基于GS-SVM的光伏阵列积灰程度评估方法研究  

EVALUATION METHOD OF ASH ACCUMULATION DEGREE OF PHOTOVOLTAIC ARRAY BASED ON GS-SVM

作　　者：张逴 李树成 魏东 刘素民 易建波[2] Zhang Chuo;Li Shucheng;Wei Dong;Liu Sumin;Yi Jianbo(Sichuan Shengtian New Energy Development Co.,Ltd.,Chengdu 610213,China;Key Laboratory of Wide-area Measurement and Control on Power System of Sichuan Province(University of Electronic Science and Technology of China),Chengdu 611731,China)

机构地区：[1]四川晟天新能源发展有限公司,成都610213 [2]电力系统广域测量与控制四川省重点实验室(电子科技大学),成都611731

出　　处：《太阳能学报》2024年第11期220-226,共7页Acta Energiae Solaris Sinica

基　　金：四川省科技厅重点研发计划(2021YFG0098);国家重点研发计划(2018YFB0905000)。

摘　　要：提出一种基于网格搜索(GS)和支持向量机(SVM)的阵列积灰程度评估方法,有效解决光伏阵列表面积灰与光伏出力之间的潜在经济性损益问题。通过对光伏阵列积灰以及混有其他故障的输出特性进行分析,揭示短路电流这一电气参数能反映光伏阵列积灰情况且不易受其他故障干扰。提出以短路电流、太阳辐照度以及温度3个参数作为输入特征量,含GS超参数优化技术的SVM积灰程度评估模型。经过实验测试分析,证明该方法准确率高于DecisionTree、GS-DecisionTree以及XGBoost评估方法,且应用中对数据采集量和训练样本量的需求低,易于推广应用。A method for evaluating the degree of array ash accumulation based on grid search(GS)and support vector machine(SVM)is proposed to effectively solve the potential economic gain/loss problem between the surface ash accumulation of PV arrays and the PV output power.By analyzing the output characteristics of PV arrays with ash accumulation and mixed with other faults,it is revealed that the electrical parameter of short-circuit current is able to reflect the ash accumulation of PV arrays and is not easily disturbed by other faults.An SVM model for assessing the degree of ash accumulation is proposed using three parameters,namely,short-circuit current,solar irradiance,and temperature,as input feature quantities and incorporating GS hyperparametric optimization techniques.After the experimental test and analysis,it is proved that the accuracy of this method is higher than DecisionTree,GS-DecisionTree,and XGBoost evaluation methods,and the application is easy to be popularized because of the low demand for the amount of data collection and the amount of training samples.

关 键 词：光伏组件 短路电流 网格搜索 支持向量机 积灰评估 

分 类 号：TM715[电气工程—电力系统及自动化]