Prediction of Solar Radiation in Qinghai Lake Area Based on BiLSTM-Attention Method

Abstract

Short-term solar radiation prediction plays a crucial role in production and life. There is much room for improvement in the prediction accuracy and stability of traditional models. In order to solve this problem, this paper uses a method based on deep learning to predict solar radiation. A short-term solar radiation prediction model is established for Qinghai Lake combined with a bidirectional long-term memory network and attention mechanism. Based on the historical solar radiation in the past month and the average temperature at 1.5 meters, the solar radiation prediction in the next two weeks is made prediction. The experimental results show that the prediction model combined with the bidirectional long-term memory network and the attention mechanism is superior to the traditional prediction method in predicting accuracy, convergence speed and root mean square error and average absolute error, which can effectively improve. The accuracy and stability of the short-term solar radiation prediction model in local areas.