Study on Application of Support Vector Machine in Engine Fault Diagnosis

In view of the superiority of support vector machine and the characteristics of the fault engine.A novel fault diagnosis method based on support vector machine is presented in this paper.The method is special designed for small number sample set,and can obtain commendable generalization ability.Firstly feature information is extracted via information combination method.Then support vector machine is adopted to realize pattern recognition and correlation.Finally on the basis of this,the fault diagnosis model is established through the multi-class SVM.The method ensures the higher accuracy in the diagnosis.The results are satisfactory and it is proved that this method is effective and commendable.