Renewable Energy Load Prediction Based on Deep Learning Neural Network

Accurate generation and load forecasting is beneficial for power companies to formulate reasonable strategies so as to guarantee power system reliability. This paper proposes a source-load prediction method based on particle swarm optimization (PSO) long and short-term memory (LSTM) neural network. The PSO is used to optimize the hidden layer number, the training times, and the learning rate of the LSTM, to determine the optimal parameter combinations. The proposed PSO-LSTM is also used to establish the wind power prediction model and the customer-side load prediction model, and simulations are conducted to compare and analyze them, and the results verify that the proposed method has higher prediction accuracy than the traditional LSTM, BP, and GA-BP.