Short-term power load forecasting based on EEMD-SE and RBF optimized by genetic algorithm

Abstract

In order to improve the forecasting accuracy of short-term power load，a power load forecasting approach based on ensemble empirical mode decomposition(EEMD)-sample entropy(SE) and genetic algorithm(GA) for RBF neural network optimization is proposed in the paper. The EEMD decomposition method is used to decompose the load sequence adaptively, and according to sample entropy to combine the subsequences with similar complexity, which effectively reduces the scale of operation. Based on the difference of sub-sequence complexity, the corresponding RBF neural network model is constructed. The genetic algorithm is used to avoid the neural network falling into the local optimal and convergence problem. Then the new sub-sequences are forecasted and superimposed to obtain the final forecasting result. The simulated results show that the prediction algorithm has a satisfactory prediction effect and meets the requirements of short-term power load forecasting.

Keywords

• load forecasting
• ensemble empirical mode decomposition
• genetic algorithm
• neural network
• sample entropy