A fuzzy-neural approach to electricity load and spot price forecasting in a deregulated electricity market

Abstract

Accurate short term load forecasting is crucial to the efficient and economic operation of modem electrical power systems. With the recent effort by many governments in the development of open and deregulated power markets, research in forecasting methods is getting renewed attention. Although long term and short term electric load forecasting has been of interest to the practicing engineers and researchers for many years, spot-price prediction is a relatively new research area. This paper examines the use of a neural-fuzzy inference method for the prediction of 24 hourly load and spot price for the next day. Publicly available data of the electricity market of the state of New South Wales, Australia is used in a case study.