Abstract
Non-intrusive reduced order modeling methods (ROMs) have become increasingly popular for science and engineering applications such as predicting the field-based solutions for aerodynamic flows. A large sample size is, however, required to train the models for global accuracy. In this paper, a novel adaptive sampling strategy is introduced for these models that uses field-based uncertainty as a sampling metric. The strategy uses Monte Carlo simulations to propagate the uncertainty in the prediction of the latent space of the ROM obtained using a multi-task Gaussian process to the high-dimensional solution of the ROM. The high-dimensional uncertainty is used to discover new sampling locations to improve the global accuracy of the ROM with fewer samples. The performance of the proposed method is demonstrated on the environment model function and compared to one-shot sampling strategies. The results indicate that the proposed adaptive sampling strategies can reduce the mean relative error of the ROM to the order of \(8 \times 10^{-4}\) which is a 20% and 27% improvement over the Latin hypercube and Halton sequence sampling strategies, respectively at the same number of samples.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Aute, V., Saleh, K., Abdelaziz, O., Azarm, S., Radermacher, R.: Cross-validation based single response adaptive design of experiments for Kriging metamodeling of deterministic computer simulations. Struct. Multidiscip. Optim. 48, 581–605 (2013). https://doi.org/10.1007/s00158-013-0918-5
Benner, P., Gugercin, S., Willcox, K.: A survey of projection-based model reduction methods for parametric dynamical systems. SIAM Rev. 57, 483–531 (2015). https://doi.org/10.1137/130932715
Bliznyuk, N., Ruppert, D., Shoemaker, C., Regis, R., Wild, S., Mugunthan, P.: Bayesian calibration and uncertainty analysis for computationally expensive models using optimization and radial basis function approximation. J. Comput. Graph. Stat. 17, 270–294 (2008). https://doi.org/10.1198/106186008X320681
Bonilla, E.V., Chai, K., Williams, C.: Multi-task Gaussian process prediction. In: Advances in Neural Information Processing Systems, Vancouver, Canada, 3–6 December 2007, vol. 20 (2007)
Braconnier, T., Ferrier, M., Jouhaud, J.C., Montagnac, M., Sagaut, P.: Towards an adaptive POD/SVD surrogate model for aeronautic design. Comput. Fluids 40, 195–209 (2011). https://doi.org/10.1016/j.compfluid.2010.09.002
Decker, K., Iyengar, N., Rajaram, D., Perron, C., Mavris, D.: Manifold alignment-based nonintrusive and nonlinear multifidelity reduced-order modeling. AIAA J. 61, 454–474 (2023). https://doi.org/10.2514/1.J061720
Du, X., Wang, J., Martins, J.R.: A fully automated adaptive sampling strategy for reduced-order modeling of flow fields. In: AIAA SciTech 2023 Forum, National Harbor, MD, 23–27 January 2023 (2023). https://doi.org/10.2514/6.2023-0534
Franz, T., Zimmermann, R., Görtz, S.: Adaptive sampling for nonlinear dimensionality reduction based on manifold learning. Model. Simul. Appl. 17, 255–269 (2017). https://doi.org/10.1007/978-3-319-58786-8_16
Fuhg, J.N., Fau, A., Nackenhorst, U.: State-of-the-art and comparative review of adaptive sampling methods for Kriging. Arch. Comput. Meth. Eng. 28, 2689–2747 (2021). https://doi.org/10.1007/s11831-020-09474-6
Gardner, J.R., Pleiss, G., Bindel, D., Weinberger, K.Q., Wilson, A.G.: GPyTorch: blackbox matrix-matrix Gaussian process inference with GPU acceleration. In: Proceedings of the 32nd International Conference on Neural Information Processing Systems, Montreal, Canada, 3–8 December 2018, pp. 7587–7597 (2018)
Guénot, M., Lepot, I., Sainvitu, C., Goblet, J., Coelho, R.F.: Adaptive sampling strategies for non-intrusive POD-based surrogates. Eng. Comput. (Swansea, Wales) 30, 521–547 (2013). https://doi.org/10.1108/02644401311329352
Halder, R., Fidkowski, K.J., Maki, K.J.: Non-intrusive reduced-order modeling using convolutional autoencoders. Int. J. Numer. Meth. Eng. 123, 5369–5390 (2022). https://doi.org/10.1002/nme.7072
Jiang, P., Shu, L., Zhou, Q., Zhou, H., Shao, X., Xu, J.: A novel sequential exploration-exploitation sampling strategy for global metamodeling. IFAC-PapersOnLine 48, 532–537 (2015). https://doi.org/10.1016/j.ifacol.2015.12.183
Karcher, N., Franz, T.: Adaptive sampling strategies for reduced-order modeling. CEAS Aeronaut. J. 13, 487–502 (2022). https://doi.org/10.1007/s13272-022-00574-6
Liu, H., Ong, Y.S., Cai, J.: A survey of adaptive sampling for global metamodeling in support of simulation-based complex engineering design. Struct. Multidiscip. Optim. 57, 393–416 (2018). https://doi.org/10.1007/s00158-017-1739-8
Metropolis, N., Rosenbluth, A.W., Rosenbluth, M.N., Teller, A.H., Teller, E.: Equation of state calculations by fast computing machines. J. Chem. Phys. 21, 1087–1092 (1953). https://doi.org/10.1063/1.1699114
Niederreiter, H.: Random number generation and Quasi-Monte Carlo methods. SIAM (1992). https://doi.org/10.1137/1.9781611970081.fm
Paszke, A., et al.: PyTorch: an imperative style, high-performance deep learning library. In: 33rd Conference on Neural Information Processing Systems, NeurIPS 2019, Vancouver, Canada, 8–14 December 2019 (2019)
Rajaram, D., Perron, C., Puranik, T.G., Mavris, D.N.: Randomized algorithms for non-intrusive parametric reduced order modeling. AIAA J. 58, 5389–5407 (2020). https://doi.org/10.2514/1.J059616
Sabater, C., Stürmer, P., Bekemeyer, P.: Fast predictions of aircraft aerodynamics using deep-learning techniques. AIAA J. 60, 5249–5261 (2022). https://doi.org/10.2514/1.J061234
Storn, R., Price, K.: Differential evolution - a simple and efficient heuristic for global optimization over continuous spaces. J. Glob. Optim. 11, 341–359 (1997). https://doi.org/10.1023/A:1008202821328
Tang, B.: Orthogonal array-based Latin hypercubes. J. Am. Stat. Assoc. 88(424), 1392–1397 (1993). https://doi.org/10.1080/01621459.1993.10476423
Yang, M., Xiao, Z.: POD-based surrogate modeling of transitional flows using an adaptive sampling in Gaussian process. Int. J. Heat Fluid Flow 84, 108596 (2020). https://doi.org/10.1016/j.ijheatfluidflow.2020.108596
Yu, J., Yan, C., Guo, M.: Non-intrusive reduced-order modeling for fluid problems: a brief review. Proc. Inst. Mech. Eng. Part G J. Aerosp. Eng. 233, 5896–5912 (2019). https://doi.org/10.1177/0954410019890721
Acknowledgments
This work was supported in part by the U.S. National Science Foundation (NSF) award number 2223732 and by the Icelandic Centre for Research (RANNIS) award number 239858.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Dikshit, A., Leifsson, L., Koziel, S., Pietrenko-Dabrowska, A. (2024). Adaptive Sampling for Non-intrusive Reduced Order Models Using Multi-task Variance. In: Franco, L., de Mulatier, C., Paszynski, M., Krzhizhanovskaya, V.V., Dongarra, J.J., Sloot, P.M.A. (eds) Computational Science – ICCS 2024. ICCS 2024. Lecture Notes in Computer Science, vol 14836. Springer, Cham. https://doi.org/10.1007/978-3-031-63775-9_8
Download citation
DOI: https://doi.org/10.1007/978-3-031-63775-9_8
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-63774-2
Online ISBN: 978-3-031-63775-9
eBook Packages: Computer ScienceComputer Science (R0)