MOEA/D with gradient-enhanced kriging for expensive multiobjective optimization

摘要

In many real-world engineering design optimization problems, objective function evaluations are very time costly andoften conducted by solving partial differential equations. Gradients of the objective functions can be obtained as abyproduct. Naturally, these problems can be solved more efficiently if gradient information is used. This paper studies howto do expensive multiobjective optimization when gradients are available. We propose a method, called MOEA/D–GEK,which combines MOEA/D and gradient-enhanced kriging. The gradients are used for building kriging models. Experimentalstudies on a set of test instances and an engineering problem of aerodynamic design optimization for a transonicairfoil show the high efficiency and effectiveness of our proposed method.