基于小波能谱熵的串联型故障电弧检测方法

Series fault arc detection method based on wavelet energy spectrum entropy

低压线路中的串联故障电弧检测多以提取电流信号的故障特征为主,实际中,电流故障特征难以与非线性负载的负荷电流特征进行区分。相比之下,通过识别负载端故障电压特征更易建立统一故障判据。本文通过建立电弧分段仿真模型分析了电弧电阻对负载端电压故障特征的影响,从选择最优小波分解层数与小波基函数出发,提出了一种利用小波能谱熵的电弧故障检测方法,该方法利用故障电弧电压对负载端电压造成的畸变进行故障检测,利用小波能谱熵克服了故障特征频带难以确定的问题。实测及对比实验表明,该方法可有效识别各类负载线路的串联电弧故障,其检测准确率达98%以上。

The arc fault detection of series fault in low voltage line is mainly based on the fault characteristics of current signal.In practice,the current fault characteristics are difficult to distinguish from the load current characteristics of nonlinear loads.By contrast,it is easier to establish a unified fault criterion by identifying the fault voltage characteristics of the load side.In this paper,the influence of arc resistance on the fault characteristics of load voltage is analyzed by establishing the arc subsection simulation model.Starting from selecting the optimal wavelet decomposition level and wavelet basis function,an arc fault detection method based on wavelet energy spectrum entropy is proposed.This method uses the fault arc voltage to detect the distortion caused by the load voltage,and uses the wavelet energy spectrum entropy to overcome the problem that the fault characteristic frequency band is difficult to determine.The measured and comparative tests show that this method can effectively identify the series arc faults of various load lines,and the detection accuracy is more than 98%.