Combinational Forecast for Transformer Faults Based on Support Vector Machine

The forecast to dissolved gases in a transformer is useful to the forecast of the transformer’s faults.A combinational model and relative solving steps for power transformer fault forecast are proposed on the basis of support vector machine(SVM) theory.During the process of the forecast,firstly several single forecast approaches,such as linear model,exponent model,exponentiation model,non-equal-gap grey predictive model and non-equal-gap grey verhulst predictive model,are used to form a model group,and a set of data in time sequence on each dissolved gas are fitted by the model group.Secondly,the fitted results by different traditional predictive models in time sequence act as the input of the support vector machine regression(SVMR) model,and the changeable weights of the input quantities,which are the important part of the combinational SVMR prediction model,can be obtained by relative SVMR approach based on known input and output samples.In the paper,the procedure of the combinational prediction on transformer faults based on SVMR is discussed in details.Two examples on dissolved gas data of transformers in time sequence have proven that the proposed model can give good results on both the fitting to the known data in time sequence and the extrapolation to the data to be predicted.Therefore,the problem of good fitting and bad extrapolation in traditional predictive approaches is solved to some extent.Moreover,compared with other predictive approaches,both single model and other combinational model,the proposed combinational forecast model has higher forecast accuracy.