Thermal Modeling with Surrogate Model-Based Optimization of Direct Oil Cooling Heat Transfer Coefficient for HEV Motor

Abstract

The traction motor in parallel 2 hybrid electric vehicles adopts a cooling method involving automatic transmission fluid (ATF). The scattered ATF directly cools the heat sources of the electric motor. A thermal model based on the lumped parameter network is developed to reflect the direct oil cooling effect. In the developed oil-cooled lumped parameter thermal network (LPTN), the cooling by the ATF acts as a current source. The oil scattered through the motor shaft has high cooling efficiency, but it is difficult to theoretically calculate the nonlinear fluid behavior. Because the flow rate and rotational speed affect the ATF cooling performance, the convective heat transfer coefficient (HTC) of the ATF is reflected in the direct oil cooling thermal model of the traction motor via the proposed correlation process. In this process, a kriging surrogate model-based optimization is performed to determine the convective HTC of the ATF under specific flow and load conditions. The objective function of the optimization is to minimize the difference between the temperature predicted by the thermal model and experimentally measured temperature. As the kriging surrogate model has high prediction accuracy for nonlinear performance, the direct oil-cooled LPTN developed during the optimization process provides a useful thermal model.