Power Control ED
Power Control ED
=
=
m
i
Gi i T
P C C
1
( ) ( ) 0 , , ,
3 2
1 1
= =
= =
Gm G G loss
n
i
Di
m
i
Gi Gi
P P P P P P P f
max min
Gi Gi Gi
P P P s s
max
ij ij
P P s
max min
i i i
V V V s s
Assume bus 1 is slack bus, only m-1
independent power variable
Neglect line flow limit
in this analysis
Neglect voltage limit due
to weak coupling effect
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
5
ED Simplified Formulation
To minimize cost of operation,
Subject to:
Power flow equation
Generation limit
( )
=
=
m
i
Gi i T
P C C
1
( ) ( ) 0 , , ,
3 2
1 1
= =
= =
Gm G G loss
n
i
Di
m
i
Gi Gi
P P P P P P P f
max min
Gi Gi Gi
P P P s s
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
6
Solution to ED
The problem can categorized into nonlinear
optimization problem.
Lagrangian of the problem is,
is called Lagrange multiplier.
Optimality condition:
( )
( ) 0 . .
min
= x g t s
x f
( ) ( ) ( ) x g x f x L = , min
( )
( )
0
,
0
,
=
c
c
=
c
c
x L
x
x L
* *, x
Optimal
Solution
\
|
=
=
D
m
i
Gi T T
P P C C
1
~
( )
0 0
~
, , 2 , 1 , 0 0
~
1
= =
c
c
= =
c
c
=
c
c
=
D
m
i
Gi
T
Gi
Gi i
Gi
T
P P
C
m i
P
P C
P
C
( )
m i
P
P C
IC
Gi
Gi i
i
, , 2 , 1 = =
c
c
=
( )
=
=
m
i
Gi i T
P C C
1
0
1
=
=
D
m
i
Gi
P P
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
9
Case 1: Alternative Solution Procedure
minimize
Subject to:
From this constraint, we can reduce number of
variable by substitution to the objective function,
The problem becomes unconstrained problem,
to minimize
( )
=
=
m
i
Gi i T
P C C
1
0
1
=
=
D
m
i
Gi
P P
1 2 1
=
Gm G G D Gm
P P P P P
( ) ( ) ( ) ( )
1 2 1 1 1 2 2 1 1
+ + + + =
Gm G G D m Gm m G G T
P P P P C P C P C P C C
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
10
Necessary condition:
Let incremental cost,
The solution is to have all ICs the same for all
units.
0 =
c
c
Gi
T
P
C
0 =
c
c
c
c
+
c
c
Gi
Gm
Gm
m
Gi
i
P
P
P
C
P
C
1 , , 2 , 1 =
c
c
=
c
c
m i
P
C
P
C
Gm
m
Gi
i
1 , , 2 , 1 =
c
c
= m i
P
C
IC
Gi
i
i
-1
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
11
Case 1: Solution Interpretation
Incremental Cost is the slope of fuel-cost curve.
Unit of IC is dollar per MWh.
IC tells how much it costs to operate this unit for the next
MW power.
Example: two units with IC1 > IC2.
This means that, for additional 1 MW power,
Operating unit 1 cost more than operating unit 2.
To minimize operation cost, it is reasonable to reduce power
output of unit 1 and increase that of unit 2.
At optimal, additional cost from unit 1 and 2 should be the same.
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
12
Example 11.8
Two generators with the following cost curve.
( )
2
1 1 1 1
01 . 0 45 900
G G G
P P P C + + = ( )
2
2 2 2 2
003 . 0 43 2500
G G G
P P P C + + =
( )
1
1
1 1
1
02 . 0 45
G
G
G
P
P
P C
IC + =
c
c
=
( )
2
2
2 2
2
006 . 0 43
G
G
G
P
P
P C
IC + =
c
c
=
MW 700
2 1
= = +
D G G
P P P
( )
1 1 2 1
700 006 . 0 43 02 . 0 45
G G
P P IC IC + = + =
MW 6 . 84
1
=
G
P MW 4 . 615
2
=
G
P
/MWh 69 . 46
2 1
= = IC IC
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
13
0 100 200 300 400 500 600 700
42
44
46
48
50
52
54
56
58
60
Example 11.8
0 100 200 300 400 500 600 700
42
44
46
48
50
52
54
56
58
60
P1 = 84.6 MW P2 = 615.4 MW
IC1 = IC2 = 46.69
1
2
1
2
3
+
+
= Load
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
14
Case 1: Solution Outline
In general case with more than 2 generating
units, we know only
Total load
IC curves of each unit
Iterative procedure:
Step 1: Pick initial
Step 2: Find corresponding power output of each unit
Step 3: If total power < load, increase and go back
to step 2. Else, stop.
=
=
m
i
Gi i T
P C C
1
( ) ( ) 0 , , ,
3 2
1 1
= =
= =
Gm G G loss
n
i
Di
m
i
Gi Gi
P P P P P P P f
( ) |
.
|
\
|
=
=
D Gm G G loss
m
i
Gi T T
P P P P P P C C , , ,
~
3 2
1
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
17
Case 3: Optimality Condition
Lagrangian of the problem,
Optimality condition,
( )
( )
0 0
~
, , 3 , 2 , 0 1 0
~
0 0
~
1
1
1 1
1
= =
c
c
= =
|
|
.
|
\
|
c
c
c
c
=
c
c
=
c
c
=
c
c
=
D loss
m
i
Gi
T
Gi
loss
Gi
Gi i
Gi
T
G
G
G
T
P P P
C
m i
P
P
P
P C
P
C
P
P C
P
C
( ) |
.
|
\
|
=
=
D Gm G G loss
m
i
Gi T T
P P P P P P C C , , ,
~
3 2
1
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
18
Penalty Factor
Rewrite optimality condition 2,
Define penalty factor,
From optimality condition 1,
we have
( )
m i
P
P C
P
P
Gi
Gi i
Gi
loss
, , 2 , 1
1
1
= =
c
c
|
|
.
|
\
|
c
c
Gi
loss
i
P
P
L
c
c
=
1
1
1
1
= L
i
L
( )
=
c
c
1
1 1
G
G
P
P C
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
19
Case 3: Optimality Condition Revisit
Optimality condition,
Equivalently,
and
( )
( )
0 0
~
, , 3 , 2 , 0
~
0
~
1
1
1 1
1
1
= =
c
c
= =
c
c
=
c
c
=
c
c
=
c
c
=
D loss
m
i
Gi
T
Gi
Gi i
i
Gi
T
G
G
G
T
P P P
C
m i
P
P C
L
P
C
P
P C
L
P
C
m i IC L
i i
, , 2 , 1 = =
IC of unit i
0
1
=
=
D loss
m
i
Gi
P P P
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
20
Case 3: Solution Outline,
Penalty Factor Known
Assume that the following is known.
Total load
IC curves of each unit
Penalty factor of each unit (from explicit loss
expression)
Iterative procedure:
Step 1: Pick initial
Step 2: Find corresponding power output of each unit
from
Step 3: If total power < load, increase and go back
to step 2. Else, stop.
m i IC L
i i
, , 2 , 1 = =
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
21
Penalty Factor Calculation
Consider power flow equation with loss,
We can rewrite equation as follow:
Differentiate with respect to , we have
From,
( ) ( ) 0 , , ,
3 2
1 1
= =
= =
Gm G G loss
n
i
Di
m
i
Gi Gi
P P P P P P P f
( )
=
+ =
m
i
Gi Gm G G loss D G
P P P P P P P
2
3 2 1
, , ,
m i
P
P
P
P
Gi
loss
Gi
G
, , 3 , 2 1
1
=
c
c
=
c
c
Gi
P
Gi
loss
i
P
P
L
c
c
=
1
1
m i
L P
P
i Gi
G
, , 3 , 2
1
1
=
=
c
c
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
22
Use of Jacobian Matrix to Find Penalty Factor
From,
Use as the intermediaries together with chain
rule,
n i
P
P
P P
P
P P
P
P
P
P
P P
P
P P
P
P P
i
m
m
G
i
G
i
G
i
Gm
Gm
G
i
G
G
G
i
G
G
G
i
G
, , 3 , 2
1 3
3
1 2
2
1
1 3
3
1 2
2
1 1
=
c
c
c
c
+ +
c
c
c
c
+
c
c
c
c
=
c
c
c
c
+ +
c
c
c
c
+
c
c
c
c
=
c
c
u u u
u u u u
k
u
i Di Gi
P P P + =
i Gi
P P c = c
(
(
(
(
(
c
c
c
c
=
(
(
(
(
(
(
(
(
(
c
c
c
c
(
(
(
(
(
c
c
c
c
c
c
c
c
c
c
c
c
n
G
G
m
G
G
n
n
n
m
n
n m
P
P
P
P
P
P
P P P
P P P
u
u
u u u
u u u
1
2
1
1
2
1
2
2 2 2
2
1
1
We can use
( )
(
(
(
(
(
c
c
c
c
c
c
c
c
=
c
c
=
n
n n
n
q
p
pq
P P
P P
P
u u
u u
u
u
2
2
2
2
11
J
( ) | |
T
pq
u
11
J
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
23
Penalty Factor Calculation
(covered from book)
Relationship between power loss and net
power injection
Use of intermediaries
Matrix form
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
24
Relationship Between Power Loss and Net
Power Injection
No explicit equation for loss.
Try to relate loss to net power injection.
From,
Where = Generator i power
= Load power at bus i
= Net power injection from bus i
And, from power flow,
This means that:
2 G
P
G1 G2
1 G
P
3 D
P
1 D
P
2
V
1
V
3
V
G3
2 D
P
3 G
P
2
P
1
P
i Di Gi
P P P + =
Gi
P
Di
P
i
P
( )
= =
=
n
i
Di
m
i
Gi Gm G G loss
P P P P P P
1 1
3 2
, , ,
( )
=
=
n
i
i Gm G G loss
P P P P P
1
3 2
, , ,
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
25
Using Intermediaries
Our goal is to calculate,
But we know that
And, net power injection is in terms of
And, for every bus i,
We are using as the intermediaries to find
Gi
loss
i
P
P
L
c
c
=
1
1
( )
=
=
n
i
i Gm G G loss
P P P P P
1
3 2
, , ,
Gi
loss
P
P
c
c
( ) n i P
n i
, , 2 , 1 , , ,
3 2
= u u u
,1 = 0
k
u
i Di Gi
P P P + =
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
26
from Net Power Injection
From net power injection,
and
We can find,
( )
=
=
n
i
i Gm G G loss
P P P P P
1
3 2
, , ,
( ) n i P
n i
, , 2 , 1 , , ,
3 2
= u u u
n k
P P P P P
k
n
k
m
k
m
k k
loss
, , 3 , 2
1 1
=
c
c
+ +
c
c
+
c
c
+ +
c
c
=
c
c
+
u u u u u
k
loss
P
u c
c
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
27
from Chain Rule
Another way to find
Using chain rule, we have
And, from we have
i Di Gi
P P P + =
k
i
k
Gi
P P
u u c
c
=
c
c
m i
P
P
P P
k
Gi
Gi
loss
k
loss
, , 3 , 2 =
c
c
c
c
=
c
c
u u
k
loss
P
u c
c
n k
P
P
P P
P
P P
k
Gm
Gm
loss
k
G
G
loss
k
loss
, , 3 , 2
2
2
=
c
c
c
c
+ +
c
c
c
c
=
c
c
u u u
k
loss
P
u c
c
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
28
Two Equations of
So far, we have two equations representing the
same variable.
Equating two equations, we have
n k
P P P P P
k
n
k
m
k
m
k k
loss
, , 3 , 2
1 1
=
c
c
+ +
c
c
+
c
c
+ +
c
c
=
c
c
+
u u u u u
n k
P
P
P P
P
P P
k
Gm
Gm
loss
k
G
G
loss
k
loss
, , 3 , 2
2
2
=
c
c
c
c
+ +
c
c
c
c
=
c
c
u u u
n k
P P P P P P
P
P P
P
P
k
n
k
m
k
m
k k k
Gm
Gm
loss
k
G
G
loss
, , 3 , 2
1 2 1 2
2
=
c
c
+ +
c
c
+
c
c
+ +
c
c
+
c
c
=
c
c
c
c
+ +
c
c
c
c
+
u u u u u u u
n k
P P
P
P P
P
P P P
k
n
k
m
Gm
loss
k
m
G
loss
k k
, , 3 , 2 0 1 1
1
2
2 1
= =
c
c
+ +
c
c
+
|
|
.
|
\
|
c
c
c
c
+ +
|
|
.
|
\
|
c
c
c
c
+
c
c
+
u u u u u
k
loss
P
u c
c
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
29
Matrix Form
Rearranging equation to,
We can rewrite into following matrix form,
n k
P P P
P
P P
P
P P
k k
n
k
Gm
Gm
loss
k
m
G
loss
k
, , 3 , 2 1 1
1 1
2
2
=
c
c
=
c
c
+ +
c
c
+
|
|
.
|
\
|
c
c
c
c
+ +
|
|
.
|
\
|
c
c
c
c
+
u u u u u
(
(
(
(
(
c
c
c
c
=
(
(
(
(
(
(
(
(
(
c
c
c
c
(
(
(
(
(
c
c
c
c
c
c
c
c
c
c
c
c
n
Gm
loss
G
loss
n
n
n
m
n
n m
P
P
P
P
P
P
P P P
P P P
u
u
u u u
u u u
1
2
1
2
2
2 2 2
2
1
1
1
1
( )
(
(
(
(
(
c
c
c
c
c
c
c
c
=
c
c
=
n
n n
n
q
p
pq
P P
P P
P
u u
u u
u
u
2
2
2
2
11
J
We can use
( ) | |
T
pq
u
11
J
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
30
RHS Calculation
The right hand side of the matrix can be found
from,
( ) ( ) | | n k B G V V
P
k k k k k
k
, , 3 , 2 cos sin
1 1 1 1 1
1
= =
c
c
u u u u
u
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
31
Case 3: Solution Outline
Assume initial dispatch,
Step 1: Solve power flow for
At this step, we can find,
Step 2: Evaluate
and find
Step 3: Check if , DONE.
If not, go to 4.
Step 4: If then,
If then,
Go back to Step 1.
0 0
3
0
2
, , ,
Gm G G
P P P
1 G
P n i
i
, , 3 , 2 = u
( ) | |
T
pq
u
11
J
n i
P
i
, , 3 , 2
1
=
c
c
u
( ) | | ( )
(
c
c
=
(
c
c
=
i
T
pq
i i
P
P
P
L u
u
1
1
11
1
1
J
m i IC L
i i
, , 2 , 1 = =
m i IC L
i i
, , 2 , 1 = - >
o
Gi Gi
P P
m j IC L
j j
, , 2 , 1 = - < o +
Gj Gj
P P
ELEN 460:Power System Control and Operation:
Economic Dispatch (ED)
32
Case 4: With Line Losses,
With Generator Limit
The solution outline to this case is basically the
same as Case 3.
However, in Step 1 and 4, lower and upper limit
of each generating units need to be check.