Higher-order multiobjective design of MEMS

Issue title: Special Supplement Issue: Selected papers from the 13th International Workshop on Optimization and Inverse Problems in Electromagnetism (OIPE 2014)

Guest editors: Domenico Lahaye

Article type: Research Article

Authors: Di Barba, Paolo^{a; *} | Savini, Antonio^a | Wiak, Sƚawomir^b

Affiliations: [a] Department of Electrical, Computer and Biomedical Engineering, University of Pavia, Pavia, Italy | [b] Institute of Mechatronics and Information Systems, Ƚódź University of Technology, Ƚódź, Poland

Correspondence: [*] Corresponding author: Paolo Di Barba, Department of Electrical, Computer and Biomedical Engineering, University of Pavia, Via Ferrata 5, I-27100 Pavia, Italy. E-mail:paolo.dibarba@unipv.it

Abstract: An evolution strategy, approximating a single Pareto-optimal solution in an objective space exhibiting more than two objectives, is proposed as a cost-effective method (individual-based approach). Moreover, a standard NSGA-II code is modified in order to control the spread-out of unfeasible solutions in a three-dimensional objective space (population-based approach). Both methods are validated by means of a case study, i.e. the optimal design of a magnetic micro-switch.

Keywords: Evolutionary computing, finite-element analysis, MEMS, multiobjective optimal design

DOI: 10.3233/JAE-140165

Journal: International Journal of Applied Electromagnetics and Mechanics, vol. 53, no. S2, pp. S239-S247, 2017