Solving a class of multi-scale elliptic PDEs by means of Fourier-based mixed physics informed neural networks
Preprint
Abstract
Deep neural networks have garnered widespread attention due to their simplicity and flexibility in the fields of engineering and scientific calculation. In this study, we probe into solving a class of elliptic partial differential equations(PDEs) with multiple scales by utilizing Fourier-based mixed physics informed neural networks(dubbed FMPINN), its solver is configured as a multi-scale deep neural network. In contrast to the classical PINN method, a dual (flux) variable about the rough coefficient of PDEs is introduced to avoid the ill-condition of neural tangent kernel matrix caused by the oscillating coefficient of multi-scale PDEs. Therefore, apart from the physical conservation laws, the discrepancy between the auxiliary variables and the gradients of multi-scale coefficients is incorporated into the cost function, then obtaining a satisfactory solution of PDEs by minimizing the defined loss through some optimization methods. Additionally, a trigonometric activation function is introduced for FMPINN, which is suited for representing the derivatives of complex target functions. Handling the input data by Fourier feature mapping will effectively improve the capacity of deep neural networks to solve high-frequency problems. Finally, to validate the efficiency and robustness of the proposed FMPINN algorithm, we present several numerical examples of multi-scale problems in various dimensional Euclidean spaces. These examples cover both low-frequency and high-frequency oscillation cases, demonstrating the effectiveness of our approach. All code and data accompanying this manuscript will be made publicly available at \href{https://github.com/Blue-Giant/FMPINN}{https://github.com/Blue-Giant/FMPINN}.Keywords
All Related Versions
This publication has 52 references indexed in Scilit:
- A Priori Error Estimate of Deep Mixed Residual Method for Elliptic PDEsJournal of Scientific Computing, 2024
- Adaptive multi-scale neural network with Resnet blocks for solving partial differential equationsNonlinear Dynamics, 2022
- A deep First-Order System Least Squares method for solving elliptic PDEsComputers & Mathematics with Applications, 2022
- A deep domain decomposition method based on Fourier featuresJournal of Computational and Applied Mathematics, 2022
- The Deep Ritz Method: A Deep Learning-Based Numerical Algorithm for Solving Variational ProblemsCommunications in Mathematics and Statistics, 2018
- A Tutorial on Latin Hypercube Design of ExperimentsQuality and Reliability Engineering International, 2015
- An active set limited memory BFGS algorithm for bound constrained optimizationApplied Mathematical Modelling, 2011
- Flux Norm Approach to Finite Dimensional Homogenization Approximations with Non-Separated Scales and High ContrastArchive for Rational Mechanics and Analysis, 2010
- Adaptive variational multiscale methods based on a posteriori error estimation: Energy norm estimates for elliptic problemsComputer Methods in Applied Mechanics and Engineering, 2007
- Neural‐network‐based approximations for solving partial differential equationsCommunications in Numerical Methods in Engineering, 1994