基于支持向量机的感应电机状态监测

State Monitoring of Induction Motor Based on SVM

提出一种基于支持向量机的三相异步电动机状态监测的方法。利用健康电机、轴承故障、定子匝间短路和转子断条的电机在不同负载条件下的各种数据,计算三线电压和电流计算总谐波失真形成特征向量,然后将其用于支持向量机的训练,并针对两个核函数对基于支持向量机的感应电机状态监测的性能进行了评估。仿真结果表明,所提出的方法的故障识别准确率超过98%,验证了其有效性。

A method of state monitoring and fault diagnosis of three-phase asynchronous motor based on support vector machine is proposed.Using various data of healthy motor, bearing fault, stator inter-turn short circuit and rotor broken bar under different load conditions, the three-wire voltage and current are calculated to calculate the total harmonic distortion to form an eigenvector, which is then used for the support vector machine.We train and evaluate the performance of the projective SVM-based scheme against two kernel functions.The simulation results show that the fault identification accuracy of the proposed method exceeds 98%,which verifies its effectiveness.