We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
Multi-fidelity surrogate model ensemble based on feasible intervals.
- Authors
Zhang, Shuai; Liang, Pengwei; Pang, Yong; Li, Jianji; Song, Xueguan
- Abstract
Multi-fidelity surrogate models received a lot of attention in engineering optimization due to their ability to achieve the required accuracy at a lower cost. However, selecting an appropriate scale factor to improve the prediction accuracy remains a challenge. As a result, this paper proposes a novel method for determining the scale factor. Unlike previous studies, the proposed method uses feasible intervals to determine a series of scaling factors and corresponding multi-fidelity surrogate models. Then, the ensemble of multi-fidelity surrogate models is used to improve prediction accuracy. Twenty test functions and an engineering problem are used to validate the proposed model. The results show that this model outperforms the other multi-fidelity surrogate models in terms of prediction accuracy and robustness. Furthermore, the impact of various cost ratios and proportions on the performance of the proposed model is investigated. Once again, it demonstrates a higher priority than the other models. This work provides a new approach to the design and optimization of engineering problems.
- Subjects
RATIO &; proportion; ENGINEERING design; PREDICTION models
- Publication
Structural & Multidisciplinary Optimization, 2022, Vol 65, Issue 8, p1
- ISSN
1615-147X
- Publication type
Article
- DOI
10.1007/s00158-022-03329-3