Title :
Neural network load forecasting with weather ensemble predictions
Author :
Taylor, James W. ; Buizza, Roberto
Author_Institution :
Said Bus. Sch., Oxford Univ., UK
fDate :
8/1/2002 12:00:00 AM
Abstract :
In recent years, a large amount of literature has evolved on the use of artificial neural networks (ANNs) for electric load forecasting. ANNs are particularly appealing because of their ability to model an unspecified nonlinear relationship between load and weather variables. Weather forecasts are a key input when the ANN is used for forecasting. This paper investigates the use of weather ensemble predictions in the application of ANNs to load forecasting for lead times from one to ten days ahead. A weather ensemble prediction consists of multiple scenarios for a weather variable. We use these scenarios to produce multiple scenarios for load. The results show that the average of the load scenarios is a more accurate load forecast than that produced using traditional weather forecasts. We use the load scenarios to estimate the uncertainty in the ANN load forecast. This compares favorably with estimates based solely on historical load forecast errors.
Keywords :
load forecasting; neural nets; power engineering computing; weather forecasting; England; Wales; artificial neural networks; electric load forecasting; historical load forecast errors; key input; load variables; multiple load scenarios; neural network load forecasting; neural networks; uncertainty estimation; unspecified nonlinear relationship; weather ensemble predictions; weather forecasts; weather variables; Artificial neural networks; Economic forecasting; Linear regression; Load forecasting; Load modeling; Neural networks; Predictive models; Random variables; Uncertainty; Weather forecasting;
Journal_Title :
Power Systems, IEEE Transactions on
DOI :
10.1109/TPWRS.2002.800906