For a reduction of greenhouse gas emissions from vehicles on a longer term, combustion car engines will be replaced by electrical drive systems. Due to the limited capacities of the energy storage units, highly efficient use of the electrical energy is required. High efficiencies are reached for low machine temperatures, which are achieved in the compact systems by powerful convective liquid cooling systems. The design of the complex cooling channels requires the reliable and accurate computation of momentum and heat transport. For this purpose, Large-Eddy Simulation (LES) is a particularly powerful, ever more improving numerical simulation tool, provided that the fraction of computationally costly, directly resolved eddy structures is not too big and the effect from the unresolved fine structure may be modelled with sufficient accuracy. The models established today were developed and validated for flow conditions quite remote from the present applications in electro-mobility. In particular the Reynolds analogy between momentum and heat transport, which is the basis of many models to date, is not applicable to the present coolants with high molecular Prandtl numbers. For reasons of numerical stability, model parameters must furthermore be averaged along statistically homogeneous directions, which is impossible in most complicated flow geometries. The present project will therefore develop a model accounting for the specific electro-mobility related requirements from the geometric complexity, the coolant material properties, and the thermal properties of the wall materials. The model development bases on the direct numerical simulation (DNS) of generic test cases for a comprehensive and detailed analysis of the fine structure in momentum and heat transport, and on a priori as well as a posteriori LES for the model validation and improvement. The simulation results achieved by DNS and LES will be validated by experimental data acquired in the frame of the project using a heated pipe test rig for relevant operation conditions. Finally, the validated model will be implemented in an industrial code for numerical flow simulation of the industrial partner and tested for realistic electro-mobile applications. Important validation criteria, further to the predictive accuracy, are the computational efficiency and the numerical stability of the model.
|Effective start/end date||1/12/17 → 30/11/20|