Open Access
ARTICLE
Differing Mapping using Ensemble of Metamodels for Global Variable-fidelity Metamodeling
The State Key Laboratory of Digital Manufacturing Equipment and Technology, School of Mechanical Science and Engineering, Huazhong University of Science & Technology, 430074 Wuhan, PR China.
Corresponding author. Email: jiangping@hust.edu.cn; Tel: +86-027-87557742.
Computer Modeling in Engineering & Sciences 2015, 106(5), 323-355. https://doi.org/10.3970/cmes.2015.106.323
Abstract
Computational simulation models with different fidelity have been widely used in complex systems design. However, running the most accurate simulation models tends to be very time-consuming and can therefore only be used sporadically, while incorporating less accurate, inexpensive models into the design process may result in inaccurate design alternatives. To make a trade-off between high accuracy and low expense, variable fidelity (VF) metamodeling approaches that aim to integrate information from both low fidelity (LF) and high-fidelity (HF) models have gained increasing popularity. In this paper, a Difference Mapping Framework using Ensemble of Metamodels (DMF-EM) for global VF metamodeling is proposed. In DMF-EM, a tuned model is created to bring the low fidelity model as close as possible to high fidelity model. Then, a VF metamodel is obtained by calibrating the tuned model using scaling function that is used to map the difference between the high fidelity model and the tuned model. Since the nature of the scaling function is not a priori, it is fitted using ensemble of metamodels to decrease the risk of adopting inappropriate metamodels. As a demonstration, the proposed approach is compared to existing methods using several numerical cases and two engineering examples. Results illustrate that the proposed DAD-VFM approach is more accurate and robust, that is needed in metamodel-based engineering design problems.Keywords
Cite This Article
This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.