Construction of nested maximin designs based on successive local enumeration and modified novel global harmony search algorithm

ABSTRACT

Engineering design often involves different types of simulation, which results in expensive computational costs. Variable fidelity approximation-based design optimization approaches can realize effective simulation and efficiency optimization of the design space using approximation models with different levels of fidelity and have been widely used in different fields. As the foundations of variable fidelity approximation models, the selection of sample points of variable-fidelity approximation, called nested designs, is essential. In this article a novel nested maximin Latin hypercube design is constructed based on successive local enumeration and a modified novel global harmony search algorithm. In the proposed nested designs, successive local enumeration is employed to select sample points for a low-fidelity model, whereas the modified novel global harmony search algorithm is employed to select sample points for a high-fidelity model. A comparative study with multiple criteria and an engineering application are employed to verify the efficiency of the proposed nested designs approach.

KEYWORDS:

• Nested design of experiments
• variable fidelity
• harmony search
• successive local enumeration

Acknowledgments

The authors would like to thank the cloud system at Huazhong University of Science and Technology (HUST) for providing them with computing services.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This research work is supported by the National Natural Science Foundation of China (NSFC) [Grant Numbers 51305147, 61232008 and 51421062].