Transient Thermal Analysis of an External-Rotor Hub Motor Based on LPTN-RBF-SRSM Coupled Model

The complex geometric structure and transient load characteristics of external-rotor hub motors present significant challenges for heat dissipation. This paper investigates a 75 kW external-rotor hub motor and proposes a transient thermal analysis method that couples a lumped parameter thermal network (LPTN) with a radial basis function-stochastic response surface method (RBF-SRSM). The surrogate RBF-SRSM model is trained on a limited set of CFD simulation data and efficiently predicts the convective heat transfer coefficient (CHTC) under varying inlet temperatures and flow rates, thereby replacing time-consuming CFD calculations. The predicted CHTC is applied as a boundary condition in the LPTN, enabling rapid and accurate temperature prediction of critical components, including windings and stator cores. Experimental results show that the proposed method substantially reduces the reliance on empirical correlations in traditional LPTN approaches, while achieving high prediction accuracy and significantly reducing computation time. This research provides an efficient and accurate solution for the transient thermal analysis of external-rotor hub motors.