Computationally Efficient Modeling of Electrical Machines With Cooling Jacket

Artikel i vetenskaplig tidskrift, 2019

Modeling of electrical machines is a multiphysics problem. Depending on the phenomena of interest and the computational time constraint, this can be done at different levels of detail. In this article, the main approaches to model the thermal behavior of electrical machines with a liquid cooled casing around the stator (often referred to as cooling jacket) are analyzed and a novel approach is presented. The proposed method aims at creating computationally efficient 3-D multiphysics models of electrical machines with liquid cooled jacket. This model is based on the assumption of a fully developed flow in the cooling jacket which allows to scale the computational fluid dynamics (CFD) simulation to 1-D. The slot with a two layer concentrated winding and potting material is modeled using a composite material comprising of both the conductors and slot filler. Similarly, a unified material is used to model the end-windings. Experimental results on a traction machine for vehicle applications are presented showing good agreement with the simulations. Also, a comparison with a 3-D CFD is presented to verify the pressure drop in the pipe bend. Finally, the model is used to simulate a dynamic load cycle, which would be computationally extremely demanding with combined 3-D CFD and thermal FEA of the machine and its cooling.