PJM Reliability Pricing Model Bhavaraju 200706271 -...

PJM ©2007

PJM Reliability Pricing Model

Murty P. BhavarajuPJM InterconnectionIEEE/PES General Meeting Tampa, FLJune 27, 2007

Note: Presentation by B. Hobbs et al. on the dynamic analysis of the RPM is attached

©2007 PJMwww.pjm.com 2

Forward Procurement of Capacity Resources

• Reliability Pricing Model (RPM) is PJM’s new resource adequacy construct that will replace the existing capacity construct effective June 1, 2007.

• RPM’s FORWARD procurement of resources through a base residual auction three years prior to the delivery year:Provides long-term price signal for capacity resources.Encourages Load Serving Entities (LSEs) to make long-term bilateral contracts to hedge their locational reliability charges.Supports the Regional Transmission Expansion Planning Process (RTEPP).


Locational Value of Capacity Resources

• RPM recognizes import capability limitations in certain areas as identified in the PJM Regional Transmission Expansion Planning Process (RTEPP).

• RTEPP has currently identified 23 sub-regions as Locational Deliverability Areas (LDAs) for evaluating the locational constraints.

• LDA with import capability limit less than 105% of import capability requirement will be modeled as constrained LDA in RPM.

• RPM auction could result in higher capacity prices in the constrained areas that would encourage new generation additions or keep generating units from retiring.


Variable Resource Requirement Stabilizes Prices

• Variable Resource Requirement (VRR) is also referred to as Demand Curve.

• Defines a relationship between level of reserve and capacity price based on the net annual cost of a new combustion turbine.

• Recognizes the value of additional capacity above the reserve required to meet the reliability criterion.

• Higher price accepted when there is shortage in meeting the required capacity.

• VRR curves established for PJM Region and each constrained Locational Deliverability Area (LDA).


RTO Variable Resource Requirement Curve

0.0050.00

100.00150.00200.00250.00300.00

139723

.7141

013.0

14230

2.414

3591.8

14488

1.1146

170.5

147459

.914

8749.2

15003

8.615

1328.0

15261

7.3

Quanitity, UCAP MW

UCAP

Pric

e, $

/MW

-Day

RTO VRR Curve

RTO Variable Resource Requirement Curve


Participation in RPM

• Participation by LSEs for load served in PJM region is mandatory, except for those LSEs that have elected Fixed Resource Requirement (FRR) Alternative.– Each LSE shall be responsible for paying a Locational

Reliability Charge.– May choose to hedge Locational Reliability Charges

by offering capacity into the auction.• Participation by resource providers is subject to

the market power mitigation rules described in the tariff.


Supply in RPM Auctions

• Eligible Capacity Resources:– Existing & planned generation in PJM– Existing external generation– Bilateral contracts for unit-specific capacity resources– Existing & planned demand resources– Qualifying Transmission Upgrades

• Resources must meet the requirements specified in PJM Agreements and Business Rules.


2007/2008 RPM Auction Resource Clearing Prices, $/MW-Day

$40.80

$188.54

$197.67

$188.54

$197.67

$40.80

Resource Clearing

Price [$/MW-day]

$140.16 $147.74 $40.80 SWMAAC

$177.51 $156.87 $40.80 EMAAC

$40.80 $0.00 $40.80 RTO

Final Zonal ILR Price/

Prelim. Load Obligation

Rate [$/MW-day]

LocationalPrice Adder [$/MW-day]

System Marginal

Price [$/MW-day]LDA


EMAAC 07/08

$0.00

$50.00

$100.00

$150.00

$200.00

$250.00

36000.0 36500.0 37000.0 37500.0 38000.0 38500.0 39000.0

07/08 EMAAC VRR Curve intersect

Eastern MAAC Results


SWMMAC 07/08

$0.00

$50.00

$100.00

$150.00

$200.00

$250.00

15400.0 15600.0 15800.0 16000.0 16200.0 16400.0 16600.0 16800.0

07/08 SWMAAC VRR Curve intersect

Southwestern MAAC Results


RTO 07/08

$0.00

$50.00

$100.00

$150.00

$200.00

$250.00

$300.00

120000 121000 122000 123000 124000 125000 126000 127000 128000 129000 130000 131000

07/08 RTO VRR Curve intersect

RTO Results (Excluding FRR Obligations)


Discussion of Results

• 127.6 MW of demand response cleared • Approx. 350 MW of new capacity offered of

which 311 MW cleared• Total value of CTRs = $1.48 million/day• Resource clearing prices in Eastern MAAC and

Southwestern MAAC above net Cost of New Entry

• Uncleared MW in RTO: 1,443.5 (UCAP)


Timing of RPM Auctions

Base Residual Auction

Delivery Year

3 Years

Second Incremental Auction

Third Incremental Auction

June May4 months

13 months

First Incremental Auction

23 months

EFORd Fixed

Ongoing Bilateral Market – (shorter-term reconfiguration)

Interruptible Load for

Reliability (ILR)


RPM Auction Process

Optimization Algorithm

Supply Resource-

specific Sell Offers

Demand•Variable Resource Requirement (VRR)

Curves for Base Residual Auction•Locational Buy Bids for Incremental

Auctions

Locational Constraints

Auction Results•Resource Commitments•Resource Clearing Prices•UCAP Obligation values•Capacity Transfer Rights

Zonal CapacityPrices

ILR PricesCTR Rates


Bilateral Market

• Provides LSEs the opportunity to self-supply and hedge against the Locational Reliability Charge determined through the Base Residual and Second Incremental Auction.

• Provides resource providers an opportunity to cover any commitment shortages due to resource cancellations, delays, deratings, or EFORd increases, or decrease in nominated value of a planned demand resource.



• Allows for procurement of unit-specific resource commitments required, after accounting for self-supply, to satisfy the region’s unforced capacity obligation for a future Delivery Year (less an amount reserved for Interruptible Load for Reliability (ILR)).

• Cost of procurement is allocated to LSEs serving load in the actual Delivery Year through the Locational Reliability Charge.


Incremental Auctions: First and Third

• Allow for an incremental procurement of resource commitments for future Delivery Year to accommodate adjustments to participants’ resource positions due to resource cancellations, delays, deratings, or EFORd increases, etc.

• Buyers pay suppliers with no change in the Locational Reliability Charge assessed to LSEs during the Delivery Year.


Incremental Auctions: Second Incremental Auction

• Second Incremental Auction is held ONLY to mitigate reliability concern. – If increased obligation (due to increase in

peak load forecast) compared with capacity cleared in the Base Residual Auction > 100 MW UCAP, the auction is conducted.

• Cost of procurement is allocated to LSEs serving load during the actual Delivery Year through the Locational Reliability Charge.


Locational Deliverability Areas (LDA)

1. AE2. AEP3. APS4. BGE5. ComEd6. Dayton7. DLCO8. Dominion9. DPL10.JCPL11.MetEd12.PECO 13.Penelec14.PEPCO15.PPL16.PSEG

17.Mid-Atlantic Area Council (MAAC) Region 18.ComEd, AEP, Dayton, APS, and

Duquesne19.Eastern MAAC (PSE&G, JCP&L, PECO,

AE, DPL & RECO)20.Southwestern MAAC (PEPCO & BG&E)21.Western MAAC (Penelec, MetEd, PPL)22.PSEG northern region (north of Linden

substation); and23.DPL southern region (south of

Chesapeake and Delaware Canal)

PJM required to make a filing with FERC before adding a new LDA.


LDAs for Transition Period

All 23 LDAs2010/2011

•PJM Mid-Atlantic Region and APS•Eastern MAAC (PSE&G, JCP&L, PECO, AE, DPL, and RECO)•Southwestern MAAC (PEPCO & BG&E)•Rest of Market - ComEd, AEP, Dayton, Dominion and Duquesne

2007/2008, 2008/2009, & 2009/2010

Locational Deliverability Areas

Transition Delivery Year


Zonal Capacity Prices

Base Residual Auction Results

Second Incremental

Auction Results

ILR Certification

•Preliminary Zonal Capacity Prices•Preliminary Zonal ILR Prices•Base Zonal Capacity Transfer Right (CTR) Credit Rate

•Adjusted Zonal Capacity Prices•Final Zonal ILR Prices•Final Zonal CTR Credit Rate

Final Zonal Capacity Prices

Locational Reliability Charge = Daily Zonal UCAP Obligation * Final Zonal Capacity Price


Unforced Capacity Obligations•Zonal W/N Summer Peak – 4 yr•Preliminary RTO Peak Load Forecast•Forecast Pool Requirement

Base RTO UCAP

Obligation

Base Zonal RPM Scaling

Factors

Base Zonal UCAP

Obligation

•Zonal W/N Summer Peak – 4 yr•Forecast Pool Requirement•Forecast Zonal ILR Obligation

•Zonal W/N Summer Peak – 1 yr•Forecast Pool Requirement

•Preliminary Zonal Peak Load Forecast•Final Zonal Peak Load Forecast

Zonal Allocation of Incremental Obligation

Satisfied in 2nd IA

Final Zonal UCAP

Obligation

Final Zonal RPM Scaling

Factors

•LSE Allocation of Zonal W/N Summer Peak – 1 yr

LSE Obligation Peak Load

LSE Daily UCAP

Obligation


Second Incremental

Auction

Delivery Year


Summary of RPM Activities

Pre-Delivery Year Activity• RPM Auctions

– Base Residual Auction– 1st Incremental Auction– 2nd Incremental Auction– 3rd Incremental Auction

• Interruptible Load for Reliability Nomination

Delivery Year Activity• Auction Credits/Charges• ILR Credits• Daily Unforced Capacity

Obligations & Locational Reliability Charges

• CTR Credits• Resource Performance

Assessments• Deficiency & Penalty

Charges/Credits• Non-Unit Specific Capacity

Transactions Charges/Credits

On-going Bilateral Market

Model-based Assessment of the PJM Reliability Pricing Model

Benjamin F. Hobbs, Ming-Che Hu, and Javier Inon

Whiting School of EngineeringThe Johns Hopkins University

[email protected]

Murthy Bhavaraju

PJM Interconnection, LLC

This is a summary of work conducted at JHU sponsored by PJM Interconnection; however, the authors are solely responsible for all opinions expressed, which do not necessarily represent the

position of the sponsor.

Outline

Dynamic Model Analysis of PJM RPMQuestions askedModel assumptions & structure• Simple model of representative agent• Transient simulation with random shocksResultsConclusions

Dynamic Analysis: Questions

1. How do different curves affect….• Stability of capacity market?• Costs to consumers? • Ability to meet reserve requirement, reliability

criterion?

2. How robust are these conclusions to different assumptions about:• Generator behavior? • Demand curve parameters?

Dynamic Analysis: Basic Assumptions

Capacity additions are a dynamic process. Investment depends on:1. Forecast revenue streams

– Based on capacity and energy prices from recent auctionsMore forecast net revenue investment ↑

2. Revenue stream variability– Variations due to forecast changes and weather

Highly variable energy and capacity prices investment ↓(due to risk aversion)

3. Risk attitudes: – No hedges (incomplete market) – Risk aversion– Short-sightedness

Random shocks (weather, economic fluctuations) cause variation in returns

• Result: boom/bust cycles in investment

Use of demand curve changes the market dynamics

Dynamic Model Overview

1. Simple & transparent model simulates: • annual construction of turbine capacity,

• revenues from energy, ancillary services, & capacity markets,

• market stability in face of random demand shocks,

• consumer costs

2. Allows exploration of assumptions3. The model assesses profitability of CTs needed to

meet the reliability requirement • Other generation types and profitability not modeled• “Representative Agent” approach• Annual time step

Simulation Overview: Auction in Year y-4for Capacity Installed by Year y: Repeated for 100 years

Risk-Adjusted Forecast Profit (RAFPy)(↑ if profits higher, ↓ if profits more variable)

Maximum New Capacity Additions NCAy

Capacity Price from Demand Curve(Assume existing capacity bids 0, and NCAy bids B)

Year y-7:Profit =

PRPM + E/ASGross Margin– Fixed Cost

Year y-6:PRPM

+E/AS GM

– FC

Year y-5:PRPM

+E/AS GM

– FC

Year y-4:PRPM

+E/AS GM

– FC

Year y-3:PRPM

+E/AS GM

– FC

Year y-2:PRPM

+E/AS GM

– FC

Year y-1:PRPM

+E/AS GM

– FC

Year y:PRPM

+E/AS GM

– FC

Actual and Estimated Profits: Blue = Known at Auction in Year y-4; Brown = Estimated

Forecast Weights for Profits in y-7, …, y

Risk averse utility functionpenalizes variable profits

NCAy

1.7%

0% RAFPy

PRPM,y

0Total CAP

Risk Adjusted Forecast Profit (RAFP) Calculation

Actual and anticipated combustion turbine profits π• π = Gross Margin (Energy, A/S) + PRPM

• Gross margin based on reserve margins, 1999-2004 experience

• RAFP = single profit with same utility as 7 years of experienced/estimated profit

0

10000

20000

30000

40000

50000

60000

70000

80000

0.95 1 1.05 1.1 1.15

Ratio of (Unforced Reserve Margin) / (Target IRM)

E/AS

Net

Rev

enue

$/In

stal

led

MW

/yr

$10,000 + Simulated Scarcity Rent

Historical E/AS Net Revenues

```

Determination of Capacity Price

PCAP,y

B

Total Cap

Capacity Demand CurveCapacity Bid Curve

Existing Capacity

Max New Capacity

New CapacityChosen

PJM Analysis: Five Curves Considered

Vertical Demand

Results: Summary

1. Sloped curve stabilizes capacity payments

2. More stable payments even out investment, forecast reserves

3. More stable revenues lowers capital costs. Consumer costs (capacity, scarcity) fall:• $129/peak kW/yr for

vertical• $71/peak kW/yr for

sloped

4. Results robust 0.96

0.98

1.00

1.02

1.04

1.06

1.08

0 20 40 60 80 100

Time

Rese

rve/IR

M R

atio

.

VRR (IRM+1%)Vertical at Target IRM

0

40,000

80,000

120,000

160,000

0 20 40 60 80 100

Time

Capa

city P

rice (

$/MW

/yr)

.

VRR (IRM+1%)

Vertical at Target IRM

(values depend on assumptions)

Sample Results: Average (Risk aversion parameter = 0.7; Results depend on specific assumptions)

14

21

26

37

47

ScarcityRev.

$/kW-yr

10

10

10

10

10

E&A/SRevenue$/kW-yr

745013{17%}

3.40985. Alternate Curve with New Entry Net Cost at IRM + 4%

714212{17%}

1.79984. Alternate Curve with New Entry Net Cost at IRM + 1%

744015{17%}

1.23923. Alternate Curve with New Entry Net Cost at IRM

843925{21%}

-0.06542. Original PJM Curve, Based on VOLL

1297066 {35%}

-0.44391. Vertical Demand

Scarcity + RPM

Payment by Consumers (Peak Load

Basis)

RPM Payment $/kW-yr

Generation Profit

$/kW-yr {Return on

Equity}

Average%

Reserve over IRM

% Yearsmeet or Exceed

IRMCurve

⇒ Alternate (sloped) curves have lower consumer cost and better adequacy

Sensitivity AnalysesDemand Curve Shapes:• Where right-hand tail drops to zero• Max. P: dropped by 25%, 40%

Investment Assumptions:• Percent CT added when profit is equal to cost (base: 7%)

+ 2%• Degree of risk aversion:

– risk neutral– very risk averse

• Relative weight placed on earlier year profits in forecast:– low– high

Bidding Assumptions:• Existing capacity bids positive • Potential capacity bids various amounts B

Result: Sloped demand always preferred to vertical

5. Conclusions:Advantages of Sloped Demand

Logically reflects reality of capacity value:– If there are extra reserves, the marginal value of

capacity can be close to but not equal to zero– If reserves are short, payments should be higher

Compared to vertical demand, it lowers risk to generators. Result:– Lower required return to capital– More investment in generation – Dampened capacity cycles– Lower consumer cost

Elasticity mitigates market power in capacity market

PJM Reliability Pricing Model Bhavaraju 200706271 -...

Documents

Transcript of PJM Reliability Pricing Model Bhavaraju 200706271 -...