Hybrid DWT-DE-RNN Power Load Forecasting for Industrial Users
The short-term prediction of power load is of great significance for the planning and development of the power industry.With the reform and development of the electricity market,short-term forecasting of power load is extremely important for indus-trial manufacturing enterprises to effectively reduce energy costs.However,the actual load sequence data exhibits multiple com-plex properties,such as nonlinearity,non-stationarity,and time variation.A three-level hybrid ensemble short-term load fore-casting method consisting of discrete wavelet transform(DWT),differential evolution algorithm(DE)and radial basis function neural network(RBFNN)is proposed.DWT is used to decompose load data to obtain good power consumption characteristics;DE is used to obtain the best tunable parameters required for RBFNN prediction.The mixed integration method(DWT-DE-RBFNN)was evaluated using the 2001 load data of PJM public data set and the 2015 annual data of an industrial park in Liaoning Prov-ince.The DWT-DE-RBFNN method was compared with three other mainstream coupling methods(RBFNN,BPNN,SaDE-ELM).Statistical analysis shows that the proposed method shows better prediction accuracy on the three standard scales of MAPE,MAD and RMSE,which shows the advanced nature of the method.
万方数据
维普
