ISO New England – Results of

Ancillary Service Pilot Programs:

Jon Lowell and Henry Yoshimura

ISO New England

October 25-26, 2011

• Alternative Technology Regulation Pilot Program

• Demand Response Reserves Pilot Program

Alternative Technology Regulation

Pilot Program

Alternative Technology Regulation Pilot

Program • Why?

– Stakeholder request

– Low risk opportunity to evaluate new

technologies

– Allow new technologies to experience realistic

operation

– Minimal impact on the existing regulation

market

3

• What?

– Limited to 13 MW total to ensure no threat to reliability

– Incremental to existing regulation market

– Participants operate as “price-takers”

– Will continue until permanent market changes are approved and implemented

• When?

– Commenced 11/2008; closed to new entrants after 11/2009

• Expected to re-open Q2 2012

– Will continue until permanent market changes are approved and implemented

Principal Issues

• Can these new technologies perform as advertised? – So far, the answer is yes

• What issues have been identified? – Managing state-of-charge for storage resources with limited energy storage (i.e.

less than 2 hours)

– Cost-effective metering and communications for geographically dispersed

resources (electric vehicles, demand response)

• Benefits of “fast response” vs. “sustained response” – AGC dispatch sends new set points every 4 seconds

– Many/most of the new technologies can move from top to bottom of their

regulating range within a single 4-second cycle

– NERC compliance criteria based on 15-minute intervals

• “Fast” is better than “slow”, but: – Faster response is unlikely to reduce noncompliance risk

– Might allow reduced regulation requirement

– Current New England requirement is ~65MW for a 28,000 MW summer peak

system.

4

Impact of Response Rate

5

0

10

20

30

40

50

60

70

80

-

1,000,000

2,000,000

3,000,000

4,000,000

5,000,000

6,000,000

7,000,000

<3 3-5 6-10 11-20 >20

$/m

ile o

r $

/MW

of

Cap

pe

r h

ou

r

An

nu

al M

ileag

e o

r A

nn

ual

Co

st

Response Rate MW/min

ISO New England Regulation Market - 2010

Mileage

Payment

Payment/Mile

Pmt / MW of Delvd Reg Cap

Factors Affecting Regulation Market

Economics

• Late-2005 – implemented service payment and

separated resource-specific

opportunity costs from the

Regulation Clearing Price

• Early-2007 – Bidders fully adjusted to market

design

– minor improvements in selection

algorithm

• 2009 – Reduction in requirements

– Lower gas prices reduced

opportunity costs

• Future? – Uniform clearing price

– New technologies

6

0

10

20

30

40

50

60

70

80

90

0

20

40

60

80

100

120

140

160

2005 2006 2007 2008 2009 2010

Re

g P

mts

$m

Re

g R

eq

m M

W

Regulation Market

Reg Requirement MW Reg Pmts to Resources

Pilot Program – Interested Technologies

• Currently participating – Flywheels

– Residential electric thermal storage

• Qualified to participate, but not online – NaS batteries

– Commercial/industrial load control

– HVDC power electronics

• Expressions of Interest – Lithium-ion batteries

– Electric vehicles

– Buoyancy energy storage

– Wind-powered microgrid

7

Where Are We Heading?

• ISO-NE is moving to eliminate the “pilot” status and allow full

participation of these new technologies in the regulation market – Timing mostly depends on scope of forthcoming rules related to FERC’s

February 2011 Regulation NOPR

• May preserve the Pilot Program infrastructure as a technology

sandbox – Innovation doesn’t occur on schedule

– Will shorten or avoid delays related to stakeholder and regulatory

processes

• Develop cost-effective approaches to metering and settlement

to accommodate geographically dispersed “smart-grid”

applications – Expensive solutions will be a significant barrier to entry

8

ISO-NE’s Regulation Market - Design

Improvement Goals

• Uniform clearing price that incorporates opportunity costs

of the marginal resource

• Cost-effectively integrating limited energy storage

resource

• Cost-effectively integrating geographically dispersed,

aggregated resources – Examples: vehicle-to-grid, demand-response

9

Demand Response Reserves Pilot

Program

Demand Response in Reserve Markets

• Demand Response is currently not eligible to provide

reserves to the wholesale electric system

– Dispatchable demands are eligible to provide reserves

• Demand Response Reserves Pilot started in 2006

• 50 MW test over a multi-year period to demonstrate

performance during reserve activation events

• Can Demand Response provide a product similar to 10

and 30-minute spinning and non-spinning reserves?

11

Participating Asset Types

• Load Reduction – The most common load reduction technology/strategy included

lighting and HVAC usage

– Asset performance was assessed by comparing actual metered

load during an event to an asset-specific estimated baseline

• Behind-the-Meter Generation – Asset performance assessed solely on the metered generation at

the time of an event

• Direct Load Control – Centralized control of a specific end-use across a large number

of small customers – e.g., residential air conditioner curtailment

– Asset performance was assessed by comparing actual metered

load during an event to a control group

12

Participating Industries

• Grocery Stores

• Manufacturing

• Large Retail

• Education Sector

• Wastewater Treatment Facilities

• Aggregated Air Conditioning Curtailment

• Behind-the-Meter Generators (regardless of their industry

of origin)

13

Demand Response Reserves Pilot Summary

• Pilot ran from October 2006

to May 2010

– Asset types include load

reductions, Behind the Meter

Generation, direct load

control (i.e. aggregated air

conditioner curtailment)

• 109 assets participated in at

least one season; 35 assets

participated in all seasons

– Assets enrolled in six

seasons; 26.4 MW per

season on average

– 107 events

14

0%

20%

40%

60%

80%

100%

120%

140%

160%

180%

10

/06

/20

06

10

/30

/20

06

12

/06

/20

06

01

/16

/20

07

02

/15

/20

07

03

/06

/20

07

03

/22

/20

07

04

/13

/20

07

04

/26

/20

07

05

/30

/20

07

06

/08

/20

07

06

/22

/20

07

07

/05

/20

07

07

/25

/20

07

08

/07

/20

07

08

/13

/20

07

08

/15

/20

07

09

/05

/20

07

09

/26

/20

07

11

/06

/20

07

11

/29

/20

07

12

/28

/20

07

01

/22

/20

08

02

/29

/20

08

03

/26

/20

08

04

/17

/20

08

05

/01

/20

08

06

/12

/20

08

06

/23

/20

08

07

/02

/20

08

07

/16

/20

08

07

/28

/20

08

08

/12

/20

08

09

/02

/20

08

09

/17

/20

08

06

/03

/20

09

06

/08

/20

09

06

/25

/20

09

07

/08

/20

09

07

/17

/20

09

08

/07

/20

09

08

/17

/20

09

08

/28

/20

09

09

/10

/20

09

10

/07

/20

09

11

/02

/20

09

12

/07

/20

09

01

/15

/20

10

02

/16

/20

10

03

/05

/20

10

03

/22

/20

10

04

/13

/20

10

04

/27

/20

10

05

/17

/20

10

Total Load Relief as Percent of DRR Contract Amount (Mean MW) Session Average Linear Trend

DR Reserve Pilot Performance

15

DRRP Performance of Load Reduction

16

17

18

DRRP Performance of Behind-the-Meter Generation

19

Comparing the Performance of Pilot Program

Assets and Generation Resources

• Participating pilot program assets showed less reliability

than generation resources (i.e., in-front of meter

generators)

• In-front-of-the-meter generation resources showed a

moderate increase in reliability since the pilot program

started, while pilot program assets showed a decrease in

reliability during the same time frame

20

Recommendations for Further Research

• Investigate / implement: – Performance erosion

– Audit day behavior

– Weather-based performance metrics

– Penalties for over-performance

– Tools to assist asset providers to set goals and

maintain performance

– Change rules regarding re-submission of data

– Special metering requirements for behind-the-meter

generation assets

21