D. Maratukulam

ANNSTLF-Artificial Neural Network Short-Term Load Forecaster- generation three

This paper describes the third generation of an hourly short-term load forecasting system known as ANNSTLF (Artificial Neural Network Short-Term Load Forecaster). This forecaster has received wide acceptance by the electric utility industry and is being used by 35 utilities across the US and Canada. The third generation architecture is substantially changed from the previous generation. It includes only two ANN forecasters, one predicts the base load and the other forecasts the change in load. The final forecast is computed by adaptive combination of these two forecasts. The effect of humidity and wind speed are considered through a linear transformation of temperature. A novel weighted interpolation scheme is developed for forecasting of holiday loads, giving improved accuracy. The holiday peak load is first estimated and then the ANNSTLF forecast is re-shaped with the new peak forecast. The performance on data from ten different utilities is reported and compared to the previous generation.

ANNSTLF-a neural-network-based electric load forecasting system

A key component of the daily operation and planning activities of an electric utility is short-term load forecasting, i.e., the prediction of hourly loads (demand) for the next hour to several days out. The accuracy of such forecasts has significant economic impact for the utility. This paper describes a load forecasting system known as ANNSTLF (artificial neural-network short-term load forecaster) which has received wide acceptance by the electric utility industry and presently is being used by 32 utilities across the USA and Canada. ANNSTLF can consider the effect of temperature and relative humidity on the load. Besides its load forecasting engine, ANNSTLF contains forecasters that can generate the hourly temperature and relative humidity forecasts needed by the system. ANNSTLF is based on a multiple ANN strategy that captures various trends in the data. Both the first and the second generation of the load forecasting engine are discussed and compared. The building block of the forecasters is a multilayer perceptron trained with the error backpropagation learning rule. An adaptive scheme is employed to adjust the ANN weights during online forecasting. The forecasting models are site independent and only the number of hidden layer nodes of ANNs need to be adjusted for a new database. The results of testing the system on data from ten different utilities are reported.

Analysis of the value for unit commitment of improved load forecasts

Load forecast errors can yield suboptimal unit commitment decisions. The economic cost of inaccurate forecasts is assessed by a combination of forecast simulation, unit commitment optimization, and economic dispatch modeling for several different generation/load systems. The forecast simulation preserves the error distributions and correlations actually experienced by users of a neural net-based forecasting system. Underforecasts result in purchases of expensive peaking or spot market power; overforecasts inflate start-up and fixed costs because too much capacity is committed. The value of improved accuracy is found to depend on load and generator characteristics; for the systems considered here, a reduction of 1% in mean absolute percentage error (MAPE) decreases variable generation costs by approximately 0.1%-0.3% when MAPE is in the range of 3%-5%. These values are broadly consistent with the results of a survey of 19 utilities, using estimates obtained by simpler methods. A conservative estimate is that a 1% reduction in forecasting error for a 10,000 MW utility can save up to

An artificial neural network hourly temperature forecaster with applications in load forecasting

1.6 million annually.

Artificial neural networks for short-term energy forecasting: Accuracy and economic value

Many power system short-term load forecasting techniques use forecast hourly temperatures in generating a load forecast. Some utility companies, however, do not have access to a weather service that provides these forecasts. To fill this need, a temperature forecaster, based on artificial neural networks, has been developed that predicts hourly temperatures up to seven days ahead. The prediction is based on forecast daily high and low temperatures and other information that would be readily available to any electric utility. The forecaster has been evaluated using data from eight utilities in the USA. The mean absolute error of one day ahead forecasts for these utilities is 1.48/spl deg/F. The forecaster is implemented at several electric utilities and is being used in production environments.

Recent results in secondary voltage control of power systems

Abstract Sixteen electric utilities surveyed state that use of ANNs significantly reduced errors in daily electric load forecasts, while only three found otherwise. Data for five gas utilities reinforces this result: the mean absolute percentage error (MAPE) for ANN daily gas demand forecasts was 6.4%, a 1.9% improvement over previous methods. Yet ANNs were not always best, implying opportunities for further improvement. The economic value of error reduction for electric utilities was assessed by examining operating decisions. For 19 utilities surveyed, an average of

A decomposed state estimation technique suitable for parallel processor implementation

800 000/year per utility is estimated to be saved from use of ANN-based forecasts. Most benefits resulted from improved generating unit scheduling; the utilities estimated such benefits to be up to

Hopf bifurcations in a SMIB power system with SSR

143 annually per peak MW of demand for each 1% improvement in MAPE. This estimated worth of accuracy improvement (roughly 0.1% of annual generation O&M costs) is confirmed by solving generation scheduling and dispatch models under various levels of forecast accuracy.

Advanced static compensator for flexible AC transmission

Recent results are reported on a pilot-point-based secondary voltage control in power systems. Similarities and differences compared to the optimal power flow methodology are described, and theoretical background for secondary voltage controller design is reviewed. Two complementary design problem formulations are reported, together with simulations on the New England 39 bus and Bonneville Power Administration 338 bus power system models. The results show the feasibility of the pilot-point approach and reveal the nature of the tradeoffs involved in secondary voltage controller designs. >

Bulk transmission system loss analysis

The authors present an approach to the linear programming state estimation problem that is suitable for large-scale interconnected power systems. It uses the Dantzig-Wolfe decomposition algorithm. The performance of the method is enhanced by using a multiple column strategy and several other modifications to the original algorithm. Scaling techniques are used to guarantee a reliable solution. Test results on standard test systems are used to illustrate the usefulness of the method and its viability for parallel processor implementation. >