Study on the Prediction of Consumer Electricity Consumption Based on Hybrid Model

Aiming at the characteristic that energy consumption data is a complex nonlinear time series, a hybrid energy consumption prediction model based on Prophet-LSTM is proposed. Firstly, the Prophet model incorporates holiday factors, power cycle models, and trend change functions into the modeling process to improve the robustness of the model and obtain better prediction results. Secondly, a convolutional neural network (CNN) is introduced to extract features of the data to improve the predictive ability of the model on the basis of the long and short-term memory neural network (LSTM). Finally, the initial weights are respectively assigned to the predicted values of the Prophet and LSTM models, and the optimal weight distribution is obtained through grid optimization. Finally, the two models' predicted values are added to get the final predicted value. The simulation results show that the hybrid model based on Prophet-LSTM proposed in this paper can achieve ideal prediction results, and the prediction accuracy of a single model is significantly improved.