Short-term Electricity Load Prediction Based on Machine Learning and Improved LSTM Model

With the continuous development of computers, artificial intelligence is gradually integrated into our lives, and electricity is one of the most important energy sources in today's era. Power load related problems also lead to a lot of power loss, using machine learning to provide reliable power load prediction is of great significance. In this paper, LSTM (Long Short Term Memory Model) is used to forecast power load. The data comes from parts of Linyi City, combined with weather factors such as temperature and sunshine and the daily electricity consumption of continuously developing new energy vehicles to predict the total electricity consumption. LSTM model is used for power load prediction, and the learning rate and iteration times of the model are optimized and adjusted. The short-term power load prediction algorithm of LSTM model with higher accuracy and stronger applicability and stability is obtained by fully considering many influencing factors such as climate and electricity consumption of new energy vehicles that have an impact on power load.