A Least Squares Support Vector Machines (LS-SVM) approach for predicting critical flashover voltage of polluted insulators

Least Squares Support Vector Machines (LS-SVM) are a class of kernel machines emphasizing on primal-dual aspects in a constrained optimization framework. LS-SVMs aim at extending methodologies typical of classical support vector machines for problems beyond classification and regression. This paper describes a methodology that was developed for the prediction of the critical flashover voltage of polluted insulators by using a Least Squares Support Vector Machines (LS-SVM). The methodology uses as input variables characteristics of the insulator such as diameter, height, creepage distance, form factor and equivalent salt deposit density. The estimation of flashover performance of polluted insulators is based on field experience and laboratory tests are invaluable as they significantly reduce the time and labour involved in insulators design and selection. The majority of the variables to be predicted are dependent upon several independent variables. The results from this work are useful to predict the contamination severity, critical flashover voltage as a function of contamination severity, arc length, and especially to predict the flashover voltage. The validity of the approach was examined by testing several insulators with different geometries. Moreover the performance of the proposed approach with other intelligence method based on ANN is compared. It can be concluded that the LS-SVM approach has better generalization ability that assist the measurement and monitoring of contamination severity, flashover voltage and leakage current.