A cusp-capturing PINN for elliptic interface problems

DOI10.1016/J.JCP.2023.112359MaRDI QIDQ6095097zbMATH OpenOpenAlexFDO

Authors Yu-Hau Tseng, Te-Sheng Lin, Wei-Fan Hu, Ming-Chih Lai

Publication date 27 November 2023

Published in Journal of Computational Physics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2210.08424

zbMATH Keywords

cusp capturingelliptic interface problemfluid structure interaction problemlevel set function

Mathematics Subject Classification ID

Elliptic equations and elliptic systems (35Jxx) Numerical methods for partial differential equations, boundary value problems (65Nxx) Artificial intelligence (68Txx)

Abstract: In this paper, we propose a cusp-capturing physics-informed neural network (PINN) to solve discontinuous-coefficient elliptic interface problems whose solution is continuous but has discontinuous first derivatives on the interface. To find such a solution using neural network representation, we introduce a cusp-enforced level set function as an additional feature input to the network to retain the inherent solution properties; that is, capturing the solution cusps (where the derivatives are discontinuous) sharply. In addition, the proposed neural network has the advantage of being mesh-free, so it can easily handle problems in irregular domains. We train the network using the physics-informed framework in which the loss function comprises the residual of the differential equation together with certain interface and boundary conditions. We conduct a series of numerical experiments to demonstrate the effectiveness of the cusp-capturing technique and the accuracy of the present network model. Numerical results show that even using a one-hidden-layer (shallow) network with a moderate number of neurons and sufficient training data points, the present network model can achieve prediction accuracy comparable with traditional methods. Besides, if the solution is discontinuous across the interface, we can simply incorporate an additional supervised learning task for solution jump approximation into the present network without much difficulty.

Recommendations

• A Discontinuity and Cusp Capturing PINN for Stokes Interface Problems with Discontinuous Viscosity and Singular Forces
• An immersed interface neural network for elliptic interface problems
• A discontinuity capturing shallow neural network for elliptic interface problems
• Interface PINNs (I-PINNs): a physics-informed neural networks framework for interface problems
• A mesh-free method using piecewise deep neural network for elliptic interface problems

Cites work

• A boundary condition capturing method for Poisson's equation on irregular domains
• A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output from a Computer Code
• A discontinuity capturing shallow neural network for elliptic interface problems
• A hybrid immersed boundary and immersed interface method for electrohydrodynamic simulations
• A mesh-free method for interface problems using the deep learning approach
• A mesh-free method using piecewise deep neural network for elliptic interface problems
• A shallow Ritz method for elliptic problems with singular sources
• A simple implementation of the immersed interface methods for Stokes flows with singular forces
• A Simple Mesh Generator in MATLAB
• An efficient meshfree method based on Pascal polynomials and multiple-scale approach for numerical solution of 2-D and 3-D second order elliptic interface problems
• Application of a level set method for simulation of droplet collisions
• Approximation by superpositions of a sigmoidal function
• Controlling droplet bouncing and coalescence with surfactant
• Deep Nitsche Method: Deep Ritz Method with Essential Boundary Conditions
• Deep unfitted Nitsche method for elliptic interface problems
• Evaluating Derivatives
• Fitted and Unfitted Finite-Element Methods for Elliptic Equations with Smooth Interfaces
• scientific article; zbMATH DE number 3579922 (Why is no real title available?)
• scientific article; zbMATH DE number 1405266 (Why is no real title available?)
• INN: interfaced neural networks as an accessible meshless approach for solving interface PDE problems
• Local meshless methods for second order elliptic interface problems with sharp corners
• Multilayer feedforward networks are universal approximators
• Numerical analysis of blood flow in the heart
• On the convergence of physics informed neural networks for linear second-order elliptic and parabolic type PDEs
• On the limited memory BFGS method for large scale optimization
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Properties of Discrete Delta Functions and Local Convergence of the Immersed Boundary Method
• Solving elliptic interface problems with jump conditions on Cartesian grids
• The deep Ritz method: a deep learning-based numerical algorithm for solving variational problems
• The immersed boundary method
• The Immersed Interface Method for Elliptic Equations with Discontinuous Coefficients and Singular Sources
• Three-dimensional elliptic solvers for interface problems and applications
• xGFM: recovering convergence of fluxes in the ghost fluid method

Cited in

(27)

• A Discontinuity and Cusp Capturing PINN for Stokes Interface Problems with Discontinuous Viscosity and Singular Forces
• CEENs: causality-enforced evolutional networks for solving time-dependent partial differential equations
• A hybrid neural-network and MAC scheme for Stokes interface problems
• Interface PINNs (I-PINNs): a physics-informed neural networks framework for interface problems
• Solving parametric elliptic interface problems via interfaced operator network
• Phase field smoothing-PINN: a neural network solver for partial differential equations with discontinuous coefficients
• On physics-informed neural networks training for coupled hydro-poromechanical problems
• An accurate and efficient continuity-preserved method based on randomized neural networks for elliptic interface problems
• An immersed interface neural network for elliptic interface problems
• Energy-based adaptive deep unfitted Nitsche method for elliptic interface problems
• IG-PINNs: interface-gated physics-informed neural networks for solving elliptic interface problems
• Adaptive pseudo-solution-assisted PINNs for multiscale Darcy and Darcy-Forchheimer equations
• Subspace method based on neural networks for solving the partial differential equation
• DENNs: discontinuity-embedded neural networks for fracture mechanics
• XI-DeepONet: an operator learning method for elliptic interface problems
• Adaptive deep physics-informed neural network with dual-nested activation for solving complex partial differential equations
• Dirichlet-Neumann learning algorithm for solving elliptic interface problems
• The hard-constraint PINNs for interface optimal control problems
• A multiple transferable neural network method with domain decomposition for elliptic interface problems
• General enrichments of stable GFEM for interface problems: theory and extreme learning machine construction
• Accurate adaptive deep learning method for solving elliptic problems
• Novel and general discontinuity-removing PINNs for elliptic interface problems
• Extended physics-informed extreme learning machine for linear elastic fracture mechanics
• Two novel discontinuity-removing PINNs for solving variable coefficient elliptic interface problems on curved surfaces
• Continuity-preserved deep learning method for solving elliptic interface problems
• BiLO: bilevel local operator learning for PDE inverse problems
• A novel singularity- and discontinuity-capturing PINN for time-fractional diffusion equations involving initial singularities and interfaces on complex curved surfaces

This page was built for publication: A cusp-capturing PINN for elliptic interface problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6095097)