Partial discharge diagnosis method using electromagnetic wave mode transformation in gas insulated switchgear

2007 
Though the gas insulated switchgear (GIS) insulation diagnosis method using ultra high frequency (UHF) band signal is widely used as a method to detect partial discharge (PD) sensitively, diagnosis methods for the electromagnetic wave mode propagation behavior in GIS tanks have not been investigated enough. Especially, diagnosis methods based on the electromagnetic wave distribution of the circumferential direction have been rarely discussed. In the preceding studies by the authors comparing investigations were made by experiments and analyses on the electromagnetic field distribution patterns of the GIS circumferential direction in front of or behind the L-shape or T-shape structures. It was found that the electromagnetic field distributions that are characteristic of mode transformations are seen when the electromagnetic waves of the PD signals are propagated through the L-shape and/or T-shape structures in the GISs. In this study, the diagnosis algorithm is constructed by extracting the characteristics in the output of the high order mode electromagnetic waves transmitted through the L-shape part and/or T-shape part. If the output distributions of the signals detected before and/or after the transmitting have those characteristics, the signals can be determined to be the ones propagated inside and the information on the circumferential direction position of the signal source can be obtained. A simple diagnosis method with high precision has been developed as a GIS-PD diagnosis technology in the field based on the diagnosis algorithm as above. The diagnoses in the field can be classified into the first order diagnosis for high sensitivity detection and the second order diagnosis for high precision judgment by using the algorithm to do more reliable evaluation.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    7
    References
    33
    Citations
    NaN
    KQI
    []