Multiobjective design optimization and Pareto front analysis of a radial eddy current coupler

Article type: Research Article

Authors: Canova, Aldo^a | Freschi, Fabio^{a; *}

Affiliations: [a] Politecnico di Torino, Dipartimento di Ingegneria Elettrica, corso Duca degli Abruzzi, 24 – 10129 Torino, Italy

Correspondence: [*] Corresponding author. E-mail: fabio.freschi@polito.it

Abstract: The paper proposes a multiobjective optimization to design a radial eddy current coupler, where the torque is maximized while the momentum of inertia is minimized to improve the dynamic behavior of the device. An analytical solution of the electromagnetic problem is used to speed up the computation of objectives and an optimization algorithm based on the emulation of the immune system behavior, referred to as Vector Immune System, is adopted to find tradeoff solutions between objectives. The solutions found so far are compared with solutions obtained by a Genetic Algorithm with fuzzy scalarization of objectives. It shows a faster convergence with a better spread of trade-off solutions available for a higher level decision making. A deep analysis of these Pareto optimal configurations is also provided, trying to discover common properties of optimal solutions.

Keywords: Multiobjective optimization, artificial immune systems, analytical models, eddy current devices

DOI: 10.3233/JAE-2010-1079

Journal: International Journal of Applied Electromagnetics and Mechanics, vol. 32, no. 4, pp. 219-236, 2010