
[I]=[Y][V] ………………………….(1)
[I] is branch current matrix
[Y] is the admittance matrix
[V] is bus voltage
• The admittance matrix can be determined
from the network by
 Problem formulation
• Writing the current equations
I1=(V1-V3)Ya+(V1-V2)Yb+V1Y1
I2=(V2-V3)Yc+(V2-V1)Yb+V2Y2
I3=(V3-V1)Ya+(V3-V2)Yc+V3Y3

• Then write the currents in terms of net input

power and voltage at the bus
S P  jQ P  jQ j
I   e  YV
V V V
 Problem formulation cont…
• If we ignore the reactive power
P1
I1   Y11 V1  Y12V2  Y13V3
V1

I2 
P2
 Y21V1  Y22V2  Y23V3 (2)
V2
P3
I3   Y31V1  Y32V2  Y33V3
V3

• To solve this, rearrange the equations as

1  P1 
V1    Y12 V2  Y13V3 
Y11  V1 
1  P2 
V2    Y21V1  Y23V3 
Y22  V2 
1  P3 
V3    Y31V1  Y32 V2 
Y33  V3 
 Types of buses in PF analysis
• The following types of buses and
corresponding quantities are assumed in PF
problem
Simple DC example
• Formulate the power flow equation for the
given network and solve them using numerical
methods

• The bus admittance matrix of the network is

 Cont…
• Hence the currents in terms of the bus
admittance matrix is

• Write the currents in terms of injected powers

 Example cont..
• Equating the two equations
 1 .5
I2   4(1.0)  9E 2  5E 3
E2
1.2
I3   10(1.0)  5E 2  15E 3
E3
P1  E1I1  14  4E 2  10E 3

• The three unknowns to be calculated are P1,

E2 and E3
 Solution methods
• Gauss-Seidel Iteration method
– Steps in GS method
• Assume initial value of unknown voltages
• Calculate for the unknown bus using the assumed
values, each time using newly calculated values
• Repeat the process until the change in voltages is
below a given minimum
 Example of GS method
• Solve for the voltages E2, E1 and Power P1 for
the following flow problem
 1 .5
I2   4(1.0)  9E 2  5E 3
E2
1.2
I3   10(1.0)  5E 2  15E 3
E3
P1  E1I1  14  4E 2  10E 3

• Let E2=E3=1, initial assumption. Rewriting

E2new = 10.2/9=1.133
 Example GS method
• For E3

E3=(8.5+5*1.133)/15=0.9443

• For P1
P1=14-4*1.133-10*0.9443=0.025
 Newton-Raphson method
• Considering eq(2) I1 
P1
V1
 Y11 V1  Y12V2  Y13V3

P2
I2   Y21V1  Y22V2  Y23V3
V2
P3
I3   Y31V1  Y32V2  Y33V3
V3

• We can rewrite it as
P1  V1 Y11 V1  Y12 V2  Y13V3 
P2  V2 Y21V1  Y22 V2  Y23V3 
P3  V3 Y31V1  Y32V2  Y33V3 
 NR method cont…
• In a Newton Raphson method, new solution is
generated according to the following rule

• Where is nonlinear function to be solved

• For a system of nonlinear equations with n
variables, the iteration will be

• Where J is the Jacobean matrix

 NR method cont…

• Hence the the Jacobean in the PF equation is

given by

• Which gives

V  V  J 1P
The power flow equation with P and Q
• When considering the reactive power, the PF
problem can be formulated in two ways
• Using Polar coordinates
• Using Rectangular coordinates
 PF in polar coordinates
• Consider the current equation in matrix form
I=YV
• The same equation when written considering the
net power injection in the buses
P  jQ
I  YV
V

• In the polar form of PF, each voltage is written in

polar form as
Flow chart of NR method of PF
 Polar form of PF
• Using the last two equations, the net power
injection at the buses is given by

• Where Gij is the conductance of line ij and Bij

is the suceptance of line ij
 PF in Rectangular coordinates
• In rectangular form, The voltage is
represented as

• Hence the power flow equation becomes

Example: Three bus system

• For the following system, solve the power flow

problem using NR method

• Bus admittance matrix is

 15  5  10
Y    5 15  10
 10  10 20 
• Bus power relations are

• The Jacobian matrix is

• Since V1= is reference, V1=0, and the

Jacobean matrix reduces to
• The elements of the Jacobian matrix are then

• Taking V2=V3=1.0 and using the Y matrix, the

Jacobian matrix becomes

P2=V2(Y21V1+Y22V2+Y23V3)-P2old=-1.1
P3=V3(Y31V1+Y32V2+Y33V3)-P3old=1.3
• Then the first iteration of the NR method
becomes
V2 n  V2 o  1 P3 
 V    V   J P 
 3 n   3o   3

• MATLAB program for this will be

– Write user defined function newtonraph
– Call using main function pfnewtonraph
 5. Security Constrained Optimal Power Flow

• OPF Is an extended power flow problem with objectives

of
– Minimum total generation cost, system loss, bus voltage
deviation, emission of generating units, number of control
actions, and load shedding
– Fulfill power flow equations, security and equipment
operating limits
• Control variables are
– generator real power output and voltages
– transformer tap settings
– phase shifters
– switched capacitors,
– reactors
 Problem formulation

• Subject to
 Problem formulation….
• The first two constraints are PF equations
• Using the rectangular form of PF, the problem
formulation becomes

• Subject to
Problem formulation cont…
 Example
• Formulate the OPF problem for the following
DC network to minimize loss