Sand dust influence degree analysis method based on real-time monitoring of photovoltaic panels

Wang Ningbo,Ma Ming,Cao Yinli,Shen Runjie,He Bin,Han Xushan,Ma Yanhong,Li Xiaohu,Zhang Peng,Han Zifen,Zhang Jianmei,Zhou Qiang,Zhao Long,Wang Mingsong,Lyu Qingquan,Wang Dingmei,Chen Zhao,Zhang Yanli,Zhang Ruixiao,Zhou Shiyuan,Zhang Jinping,Huang Rong,Li Jin,Zhang Zhenzhen,Gao Pengfei
2018-01-01
Abstract:The invention discloses a sand dust influence degree analysis method based on real-time monitoring of photovoltaic panels. The method comprises: defining influence coefficients on the power generationefficiency of photovoltaic panels by different sand dust accumulation under the same light intensity; on the basis of a comparison experiment, calibrating respective power generation efficiencies ofthe photovoltaic panels with different accumulated sand dust amounts and then acquiring evaluation indexes corresponding to sand dust layers with different thicknesses; calibrating power generation efficiency coefficients of the photovoltaic panels with different accumulated sand dust amounts; acquiring small data sets of about thousands of pictures with calibration information; acquiring an imageclassifier by using a Keras-based deep learning framework; and shooting lots of pictures of actual photovoltaic panels within different periods of time and acquiring real-time sand dust influence degree indexes of photovoltaic panels at all positions in a photovoltaic electric field by combining position information. With the disclosed method, the evaluation index of the sand dust influence degree can be monitored in real time; and the method can be applied to the image classification model.
What problem does this paper attempt to address?