AUTHOR=Huang Ruanming , Wang Xiaohui , Fei Fei , Li Haoen , Wu Enqi 

TITLE=Forecast Method of Distributed Photovoltaic Power Generation Based on EM-WS-CNN Neural Networks

JOURNAL=Frontiers in Energy Research

VOLUME=Volume 10 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2022.902722

DOI=10.3389/fenrg.2022.902722

ISSN=2296-598X

ABSTRACT=In order to cope with the challenges brought by large-scale distributed photovoltaic power generation related to production and consumers to power grid dispatch, a maximum expected sample weighted convolutional neural network (expectations-maximization weighted Samples-convolutional neural network，EM-WS-CNN) is proposed to forecast the distributed photovoltaic output.Firstly, the distance correlation coefficient and principal component analysis method are used to extract comprehensive meteorological factors from the original meteorological data, and then the five statistical indicators of the comprehensive meteorological factors and historical power data are used as clustering features, and the historical data is divided into different groups using the maximum expected clustering. Weather type, weighted training samples based on membership matrix. Finally, the weighted training data is used to construct the EM-WS-CNN model. The above method is compared with the CNN model in the experimental analysis, and the results show that the proposed method has higher accuracy and robustness.