FKFS Events

2026 Stuttgart International Symposium
on Automotive and Engine Technology

8 - 9 July 2026

Session: E-Powertrain III | | 17:00 - 17:30

Closed-Loop Real-Time Simulation of the Thermal and Electromagnetic Behavior in Electric Drive Trains: Optimizing Model Accuracy for Virtual Prototyping

Fabian Jonczyk, RWTH Aachen Univ.

This article deals with the coupled electromagnetic and thermal simulation of a permanently excited synchronous machine (PSM) in the context of virtual prototyping in a real-time hardware-in-the-loop (HiL) environment. Thermal simulation is often neglected due to the increased complexity of a coupled simulation. This neglect leads to inaccurate simulations and incomplete design optimizations. To ensure the efficiency and reliability of such machines in real operating conditions, a specific understanding of thermal behavior is essential.
 

The objective of this development is to enable a precise and realistic real-time simulation that represents the electromagnetic as well as the thermal behavior. The electromagnetic simulation is executed on a Field-Programmable-Gate-Array (FPGA) and parameterized by Finite Element Method (FEM) simulations. The thermal model is based on a Lumped-Parameter-Thermal-Network (LPTN), which is based on physical laws, geometry parameters and material specifications. The thermal simulation considers iron and copper losses to ensure an accurate representation of the heat sources in the system. Taking these losses into account is essential to correctly model thermal behavior under different load scenarios.
 

The coupled real-time simulation of transient thermal behavior offers a significant advantage for the development of e-machines by taking thermal effects into account at an early stage and testing controllers before a physical prototype is built. This leads to a targeted optimization of the machine design and a reduction in development times, particularly in the development of control units. Validations with real measurement data will further confirm the model quality and increase the efficiency of the development processes.