Application and research of short-term Power load forecasting based on LSTM

With the continuous development of the power industry, the number and types of power data became more and more large. As an important part of power grid planning, the improvement of the accuracy of short-term power load forecasting is of great significance for the reliable and efficient operation of the system. According to the nonlinear characteristics of short-term load forecasting, a forecasting method based on long-term memory neural network (LSTM) is proposed. Through the construction of LSTM model, the actual power load data of a city in China is selected for experiment, and the final results were obtained. Compared with the experimental results of Elman, BP and LSSVM, it is proved that LSTM has better prediction accuracy and smaller error, which proves the validity and feasibility of this method.