Abstract—The goal of the robust optimization is to obtain the optimal solution while ensuring that the objective function value is not too sensitive to the uncertainties and the constraints are still feasible under the worst case of the variations of the uncertainty. The effectuation of engineering applications robust design optimization relies on the expensive simulation analysis, which is so time consuming that experimenters turned to mathematical models. In this work, a Co-Kriging multi-fidelity surrogate model assisted robust optimization approach is proposed to improve the efficiency of the robustness optimization. In the developed approach, the Co-Kriging multi-fidelity surrogate model is constructed to integrate the sample date from both low-fidelity (LF) and high-fidelity (HF) models. What is more, the concurrent treatment of the uncertainties from the multi-fidelity surrogate model, design variables, and noise parameters are investigated. The effectiveness and merits of the developed approach are illustrated on a benchmark numerical case.
Index Terms—Co-kriging, multi-fidelity surrogate model, robust optimization, uncertainty quantification.
Hansi Xu, Qi Zhou, Ping Jiang and Tingli Xie are with Huazhong University of Science and Technology, China (e-mail: frisk@hust.edu.cn, qizhouhust@gmail.com, jiangping@hust.edu.cn, xietingli0727@gmail.com).
[PDF]
Cite: Hansi Xu, Qi Zhou, Ping Jiang, and Tingli Xie, "A Co-kriging Multi-fidelity Surrogate Model Assisted Robust Optimization Approach," International Journal of Modeling and Optimization vol. 9, no. 3, pp. 155-159, 2019.
