Study on the fault of power-shift steering transmission based on SVM structural risk

Support vector machine (SVM) is an efficient method for data mining of oil analysis. The principle and structural risk of SVM are described in this paper. And the structural risk is studied using oil analysis data. During the process, parameters determination is a very important part because parameters have great influence on the performance of SVM. We select the Radial Basis Function (RBF) as the kernel function of SVM and study the influence of parameters s and C for SVM structural risk. The recognition rate of SVM model is influenced by SVM structural risk. The recognition rate is studied through an experimentation research.