DCAMM seminar: Topology and Shape Optimization for Electrical Machines

Topology and Shape Optimization for Electrical Machines considering EMAG and Structural Modeling

Speaker: Claus B.W. Pedersen, Technology Director in SIMULIA – Multiphysics/Structures Science & Technology (Dassault Systèmes)
Co-authors: Yi Zhou, Stefan Reitzinger and Marcel Hoffarth

Date: Monday 30 November 2026
Time: 10:30–11:30
Place: Room Ohm (Ø29-608-3) and Zoom

For those in Sønderborg, you can register for Zoom participation here.

Abstract

The present optimization workflow considers initially topology optimization for the conceptual designing of E-machines followed up by a subsequential shape optimization for the fine-tuning optimization of the E-machines where both optimization approaches simultaneously consider electromagnetic and structural properties. The optimization workflow also considers several operation points for the E-machine where an operating point for the electrical machine is defined by the current magnitude and corresponding load angle. Additionally, each load angle for the injected current at a given operation point is also a primary design variable modified in each optimization iteration.

Concurrently, the electromagnetic design responses and the mechanical design responses are applied as objective functions and/or as constraints. The most common electromagnetic design responses are the average torque, torque rippling, coil voltages and spatial harmonics of the lumped tooth forces including radial and tangential components whereas the mechanical design responses are usual stiffness represented by displacements and strength using stress criteria. The iterative optimization workflow applies mathematical programming and the adjoint sensitivities of both the EMAG and structural design responses. The elemental relative densities represent the topology whereas the nodal positions of the finite elements represent the non-parametric shape. Both the topology optimization and non-parametric shape optimization apply a regularization for imposing a length-scale into the optimization formulation.

We will show topology optimization and non-parametric shape optimization for a PMSM (Permanent Magnet Synchronous Machine) to illustrate that the present workflow can address the numerous design requirements.