Research on External Insulation State Detection Method Based on Information Fusion

The external insulation fault of power system will cause huge loss, so it’s very important to accurately judge and predict the external insulation state. The existing detection methods of external insulation state are based on a single signal generated by discharge, which is prone to misjudgment and is not conducive to fault judgment and prediction. In this paper, the partial discharge experiment of contaminated insulator is carried out. Through the collection of the ultraviolet and infrared signals released in the discharge, data transmission and information fusion, it can provide support for the judgement of the state of the external insulation. The adaptive network-based fuzzy inference system (ANFIS) is used for information fusion of two kinds of information to diagnosis insulator running state. The accuracy of the information fusion diagnosis can reach 92%, which meets the requirements of the external insulation detection of the power system.