Condition assessment strategies of composite insulator based on statistic methods

Composite insulator has significant advantages over porcelain and glass insulator for its light weight and anti-pollution flashover performance and has been widely used in overhead power transmission line systems. These advantages would become more remarkable at higher voltage level. In China, silicone composite insulator has been generally regarded as the first choice in UHV and UHVDC overhead transmission systems. However, as a polymer material, the aging threat of silicone rubber has always been a major concern in the application of composite insulators. It is not just important but also urgent to develop specific strategies to evaluate the operation condition of composite insulators and corresponding maintenance methods. In this paper, 65 composite insulator specimens have been collected and studied. A series of tests and measurements have been conducted and all test results have been gathered and filtered based on statistic methods to form the final condition assessment strategies. In this procedure, Fisher Discriminant analysis, Bayes Discriminant analysis and binary-logistic regression method have been applied to establish the relations between the operation condition of composite insulators and the measured data. A three-group and a two-group classification strategy has been developed in this paper to assess the operation condition of insulators.