Prediction of photovoltaic power generation probability based on gaussian mixture model
Conference: ECITech 2022 - The 2022 International Conference on Electrical, Control and Information Technology
03/25/2022 - 03/27/2022 at Kunming, China
Proceedings: ECITech 2022
Pages: 6Language: englishTyp: PDF
Authors:
Pan, Zhiyuan; Liu, Chaonan; Zhao, Yishu; Wang, Jing; Liu, Jing (State Grid of China Technology College, Jinan, Shandong, China)
Wang, Wei (Shandong University of Science and Technology College of Intelligent Equipment, Tai’an, Shandong, China)
Wang, Zhiling (State Grid Chongqing Electric Power Company urban power supply Branch, Chongqing, China)
Abstract:
In view of the strong randomness and uncertainty of photovoltaic output affected by the environment, this paper proposes a method for predicting photovoltaic output based on a Gaussian mixture model. Based on the factors that affect photovoltaic output, clustering algorithm is used to classify historical data, the improved BP neural network is used to predict the classified pv data. From the prediction error, the Gaussian mixture model is established by the EM method, and finally the probability model of photovoltaic output is obtained. The simulation results show that the PV output prediction performance of the proposed method is better than a single Gaussian distribution, which proves the effectiveness of the proposed method.