FAULT DIAGNOSIS METHOD OF ROTATING MACHINERY BASED ON VOLTERRA SERIES AND SVM

A new fault diagnosis method of rotating machinery based on Volterra series and support vector machine(SVM) is proposed.In the proposed method,the Volterra kernels are identified in the four conditions,i.e.normal,rotor crack,rotor rub,and pedestal looseness,by the quantum particle swarm optimization(QPSO) algorithm.Then the first order Volterra kernels and front three order Volterra kernels are respectively input into the SVM classifier for training.The experiment result shows that the proposed method is effective.When the type of fault is hardly distinguished with the first order Volterra kernels,the higher-order Volterra kernels can be used for classification.The proposed method has obvious predominance in the fault diagnosis of rotating machine.