Stochastic topology optimization based on level-set method

Abstract

Purpose

Conventional level-set method tends to fall into local optima because optimization is conducted based on gradient method. The purpose of this paper is to develop a novel topology optimization in which simulated annealing (SA) is introduced to overcome the difficulties in level-set method.

Design/methodology/approach

Level-set based topology optimization for two-dimensional optimization problem.

Findings

It is shown in the numerical examples, where conventional and present methods are applied to shape optimization of ferrite inductor and Interior Permanent Magnetic (IPM)-motor, the present method can find solutions with better performance than those obtained by the conventional method.

Originality/value

SA is introduced to improve the search performances of level-set method.

Citation

Hidaka, Y., Sato, T., Watanabe, K. and Igarashi, H. (2014), "Stochastic topology optimization based on level-set method", COMPEL - The international journal for computation and mathematics in electrical and electronic engineering, Vol. 33 No. 6, pp. 1904-1919. https://doi.org/10.1108/COMPEL-11-2013-0375