Short-term power load forecasting based on CEEMDAN-CNN-LSTM hybrid modeling

In the analysis of power supply and demand situation, load forecasting is the basis for the safe and stable operation of power system. In this paper, a CEEMDAN-CNN-LSTM short-term power load forecasting method is proposed. Firstly, CEEMDAN is used to decompose the original power load signal and construct a filtering algorithm to solve the optimal noise reduction smoothing model. Then, each IMF component is input into the CNN-LSTM model for prediction, and the final load prediction value is obtained by cumulative summation of the predicted values. Taking the Belgian data as an example, the CEEMDAN-CNN-LSTM model is compared with other prediction models. The results show that the CEEMDAN-CNN-LSTM model has a lower error within 24 hours, indicating that the method has a higher prediction accuracy.