Forecasting System Based On Wavelet Transform And Pso-Svm

A new method for load forecasting based on LS-SVM, PSO and Wavelet Transform is proposed. The wavelet transform is adopted to decompose the historical data, so the approximate part and several detail parts are obtained. The results of Wavelet Transform are predicted by a separate LS-SVM predictor. PSO is employed to determine these parameters of SVM model. The novel forecast model integrates the advantage of WT, PSO and LS-SVM. Compared with other predictors, this forecast model has greater generality ability and higher accuracy.