Lattice-Boltzmann based optimization method integrating design of electric machines and gearboxes in electrified power trains for future vehicles
||Lattice-Boltzmann based optimization method integrating design of electric machines and gearboxes in electrified power trains for future vehicles|
||SNIC Medium Compute|
||Huadong Yao <email@example.com>|
||Chalmers tekniska högskola|
||2020-06-28 – 2021-07-01|
||20306 20303 20304|
Electrification of vehicles has a good momentum today. However, energy consumption of electrified vehicles is not very much in focus, instead in-efficient electrified vehicles are appreciated on the cost of more energy efficient ICE vehicles. There is talk about a ‘nordic energy mix’, but still almost every single extra kWh of electric energy put in an electrified vehicle in Scandinavia, must be generated by fossil fuels. This needs to come to an end when electrified vehicles become more and more common. In a recently conducted research project [Vehicle – Energimyndigheten project 41213] it was identified that the gear-box has the same loss as the electrical machine. The gearbox is usually a two-stage gearbox, with a gear ratio of 8-12. If a single stage gear-box could be used, the efficiency would go up, however, then the electric machine becomes larger. No gear at all gives best efficiency, but a very large electrical machine. In principle, 1500 rpm at top speed with a power of 100-150 kW is what is needed.
The purpose of this project is to optimize the electric machine-gear package to find a system solution that has the highest efficiency, but still reasonable size. The hypothesis is that by a careful design of the gearbox, selection of more viscous oil, but still suitable to maintain the life-time of the gear in combination with an adapted machine design will lead to important loss savings in the transmission of an electrified vehicle.
A set of base electrical machines will be identified, focusing on the environmentally sound machines without rare earth metals. For a given requirement (100 kW) veracious electric machines will be deigned and the outcome will be communicated to the Fluid Dynamics division. Based on the feed-back from their side modifications will take place.
The Fluid Dynamics Division will study the efficiencies of the designs from the Electrical Power Engineering Division. The performance of every design (torques, losses and temperatures, etc.) at a set of oil viscosities and volumes within the gearbox will be accessed. The oil type and volume for the best efficiency will be identified. These results will be fed back Electrical Power Engineering Division to down-select the designs and to conduct further optimization. Numerical investigations will be conducted to predict the design performance. The numerical method is the Lattice-Boltzmann method. The numerical tool will be the commercial software OMNIS. Meanwhile, to resolve specific physical mechanisms (e.g., temperature, air bubbles and vibration), another numerical tool based on the open-source code LUMA will be developed in collaboration with Dr. Adrian Harwood and Dr. Alistair Revell at the University of Manchester.