MOO for Surrogate Model Based Traction Inverter Systems

Beschreibung

In power electronics, especially in the area of sustainable mobility through electric vehicles (EVs), the aim is to
have power-efficient and compact power electronic systems. Electric vehicles commonly use traction inverters which are
generally highly complex systems with a large number of parameters. In the context of Multi-Objective Optimization (MOO)
it is indispensable to have surrogate models which are cheap to evaluate, because sufficiently many Spice or 3D system
level simulation are most likely not feasible. The aim of this work is to demonstrate the capabilities of Surrogate-Based
Optimization for complex traction inverter systems having a large number of design parameters.

Zeitraum	17 Sep. 2024
Bekanntheitsgrad	International