基于决策树的光伏组件故障诊断方法研究

摘要

The output characteristics of PV modules when they are in abnormal degradation or partial shading are analyzed . A novel online fault diagnosis method for PV modules with abnormal degradation or partial shading based on decision tree ( DT) is proposed. The change of fill factor ( FF) , the slope factor ( K) and the ratio of output current (Im/Isc) under the two different faults are analyzed. The voltage of maximum power point (Um), the cur-rent of maximum power point (Im), open circuit voltage (Uoc), short circuit current (Isc), FF, K and Im/Isc are provided as the attribute collection which can be used by the DT algorithm. When the attribute data needed is ac-quired, the DT algorithm can determine the PV modules' state. The experimental results show the feasibility and effectiveness of the proposed method.%分析了光伏组件在局部阴影或异常老化状态下的输出特性,提出了一种基于决策树算法的光伏组件在线诊断局部阴影或异常老化的判断方法.同时分析了在这两种状态下填充因子FF、斜率因子K和输出电流比Im/Isc的变化规律,结合光伏组件的四个输出参数(最大功率点电压Um和电流Im、开路电压Uoc和短路电流Is)一起作为属性集合,用于提供给决策树生成算法自由选择合适的属性生成故障诊断决策树.实际应用中,只要获得需要的属性数据即可通过生成的决策树诊断出光伏组件的工作状态.实验结果证明了该方法的可行性和有效性.