Robust shape optimization of electromechanical devices

Abstract

This paper presents the robust shape optimization of electromechanical devices considering the uncertainties of design variables based on numerical optimization technique and finite element method (FEM). In the formulation of robust optimization, the multiobjective function is composed of the mean and the standard deviation of original objective function, while the constraints are supplemented by adding penalty term to original constraints. The sequential quadratic programming (SQP) is applied to solve the robust optimization problem. The results of robust shape optimization considering manufacturing errors are compared with those of conventional shape optimization. © 1999 IEEE.