CHAPTER 2

ECONOMIC DISPATCH
2.1 Introduction
2 2 Operating Cost of a Thermal Plant
2.2
2.3 Economic Dispatch
2.4 Allocation of Load among
g Units within the Same
Plants
2.5 Allocation of Load among Plant Incorporating Loss
 ED-Introduction
ED Introduction
• Power system operation worldwide
– Experiencing dramatic change due to the ongoing
of industry
industry. These visible changes have
• Shifting of responsibilities
• Changes
g in the area of influence
• Shift in the operating objectives and strategies
• Distribution of work
• Eg. tnb
• Power system operation from classical
perspective view – single entity
 ED Introduction
ED-Introduction

• Economic power system operation is important

for profit return on the capital investment
• Capital investment comprises
– Cost of land
land, buildings
– Cost of generating equipment
– Cost of transmission system
– Cost of distribution system, etc
 ED Introduction
ED-Introduction

• Reasons for power companies have to achieve

the maximum efficiency of power system
operation
– Rates
R t fixed
fi d by
b regulatory
l t bodies
b di
• TNB new tariff for all sectors
– Important of conservation of fuel
– Minimizes the cost of kWh to the consumer
 ED Introduction
ED-Introduction

• Economics operation involving power

generation and delivery can divided into two
parts:
– Dealing
D li with ith minimum
i i costt off power production
d ti
called economic dispatch (ED)
– Dealing with minimum loss (ML) delivery of the
generated power to the loads
 ED Introduction
ED-Introduction

• Economic Dispatch
– Determines
D t i th
the power output
t t off each
h plants
l t which
hi h
can minimize the overall cost of fuel needed to
serve the system load
– Focuses upon coordinating the production costs at
all power plants operating on the system
 ED Introduction
ED-Introduction

• Minimum Loss problem

– Can
C assume many formsf depending
d di on howh control
t l
of the power in the system is exercised.
•B th ED and
Both d ML can be
b solved
l d by
b means OPF
program
 ED-Introduction
ED Introduction
• Opt
Optimal
a Power
o e Flow
o (OPF)
(O )
– Used to optimise the power flow solution of a large scale
power system
– Done by minimising selected objective functions while
maintaining an acceptable system performance, i.e
generator capability limits
– Objective functions also known as cost function may present
as economic costs, system security
• Several power flow program
– Gauss Seidel
– Newton Raphson
– Fast Decoupled
 ED- Operating cost
of a Thermal Plant

• Factors influencing power generation at minimum cost

– Operating efficiencies of generators
– Fuel cost
– Transmission losses
• Efficient generator in the system does not guarantee
minimum cost
– Located in an area where fuel cost is high
– Higher transmission losses due to the far distance location
 ED- Operating cost
of a Thermal Plant

• Hence, the operating cost plays an important

role in the economic scheduling
• Input to thermal plant is generally measured in
Bt /h and
Btu/h d the
th output
t t iis measuredd iin MW
 ED- Operating cost
of a Thermal Plant

• Fig 2.1(a) shows a simplified input-output curve of a

thermal unit known as heat-rate curve
– Heat rate – as the ratio of fuel input to the corresponding
power output ( million Btu/MWh)
• Fig 2 1 (b) shows the results in fuel
2.1 fuel-cost
cost curve
– Converting the ordinate of heat-rate curve from Btu/h to $/h
• Btu/h x $/Btu
 ED- Operating cost
of a Thermal Plant

Fuel Cost
Input Ci
Bt /h
Btu/h $/h

Pi, MW Pi, MW
Fig 2.1
(a) Heat-rate
Heat rate curve (b) Fuel-cost
F l t curve
 ED- Operating cost
of a Thermal Plant

• In practical case, the fuel cost of generator i

can be represented as quadratic function of
real power generation as shown in (2.1)

Ci = α i + βi Pi + γ i Pi $ / h
2
(2 1)
(2.1)
 ED- Operating cost
of a Thermal Plant

• An important characteristic is obtained by

plotting the derivative of the fuel
fuel-cost
cost curve
versus the real power
– Known
K as the
th incremental
i t l fuel-cost
f l t curve
– Measure of how costly it will be to produce the next
increment of power
 ED- Operating cost
of a Thermal Plant

• IFC can be denoted by

dC i
= 2 γ i Pi + β i $ / MWh (2.2)
dP i

– where α , β , γ = constants
t t
i i i

Pi = output of generator i in MW
 ED- Operating cost
of a Thermal Plant

λi
$/MWh

Pi, MW
Fig 2.2 Typical incremental fuel-cost curve
 Economic Dispatch

• The name given to the process of apportioning

the total load on a system between the various
generating plants to achieve the greatest
economy of operation
• Simplest form of ED
• Complicated facet of ED
 Economic Dispatch
• Simplest form of ED
– Minimising only the fuel cost through determining of
generation
• Among a set of online (committed) units in the same
thermal power plants
• Among a number of committed thermal power plant of
different locations subject to the constraints that total
generation equal to the demand or equal to the demand
plus transmission losses
– Conventional ED for verticallyy integrated
g power
p
system
 Economic Dispatch
• Co
Complicated
p cated facet
acet o
of ED co
considers
s de s
– Individual generation units upper and lower limits
– Conflicts among economy, security and emission
(environmental)
– Minimising the total cost considering fuel, labour and
maintenances
– Optimal power solution considering both active and reactive
control variables
– Deregulated II.e
e competitive market structure
( many GenCo. Competing through ISO)
 Economic Dispatch

• Discussion only limit on the following

cases
– Generator within a plant (generation limits
and transmission loss neglected
– Generators in two or more plants apart from
one another (generation limits neglected but
transmission loss considered)
 Economic Dispatch

• Hence, it is requirement to understand the

following quantities or concepts
– Fuel cost
– System
S t llambda
bd or system
t marginal
i l price
i or system
t
incremental cost λ
– Loss
L or B coefficients
ffi i t
– Penalty factors
 Economic Dispatch

• ED problem involves the solution of two

different problems

– Unit commitment or pre-dispatch

– On-line economic dispatch

 Economic Dispatch

• ED problem
– Unit
U it commitment
it t or pre-dispatch
di t h
• Select optimally out of the available generating sources to
operate for meeting the expected load
• Provide a specified margin of operating reserve over a
specified period of time
 Economic Dispatch

• ED problem
– On-line
O li economic
i dispatch
di t h
• Required to distribute the load among the generating
units
• Distribution that minimise the total cost of supplying the
minute to minute requirement of the system
 Economic Dispatch-
Simulations

• Load flow study for particular demand

– the
th generation
ti att allll the
th generator
t buses
b (regulated
( l t d
buses) are fixed
– The generation at slack or swing bus is not fixed
where allow the generation to take value within
certain limits
 Economic Dispatch-
Practical

• Economic load dispatch

– Generations are allowed to take values again within
certain limits to meet a particular load demand
within minimum fuel consumption
– Is really the solution of a large number of load flow
problems
– Able
Abl to
t choose
h the
th one that
th t can minimise
i i i the
th costt
of generation
 Economic Dispatch-
Practical

• Economic load dispatch

– Cost of generation
• Total cost of generation is a function of the individual
generation
• Depend upon the system constraint
• Not fixed for a particular load demand but depends upon
tthe
e ope
operating
at g constraints
co st a ts o of tthe
e sou
sources
ces
– Understand the various constraints before taking up
the ELD problem
 EDP Constraints
EDP-Constraints

• Two types of constraints

–Equality constraints

–Inequality constraints
 Economic Dispatch

• Two types of constraints

– Equality constraints
• These constraints arise out of the necessity for the power system to
satisfy load balance
– At any bus i, if Psi and Qsi correspond to the scheduled
generation, Pdi and Qdi are the load demands the following
equations must be satisfied at that bus
• Psi –P
PDi –Pi
Pi = Gi=0
0

• Qsi-QDi-Qi=Hi=0
 Economic Dispatch

• Inequality constraints
– A number of other constraints due to physical and operation
limitations of the units components will arise in economic
scheduling
• Hard ware type
– Those are definite and specific like the tapping rage off and on-
load tap changing transformer
• Software type
– Those have some flexibility associated with them like nodal
voltage and phase angles between the nodal voltages
 Economic Dispatch

– Efficiently handled by penalty function methods

» Generation constraints – thermal limit
» Voltage constraints – vary certain limit
» Transformer tap settings- tapping limit
» Transmission line constraints – thermal capability
limit of circuit
» etc
 ED-Allocation of Load
Among Units within the
Same Plant
• Early attempt of ED
– Supply power from the most efficient plants at light loads
– Continue supply until the maximum efficiency of that plant
was reached
– Next plant would start for further increase in load
– Third plant would be called upon until the maximum
efficiency of the second plant was reached
– However
H thi
this method
th d fails
f il to
t minimise
i i i the
th costt
 ED-Allocation of Load
Among Units within the
Same Plant
• In the context of allocation of load among two units
within
ithi a plant
l t
– For instance, two plants with different incremental fuel cost
(IFC)
– For greater reduction in operating cost
• The load should be transferred higher incremental cost unit to lower
incremental cost unit
• The transfer should be continued until the incremental fuel costs of
the two units are equal
– For ED the units must operate at the same IFC
 ED-Allocation of Load
Among Units within the
Same Plant
• An ED schedule for assigning loads to each unit
in a plant can be prepared by
– Assuming various values of total plant output
– Calculating the corresponding incremental fuel cost
λ of the plant
– Substituting the value of λ for λi in the equation for
the incremental fuel cost of each unit to calculate
its output
 ED-Allocation of Load
Among Units within the
Same Plant
•F
For ttwo plant
l t with
ith ttwo units
it operating
ti under
d
ELD the λ of the plant equals λi of each unit, so

dC 1 dC 2
λ = = 2 γ 1 P1 + β 1 ; λ = = 2 γ 2 P2 + β 2 (2.3)
dP1 dP 2
and
P1 + P2 = PD (2.4)
or
k

∑i =1
Pi = PD (2 5)
(2.5)
 ED-Allocation of Load
Among Units within the
Same Plant
– Where PD is the total load demand
• Solving
g for P1 and P2 we obtain
λ − β1 λ − β2
P1 = ; P2 = (2.6)
2γ 1 2γ 2

• And for Pi
λ − β
Pi = i
(2.7)
2γ i
 ED-Allocation of Load
Among Units within the
Same Plant
k
λ − βi
∑i =1 2γ i
= PD (2.8)
• Thus an analytical solution can be obtained for
λ as
k
β i
PD + ∑
i=1 2γ i
λ = k
(2.9)
1
∑
i=1 2γ i
 ED-Allocation of Load
Among Units within the
Same Plant
• Example 2.12 1 (Example 7.4
7 4 page 271 Hadi Saadat)
The fuel cost functions for three thermal plants in $/h are given
by
C1= 500 + 5.3P1 +0.004P12
C2= 400 + 5.5P2 +0.006P22
C3= 200 + 5.8P3
5 8P3 +0.009P3
+0 009P32
where P1,P2 and P3 are in MW. The total load, PD is 800MW.
Neglecting line losses and generation limits, find the optimal
dispatch and the total cost in $/h by
b
(a) analytical method
((b)) graphical
g p demonstration
(c) iterative technique using gradient method (assume λ)
 ED-Allocation of Load
Among Units within the
Same Plant
• Example 2.1
21
• By neglecting the losses and generating limit,
tthe
e tota
total input
put po
power
e equa
equal tthe
e tota
total de
demand
a d
P1 + P2 + P3 = 800
Solving
g the lambda using g eqn.
q (2.9):
( )
λ = 8.5 $/MWh
Substituting for λ in eqn. (2.7), the ED:
P1 400 MW
P1=
P2 = 250 MW
P3 = 150 MW
 ED-Allocation of Load
Among Units within the
Same Plant
• Example 2.1-Graph
2 1 Graph demo
 ED-Allocation of Load
Among Units within the
Same Plant
• Example 2.2
22
There are two generators of 100MW each with incremental
characteristics:
λ1=2 +0.012P1
λ2=1.5 +0.015P2
Minimum load on each unit is 10MW, total load to be
supplied is 150MW. Determine the economic operating
schedule
(a) analytical method
(b) graphical demonstration
 ED-Allocation of Load
Among Units within the
Same Plant
• E
Examplel 2.2
22
b) analytical method
By neglecting the losses and generating limit
limit, the total input
power equal the total demand
P1 + P2 = 150
Solving the lambda using eqn. (2.9):
λ = λ1= λ2= 2.78 $/MWh
Substituting for λ in eqn.
eqn (2.7),
(2 7) the ED:
P1= 65 MW
P2 = 85 MW
 ED-Allocation of Load
Among Units within the
Same Plant
• Example
E l 22.2
2
b) Graph demo
 ED-Allocation of Load
Among Units within the
Same Plant
• Example 2.3
Determine the saving in fuel cost in dollars per hour
for the economic dispatch of a total load of 150 MW
between the two units of the plants described in
Example 2.2 compared with equal allocation of the
same total load
 ED-Allocation of Load
Among Units within the
Same Plant
• Example 2.4
24
The fuel inputs per hour of plant 1 and plant 2 are given
as
C1=0.2P12 + 40P1 + 120 $/h
C2=0.25P22 + 30P2 + 150 $/h
Determine the economic operating schedule and the
corresponding cost of generation if the maximum and
minimum loading on each units is 100 MW and 25 MW,
the demand is 180 MW and the transmission losses are
neglected. If the load is equally shared by both the units,
determine the saving obtained by loading the units as per
equal incremental production cost
 ED-Allocation of Load
Among Plant
Incorporating Loss
• When transmission distances are very small
and load density is very high
– transmission losses may be neglected
– The optimal of generation is achieved with all plants
operating at equal incremental production cost
• In large interconnected network, long transmission
distances with low load density area
– Transmission losses may be 5% to 10% of the total load
– Essential to take into account the losses when determining
the economic allocation of load among the plants
 ED-Allocation of Load
Among Plant
Incorporating Loss
– Although the incremental cost at one plant bus may lower
than that of another plant for a given allocation of load
among the plants but
• Much farther from the load centre
• Losses in transmission is so high
g
– Economic may dictate lowering the load of this plant and
increasing the load of other plant although higher
incremental cost
– Thus, it requires coordinating transmission losses into the
scheduling of the output of each plant for maximum
economic
 ED-Allocation of Load
Among Plant
Incorporating Loss
• For a system with k generating units
k
Ct = C1 + C2 + ...................... + Ck = ∑Ci (2.10)
i =1
– Where Ct= total fuel cost for all g
generating
g units
• The total MW power input to the network from
all the units is the sum
k
P1 + P2 + ...................+ Pk = ∑ Pi (2.11)
(2 11)
i =1
 ED-Allocation of Load
Among Plant
Incorporating Loss
– Where P1,P2, ….Pk = the individual outputs of the units
injected to the network. The total fuel cost C of the system is
a function of all the power plants output
• The economic dispatch problem including
transmission losses is defined as

k
Min C t = ∑
i = 1
Ci

k
Subject to PD + PL − ∑
i =1
P1 = 0 ( 2.12)
 ED-Allocation of Load
Among Plant
Incorporating Loss
– Where PL is the total transmission system loss which is
assumed to be a function of generation
generation.
– On common practice for including the effect of transmission
p
losses is to express the total transmission loss as a quadratic
q
function of the generator power outputs
• The simplest function is
k k
PL = ∑∑PB
i =1 j =1
i ij Pj (2.13)
 ED-Allocation of Load
Among Plant
Incorporating Loss
• A more generall formula
f l containing
t i i a linear
li term
t andd
constant term, referred to as Kron’s loss formula is
k k k
PL = ∑∑ Pi Bij Pj + ∑ B0i Pi + B00 (2.14)
i =1 j =1 i =1

• Pi and Pj = the source loadings

• Bij = transmission loss coefficients or B coefficients
 ED-Allocation of Load
Among Plant
Incorporating Loss
• In the case of two generating plants,
transmission loss equation can be written as

PL = B11 P1 + 2B12 P1 P2 + B22 P2 + B10 P1 + B20 P2 + B00 (2.15)

2 2

– Which can rearrange into the equivalent form

⎡ B11 B12 ⎤ ⎡ P1 ⎤ ⎡ B10 ⎤

PL = [P1 P2 ]⎢ ⎥ ⎢ ⎥ + [P1 P2 ]⎢ ⎥ + B 00 (2.16)
⎣ B 21 B 22 ⎦ ⎣ P2 ⎦ ⎣ B 20 ⎦
 ED-Allocation of Load
Among Plant
Incorporating Loss
• Making
M ki use off the
th Langragian
L i multiplier
lti li λ

⎛ k
⎞
L = C t + λ ⎜ PD + PL − ∑ Pi ⎟ (2 17)
(2.17)
⎝ i =1 ⎠
• The augmented cost function L is often called the
Langrangian.
• The Langrange multlipier λ is the effective incremental
fuel cost of the system when transmission line losses
are taken into account
 ED-Allocation of Load
Among Plant
Incorporating Loss

• For minimum cost we require the derivative of

L with respect to each Pi to equal zero
• Hence,
Hence
∂L ∂ ⎡ ⎛ k
⎞⎤
= ⎢(C1 + C2 + ........+ Ck ) + λ⎜ PL + PD − ∑ Pi ⎟⎥ = 0 (2.18)
(2 18)
∂Pi ∂Pi ⎣ ⎝ i =1 ⎠⎦
 ED-Allocation of Load
Among Plant
Incorporating Loss
• Since PD is fixed and the fuel cost of any one
unit varies only of the power output of that unit
is varied, equation (2.18) yields

∂L ∂Ci ⎛ ∂PL ⎞
= + λ ⎜⎜ − 1⎟⎟ = 0 (2.19)
∂Pi ∂Pi ⎝ ∂Pi ⎠
∂Ci ∂PL
+λ = λ (coordination equation)
∂Pi ∂Pi
 ED-Allocation of Load
Among Plant
Incorporating Loss
• For each of the generating unit outputs P1, P2,………Pk.
Because Ci depends on only Pi, the partial derivative
of Ci can be replaced by the full derivative and
equation (2.19) then gives
⎛ ⎞
⎜ ⎟
⎜ 1 ⎟ dC i
λ = (2.20)
⎜ ∂ P L ⎟ dP i
⎜1− ⎟
⎝ ∂ P i ⎠

– For every value of i

 ED-Allocation of Load
Among Plant
Incorporating Loss

• This equation is often written in the form

dCi
λ = Li (2.21)
dPi
• Which is approximately the cost in $/h to
i
increase the
h totall delivered
d li d load
l d by
b 1 MW
 ED-Allocation of Load
Among Plant
Incorporating Loss
• Where Li is called the penalty factor of plant i and is
given by
1
L = (2.22)
i
∂PL
1 −
∂ Pi
• The result of equation that minimum fuel cost is
obtained when the incremental fuel cost of each unit
multiplied by its penalty factor is the same for all
generating units in the system
 ED-Allocation of Load
Among Plant
Incorporating Loss
• For system of three units, not necessarily in the
same power plant equation (2.22) yields

dCi dC2 dC3

λ = L1 = L2 = L3 ((2.23))
dPi dP2 dP3

∂ PL
• The penalty factor Li depends on which is a
∂ Pi

measure of sensitivity of the transmission

system losses to the changes in Pi alone.
alone
 ED-Allocation of Load
Among Plant
Incorporating Loss
• Example 2.5
A two bus system is shown in Figure below. If a load of 125 MW is transmitted from
plant 1 to the load, a loss of 15.625 MW is incurred. Determine the generation
schedule and the load demand if the cost of received power is $24/MWh.
$24/MWh Solve the
problem using coordination equations and the penalty factor

λ1 = 0.025P1 + 15
λ2 = 0.05P2 + 20
1 2 G2
G1

Load
 ED-Allocation of Load
Among Plant
Incorporating Loss
• Example 2.6
A plant system with two power plants is operating
under economic dispatch. The plant outputs are
P1=140 MW and P2= 250 MW. The loss coefficients
are B11=0.001
B11 0 001 MW-1, B12=-0.0001
B12 0 0001 MW-1,
B22=0.0013 MW-1. If the system incremental cost is
13 $/MWh,
$/MWh find
i) the penalty factors and incremental fuel costs of
both plants
ii) the total system load