Improving Multi-Objective Genetic Algorithm Efficiency for Computational Expensive Problems Adopting Online Variable-Fidelity Metamodel

Multi-objective genetic algorithms (MOGAs) are effective ways for obtaining Pareto solutions of multi-objective optimization problems. However, the high computational cost of MOGAs limits their applications to practical engineering optimization problems involving computational expensive simulations. To address this issue, a variable-fidelity metamodel (VFM) assisted MOGA approach is proposed, in which VFM is embedded in the computation process of MOGA to replace expensive simulation models. The VFM is updated in the optimization process considering the cost of simulation models with different fidelity and the effects of the VFM uncertainty. A numerical example and an engineering case are used to demonstrate the applicability and efficiency of the proposed approach. The results show that the proposed approach can obtain Pareto solutions with high quality and it outperforms the other three existing approaches in terms of computational efficiency.