An output mapping variable fidelity metamodeling approach based on nested Latin hypercube design for complex engineering design optimization
- 56 Downloads
Engineering design is usually a daunting optimization task which often involving time-consuming, even computation-prohibitive process. Variable fidelity metamodeling has been developed as an efficient approach to alleviate this issue, owing to its capacity of achieving accurate metamodels within limited sample size. An output mapping modeling method is proposed in this paper as an alternative variable fidelity metamodeling method in which the low fidelity outputs is directly mapped to the high fidelity outputs through least square support vector regression. Furthermore, a nested Latin hypercube design method is developed for the output mapping modeling, in which a simple sampling design is treated as a building block or DOE seed to be translated and propagated through the design space, resulting in a final sampling configuration. Effectiveness of the proposed method are demonstrated by several numerical functions and two engineering design problem, in which different sample sizes, predictive accuracies and robustness are considered.
KeywordsVariable fidelity metamodeling Output mapping Design of experiment Latin hypercube design Least square support vector regression
This research was supported by the National Nature Science Foundation of China under grant NO.51505439, the Research Fund for the Doctoral Program of Higher Education of China under grant No.2014M562085 and the Fundamental Research Funds for the Central Universities, CUG: Grant no. CUGL150821.
- 5.Koziel S, Ciaurri DE, Leifsson L (2011) Surrogate-based methods. In: Computational optimization, methods and algorithms. Springer, pp 33–59Google Scholar
- 6.Leifsson L, Koziel S (2011) Variable-fidelity aerodynamic shape optimization. In: Koziel S, Yang XS (eds) Computational optimization, methods and algorithms, vol 356. Studies in Computational Intelligence, pp 179–210Google Scholar
- 7.Alexandrov NM, Lewis RM (2000) Optimization with variable-fidelity models applied to wing design. In: 38th Aerospace sciences meeting & exhibit. AIAA, RenoGoogle Scholar
- 17.Zheng J, Gao L, Jiang P Block latin hypercube design for variable-fidelity meta-modeling. In: 43rd International conference on computers and industrial engineering 2013, CIE 2013, October 16, 2013 - October 18, 2013, Hong Kong, Hong kong, 2013. Proceedings of international conference on computers and industrial engineering. CIE. Curran Associates Inc., pp 653–659Google Scholar
- 29.Zheng J, Shao X, Gao L, Jiang P, Qiu H (2014) Difference mapping method using least square support vector regression for variable-fidelity metamodelling. Eng Optimiz, 1–18. https://doi.org/10.1080/0305215x.2014.918114
- 31.Gano SE, Renaud JE (2004) Variable fidelity optimization using a Kriging based scaling function. In: 10th AIAA/ISSMO multidisciplinary analysis and optimization conference. AlbanyGoogle Scholar
- 36.Balabanov V, Venter G (2004) Multi-fidelity optimization with high-fidelity analysis and low-fidelity Gradients. In: 10th AIAA/ISSMO multidisciplinary analysis and optimization conference. AlbanyGoogle Scholar