A Short-term Photovoltaic Output Prediction Method Based on Improved PSO-RVM Algorithm

摘要

In order to improve the accuracy and efficiency of photovoltaic system output prediction effectively, realize short-term photovoltaic (PV) output forecasting, and to provide auxiliary reference basis for grid dispatching and planning, a short-term PV output prediction method based on improved PSO-RVM algorithm is established. Firstly, the improved fuzzy clustering method is used to obtain the similar day, the wavelet analysis method is used to decompose the similar day PV output by high frequency and low frequency. Then, the improved PSO-RVM algorithm is used to predict the wavelet component of the prediction day respectively. Finally, the predicted results are added together to obtain the predicted values of PV output. The simulation results show that the proposed method can significantly improve the short-term PV output prediction accuracy.