Power System Load Forecasting Model Based on WT-CNN-LSTM Hybrid Neural Network
As the proportion of electricity in China's energy continues to increase,electricity forecasting plays an irreplaceable role in modern energy management.Due to the diversification of the electric power structure and the complexity of influencing factors,the traditional prediction model has limitations in electric load forecasting.This paper combines the wavelet transform WT and neural network CNN-LSTM,and applies the WT-CNN-LSTM hybrid neural network to power system load forecasting,and conducts comparative experiments with the traditional machine learning model and time series forecasting model.The results show that the WT-CNN-LSTM neural network has a higher accuracy in electric load forecasting,and can provide reference basis for the operation and planning of the power system.
Power System Load ForecastingCNN-LSTM Hybrid Neural NetworkWavelet VariationBig Data
万方数据
维普
