A Three-Phase Asynchronous Motor Inter-Turn Short Circuit Fault Diagnosis Method Using Finite Element Model and GWO-ELM

Three-phase asynchronous motors are widely used in modern industry, and their operational state is closely related to the overall system condition. Typically, asynchronous motors operate under normal conditions, making it difficult to obtain fault data, which limits the application of artificial intelligence methods in fault diagnosis. In this paper, a finite element model of a three-phase asynchronous motor is developed to generate sufficient fault data. After validating the model's effectiveness, fault data are generated using the finite element model. To address the negative effects of the random hidden layer parameter values on the fault diagnosis accuracy of the Extreme Learning Machine (ELM), the Grey Wolf Optimization (GWO) algorithm is employed to optimize the initialization parameters of ELM for improving the accuracy of fault diagnosis. The fault diagnosis results on the test set validate the effectiveness of the GWO in optimizing the ELM.