In order to solve the safety hazards that may arise from the continuous abnormal heating of photovoltaic panels,as well as the deviation problem in the existing situation perception,a method for identifying the abnormal heating situation of photovoltaic panels is proposed.Integrating temperature data points from multiple time periods using sliding windows,constructing a time series photovoltaic cell temperature data matrix,and analyzing the thermal coupling mapping relationship between time data and temperature data.Based on the mapping relationship constraint machine learning algorithm,calculate the abnormal heating score,and determine the abnormal situation by taking the abnormal score value.The experiment shows that the trend display of photovoltaic cell group map-ping data is accurate,there are no omissions,and the error is low.
关键词
时间序列/映射感知/光伏电池组/升温异常态势/热耦合关系/自动化识别
Key words
time series/mapping perception/photovoltaic battery pack/abnormal warming trend/thermal coupling relation-ship/automated identification
维普
万方数据