Robust shape optimization of electromechanical devices

This paper presents the robust shape optimization of electromechanical devi ces considering the uncertainties of design variables based on numerical op timization technique and finite element method (FEM). In the formulation of robust optimization, the multiobjective function is composed of the mean a nd the standard deviation of original objective function, while the constra ints are supplemented by adding penalty term to original constraints. The s equential quadratic programming (SQP) is applied to solve the robust optimi zation problem. The results of robust shape optimization considering manufa cturing errors are compared with those of conventional shape optimization.