Fiber detecting of high voltage insulator contamination grades based on PSO-SVM

A novel method that integrates fiber technology with support vector machine classifiers to detect the contamination grades of high voltage insulators is presented i n this paper. Based on laboratory simulation experiments of the contaminated silex sensor and insulator, under condition of the complicated nonlinear relationship between the luminous flux attenuation, the contamination grades of insulator, the environment humidity and ash density, the least squares support vector machine (LSSVM) pattern recognition model of detection of the contamination grades is constructed by means of particle swarm optimization (PSO) arithmetic to optimize the parameters of the model. The method takes advantages of the minimum structure risk of SVM and the quickly globally optimizing ability of particle swarm, and the mapping relation between the luminous flux attenuation, the environment humidity, ash density and contamination grades is built quickly by learning from sample data. Then the contamination grade of insulator online detecting system is developed based on the fiber technology. And the application effect proved the feasibility of the method.