Surrogate-guided multi-objective optimization (SGMOO) using an efficient online sampling strategy
作者：
Highlights：
• A multi-objective-based data mining strategy is presented to attain the valuable samples.
• A single-objective-based sampling approach is proposed to enhance the local infilling performance.
• Meta-heuristic search is retained and elite parents are selected out from the expensive sample set.
• A novel prescreening criterion considering hypervolume and space infilling performance is proposed.
• SGMOO has an impressive performance on mathematical cases and the shape design of BWBUGs.
摘要
•A multi-objective-based data mining strategy is presented to attain the valuable samples.•A single-objective-based sampling approach is proposed to enhance the local infilling performance.•Meta-heuristic search is retained and elite parents are selected out from the expensive sample set.•A novel prescreening criterion considering hypervolume and space infilling performance is proposed.•SGMOO has an impressive performance on mathematical cases and the shape design of BWBUGs.
论文关键词：Radial Basis Function,Computationally expensive,Multi-objective optimization,Online sampling,Surrogate models
论文评审过程：Received 31 July 2020, Revised 28 February 2021, Accepted 2 March 2021, Available online 6 March 2021, Version of Record 9 March 2021.
论文官网地址：https://doi.org/10.1016/j.knosys.2021.106919
