MEETING NOTES

CALIFORNIA ENERGY COMMISSION

DEMAND ANALYSIS WORKING GROUP (DAWG)

Hearing Room B

California Energy Commission

1516 Ninth Street

Sacramento, CA 95814

Tuesday, October 23, 2012

10:00 am-4pm PDT

AGENDA

10:00 Welcome and Introductions -- Dickerson

10:05 Agenda Overview – Dickerson

10:10 Updates

·  Incremental-Uncommitted Energy Efficiency – Kavalec

There was a slight adjustment to the ‘high’ case, about 20MW, and the drafts currently on the website have been updated. The analyses can be found at:

http://www.energy.ca.gov/2012_energypolicy/documents/index.html

·  2013 Integrated Energy Policy Report – Kavalec

The forms and instructions workshop regarding submission of utility data and forecasts for the 2013 IEPR workshop at CEC will be held on October 30, 2012 @ CECThe link to the workshop is:

http://www.energy.ca.gov/calendar/events/index.php?com=detail&eID=1764

Next, a meeting to discuss econometric-demographic inputs to the 2013 forecast will be held on January 15, 2013. The main vendors of econ-demo data will be presenting (Global Insight and Economy.com/Moody’s). There will be a panel on economics within CA, one on demography (population and migration trends) within CA and final a business panel with business leaders describing the state’s business climate (Chamber of Commerce, CA Building Assn., etc.)

In late February 2012 there will be a workshop on assumptions and inputs going into the forecast, rate projections, etc. Both electricity and natural gas will be discussed. A firm date has not yet been established.

The preliminary forecast will be released in Spring 2013 with a revised forecast in late Summer 2013. Similarly to the last cycle, there will be meetings with the DF pup to compare CEC and utility forecasts. Also, the process with the expert panel will be modified so that their assessment is developed in a more structured manner.

This cycle the forecasts will be developed at the climate zone level – this is a first step toward producing forecasts at a much more disaggregated level. The CEC’s forecasting climate zones will be used (there are 16 forecasting climate zones).

Several modules of the CEC’s forecasting model have been set up to forecast at the climate zone level. However, the input data are designed at the planning zone level. Thus, additional uncertainty is being introduced. The forecasters are working on ways to mitigate this uncertainty.

Ideally the CPUC Goals Study will be finished in time for the CEC forecasters to produce a new incremental-uncommitted forecast. If not, the current estimates will be used, possibly updated somewhat.

·  Update on LTPP Status – Skinner/Young

The LTPP determines needs for new resources and authorizes procurement for the IOUs. Typically there is a 10 year planning horizon. However, this cycle, an additional 10 years is being analyzed – for policy issues only (not procurement) in Track 2.

The current status of LTPP is: the 2012 LTPP cycle is proceeding in three tracks. Track 1 is determining local capacity area (LCA) needs in the LA Basin and Big Creek Ventura, especially given the extensive once through cooling (OTC) retirements in the next nine years. A proposed decision is expected by the end of the year. A second track is to determine system needs. In the past, tables with loads and resources were used. This cycle, a set of scenarios was defined for modeling. Track 2 is two phases – define standardized planning assumptions and scenarios (PD expected by the end of 2012). Next year (2013) Track 2 will address operational flexibility modeling. That is, the needs for flexibility given the renewables that are coming on the system. The estimate of the net short position (resources that need to be obtained/purchased) is expected by the end of 2013. The third track of the LTPP is looking at IOU bundled customer procurement rules.

In Track 1 the need for resources is determined largely based on a study by CAISO that specifically examined the impact of OTC retirements. The CAISO did not include any of the adjustments for energy efficiency estimated by the Energy Commission, beyond those that are included in the base forecast. However, the efficiency in the base forecast is included. In Track 2, which looks at system needs, there are scenarios. A key assumption is the amount of energy efficiency savings beyond that which is avoided in the forecasts. The CPUC will be adopting a low- mid- and high- set of assumptions, based on the adjustments (previously known as incremental-uncommitted energy efficiency). The low case adopts the low-savings energy efficiency adjustments, the mid case adopts mid-level savings prepared by Energy Commission and the high case adopts high energy efficiency savings from the Energy Commission plus naturally occurring savings plus a low level of Big Bold Energy Efficiency Strategy (BBEES) savings. The BBEES savings estimates are from the 2008 CPUC energy efficiency study by Itron. Since these savings are uncertain, only the low level of BBEES savings discussed in that report are being used.

LTPP would like to obtain energy efficiency savings at lower levels of disaggregation, e.g., busbar. However, there is a tradeoff in terms of accuracy at these lower levels of disaggregation. LTPP would also like there to be qualitative assessments of certainty for the estimates. For example, zero-net-energy (ZNE) savings are uncertain, but savings from not-yet-funded energy efficiency from IOU programs is relatively more certain. So even attaching uncertainty to different types of efficiency (e.g., “10% of the new construction savings are likely to occur”) would be helpful.

Question – which scenarios are truly being used for procurement and which are being used for policy discussion? In concept, all of the scenarios could be given equal weight, but this depends in large part on party comments. Parties tend to feel that the base case should be used, and may carry more weight than the others. The CPUC will determine which scenarios will be considered more heavily. If all four cases, for example, show a need for new resources, the CPUC would consider that. Whereas if only one case showed a need for resources that would also be a consideration.

10:30 Use of Managed Forecast in CAISO Planning for Transmission Planning Process (beginning in 2013) - Billinton

·  Current CAISO activities

·  Overall effect of energy efficiency (committed and uncommitted) by 2021

·  CPUC and CPUC interpretation of continuation of Energy Efficiency

·  Incorporation of incremental-uncommitted energy efficiency in the March 2013 Forecast

Transmission planning process is an annual process that occurs in three stages. Phase 1 is for planning assumptions (February of a given year the analyses are posted), where stakeholders can comment on reports such as the Reliability Assessment (methodology) The study plan is finalized in March each year which is when studies are initiated. That is the timeframe where assumptions are needed. Key inputs to planning are reliability standards and criteria for what is needed to meet the planning and operational criteria. Generation assumptions take into account resources that are due to come online, and load assumptions come largely from the Energy Commission forecast. The criteria are North American Reliability Standards -- national standards. Transmission planning criteria are described in scenarios that are expressed in terms of normal operations (performance and boundary conditions for normal operations). Scenarios are also developed for different types of operational criteria – contingencies if different parts of the system are not operating normally. The WECC regional criteria are part of the Western interconnection. Within WECC criteria are different regional standards, e.g., for adjacent 500kV circuits. The studies are typically done in two terms, 1-5 years and 5-10 years. Different analyses can require more or fewer years in each breakout. Conditions such as dispatch conditions, load levels, etc. that can affect performance are modeled. When trying to plan the system, the plans are made for boundary conditions. In essence this is a three dimensional diagram: what is the load, what is the available generation and what are the available interconnections. The forecasts are based on peak (but under different conditions such as full peak, light peak).

The CAISO’s analyses are used in the CPUC’s LTPP. TPP (transmission planning process) is not an official proceeding at the CPUC. Rather, it affects CAISO’s tariff. The TPP results are fed into the LTPP, for local area requirements in LTPP Track 1. The system has to meet NERC standards. If analyses suggest that there are areas that do not meet the standards, corrective action plans must be developed. These needs can either be assessed on an annual basis, or near- (1-5) and long (5-10) year terms. A typical approach, and one that was implemented this year, is to study year 2. That is because when the plan is published, year 1 is already underway. The results are published approximately August 15. In September workshops are held for stakeholders. Participating Transmission Owners (PTOs) present their proposed solutions . In mid-October general stakeholders were invited to propose solutions. At this time in the cycle, October, CAISO is assessing whether the proposed solutions meet the planning criteria. In January, CAISO will present study results assessing whether the proposed solutions meet the criteria. ISO recommends upgrades and the ISO board approves the plan. Participating transmission facility owners must comply with tariff requirements. In the open process once a deficiency has been identified, it is theoretically possibly that demand side resources could be proposed but that has not happened yet.

Question, since the state has a loading order, with energy efficiency first in the loading order, are there any related requirements that demand side options such as efficiency are considered first for transmission planning? No the system doesn’t work quite that way. For large areas the requirements are focused on resource adequacy. When looking at the local level, the criteria have to do with contingencies. Can the system respond like a generator so it can maintain reliability – can it respond quickly enough?

The planning begins with the preliminary Energy Commission forecast. The revised forecast is incorporated later. The Energy Commission forecast is translated to the bus level. When the local area planning is analyzed, it is on a 1 in 10 scenario. The bulk forecast uses a 1 in 5 scenario for the system peak which looks at the likelihood of the northern and southern parts of the state peaking at the same time. For economic studies, the process examines a 1 in 2 scenario. The weather-normalized events form the scenarios, e.g., 1 chance in 2 of a weather event occurring (this would be a 50-50 chance of the requirements being higher or lower). This is largely based on the 6-3-1 term analytic used in forecasting (60% of today’s temp, 30% of yesterday’s and 10% of two days previous for maximum high temperatures); weekday vs. weekend are also included. Most analyses also include a minimum temperature component, not just maximum.

The base forecast includes committed energy efficiency that is in the base Energy Commission/IEPR forecast. The uncommitted energy efficiency is not included since there is uncertainty regarding the timing and location. Knowing the location and impact is critical, especially at the busbar level in order for the effects to be included in planning.

Last year there was a sensitivity analysis for once through cooling (OTC) that used the incremental uncommitted energy efficiency at the busbar level (prepared by Energy Commission). The OTC studies are just starting right now for this cycle. Energy Commission is planning to deliver a similar analysis of incremental-uncommitted energy efficiency just after the beginning of the year, and this will be included in a sensitivity analysis in the 2013 transmission plan. Last year this was done as an addendum.

CPUC described that portions of incremental-uncommitted energy is not necessarily “uncertain.” The impacts described in the CPUC Energy Efficiency Goals study are based on 30 years of experience with energy efficiency and CPUC plans to continue promoting similar levels of efficiency. The current assumptions/projections about energy savings, however, are very conservative. Compared to other impacts that are considered in the forecasts (economic and demographic drivers, impacts of electric vehicles, etc.), the efficiency included in the goals study and therefore the incremental energy efficiency adjustment is comparatively certain. That having been said, failing to include incremental energy efficiency adjustments (that is, program and code activity not yet “on the books” and included in the CEC’s base forecast) is in essence making an assumption that no energy efficiency impacts of those types would occur. Making this assumption is, therefore, not in line with state policy since the state does intend to continue energy efficiency programs and codes and standards implementation.

Since CAISO does its planning at a lower level of disaggregation, it’s not only uncertainty about whether the additional energy efficiency would occur but where and when it would occur that poses an impediment to including incremental energy efficiency in the transmission planning process and assessments.

A proposed next step for determining how to include incremental uncommitted energy efficiency more effectively in CAISO’s process would be to determine the variability of delivered impacts at the busbar level. This would provide an indication regarding whether projections of impacts likely to occur at the bus level are stable and reliable over time, and can be forecast based on past accomplishments at that level of disaggregation.

Last year, for the sensitivity analysis that was prepared by CAISO Energy Commission used an analytic method to spread results to the busbar level. A data request was prepared for the utilities seeking the relative proportion of industrial, residential and commercial peak energy use at each busbar. These values were used to spread the impacts of energy efficiency at the busbar level by these major customer classes for each service area. The Energy Commission’s mid-case scenario for incremental energy efficiency was used. Improvements upon the method under discussion for this year could involve matching this top-down approach with a bottom-up approach, based on the availability of energy efficiency impact data that can be tied to certain locations. The build up could be, for example, to a regional level. This could work in conjunction with CEC’s decision to produce the 2013 forecast at the climate zone level.

An issue to be considered is that the size of busses varies quite widely between utilities in terms of how they are modeled. For example, the way PG&E’s system is configured there are about 1400-1500 busses. SCE’s system has about the same load but only 140 busses – the average bus, therefore, in SCE’s territory is about 10 times bigger than in PG&E’s territory, because the analysis is rolled up to a higher level. There are busses in SCE territory approaching 1000 MW; San Diego’s system is 3000-4000 total. The PG&E system goes to the 60-70 kV level. In SCE’s service territory, busses are at the 230 kV level. The hardware is configured differently between the two systems.