A Fluid-Thermal Coupled Model for Estimating Temperature of Powertrain Based on Multi-Condition Experiments

It is known that it is difficult to precisely predict the temperature in motors as various complex factors. For the safety of powertrains in electric vehicles, a fluid-thermal coupled model, based on an optimization method for temperature experimental calibration and verification in multiple operating conditions, is proposed to address the issue of inadequate online temperature monitoring in mass production. The proposed method can fully consider the fluid-thermal coupling between the motor, controller, reducer, and cooling system. By calibrating and correcting heat-flow parameters through a single-objective optimization based on temperature rise curves measured at 8 rated operating points, it is ensured that the error between the predicted and measured temperatures at each operating point is always less than 10%. The maximum calibration error at rated working points is found to be 9.5°C, indicating that this method can effectively capture actual physical thermal parameters of the system. The effectiveness and robustness of this model have been verified through tests involving 2 rated points, 2 peak points, 3 cycles conditions. The results show that under these conditions, the maximum temperature prediction errors are within acceptable limits: 5.8°C for rated working points, 9.0°C for peak working points, and 2.5°C for cycles. This model can be applied for online temperature protection of powertrains in mass production due to its high precision and robustness properties.