Demand Response Spinning Reserve Demonstration - Phase 2 Findings from the Summer of 2008

The Demand Response Spinning Reserve project is a pioneering demonstration showing that existing utility load-management assets can provide an important electricity system reliability resource known as spinning reserve. Using aggregated demand-side resources to provide spinning reserve as demonstrated in this project will give grid operators at the California Independent System Operator (CA ISO) and Southern California Edison (SCE) a powerful new tool to improve reliability, prevent rolling blackouts, and lower grid operating costs.

In the first phase of this demonstration project, we target marketed SCE's air-conditioning (AC) load-cycling program, called the Summer Discount Plan (SDP), to customers on a single SCE distribution feeder and developed an external website with real-time telemetry for the aggregated loads on this feeder and conducted a large number of short-duration curtailments of participating customers' air-conditioning units to simulate provision of spinning reserve. In this second phase of the demonstration project, we explored four major elements that would be critical for this demonstration to make the transition to a commercial activity:

1. We conducted load curtailments within four geographically distinct feeders to determine the transferability of target marketing approaches and better understand the performance of SCE's load management dispatch system as well as variations in the AC use of SCE's participating customers;
2. We deployed specialized, near-real-time AC monitoring devices to improve our understanding of the aggregated load curtailments we observe on the feeders;
3. We integrated information provided by the AC monitoring devices with information from SCE's load management dispatch system to measure the time required for each step in the curtailment process; and
4. We established connectivity with the CA ISO to explore the steps involved in responding to CA ISO-initiated requests for dispatch of spinning reserve.

The major findings from the second phase of this demonstration are:

1. Demand-response resources can provide full response significantly faster than required by NERC and WECC reliability rules.
2. The aggregate impact of demand response from many small, individual sources can be estimated with varying degrees of reliability through analysis of distribution feeder loads.
3. Monitoring individual AC units helps to evaluate the efficacy of the SCE load management dispatch system and better understand AC energy use by participating customers.
4. Monitoring individual AC units provides an independent data source to corroborate the estimates of the magnitude of aggregate load curtailments and gives insight into results from estimation methods that rely solely on distribution feeder data.