利用物理和数据驱动的光伏性能退化建模方法  

Physics-based data-driven modeling method for power generation degradation of photovoltaic power plants

作　　者：王宇钖 陈志聪 吴丽君 俞金玲 程树英 林培杰 WANG Yuyang;CHEN Zhicong;WU Lijun;YU Jinling;CHENG Shuying;LIN Peijie(Institute of Micro-Nano Devices and Solar Cells,College of Physics and Information Engineering,Fuzhou University,Fuzhou,Fujian 350108,China)

机构地区：[1]福州大学物理与信息工程学院,微纳器件与太阳能电池研究所,福建福州350108

出　　处：《福州大学学报（自然科学版）》2024年第5期513-519,共7页Journal of Fuzhou University(Natural Science Edition)

基　　金：国家自然科学基金资助项目(62271151);福建省自然科学基金资助项目(2021J01580);福建省科技厅引导性基金资助项目(2022H0008)。

摘　　要：为提高户外光伏电站现场退化评估的准确性和可靠性,提出一种物理和数据驱动的光伏组件性能退化模型.研究户外光伏组件受静态温度、循环温度、相对湿度和紫外线影响的特性,并综合动态应力函数,利用累积损失模型对多应力下光伏电站性能退化进行建模.此外,退化模型的未知参数通过遗传算法来提取.使用美国国家太阳辐射数据库的长期数据对该模型进行训练和测试.将性能退化实际值和模型计算值进行对比,结果表明,本研究所提出模型的相对误差更低,验证了该方法的可行性.In order to improve the accuracy and reliability of evaluating the on-site degradation of outdoor photovoltaic power plants,a physics-and data-driven performance degradation model for photovoltaic modules is proposed in this paper.By studying the characteristics of outdoor photovoltaic modules affected by static temperature,cycle temperature,relative humidity and ultraviolet rays,we synthesized the dynamic stress function and used the cumulative loss model to model the performance degradation of photovoltaic power plants under multiple stresses,and used the genetic algorithm to extract unknown parameters of the model.The model has been trained and tested using long-term data from the US National Solar Radiation Database.The comparison results between the actual performance degradation and the calculated value of the model show that the model proposed in this paper has a lower relative error,which proves the feasibility of the proposed method.

关 键 词：光伏电站 光伏退化 数据驱动 优化算法 

分 类 号：TM615[电气工程—电力系统及自动化] TP181[自动化与计算机技术—控制理论与控制工程]