GIS在线检测中局部放电和外部干扰信号的识别

Distinguish between partial discharge and external interference signal for GIS on-line monitoring

采用小波时频分析的方法对发生在气体绝缘金属开关设备(GIS)内部局部放电而激发的电磁波和外界的电晕干扰进行了分析,讨论了区分和识别外部和内部放电及外界干扰的可能性。在单一信号存在的情况下,发生在GIS变电所的电晕干扰一般在130MHz以下,频宽小于70MHz,而内部放电激发的电磁波信号无论何种缺陷以及类型如何,其频宽至少是电晕干扰的2倍以上,所以,从信号的带宽上可以判别是内部放电还是外部干扰。当电晕和放电信号共存时,可以通过信号的带宽和信号到达的前沿综合判断。结果表明,该方法简单有效,可以满足在线检测中信号识别的要求。

The electromagnetic wave excitated by partial discharge inside CIS（Gas-Insulated Switchgear） and external corona- generated interference are analyzed using the wavelet transform in time -frequency- domain. The feasibility of distinguishing between internal discharge and external interference is discussed. When only single signal exists, the frequency of corona- generated interference in substation is commonly under 130MHz with less than 70MHz bandwidth. However, the bandwidth of whatever electro-magnetic wave excited by partial discharge inside GIS is at least double times of that of corona- generated interference. When both signals exist,they can be synthetically distinguished by band- width and advancing front. The result shows that the method is easy and effective,meeting the requirements of signal distinguish for on-line monitoring.