Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
169 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Nonlinear-manifold reduced order models with domain decomposition (2312.00713v1)

Published 1 Dec 2023 in math.NA and cs.NA

Abstract: A nonlinear-manifold reduced order model (NM-ROM) is a great way of incorporating underlying physics principles into a neural network-based data-driven approach. We combine NM-ROMs with domain decomposition (DD) for efficient computation. NM-ROMs offer benefits over linear-subspace ROMs (LS-ROMs) but can be costly to train due to parameter scaling with the full-order model (FOM) size. To address this, we employ DD on the FOM, compute subdomain NM-ROMs, and then merge them into a global NM-ROM. This approach has multiple advantages: parallel training of subdomain NM-ROMs, fewer parameters than global NM-ROMs, and adaptability to subdomain-specific FOM features. Each subdomain NM-ROM uses a shallow, sparse autoencoder, enabling hyper-reduction (HR) for improved computational speed. In this paper, we detail an algebraic DD formulation for the FOM, train HR-equipped NM-ROMs for subdomains, and numerically compare them to DD LS-ROMs with HR. Results show a significant accuracy boost, on the order of magnitude, for the proposed DD NM-ROMs over DD LS-ROMs in solving the 2D steady-state Burgers' equation.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (44)
  1. A discontinuous Galerkin reduced basis element method for elliptic problems. ESAIM Math. Model. Numer. Anal., 50(2):337–360, 2016. doi: 10.1051/m2an/2015045. URL https://doi.org/10.1051/m2an/2015045.
  2. A. C. Antoulas. Approximation of Large-Scale Dynamical Systems, volume 6 of Advances in Design and Control. Society for Industrial and Applied Mathematics (SIAM), Philadelphia, PA, 2005. doi: 10.1137/1.9780898718713. URL https://doi.org/10.1137/1.9780898718713.
  3. Model reduction of bilinear systems in the Loewner framework. SIAM J. Sci. Comput., 38(5):B889–B916, 2016. URL https://doi.org/10.1137/15M1041432.
  4. Interpolatory Model Reduction, volume 21 of Computational Science & Engineering. Society for Industrial and Applied Mathematics (SIAM), Philadelphia, PA, 2020. doi: 10.1137/1.9781611976083. URL https://doi.org/10.1137/1.9781611976083.
  5. The Schwarz alternating method for the seamless coupling of nonlinear reduced order models and full order models. arXiv:2210.12551, 2022. doi: 10.48550/ARXIV.2210.12551. URL https://doi.org/10.48550/ARXIV.2210.12551.
  6. Neural-network-augmented projection-based model order reduction for mitigating the Kolmogorov barrier to reducibility. J. Comput. Phys., 492:Paper No. 112420, 20, 2023. doi: 10.1016/j.jcp.2023.112420. URL https://doi.org/10.1016/j.jcp.2023.112420.
  7. P. Benner and T. Breiten. Two-sided projection methods for nonlinear model order reduction. SIAM J. Sci. Comput., 37(2):B239–B260, 2015. doi: 10.1137/14097255X. URL http://dx.doi.org/10.1137/14097255X.
  8. P. Benner and T. Breiten. Chapter 6: Model order reduction based on system balancing. In P. Benner, A. Cohen, M. Ohlberger, and K. Willcox, editors, Model Reduction and Approximation: Theory and Algorithms, Computational Science and Engineering, pages 261–295, Philadelphia, 2017. SIAM. doi: 10.1137/1.9781611974829.ch6. URL https://doi.org/10.1137/1.9781611974829.ch6.
  9. Iterative methods for model reduction by domain decomposition. Comput. & Fluids, 38(6):1160–1167, 2009. doi: 10.1016/j.compfluid.2008.11.008. URL https://doi.org/10.1016/j.compfluid.2008.11.008.
  10. Conservative model reduction for finite-volume models. Journal of Computational Physics, 371:280–314, 2018. doi: 10.1016/j.jcp.2018.05.019. URL https://doi.org/10.1016/j.jcp.2018.05.019.
  11. The GNAT method for nonlinear model reduction: Effective implementation and application to computational fluid dynamics and turbulent flows. Journal of Computational Physics, 242:623 – 647, 2013. doi: 10.1016/j.jcp.2013.02.028. URL http://dx.doi.org/10.1016/j.jcp.2013.02.028.
  12. Local lagrangian reduced-order modeling for the rayleigh-taylor instability by solution manifold decomposition. Journal of Computational Physics, 472:111655, 2023. doi: 10.1016/j.jcp.2022.111655. URL https://doi.org/10.1016/j.jcp.2022.111655.
  13. Y. Choi and K. Carlberg. Space–time least-squares petrov–galerkin projection for nonlinear model reduction. SIAM Journal on Scientific Computing, 41(1):A26–A58, 2019. doi: 10.1137/17M1120531. URL https://doi.org/10.1137/17M1120531.
  14. Space–time reduced order model for large-scale linear dynamical systems with application to boltzmann transport problems. Journal of Computational Physics, 424:109845, 2021. doi: 10.1016/j.jcp.2020.109845. URL https://doi.org/10.1016/j.jcp.2020.109845.
  15. Reduced order models for lagrangian hydrodynamics. Computer Methods in Applied Mechanics and Engineering, 388:114259, 2022. doi: 10.1016/j.cma.2021.114259. URL https://doi.org/10.1016/j.cma.2021.114259.
  16. A fast and accurate domain-decomposition nonlinear manifold reduced order model. arXiv:2305.15163v1, 2023. doi: 10.48550/arXiv.2305.15163. URL https://doi.org/10.48550/arXiv.2305.15163.
  17. Port reduction in parametrized component static condensation: approximation and a posteriori error estimation. Internat. J. Numer. Methods Engrg., 96(5):269–302, 2013. doi: 10.1002/nme.4543. URL https://doi.org/10.1002/nme.4543.
  18. Adaptive port reduction in static condensation. IFAC Proceedings Volumes, 45(2):695–699, 2012. doi: 10.3182/20120215-3-AT-3016.00123. URL https://doi.org/10.3182/20120215-3-AT-3016.00123. 7th Vienna International Conference on Mathematical Modelling.
  19. Data-driven model order reduction of quadratic-bilinear systems. Numer. Linear Algebra Appl., 25(6):e2200, 2018. doi: 10.1002/nla.2200. URL http://dx.doi.org/10.1002/nla.2200.
  20. C. Gu. QLMOR: A projection-based nonlinear model order reduction approach using quadratic-linear representation of nonlinear systems. Computer-Aided Design of Integrated Circuits and Systems, IEEE Transactions on, 30(9):1307–1320, sept. 2011. doi: 10.1109/TCAD.2011.2142184. URL https://doi.org/10.1109/TCAD.2011.2142184.
  21. M. Gubisch and S. Volkwein. Chapter 1: Proper Orthogonal Decomposition for linear-quadratic optimal control. In P. Benner, A. Cohen, M. Ohlberger, and K. Willcox, editors, Model Reduction and Approximation: Theory and Algorithms, Computational Science and Engineering, pages 3–64, Philadelphia, 2017. SIAM. doi: 10.1137/1.9781611974829.ch1. URL https://doi.org/10.1137/1.9781611974829.ch1.
  22. B. Haasdonk. Chapter 2: Reduced basis methods for parametrized PDEs - a tutorial introduction for stationary and instationary problems. In P. Benner, A. Cohen, M. Ohlberger, and K. Willcox, editors, Model Reduction and Approximation: Theory and Algorithms, Computational Science and Engineering, pages 65–136. SIAM, Philadelphia, 2017. doi: 10.1137/1.9781611974829.ch2. URL https://doi.org/10.1137/1.9781611974829.ch2.
  23. D. Hartman and L. K. Mestha. A deep learning framework for model reduction of dynamical systems. In 2017 IEEE Conference on Control Technology and Applications (CCTA), pages 1917–1922, 2017. doi: 10.1109/CCTA.2017.8062736. URL https://doi.org/10.1109/CCTA.2017.8062736.
  24. M. Hinze and S. Volkwein. Proper orthogonal decomposition surrogate models for nonlinear dynamical systems: Error estimates and suboptimal control. In P. Benner, V. Mehrmann, and D. C. Sorensen, editors, Dimension Reduction of Large-Scale Systems, Lecture Notes in Computational Science and Engineering, Vol. 45, pages 261–306, Heidelberg, 2005. Springer-Verlag. doi: 10.1007/3-540-27909-1_10. URL http://doi.org/10.1007/3-540-27909-1_10.
  25. Domain-decomposition least-squares Petrov-Galerkin (DD-LSPG) nonlinear model reduction. Comput. Methods Appl. Mech. Engrg., 384:Paper No. 113997, 41, 2021. doi: 10.1016/j.cma.2021.113997. URL https://doi.org/10.1016/j.cma.2021.113997.
  26. A static condensation reduced basis element method: approximation and a posteriori error estimation. ESAIM Math. Model. Numer. Anal., 47(1):213–251, 2013. doi: 10.1051/m2an/2012022. URL https://doi.org/10.1051/m2an/2012022.
  27. A reduced basis hybrid method for the coupling of parametrized domains represented by fluidic networks. Comput. Methods Appl. Mech. Engrg., 221/222:63–82, 2012. doi: 10.1016/j.cma.2012.02.005. URL https://doi.org/10.1016/j.cma.2012.02.005.
  28. A one-shot overlapping Schwarz method for component-based model reduction: application to nonlinear elasticity. Comput. Methods Appl. Mech. Engrg., 404:Paper No. 115786, 32, 2023. doi: 10.1016/j.cma.2022.115786. URL https://doi.org/10.1016/j.cma.2022.115786.
  29. K. Kashima. Nonlinear model reduction by deep autoencoder of noise response data. In 2016 IEEE 55th Conference on Decision and Control (CDC), pages 5750–5755, 2016. doi: 10.1109/CDC.2016.7799153. URL https://doi.org/10.1109/CDC.2016.7799153.
  30. Efficient nonlinear manifold reduced order model. arXiv preprint arXiv:2011.07727, 2020. doi: 10.48550/arXiv.2011.07727. URL https://doi.org/10.48550/arXiv.2011.07727.
  31. Efficient space–time reduced order model for linear dynamical systems in python using less than 120 lines of code. Mathematics, 9(14):1690, 2021. doi: 10.3390/math9141690. URL https://doi.org/10.3390/math9141690.
  32. A fast and accurate physics-informed neural network reduced order model with shallow masked autoencoder. J. Comput. Phys., 451:Paper No. 110841, 29, 2022. doi: 10.1016/j.jcp.2021.110841. URL https://doi.org/10.1016/j.jcp.2021.110841.
  33. K. Lee and K. T. Carlberg. Model reduction of dynamical systems on nonlinear manifolds using deep convolutional autoencoders. J. Comput. Phys., 404:108973, 32, 2020. doi: 10.1016/j.jcp.2019.108973. URL https://doi.org/10.1016/j.jcp.2019.108973.
  34. D3M: A deep domain decomposition method for partial differential equations. IEEE Access, 8:5283–5294, 2020a. doi: 10.1109/ACCESS.2019.2957200. URL https://doi.org/10.1109/ACCESS.2019.2957200.
  35. A deep domain decomposition method based on fourier features. Journal of Computational and Applied Mathematics, 423:114963, 2023. doi: 10.1016/j.cam.2022.114963. URL https://doi.org/10.1016/j.cam.2022.114963.
  36. Deep domain decomposition method: Elliptic problems. In J. Lu and R. Ward, editors, Proceedings of The First Mathematical and Scientific Machine Learning Conference, volume 107 of Proceedings of Machine Learning Research, pages 269–286. PMLR, 20–24 Jul 2020b. URL https://proceedings.mlr.press/v107/li20a.html.
  37. Y. Maday and E. M. Rønquist. A reduced-basis element method. J. Sci. Comput., 17(1-4):447–459, 2002. doi: 10.1023/A:1015197908587. URL https://doi.org/10.1023/A:1015197908587.
  38. Y. Maday and E. M. Rønquist. The reduced basis element method: application to a thermal fin problem. SIAM J. Sci. Comput., 26(1):240–258, 2004. doi: 10.1137/S1064827502419932. URL https://doi.org/10.1137/S1064827502419932.
  39. A framework for the solution of the generalized realization problem. Linear Algebra Appl., 425(2-3):634–662, 2007. doi: 10.1016/j.laa.2007.03.008. URL https://doi.org/10.1016/j.laa.2007.03.008.
  40. M. Ohlberger and S. Rave. Reduced basis methods: Success, limitations and future challenges. Proceedings of the Conference Algoritmy, pages 1–12, 2016. URL http://www.iam.fmph.uniba.sk/amuc/ojs/index.php/algoritmy/article/view/389.
  41. L. Prechelt. Automatic early stopping using cross validation: quantifying the criteria. Neural networks, 11(4):761–767, 1998. doi: 10.1016/S0893-6080(98)00010-0. URL https://doi.org/10.1016/S0893-6080(98)00010-0.
  42. Reduced Basis Methods for Partial Differential Equations. An Introduction, volume 92 of Unitext. Springer, Cham, 2016. doi: 10.1007/978-3-319-15431-2. URL https://doi.org/10.1007/978-3-319-15431-2.
  43. K. Smetana and T. Taddei. Localized model reduction for nonlinear elliptic partial differential equations: localized training, partition of unity, and adaptive enrichment. arXiv:2202.09872v1, 2022. doi: 10.48550/ARXIV.2202.09872. URL https://doi.org/10.48550/ARXIV.2202.09872.
  44. Domain decomposition learning methods for solving elliptic problems. arXiv preprint arXiv:2207.10358, 2022. doi: 10.48550/arXiv.2207.10358. URL https://doi.org/10.48550/arXiv.2207.10358.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now