High-fidelity Data Generation for the Motor Drive System Design

Ryoko IMAMURA
Simulation Platform Specialist
Powersys

概要

In the era of machine learning (ML) and data-driven design, engineers must develop the ability to orchestrate simulations across multiple dimensions, including physical fidelity, computational speed, and large-scale data handling. In this presentation, we will explore common problem sets in motor-inverter drive systems, such as loss prediction incorporating PWM effects, trade-offs in modulation techniques, and bearing current estimation. We will highlight how the interaction between motors and inverters adds both complexity and insight to simulation challenges. We will then explore how JMAG and power electronics simulation tools enable engineers to tackle these complex problems by scaling solvers and leveraging advanced computing environments. Finally, we will illustrate how JMAG supports engineers in developing workflow pipelines for high-fidelity physics data generation to achieve successful motor drive system design.