Short-term Load Forecasting Based on Modal Decomposition and Self Attention Mechanism
Short-term power load forecasting plays an important role in the safe operation of power system.The fluctuation and non-stationarity of power load data has always been a difficulty in load forecasting,and the forecasting effect of directly building a forecasting model is poor.To this end,a power load prediction method based on Adaptive Noise Complete Ensemble Empirical Mode Decomposition(CEEMDAN)and Bidirectional Long Short-Term Memory Network(BiLSTM)combined with Self-Attention Mechanism(SAM)is proposed.Firstly,CEEMDAN algorithm is used to decompose the power load data into multiple eigen modal components to reduce the volatility of the original load data;Then,the SAM-BILSTM net-work prediction model is constructed for each load component;Finally,the power load forecasting results are obtained by superposition and reconstruction of the component forecasting results.The experimental results show that CEEMDAN-SAM BiLSTM algorithm improves the forecasting accuracy by 2.78％,1.99％,1.28％and 0.96％respectively compared with SVR,DNN,LSTM and BiLSTM,and effectively im-proves the load forecasting accuracy.
万方数据
维普
