LMP Preliminaries

(Ref: D. Kirschen and the University of Washington,

www2.ee.washington.edu/research/real/Library/Teaching/06-
OPF.pptx)
Economic dispatch

L
A B C

Objective: minimize the cost of generation

Constraints
Equality constraint: load generation balance
Inequality constraints: upper and lower limits on
generating units output

2
Limitations of economic dispatch
B C

Network

A D

Generating units and loads are not all connected to

the same bus
The economic dispatch may result in unacceptable
flows or voltages in the network

3
Example of network limitation
B
A
LA LB

Maximum flow on each line: 100MW

CA CB

50 $/MWh 100$/MWh
PA PB
4
PAMAX PBMAX
Acceptable ED solution
100 MW 0 MW
300 MW
B
A

LA = 100 MW 100 MW LB = 200MW

The solution of this (trivial) economic dispatch is:

The flows on the lines are below the limit

The economic dispatch solution is acceptable

5
Unacceptable ED solution
200 MW 0 MW
500 MW
B
A

LA = 100 MW 200 MW LB = 400MW

The solution of the economic dispatch is:

The resulting flows exceed their limit

The economic dispatch solution is not acceptable

6
Modified ED solution
100 MW 200 MW
300 MW
B
A

LA = 100 MW 100 MW LB = 400MW

In this simple case, the solution of the economic dispatch

can be modified easily to produce acceptable flows.

This could be done mathematically by adding the following

inequality constraint:

However, adding inequality constraints for each problem

is not practical in more complex situations
We need a more general approach
7
Optimal Power Flow (OPF) - Overview
Optimization problem
Classical objective function
Minimize the cost of generation
Equality constraints
Power balance at each node - power flow equations
Inequality constraints
Network operating limits (line flows, voltages)
Limits on control variables

8
Mathematical formulation of the OPF (1)
Decision variables (control variables)
Active power output of the generating units
Voltage at the generating units
Position of the transformer taps
Position of the phase shifter (quad booster) taps
Status of the switched capacitors and reactors
Control of power electronics (HVDC, FACTS)
Amount of load disconnected
Vector of control variables:

9
Mathematical formulation of the OPF (2)
State variables
Describe the response of the system to changes in the control variables
Magnitude of voltage at each bus
Except generator busses, which are control variables
Angle of voltage at each bus
Except slack bus
Vector of state variables:

10
Mathematical formulation of the OPF (3)
Parameters
Known characteristics of the system
Assumed constant
Network topology
Network parameters (R, X, B, flow and voltage limits)
Generator cost functions
Generator limits

Vector of parameters:

11
Mathematical formulation of the OPF (4)
Classical objective function:
Minimize total generating cost:

Many other objective functions are possible:

Minimize changes in controls:

Minimize system losses

12
Mathematical formulation of the OPF (5)
Equality constraints:
Power balance at each node - power flow equations

Compact expression:

13
Mathematical formulation of the OPF (6)
Inequality constraints:
Limits on the control variables:

Operating limits on flows:

Operating limits on voltages

Compact expression:

14
Compact form of the OPF problem

Subject to:

15
OPF Challenges
Size of the problem
1000s of lines, hundreds of controls
Which inequality constraints are binding?
Problem is non-linear
Problem is non-convex
Some of the variables are discrete
Position of transformer and phase shifter taps
Status of switched capacitors or reactors

16
Solving the OPF using gradient methods
Build the Lagrangian function

The gradient of the Lagrangian indicates the direction of steepest

ascent:

Move in the opposite direction to the point with the largest gradient
Repeat until

17
Linearizing the OPF problem
Use the power of linear programming
Objective function
Use linear or piecewise linear cost functions
Equality constraints
Use dc power flow instead of ac power flow
Inequality constraints
dc power flow provides linear relations between injections (control variables)
and MW line flows

18
Sequential LP OPF
Consequence of linear approximation
The solution may be somewhat sub-optimal
The constraints may not be respected exactly
Need to iterate the solution of the linearized problem
Algorithm:
1. Linearize the problem around an operating point
2. Find the solution to this linearized optimization
3. Perform a full ac power flow at that solution to find the new operating point
4. Repeat

19
Advantages and disadvantages
Advantages of LPOPF method
Convergence of linear optimization is guaranteed
Fast
Reliable optimization engines are available
Used to calculate nodal prices in electricity markets
Disadvantages
Need to iterate the linearization
Reactive power aspects (VAr flows, voltages) are much harder to linearize
than the active power aspects (MW flows)

20
DC Power Flow
Approximation
Power Flow Equations
N
PkI - Vk Vi[Gki cos(q k - q i ) + B ki sin(q k - q i )] = 0
i=1
N
QkI - Vk Vi[Gki sin(q k - q i ) - Bki cos(q k - q i )] = 0
i=1

Set of non-linear simultaneous equations

Need a simple linear relation for fast and intuitive
analysis
dc power flow provides such a relation but requires a
number of approximations
Neglect Reactive Power
N
PkI - Vk Vi[Gki cos(q k - q i ) + B ki sin(q k - q i )] = 0
i=1
N
QkI - Vk Vi[Gki sin(q k - q i ) - Bki cos(q k - q i )] = 0
i=1

N
PkI - Vk Vi[Gki cos(q k - q i ) + B ki sin(q k - q i )] = 0
i=1
Neglect Resistance of the Branches
N
PkI - Vk Vi[Gki cos(q k - q i ) + B ki sin(q k - q i )] = 0
i=1

N
PkI - Vk Vi Bki sin(q k - q i ) = 0
i=1
Assume All Voltage Magnitudes = 1.0 p.u.
N
PkI - Vk Vi Bki sin(q k - q i ) = 0
i=1

N
PkI - Bki sin(q k - q i ) = 0
i=1
Assume all angles are small
N
PkI - Bki sin(q k - q i ) = 0
i=1

If a is small: sina a (a in radians)

N N
(q k - q i )
PkI - Bki (q k - q i ) = 0 or PkI - =0
i=1 i=1 x ki
Interpretation
N
(q k - q i )
PkI - =0
i=1 x ki P1I P2I
P12
q1 q2
N
x12
PkI - Pki = 0 x13 x 23
i=1 P31 P23
(q - q i ) q3
Pki = k
x ki P3I

(q1 - q 2 ) (q 2 - q 3 ) (q 3 - q1 )
P12 = ; P23 = ; P31 =
x12 x 23 x13
Why is it called dc power flow?
Reactance plays the role of resistance in dc circuit
Voltage angle plays the role of dc voltage
Power plays the role of dc current

(q k - q i ) (Vk - Vi )
Pki = I ki =
x ki R ki
Example of LPOPF
Solving the full non-linear OPF problem by hand is too difficult, even
for small systems
We will solve linearized 3-bus examples by hand

29
Example
A B CA

1 2 1000 Rs/MWh
PA

PAMAX=390MW

3
CB
450 MW

Economic dispatch:

PA = PAmax = 390 MW 2000 Rs/MWh

PB
PB = 60 MW PBMAX= 150 MW

30
Flows resulting from the economic dispatch
390MW 60MW
A B

Fmax = 200MW

1 2

Fmax = 260MW Fmax = 200MW

450 MW

Assume that all the lines have the same reactance

Do these injection result in acceptable flows?

31
Calculating the flows using superposition
Because we assume a linear model, superposition is
applicable

390 MW 60 MW

1 2

3
450 MW

390 MW 60 MW

1 2 1 2

3 3
390 MW
60 MW
32
Calculating the flows using superposition (1)

390 MW

FA FB
1 2

3
390 MW

FA = 2 x FB because the path 1-2-3 has twice the reactance

of the path 1-3

FA + FB = 390 MW

FA = 260 MW
FB = 130 MW

33
Calculating the flows using superposition (2)

60 MW

FD FC
1 2

3
60 MW

FC = 2 x FD because the path 2-1-3 has twice the reactance

of the path 2-3

FC + FD = 60 MW

FC = 40 MW
FD = 20 MW

34
 Calculating the flows using superposition (3)
390 MW 60 MW

130 MW 20 MW
260 MW 40 MW
1 2 1 2

3 3
390 MW 60 MW

390 MW 60 MW
110 MW
280 MW 170 MW
1 2
Fmax = 260 MW
3
450 MW

35
Correcting unacceptable flows
390 MW 60 MW

280 MW
1 2

Must reduce flow F1-3 by 20 MW 3

450 MW

Must use a combination of reducing the injection at bus 1 and

increasing the injection at bus 2 to keep the load/generation balance
Decreasing the injection at 1 by 3 MW reduces F1-3 by 2 MW
Increasing the injection at 2 by 3 MW increases F1-3 by 1 MW
A combination of a 3 MW decrease at 1 and 3 MW increase at 2
decreases F1-3 by 1 MW
To achieve a 20 MW reduction in F1-3 we need to shift 60 MW of
injection from bus 1 to bus 2
36
Check the solution using superposition
330 MW 120 MW

110 MW 40 MW
220 MW 80 MW
1 2 1 2

3 3
330 MW 120 MW

330 MW 120 MW
70 MW
260 MW 190 MW
1 2
Fmax = 260 MW
3
450 MW

37
Comments (1)
The OPF solution is more expensive than the ED solution
CED = 1000 x 390 + 2000 x 60 = Rs 5,10,000
COPF = 1000 x 330 + 2000 x 120 = Rs 5,70,000
The difference is the cost of security
Csecurity = COPF - CED = Rs 6,00,00
The constraint on the line flow is satisfied exactly
Reducing the flow below the limit would cost more

38
Comments (2)
We have used an ad hoc method to solve this problem
In practice, there are systematic techniques for calculating the
sensitivities of line flows to injections
These techniques are used to generate constraint equations that are
added to the optimization problem

39
Security Constrained OPF (SCOPF)
Conventional OPF only guarantees that the operating constraints are
satisfied under normal operating conditions
All lines in service
This does not guarantee security
Must consider N-1 contingencies

40
Example: base case solution of OPF
330MW 120MW
A 70 MW B

1 2

260 MW 190 MW

450 MW

41
Example: contingency case
330MW 120MW
A 0 MW B

1 2

330 MW 120 MW

450 MW

Unacceptable because overload of line 1-3 could lead to

a cascade trip and a system collapse

42
Formulation of the Security Constrained OPF

Power flow equations

Subject to:
for the base case

Operating limits for

the base case

Power flow equations

for contingency k
"k
Operating limits
for contingency k

Subscript 0 indicates value of variables in the base case

Subscript k indicates value of variables for contingency k
43
Preventive security formulation
subject to:

This formulation implements preventive security because

the control variables are not allowed to change after the
contingency has occurred:
uk = u0 "k
44
Corrective security
subject to:
formulation

uk - u0 Du max

This formulation implements corrective security because the

control variables are allowed to change after the contingency
has occurred
The last equation limits the changes that can take place to
what can be achieved in a reasonable amount of time
The objective function considers only the value of the control
variables in the base case
45
Size of the SCOPF problem
Example - European transmission network:
13,000 busses 13,000 voltage constraints
20,000 branches 20,000 flow constraints
N-1 security 20,000 contingencies
In theory, we must consider
20,000 x (13,000 + 20,000) = 660 million inequality constraints
However:
Not all contingencies create limit violations
Some contingencies have only a local effect

46
Limitations of N-1 criterion
Not all contingencies have the same probability
Long lines vs. short lines
Good weather vs. bad weather
Not all contingencies have the same consequences
Local undervoltage vs. edge of stability limit
N-2 conditions are not always not credible
Non-independent events
Does not ensure a consistent level of risk
Risk = probability x consequences

47
Probabilistic security analysis
Goal: operate the system at a given risk level
Challenges
Probabilities of non-independent events
Electrical failures compounded by IT failures
Estimating the consequences
What portion of the system would be blacked out?
What preventive measures should be taken?
Vast number of possibilities

48
Need for a re-look at Transmission Pricing
Inherent non-linearities in Power System Operation
Need to differentiate between physical operation and financial
transactions

Three Basic Mechanisms are required for efficient market operation

SRMC based pricing regime to encourage efficient network usage
Financial hedging contracts to minimize exposure to price volatility
Mechanism for network fixed cost recovery

These together should create efficient long run investment signals while
minimizing distortion to short run operations
LMP - Preliminaries
The Integrated Forward Market
Co-optimizes both energy and ancillary services (thus the use of the term
Integrated)
Incorporates a Full Network Model
Clears congestion with no distinction between the currently defined terms of
Inter-zonal and Intra-zonal congestion
Accepts self-scheduling of resources (there are no bids, just MW schedules)
Includes a Security Constrained Unit Commitment (SCUC) with
Start-up Cost
Minimum Load Cost
Incremental energy cost
Capacity Bid for A/S
LMP - Preliminaries
Locational marginal pricing (LMP) is a mechanism for using market-
based prices for managing transmission congestion
Prices are determined by the bids/offers submitted by market participants
The charge for transmission usage is the incremental cost of the redispatch required to
accommodate that transmission usage
If there is no transmission congestion, the charge for transmission usage would be zero
(except for other charges to recover portions of the embedded cost of the transmission
grid, etc.)
LMP - Preliminaries
Locational marginal prices differ by location when transmission
congestion occurs
Transmission congestion prevents energy from low-cost generation
from meeting all loads and clearing the market
Some of this low-cost generation must be constrained down/off,
lowering the market price of generation in the unconstrained area
Higher-cost generation must be constrained on to serve load in the
constrained area, raising the market price of generation in the
constrained area
LMP - Definitions
Locational Marginal Price (LMP) is the marginal cost of supplying, at
least cost, the next increment of electric demand at a specific location
(node) on the electric power network, taking into account both supply
(generation/import) bids and demand (load/export) offers and the
physical aspects of the transmission system including transmission and
other operational constraints
Important concepts to note
Marginal cost
Least cost
Supply bids and demand offers
Physical aspects of the transmission system
Define the term Nodal Price to be an LMP at a specific node
The term LMP can also be generically used for the price of a load aggregation point
LMP - Definitions

The definition and calculation of an LMP is based on

Economic theory
Marginal cost (aka variable cost)
Least cost
Bids (supply curves) and Offers (demand curves)
Power system operational practices
Physical aspects of the transmission system
Operating the power system based on operating criteria
Operating procedures and constraints
LMP - Definitions
Marginal cost concept
Based on the marginal cost of the total system cost
The total system cost is similar to social surplus (aka total surplus)
Social surplus is a standard economic theory term and is calculated from
Supplier surplus (based on supply curves)
Consumer surplus (based on demand curves)
LMP - Definitions
To achieve efficiency and match supply quantities with demand quantities and create a price, the total surplus is
maximized
Maximizes the benefits to both suppliers and consumers
Note that P(Q*) represents the price for
the next increment of supply to be provided
the next increment of demand to be consumed
P(Q*) is the marginal clearing price
LMP - Definitions
Note that at a clearing quantity and clearing price, the Total Surplus is equal to total consumer benefit (area under the
demand offer) minus the total cost to supply (area under the supply bid)
The values of Q* and P(Q*) can be determined from an optimization formulation
Objective function
Maximize: Total Surplus
Constraints
Total supply quantity = Total demand quantity
LMP - Definitions
Objective function can be written as

Max ( Total Surplus )

Max ( total consumer benefit minus the total cost to supply)
Max ( area under the demand offer minus the area under the supply bid )
Min ( area under the supply bid minus area under the demand offer )
The order of subtraction is flipped around so Max changes to Min
area under the supply bid minus area under the demand offer is equal to negative of the Total Surplus
Min (area under the supply bid plus (the negative of the area under the demand offer )
The negative of the area under the demand offer can be modeled assuming the demand MWhs are
negative MWhs
The reference direction is an injection into the system and is positive, thus demand or withdrawal from the
system is negative
LMP - Definitions
LMP calculation is based on
Objective Function
Min ( negative of the Total Surplus )
Min ( Total System Cost )
Note that Total System Cost is similar to Total Surplus and in this presentation it is defined as the
negative of the Total Surplus
Subject to:
Total supply = total load (this ignores losses)
Adhering to all operational constraints
e.g., interface flow limits
The LMPs are the marginal prices, which are price sensitivities that are
produced at the solution of the optimization problem
LMP - Definitions
The calculation of the LMPs are much more complicated than the
simple one supply curve, one demand curve example because
transmission constraints cannot be violated
If transmission constraints were not enforced then all supply
curves could be aggregated and all demand curves could be
aggregated and the simple methodology could be applied to
determine the system wide clearing quantity and price
Since transmission constraints are enforced, marginal prices are on
a per node basis rather than a system wide basis
LMP - Example
Simple two bus example with a generator (supply) bid and a load
(demand) offer
Assume no losses
Assume the line is not enforced so that there will not be congestion

Rs/MWh
Rs/MWh 5000

1000

- MW 100
200 MW
LMP - Example
Load is willing to pay up to Rs 5000 / MWh for 100 MW of load
Generation is willing to sell up to 200 MW at no less than Rs1000 /
MWh
Optimize
Minimize the objective function
Minimize (Total system cost)
Such that total supply equals to total load
LMP - Example
Supply cost = area under generation bid
Rs 1000 / MWh Supply MW
Demand offer = area under load offer
Rs 5000 / MWh (-Demand MW )
Note the negative sign on the Demand MW
Optimization formulation
Min (Rs1000/MWh Supply MW + Rs 5000/MWh (-Demand MW) )
Such that Supply MW = Demand MW
The solution comes out to be
Supply MW = Demand MW = 100 MW
Total system cost is
Rs1000/MWh 100 MW + Rs 5000/MWh (-100 MW)
1,00,000 5,00,000 = - Rs 4,00,000/h
LMP - Example
What is the LMP at the load node?
Note that since the line constraint is not enforced and losses are
ignored it is effectively the same as the load and generation being
located at the same bus
The supply curve and the demand curve can be put onto the same
graph to determine the marginal clearing price, but lets try to
determine the marginal price that way
LMP - Example
Add one additional MW to the load bus
This MW is not associated with the load bid
There is no bid for this additional MW, it is fixed
Need to ensure that the total generation equals total load select 2 options to
ensure this
Option 1: This 1 MW can be served by the generator
Total generation = 100 + 1 = 101 MW
Total load = 100 + 1 = 101 MW
Option 2: This 1 MW can be served by reducing the load bid by 1 MW and
leaving the generation at the same dispatch level
Total generation = 100 MW
Total load = (100 1) + 1 = 100 MW
LMP - Example
Total system costs for each option
Option 1:
Supply cost = Rs1000 / MWh 101 MW = Rs1,01,000/h
Demand cost = Rs 5000 / MWh (-100 MW) = -Rs 5,00,000/h
Total cost = 1,01,000 5,00,000 = - Rs 3,99,000/h
Option 2:
Supply cost = Rs1000 / MWh 100 MW = Rs1,00,000/h
Demand cost = Rs 5000 / MWh (-99 MW) = -Rs 4,95,000/h
Total cost = 1,00,000 4,95,000 = - Rs 3,95,000/h
LMP - Example
What is the minimum of the two changes in total system cost
Total system cost is minimized my selecting option 1
-Rs 3,99,000/h is less than Rs 3,95,000/h
Any other combination of increasing generation by x (0 < x < 1) MW and decreasing the load bid
by (1-x) MW will lead to a total system cost greater than Rs 3,99,000/h and less than Rs
3,95,000/h
Change in cost for option 1
- Rs 3,99,000/h (- Rs 4,00,000/h) = Rs1000/h

LMP for load bus is

Rs 1000/h divided by 1 MW = Rs1000 /MWh
Same solution as plotting the two curves on the same graph
LMP Multi-Node Networks
LMPs are determined from the result of a security-constrained
least-cost dispatch
Optimal Power Flow (OPF)
This is part of the Integrated Forward Market design
LMPs may differ between locations due to
Transmission congestion
Losses
Under Locational Marginal Pricing, the charge for transmission use
in terms of congestion and losses is the difference between the
LMPs at the receipt and delivery locations
LMP Multi-Node Networks
Unconstrained conditions (no congestion)
Single Market Clearing Price (assuming no losses)
This is unconstrained economic dispatch
Constrained conditions (congestion exists)
The marginal cost of energy varies by location as low cost supply cannot reach all demand
LMPs reflect increased cost to deliver energy when inadequate transmission exists
LMPs can be quite different from the single unconstrained economic dispatch price due to
costs to deliver energy from the marginal generating units to load buses
Decomposition of LMP
(Reference: An Energy Reference Bus Independent LMP Decomposition
Algorithm, IEEE Transactions on Power Systems, Cheng and Overbye,
August, 2006)
OPF Formulation
Eq(1) is a real power balance equation at
energy reference bus r (which may be
different from voltage angle reference)
Eq(2) is reactive power balance
Eq(3) and Eq(4) are for buses other than
reference bus
Eq (5) models the transmission system flow
constraints (either transmission lines,
transformers, or flowgates)
Eq(6) models the upper and lower variable
limits (e.g., bus voltage limits, generators real
and reactive power output limits, etc)
X includes the voltage magnitudes and voltage
angles of all buses, except for the actual system
slack bus
Lagrangian of the OPF
The LMP at a particular bus location is
defined as the minimal cost for
supplying an additional increment of
load at the bus location without
violating the system operation
security constraints
The real power LMP and the reactive
power LMP are the Lagrange
multipliers (p , and q), associated
with the corresponding real and
reactive power balance equations
They are the by-products of the OPF
problem solution and are often called
the shadow prices
Transmission Sensitivity Matrix
Jacobian Matrix is: Differentiate (3) and (4) wrt the
active and reactive power
demand (at all buses except the
energy reference bus), we get:

Transmission sensitivity matrix is used to measure

the marginal variations of power flows on
transmission lines with respect to small variation of
real power load at the specified bus, assuming the
generation at the energy reference bus is
responsible for the real power compensation
Hence, the transmission sensitivity matrix depends
upon the energy reference bus.
Loss Sensitivity Vector (1)
Loss Sensitivity is defined as
The loss sensitivity factor
specifies how the real power
losses change with respect to
the load variation at bus k , again
assuming the r bus generation is
responsible for the load
compensation
Loss Sensitivity Vector (2)
Differentiating (1) wrt real power Loss Sensitivity Factor is defined
demand at non-reference buses as

Values in the loss sensitivity vector again depend upon the energy
reference bus selection
Loss sensitivity of energy reference bus is zero
LMP Decomposition into Energy, Loss and
Congestion Components
Standard LMP decomposition
into its three components,
making the reasonable
assumption that the reactive
power output at the reference
bus does not hit its limits. Rearranging, the above can be
written as:
With this assumption, from (9),
it is clear that qr = 0
Then (10) can simplified as:
LMP Decomposition (2)
Eq (17) can be rewritten as This gives the decomposition of
LMP as:

Energy Component

Loss Component

Congestion Component
Interpretation
Both the loss sensitivity vector and the transmission sensitivity matrix depend on
the selection of reference bus.
Hence, the conventional decomposition of the LMP depends upon the reference
bus.
Another undesired feature of conventional LMP decomposition is that not only
do the component values themselves depend upon the reference bus, but even
the differences in these component values between any two buses depend upon
the reference bus selection
The differences of LMP congestion/loss components of two buses represent the
marginal congestion/loss costs of an additional transaction between these two
buses
Such differences can certainly affect the perception of fairness in pricing losses
and congestion and can possibly send out inaccurate price signals
Therefore
An alternative mechanism for decomposition must be used to reflect
price of losses and congestion better

Next Session:

Alterative Decomposition by the Cheng and Overbye, and by

Orfanogianni and Gross (Reference Independent Decomposition)