Load as a Resource

The Demand Response Spinning Reserve project is a pioneering demonstration of how using existing utility load-management assets can provide an important electricity system reliability resource known as spinning reserve. Providing spinning reserve with aggregated demand-side resources such as those illustrated in this report will give grid operators at California Independent System Operator (CAISO) and Southern California Edison (SCE) a powerful, new tool to improve system reliability, prevent rolling blackouts, and lower system operating costs.

The work completed to date to demonstrate the use of demand-response as spinning reserve has produced important programmatic and technical insights, including:

• Target-marketing a utility's air-conditioning load-cycling program to customers served by a single distribution feeder can be a successful strategy. SCE successfully recruited a high proportion (nearly one-third) of eligible customers to participate in the demonstration.
• Repeated curtailment of these customers' air-conditioning in a manner similar to the deployment of spinning reserve can be accomplished without a single customer complaint. SCE curtailed these customers' air-conditioning units 37 times during the final portion of Southern California's cooling season for durations lasting from five to nearly 20 minutes, and did not receive any customer complaints regarding the curtailments.
• Real-time visibility of load curtailments can be achieved through an open data platform and secure website. The project team demonstrated a highly flexible, open yet secure data-integration, archival, and presentation platform that allowed external audiences (e.g., electricity grid operators) to see curtailments in real time. Using such a platform and website could significantly lower the costs of this service relative to current practices.
• Analysis methods developed for this project could one day be used to predict magnitude of load curtailments as a function of weather and time of day. The project team developed statistical methods to estimate the load that would have been observed without a curtailment and means for comparing this estimated load to actual loads observed during curtailments. The team also conducted exploratory analyses that confirmed the existence of a relationship between the magnitude of the load curtailment, and ambient weather conditions and to a lesser, but still suggestive extent, time of day.
• Load curtailments can be fully implemented much faster than ramping up of spinning reserve from thermal generation. The project team measured full load response in less than 20 seconds and identified technical opportunities to further increase the response speed.

Demand response is the largest underutilized reliability resource in North America. Historic demand response programs have focused on reducing overall electricity consumption (increasing efficiency) and shaving peaks but have not typically been used for immediate reliability response. Many of these programs have been successful but demand response remains a limited resource. The Federal Energy Regulatory Commission (FERC) report, Assessment of Demand Response and Advanced Metering (FERC 2006) found that only five percent of customers are on some form of demand response program. Collectively they represent an estimated 37,000 MW of response potential. These programs reduce overall energy consumption and they also reduce stress on the power system at times of peak loading.

More recently demand response has begun to be considered, and in some cases actually used, to directly supply reliability services to the power system. Rather than reducing overall power system stress by reducing peak loading over multiple hours these programs are targeted to immediately respond to specific reliability events. This is made possible by advances in communications and controls and has benefits for the power system and the load.

Unfortunately, preconceptions concerning load response capabilities, coupled with misunderstandings of power system reliability needs, are limiting the use of responsive loads. In many places loads are prohibited from providing the most valuable reliability services in spite of their being evidence that their response can be superior to that of generators. This is denying the power system of a valuable reliability resource. It is also denying loads the ability to sell valuable services.

This report addresses a number of common misconceptions concerning responsive load and power system reliability interactions. It is structured as a set of short questions and answers and is intended for power system operators, planners, regulators, load owners, and other interested parties.

The report is organized into three chapters. Chapter 1 is this introduction. Chapter 2 contains questions and answers on demand response and power system reliability. Chapter 3 provides conclusions and recommendations.

Under real time pricing (RTP) tariffs, electricity consumers are charged prices that vary over short time intervals, typically hourly, and are quoted one day or less in advance to reflect contemporaneous marginal supply costs. RTP differs from conventional retail tariffs, which are based on prices that are fixed for months or years at a time to reflect average, embedded supply costs. In recent years, a resurgence of interest in RTP has occurred. Economists recognize that providing electricity consumers with price incentives to reduce their usage when wholesale prices rise would improve the performance of wholesale electricity markets in two important ways: mitigating suppliers' ability to exercise market power and dampening price volatility. Policymakers engaged in electric utility resource planning have also recognized that, by reducing peak demand, RTP could play an important role in a portfolio of strategies for cost-effectively meeting utility load obligations. While other mechanisms can be used to induce price responsive demand and/or reduce peak demand, many economists argue that RTP represents the most direct and efficient approach, and therefore it should be the primary focus of policymakers' efforts to improve the performance of wholesale and retail electricity markets (Borenstein et al. 2002).

While clearly appealing from a theoretical perspective, questions remain about the extent to which RTP can ultimately affect wholesale market performance and utility resource planning. First, assuming that RTP is offered on a voluntary basis, how many customers would choose to enroll in RTP, given the additional risks and transaction costs compared to traditional, fixed price retail supply service? Second, even if a sizable number of customers did choose to enroll, to what extent, and how consistently, would a diverse population of participants respond to the prices they face? Some insight into these issues can be gleaned from experiences with several prominent RTP programs frequently featured in the literature. However, to understand the potential role of RTP in settings with substantially different types of customers and/or different market and regulatory conditions, policymakers require a wider base of experience.

This study attempts to address some of these information gaps through an in-depth case study of 149 large commercial and industrial customer accounts served by Niagara Mohawk Power Corporation (NMPC). In October 1998, with the commencement of retail access in New York, NMPC replaced the existing time-of-use (TOU) tariff for large customers with peak demand in excess of two megawatts with a day-ahead, market-based RTP rate design. This new default SC-3A service, called "Option 1", recovers fixed costs (e.g., transmission and distribution) largely through demand charges and prices electric commodity at hourly-varying prices indexed to the NYISO day-ahead market. Hourly prices for the next day are transmitted to customers by 4pm.

This document presents a research plan to increase the use of demand response and enhance power system reliability. This research plan is based only upon interviews and discussions with California Independent System Operator (CAISO) staff.

This document presents a research plan to increase the use of demand response and enhance power system reliability. This research plan is based only upon interviews and discussions with California Independent System Operator (CAISO) staff.

The Oak Ridge National Laboratory (ORNL), at the request of the California Energy Commission and the U.S. Department of Energy, investigating opportunities for electrical load to provide the ancillary service of spinning reserve to the electric grid. The load would provide this service by stopping for a short time when there is a contingency on the grid such as a transmission line or generator outage. There is a possibility that a significant portion of the California Independent System Operator's (CAISO's) spinning reserve requirement could be supplied from the California Department of Water Resources (CDWR) pumping load.

Responsive load is the most underutilized reliability resource available to the power system today. It is currently not used at all to provide spinning reserve. Historically there were good reasons for this, but recent technological advances in communications and controls have provided new capabilities and eliminated many of the old obstacles. North American Electric Reliability Council (NERC), Federal Energy Regulatory Commission (FERC), Northeast Power Coordinating Council (NPCC), New York State Reliability Council (NYSRC), and New York Independent System Operator (NYISO) rules are beginning to recognize these changes and are starting to encourage responsive load provision of reliability services. This report provides detailed results from one example technology, the Carrier Comfort Choice responsive thermostats deployed in the Long Island Power Authority (LIPA) LIPA edge program to provide peak demand reduction through central control of residential and small commercial air-conditioners.

This report describes a framework for evaluating candidate participants in a distributed resource aggregation program, as envisioned in the Bonneville Power Administration's (BPA) EnergyWeb initiative. The framework includes definition of system goals, relevant material that characterizes a distributed energy resource (DER) participant, rules for evaluating candidate participants, and a process that utilizes this information to produce a list of the most attractive candidates.

Summer 2002 was the second year of operation for the New York Independent System Operator's (NYISO) suite of Price Responsive Load (PRL) Programs: the Day-Ahead Demand Response Program, the Emergency Demand Response Program, and the third year of operation for the Installed Capacity Program/Special Case Resources program. It also marked the second year that the New York State Energy Research Authority (NYSERDA) provided funding to support participation in these programs. NYISO and NYSERDA commissioned Neenan Associates, augmented by significant professional staff resources provided by the Consortium for Electric Reliability Technology Solutions with U.S. Department of Energy funding, to conduct a comprehensive evaluation of the performance of the 2002 PRL programs.

This report summarizes the feasibility of providing spinning reserves from packaged through the wall air conditioning (PTAC) units. Spinning reserves, together with non-spinning reserves, compose the contingency reserves; the essential resources that the power system operator uses to restore the generation and load balance and maintain bulk power system reliability in the event of a major generation or transmission outage. Spinning reserves are the fastest responding and most expensive reserves. Many responsive load technologies could (and we hope will) be used to provide spinning reserve. It is also easier for many loads (including air conditioning loads) to provide the relatively shorter and less frequent interruptions required to respond to contingencies than it is for them to reduce consumption for an entire peak period. Oak Ridge National Laboratory (ORNL) is conducting research on obtaining spinning reserve from large pumping loads and from residential and small commercial thermostat controlled heating, ventilation and air conditioning (HVAC) units. The technology selected for this project, Digi-Log's retrofit PTAC controller, offers significant advantages.

To evaluate the availability of spinning reserve capacity from responsive heating and air conditioning loads, ORNL obtained data from a number of units operating over a year at a motel in the TVA service territory. A total of 24 PTAC units in as many rooms were fitted with Digi- Log's supervisory control unit that could be controlled from the motel front desk. Twelve of the rooms formed the group in which the controller was controlled from the hotel front desk only. The remaining twelve rooms were controlled by the occupant and formed the uncontrolled group. This enables us to evaluate the spinning reserve capacity from PTACS that were operating normally and from those under active energy management.

Price-responsive load (PRL) programs vary significantly in overall design, the complexity of relationships between program administrators, load aggregators, and customers, and the availability of "enabling technologies." Enabling technologies include such features as web-based power system and price monitoring, control and dispatch of curtailable loads, communications and information systems links to program participants, availability of interval metering data to customers in near real time, and building/facility/end-use automation and management capabilities. Two state agencies—NYSERDA in New York and the CEC in California—have been conspicuous leaders in the demonstration of demand response (DR) programs utilizing enabling technologies. In partnership with key stakeholders in these two states (e.g., grid operator, state energy agencies, and program administrators), Lawrence Berkeley National Laboratory (LBNL) and Pacific Northwest National Laboratory (PNNL) surveyed 56 customers who worked with five contractors participating in CEC or NYSERDA-sponsored DR programs. We combined market research and actual load curtailment data when available (i.e., New York) or customer load reduction targets in order to explore the relative importance of contractor's program design features, sophistication of control strategies, and reliance on enabling technologies in predicting customer's ability to deliver load reductions in DR programs targeted to large commercial/industrial customers.

We found preliminary evidence that DR enabling technology has a positive effect on load curtailment potential. Many customers indicated that web-based energy information tools were useful for facilitating demand response (e.g., assessing actual performance compared to load reduction contract commitments), that multiple notification channels facilitated timely response, and that support for and use of backup generation allowed customers to achieve significant and predictable load curtailments. We also found that 60-70% of the customers relied on manual approaches to implementing load reductions/curtailments, rather than automated load control response. The long-term sustainability of customer load curtailments would be significantly enhanced by automated load response capabilities, such as optimizing EMCS systems to respond to day-ahead energy market prices or load curtailments in response to system emergencies.

This report describes a market and load research study on a small group of participants in the NYISO Emergency Demand Response Program (EDRP) and the NYSERDA Peak Load Reduction and Enabling Technology Programs. In-depth interviews were conducted with 14 individual customers that participated in the NYISO EDRP program through New York State Electric and Gas (NYSEG), AES NewEnergy, and through eBidenergy/ConsumerPowerLine. These contractors used funding from NYSERDA to apply enabling technologies that were hypothesized to improve customers' ability to curtail load. Both NYSEG and eBidenergy/ConsumerPowerLine offered their customers access to their hourly load data on a day-after basis and, during curtailment events, on a near-real-time basis. Phone interviews were conducted with most customers, however 25% of customers provided initial responses to the survey protocol via email. We then combined the market research information with load data during the curtailment events of August 7-10, 2001 to evaluate the impact of technology on curtailment responses.

This paper reports on work the Consortium for Electric Reliability Technology Solutions (CERTS) has been pursuing to hasten the arrival of meaningful load participation in competitive electricity markets. The activities include: experimental economic analysis of the effect of price responsive load in reducing market prices and price volatility; assessments of emerging demand response programs and technologies for enabling customer participation in electricity markets, and demonstrations of load in providing ancillary services (notably, spinning reserve).

Utility load management programs, including direct load control and interruptible load programs, constitute a large installed base of controllable loads that are employed by utilities as system reliability resources. In response to energy supply shortfalls expected during the summer of 2001, the California Public Utilities Commission in spring 2001 authorized new utility load management programs as well as revisions to existing programs. This report provides an independent review of the designs of these new programs for a large utility (Southern California Edison) and suggests possible improvements to enhance the "price responsiveness" of the customer actions influenced by these programs. The report also proposes a new program to elicit a mass-market demand response to utility price signals.

Utility load management programs—including direct load control and interruptible load programs - were employed by utilities in the past as system reliability resources. With electricity industry restructuring, the context for these programs has changed; the market that was once controlled by vertically integrated utilities has become competitive, raising the question: can existing load management programs be modified so that they can effectively participate in competitive energy markets? In the short run, modified and/or improved operation of load management programs may be the most effective form of demand-side response available to the electricity system today. However, in light of recent technological advances in metering, communication, and load control, utility load management programs must be carefully reviewed in order to determine appropriate investments to support this transition.

This report investigates the feasibility of and options for modifying an existing utility load management system so that it might provide reliability services (i.e. ancillary services) in the competitive markets that have resulted from electricity industry restructuring. The report is a case study of Southern California Edison's (SCE) load management programs. SCE was chosen because it operates one of the largest load management programs in the country and it operates them within a competitive wholesale electricity market. The report describes a wide range of existing and soon-to-be-available communication, control, and metering technologies that could be used to facilitate the evolution of SCE's load management programs and systems to provision of reliability services. The fundamental finding of this report is that, with modifications, SCE's load management infrastructure could be transitioned to provide critical ancillary services in competitive electricity markets, employing currently or soon-to-be available load control technologies.

California's restructured electricity markets opened on 1 April 1998. The former investor owned utilities were functionally divided into generation, transmission, and distribution activities, all of their gas-fired generating capacity was divested, and the retail market was opened to competition. To ensure that small customers shared in the expected benefit of lower prices, the enabling legislation mandated a 10% rate cut for all customers, which was implemented in a simplistic way that fossilised 1996 tariff structures. Rising fuel and environmental compliance costs, together with a reduced ability to import electricity, numerous plant outages, and exercise of market power by generators drove up wholesale electricity prices steeply in 2000, while retail tariffs remained unchanged. One of the distribution/supply companies entered bankruptcy in April 2001, and another was insolvent. During this period, two sets of interruptible load programs were in place, longstanding ones organized as special tariffs by the distribution/supply companies and hastily established ones run directly by the California Independent System Operator (CAISO). The distribution/supply company programs were effective at reducing load during the summer of 2000, but because of the high frequency of outages required by a system on the brink of failure, customer response declined and many left the tariff. The CAISO programmes failed to attract enough participation to make a significant difference to the California supply demand imbalance. The poor performance of direct load participation in California's markets reinforces the argument for accurate pricing of electricity as a stimulus to energy efficiency investment and as a constraint on market volatility.

Recent electricity price spikes are painful reminders of the value that meaningful demand-side responses could bring to the restructuring US electricity system. Review of the aggregate offers made by suppliers confirms that even a modest increase in demand elasticity could dramatically reduce these extremes in price volatility. There is a strong need for increased customer participation in markets to enhance system reliability and reduce price volatility. Indeed, allowing customers to manage their loads in response to system conditions might be thought of as the ultimate reliability resource.

This scoping report provides a three-part assessment of the current status of efforts to enhance the ability of customer's load to participate in competitive markets with a specific focus on the role of customer loads in enhancing electricity system reliability. First, this report considers the definitions of electricity-reliability-enhancing ancillary services and a preliminary assessment of the ability of customer's loads to provide these services. Second, is a review a variety of programs in which load has been called on as a system reliability resource. These experiences, drawn from both past and current utility and ISO programs, focus on programs triggered by system condition (e.g., forecast reserves fall below a threshold), rather than those triggered by price (e.g., real-time prices). Third, the report examines the status of the underlying metering, communication, and control technologies required to enable customer loads to participate in competitive electricity markets. Following the three-part assessment, we offer preliminary thoughts on directions for future research.

Recent electricity price spikes are painful reminders of the value that meaningful demand-side responses could bring to the restructuring US electricity system. Review of the aggregate offers made by suppliers confirms that even a modest increase demand elasticity could dramatically reduce these extremes in price volatility. We submit that dramatically increased customer participation in these markets to enhance system reliability and reduce price volatility is sorely needed. Indeed, allowing customers to manage their loads in response to system conditions might be thought of as the ultimate reliability resource. Toward this end, this paper outlines an agenda for public-interest R&D in support of this objective that includes the following key elements:

1. Addressing the information needs and control requirements of system operators;
2. Repositioning existing utility load management assets;
3. Accelerating the transfer emerging program experiences;
4. Pioneering promising new program design concepts;
5. Incorporating grid reliability considerations into the design and operation of end-use technologies; and
6. Disseminating technology and programmatic solutions through demonstrations