光伏系统直流串联电弧故障检测研究

搭建了光伏系统电弧故障实验平台,采集系统运行时的电流作为电弧检测的依据,利用时频域分析相结合的方法对电弧故障进行检测。分析发现系统正常运行和发生电弧故障时的电流信号在时域上的平均值变化特征明显,而在频域上利用小波对电弧电流及正常电流做5层分解,利用d5频段能量可有效对正常、阴影遮挡及故障运行状态进行区分。最后,采用将时频域特征结合利用阈值对电弧故障进行二次判断的方法对电弧故障进行检测,检测结果证明了该方法的有效性与可行性。

Research on Photovoltaic System Series DC Arc Fault Detection

Through the building photovoltaic system arc fault experiment platform, the current collection system runtime as the basis of arc detection, using time and frequency domain analysis method to detect arc fault.Analysis found that the system run normally and arc fault occurs on the current signal in time domain of the average change significantly, and in the frequency domain by using the wavelet decomposition of arc current and the normal current 5 layer, using the d5 spectrum energy can be effective to normal, shadows shade and failure to distinguish the running state, the finally adopt combines the time-frequency domain features using the threshold for secondary arc fault detection method to detect arc fault, test results show that the method is effective and feasible.