不同生弧原因下串联型故障电弧实验研究

Experimental Study on Series Fault Arc Under Different Arcs

串联型故障电弧是供电系统的常见故障,会严重影响供电可靠性,甚至引发供电安全事故。在搭建低压串联型故障电弧实验平台的基础上,以三相异步电机和三相变频器为实验负载,针对接触松动和机械振动两种原因引起的串联型故障电弧开展实验研究。利用集合经验模态分解EEMD(ensemble empirical mode decomposition)方法将实验电流信号分解为若干本征模态函数IMF (intrinsic modal function);选取前4阶IMF组成初始向量矩阵,对其进行奇异值分解SVD(singular value decomposition);构建以奇异值平均值为特征量的特征向量,从而获得了不同生弧方式下串联型故障电弧电流信号在奇异值大小上的变化规律。研究结果表明,对不同生弧原因引起的故障电流信号,将经EEMD分解得到的前4阶IMF分量的奇异值作为串联型电弧故障的特征量是可行的。

Series fault arc is a common fault in a power supply system,which will seriously affect the reliability of power supply and even lead to security incidents.Based on a low-voltage series fault arc test platform,a series fault arc experiment was carried out on two types of series fault arc resulting from loose contact and mechanical vibration,respectively,where a three-phase asynchronous motor and a three-phase inverter were used as experimental load.The ensemble empirical mode decomposition(EEMD)method was used to decompose the experimental current signal into several intrinsic modal functions(IMFs),and the first four orders of IMFs were chosen to form an initial vector matrix,which was further decomposed by singular value decomposition(SVD).Afterwards,the eigenvector with the average singular value as the characteristic quantity was constructed,thereby obtaining the variation rule for the series fault arc current signal caused by different arcs against the singular value.Results show that the singular value of the first four orders of IMF components obtained by using EEMD to decompose the fault current signal caused by different arcs is feasible as the characteristic quantity of series fault arc.