Open AccessPrediction of photovoltaic power generation based on Bayesian neural network with grey correlation
Author(s) -
Jiaxiong Zhu,
Zhigang Xiao,
Chengjian Feng
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2005/1/012069
Subject(s) - photovoltaic system , randomness , computer science , robustness (evolution) , artificial neural network , bayesian probability , grid , data mining , algorithm , artificial intelligence , mathematics , statistics , engineering , biochemistry , chemistry , geometry , electrical engineering , gene
Photovoltaic power generation depends on weather conditions, which has the characteristics of randomness, volatility and intermittence. In order to ensure the security and stability of the grid connected process and make full use of the advantages of photovoltaic power generation, this paper proposes Bayesian optimization algorithm based on grey correlation to predict the power of photovoltaic power generation. The gray correlation selects the similar correlation data from the original data set, and normalizes the data sample set to eliminate the influence of data size caused by different data dimensions. Bayesian optimization algorithm can effectively prevent over fitting by adding regularization term and super parameter α, β. Experiments show that the photovoltaic power prediction method based on Bayesian optimization algorithm under grey correlation has higher accuracy and robustness, and can provide more comprehensive information for power grid dispatching and control.