Short-term power load forecasting based on empirical mode decomposition and ANN

In order to predict effectively the short-term power load, which is inherently non-stationary and has a certain periodicity and randomness by itself, a novel approach for short-term load forecasting is presented based on empirical mode decomposition (EMD) and artificial neural network (ANN). The load sequence is decomposed into some intrinsic mode functions (IMF), and the local characteristic information of load can be reflected by them respectively. The regularity of the IMFs is analyzed to explore the degree of different factor influence on the IMFs, and the characteristics of load are concluded. Then these IMFs are forecasted by appropriate artificial neural networks.After the synthesis of the forecasted results from the IMFs, the final forecasting result of the load sequence is obtained.The simulation results show that the proposed method possesses higher precision and better adaptability than the traditional BP neural network method.