A Short-term Load Forecasting Method under Dual Network for Forty-eight moments

Power load forecasting is very important to ensure the whole society's electricity consumption. This paper presents a load forecasting method based on CNN-LSTM at 48 time points. By constantly adjusting the depth and iteration times of CNN and LSTM, the optimal prediction network is obtained. The experimental results show that when CNN depth, LSTM depth and iteration times are respectively 1, 3 and 200, the MAPE value and corresponding Loss value of the prediction model corresponding to 48 time points are 9.35% and 0.05%, respectively.