doi: 10.3934/dcdsb.2022084
Online First

Online First articles are published articles within a journal that have not yet been assigned to a formal issue. This means they do not yet have a volume number, issue number, or page numbers assigned to them, however, they can still be found and cited using their DOI (Digital Object Identifier). Online First publication benefits the research community by making new scientific discoveries known as quickly as possible.

Readers can access Online First articles via the “Online First” tab for the selected journal.

A non-intrusive model order reduction approach for parameterized time-domain Maxwell's equations

1. 

School of Mathematics, Southwestern University of Finance and Economics, Chengdu, Sichuan 611130, China

2. 

School of Mathematical Sciences, University of Electronic Science and Technology of China, Chengdu, Sichuan 611731, China

3. 

Université Côte d'Azur, Inria, CNRS, LJAD, Sophia Antipolis Cedex, France

*Corresponding author: Ting-Zhu Huang, Liang Li

Received  October 2021 Revised  March 2022 Early access May 2022

Fund Project: The first author is supported by the NSFC (Grant No. 12101511). The second author is supported by the NSFC (Grant No. 61772003) and the Key Projects of Applied Basic Research in Sichuan Province (Grant No. 2020YJ0216)

We present a non-intrusive model order reduction (NIMOR) approach with an offline-online decoupling for the solution of parameterized time-domain Maxwell's equations. During the offline stage, the training parameters are chosen by using Smolyak sparse grid method with an approximation level $ L $ ($ L\geq1 $) over a target parameterized space. For each selected parameter, the snapshot vectors are first produced by a high order discontinuous Galerkin time-domain (DGTD) solver formulated on an unstructured simplicial mesh. In order to minimize the overall computational cost in the offline stage and to improve the accuracy of the NIMOR method, a radial basis function (RBF) interpolation method is then used to construct more snapshot vectors at the sparse grid with approximation level $ L+1 $, which includes the sparse grids from approximation level $ L $. A nested proper orthogonal decomposition (POD) method is employed to extract time- and parameter-independent POD basis functions. By using the singular value decomposition (SVD) method, the principal components of the reduced coefficient matrices of the high-fidelity solutions onto the reduced-order subspace spaned by the POD basis functions are extracted. Moreover, a Gaussian process regression (GPR) method is proposed to approximate the dominating time- and parameter-modes of the reduced coefficient matrices. During the online stage, the reduced-order solutions for new time and parameter values can be rapidly recovered via outputs from the regression models without using the DGTD method. Numerical experiments for the scattering of plane wave by a 2-D dielectric cylinder and a multi-layer heterogeneous medium nicely illustrate the performance of the NIMOR method.

Citation: Kun Li, Ting-Zhu Huang, Liang Li, Ying Zhao, Stéphane Lanteri. A non-intrusive model order reduction approach for parameterized time-domain Maxwell's equations. Discrete and Continuous Dynamical Systems - B, doi: 10.3934/dcdsb.2022084
References:
[1]

F. Alsayyari, Z. Perkó, M. Tiberga, J. L. Kloosterman and D. Lathouwers, A fully adaptive nonintrusive reduced-order modelling approach for parametrized time-dependent problems, Comput. Methods Appl. Mech. Engrg., 373 (2021), Paper No. 113483, 21 pp. doi: 10.1016/j.cma.2020.113483.

[2]

C. AudouzeF. D. Vuyst and P. B. Nair, Nonintrusive reduced-order modeling of parametrized time-dependent partial differential equations, Numer. Methods Partial Differential Equations, 29 (2013), 1587-1628.  doi: 10.1002/num.21768.

[3]

M. BarraultY. MadayN. C. Nguyen and A. T. Patera, An 'empirical interpolation' method: Application to efficient reduced-basis discretization of partial differential equations, C. R. Math., 339 (2004), 667-672.  doi: 10.1016/j.crma.2004.08.006.

[4]

P. BennerS. Gugercin and K. Willcox, A survey of projection-based model reduction methods for parametric dynamical systems, SIAM Rev., 57 (2015), 483-531.  doi: 10.1137/130932715.

[5]

M. BernackiL. FezouiS. Lanteri and S. Piperno, Parallel discontinuous Galerkin unstructured mesh solvers for the calculation of three-dimensional wave propagation problems, Appl. Math. Model., 30 (2006), 744-763. 

[6]

F. CasenaveA. Ern and T. Leliévre, A nonintrusive reduced basis method applied to aeroacoustic simulations, Adv. Comput. Math., 41 (2015), 961-986.  doi: 10.1007/s10444-014-9365-0.

[7]

R. ChakirY. Maday and P. Parnaudeau, A non-intrusive reduced basis approach for parametrized heat transfer problems, J. Comput. Phys., 376 (2019), 617-633.  doi: 10.1016/j.jcp.2018.10.001.

[8]

S. Chaturantabut and D. C. Sorensen, Nonlinear model reduction via discrete empirical interpolation, SIAM J. Sci. Comput., 32 (2010), 2737-2764.  doi: 10.1137/090766498.

[9]

W. Chen, Q. Wang, J. S. Hesthaven and C. Zhang, Physics-informed machine learning for reduced-order modeling of nonlinear problems, J. Comput. Phys., 446 (2021), Paper No. 110666, 28 pp. doi: 10.1016/j.jcp.2021.110666.

[10]

N. Dal SantoS. Deparis and L. Pegolotti, Data driven approximation of parametrized pdes by reduced basis and neural networks, J. Comput. Phys., 416 (2020), 109550.  doi: 10.1016/j.jcp.2020.109550.

[11]

S. FrescaL. Dede and A. Manzoni, A comprehensive deep learning-based approach to reduced order modeling of nonlinear time-dependent parametrized pdes, J. Sci. Comput., 87 (2021).  doi: 10.1007/s10915-021-01462-7.

[12]

T. Gerstner and M. Griebel, Numerical integration using sparse grids, Numer. Algorithms, 18 (1998).  doi: 10.1023/A:1019129717644.

[13]

M. Guo and J. S. Hesthaven, Data-driven reduced order modeling for time-dependent problems, Comput. Methods Appl. Mech. Engrg., 345 (2019), 75-99.  doi: 10.1016/j.cma.2018.10.029.

[14]

J. S. Hesthaven, G. Rozza and B. Stamm, et al., Certified Reduced Basis Methods for Parametrized Partial Differential Equations, Springer, Cham; BCAM Basque Center for Applied Mathematics, Bilbao, 2016. doi: 10.1007/978-3-319-22470-1.

[15]

J. S. Hesthaven and U. Stefano, Non-intrusive reduced order modeling of nonlinear problems using neural networks, J. Comput. Phys., 363 (2018), 55-78.  doi: 10.1016/j.jcp.2018.02.037.

[16]

J. S. Hesthaven and T. Warburton, Nodal Discontinuous Galerkin Methods: Algorithms, Analysis, and Applications, Springer, New York, 2008. doi: 10.1007/978-0-387-72067-8.

[17]

K. L. JuddL. MaliarS. Maliar and R. Valero, Smolyak method for solving dynamic economic models: Lagrange interpolation, anisotropic grid and adaptive domain, J. Econom. Dynam. Control, 44 (2014), 92-123.  doi: 10.1016/j.jedc.2014.03.003.

[18]

E. N. KaratzasF. Ballarin and G. Rozza, Projection-based reduced order models for a cut finite element method in parametrized domains, Comput. Math. Appl., 79 (2020), 833-851.  doi: 10.1016/j.camwa.2019.08.003.

[19]

M. Kast, M. Guo and J. S. Hesthaven, A non-intrusive multifidelity method for the reduced order modeling of nonlinear problems, Comput. Methods Appl. Mech. Engrg., 364 (2020), 112947, 28 pp. doi: 10.1016/j.cma.2020.112947.

[20]

K. Kunisch and S. Volkwein, Optimal snapshot location for computing POD basis functions, M2AN Math. Model. Numer. Anal., 44 (2010), 509-529.  doi: 10.1051/m2an/2010011.

[21]

O. Lass and S. Volkwein, POD-Galerkin schemes for nonlinear elliptic-parabolic systems, SIAM J. Sci. Comput., 35 (2013), 1271-1298.  doi: 10.1137/110848414.

[22]

K. Li, T.-Z. Huang, L. Li and S. Lanteri, Non-intrusive reduced-order modeling of parameterized electromagnetic scattering problems using cubic spline interpolation, J. Sci. Comput., 87 (2021), Paper No. 52, 29 pp. doi: 10.1007/s10915-021-01467-2.

[23]

K. LiT.-Z. HuangL. Li and S. Lanteri, POD-based model order reduction with an adaptive snapshot selection for a discontinuous Galerkin approximation of the time-domain Maxwell's equations, J. Comput. Phys., 396 (2019), 106-128.  doi: 10.1016/j.jcp.2019.05.051.

[24]

K. LiT.-Z. HuangL. LiS. LanteriL. Xu and B. Li, A reduced-order discontinuous Galerkin method based on POD for electromagnetic simulation, IEEE Trans. Antennas and Propagation, 66 (2018), 242-254. 

[25]

S. L. Lohr, Sampling: Design and Analysis, 2$^{nd}$ edition, Brooks/Cole, Cengage Learning, Boston, MA, 2010.

[26]

S. LorenziA. CammiL. Luzzi and G. Rozza, POD-Galerkin method for finite volume approximation of Navier–Stokes and RANS equations, Comput. Methods Appl. Mech. Engrg., 311 (2016), 151-179.  doi: 10.1016/j.cma.2016.08.006.

[27]

D. LoukrezisU. Römer and H. D. Gersem, Assessing the performance of Leja and Clenshaw-Curtis collocation for computational electromagnetics with random input data, Int. J. Uncertain. Quantif., 9 (2019), 33-57.  doi: 10.1615/Int.J.UncertaintyQuantification.2018025234.

[28]

Z. Luo, Proper orthogonal decomposition-based reduced-order stabilized mixed finite volume element extrapolating model for the nonstationary incompressible Boussinesq equations, J. Math. Anal. Appl., 425 (2015), 259-280.  doi: 10.1016/j.jmaa.2014.12.011.

[29]

Z. Luo and J. Gao, A POD reduced-order finite difference time-domain extrapolating scheme for the 2D Maxwell equations in a lossy medium, J. Math. Anal. Appl., 444 (2016), 433-451.  doi: 10.1016/j.jmaa.2016.06.036.

[30]

Z. Luo and W. Jiang, A reduced-order extrapolated technique about the unknown coefficient vectors of solutions in the finite element method for hyperbolic type equation, Appl. Numer. Math., 158 (2020), 123-133.  doi: 10.1016/j.apnum.2020.07.025.

[31]

Z. LuoH. LiY. Zhou and X. Huang, A reduced FVE formulation based on POD method and error analysis for two-dimensional viscoelastic problem, J. Math. Anal. Appl., 385 (2012), 310-321.  doi: 10.1016/j.jmaa.2011.06.057.

[32]

Z. LuoQ. Ou and Z. Xie, Reduced finite difference scheme and error estimates based on POD method for non-stationary Stokes equation, Appl. Math. Mech., 32 (2011), 847-858.  doi: 10.1007/s10483-011-1464-9.

[33]

Z. Luo and F. Teng, A reduced-order extrapolated finite difference iterative scheme based on POD method for 2D sobolev equation, Appl. Math. Comput., 329 (2018), 374-383.  doi: 10.1016/j.amc.2018.02.022.

[34]

Z. LuoF. Teng and H. Xia, A reduced-order extrapolated Crank-Nicolson finite spectral element method based on POD for the 2D non-stationary boussinesq equations, J. Math. Anal. Appl., 471 (2019), 564-583.  doi: 10.1016/j.jmaa.2018.10.092.

[35]

Z. Luo and H. Ren, A reduced-order extrapolated finite difference iterative method for the riemann-liouville tempered fractional derivative equation, Appl. Numer. Math., 157 (2020), 307-314.  doi: 10.1016/j.apnum.2020.05.028.

[36]

Z. Luo and J. Shiju, A reduced-order extrapolated Crank-Nicolson collocation spectral method based on proper orthogonal decomposition for the two-dimensional viscoelastic wave equations, Numer. Methods Partial Differential Equations, 36 (2020), 49-65.  doi: 10.1002/num.22397.

[37]

B. PeherstorferK. Willcox and M. Gunzburger, Survey of multifidelity methods in uncertainty propagation, inference, and optimization, SIAM Rev., 60 (2018), 550-591.  doi: 10.1137/16M1082469.

[38]

P. PhalippouS. BouabdallahP. BreitkopfP. Villon and M. Zarroug, 'On-the-fly' snapshots selection for proper orthogonal decomposition with application to nonlinear dynamics, Comput. Methods Appl. Mech. Engrg., 367 (2020), 113120.  doi: 10.1016/j.cma.2020.113120.

[39]

C. E. Rasmussen and C. K. I. Williams, Gaussian Processes for Machine Learning, MIT press, 2006.

[40]

N. SchmittC. ScheidJ. Viquerat and S. Lanteri, Simulation of three-dimensional nanoscale light interaction with spatially dispersive metals using a high order curvilinear DGTD method, J. Comput. Phys., 373 (2018), 210-229.  doi: 10.1016/j.jcp.2018.06.033.

[41]

G. Stabile and G. Rozza, Finite POD-Galerkin stabilised reduced order methods for the parametrised incompressible Navier-Stokes equations, Comput. & Fluids, 173 (2018), 273-284.  doi: 10.1016/j.compfluid.2018.01.035.

[42]

X. SunX. Pan and J.-I. Choi, Non-intrusive framework of reduced-order modeling based on proper orthogonal decomposition and polynomial chaos expansion, J. Comput. Appl. Math., 390 (2021), 113372.  doi: 10.1016/j.cam.2020.113372.

[43]

S. UllmannM. Rotkvic and J. Lang, POD-Galerkin reduced-order modeling with adaptive finite element snapshots, J. Comput. Phys., 325 (2016), 244-258.  doi: 10.1016/j.jcp.2016.08.018.

[44]

F. Vidal-CodinaN. C. Nguyen and J. Peraire, Computing parametrized solutions for plasmonic nanogap structures, J. Comput. Phys., 366 (2018), 89-106.  doi: 10.1016/j.jcp.2018.04.009.

[45]

J. Viquerat and S. Lanteri, Simulation of near-field plasmonic interactions with a local approximation order discontinuous Galerkin time-domain method, Photonics Nanostruct., 18 (2016), 43-58. 

[46]

Q. WangJ. S. Hesthaven and D. Ray, Non-intrusive reduced order modeling of unsteady flows using artificial neural networks with application to a combustion problem, J. Comput. Phys., 384 (2019), 289-307.  doi: 10.1016/j.jcp.2019.01.031.

[47]

Q. Wang, N. Ripamonti and J. S. Hesthaven, Recurrent neural network closure of parametric POD-Galerkin reduced-order models based on the Mori-Zwanzig formalism, J. Comput. Phys., 410 (2020), 109402, 32 pp. doi: 10.1016/j.jcp.2020.109402.

[48]

C. K. I. Williams and C. E. Rasmussen, Gaussian processes for regression, In Proceedings of the 8th International Conference on Neural Information Processing Systems, MIT, (1995), 514–520.

[49]

D. XiaoF. FangC. Pain and I. Navon, A parameterized non-intrusive reduced order model and error analysis for general time-dependent nonlinear partial differential equations and its applications, Comput. Methods Appl. Mech. Engrg., 317 (2017), 868-889.  doi: 10.1016/j.cma.2016.12.033.

[50]

D. XiaoZ. LinF. FangC. C. PainL. M. NavonP. Salinas and A. Muggeridge, Non-intrusive reduced-order modeling for multiphase porous media flows using smolyak sparse grids, Internat. J. Numer. Methods Fluids, 83 (2017), 205-219.  doi: 10.1002/fld.4263.

[51]

B. Xu and X. Zhang, An efficient high-order compact finite difference scheme based on proper orthogonal decomposition for the multi-dimensional parabolic equation, Adv. Differ. Equ., 2019 (2019), 341.  doi: 10.1186/s13662-019-2273-3.

[52]

R. YondoE. Andrés and E. Valero, A review on design of experiments and surrogate models in aircraft real-time and many-query aerodynamic analyses, Prog. Aerosp. Sci., 96 (2018), 23-61. 

[53]

J. YuC. YanZ. JiangW. Yuan and S. Chen, Adaptive non-intrusive reduced order modeling for compressible flows, J. Comput. Phys., 397 (2019), 108855.  doi: 10.1016/j.jcp.2019.07.053.

[54]

X. Zhang and P. Zhang, A reduced high-order compact finite difference scheme based on proper orthogonal decomposition technique for KdV equation, Appl. Math. Comput., 339 (2018), 535-545.  doi: 10.1016/j.amc.2018.07.017.

[55]

Y. ZhouY. ZhangY. Liang and Z. Luo, A reduced-order extrapolated model based on splitting implicit finite difference scheme and proper orthogonal decomposition for the fourth-order nonlinear rosenau equation, Appl. Numer. Math., 162 (2021), 192-200.  doi: 10.1016/j.apnum.2020.12.020.

show all references

References:
[1]

F. Alsayyari, Z. Perkó, M. Tiberga, J. L. Kloosterman and D. Lathouwers, A fully adaptive nonintrusive reduced-order modelling approach for parametrized time-dependent problems, Comput. Methods Appl. Mech. Engrg., 373 (2021), Paper No. 113483, 21 pp. doi: 10.1016/j.cma.2020.113483.

[2]

C. AudouzeF. D. Vuyst and P. B. Nair, Nonintrusive reduced-order modeling of parametrized time-dependent partial differential equations, Numer. Methods Partial Differential Equations, 29 (2013), 1587-1628.  doi: 10.1002/num.21768.

[3]

M. BarraultY. MadayN. C. Nguyen and A. T. Patera, An 'empirical interpolation' method: Application to efficient reduced-basis discretization of partial differential equations, C. R. Math., 339 (2004), 667-672.  doi: 10.1016/j.crma.2004.08.006.

[4]

P. BennerS. Gugercin and K. Willcox, A survey of projection-based model reduction methods for parametric dynamical systems, SIAM Rev., 57 (2015), 483-531.  doi: 10.1137/130932715.

[5]

M. BernackiL. FezouiS. Lanteri and S. Piperno, Parallel discontinuous Galerkin unstructured mesh solvers for the calculation of three-dimensional wave propagation problems, Appl. Math. Model., 30 (2006), 744-763. 

[6]

F. CasenaveA. Ern and T. Leliévre, A nonintrusive reduced basis method applied to aeroacoustic simulations, Adv. Comput. Math., 41 (2015), 961-986.  doi: 10.1007/s10444-014-9365-0.

[7]

R. ChakirY. Maday and P. Parnaudeau, A non-intrusive reduced basis approach for parametrized heat transfer problems, J. Comput. Phys., 376 (2019), 617-633.  doi: 10.1016/j.jcp.2018.10.001.

[8]

S. Chaturantabut and D. C. Sorensen, Nonlinear model reduction via discrete empirical interpolation, SIAM J. Sci. Comput., 32 (2010), 2737-2764.  doi: 10.1137/090766498.

[9]

W. Chen, Q. Wang, J. S. Hesthaven and C. Zhang, Physics-informed machine learning for reduced-order modeling of nonlinear problems, J. Comput. Phys., 446 (2021), Paper No. 110666, 28 pp. doi: 10.1016/j.jcp.2021.110666.

[10]

N. Dal SantoS. Deparis and L. Pegolotti, Data driven approximation of parametrized pdes by reduced basis and neural networks, J. Comput. Phys., 416 (2020), 109550.  doi: 10.1016/j.jcp.2020.109550.

[11]

S. FrescaL. Dede and A. Manzoni, A comprehensive deep learning-based approach to reduced order modeling of nonlinear time-dependent parametrized pdes, J. Sci. Comput., 87 (2021).  doi: 10.1007/s10915-021-01462-7.

[12]

T. Gerstner and M. Griebel, Numerical integration using sparse grids, Numer. Algorithms, 18 (1998).  doi: 10.1023/A:1019129717644.

[13]

M. Guo and J. S. Hesthaven, Data-driven reduced order modeling for time-dependent problems, Comput. Methods Appl. Mech. Engrg., 345 (2019), 75-99.  doi: 10.1016/j.cma.2018.10.029.

[14]

J. S. Hesthaven, G. Rozza and B. Stamm, et al., Certified Reduced Basis Methods for Parametrized Partial Differential Equations, Springer, Cham; BCAM Basque Center for Applied Mathematics, Bilbao, 2016. doi: 10.1007/978-3-319-22470-1.

[15]

J. S. Hesthaven and U. Stefano, Non-intrusive reduced order modeling of nonlinear problems using neural networks, J. Comput. Phys., 363 (2018), 55-78.  doi: 10.1016/j.jcp.2018.02.037.

[16]

J. S. Hesthaven and T. Warburton, Nodal Discontinuous Galerkin Methods: Algorithms, Analysis, and Applications, Springer, New York, 2008. doi: 10.1007/978-0-387-72067-8.

[17]

K. L. JuddL. MaliarS. Maliar and R. Valero, Smolyak method for solving dynamic economic models: Lagrange interpolation, anisotropic grid and adaptive domain, J. Econom. Dynam. Control, 44 (2014), 92-123.  doi: 10.1016/j.jedc.2014.03.003.

[18]

E. N. KaratzasF. Ballarin and G. Rozza, Projection-based reduced order models for a cut finite element method in parametrized domains, Comput. Math. Appl., 79 (2020), 833-851.  doi: 10.1016/j.camwa.2019.08.003.

[19]

M. Kast, M. Guo and J. S. Hesthaven, A non-intrusive multifidelity method for the reduced order modeling of nonlinear problems, Comput. Methods Appl. Mech. Engrg., 364 (2020), 112947, 28 pp. doi: 10.1016/j.cma.2020.112947.

[20]

K. Kunisch and S. Volkwein, Optimal snapshot location for computing POD basis functions, M2AN Math. Model. Numer. Anal., 44 (2010), 509-529.  doi: 10.1051/m2an/2010011.

[21]

O. Lass and S. Volkwein, POD-Galerkin schemes for nonlinear elliptic-parabolic systems, SIAM J. Sci. Comput., 35 (2013), 1271-1298.  doi: 10.1137/110848414.

[22]

K. Li, T.-Z. Huang, L. Li and S. Lanteri, Non-intrusive reduced-order modeling of parameterized electromagnetic scattering problems using cubic spline interpolation, J. Sci. Comput., 87 (2021), Paper No. 52, 29 pp. doi: 10.1007/s10915-021-01467-2.

[23]

K. LiT.-Z. HuangL. Li and S. Lanteri, POD-based model order reduction with an adaptive snapshot selection for a discontinuous Galerkin approximation of the time-domain Maxwell's equations, J. Comput. Phys., 396 (2019), 106-128.  doi: 10.1016/j.jcp.2019.05.051.

[24]

K. LiT.-Z. HuangL. LiS. LanteriL. Xu and B. Li, A reduced-order discontinuous Galerkin method based on POD for electromagnetic simulation, IEEE Trans. Antennas and Propagation, 66 (2018), 242-254. 

[25]

S. L. Lohr, Sampling: Design and Analysis, 2$^{nd}$ edition, Brooks/Cole, Cengage Learning, Boston, MA, 2010.

[26]

S. LorenziA. CammiL. Luzzi and G. Rozza, POD-Galerkin method for finite volume approximation of Navier–Stokes and RANS equations, Comput. Methods Appl. Mech. Engrg., 311 (2016), 151-179.  doi: 10.1016/j.cma.2016.08.006.

[27]

D. LoukrezisU. Römer and H. D. Gersem, Assessing the performance of Leja and Clenshaw-Curtis collocation for computational electromagnetics with random input data, Int. J. Uncertain. Quantif., 9 (2019), 33-57.  doi: 10.1615/Int.J.UncertaintyQuantification.2018025234.

[28]

Z. Luo, Proper orthogonal decomposition-based reduced-order stabilized mixed finite volume element extrapolating model for the nonstationary incompressible Boussinesq equations, J. Math. Anal. Appl., 425 (2015), 259-280.  doi: 10.1016/j.jmaa.2014.12.011.

[29]

Z. Luo and J. Gao, A POD reduced-order finite difference time-domain extrapolating scheme for the 2D Maxwell equations in a lossy medium, J. Math. Anal. Appl., 444 (2016), 433-451.  doi: 10.1016/j.jmaa.2016.06.036.

[30]

Z. Luo and W. Jiang, A reduced-order extrapolated technique about the unknown coefficient vectors of solutions in the finite element method for hyperbolic type equation, Appl. Numer. Math., 158 (2020), 123-133.  doi: 10.1016/j.apnum.2020.07.025.

[31]

Z. LuoH. LiY. Zhou and X. Huang, A reduced FVE formulation based on POD method and error analysis for two-dimensional viscoelastic problem, J. Math. Anal. Appl., 385 (2012), 310-321.  doi: 10.1016/j.jmaa.2011.06.057.

[32]

Z. LuoQ. Ou and Z. Xie, Reduced finite difference scheme and error estimates based on POD method for non-stationary Stokes equation, Appl. Math. Mech., 32 (2011), 847-858.  doi: 10.1007/s10483-011-1464-9.

[33]

Z. Luo and F. Teng, A reduced-order extrapolated finite difference iterative scheme based on POD method for 2D sobolev equation, Appl. Math. Comput., 329 (2018), 374-383.  doi: 10.1016/j.amc.2018.02.022.

[34]

Z. LuoF. Teng and H. Xia, A reduced-order extrapolated Crank-Nicolson finite spectral element method based on POD for the 2D non-stationary boussinesq equations, J. Math. Anal. Appl., 471 (2019), 564-583.  doi: 10.1016/j.jmaa.2018.10.092.

[35]

Z. Luo and H. Ren, A reduced-order extrapolated finite difference iterative method for the riemann-liouville tempered fractional derivative equation, Appl. Numer. Math., 157 (2020), 307-314.  doi: 10.1016/j.apnum.2020.05.028.

[36]

Z. Luo and J. Shiju, A reduced-order extrapolated Crank-Nicolson collocation spectral method based on proper orthogonal decomposition for the two-dimensional viscoelastic wave equations, Numer. Methods Partial Differential Equations, 36 (2020), 49-65.  doi: 10.1002/num.22397.

[37]

B. PeherstorferK. Willcox and M. Gunzburger, Survey of multifidelity methods in uncertainty propagation, inference, and optimization, SIAM Rev., 60 (2018), 550-591.  doi: 10.1137/16M1082469.

[38]

P. PhalippouS. BouabdallahP. BreitkopfP. Villon and M. Zarroug, 'On-the-fly' snapshots selection for proper orthogonal decomposition with application to nonlinear dynamics, Comput. Methods Appl. Mech. Engrg., 367 (2020), 113120.  doi: 10.1016/j.cma.2020.113120.

[39]

C. E. Rasmussen and C. K. I. Williams, Gaussian Processes for Machine Learning, MIT press, 2006.

[40]

N. SchmittC. ScheidJ. Viquerat and S. Lanteri, Simulation of three-dimensional nanoscale light interaction with spatially dispersive metals using a high order curvilinear DGTD method, J. Comput. Phys., 373 (2018), 210-229.  doi: 10.1016/j.jcp.2018.06.033.

[41]

G. Stabile and G. Rozza, Finite POD-Galerkin stabilised reduced order methods for the parametrised incompressible Navier-Stokes equations, Comput. & Fluids, 173 (2018), 273-284.  doi: 10.1016/j.compfluid.2018.01.035.

[42]

X. SunX. Pan and J.-I. Choi, Non-intrusive framework of reduced-order modeling based on proper orthogonal decomposition and polynomial chaos expansion, J. Comput. Appl. Math., 390 (2021), 113372.  doi: 10.1016/j.cam.2020.113372.

[43]

S. UllmannM. Rotkvic and J. Lang, POD-Galerkin reduced-order modeling with adaptive finite element snapshots, J. Comput. Phys., 325 (2016), 244-258.  doi: 10.1016/j.jcp.2016.08.018.

[44]

F. Vidal-CodinaN. C. Nguyen and J. Peraire, Computing parametrized solutions for plasmonic nanogap structures, J. Comput. Phys., 366 (2018), 89-106.  doi: 10.1016/j.jcp.2018.04.009.

[45]

J. Viquerat and S. Lanteri, Simulation of near-field plasmonic interactions with a local approximation order discontinuous Galerkin time-domain method, Photonics Nanostruct., 18 (2016), 43-58. 

[46]

Q. WangJ. S. Hesthaven and D. Ray, Non-intrusive reduced order modeling of unsteady flows using artificial neural networks with application to a combustion problem, J. Comput. Phys., 384 (2019), 289-307.  doi: 10.1016/j.jcp.2019.01.031.

[47]

Q. Wang, N. Ripamonti and J. S. Hesthaven, Recurrent neural network closure of parametric POD-Galerkin reduced-order models based on the Mori-Zwanzig formalism, J. Comput. Phys., 410 (2020), 109402, 32 pp. doi: 10.1016/j.jcp.2020.109402.

[48]

C. K. I. Williams and C. E. Rasmussen, Gaussian processes for regression, In Proceedings of the 8th International Conference on Neural Information Processing Systems, MIT, (1995), 514–520.

[49]

D. XiaoF. FangC. Pain and I. Navon, A parameterized non-intrusive reduced order model and error analysis for general time-dependent nonlinear partial differential equations and its applications, Comput. Methods Appl. Mech. Engrg., 317 (2017), 868-889.  doi: 10.1016/j.cma.2016.12.033.

[50]

D. XiaoZ. LinF. FangC. C. PainL. M. NavonP. Salinas and A. Muggeridge, Non-intrusive reduced-order modeling for multiphase porous media flows using smolyak sparse grids, Internat. J. Numer. Methods Fluids, 83 (2017), 205-219.  doi: 10.1002/fld.4263.

[51]

B. Xu and X. Zhang, An efficient high-order compact finite difference scheme based on proper orthogonal decomposition for the multi-dimensional parabolic equation, Adv. Differ. Equ., 2019 (2019), 341.  doi: 10.1186/s13662-019-2273-3.

[52]

R. YondoE. Andrés and E. Valero, A review on design of experiments and surrogate models in aircraft real-time and many-query aerodynamic analyses, Prog. Aerosp. Sci., 96 (2018), 23-61. 

[53]

J. YuC. YanZ. JiangW. Yuan and S. Chen, Adaptive non-intrusive reduced order modeling for compressible flows, J. Comput. Phys., 397 (2019), 108855.  doi: 10.1016/j.jcp.2019.07.053.

[54]

X. Zhang and P. Zhang, A reduced high-order compact finite difference scheme based on proper orthogonal decomposition technique for KdV equation, Appl. Math. Comput., 339 (2018), 535-545.  doi: 10.1016/j.amc.2018.07.017.

[55]

Y. ZhouY. ZhangY. Liang and Z. Luo, A reduced-order extrapolated model based on splitting implicit finite difference scheme and proper orthogonal decomposition for the fourth-order nonlinear rosenau equation, Appl. Numer. Math., 162 (2021), 192-200.  doi: 10.1016/j.apnum.2020.12.020.

Figure 1.  1-D Smolyak sparse grids with approximation levels 0, 1, 2, 3
Figure 2.  2-D Smolyak sparse grid with approximation level 3 and full tensor product grid
Figure 3.  Scattering of plane wave by a dielectric disk: the convergence histories of $ \overline{e}_{ {\bf E}, {\rm{Pro}}^{(i)}} $, and $ \overline{e}_{ {\bf E}, {\rm{NIMOR}}^{(i)}} $ (a), $ \overline{e}_{ {\bf H}, {\rm{Pro}}^{(i)}} $, and $ \overline{e}_{ {\bf H}, {\rm{NIMOR}}^{(i)}} $ (b) ($ i = 1, 2 $) on the testing set $ \mathcal{T}_{te}\times\mathcal{P}_{te} $ with vary truncation tolerances $ \rho_\theta $, where $ \overline{e}_{ {\bf u}, {\rm{Pro}}^{(i)}} $ is the average projection error of $ {\rm{NIMOR}}^{(i)} $ method for $ {\bf u} $
Figure 4.  Scattering of plane wave by a dielectric disk: the $ 5 $-th, $ 10 $-th, $ 15 $-th, and $ 20 $-th exact and approximation reduced-order cofficients of $ E_z $ based on $ {\rm{NIMOR}}^{(1)} $
Figure 5.  Scattering of plane wave by a dielectric disk: the $ 5 $-th, $ 10 $-th, $ 15 $-th, and $ 20 $-th exact and approximation reduced-order cofficients of $ H_y $ based on $ {\rm{NIMOR}}^{(1)} $
Figure 6.  Scattering of plane wave by a dielectric disk: comparison of the 1-D x-wise distribution along $ y = 0 $ of the real part of $ H_y $ (left) and $ E_z $ (right) of four test points: $ \theta^{(1)} = 1.215 $ (1-th row), $ \theta^{(2)} = 2.215 $ (2-th row), $ \theta^{(3)} = 3.215 $ (3-th row) and $ \theta^{(4)} = 4.215 $ (4-th row)
Figure 7.  Scattering of plane wave by a dielectric disk: comparison of relative projection and NIMOR errors for $ {\bf E} $ (left) and $ {\bf H} $ (right) four the testing parameters
Figure 8.  Scattering of a plane wave by a multi-layer heterogeneous medium: geometry of the multi-layer medium
Figure 9.  Scattering of a plane wave by a multi-layer heterogeneous medium: the convergence histories of $ \overline{e}_{ {\bf E}, {\rm{Pro}}^{(i)}} $, and $ \overline{e}_{ {\bf E}, {\rm{NIMOR}}^{(i)}} $ (a), $ \overline{e}_{ {\bf H}, {\rm{Pro}}^{(i)}} $, and $ \overline{e}_{ {\bf H}, {\rm{NIMOR}}^{(i)}} $ (b) ($ i = 1, 2 $) on the testing set $ \mathcal{T}_{te}\times\mathcal{P}_{te} $ with vary truncation tolerances $ \rho_\theta $, where $ \overline{e}_{ {\bf u}, {\rm{Pro}}^{(i)}} $ is the average projection error of $ {\rm{NIMOR}}^{(i)} $ method for $ {\bf u} $
Figure 10.  Scattering of plane wave by a multi-layer heterogeneous medium: the 3th, 6th, 9th, and 12th exact ($ - $) and approximation ($ * $) time-modes for $ E_z $ (left) and $ H_y $ (right) (the 1th mode: black, the 3th mode: red, the 5th mode: brown; the 7th mode: blue; the 9th mode: green)
Figure 11.  Scattering of plane wave by a multi-layer heterogeneous medium: comparison of the 1-D x-wise distribution along $ y = 0 $ of the real part of $ H_y $ (left) and $ E_z $ (right) of four test points: $ \theta^{(1)} = \{(5.125, 3.375, 2.125, 1.375)\} $ (1-th row), $ \theta^{(2)} = \{(5.425, 3.625, 2.425, 1.625)\} $ (2-th row), $ \theta^{(3)} = \{(5.125, 3.625, 2.125, 1.625)\} $ (3-th row) and $ \theta^{(4)} = \{(5.425, 3.375, 2.425, 1.375)\} $ (4-th row)
Figure 12.  Scattering of plane wave by a multi-layer heterogeneous medium: comparison of relative projection and NIMOR errors for $ {\bf E} $ (left) and $ {\bf H} $ (right) four the testing parameters
Table 1.  Comparison of the number of sampling points using Smolyak sparse gird and full tensor product grid. Here $ M_1 $ is the number of sampling points using Smolyak sparse gird, and $ M_2 $ is the number of sampling points using full tensor product grid
Dimension size $ p $ Approximation level $ l $ $ M_1 $ $ M_2 $ $ \dfrac{M_2}{M_1} $
2 13 25 1.923
2 3 29 81 2.793
5 145 1089 7.510
2 61 3125 51.230
5 3 241 59049 $ 2.450\times10^2 $
5 2433 $ 3.914\times10^7 $ $ 1.609\times10^4 $
2 221 $ 9.766\times10^6 $ $ 4.419\times10^4 $
10 3 1581 $ 3.487\times10^9 $ $ 2.206\times10^6 $
5 41265 $ 1.532\times10^{15} $ $ 3.713\times10^{10} $
Dimension size $ p $ Approximation level $ l $ $ M_1 $ $ M_2 $ $ \dfrac{M_2}{M_1} $
2 13 25 1.923
2 3 29 81 2.793
5 145 1089 7.510
2 61 3125 51.230
5 3 241 59049 $ 2.450\times10^2 $
5 2433 $ 3.914\times10^7 $ $ 1.609\times10^4 $
2 221 $ 9.766\times10^6 $ $ 4.419\times10^4 $
10 3 1581 $ 3.487\times10^9 $ $ 2.206\times10^6 $
5 41265 $ 1.532\times10^{15} $ $ 3.713\times10^{10} $
Table 2.  Scattering of plane wave by a dielectric disk: settings for the training, and testing datasets
Data set Training set Testing set
Parameter sample points 65, uneven (Smolyak mehod) 40, random (LHS method)
Time sample points 263, uniform 263, uniform
Size 17095 10520
Data set Training set Testing set
Parameter sample points 65, uneven (Smolyak mehod) 40, random (LHS method)
Time sample points 263, uniform 263, uniform
Size 17095 10520
Table 3.  Scattering of plane wave by a dielectric disk: the average projection and NIMOR errors on the testing set
Average relative errors $ \overline{e}_{ {\bf E}, {\rm{Pro}}} $ $ \overline{e}_{ {\bf E}, {\rm{NIMOR}}} $ $ \overline{e}_{ {\bf H}, {\rm{Pro}}} $ $ \overline{e}_{ {\bf H}, {\rm{NIMOR}}} $
$ {\rm{NIMOR}}^{(1)} {\rm{method}} $ $ 1.170\times 10^{-2} $ $ 1.500\times 10^{-2} $ $ 1.065\times 10^{-2} $ $ 1.447\times 10^{-2} $
$ {\rm{NIMOR}}^{(2)} {\rm{method}} $ $ 1.173\times 10^{-2} $ $ 1.279\times 10^{-2} $ $ 1.069\times 10^{-2} $ $ 1.198\times 10^{-2} $
Average relative errors $ \overline{e}_{ {\bf E}, {\rm{Pro}}} $ $ \overline{e}_{ {\bf E}, {\rm{NIMOR}}} $ $ \overline{e}_{ {\bf H}, {\rm{Pro}}} $ $ \overline{e}_{ {\bf H}, {\rm{NIMOR}}} $
$ {\rm{NIMOR}}^{(1)} {\rm{method}} $ $ 1.170\times 10^{-2} $ $ 1.500\times 10^{-2} $ $ 1.065\times 10^{-2} $ $ 1.447\times 10^{-2} $
$ {\rm{NIMOR}}^{(2)} {\rm{method}} $ $ 1.173\times 10^{-2} $ $ 1.279\times 10^{-2} $ $ 1.069\times 10^{-2} $ $ 1.198\times 10^{-2} $
Table 4.  Scattering of plane wave by a dielectric disk: computational times of $ {\rm{NIMOR}}^{(i)} $ ($ i = 1, 2 $) and DGTD methods in terms of CPU time. The unit of time cost is second
Method Offlin stage
(Snapshots, Nested POD, GRP training)
Online stage
(one run for new paramter)
DGTD - $ 4.254\times10^{2} $
$ {\rm{NIMOR}}^{(1)} $ $ 1.444\times10^{4} $ 3.8
$ {\rm{NIMOR}}^{(2)} $ $ 2.793\times10^{4} $ 3.1
Method Offlin stage
(Snapshots, Nested POD, GRP training)
Online stage
(one run for new paramter)
DGTD - $ 4.254\times10^{2} $
$ {\rm{NIMOR}}^{(1)} $ $ 1.444\times10^{4} $ 3.8
$ {\rm{NIMOR}}^{(2)} $ $ 2.793\times10^{4} $ 3.1
Table 5.  Scattering of a plane wave by a multi-layer heterogeneous medium: the distribution and range of material parameters
Layer $ i $ $ \mathcal{P}^{(i)} $ $ \mu_{r, i} $ $ r_i $
1 $ \varepsilon_{r, 1}\in[5.0, 5.6] $ 1 0.15
2 $ \varepsilon_{r, 2}\in[3.25, 3.75] $ 1 0.3
3 $ \varepsilon_{r, 3}\in[2.0, 2.5] $ 1 0.45
4 $ \varepsilon_{r, 4}\in[1.25, 1.75] $ 1 0.6
Layer $ i $ $ \mathcal{P}^{(i)} $ $ \mu_{r, i} $ $ r_i $
1 $ \varepsilon_{r, 1}\in[5.0, 5.6] $ 1 0.15
2 $ \varepsilon_{r, 2}\in[3.25, 3.75] $ 1 0.3
3 $ \varepsilon_{r, 3}\in[2.0, 2.5] $ 1 0.45
4 $ \varepsilon_{r, 4}\in[1.25, 1.75] $ 1 0.6
Table 6.  Scattering of plane wave by a multi-layer heterogeneous medium: settings for the training, and testing datasets
Data set Training set Testing set
Parameter sample points 137, uneven (Smolyak mehod) 40, random (LHS method)
Time sample points 253, uniform 253, uniform
Size 34661 10120
Data set Training set Testing set
Parameter sample points 137, uneven (Smolyak mehod) 40, random (LHS method)
Time sample points 253, uniform 253, uniform
Size 34661 10120
Table 7.  Scattering of plane wave by a multi-layer heterogeneous medium: the average projection and NIMOR errors on the testing set
Average relative errors $ \overline{e}_{ {\bf E}, {\rm{Pro}}} $ $ \overline{e}_{ {\bf E}, {\rm{NIMOR}}} $ $ \overline{e}_{ {\bf H}, {\rm{Pro}}} $ $ \overline{e}_{ {\bf H}, {\rm{NIMOR}}} $
$ {\rm{NIMOR}}^{(1)} {\rm{method}} $ $ 3.668\times 10^{-3} $ $ 6.011\times 10^{-3} $ $ 4.419\times 10^{-3} $ $ 6.695\times 10^{-3} $
$ {\rm{NIMOR}}^{(2)} {\rm{method}} $ $ 3.665\times 10^{-3} $ $ 4.744\times 10^{-3} $ $ 4.405\times 10^{-3} $ $ 5.229\times 10^{-3} $
Average relative errors $ \overline{e}_{ {\bf E}, {\rm{Pro}}} $ $ \overline{e}_{ {\bf E}, {\rm{NIMOR}}} $ $ \overline{e}_{ {\bf H}, {\rm{Pro}}} $ $ \overline{e}_{ {\bf H}, {\rm{NIMOR}}} $
$ {\rm{NIMOR}}^{(1)} {\rm{method}} $ $ 3.668\times 10^{-3} $ $ 6.011\times 10^{-3} $ $ 4.419\times 10^{-3} $ $ 6.695\times 10^{-3} $
$ {\rm{NIMOR}}^{(2)} {\rm{method}} $ $ 3.665\times 10^{-3} $ $ 4.744\times 10^{-3} $ $ 4.405\times 10^{-3} $ $ 5.229\times 10^{-3} $
Table 8.  Scattering of plane wave by a multi-layer heterogeneous medium: computational times of $ {\rm{NIMOR}}^{(i)} $ ($ i = 1, 2 $) and DGTD methods in terms of CPU time. The unit of time cost is second
Method Offlin stage
(Snapshots, Nested POD, GRP training)
Online stage
(one run for new paramter)
DGTD - $ 4.513\times10^{2} $
$ {\rm{NIMOR}}^{(1)} $ $ 1.912\times10^{4} $ 3.9
$ {\rm{NIMOR}}^{(2)} $ $ 4.221\times10^{4} $ 3.3
Method Offlin stage
(Snapshots, Nested POD, GRP training)
Online stage
(one run for new paramter)
DGTD - $ 4.513\times10^{2} $
$ {\rm{NIMOR}}^{(1)} $ $ 1.912\times10^{4} $ 3.9
$ {\rm{NIMOR}}^{(2)} $ $ 4.221\times10^{4} $ 3.3
[1]

Gang Bao, Bin Hu, Peijun Li, Jue Wang. Analysis of time-domain Maxwell's equations in biperiodic structures. Discrete and Continuous Dynamical Systems - B, 2020, 25 (1) : 259-286. doi: 10.3934/dcdsb.2019181

[2]

Lars Grüne, Luca Mechelli, Simon Pirkelmann, Stefan Volkwein. Performance estimates for economic model predictive control and their application in proper orthogonal decomposition-based implementations. Mathematical Control and Related Fields, 2021, 11 (3) : 579-599. doi: 10.3934/mcrf.2021013

[3]

Charles L. Epstein, Leslie Greengard, Thomas Hagstrom. On the stability of time-domain integral equations for acoustic wave propagation. Discrete and Continuous Dynamical Systems, 2016, 36 (8) : 4367-4382. doi: 10.3934/dcds.2016.36.4367

[4]

Mojtaba F. Fathi, Ahmadreza Baghaie, Ali Bakhshinejad, Raphael H. Sacho, Roshan M. D'Souza. Time-resolved denoising using model order reduction, dynamic mode decomposition, and kalman filter and smoother. Journal of Computational Dynamics, 2020, 7 (2) : 469-487. doi: 10.3934/jcd.2020019

[5]

Jing Li, Panos Stinis. Model reduction for a power grid model. Journal of Computational Dynamics, 2022, 9 (1) : 1-26. doi: 10.3934/jcd.2021019

[6]

Oleg Yu. Imanuvilov, Masahiro Yamamoto. Calderón problem for Maxwell's equations in cylindrical domain. Inverse Problems and Imaging, 2014, 8 (4) : 1117-1137. doi: 10.3934/ipi.2014.8.1117

[7]

Meixin Xiong, Liuhong Chen, Ju Ming, Jaemin Shin. Accelerating the Bayesian inference of inverse problems by using data-driven compressive sensing method based on proper orthogonal decomposition. Electronic Research Archive, 2021, 29 (5) : 3383-3403. doi: 10.3934/era.2021044

[8]

Björn Birnir, Niklas Wellander. Homogenized Maxwell's equations; A model for ceramic varistors. Discrete and Continuous Dynamical Systems - B, 2006, 6 (2) : 257-272. doi: 10.3934/dcdsb.2006.6.257

[9]

Lu Zhao, Heping Dong, Fuming Ma. Time-domain analysis of forward obstacle scattering for elastic wave. Discrete and Continuous Dynamical Systems - B, 2021, 26 (8) : 4111-4130. doi: 10.3934/dcdsb.2020276

[10]

Khalid Latrach, Hatem Megdiche. Time asymptotic behaviour for Rotenberg's model with Maxwell boundary conditions. Discrete and Continuous Dynamical Systems, 2011, 29 (1) : 305-321. doi: 10.3934/dcds.2011.29.305

[11]

Amine Laghrib, Lekbir Afraites, Aissam Hadri, Mourad Nachaoui. A non-convex PDE-constrained denoising model for impulse and Gaussian noise mixture reduction. Inverse Problems and Imaging, , () : -. doi: 10.3934/ipi.2022031

[12]

M. Sumon Hossain, M. Monir Uddin. Iterative methods for solving large sparse Lyapunov equations and application to model reduction of index 1 differential-algebraic-equations. Numerical Algebra, Control and Optimization, 2019, 9 (2) : 173-186. doi: 10.3934/naco.2019013

[13]

S. S. Krigman. Exact boundary controllability of Maxwell's equations with weak conductivity in the heterogeneous medium inside a general domain. Conference Publications, 2007, 2007 (Special) : 590-601. doi: 10.3934/proc.2007.2007.590

[14]

Quyen Tran, Harbir Antil, Hugo Díaz. Optimal control of parameterized stationary Maxwell's system: Reduced basis, convergence analysis, and a posteriori error estimates. Mathematical Control and Related Fields, 2022  doi: 10.3934/mcrf.2022003

[15]

Corinna Burkard, Roland Potthast. A time-domain probe method for three-dimensional rough surface reconstructions. Inverse Problems and Imaging, 2009, 3 (2) : 259-274. doi: 10.3934/ipi.2009.3.259

[16]

Hongjun Gao, Fei Liang. On the stochastic beam equation driven by a Non-Gaussian Lévy process. Discrete and Continuous Dynamical Systems - B, 2014, 19 (4) : 1027-1045. doi: 10.3934/dcdsb.2014.19.1027

[17]

Yan Wang, Lei Wang, Yanxiang Zhao, Aimin Song, Yanping Ma. A stochastic model for microbial fermentation process under Gaussian white noise environment. Numerical Algebra, Control and Optimization, 2015, 5 (4) : 381-392. doi: 10.3934/naco.2015.5.381

[18]

W. Wei, H. M. Yin. Global solvability for a singular nonlinear Maxwell's equations. Communications on Pure and Applied Analysis, 2005, 4 (2) : 431-444. doi: 10.3934/cpaa.2005.4.431

[19]

Annalena Albicker, Roland Griesmaier. Monotonicity in inverse scattering for Maxwell's equations. Inverse Problems and Imaging, , () : -. doi: 10.3934/ipi.2022032

[20]

Kaiping Liu, Haitao Che, Haibin Chen, Meixia Li. Parameterized S-type M-eigenvalue inclusion intervals for fourth-order partially symmetric tensors and its applications. Journal of Industrial and Management Optimization, 2022  doi: 10.3934/jimo.2022077

2021 Impact Factor: 1.497

Metrics

  • PDF downloads (104)
  • HTML views (89)
  • Cited by (0)

[Back to Top]