Short-term Forecasting of PV Power Based on the Fuzzy Clustering Algorithm and Support Vector Machine in Smart Distribution Planning

This paper presents a photovoltaic (PV) short-term power prediction method based on fuzzy clustering and support vector machines. Using the meteorological information to establish a fuzzy similarity matrix, a set of historical day sample sets most similar to the forecast day is obtained through classification recognition, and the meteorological factors of the prediction date are used as input samples of the prediction model. Thus, the PV power generation prediction model was established. According to the actual measure data, the proposed model is verified. The results show that the method has high prediction accuracy and has better reference value for PV power generation prediction.