Demand response in U.S. electricity markets: Empirical evidence

Empirical evidence concerning demand response (DR) resources is needed in order to establish baseline conditions, develop standardized methods to assess DR availability and performance, and to build confidence among policymakers, utilities, system operators, and stakeholders that DR resources do offer a viable, cost-effective alternative to supply-side investments. This paper summarizes the existing contribution of DR resources in U.S. electric power markets. In 2008, customers enrolled in existing wholesale and retail DR programs were capable of providing ∼38,000 MW of potential peak load reductions in the United States. Participants in organized wholesale market DR programs, though, have historically overestimated their likely performance during declared curtailments events, but appear to be getting better as they and their agents gain experience. In places with less developed organized wholesale market DR programs, utilities are learning how to create more flexible DR resources by adapting legacy load management programs to fit into existing wholesale market constructs. Overall, the development of open and organized wholesale markets coupled with direct policy support by the Federal Energy Regulatory Commission has facilitated new entry by curtailment service providers, which has likely expanded the demand response industry and led to product and service innovation.     

Demand response modeling considering Interruptible/Curtailable loads and capacity market programs

Recently, a massive focus has been made on demand response (DR) programs, aimed to electricity price reduction, transmission lines congestion resolving, security enhancement and improvement of market liquidity. Basically, demand response programs are divided into two main categories namely, incentive-based programs and time-based programs. The focus of this paper is on Interruptible/Curtailable service (I/C) and capacity market programs (CAP), which are incentive-based demand response programs including penalties for customers in case of no responding to load reduction. First, by using the concept of price elasticity of demand and customer benefit function, economic model of above mentioned programs is developed. The proposed model helps the independent system operator (ISO) to identify and employ relevant DR program which both improves the characteristics of the load curve and also be welcome by customers. To evaluate the performance of the model, simulation study has been conducted using the load curve of the peak day of the Iranian power system grid in 2007. In the numerical study section, the impact of these programs on load shape and load level, and benefit of customers as well as reduction of energy consumption are shown. In addition, by using strategy success indices the results of simulation studies for different scenarios are analyzed and investigated for determination of the scenarios priority.     

Variability of electricity load patterns and its effect on demand response: A critical peak pricing experiment on Korean commercial and industrial customers

To the extent that demand response represents an intentional electricity usage adjustment to price changes or incentive payments, consumers who exhibit more-variable load patterns on normal days may be capable of altering their loads more significantly in response to dynamic pricing plans. This study investigates the variation in the pre-enrollment load patterns of Korean commercial and industrial electricity customers and their impact on event-day loads during a critical peak pricing experiment in the winter of 2013. Contrary to conventional approaches to profiling electricity loads, this study proposes a new clustering technique based on variability indices that collectively represent the potential demand–response resource that these customers would supply. Our analysis reveals that variability in pre-enrollment load patterns does indeed have great predictive power for estimating their impact on demand–response loads. Customers in relatively low-variability clusters provided limited or no response, whereas customers in relatively high-variability clusters consistently presented large load impacts, accounting for most of the program-level peak reductions. This study suggests that dynamic pricing programs themselves may not offer adequate motivation for meaningful adjustments in load patterns, particularly for customers in low-variability clusters.     

An exploratory analysis of California residential customer response to critical peak pricing of electricity

This paper summarizes the results from an exploratory analysis of residential customer response to a critical peak pricing (CPP) experiment in California, in which 15 times per year participating customers received high price signals dispatched by a local electricity distribution company. The high prices were about three times the on-peak price for the otherwise applicable time-of-use rate. Using hourly load data collected during the 15-month experiment, we find statistically significant load reduction for participants both with and without automated end-use control technologies. During 5-h critical peak periods, participants without control technology used up to 13% less energy than they did during normal peak periods. Participants equipped with programmable communicating thermostats used 25% and 41% less for 5 and 2 h critical events, respectively. Thus, this paper offers convincing evidence that the residential sector can provide substantial contributions to retail demand response, which is considered a potential tool for mitigating market power, stabilizing wholesale market prices, managing system reliability, and maintaining system resource adequacy.     

Optimal real time pricing in an agent-based retail market using a comprehensive demand response model

In this paper, a weighted combination of different demand vs. price functions referred to as Composite Demand Function (CDF) is introduced in order to represent the demand model of consuming sectors which comprise different clusters of customers with divergent load profiles and energy use habitudes. Derived from the mathematical representations of demand, dynamic price elasticities are proposed to demonstrate the customers’ demand sensitivity with respect to the hourly price. Based on the proposed CDF and dynamic elasticities, a comprehensive demand response (CDR) model is developed in this paper for the purpose of representing customer response to time-based and incentive-based demand response (DR) programs. The above model helps a Retail Energy Provider (REP) agent in an agent-based retail environment to offer day-ahead real time prices to its customers. The most beneficial real time prices are determined through an economically optimized manner represented by REP agent’s learning capability based on the principles of Q-learning method incorporating different aspects of the problem such as price caps and customer response to real time pricing as a time-based demand response program represented by the CDR model. Numerical studies are conducted based on New England day-ahead market’s data to investigate the performance of the proposed model.     

Robust bidding and offering strategies of electricity retailer under multi-tariff pricing

In this paper, an electricity retailer seeks to determine selling price for end-user consumers under fixed pricing (FP), time-of-use pricing (TOU) and real-time pricing (RTP). Furthermore, in order to provide power exchange between the retailer and the power market, bidding and offering curves should be prepared to bid and offer to the day-ahead market. Therefore, this paper proposes a robust optimization approach (ROA) to obtain optimal bidding and offering strategies for the retailer. To achieve this, ROA is used for uncertainty modeling of power market prices in which the minimum and maximum limits of prices are considered for uncertainty modeling. Lower and upper bounds of price is consecutively subdivided into sequentially nested subintervals which allows formulating robust mixed-integer linear programming (RMIP) problem. The proposed RMIP model helps retailer to select a robust decision in the presence of market price uncertainty. Furthermore, the bidding and offering curves of the retailer are obtained from sufficient data through solving these problems. Meanwhile, the uncertainty of customers demand and variable climate condition are modeled based on stochastic programming. To validate the proposed robust optimization model, three case studies are evaluated and the results are compared.     

Option value of electricity demand response

As electricity markets deregulate and energy tariffs increasingly expose customers to commodity price volatility, it is difficult for energy consumers to assess the economic value of investments in technologies that manage electricity demand in response to changing energy prices. The key uncertainties in evaluating the economics of demand–response technologies are the level and volatility of future wholesale energy prices. In this paper, we demonstrate that financial engineering methodologies originally developed for pricing equity and commodity derivatives (e.g., futures, swaps, options) can be used to estimate the value of demand-response technologies. We adapt models used to value energy options and assets to value three common demand–response strategies: load curtailment, load shifting or displacement, and short-term fuel substitution—specifically, distributed generation. These option models represent an improvement to traditional discounted cash flow methods for assessing the relative merits of demand-side technology investments in restructured electricity markets.     

Industrial customer response to wholesale prices in the restructured Texas electricity market

This paper estimates the demand responsiveness of the 20 largest industrial energy consumers in the Houston area to wholesale price signals in the restructured Electric Reliability Council of Texas (ERCOT) market. Statistical analysis of their load patterns employing a Symmetric Generalized McFadden cost function model suggests that ERCOT achieved limited success in establishing a market that facilitates demand response from the largest industrial energy consumers in the Houston area to wholesale price signals in its second year of retail competition. The muted price response is at least partially because energy consumers who opt to offer their “interruptibility” to the market as an ancillary service are constrained in their ability to respond to wholesale energy prices.     

Flexible demand response programs modeling in competitive electricity markets

In recent years, extensive researches have been conducted on implementation of demand response programs (DRPs), aimed to electricity price reduction, transmission lines congestion resolving, security enhancement and improvement of market liquidity. Basically, DRPs are divided into two main categories namely, incentive-based programs (IBPs) and time-based rate programs (TBRPs). Mathematical modeling of these programs helps regulators and market policy makers to evaluate the impact of price responsive loads on the market and system operational conditions. In this paper, an economic model of price/incentive responsive loads is derived based on the concept of flexible price elasticity of demand and customer benefit function. The mathematical model for flexible price elasticity of demand is presented to calculate each of the demand response (DR) program’s elasticity based on the electricity price before and after implementing DRPs. In the proposed model, a demand ratio parameter has been introduced to determine the appropriate values of incentive and penalty in IBPs according to the level of demand. Furthermore, the importance of determining optimum participation level of customers in different DRPs has been investigated. The proposed model together with the strategy success index (SSI) has been applied to provide an opportunity for major players of the market, i.e. independent system operator (ISO), utilities and customers to select their favorite programs that satisfy their desires. In order to evaluate the performance of the proposed model, numerical studies are conducted on the Iranian interconnected network load profile on the annual peak day of the year 2007.     

A bilevel model for electricity retailers' participation in a demand response market environment

Demand response programmes are seen as one of the contributing solutions to the challenges posed to power systems by the large-scale integration of renewable power sources, mostly due to their intermittent and stochastic nature. Among demand response programmes, real-time pricing schemes for small consumers are believed to have significant potential for peak-shaving and load-shifting, thus relieving the power system while reducing costs and risk for energy retailers. This paper proposes a game theoretical model accounting for the Stackelberg relationship between retailers (leaders) and consumers (followers) in a dynamic price environment. Both players in the game solve an economic optimisation problem subject to stochasticity in prices, weather-related variables and must-serve load. The model allows the determination of the dynamic price-signal delivering maximum retailer profit, and the optimal load pattern for consumers under this pricing. The bilevel programme is reformulated as a single-level MILP, which can be solved using commercial off-the-shelf optimisation software. In an illustrative example, we simulate and compare the dynamic pricing scheme with fixed and time-of-use pricing. We find that the dynamic pricing scheme is the most effective in achieving load-shifting, thus reducing retailer costs for energy procurement and regulation in the wholesale market. Additionally, the redistribution of the saved costs between retailers and consumers is investigated, showing that real-time pricing is less convenient than fixed and time-of-use price for consumers. This implies that careful design of the retail market is needed. Finally, we carry out a sensitivity analysis to analyse the effect of different levels of consumer flexibility.     

An integrated ant colony optimization approach to compare strategies of clearing market in electricity markets: Agent-based simulation

In this paper, an innovative model of agent based simulation, based on Ant Colony Optimization (ACO) algorithm is proposed in order to compare three available strategies of clearing wholesale electricity markets, i.e. uniform, pay-as-bid, and generalized Vickrey rules. The supply side actors of the power market are modeled as adaptive agents who learn how to bid strategically to optimize their profit through indirect interaction with other actors of the market. The proposed model is proper for bidding functions with high number of dimensions and enables modelers to avoid curse of dimensionality as dimension grows. Test systems are then used to study the behavior of each pricing rule under different degrees of competition and heterogeneity. Finally, the pricing rules are comprehensively compared using different economic criteria such as average cleared price, efficiency of allocation, and price volatility. Also, principle component analysis (PCA) is used to rank and select the best price rule. To the knowledge of the authors, this is the first study that uses ACO for assessing strategies of wholesale electricity market.     

Aggregate industrial energy consumer response to wholesale prices in the restructured Texas electricity market

The aggregate response of consumers to wholesale price signals is very limited in the restructured Electric Reliability Council of Texas (ERCOT) market. An overall average own-price elasticity of demand of − 0.000008 for industrial energy consumers served at transmission voltage is estimated using a Symmetric Generalized McFadden cost function model. To date, ERCOT has sought to promote demand response to price signals without reliance on “stand alone” demand response programs, but with a market structure that is designed to facilitate economic demand response. This very limited responsiveness to wholesale price signals may prove problematic in light of policy decisions to pursue an “energy only” resource adequacy mechanism for ERCOT.     

The residential demand for electricity by time-of-use: A survey of twelve experiments with peak load pricing

Information on customer response to time-of-use (TOU) rates plays a major part in utility resource planning, particularly in the design of cost-based rate structures and cost-effective load management programs. Several elasticity concepts have been used by economists to analyze customer response to TOU rates. We discuss the interrelationships between various concepts and show that total uncompensated price elasticities are the appropriate measure of customer response.Evidence from twelve pricing experiments involving about 7000 customers indicates that residential peak-period electricity consumption is generally price-sensitive. TOU rates generally reduce peak period electricity use, as well as daily use. Response is generally higher for high use customers.Short-run own-price elasticities of peak consumption range from nil to ?0.45. Off-peak elasticities lie in a similar range, but are less statistically significant. This wide range indicates that elasticities are not fixed constants but vary parametrically with several conditioning variables such as level of total (daily) electricity use, composition of appliance portfolio and duration of pricing periods. If proper allowance is made for these interactions, it may be possible to transfer elasticities between utility service areas, thus obviating the expensive and time-consuming need for every utility to conduct its own experiment.     

Impacts of day-ahead versus real-time market prices on wholesale electricity demand in Texas

The somewhat recent nodal market structure in Texas impacts wholesale day-ahead market (DAM) and real-time market (RTM) prices. However, comparative insights on consumer responses to both these prices have not received attention. This paper attempts to fill this void by developing a system-wide demand response model to better understand price elasticities under DAM and RTM pricing. These insights may also assist grid operators to develop improved short-term forecasts of electricity demand. Using a large dataset from the Electric Reliability Council of Texas and a hierarchical Bayesian population model, we offer new insights on how DAM and RTM pricing shapes demand for electricity, and the related consequences for maintaining a reliable electricity market.     

Efficient pricing and investment in electricity markets with intermittent resources

Facing growing technological and environmental challenges, the electricity industry needs effective pricing mechanism to promote efficient risk management and investment decisions. In a restructured electricity market with competitive wholesale prices and traditionally regulated retail rates, however, there are technical and institutional barriers that prevent dynamic pricing with price responsive demand. In regions with limited energy storage capacity, intermittent renewable resources present special challenges. This could adversely affect the effectiveness of public policies causing inefficient investments in energy technologies. In this paper, we present an updated economic model of pricing and investment in restructured electricity market and use the model in a simulation study for an initial assessment of renewable energy strategy and alternative pricing mechanisms. A key objective of the study is to shed light on the policy issues so that effective decisions can be made to improve efficiency.     

Customer-economics of residential photovoltaic systems (Part 1): The impact of high renewable energy penetrations on electricity bill savings with net metering

Residential photovoltaic (PV) systems in the US are often compensated at the customer's underlying retail electricity rate through net metering. Given the uncertainty in future retail rates and the inherent links between rates and the customer–economics of behind-the-meter PV, there is growing interest in understanding how potential changes in rates may impact the value of bill savings from PV. In this article, we first use a production cost and capacity expansion model to project California hourly wholesale electricity market prices under two potential electricity market scenarios, including a reference and a 33% renewables scenario. Second, based on the wholesale electricity market prices generated by the model, we develop retail rates (i.e., flat, time-of-use, and real-time pricing) for each future scenario based on standard retail rate design principles. Finally, based on these retail rates, the bill savings from PV is estimated for 226 California residential customers under two types of net metering, for each scenario. We find that high renewable penetrations can drive substantial changes in residential retail rates and that these changes, together with variations in retail rate structures and PV compensation mechanisms, interact to place substantial uncertainty on the future value of bill savings from residential PV.     

Energy efficiency as a resource in the ISO New England Forward Capacity Market

ISO New England, which oversees New England’s bulk electric power system and wholesale electricity markets, recently established a Forward Capacity Market (FCM) that will pay suppliers to ensure sufficient capacity is available to meet future peak loads. Under the FCM, ISO New England projects the needs of the power system 3 years in advance and then holds an annual auction to purchase the resources necessary to satisfy the future regional requirements. This market is groundbreaking in that it was the first to allow energy efficiency and other demand resources to compete directly with generators. In the first auction, held in February 2008, demand resources contributed substantially to eliminating the need for new generating capacity in the near term and to providing low-cost resources to the region’s ratepayers. Two additional successful auctions have now been held. Participating in the FCM requires a considerable and complex bid, financial assurance, and claim activities. Meeting new intensive measurement, tracking, and verification requirements adds new costs. For efficiency portfolio administrators, participation raises policy questions regarding ownership of capacity credits, appropriate disposition of revenues, increasing emphasis on peak savings, and whether traditionally short-term budget cycles should change to enable the longer-term planning necessary to bid resources several years into the future. On the other hand, revenues from the FCM can provide needed funding for additional efficiency investments. This paper describes the FCM, examines the experience and trade-offs involved in participating for efficiency programs, and reviews the benefits of such participation for the program and the region, including the positive value from increased exposure of the part that efficiency can play in our energy mix.     

Extending the bidding format to promote demand response

We propose an extended bidding structure to allow more realistic demand characteristics and behaviors to be expressed via flexible bids. In today's ISO-run energy markets, demand bid formats are all separable over time. However, a significant and growing segment of demand can be shifted across time and therefore has no way to bid its true valuation of consumption. We propose additional bid types that allow deferrable, adjustable and storage-type loads to better express their value, and thus elicit demand response in the most natural way – via direct participation in the market. We show that the additional bid types are easily incorporated into the existing market with no technological barrier and that they preserve the market's efficiency and incentive-compatibility properties. Using real market data, we give a numerical demonstration that the extended bid format could substantially increase social welfare, and also present additional insight on storage expansion scenarios.     

The integration of Price Responsive Demand into Regional Transmission Organization (RTO) wholesale power markets and system operations

A number of states and utilities are pursuing demand response based on dynamic and time-differentiated retail prices and utility investments in Advanced Metering Infrastructure (AMI), often as part of Smart Grid initiatives. These developments could produce large amounts of Price Responsive Demand, demand that predictably responds to changes in wholesale prices. Price Responsive Demand could provide significant reliability and economic benefits. However, existing RTO tariffs present potential barriers to the development of Price Responsive Demand. Effectively integrating Price Responsive Demand into RTO markets and operations will require changes in demand forecasting, scarcity pricing reform, synchronization of scarcity pricing with capacity markets, tracking voluntary hedging by price responsive loads, and a non-discriminatory approach in curtailments in capacity emergencies. The article describes changes in RTO policies and systems needed incorporate Price Responsive Demand.     

Dynamic electricity pricing—Which programs do consumers prefer?

Dynamic pricing is being discussed as one method of demand side management (DSM) which could be crucial for integrating more renewable energy sources into the electricity system. At the same time, there have been very few analyses of consumer preferences in this regard: Which type of pricing program are consumers most likely to choose and why? This paper sheds some light on these issues based on two empirical studies from Germany: (1) A questionnaire study including a conjoint analysis-design and (2) A field experiment with test-residents of a smart home laboratory. The results show that consumers are open to dynamic pricing, but prefer simple programs to complex and highly dynamic ones; smart home technologies including demand automation are seen as a prerequisite for DSM. The study provides some indications that consumers might be more willing to accept more dynamic pricing programs if they have the chance to experience in practice how these can be managed in everyday life. At the same time, the individual and societal advantages of such programs are not obvious to consumers. For this reason, any market roll-out will need to be accompanied by convincing communication and information campaigns to ensure that these advantages are perceived.