System Driven Thermal Prediction of E-Machine Temperatures

When designing an electric machine or adapting an existing design to a new use case, sizing the thermal solution for expected duty cycles is a major consideration. To define an appropriate duty cycle is often a challenge. In the simplest form, this could be speed and torque demand curves, but in increasingly complex and interconnected systems, we don’t always have direct access to this information. A vehicle designer might define the duty cycle as speed and acceleration requirements, or a wind turbine engineer might define the mechanical loads based on weather conditions. Cooling solutions in larger systems are also often integrated with other sub-systems or components. The impact of these components is difficult to model accurately in isolated thermal simulations of the e-machine.

By tying machine performance, mechanically and thermally, into larger system-level simulations, a machine’s behaviour can be tied to duty cycles defined in other design domains. In this presentation, we will show the configuration of a vehicle system, including a detailed e- machine model and dedicated thermal network, to provide accurate results based on vehicle centric duty cycles. Being able to test thermal behaviour in this environment can dramatically reduce the design cycle time compared to traditional methods such as CFD simulations while maintaining a high level of confidence in the resulting design.

Using this complete system simulation, we can quickly determine the scale of the necessary cooling solution for the e-machine in the vehicle as well as the performance metrics of both the machine and the entire vehicle system. All critical data for a machine designer working to integrate with complex systems.