van Luenen, Jelte William (Eindhoven University of Technology), Borsboom, Olaf (Eindhoven University of Technology), Bayazit, Göksenin Hande (Eindhoven University of Technology), Ilhan Caarls, Esin (Eindhoven University of Technology), Lomonova, Elena A. (Eindhoven Univ of Technology), Salazar, Mauro (Eindhoven University of Technology)

A Bayesian Optimization Framework for Electric Machine Design in Powertrains

Scheduled for presentation during the Regular Session "Modelling and optimal design of electric powertrains" (WeB1), Wednesday, June 18, 2025, 14:20−14:40, Kapel

AAC 2025 11th IFAC International Symposium on Advances in Automotive Control, June 15-18, 2025, Eindhoven, Netherlands

This information is tentative and subject to change. Compiled on January 21, 2026

Abstract

This paper presents a framework for the design optimization of electric vehicle powertrains utilizing physics-based models to estimate electric machine performance. Specifically, the framework employs analytical and high-fidelity finite element models to parameterize the electric motor. This allows for detailed analysis of the geometry. The effect of the rotor geometry parameters is studied by comparing two optimized powertrains, one including powertrain components and external machine dimensions, and one additionally including internal machine parameters in its decision variables. By including the magnet size as a decision variable, the optimizer reduces the energy consumption by $0.3$% over the WLTP cycle, but with a significantly longer machine and a smaller diameter, emphasizing that including internal geometry parameters on high-level powertrain optimization is impactful.