Short-term Prediction of Photovoltaic Power based on Combination of FCM and CG-DBN

Aiming at nonlinearity, large fluctuation and instability of photovoltaic output power, this paper proposes a combined prediction model based on similar-day clustering and improved deep belief network (DBN) with conjugate gradient (CG). Fuzzy C-means clustering algorithm (FCM) is firstly used to perform similar-day clustering from raw data according to the membership degree. Then the CG-DBN prediction model is built based on the obtained categories to predict the short-term PV output. The scheme is applied to Longyou power station in Zhejiang province. The prediction results are compared with the traditional prediction model and prediction performance of the FCM and CG-DBN combination forecast model is better than other models.