A Machine-Learning Method for Time-Dependent Wave Equations over Unbounded Domains

Time-dependent wave equations represent an important class of partial differential equations (PDE) for describing wave propagation phenomena, which are often formulated over unbounded domains. Given a compactly supported initial condition, classical numerical methods reduce such problems to bounded domains using artificial boundary condition (ABC). In this work, we present a machine-learning method to solve this type of equations as an alternative to ABCs. Specifically, the mapping from the initial conditions to the PDE solution is represented by a neural network, trained using wave packets that are parameterized by their band width and wave numbers. The accuracy is tested for both the second-order wave equation and the Schrödinger equation, including the nonlinear Schrödinger equation. We examine the accuracy from both interpolations and extrapolations. For initial conditions lying in the training set, the learned map has good interpolation accuracy, due to the approximation property of deep neural networks. The learned map also exhibits some good extrapolation accuracy. We also demonstrate the effectiveness of the method for problems in irregular domains. Overall, the proposed method provides an interesting alternative for finite-time simulation of wave propagation.

[1]  G. M.,et al.  Partial Differential Equations I , 2023, Applied Mathematical Sciences.

[2]  A. Majda,et al.  Absorbing boundary conditions for the numerical simulation of waves , 1977 .

[3]  M. Gregory Forest,et al.  Geometry and Modulation Theory for the Periodic Nonlinear Schrodinger Equation , 1986 .

[4]  Gérard A. Maugin,et al.  Electrodynamics of Continua I: Foundations and Solid Media , 1989 .

[5]  H. Yoshida.  Construction of higher order symplectic integrators , 1990 .

[6]  Shibata.  Absorbing boundary conditions for the finite-difference time-domain calculation of the one-dimensional Schrödinger equation. , 1991, Physical review. B, Condensed matter.

[7]  Jean-Pierre Berenger,et al.  A perfectly matched layer for the absorption of electromagnetic waves , 1994 .

[8]  Weizhu Bao,et al.  AN ARTIFICIAL BOUNDARY CONDITION FOR TWO‐DIMENSIONAL INCOMPRESSIBLE VISCOUS FLOWS USING THE METHOD OF LINES , 1996 .

[9]  Dimitrios I. Fotiadis,et al.  Artificial neural networks for solving ordinary and partial differential equations , 1997, IEEE Trans. Neural Networks.

[10]  C. Dafermos.  Hyberbolic Conservation Laws in Continuum Physics , 2000 .

[11]  John L. Tassoulas,et al.  CONTINUED-FRACTION ABSORBING BOUNDARY CONDITIONS FOR THE WAVE EQUATION , 2000 .

[12]  W. E,et al.  Matching conditions in atomistic-continuum modeling of materials. , 2001, Physical review letters.

[13]  Patrick Joly,et al.  FICTITIOUS DOMAINS, MIXED FINITE ELEMENTS AND PERFECTLY MATCHED LAYERS FOR 2-D ELASTIC WAVE PROPAGATION , 2001 .

[14]  D K Smith,et al.  Numerical Optimization , 2001, J. Oper. Res. Soc..

[15]  P. Markowich,et al.  On time-splitting spectral approximations for the Schrödinger equation in the semiclassical regime , 2002 .

[16]  Matthias Ehrhardt,et al.  Discrete transparent boundary conditions for the Schrödinger equation: fast calculation, approximation, and stability , 2003 .

[17]  王东东,et al.  Computer Methods in Applied Mechanics and Engineering , 2004 .

[18]  Leslie Greengard,et al.  Fast evaluation of nonreflecting boundary conditions for the Schrödinger equation in one dimension , 2004 .

[19]  Xiantao Li,et al.  Variational Boundary Conditions for Molecular Dynamics Simulations of Solids at Low Temperature , 2005 .

[20]  Gang Bao,et al.  Convergence Analysis of the Perfectly Matched Layer Problemsfor Time-Harmonic Maxwell's Equations , 2005, SIAM J. Numer. Anal..

[21]  Eduard G. Karpov,et al.  A Green's function approach to deriving non‐reflecting boundary conditions in molecular dynamics simulations , 2005 .

[22]  Julien Diaz,et al.  A time domain analysis of PML models in acoustics , 2006 .

[23]  Takashi Nakatsukasa,et al.  Real‐time, real‐space implementation of the linear response time‐dependent density‐functional theory , 2006 .

[24]  Christophe Besse,et al.  A Review of artificial boundary conditions for the Schrödinger equation , 2007 .

[25]  H. Robbins.  A Stochastic Approximation Method , 1951 .

[26]  Dan Givoli,et al.  Computational Absorbing Boundaries , 2008 .

[27]  S. Marburg,et al.  Computational acoustics of noise propagation in fluids : finite and boudary element methods , 2008 .

[28]  William E. Schiesser,et al.  Linear and nonlinear waves , 2009, Scholarpedia.

[29]  S. Goldsack,et al.  IN REAL-TIME , 2008 .

[30]  Christophe Besse,et al.  Communi-cations Computational methods for the dynamics of the nonlinear Schr̈odinger / Gross-Pitaevskii equations , 2013 .

[31]  Weizhong Dai,et al.  A generalized finite-difference time-domain scheme for solving nonlinear Schrödinger equations , 2013, Comput. Phys. Commun..

[32]  Xiaonan Wu,et al.  Artificial Boundary Method , 2013 .

[33]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[34]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[35]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[36]  E Weinan,et al.  The Deep Ritz Method: A Deep Learning-Based Numerical Algorithm for Solving Variational Problems , 2017, Communications in Mathematics and Statistics.

[37]  E Weinan,et al.  Deep Learning-Based Numerical Methods for High-Dimensional Parabolic Partial Differential Equations and Backward Stochastic Differential Equations , 2017, Communications in Mathematics and Statistics.

[38]  Arnulf Jentzen,et al.  Solving high-dimensional partial differential equations using deep learning , 2017, Proceedings of the National Academy of Sciences.

[39]  Jiwei Zhang,et al.  An Efficient Second-Order Finite Difference Method for the One-Dimensional Schrödinger Equation with Absorbing Boundary Conditions , 2018, SIAM J. Numer. Anal..

[40]  Jorge Nocedal,et al.  Optimization Methods for Large-Scale Machine Learning , 2016, SIAM Rev..

[41]  Justin A. Sirignano,et al.  DGM: A deep learning algorithm for solving partial differential equations , 2017, J. Comput. Phys..

[42]  He Ming Yao,et al.  Machine-Learning-Based PML for the FDTD Method , 2019, IEEE Antennas and Wireless Propagation Letters.

[43]  E Weinan,et al.  Machine Learning Approximation Algorithms for High-Dimensional Fully Nonlinear Partial Differential Equations and Second-order Backward Stochastic Differential Equations , 2017, J. Nonlinear Sci..

[44]  George Em Karniadakis,et al.  Systems biology informed deep learning for inferring parameters and hidden dynamics , 2019, bioRxiv.

[45]  Paris Perdikaris,et al.  Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations , 2019, J. Comput. Phys..

[46]  Philippe von Wurstemberger,et al.  Overcoming the curse of dimensionality in the approximative pricing of financial derivatives with default risks , 2019, Electronic Journal of Probability.

[47]  Xiantao Li,et al.  Absorbing boundary conditions for the time-dependent Schrödinger-type equations in R 3 . , 2019 .

[48]  Yingshi Chen,et al.  Learning Unsplit-field-based PML for the FDTD Method by Deep Differentiable Forest , 2020, ArXiv.

[49]  Wei Cai,et al.  A Phase Shift Deep Neural Network for High Frequency Approximation and Wave Problems , 2019, SIAM Journal on Scientific Computing.

[50]  Xiaotao Li,et al.  Absorbing boundary conditions for the time-dependent Schrödinger-type equations in R3 , 2019, Physical review. E.

[51]  L. Dal Negro,et al.  Physics-informed neural networks for inverse problems in nano-optics and metamaterials. , 2019, Optics express.

[52]  W. E,et al.  Machine Learning and Computational Mathematics , 2020, Communications in Computational Physics.

[53]  Lexing Ying,et al.  Solving parametric PDE problems with artificial neural networks , 2017, European Journal of Applied Mathematics.

[54]  Philippe von Wurstemberger,et al.  Numerical simulations for full history recursive multilevel Picard approximations for systems of high-dimensional partial differential equations , 2020, Communications in Computational Physics.

[55]  Gang Bao,et al.  Weak Adversarial Networks for High-dimensional Partial Differential Equations , 2019, J. Comput. Phys..

[56]  George Em Karniadakis,et al.  Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations , 2020, Science.

[57]  He Ming Yao,et al.  Enhanced PML Based on the Long Short Term Memory Network for the FDTD Method , 2020, IEEE Access.

[58]  global sci,et al.  Practical Absorbing Boundary Conditions for Wave Propagation on Arbitrary Domain , 2020, Advances in Applied Mathematics and Mechanics.

[59]  George Em Karniadakis,et al.  NSFnets (Navier-Stokes flow nets): Physics-informed neural networks for the incompressible Navier-Stokes equations , 2020, J. Comput. Phys..

[60]  Liu Yang,et al.  B-PINNs: Bayesian Physics-Informed Neural Networks for Forward and Inverse PDE Problems with Noisy Data , 2020, J. Comput. Phys..

[61]  Minxin Chen,et al.  MIM: A deep mixed residual method for solving high-order partial differential equations , 2020, J. Comput. Phys..