Forecasting power market clearing price using neural networks

Deregulation of the electric power industry worldwide raises many challenging issues. Forecasting the hourly market clearing prices (MCPs) in the daily power markets is the most essential task and a basis for any decision making. One approach to predict the market behaviors is to use the historical prices, quantities and other information to forecast the future prices. The basic idea is to use history and other estimated factors in the future to "fit" and "extrapolate" the prices. Aiming at this challenging task, we developed a neural network method to forecast the MCPs for the California day-ahead energy markets. The structure of the neural network we used is a three-layer backpropagation network. The historical MCPs and quantities of California day-ahead energy market, the ISO load forecasts and other public information that may influence the markets are used for training, validating and forecasting test. Preliminary results show that our method is promising.