LSTM load forecasting algorithm based on time-sharing somatosensory

With economic development, people's living standards have gradually improved, and the use of air conditioners and other electrical equipment has increased year by year, which directly leads to the increasing impact of meteorological factors on power load. The temperature adjustment load caused by the extensive use of air conditioners in summer has proved to be closely related to meteorological factors such as temperature, humidity, and wind speed. As a basic model, LSTM can maximize the time-series and non-linear relationship between data when predicting power load data. However, this network tends to ignore the influencing factors that lead to sudden changes in load data. This paper conducts comprehensive modeling analysis based on minute-level meteorological factors, obtains the change value of load through the change of body temperature, and obtains the time series forecast value of load through LSTM, and then obtains the final forecast value. The forecast results show that this method can effectively improve the accuracy of short-term load forecasting and is an effective load forecasting method.