Prediction of Low Voltage User Electricity Consumption Behavior Based on Conv-Lstm Network Algorithm

The electrification level of residents' lives in China continues to improve, and various types of electrical equipment enrich the lives of Chinese residents, which also increases the demand for electricity and the duration of electricity consumption for residential users. A large number of residential users concentrate on electricity consumption during a certain period of time, which increases the load on the power grid and leads to load shortages, resulting in power outages and restrictions. Therefore, it is necessary to predict and analyze the electricity consumption behavior of low-voltage users. This article mainly uses Conv-LSTM network algorithm to predict and analyze the electricity consumption behavior of low-voltage users. The prediction results of user electricity consumption are obtained through methods such as building prediction models, correcting outliers, and normalizing data.