Two-dimensional steady-state thermal analytical model of permanent-magnet synchronous machines operating in generator mode

Purpose Thermal analysis of electrical machines is usually performed by using numerical methods or lumped parameter thermal networks depending on the desired accuracy. The analytical prediction of temperature distribution based on the formal resolution of thermal partial differential equations (PDEs) by the harmonic modeling technique (or the Fourier method) is uncommon in electrical machines. Therefore, this paper aims to present a two-dimensional (2D) analytical model of steady-state temperature distribution for permanent-magnet (PM) synchronous machines (PMSM) operating in generator mode. Design/methodology/approach The proposed model is based on the multi-layer models with the convolution theorem (i.e. Cauchy's product theorem) by using complex Fourier's series and the separation of variables method. This technique takes into the different thermal conductivities of the machine parts. The heat sources are determined by calculating the different power losses in the PMSM with the finite-element method (FEM). Findings To validate the proposed analytical model, the analytical results are compared with those obtained by thermal FEM. The comparisons show good results of the proposed model. Originality/value A new 2D analytical model based on the PDE in steady-state for full prediction of temperature distribution in the PMSM takes into account the heat transfer by conduction, convection and radiation.

Automated summary

The paper presents a 2D analytical model for predicting steady-state temperature distribution in permanent-magnet synchronous machines, considering heat transfer through conduction, convection, and radiation. The model is based on complex Fourier series and the separation of variables method, taking into account different thermal conductivities of machine parts and determining heat sources through power losses calculated with FEM. Validation of the model shows good results compared to thermal FEM analysis.