Application of clustering analysis in the prediction of photovoltaic power generation based on neural network

2017 International Conference on New Energy and Future Energy System

Cheng, K.^1 ; Guo, L.M.^1 ; Wang, Y.K.^1 ; Zafar, M.T.^1

School of Power and Energy, North-western Polytechnical University, Xi'an, Shaanxi province, 710072, China^1

关键词: BP neural networks;    Clustering analysis;    Forecasting modeling;    Historical data;    Number of datum;    Photovoltaic power generation;    Prediction model;    PV power generation;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/93/1/012024/pdf
DOI  :  10.1088/1755-1315/93/1/012024

来源: IOP

【 摘 要 】

In order to select effective samples in the large number of data of PV power generation years and improve the accuracy of PV power generation forecasting model, this paper studies the application of clustering analysis in this field and establishes forecasting model based on neural network. Based on three different types of weather on sunny, cloudy and rainy days, this research screens samples of historical data by the clustering analysis method. After screening, it establishes BP neural network prediction models using screened data as training data. Then, compare the six types of photovoltaic power generation prediction models before and after the data screening. Results show that the prediction model combining with clustering analysis and BP neural networks is an effective method to improve the precision of photovoltaic power generation.