GIS Insulation State Evaluation Based on Multi-source Information Fusion

摘要

With the wide application of Gas Insulated Switchgear (GIS) in the field of ultra-high voltage, the identification and evaluation of insulation state in GIS equipment has been a hot topic in this field, Therefore, how to effectively integrate the multi-source information representing GIS insulation state and establish a scientific state evaluation model is a major problem to be solved. In this paper, the insulation degradation process of typical insulation defect inside GIS were simulated, and ultra-high frequency (UHF) and ultrasonic information of defect in different degradation stages were obtained. At the same time, the genetic algorithm is used to further characterize the PD severity assessment feature, and the PD assessment index set based on UHF and ultrasonic information was constructed. Based on this, the fuzzy hierarchy analysis was used as the evaluation method framework, the factor analysis weighting method and fuzzy evaluation function based on mutual information was used to obtain the preliminary evaluation result. Finally, the preliminary evaluation result was used to form a complete comprehensive evaluation model. The evaluation scheme constructed in this paper can effectively complement the multi-source information, thereby improving the reliability of the GIS insulation state evaluation result.