Multiobjective optimization based on response surface model and its application to engineering shape design

A combined method is presented to deal with the practical engineering problems of multiobjective optimization. The Nondominated Sorting Genetic Algorithm II (NSGA-II) is adopted as a searching tool for the Pareto-optimal solutions, which is improved by using a new crossover operator. The response surface model (RSM) based on the radial basis function is used to reduce the computational effort. The application of the method to the shape optimization process of a permanent magnet assembly for magnetic resonance imaging devices is described, and the numerical results show that the method is feasible and efficient for the engineering shape optimization. Index Terms--Genetic algorithm, main magnet of MRI, multiobjective optimization, shape optimization.