摘要
提出准确预测光伏组件现场性能退化率的分区预测方法。首先系统识别和选取影响光伏组件性能退化的综合性因素,再基于加速寿命试验确定各综合性影响因素对光组件性能退化的影响权重,进而构建光伏组件现场性能退化率的预测模型;其次综合考虑性能退化影响因素的影响权重和时间权重,进行性能退化影响因素的地理区域聚类,得到性能退化程度最具一致性的地理区域划分结果,从而实现光伏组件性能退化率的分区预测。广东、海南两省所处地理区域类别的光伏组件现场性能退化率预测结果表明,预测值与实际值的相对误差为6.67%,能满足工程应用的要求。
This study proposes a zoning prediction method for accurately evaluating the field performance degradation rate of PV modules.After systematically identifying and selecting the comprehensive factors affecting the performance degradation of PV modules,the weights of each factor on the performance degradation are determined based on the accelerated life test,and then a prediction model of the field performance degradation rate of photovoltaic modules is constructed.With the geographic clustering of the factors considering those influence weights and time weights,the geographical area divisions with the most consistent performance degradation degree are obtained.Therefore,the field degradation rate prediction can be performed based on the clustering regional categories.The prediction results of the field performance degradation rate of PV modules in the geographical region of Guangdong and Hainan provinces show that the relative error between the predicted value and the actual value is 6.67%.
作者
熊玉辉
孙帅帅
李少帅
徐锛
刘卫东
Xiong Yuhui;Sun Shuaishuai;Li Shaoshuai;Xu Ben;Liu Weidong(Advanced Manufacturing School,Nanchang University,Nanchang 330036,China;PowerChina Jiangxi Electric Power Engineering Co.,Ltd.,Nanchang 330046,China;GoerTek,Qingdao 266104,China)
出处
《太阳能学报》
EI
CAS
CSCD
北大核心
2023年第10期182-190,共9页
Acta Energiae Solaris Sinica
基金
国家自然科学基金面上项目(72071099)。
关键词
光伏组件
退化
现场寿命
预测
聚类算法
PV modules
degradation
field life
forecasting
clustering algorithms
