Real-time gas insulation state evaluation model of the gas insulated switchgear

Gas insulated switchgear (GIS) has been applied greatly in the ultra-high voltage (UHV) power grids because of its advantages of high reliability, compact structure, small footprint, low electromagnetic pollution and so on. Therefore, the real-time monitoring of its operation status is important for ensuring the reliable and safe operation of the power grids. In this paper, we propose a novel model of real-time gas insulation state evaluation for the GIS, presenting the distributed iterative self-organizing data analysis techniques algorithm (ISODATA) clustering algorithm and distributed fuzzy K-Nearest neighbor (FKNN) classification algorithm which are implemented on the Spark cluster platform. Experimental results show that the proposed approach has a much better efficiency and accuracy than the FKNN algorithm in the single machine environment.