Solar Power Generation Forecast Based on LSTM

Photovoltaic power generation is an effective way to use solar energy, which is a recognized ideal renewable energy source. However, photovoltaic that is susceptible to weather conditions is unstable, and will adversely affect the power grid. Therefore, it is necessary to improve the accuracy of solar power generation. This paper uses the LSTM model to predict solar power generation. At the same time, the data is reduced by using PCA to reduce the training duration of the model and improve the generalization ability of the model. Compared with other models, simulation experiment shows that the LSTM model is better.