基于非线性信号的光伏组件表面清洁度识别技术

Surface Cleanliness Identification of Photovoltaic Modules Based on Non-Linear Signals

光伏组件表面清洁度分析过程中,容易受到非线性信号影响,导致识别结果不精准;为了解决这个问题,提出了光伏组件表面清洁度非线性自回归识别技术;分析脏污、热斑效应对光伏组件发电量影响,获取光伏组件表面时程响应非线性信号;模拟光伏组件表面的时域非线性不清洁问题,分析非线性信号单元,从时程响应中提取相应的非线性特征,消减环境不确定因素的干扰;通过线性函数过滤时程响应线性信号,计算概率化条件方差,确定非线性自回归识别指标;构建非线性自回归脏污、热斑效应识别结构,通过非线性自回归I-V曲线识别脏污,利用非线性自回归损失函数识别热斑效应;由实验结果可知,使用所研究技术识别的脏污I-V特性显示,当电压为0时,短路电流为0.41 A,当电流为0时,开路电压为19.5 V;识别的热斑效应I-V特性显示,与正常组件相比,受热斑效应影响的开路电压、短路电流都有所下降,最大开路电压分别为100、80和55 V,与实际数据一致,具有精准识别效果。

The process of analyzing the surface cleanliness of photovoltaic modules is easily affected by nonlinear signals,leading to inaccurate identification results.In order to solve this problem,a nonlinear autoregressive identification technique for PV module surface cleanliness is proposed.The influence of dirt and hot spot effect on the power generation of PV modules is analyzed to obtain the time-domain response nonlinear signals on the surface of PV modules.Simulate the time-domain nonlinear unclean liness problem on the surface of PV modules,analyze the nonlinear signal units,extract the corresponding nonlinear features from the time-range response,and abate the interference of environmental uncertainties.Filter the time-range response linear signal by linear function,calculate probabilistic conditional variance,and determine the nonlinear autoregressive identification index.The nonlinear autoregressive dirty and hot spot effect recognition structure is constructed,and the dirty is recognized by the nonlinear autoregressive I-V curve,and the hot spot effect is recognized using the nonlinear autoregressive loss function.As can be seen from the experimental results,the dirty I-V characteristics identified using the studied technique show that when the voltage is 0,the short-circuit current is 0.41 A,and when the current is 0,the open-circuit voltage is 19.5 V;the identified hot spot effect I-V characteristics show that compared with the normal components,the open-circuit voltage and the short-circuit current affected by the hot spot effect have decreased,and the maximum open-circuit voltages are respectively 100,80 and 55 V,which is consistent with the actual data and has an accurate recognition effect.