An Attempt to Forecast Price Spikes in Electric Power Markets

Price forecasting has become one of the main focus of electric power market research efforts. Existing price forecasting methods have shown a good ability to forecast normal electricity prices. However, none of the research efforts have demonstrated to be effective in forecasting price spikes in electric power markets. This paper presents a methodology to forecast price and spikes using system identification approach and radial basis functions neural networks. The results obtained using PJM interconnection data with mean absolute percentage error (MAPE) of less than 5% demonstrate the feasibility and accuracy of the proposed method.