Top Banner
MAY 1 2013 Laurie Corcoran SENIOR ANALYST, DEMAND RESOURCE STRATEGY Proposed Transition Period Solution To Address Accuracy Of Demand Response Baselines During Facility Outages
25

MAY 1 2013

Jan 31, 2016

Download

Documents

Kira

MAY 1 2013. Laurie Corcoran. SENIOR ANALYST, Demand Resource Strategy. Proposed Transition Period Solution To Address Accuracy Of Demand Response Baselines During Facility Outages. Background. - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: MAY 1 2013

MAY 1 2013

Laurie CorcoranSENIOR ANALYST, DEMAND RESOURCE STRATEGY

Proposed Transition Period Solution To Address Accuracy Of Demand Response Baselines During Facility Outages

Page 2: MAY 1 2013

Background

• The Demand Response Baseline (“baseline”) is an estimate of a demand response asset’s expected or normal load for each interval of the day absent demand reductions from participating in the capacity and/or energy markets.

• Interval meter data from previous days is used in determining an asset’s baseline.

• To measure demand reductions resulting from market participation, the asset’s actual load is compared to the baseline in each dispatch interval.

2

Page 3: MAY 1 2013

Problem

• Currently, the market rules require meter data from days on which an asset is on a scheduled reduction (e.g., scheduled maintenance of energy consuming equipment) or forced reduction (e.g., distribution or transmission outage) to be used in the baseline computation.

• Demand response providers cannot notify the ISO of scheduled or forced reductions.

• As a result, meter data from days when the asset has no or substantially lower load due to a scheduled or forced reduction may be included in the baseline calculation, which impacts the accuracy of the baseline once the asset’s load returns to normal.

3

Page 4: MAY 1 2013

History

• ISO-NE has been working on a solution for the transition period that addresses baseline accuracy during outages. – Implementing a solution for the transition period is difficult :

• IT support is very limited• Must account for multiple assets behind the same meter which may be

interacting with one another– Under full-integration the requirement for an asset to be measured at

the Retail Delivery Point solves the most difficult issue.

• April 10, 2013 Markets Committee meeting– Markets Committee referred the issue of Demand Response Baseline

accuracy during and following scheduled or forced reductions to the Demand Resources Working Group (DRWG).

4

Page 5: MAY 1 2013

Assessment Of Current Issues With Baseline Accuracy During Outages• The ISO assessed whether changes were needed to preserve

baseline accuracy during and following three types of outages:– Planned outages– Scheduled reductions– Forced reductions

• The following slide shows a table summarizing the assessment of the current impact on baseline accuracy of each of the three types of outages.

• As shown in the table it was determined that no change is necessary to maintain baseline accuracy during planned outages.

• The proposed approach for the transition period addresses baseline accuracy issues during forced reductions and scheduled reductions.

5

Page 6: MAY 1 2013

Summary Of Current Issues With Baseline Accuracy Of Load Reducing Assets During Outages

6

Planned Outage Scheduled Reduction Forced Reduction

DescriptionThe equipment used to reduce demand is unavailable, e.g., a generator behind the meter is on a planned outage.

Scheduled maintenance of energy consuming equipment or facility shutdown for maintenance/vacation. The demand of the facility is already reduced.

No electrical service to facility. Distribution or transmission line serving the asset is out of service.

Consumption during outage Greater than or the same as pre-outageSignificantly lower than pre-outage or zero

Zero

Effect of current methodology on baseline accuracy

Baseline will stay the same or be increased Baseline will decline Baseline will decline

Effect of outage on grid Requires the same or additional generationLess generation is required to serve this load

Less generation is required to serve this load

Effect on grid during OP-4/AuditUnable to deliver expected reduction, additional generation is procured

Larger than expected reduction is delivered, less generation is procured

Larger than expected reduction is delivered, less generation is procured

Accuracy of current baseline methodology after outage

Baseline is the same or higher than pre-outage

Baseline is lower than pre-outage Baseline is lower than pre-outage

Does symmetric baseline adjustment address baseline inaccuracy during and after outage?

YesDoes not address baseline inaccuracy during an outage

Does not address baseline inaccuracy during an outage

Does performance calculated for OP-4/capacity audit events during and after outage accurately capture delivered reduction?

Yes No, reduction will be underestimated No, reduction will be underestimated

Is change needed to ensure the accuracy of the baseline and of OP-4/capacity audit performance calculations?

No Yes Yes 

Page 7: MAY 1 2013

Proposed Approach For The Transition Period

• As part of the assessment of the issue, ISO-NE determined that minimal or no IT resources would be available to work on this project. The proposed solutions have been developed in recognition of this constraint.

• Separate solutions have been developed to address baseline accuracy issues during scheduled or forced reductions for:– Load reducing assets– Incremental DG assets

• During periods with scheduled or forced reductions, when demand will be reduced regardless of price, participants should not submit Demand Reduction Offers.

7

Page 8: MAY 1 2013

PROPOSED SOLUTIONS FOR LOAD REDUCING ASSETS DURING SCHEDULED OR FORCED REDUCTIONS

Page 9: MAY 1 2013

Scheduled Reductions• During a scheduled reduction, the load at the facility is

substantially reduced or zero.

• If meter data from scheduled reduction days is included in the baseline calculation, the baseline will decline.

• Performance calculations based on the degraded baseline will not accurately reflect the reduced demand on the grid.

• A symmetric baseline adjustment applied during a scheduled reduction will further degrade the baseline and performance calculated using this degraded baseline will not accurately reflect the reductions delivered during OP-4 events or capacity audits.

9

Page 10: MAY 1 2013

Scheduled Reductions (cont.)• Proposed Solution: To preserve the accuracy of the baseline and

performance calculations, the recommendation is to:– submit the values from the last unadjusted baseline calculated

prior to the day of the scheduled reduction for intervals without OP-4 events or capacity audits, and,

– submit actual meter data for intervals with OP-4 events or capacity audits. On days with OP-4 events or capacity audits, performance will be determined by taking the difference between that day’s unadjusted baseline and actual meter data.

• Proposed Notification: Notification of a scheduled reduction must be submitted to the ISO at least 15 days ahead of the start of the reduction. Scheduled reductions must be a minimum of a single calendar day, and shall not exceed a total of 14 calendar days per Capacity Commitment Period.

10

Page 11: MAY 1 2013

Forced Reductions• During a forced reduction there is no electrical service to the

facility and the load is zero.

• If meter data from forced reduction days is included in the baseline calculation the baseline will decline.

• Performance calculations based on the degraded baseline will not accurately reflect the reduced demand on the grid.

• A symmetric baseline adjustment applied during a forced reduction will further degrade the baseline and performance calculated using this degraded baseline will not accurately reflect the reductions delivered during OP-4 events or capacity audits.

11

Page 12: MAY 1 2013

Forced Reductions (cont.)• Proposed Solution: To preserve the accuracy of the baseline and

performance calculations, the recommendation is to:– submit the values from the last unadjusted baseline calculated

prior to the day of the forced reduction for intervals without OP-4 events or capacity audits, and,

– submit actual meter data for intervals with OP-4 events or capacity audits. On days with OP-4 events or capacity audits, performance will be determined by taking the difference between that day’s unadjusted baseline and actual meter data.

• Proposed Notification: Notification of a forced reduction should be submitted to the ISO for any reductions in demand that occur as a result of actions outside the control of the asset (where there is no electrical service to the facility). The forced reduction notification may be submitted or revised during the resettlement process.

12

Page 13: MAY 1 2013

Impact Of Proposed Approach To Preserving Baseline Accuracy Of Load Reducing Assets During Scheduled Or Forced Reductions

13

Planned Outage Scheduled Reduction Forced Reduction

DescriptionThe equipment used to reduce demand is unavailable, e.g., a generator behind the meter is on a planned outage.

Scheduled maintenance of energy consuming equipment or facility shutdown for maintenance/vacation. The demand of the facility is already reduced.

No electrical service to facility. Distribution or transmission line serving the asset is out of service.

Meter data submission No change from current

During capacity event/audit interruption intervals, submit actual meter data (which will be minimal or 0). During all other intervals, submit baseline values

During capacity event/audit interruption intervals, submit actual meter data (which will be 0). During all other intervals, submit baseline values

Effect of proposed approach on baseline accuracy

None - no change is proposed for assets on a planned outage. The baseline accurately reflects consumption.

Baseline accuracy during and after outages is preserved

Baseline accuracy during and after outages is preserved

Effect of proposed approach on baseline adjustment

No change from currentNo baseline adjustment is applied on capacity event/audit days

No baseline adjustment is applied on capacity event/audit days

Performance calculation for capacity/audit events

No change from current No change from current No change from current

Accuracy of performance calculated during capacity events/audits with proposed change

Performance accurately reflects MWhs of reduction

Performance accurately reflects MWhs of reduction

Performance accurately reflects MWhs of reduction

Page 14: MAY 1 2013

PROPOSED SOLUTION FOR INCREMENTAL DG ASSETS

Page 15: MAY 1 2013

Baseline Accuracy Issues For Incremental Distributed Generation (DG) Assets During Outages

• Developing an approach to address baseline accuracy issues for incremental Distributed Generation (DG) assets during outages, in which performance is based on incremental output, is more complex than developing an approach for load reducing assets.

• Incremental DG assets: – produce output all the time– produce incrementally more output than the normal (baseline)

amount when dispatched– do not have a baseline adjustment applied on event days– on outage will incur a negative deviation equal to the expected

increment that is not delivered plus the amount of the baseline

15

Page 16: MAY 1 2013

Baseline Accuracy Issues For Incremental Distributed Generation (DG) Assets During Outages (cont.)• Stakeholders have suggested the following approaches:

– cap the negative deviation, – develop a baseline adjustment for incremental DG assets, or,– develop another approach to appropriately address the specific issues

of incremental DG baselines.

• The approaches suggested by stakeholders would require significant IT resources. Further, capping of negative deviations is not justified because failure of the DG asset requires the wholesale power system to dispatch, all other things equal, other energy resources to supply both the incremental and baseline output of the failed DG asset.

16

Page 17: MAY 1 2013

Incremental DG Asset Scenarios• Assume two assets located at the same facility, an incremental DG asset

and a load reducing asset. Both assets are associated with the same resource.

• Terms used:– UBL = Unadjusted Baseline– Normal operations = no event occurs in the interval– RDP = Retail Delivery Point– TFL = Total Facility Load, the sum of the load at the RDP and all output

behind the RDP (the total electricity used at a facility inclusive of that purchased from the grid and that produced on-site)

– LR = load reducing asset metered at RDP– LR UBL = UBL for LR based on TFL data– LR Reduction = reduction in consumption by LR– LR Actual = TFL in real-time– Load at RDP = meter reading at RDP in real-time – DG = an incremental DG asset located behind the RDP– DG UBL = UBL for DG based on output data

17

Page 18: MAY 1 2013

Incremental DG Asset Scenarios (cont.)

• The scenarios show a single interval

• During normal operations LR UBL is 11 MW

• During normal operations DG UBL is 7 MW

• During normal operations demand at the facility (LR UBL) is served by:– 7 MW of DG– 4 MW from the grid

• The associated resource has a 4 MW obligation

• Historically DG has provided 3 MW of additional output (10 MW total) when dispatched

• Historically LR has provided 1 MW of reduction when dispatched

• The negative deviation for the DG is capped at -3 MWs

18

Page 19: MAY 1 2013

Incremental DG Asset Scenarios (cont.)

19

Scenario 1 – normal operations

Scenario 2 – the associated resource is dispatched for an audit•LR delivers1 MW•DG delivers 3 MW

Scenario 3 – the associated resource is dispatched for an audit, DG is on outage

• LR delivers1 MW• DG has zero output and incurs a negative deviation during the audit

Page 20: MAY 1 2013

Incremental DG Asset Scenarios (cont.)

20

The difference between scenarios 2 & 3 is the DG in scenario 3 is on an outage and has zero output. The failure of the DG asset in scenario 3 requires the wholesale power system to dispatch other energy resources to supply the 10 MW load at the RDP. Capping the negative deviation at -3 MWs does not account for the full impact on the wholesale power system of the failure of the DG asset.

1 2 3LR UBL 11.0 11.0 11.0LR Reduction 0.0 1.0 1.0LR Actual 11.0 10.0 10.0DG UBL 7.0 7.0 7.0DG Expected Increment During Event 0.0 3.0 3.0DG Expected Output 7.0 10.0 10.0DG Actual Output 7.0 10.0 0.0Load at RDP 4.0 0.0 10.0Generation Dispatched To Meet Demand At RDP 4.0 0.0 10.0DG Negative Deviation (Capped at -3 MW) 0.0 0.0 -3.0

Page 21: MAY 1 2013

Incremental DG Asset Scenarios (cont.)

21

0.0

2.0

4.0

6.0

8.0

10.0

12.0

1 2 3

Load at RDP

DG Actual Output

LR UBL

DG UBL

Page 22: MAY 1 2013

Approach To Baseline Inaccuracy Issues For Incremental DGs • Proposed Solution: Provide the opportunity for participants to

retire the incremental DG assets behind the RDP and register a single asset with a baseline based on load at the RDP.

• The baseline associated with the existing metering configuration “Load Reduction - DG Used to Reduce Load at another on-site Asset” is based on Total Facility Load (the total electricity used at a facility inclusive of that purchased from the grid and that produced on-site). If this metering configuration is associated with a baseline based on Facility Metered Load (the electricity purchased from the grid) for a single asset at the RDP, a comparable result to metering a Demand Response Asset at the RDP, (which is required in full integration) is achieved.

22

Page 23: MAY 1 2013

NEXT STEPS

Page 24: MAY 1 2013

Process

• DRWG feedback on proposed solutions

24

Page 25: MAY 1 2013

25