Calibration of thermal models for electric traction drives using optimization algorithms

Project goal

The use of electric traction drives in electric vehicles requires a high overload capacity. This means that the drives are sometimes operated far outside their nominal range, which is why thermal protection of all thermally critical components is mandatory. The achievable performance of electric drives depends directly on the accuracy of the thermal models used for thermal protection.

However, the parameterization of the thermal models turns out to be a challenging task:

  •     Due to the high utilization of traction drives, the losses and thermal networks of the components to be protected are often highly nonlinear.
  •     The losses are usually not directly measurable, for example rotor losses or losses in the DC link capacitor.
  •     At one operating point, different parameter sets can usually lead to the same temperature curves.

In the planned development project, the calibration of thermal models for traction drives is to be investigated using optimization algorithms. The use of optimization algorithms is known from the field of combustion engines and from the design of electrical machines. In the calibration of thermal models for traction drives, however, such algorithms have been used only sporadically and have not yet been investigated in general.

The East Bavarian University of Applied Sciences Amberg-Weiden is working on the research project as part of a cooperative doctorate with the Technical University of Darmstadt.

Project duration

September 2022 to August 2026 (48 months)