4th Year Lectures

EPM 401A
Electrical Power Systems 2A
(Economic Dispatch)
Dr. Mostafa Elshahed

Electrical Power Dept., Faculty of

Engineering, Cairo University, Egypt
 Economic Operation of Power Systems

Optimal System Operation

Economic Dispatch
Unit Commitment

By Dr. Mostafa Elshahed

 Power System Analysis - Stevenson, Grainger, Ch13.
 Electrical Power Systems, Das, Ch 16.

2
 Economic Dispatch

 Given load

L
 Given set of units on-line A B C

 How much should each unit generate to meet this load at

minimum cost?

3
 Generating Unit Cost Curves
I/O curve of a generating unit:

 The input to the thermal plant is generally measured in Btu/hr, and

the output is measured in MW.

 Obtained experimentally. Input energy rate

BTU/hr
Kcal/hr

P MW
4
 Generating Unit Cost Curves

Fuel-cost curve of a generating unit:

 The BTU can be converted to money $ and thus this curve is a
representation of the operating cost of the unit.
 In most of units, the fuel cost can be represented as 2nd order
continuous quadratic equation as: F = aP2+ b P + c

where a, b, and c are the constants of the Fuel-Cost

equation. $

 This is a Smooth Cost Function.

P MW 5
 Generating Unit Cost Curves

 The upper limit is set by thermal limits on the turbine generator unit.

 The lower limit usually comes from thermodynamic and/or practical

considerations, e.g. the fuel burning rate has to be above a certain
value or the flame goes out.

Fuel-Cost
$

P MW
P min P max
6
 Generating Unit Cost Curves
The heat rate curve: is the I/O curve divided by MW.

The most
efficient load

The unit efficiency: is the reciprocal of the heat rate.

7
 Incremental cost curve
 The incremental fuel-cost curve (the incremental $/MWh as a
function of MW) is first derivative of the fuel-cost curve versus
the real power.
If F = aP2+ bP + c $/hr, Incremental cost = 2aP+b $/MWhr

 The additional cost per

hour to increase the
output power of the unit
by 1 MW

 The saving in cost per

hour for reducing the
output power of the unit
by 1 MW.
8
 Incremental cost curve

- For 2nd order quadratic cost equation, the incremental cost curve
will be linear.
- If the cost curve is cubic or higher order, the incremental cost
curve will not be linear.
- The incremental cost curve is usually approximated by a straight
line.

Incremental
Fuel cost

$/MWhr

P MW 9
 Curve Fitting: Least Square Method
The method of least squares is used to model numerical
data obtained from measurements by adjusting the
parameters of a model so as to get an optimal fit of the data.
The best fit is that instance of the model for which the sum of
squared residuals has its least value, a residual being the
difference between an measured value and the value given
by the model.

aPi + bPi + c
2

10
 Least Square Method
n
 e = ∑ [aPi + bPi + c − F ( Pi )]2
2

i =1

To get the sum, e, of squared residuals is a minimum:

∂e 
n n n n

∑ ∑ ∑ ∑
n
= ∑ 2[aPi + bPi + c − F ( Pi )]( Pi ) = 0 + + = ⋅
2 2 4 3 2 2
a Pi b Pi c P i Pi F ( P )
i 
∂a i =1 i =1 i =1 i =1 i =1


∂e 
n n n n

= ∑ 2[aPi + bPi + c − F ( Pi )]( Pi ) = 0  ⇒∴ a ∑ Pi + b∑ Pi + c ∑ Pi = ∑ Pi ⋅ F ( Pi ) 
n
3 2
∴
2

∂b i =1  i =1 i =1 i =1 i =1 
∂e 
n n n n
= ∑ 2[aPi + bPi + c − F ( Pi )](1) = 0 
n
a ∑ Pi + b∑ Pi + ∑ c = ∑ F ( Pi )
2 2

∂c i =1 
 i =1 i =1 i =1 i =1 

Solving these three equations to get a, b, c:

 n 4 n n
  n 2 
∑ Pi ∑P ∑ ∑ i ⋅
3 2
i Pi  P F ( P )
i 
 i =n1 i =1 i =1
 a   n i =1

 
n n
 P3 Pi  b  =  ∑ Pi ⋅ F ( Pi ) 
∑ ∑P ∑
2

i =1
i
i =1
i
i =1
  i =1 
 n 2 n  
 c   n 
∑ Pi ∑P n   ∑ F ( Pi ) 
 i =1   i =1 
i
i =1
11
Ex.: Least Square Method
• The data for the input / output curve for the generating unit in a thermal power
station :
MW 70 75 112.5 150
MBTU/hr 574 611 896 1190
• Then using Least Square Method to fit these data into:
F ( Pi ) = aPi + bPi + c
2

Solution: Calculate all quantities required as shown:

Pi Pi 2 Pi 3 Pi 4 F ( Pi ) Pi ⋅ F (P) Pi 2 ⋅ F ( Pi )
70 574
75 611
112.5 896
150 1190
407.5 45681.25 5563703.125 722081289.1 3271 365305 44364475

12
 Ex.: Least Square Method

Pi Pi 2 Pi 3 Pi 4 F ( Pi ) Pi ⋅ F (P) Pi 2 ⋅ F ( Pi )
70 574
75 611
112.5 896
150 1190
407.5 45681.25 5563703.125 722081289.1 3271 365305 44364475
 n 4 n n
  n 2 
∑ Pi ∑P ∑ ∑ i ⋅
3 2
i Pi  P F ( P )
i 
 i =n1 i =1 i =1
 a   n i =1

Pi  b  =  ∑ Pi ⋅ F ( Pi ) 
n n
 P3
∑ ∑P ∑
2

i =1
i
i =1
i
i =1
  i =1 
 n 2 n   c   n 
∑ Pi ∑P n   ∑ F ( Pi ) 
 i =1   i =1 
i
i =1

 a=3.28×10-3 , b=6.9794, c=69.2251

13
Economic Dispatch: Formulation
Initially we'll ignore generator limits and the losses:
We are now in a position to formulate our optimization problem:
G1

 
n P1

min  F = ∑ Fi 
G2
P2
 i =1  n
Subject to Equality
Constraint: ∑P = P
i =1
i D Gn
Pn PD

This optimizing problem with equality constraint can be solved by

the classical method of Lagrange multiplier.
In this method, the cost function is replaced by an augmented
Lagrangian function.

14
Economic Dispatch: Formulation

Now this Lagrangian objective function has to be

minimized with no constraints on it as follows:
n
 n  λ: Lagrange multiplier
ℑ = ∑ Fi − λ ∑ Pi − PD 
n: No. of units
i =1  i =1 
𝐹𝐹𝑖𝑖 : Cost of unit i
ℑ = function( P1 , P2 ,.......Pn , λ ) 𝑃𝑃𝑖𝑖 : Power generated from
unit i
∂ℑ ∂F1  𝑃𝑃𝐷𝐷 =Total demand power
= −λ = 0
∂P1 ∂P1
 (1) n equations
∂ℑ ∂F2  ∂Fi i=1,2…n
= − λ = 0 ⇒ =λ
∂P2 ∂P2  ∂Pi
 
∂ℑ  n

∑P −P
one equation
=0 i D =0 (2)
∂λ i =1 15
Formulation of Lagrange Multiplier Method

 Equation (1) is called Coordination Equation, it says that at the minimum cost
operating point, the incremental cost for all the generating units must be equal.

 This condition is known as Equal Incremental Cost Rule.

 Equation (2) is known as Power Balance Equation.

16
Concept of Economical Division of Plant Load

 Two Units, initially, IC2 > IC1

 Decrease the output power of unit 2 by ΔP and increase
output power in unit 1 by ΔP.
 This change can be continued until IC1 = IC2 at which the
total cost will be minimum.

 Thus the total cost will be

minimum when the
Incremental costs are
Equal.

17
 Ex : 1

A power plant has 3 units with the following cost characteristics.

F1 = 0.5 P1 + 215 P1 + 5000 LE / hr

2

Find the economic scheduling

F2 = 1.0 P2 + 270 P2 + 5000 LE / hr
2
for a total load of 200 MW
F3 = 0.7 P3 + 160 P3 + 9000 LE / hr
2

SOLUTION : dF1
Knowing the cost = 1.0 P1 + 215 LE / MWhr
characteristics, incremental dP1
cost characteristics are dF2
= 2.0 P2 + 270 LE / MWhr
obtained: dP2
dF3
= 1.4 P3 + 160 LE / MWhr
dP3 18
 Ex : 1

Using the equal incremental cost rule:

1.0 P1 + 215 = λ
λ − 215 λ − 270 λ − 160
2.0 P2 + 270 = λ P1 = , P2 = , P3 =
1.0 2.0 1.4
1.4 P3 + 160 = λ

Since P1 + P2 + P3 = 200 we have:

λ − 215 λ − 270 λ − 160
+ + = 200 λ = 300 LE / MWhr
1.0 2.0 1.4

Then: P1 = 85MW , P2 = 15MW , P3 = 100 MW

19
Economic Dispatch with Inequality Constraints

 n

min  F = ∑ Fi 
 i =1  n

Equality Constraints: ∑P −P
i =1
i D =0

Inequality Constraints: Pi ,min ≤ Pi ≤ Pi ,max

In Economic Dispatch Solution:

• Only involve equality constraints, then get generation levels P1,
P2, …, Pn .
• For generation levels above maximum value or below
minimum, these generation levels are settled as constant values
at these limits.
• Economic Dispatch solution is repeated with only other variable
sets.
20
 Graphical Solution
 The units' data are simply plotted adjacent to each other.
 Then, a value for λ is chosen and the generations are added.
 If the total generation is equal to the total load demand "PD" then the
optimal solution has been found.
 Otherwise, a new value for λ is chosen and the process is repeated.
 The limitations of each unit are included as vertical lines since the
rules must not include generation beyond unit capabilities. The unit
is simply fixed at the value crossed.

21
 Ex : 2
A power plant has 3 units with the following cost characteristics.
The maximum and minimum loads allowable on each unit are
150 and 39 MW.
F1 = 0.5 P1 + 215 P1 + 5000 LE / hr
2

Find the economic scheduling

F2 = 1.0 P2 + 270 P2 + 5000 LE / hr
2
for a total load of 200 MW
F3 = 0.7 P3 + 160 P3 + 9000 LE / hr
2

SOLUTION : dF1
Knowing the cost = 1.0 P1 + 215 LE / MWhr
characteristics, incremental dP1
cost characteristics are dF2
= 2.0 P2 + 270 LE / MWhr
obtained: dP2
dF3
= 1.4 P3 + 160 LE / MWhr
dP3 22
 Ex : 2

Using the equal incremental cost rule:

1.0 P1 + 215 = λ
λ − 215 λ − 270 λ − 160
2.0 P2 + 270 = λ P1 = , P2 = , P3 =
1.0 2.0 1.4
1.4 P3 + 160 = λ
Since P1 + P2 + P3 = 200 we have:
λ − 215 λ − 270 λ − 160
+ + = 200 λ = 300 LE / MWhr
1.0 2.0 1.4
Then: P1 = 85MW , P2 = 15MW , P3 = 100 MW
As P2 < P2 min, therefore P2 is set at the min value of 39 MW.
Then P1 + P3 = 200 − 39 = 161 MW (this power has to be
scheduled between only units 1 and 3).
λ − 215 λ − 160 λ = 286 LE / MWhr
+ = 161
1.0 1.4
Then: P1 = 71MW , P3 = 90 MW (within the limits). 23
 Ex : 3
There is a simple two units system that have the following
incremental fuel cost function:
dF1 dF2
= 0.008 P1 + 8, = 0.0096 P2 + 6.4
dP1 dP2
Assume that both units are operating at all times, that total load
varies from 250 to 1250 MW, and that maximum and minimum
loads on each unit are to be 625 and 100 MW, respectively.

Find:
The incremental fuel cost and the allocation of load between units.
Determine:
The saving in fuel cost for a total load of 900 MW distributed
economically compared with equal distribution of the same total
load
24
 Ex : 3
P1 MW λ1 P2 MW λ2 P1+ P2 Plant λ
100 # 8.8 100 7.36 200 Units at min
100 # 8.8 150 7.84 250 7.84
100 # 8.8 250 8.8 350 8.8
200 9.6 333 9.6 533 9.6
300 10.4 417 10.4 717 10.4
400 11.2 500 11.2 900 11.2
500 12.0 583 12.0 1083 12.0
550 12.4 625 # 12.4 1175 12.4
625 13.0 625 # 12.4 1250 13.0 Units at max
800
# indicates the output of the unit 700

Unit output MW
600
at its minimum or maximum limit 500
400
Unit 2

and the plant λ is then equal to the 300

Unit 1
200
incremental fuel cost of the unit 100

not at a limit. 200 400 600 800 1000 1200 1400

Plant output MW
25
 Ex : 3
P1 MW λ1 P2 MW λ2 P1+ P2 Plant λ
100 # 8.8 100 7.36 200 Units at min
100 # 8.8 150 7.84 250 7.84
100 # 8.8 250 8.8 350 8.8
200 9.6 333 9.6 533 9.6
300 10.4 417 10.4 717 10.4
400 11.2 500 11.2 900 11.2
500 12.0 583 12.0 1083 12.0
550 12.4 625 # 12.4 1175 12.4
625 13.0 625 # 12.4 1250 13.0 Units at max

# indicates the output of the unit

at its minimum or maximum limit
and the plant λ is then equal to the
incremental fuel cost of the unit
not at a limit.
26
 Ex : 3
For a total of 900 MW, economically distributed: From last table, or:
dF1 dF2
=
dP1 dP2
∴ 0.008 P1 + 8 = 0.0096 P2 + 6.4
and P1 + P2 = 900 MW
∴ 0.008 P1 + 8 = 0.0096(900 − P1 ) + 6.4
∴ P1 = 400 MW and P2 = 500 MW

For equal distribution of, each unit supply 450 MW:

The change in cost of unit 1:
450 450
dF1 450
∫400 dP1 1 400∫
= + = + = $570 per hour
2
dP ( 0.008 P1 8) dP1 0 .004 P1 8 P1
400

The change in cost of unit 2:

450 450
dF2 450
∫500 dP2 dP2 = 500∫ (0.0096 P2 + 6.4)dP2 = 0.0048P2 + 6.4 P2 500 = −$548 per hour
2

The net increase in cost = 570-548 = $22 per hour (during a year: $ 192,720)
27
 Inclusion of Transmission Losses

 The optimum operation for different plants situated close to

each other can be done without considering losses in
transmission system.
 If there are different plants located at different places, the
losses on the transmission system can’t be ignored (≈ 5 to 15%
load) and shall be met.
 Transmission loss depends on i) line parameters ii) bus
voltages and iii) power flow.
 This impact on losses should be included when doing the
economic dispatch.
 Determination of transmission loss requires complex
computations. However, with reasonable approximations,
losses can be included by slightly rewriting the Lagrangian
function.
28
 Inclusion of Transmission Losses
G1 G2
 G1 and G2 are similar
plants (same type, same
size, same manufacture, P1 P2
and same fuel cost).
 The load PD is met by the PD
two generators:
 If Equal Incremental Cost Rule is applied,
Then P1+ P2= PD/2
This is not economical solution, as the output power of G1
makes some transmission losses to supply the load while the
load receives the output power of G2 directly (right close to
load) without losses.
 The right solution is to supply more power from G2 (P2> P1) for
less transmission losses.
29
 Economic Dispatch Including Losses
 n

min  F = ∑ Fi 
 i =1  n

Equality Constraints: ∑P = P
i =1
i D + Ploss Ploss = function( P1 , P2 ,.......Pn )
 n n

Lagrangian objective function: ℑ = ∑ Fi − λ ∑ Pi − PD − Ploss 
i =1  i =1 
The necessary conditions for a minimum are now:
∂ℑ dF1  ∂Ploss  
= − λ 1 − 0 −  = 0 
∂P1 dP1  ∂P1   dFi
∂ℑ dF2  ∂Ploss   dPi n equations
= − λ 1 − 0 −  = 0 ⇒ =λ
∂P2 dP2  ∂P2   1 − ∂Ploss 
   ∂Pi 

one equation n  
∂ℑ
and
∂λ
=0
i =1
∑P −P i D − Ploss = 0 Li =
1
 ∂Ploss 
1 − 
The penalty factor L (L ≥ 1) for the i generator:  ∂ Pi  30
 Economic Dispatch Including Losses
The minimum fuel cost is obtained when the incremental cost of each plant
multiplied by its penalty factor is the same for all the plants in the power system.

Thus the solution for the problem:

n n
Minimize: F = ∑ Fi subject to:
i =1
∑P
i =1
i = PD + Ploss
is obtained when the following equations are satisfied:
dFi 1
Li = λ i = 1,2,3.....n, Li =
dPi  ∂Ploss 
Eq (I) 
 1 − 
 ∂Pi  
dFi  ∂Ploss  dFi ∂Ploss
= λ
 1 − 
 → + λ =λ
dPi  ∂P i  dPi ∂P i
n

and ∑P
i =1
i = PD + Ploss → Eq (II)

- Equation (I) is called Exact Coordination Equation.

∂Ploss
- is called incremental transmission loss (ITL) of i generator.
∂Pi 31
 Loss Equation and
Loss Coefficients (B- Coefficients)

General loss formulas for two generator sets:

∴ Ploss = P1 B11 + 2 P1 P2 B12 + P2 B22

2 2

 B11 B12   P1 
∴ Ploss = [P1 P2 ]
 B21 B22   P2 
 B- Coefficients are reciprocal megawatts (dimension 1/ MW).

 The B- Coefficients are constant as P1 and P2 vary, only

insofar as bus voltages at the plants maintain constant
magnitude and plant power factor remain constant.

 In large systems, calculations are based on several sets of

loss coefficients depending on load conditions.
32
 Loss Equation and
Loss Coefficients (B- Coefficients)
Square symmetrical
For general n generator sets: matrix
Bij = B ji
 B11 B12 . . B1n   P1 
B
 21 B22 . . B2 n   P2 
Ploss = [P1 P2 . . Pn ] .  
  
 .  
 Bn1 . . . Bnn   Pn 

Or n n
∂Ploss ∂ n n n
Ploss = [P ] [B ][P ] = ∑∑ Pi Bij Pj = ∑∑ P B P = 2∑ Pj Bij
T

∂Pi ∂Pi
i ij j
i =1 j =1 i =1 j =1 j =1

Ex: For three sources:

∴ Ploss = P1 B11 + P2 B22 + P3 B33 + 2 P1 P2 B12 + 2 P2 P3 B23 + 2 P3 P1 B31
2 2 2

∂P
∴ loss = 2 P1 B11 + 2 P2 B12 + 2 P3 B31
∂P 33
 Ex : 4
A system consists of two plants connected by in TL line. The only
load is located at plant 2.
When 200 MW is transmitted from plant 1 to plant 2, power loss in
the line is 16 MW.
Find the required generation for each plant and the power
received by the load when λ is $12.5 per MWhr.

Assume the incremental fuel costs can be approximated by:

dF1
= 0.010 P1 + 8.5 $/ MWhr
dP1
dF2 P1 P2
= 0.015 P2 + 9.5 $/ MWhr
dP2

If the transmission loss is neglected in economic dispatch

calculations, comment on achieved results. 34
 Ex : 4
For two plants system: ∴ Ploss = B P
11 1
2
+ 2 B P P
12 1 2 + B P
22 2
2

Since all the load is at plant 2, varying P2 can’t affect loss power.
∴ B22 = 0, B12 = 0 ∴ Ploss = B11 P1
2

When P1=200 MW, Ploss=16 MW, So:

∴ Ploss = 16 = B11 ( 200) 2 ⇒ B11 = 0.0004 MW −1
∂Ploss 1
∴ = 2 P1 B11 + 2 P2 B12 = 0.0008P1 , and then L1 =
∂P1 1 − 0.0008 P1
∂Ploss
∴ = 2 P2 B22 + 2 P1 B12 = 0, and then L2 = 1
∂P2
For economic dispatch:
dF dF 1
∴ 1 L1 = 2 L2 = λ ⇒ (0.010 P1 + 8.5) = (0.015 P2 + 9.5) × 1 = 12.5
dP1 dP2 1 − 0.0008 P1
∴ P1 = 200 MW , P2 = 200 MW , Ploss = 0.0004 × (200) 2 = 16 MW
, Pload = P1 + P2 − Ploss = 384 MW 35
 Ex : 4
If transmission loss is neglected in economic dispatch calculations:
dF1 dF2
∴ = ⇒ (0.010 P1 + 8.5) = (0.015 P2 + 9.5)
dP1 dP2
 P1 + P2 = Pload + Ploss ⇒ P1 + P2 = 384 + 0.0004 × ( P1 ) 2
Solving these two equations gives: P1 = 290 .7 MW , P2 = 127 .1 MW
The generation of unit 1 change from 200 to 290.7 MW, the increase in cost of
unit 1:
290.7 290.7 290.7
dF1 .01 2
∫
200
dP1
dP1 = ∫200(0.010 P1 + 8.5)dP1 = 2 P1 + 8.5P1 200 = $993.48 per hour
The change in cost of unit 2:
127.1 127.1 127.1
dF2 .015 2
∫
200
dP2
dP2 = ∫
200
(0.015 P2 + 9.5) dP2 =
2
P2 + 9.2 P2
200
=
−$871.39 per hour

The net saving by accounting transmission loss in economic dispatch =

993.48 - 871.39 = $122.09 per hour
36
 Methodology of Lambda-Iterative
Solution Method for Economic Dispatch Including
Losses assuming 2nd order quadratic cost equation

= ai Pi + bi Pi + ci
2
Known the quadratic cost equation for i unit is: F
i
dFi
So, the linear incremental cost of i unit is: = 2ai Pi + bi
dPi
∂Ploss n
= 2∑ Pj Bij
And the incremental transmission loss (ITL) of i unit is: is:
∂Pi j =1
Substitute in Exact Coordination Equation: dFi ∂Ploss
+λ =λ
dPi ∂Pi
n
i.e:
(2ai Pi + bi ) + 2λ ∑ Bij Pj = λ
j =1

For any particular value of λ , the above set of linear simultaneous equations are
directly solved or iteratively for the values of Pi (by assuming initial values of Pi,
then iterations are stopped when Pi converge with accepted accuracy).

37
 Steps of Lambda-Iterative
Solution Method for Economic Dispatch Including Losses

To determine the optimum scheduling of system load between n plants, the data
required are
i) system load PD ,
ii) incremental cost characteristics of the n plants (assume 2nd order quadratic
cost equation) and
iii) loss coefficients matrix (n × n).
The iterative solution procedure (Lambda-Iteration Algorithm) is:
Step 1:
For the first iteration, choose suitable initial value of λ.
one way is to assume that the transmission losses are zero and calculate initial
value of λ.
Step 2: Solving the Exact Coordination Equation for such λ: dFi L = λ
i
These set of linear simultaneous equations may be dPi
directly solved or iteratively for the values of Pi for i=1,2…n .

n
(2ai Pi + bi ) + 2λ ∑ Bij Pj = λ
j =1
38
 Steps of Lambda-Iterative
Solution Method for Economic Dispatch Including Losses

Step 3:
Calculate transmission loss using computed Pi :
n n
Ploss = [P ] [B ][P ] = ∑∑ Pi Bij Pj
T

i =1 j =1

Step 4:
Check the Power Balance Equation by:
n
Comparing ∑P
i =1
i
with PD + Ploss to check the power balance.

If the power balance is satisfied within a specified tolerance, then the present
solution is the optimal solution; otherwise update the value of λ.

39
 Steps of Lambda-Iterative
Solution Method for Economic Dispatch Including Losses

First time updating can be:

n
Value of λ is increased by a suitable step (about 5%) if: ∑P
i =1
i < PD + Ploss

n
Value of λ is decreased by a suitable step (about 5%) if:
∑P
i =1
i > PD + Ploss

Step 5:
Return to Step 2 and continue the calculations of Steps 2, 3 and 4 until the power
balance equation is satisfied with desired accuracy.

40
 Ex : 5

A power system with two plants having incremental fuel cost as:
dF1 dF2
= 1.0 P1 + 200 $/ MWhr = 1.0 P2 + 150 $/ MWhr
dP1 dP2
Loss coefficients matrix is given by B11 B12   0.001 - 0.0005
B  =
 21 B 22  - 0.0005 0.0024 
Find the optimum scheduling for a system load of 100 MW.

Applying the iterative solution procedure (Lambda-Iteration Algorithm) as follows:

Step 1:
For the first iteration, calculating λ assuming that the transmission losses are zero.
dF1 dF2 λ − 200 λ − 150
∴ = = λ ⇒ P1 = , P2 =
dP1 dP2 1.0 1.0
 P1 + P2 = PD = 100 MW ⇒ λ = 225 $ per MWhr
41
 Ex : 5
Step 2: Solving the Exact Coordination Equation for such λ:
dFi 1 dFi ∂P ∂Ploss n
×
dPi 
1 −
∂Ploss 

=λ⇒
dPi
+ λ loss = λ
∂Pi
∂Pi
= 2 Pj Bij
j =1
∑
 ∂P i 

(1.0 P1 + 200) +λ (2 P1 B11 + 2 P2 B12 ) = λ (1.0 P1 + 200) +225(0.002 P1 - 0.001P2 ) = 225

(1.0 P2 + 150) +λ (2 P2 B22 + 2 P1 B21 ) = λ (1.0 P2 + 150) +225(0.0048P2 - 0.001P1 ) = 225
Solving these two equations in P1, P2 using Cramer’s rule:
1.45P1 - 0.225P2 = 25 1.45 - 0.225   P1  25  P1  23.227 MW 
- 0.225 2.08 *  P  = 75  P  = 38.571 MW 
- 0.225P1 + 2.08P2 = 75    2    2  
 B11 B12   P1 
∴ Ploss = [P1 P2 ] = + +
2 2
P  P1 B11 2 P P B
1 2 12 P2 B22
Step 3:  21
B B 22   2 

= 0.001× (23.227) 2 − 2 × 23.227 × 38.571× 0.0005 + 0.0024 × (38.571) 2

= 3.214 MW
Step 4:Check the Power Balance Equation:
n

∑P = P + P
i =1
i 1 2 = 23.227 + 38.571 = 61.789 MW PD + Ploss = 100 + 3.214 = 103.214 MW
42
 Ex : 5
First time updating can be:
n
Value of λ is increased by a suitable step (about 15%) as: ∑P
i =1
i < PD + Ploss
∴ λ = 1.15 × 225 = 258.75 ≈ 259 $ per MWhr
Repeat Step 2: Solving the Exact Coordination Equation for such λ:
dFi ∂P
+ λ loss = λ
dPi ∂Pi
(1.0 P1 + 200) +λ (2 P1 B11 + 2 P2 B12 ) = λ (1.0 P1 + 200) + 259(0.002 P1 - 0.001P2 ) = 259
(1.0 P2 + 150) +λ (2 P2 B22 + 2 P1 B21 ) = λ (1.0 P2 + 150) + 259(0.0048 P2 - 0.001P1 ) = 259
1.518 P1 - 0.259 P2 = 59 1.518 - 0.259   P1  59   P1  48.107 
- 0.259 2.243 *  P  = 109  P  = 54.152 
- 0.259 P1 + 2.243P2 = 109    2    2  
Step 3:  B11 B12   P1 
∴ Ploss = [P1 P2 ] = + +
2 2
P  P1 B11 2 P P B
1 2 12 P2 B22
 21
B B 22   2 

= 0.001× (48.107) 2 − 2 × 48.107 × 54.152 × 0.0005 + 0.0024 × (54.152) 2

= 6.747 MW 43
 Ex : 5

Step 4: Check the Power Balance Equation

n

∑P = P + P
i =1
i 1 2 = 48.107 + 54.152 = 102.259 MW PD + Ploss = 100 + 6.747 = 106.747 MW

n
∑ Pi ≤ PD + Ploss
i =1

Value of λ is increased by a suitable step (about 5%)

With this new value of λ, coordination equations are formed and the procedure
has to be repeated……
The optimal dispatch is:
λ =263.467 $/MWhr , P1=51.2401 MW, P2 =56.066 MW, Ploss =7.297 MW

44