6. Assignment of Proceeding
Michael R. Peevey is the assigned Commissioner and David M. Gamson is the assigned ALJ in this proceeding.
1. The assumptions, processes, and criteria used for the 2011 LCR study were discussed and recommended in a CAISO stakeholder meeting, and they generally mirror those used in the 2007 through 2010 LCR studies.
2. The SCP is an attempt to reduce transactions costs associated with buying, selling, and trading capacity to meet RA requirements. In order to meet this goal, the SCP seeks to standardize the obligations of RA providers and other related terms of RA contracts. As implemented to date, the SCP places contract terms relating to availability standards and penalties in Section 40.9 of the CAISO Tariff.
3. The FERC approved the existing SCP on June 28, 2009. In that order, FERC directed the CAISO to work toward extending the SCP to currently exempt resources. At this time, certain resources whose QC are determined based on historical data (including QF resources) and demand response resources are exempt from the SCP.
4. A process for local true-up of RA capacity was adopted in D.10-03-022 for 2010 only.
5. One of the purposes of the LCR studies is to identify the local constraints in the coming year. The 2011 LCR results of the "other PG&E areas" shows that there still are a limited amount of resources in those areas.
6. Previous resource adequacy decisions, including D.06-06-064, provided that an LSE cannot be required to procure capacity that does not exist, in situations where the local area resource need is higher than existing generation capacity. This "blanket waiver" has been continued year to year.
7. In order for the SCP to be fully functional, it must be available as a commercially-viable product that LSEs can purchase, consistent with the counting rules developed by the Commission. This requires turning the SCP into a fungible product that is easily commercially traded.
8. A QC Report provided by the Energy Division in workshops led to parties' comments, and formed the basis for the Commission to consider a Qualifying Capacity Methodology Manual.
9. The classification methodology proposed by the Energy Division in
Section 3.1 of the QC Report allows for case by case determination of the dispatchability classification of individual resources, including CHP. The Energy Division's proposal allows the specific details of a resource's operational characteristics, both physical and contractual, to be considered in its classification.
10. Because the SCP availability standard already applies to dispatchable resources, the resource owner and scheduling coordinator have proper incentives to classify the resource appropriately.
11. The Energy Division's proposals to measure the QC of new
non-dispatchable resources using an approximation based on existing
non-dispatchable resources, and to modify the measurement hours, provide a reasonable approach to address a gap in previously adopted rules.
12. The CAC proposal to calculate NQC on a monthly basis as opposed to a summer-months average for the entire year, as applied to all non-dispatchable resources, is reasonable.
13. Counting rules for all resources should be operator-neutral, and should only differentiate between resources based on the operational characteristics of the resources. However, in some cases, especially with demand response, operational characteristics may be substantially correlated with the characteristics of the resource operator.
14. Line losses are included in the load forecast used for RA requirements. Demand response resources provide a means of balancing supply and demand without accruing line losses.
15. The entire structure of the local RA program uses summer peak values for not only load forecasts, but all supply resources. This conservative approach provides a significant margin of safety in the off-peak months.
16. The existing RA procurement penalty structure provides no guidance as to what happens if the LSE does not replace capacity within 10 business days after notification.
17. For system procurement deficiencies, a higher penalty for deficiencies which endure more than a few days whould serve to discourage extended non-compliance.
18. A local RA penalty level that exceeds the local RA waiver trigger price ($40/kW-yr) may, in circumstances of market power, effectively raise market prices for local RA.
19. LSEs currently submit a month ahead forecast 60 days prior to the month they are forecasting. The monthly load forecast is submitted 30 days before the filing of the Month-Ahead RA showing. This monthly load forecast is used as part of the month-ahead compliance check because it may change the load forecast which in turn changes the RA obligation. In some cases, LSEs have filed updates to their load forecasts after they file their month-ahead RA showing.
20. Currently there is no established timeline for the RA record retention process.
21. The RA Guides provide a process for LSEs to show additional local resources after the date of the year-ahead local RA filing.
1. The CAISO's 2011 LCR study should be approved as the basis for establishing local procurement obligations for 2011 applicable to Commission-jurisdictional LSEs.
2. Because the current local RA program establishes procurement obligations for the following year, LSEs should only be responsible for procurement in a local area to the level of resources that exist in the area.
3. There is a need for further discussion and record development regarding proposals for a local true-up methodology for 2011 and beyond, once there is sufficient experience gathered with the local RA True up mechanism adopted in D.10-03-022.
4. Given the local resource constraints identified by the CAISO in the "other PG&E" local areas and consequent market power concerns, it is reasonable to keep the local areas aggregated for 2011.
5. There is no foreseeable situation where there will be no need for the "blanket waiver". The "blanket waiver" should be adopted for 2011 and beyond.
6. While in theory the SCP ultimately should be extended to DR resources, there is no viable proposal to effectuate this change at this time.
7. A Qualifying Capacity Methodology Manual (QC Methodology Manual) should be adopted.
8. The Energy Division's proposal to allow for case by case determination of the dispatchability classification of individual resources should be adopted as part of the QC Methodology Manual.
9. The Energy Division's proposals to measure the QC of new non-dispatchable resources using an approximation based on existing non-dispatchable resources, and to modify the measurement hours, should be approved as part of the QC Methodology Manual.
10. The CAC proposal to calculate NQC on a monthly basis as opposed to a summer-months average for the entire year, as applied to all non-dispatchable resources, should be approved as part of the QC Methodology Manual.
11. Counting rules for all resources should be operator-neutral, and should only differentiate between resources based on the operational characteristics of the resources. Fairness requires that we decline to differentiate based on the identity of the operator.
12. It is reasonable that dispatchable DR resources with financial incentives for availability and performance comparable to those of dispatchable supply resources should be able to receive QC with a comparable testing methodology. However, unless and until it is demonstrated to us, in this or a future RA proceeding, that such a DR resource exists, we will retain our current policy that the LIPs are used to establish the QC of DR resources to the maximum extent possible.
13. DR resources should receive the benefit of avoiding line losses in calculating RA values.
14. SCE's proposal to value line losses for DR resources in calculating RA values is reasonable.
15. The current treatment of AC Cycling programs is consistent with the larger local RA program and should continue.
16. It is reasonable to both provide an incentive for timely replacement of RA procurement capacity and to provide a clear penalty if this does not occur.
17. The Energy Division's proposal that LSEs may, at the discretion of CEC staff, file changes to their load forecasts up to 25 days before the due date of the month-ahead compliance filings, is reasonable.
18. The Energy Division's proposal to keep all RA filings and related materials for three calendar years after the end of the compliance year is reasonable.
IT IS ORDERED that:
1. The California Independent System Operator's final 2011 Local Capacity Technical Analysis Final Report and Study Results is adopted as the basis for establishing local procurement obligations for 2011 applicable to
Commission-jurisdictional load-serving entities, as listed in Appendix A to this decision.
2. The "Option 2/Category C" Local Capacity Requirements set forth in the California Independent System Operator's 2011 Local Capacity Technical Analysis, Final Report and Study Results, dated May 3, 2010, are adopted as the basis for establishing local resource adequacy procurement obligations for
load-serving entities subject to this Commission's resource adequacy program requirements. The Local Capacity Requirements for 2011 are as follows:
2011 LCR Need Based on Category C with operating procedure | |||
Local Area Name |
Existing Capacity Needed |
Deficiency |
Total (MW) |
Humboldt |
188 |
17 |
205 |
North Coast / North Bay |
734 |
0 |
734 |
Sierra |
1510 |
572 |
2082 |
Stockton |
459 |
223 |
682 |
Greater Bay |
4804 |
74 |
4878 |
Greater Fresno |
2444 |
4 |
2448 |
Kern |
434 |
13 |
447 |
LA Basin |
10589 |
0 |
10589 |
Big Creek/ Ventura |
2786 |
0 |
2786 |
San Diego |
3146 |
61 |
3207 |
Total |
27094 |
964 |
28058 |
3. The local resource adequacy program and associated requirements adopted in Decision (D.) 06-06-064 for compliance year 2007, and continued in effect by D.07-06-029 and D.08-06-031 and D.09-06-028 for compliance years 2008, 2009 and 2010, respectively, are continued in effect for compliance year 2011, subject to the modifications, refinements, and local capacity requirements adopted in the ordering paragraphs in this decision.
4. The assigned Administrative Law Judge in this proceeding shall take comments on a re-evaluation of the 2010 resource adequacy local true-up adopted in Decision (D.) 10-03-022 in order to consider implementing a resource adequacy local true-up or reallocation methodology for 2011 and beyond. The local true-up method adopted in D.10-03-022 remains in place until superseded.
5. While we may, at our discretion, revisit the issue in the future, the "blanket waiver" rule that an load serving entities cannot be required to procure capacity that does not exist, in situations where the local area resource need is higher than existing generation capacity, is made permanent.
6. The following modifications to the resource adequacy requirements adopted by Decision (D.) 04-01-050; D.04-10-035; D.05-10-042 as modified by D.06-02-007, D.06-04-040, and D.06-12-037; D.06-06-064, D.06-07-031; D.07-06-029; D.08-06-031; and D.09-06-028 are adopted beginning with the 2011 resource adequacy program compliance year:
a. The Qualifying Capacity Methodology Manual in Appendix B to this decision is adopted as part of the resource adequacy program. The Energy Division shall use the Manual to calculate a 2011 net qualifying capacity list and post the results on the Energy Division's website. Each load-serving entity shall use net qualifying capacity values established according to the manual along with relevant allocations for resource adequacy (RA) credit to fulfill its resource adequacy obligation.
b. Line losses avoided by demand response (DR) resources shall be valued for the purposes of resource adequacy calculations as follows:
DR RA Value = 1.15 * DR Load Impact * (1.00/(1.00 - transmission and distribution (T&D) Line Loss Rate)) where, T&D Line Loss Rate = 3% + IOU-specific Distribution Loss Factors.
c. Full year local resource adequacy credit for Air Conditioner Cycling programs shall continue.
d. The Energy Division shall keep all resource adequacy filings and related materials for three calendar years after the end of the compliance year. The Energy Division shall generally destroy records past their retention date, but may retain records for statistical, enforcement or other purposes.
e. Load-serving entities may, at the discretion of the California Energy Commission staff, file changes to their load forecasts up to 25 days before the due date of the month-ahead compliance filings.
f. The requirement in Decision 06-06-064 that LSEs list all local resources under their control on their local RA filing is modified so that at the time of the year-ahead local filing, each LSE shall submit, in addition to its other year-ahead filings, a list of all local resources it controls (via ownership or contact) that are not listed as RA resources and committed to the CAISO up to their full NQC in the year-ahead local filing. This "additional local resource list" shall be sent to the CPUC, CAISO, and CEC. LSEs may commit resources from their additional local resource list as RA resources in order to meet residual collective local deficiencies identified by CAISO.
g. The following penalty structure for resource adequacy procurement deficiencies is adopted for violations which occur after the date of this decision:
Small Procurement Deficiency (modifying E-4195, Appendix A) |
System Procurement Deficiency (modifying D.05-10-042, COL 21 and D.06-06-064, COL 26) |
Local Procurement Deficiency (modifying D.06-06-064, COL 25 and COL 26) | |
Replaced within five-business days of the date of notification |
$1,500 first incident in calendar year; $3,000 for each incident thereafter in a calendar year |
$3.33/kilowatt (kW)-month |
$3.33/kW-month |
Replaced after five-business days from the date of notification or not replaced |
LSE pays the applicable System or local RA penalty for the deficiency |
$6.66/kW-month |
$3.33/kW-month |
7. Rulemaking 09-10-032 shall remain open.
This order is effective today.
Dated June 24, 2010, at San Francisco, California.
MICHAEL R. PEEVEY
President
DIAN M. GRUENEICH
JOHN A. BOHN
TIMOTHY ALAN SIMON
NANCY E. RYAN
Commissioners
APPENDIX A
Load-Serving Entities as Defined in Section 380(j)
Electrical Corporations
Brian Cherry (39)
Director, Regulatory Relations
Pacific Gas and Electric Company
P. O. Box 770000, B10C
San Francisco, CA 94177
Steve Rahon (902)
Director, Tariff & Regulatory Accounts
San Diego Gas & Electric Company
8330 Century Park Court, CP32C
San Diego, CA 92123-1548
Akbar Jazayeiri (338)
Director of Revenue & Tariffs
Southern California Edison Company
P. O. Box 800
2241 Walnut Grove Avenue
Rosemead, CA 91770
Electric Service Providers
Inger Goodman
Commerce Energy, Inc. (1092)
575 Anton Boulevard, Suite 650
Costa Mesa, CA 92626
Ron Cerniglia
Direct Energy Services, LLC (1341)
12 Greenway Plaza, Suite 600
Houston, TX 77046
ron.cerniglia@directenergy.com
Kerry Hughes
Direct Energy Business (1351)
7220 Avenida Encinas, Suite 120
Carlsbad, CA 92009
customerrelations@directenergy.com
Victor Gonzales
Constellation NewEnergy, Inc. (1359)
111 Market Place, Suite 500
Baltimore, MD 21202
victor.m.gonzalez@constellation.com
Kevin Boudreaux
Calpine PowerAmerica-CA, LLC (1362)
3875 Hopyard Road, Suite 345
Pleasanton, CA 94588-8558
Drake Welch
Sempra Energy Solutions (1364)
401 West A Street, Suite 500
San Diego, CA 92101-3017
Thomas Darton
Pilot Power Group, Inc. (1365)
8910 University Center Lane, Suite 520
San Diego, CA 92122
Rick C. Noger
Praxair Plainfield, Inc. (1370)
2711 Centerville Road, Suite 400
Wilmington, DE 19808
Jenny Zyak
Liberty Power Holdings LLC (1371)
1901 W. Cypress Creek Road, Suite 600
Fort Lauderdale, FL 33309
Jenny Zyak
Liberty Power Delaware LLC (1372)
1901 W. Cypress Creek Road, Suite 600
Fort Lauderdale, FL 33309
Michael Mazur
3 Phases Electrical Consulting (1373)
2100 Sepulveda Blvd., Suite 37
Manhattan Beach, CA 90266
Marcie Milner
Shell Energy (1374)
4445 Eastgate Mall, Suite 100
San Diego, CA 92121
Any electric service provider that, subsequent to the date of the order instituting this rulemaking, becomes registered to provide services within the service territory of one or more of the respondent electrical corporations through direct access transactions shall, upon such registration, become a respondent to this proceeding. Any electric service provider respondent whose registration is cancelled during the course of this proceeding shall, upon confirmation of such cancellation by the Energy Division, cease to be a respondent to this proceeding.
Community Choice Aggregators
Marin Energy Authority
John Dalessi
Staff Consultant
916-631-3210
jdalessi@navigantconsulting.com
3100 Zinfandel Drive, Ste. 600
Rancho Cordova, CA 95670
Any community choice aggregator that, subsequent to the date of the order instituting this rulemaking, files an implementation plan or becomes registered to provide services within the service territory of one or more of the respondent electrical corporations through community choice aggregation transactions shall, upon such filing or registration, become a respondent to this proceeding. Any community choice aggregator respondent that withdraws its implementation plan or whose registration is cancelled during the course of this proceeding shall, upon confirmation of such withdrawal or cancellation by the Energy Division, cease to be a respondent to this proceeding.
(END OF APPENDIX A)
APPENDIX B
Qualifying Capacity Methodology Manual
This manual describes the current net qualifying capacity (NQC) counting rules of the California Public Utilities Commission (CPUC) and the methodology for implementing these rules. Each year, CPUC staff works with the California Energy Resources Conservation and Development Commission (Energy Commission) and California Independent System Operator (California ISO) to publish an NQC list which describes the amount of capacity that can be counted from each resource toward meeting Resource Adequacy (RA) requirements in the CPUC's RA program. The qualifying capacity (QC) of each resource is set by the methodologies described in this document; then if it's QC is not fully deliverable to aggregate California ISO load, it is adjusted to its deliverable capacity resulting in its NQC. For purposes of this report, the term `resource' is used to refer to a generator that has a resource ID on the Master CAISO Control Area Generation Capability List (Generation Capability List)78 or a demand response program which may not have a resource ID.
2.1. Guide to this Document
Sections 3 through 6 describe issues relevant to a variety of resource classifications. Sections 7 through 10 provide details on the specific calculation methodologies for each of the resource types described in Section 3, Resource Classification. Section 4, Deliverability describes California ISO's methodology for assessing the deliverability of generating resources and how Deliverability Assessment impacts NQC. Section 5 lists certain data conventions used in calculating QC. Section 6 discusses the treatment of outages in QC calculations.
The appendices to this report are presented in a separate file.
CPUC staff coordinates with California ISO and Energy Commission staff each year to group resources, by California ISO scheduling resource ID (CAISO ID), into the classifications described below. Classification is based on the dispatchability and technology type of the resource. Primary guidance comes from the most recent available Generation Capability List. Classification for QC calculation does not consider Qualifying Facility status. Demand response resources are not listed on the Generation Capability List; these resources are addressed in Section 10.
First, some resources are selected and classified according to the "ISO Classification" column. Resources listed as wind are classified as wind, and solar resources are classified as solar. The wind and solar classifications receive QC according to the methodology described in Section 8. Resources listed as hydro are classified as hydro resources. Hydro resources are sub-classified by dispatchability, as described below. Each year, Energy Division and California ISO publish a preliminary NQC list of all resources, including the proposed classification of each resource. Resource owners and Scheduling Coordinators (SCs) may suggest changes to the classification of their resources; stakeholders suggesting a change should provide appropriate support for their proposed change such as confirmation from the SC that the resource is dispatchable. On this preliminary list, hydro and other remaining resources are grouped according to dispatchability. Hydro resources may be listed as either "dispatchable hydro" or "non-dispatchable hydro." Hydro resources that are dispatchable by the SC or California ISO are classified as dispatchable hydro. The remaining resources (i.e. resources that are not demand response, wind, solar, or hydro) are also grouped by dispatchability. Resources that are dispatchable by the SC or California ISO are classified as dispatchable generation. Dispatchable generation resources including dispatchable hydro resources receive QC according to the methodology described in Section 7. This classification includes a variety of technologies: steam turbines; combustion turbines; combined cycles; reciprocating engines; and dispatchable combined heat and power (CHP), biomass and geothermal. Again, status as a use limited resource does not prevent a unit from being classified as dispatchable.
Finally, the remaining resources are classified as other non-dispatchable resources. Non-dispatchable hydro and other non-dispatchable resources receive QC according to the methodology described in Section 9.
Deliverability is the ability of the output of a generating resource to be delivered to aggregate load. The only difference between QC and NQC is the deliverability of the resource to aggregate California ISO load. If a resource's QC exceeds its deliverable capacity as determined by California ISO Deliverability Assessments, its NQC is adjusted to its deliverable capacity. In many cases, a resource is fully deliverable and there is no difference between QC and NQC.
California ISO assesses the deliverability of new and existing resources two to three times per year; a Deliverability Assessment is a required part of the Large Generator Interconnection Procedures (LGIP).79 Existing resources retain priority for deliverability over new resources and resources are not expected80 to lose deliverability rights unless the resource is unable to produce its deliverable capacity for at least three consecutive years. The deliverability study provides new resources with information to understand which network upgrades are necessary to achieve full deliverability.
The ability of the output from a new generation project and existing generation to be delivered to aggregate load within California ISO during a resource shortage condition is evaluated pursuant to the ISO's LGIP and the California ISO Deliverability Assessment Methodology posted on the California ISO's website.81
The California ISO Tariff defines a generation project's deliverability as one of two discrete states: Full Capacity Deliverability Status and Energy-Only Deliverability Status. The NQC value of any Energy-Only facility is deemed to be zero.82 Therefore, a generation resource's Deliverability Study Value is typically either 100% or 0% of its QC. However, it is possible that a very few projects that submitted interconnection requests prior to the reformation of the LGIP could have a deliverability level between 100% and 0%. There is also a remote possibility that the deliverability of existing resources could degrade substantially below 100% deliverable and as a result their deliverability level would need to be reduced accordingly. As of August 6, 2009, all generation resources were deliverable to 100% of their QC value. However, at that time, there were approximately 10,000 MW of energy only interconnection requests in the current California ISO interconnection queue. The California ISO Tariff defines Energy-Only connection resources to have an NQC of zero. Therefore, it is likely that, as these resources achieve commercial operation, many of them will have an NQC equal to zero.
The base case for the deliverability study is updated each year. Deliverability studies model peak demand periods and assume that all generating resources are dispatched to meet demand. The base case also assumes that sufficient generation is available within load pockets. Dispatch and outage contingency scenarios are also studied. Generation costs are not considered in the deliverability studies. A finding of deliverability does not ensure that a resource will not experience congestion, especially during non-peak periods. The deliverability study models a five-year planning horizon.
Not all new resources use the LGIP. Some resources connected to the transmission system with nameplate capacity 20 MW or less use the Small Generator Interconnection Procedure (SGIP). The SGIP does not include a Deliverability Assessment and resources that use SGIP have an NQC equal to zero.83 Other small resources that are connected to the distribution system may use a Small Generator Interconnection Agreement (SGIA) with the distribution system owner.84 These SGIAs include deliverability assessments which are accepted by California ISO. Therefore, these resources can be deliverable up to 100% their QC.
This section lists certain conventions used by the staffs of the CPUC, California ISO, and Energy Commission in dealing with the data in the QC calculation process:
· For wind, solar, and other non-dispatchable resources, historical production data is used. This data is obtained by subpoena from CPUC to California ISO; CPUC subpoenas data for specific resource IDs in these classifications from the classification list. CPUC subpoenas hourly "Actual Settlement Quality Meter Data" which describes the production profile for each resource. The production is measured in MWh produced per hour. This data represents the average generation (MW) over each hour and does not provide any information about intra-hour variation in generation.
· New wind, solar, and other non-dispatchable resources are considered to begin operation in the first month the resource operated before the 15th day. A resource that began producing on the 16th (or later) day of a month is considered to begin operation during the following month. The first positive values in the Actual Settlement Quality Meter Data are the sign that a resource began producing. Under this convention, no distinction is made between zero values due to a discontinuation of operation versus zero production during the normal course of operation (e.g., due to lack of fuel such as wind).
This section describes how past outages may impact the QC of some resources; it does not describe how California ISO schedules and approves outages or how SCs should report outages.
Scheduled outages greater than 25% of days in a month reduce the amount of NQC that a resource can count for RA during that month; this rule is referred to as the scheduled outage criterion.85 For resource types whose NQC is derived from historical data,86 proxy data is generated to replace data during any scheduled outages of sufficient duration to trigger the scheduled outage criterion87 and for any forced outage, non-ambient derate, or temperature-related ambient derate. These resource classifications include non-dispatchable wind, solar, biomass, CHP, and geothermal resources. Outages or derates that only partially reduce the output of the resource are treated the same as outages or derates with zero output; therefore, production during an outage or derate has no impact on the calculated QC.
In order to generate the set of outages or derates to be "corrected" California ISO retrieves data from Scheduling and Logging for ISO of California (SLIC) system.88 First, CPUC provides a list of resources to California ISO to include in its query. Then, for each calendar month within the three calendar years used for calculations, California ISO queries SLIC for all outages of outage types:
· "Planned" with a duration greater than seven days,
· "Forced" of any duration, or
· "Ambient" of any duration, with the "Ambient Not Due to Temperature" attribute not selected.
Other criteria for the data query are:
· Process Status: "APPROVED", "OUT", "REQUESTED" , "SCHEDULED", or "INSERVICE" (INSERVICE status is necessary to pull historical data since status changes to INSERVICE after outage is over)
· Resource type: "GENERATOR"
· Outage mode: "DERATE"
After receiving the description of the outages and derates from California ISO, the CPUC and Energy Commission remove the data during the outages and develop replacement proxy data. For each outage or derate hour, the values for the same hour on the same calendar day for other years in the data set are averaged. This average value is inserted as the proxy value. The average includes all values in the data set, for the appropriate day and hour, which are not marked as an outage or derate. Therefore, if there were overlapping outages or derates in two out of three years (i.e. outages during two years covered some of the same hours), all three years would receive the value of the remaining year for the hours marked as outage or derate during both years. If an outage or derate exists at the same time period for all three years, that hour is excluded from the QC calculation.89
Table 1 shows an example for this calculation. The resource had an outage in year 3 including all hours of March 7. Note that the production values during the outage (i.e. in year 3) do not affect the proxy values.
Date |
Hour |
Year 1 (MWh) |
Year 2 (MWh) |
Year 3 (MWh) |
Average (MWh), Years 1 - 2 |
Average (MWh), Years 1 -3 |
Proxy Value (MWh) - Year 3 |
7-Mar |
1 |
50 |
53 |
16 |
51.5 |
39.7 |
51.5 |
7-Mar |
2 |
51 |
54 |
15 |
52.5 |
40 |
52.5 |
7-Mar |
3 |
50 |
52 |
17 |
51 |
39.7 |
51 |
7-Mar |
4 |
52 |
50 |
16 |
51 |
39.3 |
51 |
7-Mar |
5 |
55 |
53 |
17 |
54 |
41.7 |
54 |
7-Mar |
6 |
60 |
63 |
18 |
61.5 |
47 |
61.5 |
7-Mar |
7 |
70 |
65 |
16 |
67.5 |
50.3 |
67.5 |
7-Mar |
8 |
71 |
70 |
17 |
70.5 |
52.7 |
70.5 |
7-Mar |
9 |
72 |
75 |
18 |
73.5 |
55 |
73.5 |
7-Mar |
10 |
72 |
74 |
17 |
73 |
54.3 |
73 |
7-Mar |
11 |
74 |
72 |
16 |
73 |
54 |
73 |
7-Mar |
12 |
74 |
73 |
20 |
73.5 |
55.7 |
73.5 |
7-Mar |
13 |
75 |
77 |
19 |
76 |
57 |
76 |
7-Mar |
14 |
74 |
76 |
18 |
75 |
56 |
75 |
7-Mar |
15 |
76 |
72 |
19 |
74 |
55.7 |
74 |
7-Mar |
16 |
75 |
73 |
19 |
74 |
55.7 |
74 |
7-Mar |
17 |
75 |
78 |
18 |
76.5 |
57 |
76.5 |
7-Mar |
18 |
74 |
75 |
20 |
74.5 |
56.3 |
74.5 |
7-Mar |
19 |
70 |
73 |
19 |
71.5 |
54 |
71.5 |
7-Mar |
20 |
68 |
69 |
18 |
68.5 |
51.7 |
68.5 |
7-Mar |
21 |
65 |
67 |
19 |
66 |
50.3 |
66 |
7-Mar |
22 |
63 |
65 |
18 |
64 |
48.7 |
64 |
7-Mar |
23 |
60 |
62 |
18 |
61 |
46.7 |
61 |
7-Mar |
24 |
58 |
59 |
18 |
58.5 |
45 |
58.5 |
Table 1. Example of Proxy Data
Dispatchable generation resources receive NQC values based on their available capacity,90 subject to the checks described in Section 4, Deliverability. The Scheduling Coordinator (SC) of the resource submits a proposed QC value to the California ISO, along with a reference to the resource's most recent maximum power plant output (PMax) test91 that is in California ISO's master file. This information is submitted to California ISO in a standard format;92 California ISO checks the submitted value for consistency with the PMax and maximum deliverable capacity. If the proposed QC value is less than or equal to the PMax and the maximum deliverable capacity, it is accepted for the NQC value. If not, the previous NQC value is retained. The SC may coordinate with California ISO to update the PMax test or supply other information as requested by California ISO in order to determine an acceptable change to NQC. The SC may use this process to update the QC from time to time. At the time each compliance year's NQC list is published, California ISO checks that each NQC is less than or equal to the most recent PMax for the resource.
The QC of wind and solar resources is based on an exceedance methodology.93 The exceedance approach measures the minimum amount of generation produced by the resource in a certain percentage of included hours. For example, the mathematical concept of "median" is a special case of the exceedance concept, with the exceedance level set to 50%. The exceedance level used to calculate the QC of wind and solar resources is 70%. Another way to describe the exceedance level is that the 70% exceedance level of a resource's production profile is the maximum generation amount that it produces at least 70% of the time. The exceedance concept is depicted in Figure 1; while the median is not used in the wind and solar QC calculation, it is included in the diagram to provide context to the 70% exceedance. The 70% exceedance value is shown as a blue horizontal line and the median is a purple horizontal line.
Figure 1. Conceptual Diagram of Exceedance94
Intuitively, the exceedance calculation ranks all of the included hours by production and draws the initial QC from the value 70% of the way through the ranking (30% from the lowest value). In practice, this could be achieved with the percentile function in Excel, but for QC calculations the Statistical Analysis Software® (SAS)95 PROC UNIVARIATE routine is used.96 Since in many cases, the precise 70th percentile falls between two values, interpolation between the two values surrounding the 70th percentile is needed. The average, weighted by proximity to the 70th percentile, of the two values is used.97 In Figure 1, interpolation is not needed since there are exactly 100 values in the data set and the 70th percentile corresponds to a discrete value in the data.
The included hours for the wind and solar QC calculations are shown in Table 2. The included hours vary seasonally and are based on the time of system peak demand.
Jan-Mar, Nov and Dec: |
HE17 - HE2198 (4:00 p.m. - 9:00 p.m.) |
Apr-Oct: |
HE14 - HE18 (1:00 p.m. - 6:00 p.m.) |
Table 2. Included Hours for QC Calculations
36 months of production data (Actual Settlement Quality Meter Data, as described in Section 4) are used for the QC calculation. Staff uses the three most recent years of complete data available (i.e. for 2009 QC values, 2005-2007 data). As noted below, most of the following steps are repeated for each of the 36 months; then the three years are averaged to result in 12 final monthly values.
The first step in calculating QC of wind and solar resources is to calculate the 70% exceedance for each time period. This is called the Initial QC. An initial QC is calculated for each resource for each of the 36 months.
Equation 1. Initial QC
Differences in production profiles across different individual wind or solar resources are called diversity. The exceedance of the sum of a diverse group of resources is always greater than or equal to the sum of the exceedances of the individual resources (i.e. the initial QCs). Any difference between the exceedance of the sum and the sum of the initial QCs is called the diversity benefit. The total benefit of diversity is the difference between the 70% exceedance of all wind and solar resources as a group and the sum of the initial QCs of all individual resources. The system diversity benefit is calculated for each of the 36 months.
Equation 2. System Diversity Benefit
The benefits of resource diversity are allocated to all wind and solar resources on the basis of energy produced during included hours. Each resource's diversity share is calculated as the kWh produced during the included hours by that resource divided by the kWh produced by all wind and solar resources during the same time period. The resource specific diversity benefit is the product of the resource diversity share and the system diversity benefit. No resource may have a calculated QC that exceeds its maximum capacity (maximum capacity is the 1st percentile exceedance of the resources production during all hours of the month). Therefore, this process is repeated in "passes" (for each of the 36 months) until the entire system diversity benefit (for the month) is allocated to specific resources and no resources have calculated QC greater than maximum capacity. For the first pass, all resources are included, but in any passes after the first, only resources with calculated QCs from the previous pass that are less than maximum capacity. The resource diversity benefit is calculated for each resource for each of the 36 months. It is possible that some of the 36 months may require multiple passes while other months require only a single pass.
The sum of a resource diversity benefit and a corresponding initial QC is referred to as a calculated QC. As noted above, the calculated QC cannot exceed the maximum capacity. If the calculated QC would exceed the maximum capacity, the calculated QC is set to the maximum capacity and the amount of the resource diversity benefit that is beyond the maximum capacity is considered the residual resource diversity benefit. The residual resource diversity benefits of all resources are summed to become the system diversity benefit used in the following pass. For the first pass, the initial QCs are used in Equation 5 for the calculated QC of the previous pass (i.e. CalculatedQCPass-1).
Equation 5. Calculated QC for Existing Resources
Equation 6. System Diversity Benefit for Pass 2 and any later Passes
If Equation 6 yields a positive system diversity benefit, a new pass is initiated, beginning withEquation 3. Only the resources which have a calculated QC less than maximum capacity from the just completed pass are included in the calculations during the new pass.
After the proceeding steps are completed, each existing resource has 36 initial QCs and 36 corresponding resource diversity benefits. Therefore, each existing resource has 36 calculated QCs.New resources, which do not have the complete 36 months of data, have calculated QCs for any month(s) which they do have data. For each month that a new wind (solar) resource does not have an initial QC and resource diversity benefit, it receives a calculated QC value based on the performance (i.e. calculated QC) of all wind (solar) resources that existed during that month. This value is the average calculated QC as a fraction of the available capacity of all of the wind (solar) resources in that month. The available capacity is calculated as the 1st percentile exceedance value of all hours in the month. This value is multiplied by the Net Dependable Capacity (NDC) of the new resource, as recorded in the Generation Capability List.
Equation 7. Calculated QC for New Wind (Solar) Resources
Now each and every wind and solar resource has 36 QC calculated values. To calculate the final 12 monthly QC values, the three corresponding months are averaged for each resource. For example, the three January values are averaged to calculate the final January QC.
Equation 8. Final QC
The preceding description is a conceptual approach to the calculations of wind and solar QC values. In practice, the calculations are performed in a SAS® program.
Non-dispatchable generation resources not described in previous sections receive monthly QC values based on a three-year rolling average of production during certain hours, shown in Table 2. The three most recent years of available data are used; for example, 2010 QC is calculated based on 2006-2008 data. Historical production data is adjusted for scheduled outages as described in Section 6. SAS® code for these calculations is included in the Appendix.
For this calculation, each monthly value is calculated as an average of the production during the specified hours. The 36 monthly average values are calculated as:
Equation 9. Monthly Average Production for Non-Dispatchable Resources
Then, the monthly values are averaged together for all (up to three) years of available data to calculate the final QC for each month.
Equation 10. Final QC of Non-Dispatchable Resources
New non-dispatchable resources with zero complete months of available data for any month shall receive QC for that month based on multiplying the resource's NDC by the average QC as a percent of NDC of all existing resources in this classification.
Equation 11. QC for Non-Dispatchable Resources with no Available Data
In D.09-06-028, CPUC directed that the QC of DR resources will be based on the Load Impact Protocols (LIPs) adopted by D.08-04-050.99 However, the LIPs provide far more detailed information than 12 monthly QC values. The discussion of the LIPs in this Manual does not in anyway impact the requirements of any previous decision in the DR proceedings or any other uses of the LIPs besides QC calculations.
The LIPs must be followed by the entity (typically the Investor Owned Utility{IOU}) requesting that the DR program be eligible for meeting RA Requirements. That entity must work with Energy Division staff to provide at least the LIP information described below for the DR resource to receive QC values. The following table summarizes the use of LIPs for QC demonstration. Event based resources (i.e. AC cycling) are DR programs that only operate when a specific event is called while non-event based resources (i.e. Time-Of-Use rates or permanent load shifting) operate each day, regardless of whether or not a DR event is "called". Page and section references in this table refer to Attachment A to D.08-04-050.
The monthly QC of a DR resource is the average expected (ex ante) load impact measured over certain measurement hours. The measurement hours are:
RA Compliance Year |
Hours | |
2011 |
Hour Ending (HE) 15 to HE 18 (2:00 p.m. to 6:00 p.m.) | |
2012 and beyond, except for programs that have a different, fixed operational period set by CPUC decision. |
Jan-Mar, Nov and Dec: |
HE 17 to HE 21 (4:00 p.m. - 9:00 p.m.) |
Apr-Oct: |
HE 14 to HE 18 (1:00 p.m. - 6:00 p.m.) | |
Table 3. Measurement Hours for DR
The hourly estimates for each of these hours from the LIP data are averaged together. These hourly estimates must be provided according to protocols 17, 21, 22, and 23. Other protocols described in this table are required for supporting data and report formatting.
Resource Type |
Load Impact Protocols Required |
Event Based Resources. Example IOU programs: CPP CBP DBP AC Cycling OBMC |
Ex Post for Event Based Resources Protocol 7 requires impact estimates be reported in a table format. Uncertainty adjustments are not needed in the table. Protocol 8 requires reporting for the average across all participants notified on an average event day over the evaluation period. Only the hourly load drop across all participants notified on an average event day is required; no need to provide the following details: · Each day on which an event was called; · The average event day over the evaluation period · For the average across all participants notified on each day on which an event was called; · For the total of all participants notified on each day on which an event was called. Protocol 10 requires regression based methods (read section 4.2.2, pg 60 for an overview of regression analysis). Any suppliers choosing not to use regression as described in Protocol 10 must file an evaluation plan (Protocols 1-3) well in advance of the QC demonstration deadline.100 Ex Ante for Event Based Resources Protocol 17 requires that ex ante estimates should be informed by ex post whenever possible. Protocol 21 requires impact estimates be reported in a table format. Uncertainty adjustments are not needed in the table. Protocol 22 requires the use of 1-in-2 weather year for the monthly system peak day. The 1-in-10 weather year, typical event day, or an average weekday for each month are not needed for QC calculation. Protocol 23 requires ex ante estimates be based on regression methodologies (read section 6.2, pg 98 for guidance). Portfolio Impacts, if Required Protocol 24 describes methodology for estimating the impacts of multiple DR programs within a portfolio. All DR resources whose participants also participate in other DR programs (potentially operated by other entities) must follow Protocol 24; such resources should also submit an evaluation plan (Protocols 1-3). Sampling if Required Protocol 25 requires certain procedures to ensure that sampling bias is minimized. Protocol 25 is not anticipated to be required for most DR resources using LIPs only to demonstrate QC; DR resources with a small number of participating customers should provide data from all participants, obviating the need for sampling methodologies. For resources with enough participants to adopt a sampling methodology, an evaluation plan (Protocols 1-3) is required well in advance of the QC demonstration deadline. Reporting Protocols Protocol 26 lists certain sections that should be included in the evaluation reports. These reports may be limited in scope, as described above. |
Non-Event Based Resource. Example IOU programs: TOU RTP SLRP PLS |
Ex Post for Non-Event Based Resources Protocol 14 (same as Protocol 7) requires impact estimates be reported in a table format. Uncertainty adjustments are not needed in the table. Protocol 15 requires reporting for the monthly system peak day. Protocol 16 requires regression based methods (read section 5.2, pg 84 for guidance). Any suppliers choosing not to use regression as described in Protocol 10 must file an evaluation plan (Protocols 1-3) well in advance of the QC demonstration deadline. Ex Ante for Non-Event Based Resources Protocol 17 requires ex ante estimates should be informed by ex post whenever possible. Protocol 21 requires impact estimates be reported in a table format. Uncertainty adjustments are not needed in the table. Protocol 22 requires the use of 1-in-2 weather year for the monthly system peak day. The 1-in-10 weather year, average weekday, or typical event day are not needed for QC calculation. Protocol 23 requires ex ante estimates be based on regression methodologies (read section 6.2, pg 98 for guidance). Portfolio Impacts, if Required Protocol 24 describes methodology for estimating the impacts of multiple DR programs within a portfolio. All DR resources whose participants also participate in other DR programs (potentially operated by other entities) must follow Protocol 24; such resources should also submit an evaluation plan (Protocols 1-3). Sampling if Required Protocol 25 requires certain procedures to ensure that sampling bias is minimized. Protocol 25 is not anticipated to be required for most DR resources using LIPs only to demonstrate QC; DR resources with a small number of participating customers should provide data from all participants, obviating the need for sampling methodologies. For resources with enough participants to adopt a sampling methodology, an evaluation plan (Protocols 1-3) is required well in advance of the QC demonstration deadline. Evaluation Reporting Protocol 26 lists certain sections that should be included in the evaluation reports. These reports may be limited in scope, as described above. |
Table 4. Required LIPs
As noted above, in order to summarize the detailed LIP information to monthly QC values, QC is measured using the average expected (ex ante) load impact during the appropriate measurement hours shown in Table 3. CPUC staff takes the hourly estimates provided101 according to the LIPs and averages the estimates over the relevant hours.
In order for DR programs to receive local capacity credit for RA, the load impact must be broken down by local areas. However, this breakdown is not required for all months - it is only required for August. Further, for compliance purposes the CPUC aggregates PG&E's "other" local areas: Fresno, Humboldt, North Coast/North Bay, Sierra, and Stockton. These areas do not need to be broken out individually. For August, average expected (ex ante) load impact must be provided by local area as follows, for each DR program:
SDG&E |
SCE |
PG&E |
San Diego |
Big Creek/Ventura |
Greater Bay Area |
System (no local area) |
LA Basin |
Other PG&E local areas |
System (no local area) |
System (no local area) | |
Program Total |
Program Total |
Program Total |
Table 5. Local Area Breakdown for DR Resources.
For each program, the sum of system and local capacities should equal the program total capacity. Table 5 is not intended to be a format, but simply a description of the data required. If a program operates in multiple IOU territories, expected load impacts for all relevant local areas should be included.
Avoided line losses should be included along with the LIP estimates for QC calculation purposes, but not directly included in the LIP estimates. CPUC staff will "gross-up" the DR QC for avoided line losses. A single loss rate for each service area is calculated according to Equation 12. Total Line Loss Factor
Equation 12. Total Line Loss Factor
The service area specific distribution loss rate is calculated from the most recent available data submitted in each IOUs current or previous general rate case. Generally, in the rate cases the IOUs submit loss factors from each of several locations on the transmission and distribution grid. The ratio of the transmission loss factor to the secondary distribution loss factor yields the loss rate for sub-transmission and distribution, which is called the distribution loss rate.
Equation 13. Distribution Loss Rate
Finally, the QC of DR is calculated by grossing up by the loss rate.
Equation 14. Final QC of DR
Acronym |
Definition |
CAISO ID |
California ISO Scheduling Resource ID |
California ISO |
California Independent System Operator |
CEC |
California Energy Resources Conservation and Development Commission |
CPUC |
California Public Utilities Commission |
HE |
Hour Ending |
IOU |
Investor Owned Utility |
kW |
Kilowatt |
kWh |
Kilowatt-hour |
LGIP |
Large Generator Interconnection Procedures |
LIP |
Load Impact Protocol |
MW |
Megawatt |
MWh |
Megawatt-hour |
NQC |
Net Qualifying Capacity |
PMax |
Maximum Power Plant Output |
QC |
Qualifying Capacity |
RA |
Resource Adequacy |
SAS® |
Statistical Analysis Software |
SC |
Scheduling Coordinator |
SGIA |
Small Generator Interconnection Agreement |
SGIP |
Small Generator Interconnection Procedures |
SLIC |
Scheduling and Logging for ISO of California |
78 http://www.caiso.com/14d4/14d4c4ff59780.html.
79 See Appendix U of the California ISO Tariff: http://www.caiso.com/2471/2471994c26350.pdf. See also: Section 5.1.3.4 of CAISO's Business Practice Manual for Reliability Requirements: https://bpm.caiso.com/bpm/bpm/version/000000000000011.
80 The exception to this rule is reduction in deliverability caused by any degradations of the transmission system which are not repaired promptly, for example due to fires or other force majeure events.
81 http://www.caiso.com/23d7/23d7e41c14580.pdf.
82 CAISO Tariff Appendix A, Fourth Replacement Volume No. 2, Sheet No. 863: http://www.caiso.com/2471/2471974a121c0.pdf.
83 See Appendix S to the California ISO Tariff: http://www.caiso.com/2471/247198fe24690.pdf.
84 SGIA interconnections use the Wholesale Distribution Access Tariff (WDAT).
85 The scheduled outage criterion was adopted by D.06-07-031. For more information, see Section 13 of the 2010 RA Guide: http://www.cpuc.ca.gov/NR/rdonlyres/14DFD39E-40C6-4FAF-8C36-38F8708BC23A/0/RAGuide2010.doc.
87 D.09-06-028 at 29.
88 For more information about SLIC, see: http://www.caiso.com/docs/2005/10/28/200510281047542112.html.
89 See Error! Reference source not found.
90 See also, Section 5 of CAISO's Business Practice Manual for Reliability Requirements: https://bpm.caiso.com/bpm/bpm/version/000000000000011.
91 California ISO coordinates with SCs for resources to schedule PMax tests at a time selected by the SC. Generally, SCs select the timing of a PMax test to demonstrate output of the resource at or near its maximum possible output.
92 See http://www.caiso.com/1796/179697c864850.xls.
93 Adopted in D.09-06-028, Appendix C.
94 The production profile in the figure is generated randomly and is not intended to represent any particular resource or classification of resources.
95 For more information about SAS®, see http://www.sas.com/technologies/analytics/statistics/stat/index.html.
96 See Error! Reference source not found.
97 See the description of the PCTLDEF=1 at: http://support.sas.com/documentation/cdl/en/procstat/59629/HTML/default/procstat_univariate_sect028.htm.
98 HE indicates "hour ending", or the 60 minutes that end at the numbered hour, in 24 hour time. For example, HE17 indicates the 60 minutes beginning at 16:00
(i.e. 4:00 p.m.) and ending at 16:59.
99 The LIPs are detailed in Appendix A to D.08-04-050; http://docs.cpuc.ca.gov/WORD_PDF/FINAL_DECISION/81979.PDF.
100 The deadline is typically April 1.
101 If assumptions underlying the LIP estimates for a particular program are unreasonably optimistic, CPUC staff accordingly reduces the load impacts.
