Short-Term Load Forecast Method Based on à Trous Wavelet Transform and Multiple Kernels SVM
A method for short-term load forecast based on à Trous wavelet transform and multiple kernels SVM is presented in this paper.The à Trous wavelet transform is applied to discompose the load time series into approximate components and detailed components; the multiple kernels SVM is chosen to forecast the discomposed data.The multi-scale load forecasting results are obtained by synthetizing the forecasted data.The model is applied to 1-step and 2-step forecasting of actual load data.The experimental results show that the maximum error of RMSE is 1.82.In comparison with the standard BP neural network,the proposed method has higher prediction accuracy and better generalization ability.
NETL
NSTL
万方数据
维普
