Forecasting electric daily peak load based on local prediction

The forecasting of load demand has become one of the major research fields in electricity market. It has always been the essential part of an efficient power system planning and operation. This paper presents a new approach to a multivariate load forecasting by combining support vector regression with a local prediction framework which employs the correlation dimension and mutual information methods used in time-series analysis for data preprocessing. Local prediction uses only a set of K nearest neighbours in the reconstructed embedded space with considering the more relevant historical instances. The performance of the proposed predication model is evaluated on the data used in the EUNITE competition in 2001. The results show that the proposed method provides a relatively better forecasting performance in comparison with the best result found in the competition and other published papers that uses the same competition's data.