Locational Marginal Pricing

with Simulation Demo

PJM State & Member Training Dept.

PJM©2017 7/13/2017
 Agenda
Morning Session – LMP

• Production Cost
• Generation Dispatch
• LMP Components
• 5 Bus Model
• Shadow Prices
• Statistics
• LMP Simulation Demo

PJM©2017 2 7/13/2017
 Objectives

• Explain the difference between Dispatch Rate and LMP

• Explain the purpose of the three LMP components
• Explain Shadow Prices

PJM©2017 3 7/13/2017
 Agenda

Morning Session – LMP

• Production Cost
• Generation Dispatch
• LMP Components
• 5 Bus Model
• Shadow Prices
• Statistics
• LMP Simulation Demo

PJM©2017 4 7/13/2017
 Objective Function and Production Cost

PJM©2017 5 7/13/2017
 Objectives

• Explain PJM’s objective function

• Explain the difference between a Block Offer curve and

a Slope Offer Curve

• Define Production Cost

PJM©2017 6 7/13/2017
 What is Optimization?

• Optimization seeks to minimize or maximize the

value of a desired outcome:
‒ The “Objective Function”

• Achievement of this outcome must consider the

availability of resources or other limiting factors
‒ Subject to “Boundary Conditions”

PJM©2017 7 7/13/2017
 Optimization in Energy Markets

• The “objective function” in the Day Ahead Market is to minimize

total production costs
• The “boundary conditions” are extremely complex and include:
‒ Thermal / reactive limits
‒ Generator operating constraints
‒ External transaction schedules
‒ Regional reliability requirements
‒ Synchronized Reserves Requirement
‒ Regulation Requirement

PJM©2017 8 7/13/2017
 Optimization

• The objective in the Real Time Market is to maintain reliability

and minimize Start-up and No-Load costs
• PJM Real-Time Market clearing is a joint optimization between
energy, regulation, synchronized reserves and non-
synchronized reserves products
• The goal of the optimization is to minimize the total cost of
producing energy, regulation, and reserves

PJM©2017 9 7/13/2017
 Production Cost

PJM©2017 10 7/13/2017
 Production Cost

• It is the bids from suppliers (offers) that define the bid production
cost that is to be minimized to meet energy balance while not
violating other constraints in system operation, and consequently it
is those supply offers that will determine the price
in the Day-Ahead and Real-Time Energy Markets that is paid to
generation or demand resources

PJM©2017 11 7/13/2017
 Production Cost

• Generation must be placed in and out of service and operated

to achieve the lowest possible overall cost for the system

• Production Cost is the cost to operate a unit for a particular

period of time

• Two types of Production Cost are used at PJM:

‒ Hourly Production Cost
‒ Total Production Cost

PJM©2017 12 7/13/2017
 Hourly Production Cost
• Cost per hour to operate a unit assuming a startup has
already occurred
• Calculated by summing all costs which are incurred during one hour
of operation
• No-Load Cost
• Total Energy Cost per Segment

Hourly Production Cost

=
No-Load Cost
+
Sum of Total Energy Cost per Segment

PJM©2017 13 7/13/2017
 No-Load Cost

• No-load cost (or price) is the hourly fixed cost (or price),
expressed in $/hr, to run the generating unit at zero
net output
‒ Needed to create the starting point of a monotonically
increasing incremental cost curve

PJM©2017 14 7/13/2017
 Hourly Production Example

Parameter Unit X
Hot Start Price ($) 500
Intermediate Start Price ($) 1000
Cold Start Price ($) 1500
No-Load ($/hr) 500

Offer Curve
(max 10 points) MW Price ($/MWh)
segment 1 50 10
segment 2 75 50
segment 3 100 200

Eco Min (MW) 50

Eco Max (MW) 100
Min Run Time (hours) 16

PJM©2017 15 7/13/2017
 Generator Offer Curve – Using Block Offer
• Block Offer rather than a Slope Offer curve
‒ (i.e. Use Slope field in Markets Gateway is unchecked)

• The block offer is a step function rather than a line with a slope
$250
Offer Price ($/MW)

$200

$150

$100

$50

$10
$0

0 50 75 100
Eco Min Output Eco Max
PJM©2017 16 7/13/2017
 Hourly Production Cost - Using Block Offer

• Hourly No-Load = $500

• Cost of 1st segment:

‒ 50 MW * $10 = $500

• Cost of 2nd segment:

‒ (75MW – 50 MW)* $50 = $1,250

• Cost of 3rd segment (Eco Max):

‒ (100 MW – 75 MW)* $200 = $5,000

PJM©2017 17 7/13/2017
 Hourly Production Cost - Using Block Offer

• Hourly production Cost at Eco Min =

No Load + 1st Increment to Min = $500 + $500 = $1,000/hour
• Hourly production Cost at Eco Max =
No Load + 1st Segment + 2nd Segment + 3rd Segment
= $7,250/hour
• Operating Rate =
(Hourly Production Cost at Eco Max / Eco Max MW) =
$7250/hour / 100 MW = $72.50/MWH

PJM©2017 18 7/13/2017
 Total Production Cost

• Calculated by adding all of the costs

associated with starting a unit and
operating it over a period of time

• Two cost components: Total Production Cost

=
‒ Startup Costs
Startup Cost
‒ Hourly Production Costs +
(Hourly Production Cost
x
Number of Hours)

PJM©2017 19 7/13/2017
 Start Cost

• Start Cost (or price) is associated with the cost to supply steam to
operate the turbine and bring the generating unit to synchronous
speed. There are three states for Start Costs (or prices):
‒ Hot
‒ Intermediate
‒ Cold

• Start Cost is the dollars per start as determined from start fuel, total
fuel-related cost, performance factor, electrical costs, start
maintenance adder, and additional labor cost

PJM©2017 20 7/13/2017
 Total Production Cost - Using Block Offer

• Using the results from the hourly production cost problem with min
run time of 16 hours, the range of production cost is:
‒ 16 hours at Eco Min (50 MW):
• $1,000/hour * 16 hours = $16,000
‒ 16 hours at Eco Max (100 MW):
• $7,250/hour * 16 hours = $116,000
‒ Assume unit is still in a “hot” condition
• Add the Hot Start Cost ($500)
‒ Total production cost range:
 $16,500 (eco min) to $116,500 (eco max) per day

PJM©2017 21 7/13/2017
 Generator Offer Curve – Using Slope Offer Curve
• Slope Offer Curve rather than an Block curve
‒ (i.e. Use Slope field in Markets Gateway is checked)

• The Slope Offer Curve is a straight line connected to the offer points
$250
Offer Price ($/MW)

$200

$150

$100

$50

$10
$0
0 50 75 100
Eco Min Output Eco Max
22
PJM©2017 7/13/2017
 Hourly Production Cost - Using Slope Offer Curve

• Hourly No-Load = $500

• Cost of 1st segment:

‒ 50 MW * $10 = $500

• Cost of 2nd segment:

‒ ((75MW – 50 MW)* $10) + ((75MW – 50MW)*($50 - $10) / 2) = $750

• Cost of 3rd segment (Eco Max):

‒ ((100 MW – 75 MW)* $50) + ((100MW – 75MW)*($200 - $50) / 2) = $3,125

PJM©2017 23 7/13/2017
 Hourly Production Cost - Using Slope Offer Curve

• Hourly production Cost at Eco Min =

No Load + 1st Increment to Min = $500 + $500 = $1000/hour
• Hourly production Cost at Eco Max =
No Load + 1st Segment + 2nd Segment + 3rd Segment
= $4,875/hour
• Operating Rate =
(Hourly Production Cost at Eco Max / Eco Max MW) =
$4,875/hour / 100 MW = $48.75/MWH

PJM©2017 24 7/13/2017
 Total Production Cost - Using Slope Offer Curve
• Using the results from the hourly production cost problem with min
run time of 16 hours, the range of production cost is:
‒ 16 hours at Eco Min (50 MW):
• $1,000/hour * 16 hours = $16,000
‒ 16 hours at Eco Max (100 MW):
• $4,875/hour * 16 hours = $78,000
‒ Assume unit is still in a “hot” condition
• Add the Hot Start Cost ($500)
‒ Total production cost range:
 $16,500 (eco min) to $78,500 (eco max) per day

PJM©2017 25 7/13/2017
 Questions?

PJM©2017 26 7/13/2017
 Generation Dispatch

PJM State & Member Training Dept.

PJM©2017 27 7/13/2017
 Objectives

• Explain the difference between Dispatch Rate and LMP

PJM©2017 28 7/13/2017
 Agenda

• Generation Dispatch

PJM©2017 29 7/13/2017
 Dispatch Rate

Definition:
The Dispatch Rate is expressed in dollars per MWh, calculated and
transmitted to each generator to direct the output level of all
generation resources dispatched by PJM based on the incremental offer
data which was previously received from the Generators

Where PJM wants

Dispatch Rate Instruction The units to be
loaded economically

PJM©2017 30 7/13/2017
 Dispatch Rate
The Dispatch Rate is determined by the PJM economic dispatch solution as
calculated by PJM’s Security Constrained Economic Dispatch program (SCED)
$35
$30
Dispatch Rate = $25
Offer Price $25
$20
0 50 100 150 200
MW Economic Basepoint

The Economic Basepoint is the MW value sent to the generating unit that
indicates to what level the unit should be loaded based on the economic
dispatch solution and the units incremental price curve
PJM©2017 31 7/13/2017
 Transmission Losses
• Real Power (MW) Losses
‒ Power flow converted to heat in transmission equipment
‒ Heat produced by current (I) flowing through resistance (R)
‒ Losses equal to I2R
‒ Heat loss sets the “thermal rating” of equipment

𝐻𝑒𝑎𝑡 𝐷𝑖𝑠𝑖𝑝𝑎𝑡𝑒𝑑 = 𝐼 2 𝑅

PJM©2017 32 7/13/2017
 Transmission Losses

• Real Power (MW) Losses

‒ Increase with line length
• Increased R
‒ Increase with increased current flow (I)
‒ Increase at lower voltages
• Higher currents

Power = Current
* Voltage

PJM©2017 33 7/13/2017
 Transmission Losses

50 Miles

Power Out: 90.946 MW

Power In: 100 MW
Power Loss: 9.054 MW
Voltage In: 235 KV
Voltage out: 213.72 KV
Current In: 425.53 A
Current Out: 425.53 A

PJM©2017 34 7/13/2017
 Transmission Losses

10 Miles

Power Out: 98.2 MW

Power In: 100 MW
Power Loss: 1.8 MW
Voltage In: 235 KV
Voltage out: 230.74 KV
Current In: 425.53 A
Current Out: 425.53 A

PJM©2017 35 7/13/2017
 Penalty Factors Effect on Dispatch
• The Incremental Loss for bus i is used to calculate a factor that can be used to include
the effect of losses in the dispatch
• This factor is called the Loss Penalty Factor, or Penalty Factor
1
Pf i 
 PL  Change in Losses

 1  

 Pi  Change in Unit’s MW Output

• The Penalty Factors adjust the incremental cost of each generator so as to include the
effects of losses
• Penalty factors applied to each and every location
‒ Including generation, load, virtual transaction
PJM©2017 36 7/13/2017
 Penalty Factors Effect on Dispatch
• If an increase in generation results in an increase in system losses then:
‒ Penalty factor is greater than 1
‒ Units offer curve is adjusted higher
• Unit offer curve is multiplied by penalty factor
• Unit looks less attractive to dispatch

Loss Factor Penalty Factor

PL 1
0 1 Pf i   1.0
Pi  PL 

1  P 

Increase in injection will result in  i 
higher overall system losses

PJM©2017 37 7/13/2017
 Penalty Factors Effect on Dispatch
• If an increase in generation results in a decrease in system losses then:
‒ Penalty factor is less than 1
‒ Units offer curve is adjusted lower
• Unit offer curve is multiplied by penalty factor
• Unit looks more attractive to dispatch
• Total LMP would still at least equal unit’s original offer

Loss Factor Penalty Factor

PL 1
0  1 Pf i   1.0
Pi  PL 

1  P 

Increase in injection will result in  i 
lower overall system losses
PJM©2017 38 7/13/2017
 Penalty Factors Effect on Dispatch - Example # 1

19.89 $/MWh

Generating Unit # 1 Generating Unit # 2

Offer Price = $ 10.00 ----- 200 MW Offer Price = $ 10.00 ----- 200 MW
$ 20.00 ----- 300 MW $ 20.00 ----- 300 MW
$ 30.00 ----- 400 MW $ 30.00 ----- 400 MW
$ 40.00 ----- 500 MW $ 40.00 ----- 500 MW
Generating 300 MW Generating 305 MW
Penalty Factor = 1.00 Penalty Factor = 0.97
$20 * 1.00 = $20.00 $20.50 * 0.97 = $19.89

PJM©2017 39 7/13/2017
 Dispatch Optimized to Least Production Cost

PJM©2017 40 7/13/2017
 Offer Curve for Both Units
50

45

40

35

30
($/MWh)

25

20

15

10

0
0 100 200 300 400 500 600
(MW)

PJM©2017 41 7/13/2017
 Optimal Dispatch – Unit 1 Production Cost
50
Production Cost at 200 MW = $2,000
45
Production Cost at 300 MW = $2,000 + $1,000 + $500 = $3,500
40

35

30
($/MWh)

25

20

15
$500
10

5 $2,000 $1,000
0
0 100 200 300 400 500 600
(MW)

PJM©2017 42 7/13/2017
 Optimal Dispatch – Unit 2 Production Cost
50
Production Cost at 305 MW = $2,000 + $1,000 + $500 + $100 + $1.25 = $3,601.25
45

40

35

30
($/MWh)

25

20 $1.25

15
$500 $100
10

5 $1,000
$2,000
0
305
0 100 200 300 400 500 600
(MW)

PJM©2017 43 7/13/2017
 Linear Interpolation to determine the $/MW between 300 and 400 MW
𝑑𝑦 𝑟𝑖𝑠𝑒
Slope = 𝑚 = =
𝑑𝑥 𝑟𝑢𝑛
$(30−20)
𝑚= = $. 1/MW
𝑀𝑊(400−300)

Therefore:
(𝑦1 −20)
0.1 =
(305−300)

0.5 = (𝑦1 − 20)

𝑦1 = 20.5

PJM©2017 44 7/13/2017
 Alternate Dispatch 1 – Unit 2 Production Cost
50
Production Cost at 405 MW = $2,000 + $1,000 + $500 + $2000 + $500 + $150 + $1.25 = $6,151.25
45

40

35

30 $1.25
($/MWh)

25
$500
20
$150
15
$500
10

5 $1,000 $2,000
$2,000
0
0 100 200 300 400405 500 600
(MW)

PJM©2017 45 7/13/2017
 Alternate Dispatch 1 – Unit 1 Production Cost
50
Production Cost at 200 MW = $2,000
45

40

35

30
($/MWh)

25

20

15

10

5 $2,000
0
0 100 200 300 400 500 600
(MW)

PJM©2017 46 7/13/2017
 Alternate Dispatch 2 – Unit 1&2 Production Cost
50
Production Cost at 302.5 MW = $2,000 + $1,000 + $500 + $50 + $0.313 = $3,550.313
45

40

35

30
($/MWh)

25

20 $0.313

15
$500 $50
10

5 $1,000
$2,000
0
302.5
0 100 200 300 400 500 600
(MW)

PJM©2017 47 7/13/2017
 Questions?

PJM©2017 48 7/13/2017
 Locational Marginal Pricing
Components

PJM State & Member Training Dept.

PJM©2017 49 7/13/2017
 Agenda

• LMP Components
• 5 Bus Model
• Shadow Prices
• Aggregated Pricing Points
• Statistics
• LMP Simulation Demo

PJM©2017 50 7/13/2017
 What is LMP?

• Pricing method PJM uses to:

‒ price energy purchases and sales in PJM Market
‒ price transmission congestion costs to move energy within PJM RTO
‒ price losses on the bulk power system

PJM©2017 51 7/13/2017
 How does PJM Use LMP?

• Generators get paid at generation bus LMP

• Loads pay at load bus LMP
• Transactions pay differential in source and sink LMP

PJM©2017 52 7/13/2017
 Locational Marginal Price

System Marginal
Congestion
Marginal Loss
Component
Price Component

 System Marginal Price (SMP)

• Incremental price of energy for the system, given the current dispatch,
at the load weighted reference bus
• SMP is LMP without losses or congestion
• Same price for every bus in PJM (no locational aspect)
• Calculated both in day ahead and real time

PJM©2017 53 7/13/2017
 Locational Marginal Price

System Marginal
Congestion
Marginal Loss
Component
Price Component

 Congestion Component (CLMP)

• Represents price of congestion for binding constraints
‒ Calculated using the Shadow Price
• Will be zero if no constraints (Unconstrained System)
‒ Will vary by location if system is constrained
• Used to price congestion
‒ Load pays Congestion Price
‒ Generation is paid Congestion Price
• Calculated both in day ahead and real time
PJM©2017 54 7/13/2017
 Operational Limits

• Thermal Limits - Thermal limits are due to the thermal capability of

power system equipment

• Voltage Limits - Utility and customer equipment is designed to

operate at a certain supply voltage

• Stability Limits - Refers to the power system

maintaining a state of equilibrium

PJM©2017 55 7/13/2017
 Control Actions

• There are three basic types of actions that can be performed to

control the flow of power on the electric system:

 System Reconfiguration
 Transaction Curtailments
 Redispatch Generation

PJM©2017 56 7/13/2017
 When Constraints Occur...

• Delivery limitations prevent use of “next least-cost generator”

• Higher-cost generator closer to load must be used to meet demand
• Cost expressed as “security constrained redispatch cost”

PJM©2017 57 7/13/2017
 Security Constrained Re-Dispatch
Control Area
Constrained System

Low Cost
Generator
$$ High Cost Generator
$$$$

Higher cost Generator more

Transmission advantageously located relative
“Bottleneck” to transmission system limit
or Constraint
PJM©2017 58 7/13/2017
 Congestion effects on LMP and Revenues

• When the bus is upstream of a constraint

‒ Congestion Component is negative
‒ Results in negative revenues to unit

• When the bus is downstream of a constraint

‒ Congestion Component is positive
‒ Results in positive revenues to unit

PJM©2017 59 7/13/2017
 Constraints & Marginal Units

• There will always be at least one marginal unit

‒ System Energy Unit
• There will be an additional marginal unit for each binding constraint
• It is possible and, in fact likely, that there will be multiple marginal
units for a given time interval

PJM©2017 60 7/13/2017
 Locational Marginal Price
System Marginal
Congestion
Marginal Loss
Component
Price Component

Marginal Loss Component (MLMP)

• Represents price of marginal losses
‒ Transmission losses are priced according to marginal loss factors which are calculated
at a bus and represent the percentage increase in system losses caused by a small
increase in power injection or withdrawal
• Calculated using penalty factors

• Will vary by location

• Used to price losses
‒ Load pays the Loss Price
‒ Generation is paid the Loss Price
• Calculated both in day-ahead and real-time
PJM©2017 61 7/13/2017
 Marginal Loss effects on LMP and Revenues

• When the bus is electrically distant from the load

‒ Marginal Loss Component is negative
‒ Results in negative revenues to unit

• When the bus is electrically close to the load

‒ Marginal Loss Component is positive
‒ Results in positive revenues to unit

PJM©2017 62 7/13/2017
 What would you expect to see?

Congestion Component of LMP? (-)

Loss Component of LMP? (-)

30 miles

Congestion Component of LMP? (+)

Loss Component of LMP? (+)
PJM©2017 63 7/13/2017
 Agenda

• LMP Components
• 5 Bus Model
• Shadow Prices
• Aggregated Pricing Points
• Statistics
• LMP Simulation Demo

PJM©2017 64 7/13/2017
 LMP Examples
 5-Bus Model Examples

PJM©2017 65 7/13/2017
 Example # 1 - 5 Bus Transmission Grid
Generator Offers System Loads = 669 MW
System Losses = 17 MW Sundance
600 MW 200 MW
$10/MWh 230 MW
E Thermal Limit D $40/MWh
Brighton

223 MW
A
Alta Solitude
223 MW
110 MW B 223 MW C 520 MW
Park City $30/MWh
$14/MWh 100 MW
$15/MWh
PJM©2017 66 7/13/2017
 Example # 1 - 5 Bus Transmission Grid
Dispatch & Energy Flow System Loads = 669 MW
System Losses = 17 MW
230 MW Sundance
E Thermal Limit D
Brighton 200 MW
225 $40/MWh
600 MW 152 PF = 1.0247

375
223 MW

149
600 MW 305 77
$10/MWh A
PF = 1.0625 86 MW
Alta Solitude
223 MW 223 MW 520 MW
Park City B C $30/MWh
110 MW 100 MW PF = 1.0000
$14/MWh $15/MWh
PF = 1.0492 PF = 1.0492
PJM©2017 67 7/13/2017
 LMP Calculations

• System Energy Price = LMP at the Reference Bus

(where Congestion & Losses = 0)

• Reference or “Slack” Bus is

the “electrical load center” of the system

• Losses are calculated System 1

using the System Energy * 1
Energy Price & Price Pf
the Penalty Factor (Pf)
PJM©2017 68 7/13/2017
 Example # 1 - Summary
Offer Price Penalty Factor Adjusted Offer System Energy Loss Price Congestion Total LMP
Price Price
Unit
Unconstrained
System

Brighton $10.00 1.0625 $10.625 $14.69 -$0.86 $0.00 $13.83


Alta $14.00 1.0492 $14.688 $14.69 -$0.69 $0.00 $14.00

Park City $15.00 1.0492 $15.738 $14.69 -$0.69 $0.00 $14.00

Solitude $30.00 1.0000 $30.000 $14.69 $0.00 $0.00 $14.69

Sundance $40.00 1.0247 $40.988 $14.69 -$0.35 $0.00 $14.33

= + + =
PJM©2017


*
Unit Running
Unit Not Running
69
Loss and Congestion Components of LMP
are “0” at the Reference Bus
7/13/2017
 Example # 1 - 5 Bus Transmission Grid
LMPs Area Load = 669
Area Losses = 17 MW
Area Generation = 686

LMP = $13.83 230 MW LMP = $14.33

Thermal Limit Sundance
E D 200 MW
Brighton 225
$40/MWh
600 MW 152
375
PF = 1.0247

149
223 MW
600 MW 305 77
A
86 MW
$10/MWh Solitude
PF = 1.0625 Alta
LMP = $14.00 223 MW 223 MW 520 MW LMP = $14.69
Park City B C $30/MWh PF = 1.000
110 MW 100 MW
Marginal Unit
$14/MWh $15/MWh
PF = 1.0492 PF = 1.0492 Reference Bus
PJM©2017 70 7/13/2017
 Agenda

• LMP Components
• 5 Bus Model
• Shadow Prices
• Aggregated Pricing Points
• Statistics
• LMP Simulation Demo

PJM©2017 71 7/13/2017
 Binding Constraints and Shadow Prices

• Binding constraints limit the ability to improve the objective function

‒ If a binding constraint is relaxed, or made less restrictive, a better solution
is possible
• The shadow price is the marginal improvement caused by relaxing
the constraint
‒ In energy markets, a shadow price shows the savings in Bid Production Cost if
binding constraint is relaxed by 1MW
• Shadow prices tell us how much more money
we can make (or save) by improving one of our
limiting factors or boundary conditions

PJM©2017 72 7/13/2017
 Shadow Price
$60 $90
Area 1 Area 2
G1 G3 ON
ON

ON G2
Limit = 400MW
ON G4

OFF G5

Load = 200MW Load = 600MW

Total Production Cost = (600*60) + (200*90) = $54,000

PJM©2017 73 7/13/2017
 Shadow Price
$60 $90
Area 1 Area 2
G1 G3 ON
ON

ON G2
Limit = 401MW
ON G4

OFF G5

Load = 200MW Load = 600MW

Total Production Cost = (601*60) + (199*90) = $53,970

PJM©2017 74 7/13/2017
 Shadow Prices

• (Before: 400 MW limit ) Total production cost is $54,000

• (After: 401 MW limit) Total production cost is $53,970
• “Relaxing” constraint limit by 1 MW saved us $30
in total production costs
• Difference between the “Before” and
“After” case is the Shadow price = $30

PJM©2017 75 7/13/2017
 LMP Components
System
Energy
Price = System Energy Price

System
Energy
Price x Marginal loss
Sensitivity factor = Marginal Loss Component

ConstraintA
Shadow Price x DFAXA
= Congestion ComponentA

LMP
PJM©2017 76 7/13/2017
 System Energy Price X * 1.0 = System Energy Component
System Energy Component $33.11 X 1.0 = $33.11

Marginal Loss
System Energy Price X Sensitivity Factor = Marginal Loss Component
Loss Component $33.11 X -0.0315 = ($1.04)

Congestion Components Constraint Shadow Price X DFAX = Congestion Component

Constraint A -$9.96 X -0.3151 = $3.14
Constraint B -$13.88 X 0.1225 = ($1.70)
Constraint C -$26.06 X -0.2151 = $5.61
Constraint D -$5.48 X -0.0200 = $0.11

LMP = $39.23
• Which constraints does raising unit output help?
• Which constraints does raising unit output hurt?
• Is close to center of system load?
• Bonus Question – How many marginal units does this system have?
PJM©2017 77 7/13/2017
 System Energy Price X * 1.0 = System Energy Component
System Energy Component $33.11 X 1.0 = $33.11

Marginal Loss
System Energy Price X Sensitivity Factor = Marginal Loss Component
Loss Component $33.11 X -0.0315 = ($1.04)

Congestion Components Constraint Shadow Price X DFAX = Congestion Component

Constraint A -$9.96 X -0.3151 = $3.14
Constraint B -$13.88 X 0.1225 = ($1.70)
Constraint C -$26.06 X -0.2151 = $5.61
Constraint D -$5.48 X -0.0200 = $0.11

LMP = $39.23
• Which constraints does raising unit output help? Constraints A,C and D
• Which constraints does raising unit output hurt? Constraint B
• Is close to center of system load? No
• Bonus Question – How many marginal units does this system have? 5

PJM©2017 78 7/13/2017
 Example # 2 - 5 Bus Transmission Grid
Constrained System Loads + Losses = 921

230 MW
E Thermal Limit D Sundance
Brighton 200 MW
$40/MWh
600 MW Load = 300 MW
$10/MWh A
Alta Solitude
110 MW
B C 520 MW
Park City $30/MWh
$14/MWh 100 MW Loads = 300 MW
$15/MWh
PJM©2017 79 7/13/2017
 Example # 2 - 5 Bus Transmission Grid
Dispatch Solution Ignoring Thermal Limit System Loads = 900 MW
System Losses = 21 MW
230 MW
Dispatched E Thermal Limit D
at 600 MW 200 MW
258 $40/MWh
Brighton
193 Sundance

342

144
300 MW
600 MW 355 48
$10/MWh A Dispatched
110 MW
Alta Solitude
300 MW 300 MW 520 MW
110 MW
$14/MWh
Park City B C $30/MWh
100 MW
Dispatched $15/MWh
at 110 MW Dispatched
PJM©2017 100 MW80 7/13/2017
 Example # 2 - 5 Bus Transmission Grid
Actual Dispatched Generation System Loads = 900 MW 200 MW
System Losses = 15 MW $40/MWh
Dispatched 230 MW Sundance
E Thermal Limit D
at 509 MW
230 224
Brighton
509 MW 279 178

278

180

102
300 MW
600 MW 110 MW 308
A 3

100 MW
$10/MWh 101
196 MW
Alta Solitude
300 MW 300 MW Dispatched 520 MW
110 MW at 196 MW $30/MWh
$14/MWh
Park City B C
100 MW
Dispatched
$15/MWh
at 110 MW Dispatched
PJM©2017
at 100 MW81 7/13/2017
 Calculate Shadow Price and Congestion Price
Production Cost calculated
Production Cost with 230 MW across Production Cost with 231 MW across
Brighton - Sundance line Brighton - Sundance line
Unit MW Price No Load Production Cost Unit MW Price No Load Production Cost
Brighton 509 10 $399.80 $5,489.80 Brighton 512 10 $399.80 $5,519.80
Alta 110 14 $100.00 $1,640.00 Alta 110 14 $100.00 $1,640.00
Park City 100 15 $100.00 $1,600.00 Park City 100 15 $100.00 $1,600.00
Solitude 196 30 $100.00 $5,980.00 Solitude 193 30 $100.00 $5,890.00
Sundance 0 40 $100.00 $0.00 Sundance 0 40 $100.00 $0.00
915 $14,709.80 915 $14,649.80

Shadow Price = $14,679.80 -$14,739.80 = -$60.00

Bus Monitored Line DFAX Shadow Price Congestion Price

Brighton Brighton - Sundance 0.307167 -$60.00 -$18.43
Alta Brighton - Sundance 0.199167 -$60.00 -$11.95
Park City Brighton - Sundance 0.199167 -$60.00 -$11.95
Solitude Brighton - Sundance 0 -$60.00 $0.00
Sundance Brighton - Sundance -0.16367 -$60.00 $9.82

PJM©2017 82 7/13/2017
 Example # 2 – Summary
Offer Price Penalty Adjusted System Loss Price Congestion Total LMP
Factor Offer Energy Price Price
Unit
(Shadow Price * DFAX)

Brighton $10.00 1.0553 $10.5530 $30.00 -$1.57 -$18.43 $10.00


Alta $14.00 1.0449 $14.6286 $30.00 -$1.29 -$11.95 $16.76

Park City $15.00 1.0449 $15.6735 $30.00 -$1.29 -$11.95 $16.76

Solitude $30.00 1.0000 $30.0000 $30.00 $0.00 $0.00 $30.00

Sundance $40.00 1.0161 $40.6440 $30.00 -$0.47 $9.82 $39.35

= + + =
PJM©2017


*
Unit Running
Unit Not Running
83
Loss and Congestion Components of LMP
are “0” at the Reference Bus
7/13/2017
 Agenda

• LMP Components
• 5 Bus Model
• Shadow Prices
• Aggregated Pricing Points
• Statistics
• LMP Simulation Demo

PJM©2017 84 7/13/2017
 Standardized Trading Location Definitions
Rationale Aggregate Pricing Points – hubs and zones –
create a consolidated point for forward contracts to settle,
• Need standardized trading with or without physical delivery
locations to support development
of liquid forward market
• Aggregate locations serve to
consolidate liquidity
Northern
• Hubs aggregate a group of Illinois Hub
AD Hub
representative, price-correlated
buses
• Aggregated load zones were
created to smooth the impacts of
LMP volatility on (non- Western Eastern
dispatchable) load – easing state Hub Hub

consumer protection concerns

PJM©2017 85 7/13/2017
 Agenda

• LMP Components
• 5 Bus Model
• Shadow Prices
• Aggregated Pricing Points
• Statistics
• LMP Simulation Demo

PJM©2017 86 7/13/2017
 2016 PJM State of the Market Report - LMP

PJM©2017 87 7/13/2017
 2016 PJM State of the Market Report - LMP

PJM©2017 88 7/13/2017
 2016 PJM State of the Market Report - LMP

PJM©2017 89 7/13/2017
 2016 PJM State of the Market Report - LMP

PJM©2017 90 7/13/2017
 Agenda

• LMP Components
• 5 Bus Model
• Shadow Prices
• Aggregated Pricing Points
• Statistics
• LMP Simulation Demo

PJM©2017 91 7/13/2017
 Questions?

PJM©2017 92 7/13/2017
 Base Case – One Marginal Unit

PJM©2017 93 7/13/2017
 Rainy 3 - Marginal Unit – Base Case

PJM©2017 94 7/13/2017
 Base Case with New Marginal Unit

PJM©2017 95 7/13/2017
 McCoy 2 - Marginal Unit – Modified Base

PJM©2017 96 7/13/2017
 High Load Case

Constraint
1. Rainy – Dooms 500 KV line
• Shadow Price = -$371.43
PJM©2017 97 7/13/2017
 Pilgrim CC 1 - Marginal Unit – High

PJM©2017 98 7/13/2017
 Rainy 2 - Marginal Unit – High Load

PJM©2017 99 7/13/2017
 Superior Units – High Load Case

PJM©2017 100 7/13/2017

 Not Following PJM and Causing Additional Constraint

Constraint
1. Rainy – Dooms 500 KV line
• Shadow Price = -$444.06
2. Bed – Dooms 500 KV line
1. Shadow Price = -$104.52
PJM©2017 101 7/13/2017
 Superior Units – Not Following PJM

PJM©2017 102 7/13/2017

 Questions?
PJM Client Management & Services
Telephone: (610) 666-8980
Toll Free Telephone: (866) 400-8980
Website: www.pjm.com

The Member Community is PJM’s self-service portal for members to search for answers to their questions
or to track and/or open cases with Client Management & Services

PJM©2017 103 7/13/2017