Electric Utility Load Research Articles

A hybrid deep meta-ensemble networks with application in electric utility industry load forecasting

Many industrial applications concern the forecasting of large quantities of related time series. This paper presents a novel hybrid modeling framework, named as Hybrid Deep Meta-Ensemble Networks (HDME-Nets), which combines local and global forecasts using meta-ensemble technology. The proposed framework can be used to generate multiple steps ahead of point and interval forecasts for a large number of time series concurrently. Our proposed framework is composed of four modules: a set of local forecasters that model each time series individually, a global forecaster that captures cross-sectional patterns with data pooling, a feature learner that extracts features from each time series in a supervised fashion, and a meta-combiner that combines the local and global forecasts according to the extracted features. The local forecasters are fitted firstly, and the other three modules are integrated seamlessly into one neural network, which is then jointly trained with a common objective measured with a custom loss function. Through testing two public available electric utility load datasets, we find that the proposed method can achieve improved forecasting performance compared against some state-of-the-art methods.

A Dual-Heat-Pump Residential Heating System for Shaping Electric Utility Load

We present a residential heating system designed specifically to provide electric utilities with a means for achieving intra-hour load shifting. The system couples two heat pumps with a large water tank, which serves as a thermal storage unit. One heat pump uses the water tank as a source and serves residential heating load. The other uses the outdoor air as a source and heats the water tank during favorable weather conditions, thereby improving the efficiency of the first. By aggregating large numbers of such systems distributed throughout its balancing area, a utility may use this resource as a means for achieving load shaping, the intra-hour shifting of generation resources from periods of peak demand to periods of low demand. This paper focuses on the potential for using such an aggregated resource as a technological solution for realizing winter peak demand reduction and overall lower energy consumption, thereby displacing the need for capital investment in peak generation plants. The results of this paper are promising, demonstrating potential savings for both the utility and residential customers.

Assessment of high penetration of solar photovoltaics in Wisconsin

This paper provides an assessment of the large-scale implementation of distributed solar photovoltaics in Wisconsin with regard to its interaction with the utility grid, economics of varying levels of high penetration, and displaced emissions. These assessment factors are quantified using simulations with measured hourly solar radiation and weather data from the National Solar Radiation Database as primary inputs. Hourly utility load data for each electric utility in Wisconsin for a complete year were used in combination with the simulated PV output to quantify the impacts of high penetration of distributed PV on the aggregate Wisconsin electric utility load. As the penetration rate of distributed PV systems increases, both economic and environmental benefits experience diminishing returns. At penetration rates exceeding 15–20% of the aggregate utility load peak, less of the PV-energy is utilized and the contribution of the aggregate electricity generated from PV approaches a practical limit. The limit is not affected by costs, but rather by the time-distribution of available solar radiation and mismatch with the coincidence of aggregate utility electrical loads. The unsubsidized levelized cost of electricity from PV is more than four times greater than the current market price for electricity, based on time-of-use rates, in Wisconsin. At the present time, the investment in solar PV as a cost-effective means to reduce emissions from traditional electricity generation sources is not justified.

Time Series Prediction Using Support Vector Machines: A Survey

Time series prediction techniques have been used in many real-world applications such as financial market prediction, electric utility load forecasting , weather and environmental state prediction, and reliability forecasting. The underlying system models and time series data generating processes are generally complex for these applications and the models for these systems are usually not known a priori. Accurate and unbiased estimation of the time series data produced by these systems cannot always be achieved using well known linear techniques, and thus the estimation process requires more advanced time series prediction algorithms. This paper provides a survey of time series prediction applications using a novel machine learning approach: support vector machines (SVM). The underlying motivation for using SVMs is the ability of this methodology to accurately forecast time series data when the underlying system processes are typically nonlinear, non-stationary and not defined a-priori. SVMs have also been proven to outperform other non-linear techniques including neural-network based non-linear prediction techniques such as multi-layer perceptrons.The ultimate goal is to provide the reader with insight into the applications using SVM for time series prediction, to give a brief tutorial on SVMs for time series prediction, to outline some of the advantages and challenges in using SVMs for time series prediction, and to provide a source for the reader to locate books, technical journals, and other online SVM research resources.

The Development of a Warm-Weather Relative Stress Index for Environmental Applications

The heat stress index (HSI) is a new, comprehensive summer index that evaluates daily relative stress for locations throughout the United States based on deviations from the norm. The index is based on apparent temperature and other derived meteorological variables, including cloud cover, cooling degree-days, and consecutive days of extreme heat. Statistical distributions of meteorological variables are derived for 10-day periods of the annual cycle so that percentile values for each parameter can be determined. The daily percentile values for each variable are then summed, and a statistical distribution is fit to the summed frequencies. The daily HSI value is the percentile associated with the location of the daily summed value under the summation curve. The index is analyzed and spatially verified by comparing intra- and interregional results. Although stations from various climate regions have different criteria defining an excessive heat stress event, neighboring stations typically produce similar HSI results because they are usually affected by the same air mass. To test the effectiveness of the HSI, a relationship between the index results and mortality values is made. Overall, the highest mortality days are associated with the highest HSI values, but high-HSI days are not always associated with high numbers of deaths. A mortality study such as this one is just one of many potential environmental applications of the HSI. Other applications include implementing the index to correlate extreme weather conditions with resource consumption, such as electric-utility load, to determine conditions for which load levels are excessive. The ability to forecast the HSI using a variety of weather forecasting tools has also generated interest within various industries that have a need to issue weather stress advisories, watches, and warnings.

Guest Editorial Everyday Applications Of Neural Networks

EURAL-NETWORK technology has reached a degree of maturity as evidenced by an ever-increasing number of applications. Our experience, however, is that most practitioners of neural networks are familiar with only a handful of cases where neural-network technology has been reduced to practice. The objective of this special issue is presentation of some specific cases of ongoing everyday use of neural networks. Specifically excluded are neural-network applications still in the exploratory stage. While publication of extraordinary exploratory applications papers is within the scope of the IEEE TRANSACTIONS ON NEURAL NETWORKS, this special issue deals only with neural networks used on a regular basis. At minimum, the system must be at the beta test stage. Of the 53 papers received for the special issue, the 14 herein were chosen. In some important cases, papers solicited for submission were unfortunately withheld because developers or licensees wished to not disclose proprietary technology. Nevertheless, the spectrum of the everyday neural-network applications reported herein is a veritable smorgasbord of variety. Applications are reported in telecommunications, control of Publisher Item Identifier S 1045-9227(97)05726-3. steel plants, plasma etching, pattern recognition of cataloged parts, credit card fraud detection, space robot tuning, electric utility load forecasting, railway maintenance, power system security assessment, scanning electron microscope image characterization, cold mill prediction, economic forecasting and, not least, assessment of wine bottle cork quality. This sampling of applications in everyday use is in no way complete. It gives, however, a taste of the impact of neural technology in society. Indeed, impact is the metric by which all technology is ultimately measured.

Load-data access tools cultivate productivity

Electric utility load research involves the systematic collection of customers' energy usage by time-of-day, month, season, and year. At many utilities, whole-premise and end-use load data are collected from a large number of customer sites at 15-minute intervals for multiple seasons and even years. Whole premise refers to the total load at a residence, while end-use load may include several individual major appliances, treated separately. As a result, load research data, especially end-use data, is voluminous. Thus, designing a load data distribution system that can respond to data requests in a timely, effective manner is becoming an important planning and operations goal facing many electric utility load research departments. This article discusses how Arizona Public Service Co. (APS) has implemented a load data distribution and access system to improve its staff productivity and communications efficiency.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Formulation and analysis of a rule-based short-term load forecasting algorithm

The formulation of rules for the rule base and the application of such rules are discussed. The classification of the load forecast parameters into weather-sensitive and nonweather-sensitive categories is described. The rationale underlying the development of rules for both the one-day and seven-day forecast is presented. This exercise leads to the identification and estimation of parameters relating load, weather variables, day types, and seasons. Sample rules that are the product of identifiable statistical relationships and expert knowledge are examined. A self-learning process is described which shows how rules governing the electric utility load can be updated. Results from both the one-day and seven-day forecast algorithms are presented, where the seven-day forecast is generated using both accurate and predicted weather information. The monthly average load forecast errors range between 2.97% and 10.71% for the seven-day forecasts. For the one-day forecasts, the average seasonal errors range between 1.03% and 1.53%. >

A simulation procedure for sample size determination in electric utility load research

Estimating the customer demand for electricity plays a crucial role in the equitable determination of customer electric utility rates. To estimate the customer demand accurately, it is necessary to determine the appropriate sample size. Due to the unique requirements imposed by the Public Utilities Regulatory Policies Act of 1978, existing procedures to determine sample size fail to provide the optimal sample size that will satisfy the re quirements at the minimum cost. The objective of this study is to develop and describe a simulation based procedure that is capable of achieving both criteria. The effectiveness of the pro cedure is tested for selected distributions of electrical demand. The study also identifies other instances where similar sample size determination problems occur and how the newly developed procedure can be used in these situations as well.

The evolution of an expert DSS for electric utility load research

Load research conducted by electric utility companies is an activity embracing the measurement and study of electrical load to understand the trends and behavior of electric utility consumption. The results of load research are major determinants in rate design and capacity planning for the electric utility company. Accurate load research is imperative for a well managed electric utility. The information base used in load research consists of customer billing data and sample data collected under strict guidelines set by the Public Utilities Regulatory Policies Act. Selection of appropriate sampling procedures is a central function of load research. This paper describes the evolution of an integrated expert decision support system (XDSS) for load research sample design. It is an exploratory system that has a potentially large impact on the profitability of electric utility companies. It was the result of a consulting project begun 4 years ago with a medium sized utility company. Much of the underlying statistical sampling methodology in the XDSS had to be newly derived to meet the specific needs of sampling in load research.

An interactive decision support system for designing appropriate and adaptive sampling procedures in electric utility load research

Confidence interval estimation of customer demand for electricity plays a vital role in the capacity and financial planning of electric utility companies. Inaccurate or inadequate estimation could severely affect the economic efficiency of these companies. Selecting the appropriate sampling procedure is, in turn, critical to ensuring success in the accurate estimation of electrical demand. Because of the nature and diversity of demand for electricity, however, classical sampling procedures have to be extensively modified. This paper describes a decision support system (DSS) that aids electric utility companies in selecting and designing appropriate sampling plans. The DSS adapts sophisticated statistical techniques to address all aspects of the sampling procedure.

Analysis of interrelationships between photovoltaic power and battery storage for electric utility load management

The impact of photovoltaic power generation on an electric utility's load shape under supply-side peak load management conditions is explored. Results show that some utilities utilizing battery storage for peak load shaving might benefit from use of photovoltaic (PV) power, the extent of its usefulness being dependent on the specific load shapes as well as the photovoltaic array orientations. Typical utility load shapes both in the eastern (at Rayleigh, NC) and in the western (at Hesperia, CA) parts of the USA are examined for this purpose. It is concluded that while photovoltaic power generation seems to present a bigger impact on the load of the western utility, both utilities will experience considerable savings on the size of the battery system required to shave the peak loads as well as in the night-time base capacity required to charge the battery.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Coil Protection for a Utility Scale Superconducting Magnetic Energy Storage Plant

Superconducting Magnetic Energy Storage (SMES) is proposed for electric utility load leveling. Attractive costs, high diurnal energy efficiency (> 92%), and rapid response are advantages relative to other energy storage technologies. Recent industry-led efforts have produced a conceptual design for a 5000 MWh/1000 MW energy storage plant which is technically feasible at commercially attractive estimated costs. The SMES plant design includes a protection system which prevents damage to the magnetic coil if events require a rapid discharge of stored energy. This paper describes the design and operation of the coil protection system, which is primarily passive and uses the thermal capacity of the coil itself to absorb the stored electromagnetic energy.

Electric utility energy conservation and load management programmes: R & D opportunities

During recent years, electric utilities have begun a variety of new programmes to manage their future load growth. In many cases, these conservation and load management (C/LM) programmes serve as alternatives to traditional electricity supply resources. Additional development of planning, analysis and evaluation methods is needed before utilities can fully rely on these C/LM programmes as legitimate electric system resources. This paper discusses several important research topics. These topics include better use of data from evaluations of C/LM programmes to improve demand forecasting models, development of models to predict participation in future utility C/LM programmes, review of existing integrated resource planning models, assessment of alternative regulatory treatment of C/LM programmes by state public utility commissions, and identification of better ways to disseminate results on effective C/LM programmes among utilities.

Design advances in superconducting magnetic energy storage for electric utility load leveling

Superconducting Magnetic Energy Storage (SMES) has been under study for electric utility energy storage application since the early 1970's. Major challenges have included structural design to accommodate large biaxial loads over a long lifetime, design of a conductor to carry high current, design of a reliable and passive coil protection system, design for conventional fabrication and construction methods, and system cost reduction. A SMES design concept, developed by the authors of this paper, includes technically feasible solutions to all the items listed above. Plans for full development and commercialization of this technology are being prepared.

A Feasible Utility Scale Superconducting Magnetic Energy Storage Plant

This paper presents the latest design features and estimated costs of a 5000 MWh/1000 MW Superconducting Magnetic Energy Storage (SMES) plant. SMES is proposed as a commercially viable technology for electric utility load leveling. The primary advantage of SMES over other electrical energy storage technologies is its high net roundtrip efficiency. Other features include rapid availability and low maintenance and operating costs. Economic comparisons are made with other energy storage options and with combustion turbines.

Principles for Effective Load Management

This paper discusses several of the issues of concern in achieving effective electric utility load management. Included are a discussion of the distinction between 'conservation' and 'load management', definition of 'controllable load', and a discussion of several criteria necessary to achieve significant changes in system load patterns. Examples are presented of load management applied to typical utility customers.

Two-Way Utility System Communications Using AM Broadcast Radio

An electric utility load management system functions optimally in direct proportion to the reliability of the communications system employed to transmit signals to and from the utility's central control to the thousands or millions of receivers installed on customers' appliances (water heaters, air conditioners, pumps, etc) throughout the utility's service territory. The signal from the central control to a receiver on an appliance 100 miles away on the other side of a mountain range must be as readily detected as the signal to a receiver across the street from the central control.

Advanced lead-acid batteries for electric utility load leveling

Advanced lead-acid batteries for electric utility load leveling

Future outlook for lead/acid batteries in Japan

Pollution problems, coupled with the oil crisis which erupted in the 197Os, have highlighted a need for the development of technologies for creating, saving, and storing energies throughout the world. Various efforts have been, and are being, made to meet this need. Among these efforts is the worldwide work devoted to the development of advanced batteries primarily for use in electric vehicles and utility load levelling. Despite this intense work, it will still be years before some of these advanced batteries become commercially available. Thus, lead/acid batteries, which have been ranked at the top in every yearly battery production since their invention by Gaston Plant6 in 1860, are likely to maintain this position for years to come. This is primarily due to the fact that the lead/acid battery (a) is a reliable power source which supplies power of good quality, (b) is normally operated at room temperature ensuring easy handling and safe operation, and (c) is chiefly made from lead which can easily be recycled for repetitive use. This article reviews the present status of lead/acid batteries in Japan, and presents a future outlook for them.