Switching operations of a disconnecting switch (DS) or a circuit breaker (CB) in a gas insulated substation (GIS) can stimulate outside GIS strong radiated electromagnetic fields.In order to assess the health status of GIS through its measured characteristic parameters,a method of extracting the characteristic frequency-band energy of the electric field (E-field) based on wavelet packet analysis is proposed.The radiated E-field signals of CB in good condition and those of CB in which insulation fault happens are separately decomposed into 4 layers by using the bior5.5 wavelet basis function,and the normalized energies of different frequency bands are extracted;then,it's found that the frequency band below 31.25 MHz is the characteristic frequency band of radiated E-field.By comparing the energies in the characteristic frequency bands after a random CB operation with that of the typical radiated E-field obtained when GIS totally is in good condition,insulation faults of CB can be detected and the operating state of GIS could be evaluated.%气体绝缘变电站(Gas Insulated Substation,GIS)内的隔离开关(Disconnecting Switch,DS)或者断路器(Circuit Breaker,CB)的动作会在GIS外部产生强烈的辐射电磁场环境.为了通过此测得的辐射场特征参数评估GIS的健康状态,特提出利用小波包分解提取辐射电场信号特征频段能量的方法.采用biorS.5小波基函数,分别对GIS处于健康状态与GIS中CB内出现绝缘故障两种情况下CB关合动作所产生的辐射电场信号进行4层小波包分解,从而提取不同频段的归一化能量,发现31.25 MHz以下的频段为此GIS因CB动作所产生辐射电场信号的特征频段.经推广,通过对比任意一次CB动作后辐射电场在特征频段的能量与GIS正常运行状态下典型辐射电场的特征频段的能量,可以诊断CB的绝缘故障,评估GIS的健康状态.
展开▼
