An intelligent fault diagnosis method for electrical equipment using infrared images

Abstract

Infrared thermography(IRT) plays a very important role in monitoring and inspecting thermal defects of electrical equipment without shutting down, which has an important significance for the stability of power systems. Manual analysis of infrared images for detecting defects and classifying the status of equipment may take a lot of time and efforts, and may also lead to incorrect diagnosis results. To surmount the lack of manual analysis of infrared images, intelligent fault diagnosis methods for electrical equipment are proposed recently, but there are two difficulties when using these methods: one is to find the region of interest, another is to extract features which can represent the condition of electrical equipment, as it is difficult to segment infrared images due to the over-centralized distribution and low intensity contrast of infrared images, which is quite different from that of visual light images. To overcome these two difficulties, a novel intelligent fault diagnosis approach for electrical equipment is presented in this paper. Firstly the infrared image of electric equipment is clustered into five regions using K-means algorithm, then statistical characteristics in each region is extracted, and all these characteristics of five regions are combined as the image features. These features are subsequently input to a support vector machine(SVM) which is utilized as an intelligent diagnosis system. To reinforce the SVM classification performance, a parameter optimization approach is adopted. The experimental results show the efficiency of our proposed method.