Analytical Kinetic Model for Accurate Square-slot Stator Natural Frequency Prediction

This paper introduces a novel kinetic modelling technique which accurately predicts the natural frequencies of the square-slot stator and can be used in the early design stage of an electrical vehicle traction motor. The model is an improvement of the classic energy approach, with a small increase in computational cost but a much higher accuracy, especially for high order vibration modes. A three-dimensional (3D) equivalent teeth model based on open cylindrical shell (OCS) theory and the corresponding 3D face coupling method are first proposed. The derivation of the high accuracy model from the classic energy method is thoroughly explained. To cut down the computational cost, a simplification of the proposed kinetic model which utilizes the cyclic symmetry of the stator is then implemented. Further, based on this, the natural frequency of a wounded stator and a real traction motor stator considering water-cooling parts are also investigated. To validate the effectiveness of the proposed model, structural simulations and experiments are carried out. Results of the classic energy method are also provided to highlight the improvement in accuracy.