A Forecast Method of Icing Flashover Fault Based on Partial Mutual Information Method and Support Vector Machine

摘要

Icing flashover fault of transmission line is an important reason of power grid failure.Current researches mainly focus on the study of the icing flashover characteristics of insulators and determine the evaluation model of flashover voltage with different influencing factors by artificial icing tests on insulators.On one hand,the evaluation model of flashover voltage,which only considers a factor or a few factors,cannot fully reflect the insulator flashover voltage under the combination of all factors.On the other hand,due to the measurement error in the information of the transmission lines,the current characteristics model of icing flashover is difficult to be used to forecast icing flashover fault directly.Taking the development of data mining technology into account,this paper uses support vector machine(SVM)to predict the icing flashover fault.Since the dimension of the input variable has important influence on the extensive ability of SVM model,firstly,this paper uses the method of partial mutual information to select the key factors for input variables.Secondly,SVM forecast model of icing flashover is established to train and predict.Simulation results show that the forecast method based on the partial mutual information method and SVM can predict the icing flashover more effectively,which can be the reference for the ice defense of power grid.