DSD’s novel application of MPS-based software solution speeds up thermal modeling of e-machines

Powertrain engineering consultancy Drive System Design (DSD) has presented a paper detailing a novel technique to improve the thermal modeling of e-machines at this month’s SIA Powertrain Conference in Rouen, France (May 16-17).

The paper details how credible results for heat transfer can be obtained in just one week by using a combination of particle-based fluid modeling and steady state finite-element (FE) thermal modeling.

This new approach is based on the moving particle simulation (MPS) method and is the result of a collaboration between DSD and simulation-based engineering science consultant EnginSoft. It allows design iterations to be compared during the concept stage of a new electric machine, enabling enhanced optimization for e-machines and integrated EV powertrains.

Established methods for fluid flow modeling, such as computational fluid dynamics (CFD), take very long to provide results, which is a problem during the rapid iteration of a concept design, especially where bulk heat dissipation has to be taken into account, such as with an oil-cooled e-machine.

In collaboration with EnginSoft, DSD has reduced the time by using Particleworks, an advanced software solution based on MPS to generate a heat transfer coefficient (HTC) map for the complex, multi-phase flow, then applied it to an FE model of the e-machine.

“It can take as little as a few days using modest computer hardware to simulate several seconds of real-time data, compared to weeks if using finite volume CFD,” said Luca Martinelli, design engineer, DSD.

“By combining the performance benefits of MPS with appropriate modeling simplifications, it is possible to use this method as an iterative design tool, rather than just for final validation.”

The paper, jointly authored by DSD and EnginSoft, describes the growing importance of optimum e-machine cooling as power density increases and the e-machine is integrated into the powertrain.

It then explains the thermal modeling approach and motor cooling strategy before presenting the results as predicted temperature maps for different coolant flow rates. It concludes with an assessment of the benefits and proposals for future work.

Entitled Thermal Optimisation of e-drives Using Moving Particle Semi-implicit (MPS) Method, the paper was presented as part of the Powertrain Efficiency Improvement session at 15:30 on May 16, the first day of the conference.