*3.1. Support Vector Machine*

State diagnosis of main transformers, As the case of a typical nonlinear classification problem, the overall plan of SVM is the first use of a nonlinear transform the input space data is mapped to a high-dimensional feature vector space, and then in the feature space of the optimal separating hyperplane is constructed, linear classification, after the last map back to the original space Became a nonlinear classification of input space [13].

SVM settings

At present, the commonly used kernel functions are mainly polynomial kernel function, radial basis (RBF) kernel function, hyperbolic tangent (sigmoid) kernel function, and so on. This paper mainly uses the RBF kernel function to the apply to SVM model.
