Short-Term Electricity Consumption Forecasting Based On Long Short-Term Memory Network

High-precision electricity consumption forecasting can not only ensure the safe and reliable operation of the power system, but also bring considerable economic benefits. Aiming at the contradiction between the demand for high-precision electricity consumption forecasting and the insufficiency of traditional forecasting methods, this paper uses a short-term electricity consumption forecasting model based on long-term short-term memory network (LSTM) to forecast daily electricity consumption on an hourly scale in a Wanda shopping mall in Guangdong. First, analyze the characteristics of the total electricity consumption of the mall to obtain its dominant influence factors. Secondly, the LSTM-based short-term electricity consumption prediction model is used to predict the total electricity consumption of the mall on an hourly scale. At the same time, in order to improve the accuracy of the prediction, the comfortable temperature range and time characteristics are modified in the model. Finally, in order to qualitatively and quantitatively evaluate the effectiveness of the model in this paper, the classic BP and Elman models are selected as comparison benchmarks for comparison experiments. The experimental results show that the prediction model used in this paper has higher prediction accuracy than traditional prediction models.