A Fast Robust Optimization Methodology Based on Polynomial Chaos and Evolutionary Algorithm for Inverse Problems

This paper explores the potential of polynomial chaos in robust designs of inverse problems. A fast numerical methodology based on combinations of polynomial chaos expansion and evolutionary algorithm is reported in this study. With the proposed methodology, polynomial chaos expansion is used as a stochastic response surface model for efficient computations of the expectancy metric of the objective function. Additional enhancements, such as the introduction of a new methodology for expected fitness assignment and probability feasibility model, a novel driving mechanism to bias the next iterations to search for both global and robust optimal solutions, are introduced. Numerical results on two case studies are reported to illustrate the feasibility and merits of the present work.