Application of linear lazy learning approach to short-term load forecasting

Many plans on power systems strongly depend on short-term load forecasting. In this paper a novel method based on linear lazy learning approach is proposed for short-term electric load forecasting. The proposed method is successfully verified through PJM market forecasting. The model is trained by the data available for four years and the next two years data is used for validation. The results prove the ability and high-precision of the proposed approach.