arXiv subjects

Matthew J. Zahr

Publications and source records attributed to Matthew J. Zahr.

At least 19 recordsLinked to original sources

arXiv2025

Neural network-based Godunov corrections for approximate Riemann solvers using bi-fidelity learning

The Riemann problem is fundamental in the computational modeling of hyperbolic partial differential equations, enabling the development of stable and accurate upwind schemes. While exact solvers provide robust upwinding fluxes, their high computational cost necessitates approximate solvers. Although approximate solvers achieve accuracy in many scenarios, they produce inaccurate solutions in certain cases. To overcome this limitation, we propose constructing neural network-based surrogate models, trained using supervised learning, designed to map interior and exterior conservative state variables to the corresponding exact flux. Specifically, we propose two distinct approaches: one utilizing a vanilla neural network and the other employing a bi-fidelity neural network. The performance of the proposed approaches is demonstrated through applications to one-dimensional and two-dimensional partial differential equations, showcasing their robustness and accuracy.

math.NA↗

arXiv2025

An $rp$-adaptive method for accurate resolution of shock-dominated viscous flow based on implicit shock tracking

This work introduces an optimization-based $rp$-adaptive numerical method to approximate solutions of viscous, shock-dominated flows using implicit shock tracking and a high-order discontinuous Galerkin discretization on traditionally coarse grids without nonlinear stabilization (e.g., artificial viscosity or limiting). The proposed method adapts implicit shock tracking methods, originally developed to align mesh faces with solution discontinuities, to compress elements into viscous shocks and boundary layers, functioning as a novel approach to aggressive $r$-adaptation. This form of $r$-adaptation is achieved naturally as the minimizer of the enriched residual with respect to the discrete flow variables and coordinates of the nodes of the grid. Several innovations to the shock tracking optimization solver are proposed to ensure sufficient mesh compression at viscous features to render stabilization unnecessary, including residual weighting, step constraints and modifications, and viscosity-based continuation. Finally, $p$-adaptivity is used to locally increase the polynomial degree with three clear benefits: (1) lessens the mesh compression requirements near shock waves and boundary layers, (2) reduces the error in regions where $r$-adaptivity is not sufficient with the given grid topology, and (3) reduces computational cost by performing a majority of the $r$-adaptivity iterations on the coarsest discretization. A series of numerical experiments show the proposed method effectively resolves viscous, shock-dominated flows, including accurate prediction of heat flux profiles produced by hypersonic flow over a cylinder, and compares favorably in terms of accuracy per degree of freedom to $h$-adaptation with a high-order discretization.

math.NA↗

arXiv2025

Model reduction of convection-dominated viscous conservation laws using implicit feature tracking and landmark image registration

Reduced-order models (ROMs) remain generally unreliable for convection-dominated problems, such as those encountered in hypersonic flows, due to the slowly decaying Kolmogorov $n$-width of linear subspace approximations, known as the Kolmogorov barrier. This limitation hinders the accuracy of traditional ROMs and necessitates impractical amounts of training data during the offline phase. To address this challenge, we introduce a novel landmark-based registration procedure tailored for ROMs of convection-dominated problems. Our approach leverages limited training data and incorporates a nonlinear transformation of the data using a landmark-based registration technique combined with radial basis function (RBF) interpolation. During the offline phase, we align dominant convective features in a reference domain, resulting in a rapid decay of error relative to the reduced space dimension. Landmarks are generated through a three-step process: (1) detecting shocks via edge detection techniques, (2) sampling using Monte Carlo methods, and (3) domain partitioning with $k$-means clustering, where cluster centroids serve as landmarks. Accurate landmark correspondence is achieved by minimizing pairing distances for similar features. The online phase integrates standard minimum-residual ROM methodologies, extending the optimization space to include admissible domain mappings. We validate our approach on two test cases: a space-time Burgers' equation parameterized by the initial condition, and a hypersonic viscous flow over a cylinder parameterized by the Mach number. Results demonstrate the efficacy of the proposed method in overcoming the Kolmogorov barrier and enhancing the reliability of ROMs for convection-dominated problems.

math.NA↗

arXiv2025

A sharp-interface discontinuous Galerkin method for simulation of two-phase flow of real gases based on implicit shock tracking

We present a high-order, sharp-interface method for simulation of two-phase flow of real gases using implicit shock tracking. The method is based on a phase-field formulation of two-phase, compressible, inviscid flow with a trivial mixture model. Implicit shock tracking is a high-order, optimization-based discontinuous Galerkin method that automatically aligns mesh faces with non-smooth flow features to represent them perfectly with inter-element jumps. It is used to accurately approximate shocks and rarefactions without stabilization and converge the phase-field solution to a sharp interface one by aligning mesh faces with the material interface. Time-dependent problems are formulated as steady problems in a space-time domain where complex wave interactions (e.g., intersections and reflections) manifest as space-time triplet points. The space-time formulation avoids complex re-meshing and solution transfer that would be required to track moving waves with mesh faces using the method of lines. The approach is applied to several two-phase flow Riemann problems involving gases with ideal, stiffened gas, and Becker-Kistiakowsky-Wilson (BKW) equations of state, including a spherically symmetric underwater explosion problem. In all cases, the method aligns element faces with all shocks (including secondary shocks that form at time t > 0), rarefactions, and material interfaces, and accurately resolves the flow field on coarse space-time grids.

physics.flu-dyn↗

arXiv2025

An implicit shock tracking method for simulation of shock-dominated flows over complex domains using mesh-based parametrizations

A mesh-based parametrization is a parametrization of a geometric object that is defined solely from a mesh of the object, e.g., without an analytical expression or computer-aided design (CAD) representation of the object. In this work, we propose a mesh-based parametrization of an arbitrary $d'$-dimensional object embedded in a $d$-dimensional space using tools from high-order finite elements. Using mesh-based parametrizations, we construct a boundary-preserving parametrization of the nodal coordinates of a computational mesh that ensures all nodes remain on all their original boundaries. These boundary-preseving parametrizations allow the nodes of the mesh to move only in ways that will not change the computational domain. They also ensure nodes will not move between boundaries, which would cause issues assigning boundary conditions for partial differential equation simulations and lead to inaccurate geometry representations for non-smooth boundary transitions. Finally, we integrate boundary-preserving, mesh-based parametrizations into high-order implicit shock tracking, an optimization-based discontinuous Galerkin method that moves nodes to align mesh faces with non-smooth flow features to represent them perfectly with inter-element jumps, leaving the intra-element polynomial basis to represent smooth regions of the flow with high-order accuracy. Mesh-based parametrizations enable implicit shock tracking simulations of shock-dominated flows over geometries without simple analytical parametrizations. Several demonstrations of mesh-based parametrizations are provided.

physics.comp-ph↗

arXiv2025

Symmetric, Optimization-based, Cross-element Compatible Nodal Distributions for High-order Finite Elements

We present a general framework to construct symmetric, well-conditioned, cross-element compatible nodal distributions that can be used for high-order and high-dimensional finite elements. Starting from the inherent symmetries of an element geometry, we construct node groups in a systematic and efficient manner utilizing the natural coordinates of each element, while ensuring nodes stay within the elements. Proper constraints on the symmetry group lead to nodal distributions that ensure cross-element compatibility (i.e., nodes of adjacent elements are co-located) on both homogeneous and mixed meshes. The final nodal distribution is defined as a minimizer of an optimization problem over symmetry group parameters with linear constraints that ensure nodes remain with an element and enforce other properties (e.g., cross-element compatibility). We demonstrate the merit of this framework by comparing the proposed optimization-based nodal distributions with other popular distributions available in the literature, and its robustness by generating optimized nodal distributions for otherwise difficult elements (such as simplex and pyramid elements). All nodal distributions are tabulated in the optnodes package [22].

math.NA↗

arXiv2024

Preconditioned iterative solvers for constrained high-order implicit shock tracking methods

High-order implicit shock tracking (fitting) is a class of high-order numerical methods that use numerical optimization to simultaneously compute a high-order approximation to a conservation law solution and align elements of the computational mesh with non-smooth features. This alignment ensures that non-smooth features are perfectly represented by inter-element jumps and high-order basis functions approximate smooth regions of the solution without nonlinear stabilization, which leads to accurate approximations on traditionally coarse meshes. In this work, we devise a family of preconditioners for the saddle point linear system that defines the step toward optimality at each iteration of the optimization solver so Krylov solvers can be effectively used. Our preconditioners integrate standard preconditioners from constrained optimization with popular preconditioners for discontinuous Galerkin discretizations such as block Jacobi, block incomplete LU factorizations with minimum discarded fill reordering, and p-multigrid. Thorough studies are performed using two inviscid compressible flow problems to evaluate the effectivity of each preconditioner in this family and their sensitivity to critical shock tracking parameters such as the mesh and Hessian regularization, linearization state, and resolution of the solution space.

math.NA↗

arXiv2024

An augmented Lagrangian trust-region method with inexact gradient evaluations to accelerate constrained optimization problems using model hyperreduction

We present an augmented Lagrangian trust-region method to efficiently solve constrained optimization problems governed by large-scale nonlinear systems with application to partial differential equation-constrained optimization. At each major augmented Lagrangian iteration, the expensive optimization subproblem involving the full nonlinear system is replaced by an empirical quadrature-based hyperreduced model constructed on-the-fly. To ensure convergence of these inexact augmented Lagrangian subproblems, we develop a bound-constrained trust-region method that allows for inexact gradient evaluations, and specialize it to our specific setting that leverages hyperreduced models. This approach circumvents a traditional training phase because the models are built on-the-fly in accordance with the requirements of the trust-region convergence theory. Two numerical experiments (constrained aerodynamic shape design) demonstrate the convergence and efficiency of the proposed work. A speedup of 12.7x (for all computational costs, even costs traditionally considered "offline" such as snapshot collection and data compression) relative to a standard optimization approach that does not leverage model reduction is shown.

math.OC↗

arXiv2024

A space-time high-order implicit shock tracking method for shock-dominated unsteady flows

High-order implicit shock tracking (fitting) is a class of high-order, optimization-based numerical methods to approximate solutions of conservation laws with non-smooth features by aligning elements of the computational mesh with non-smooth features. This ensures the non-smooth features are perfectly represented by inter-element jumps and high-order basis functions approximate smooth regions of the solution without nonlinear stabilization, which leads to accurate approximations on traditionally coarse meshes. In this work, we extend implicit shock tracking to time-dependent problems using a slab-based space-time approach. This is achieved by reformulating a time-dependent conservation law as a steady conservation law in one higher dimension and applying existing implicit shock tracking techniques. To avoid computations over the entire time domain and unstructured mesh generation in higher dimensions, we introduce a general procedure to generate conforming, simplex-only meshes of space-time slabs in such a way that preserves features (e.g., curved elements, refinement regions) from previous time slabs. The use of space-time slabs also simplifies the shock tracking problem by reducing temporal complexity. Several practical adaptations of the implicit shock tracking solvers are developed for the space-time setting including 1) a self-adjusting temporal boundary, 2) nondimensionalization of a space-time slab, 3) adaptive mesh refinement, and 4) shock boundary conditions, which lead to accurate solutions on coarse space-time grids, even for problem with complex flow features such as curved shocks, shock formation, shock-shock and shock-boundary interaction, and triple points.

math.NA↗

arXiv2023

An adaptive model reduction method leveraging locally supported basis functions

We propose a new method, the continuous Galerkin method with globally and locally supported basis functions (CG-GL), to address the parametric robustness issues of reduced-order models (ROMs) by incorporating solution-based adaptivity with locally supported finite element basis functions. The CG-GL method combines the accuracy of locally supported basis functions with the efficiency of globally supported data-driven basis functions. Efficient output-based dual-weighted residual error estimates are derived and implemented for the CG-GL method and used to drive efficient online trial space adaptation. An empirical quadrature procedure is introduced for rapid evaluation of nonlinear terms that does not require retraining throughout the adaptation process. Two numerical experiments demonstrate the potential of the CG-GL method to produce accurate approximations with limited training and its tunable tradeoff between accuracy and computational cost.

math.NA↗

arXiv2023

An adaptive, training-free reduced-order model for convection-dominated problems based on hybrid snapshots

The vast majority of reduced-order models (ROMs) first obtain a low dimensional representation of the problem from high-dimensional model (HDM) training data which is afterwards used to obtain a system of reduced complexity. Unfortunately, convection-dominated problems generally have a slowly decaying Kolmogorov n-width, which makes obtaining an accurate ROM built solely from training data very challenging. The accuracy of a ROM can be improved through enrichment with HDM solutions; however, due to the large computational expense of HDM evaluations for complex problems, they can only be used parsimoniously to obtain relevant computational savings. In this work, we exploit the local spatial and temporal coherence often exhibited by these problems to derive an accurate, cost-efficient approach that repeatedly combines HDM and ROM evaluations without a separate training phase. Our approach obtains solutions at a given time step by either fully solving the HDM or by combining partial HDM and ROM solves. A dynamic sampling procedure identifies regions that require the HDM solution for global accuracy and the reminder of the flow is reconstructed using the ROM. Moreover, solutions combining both HDM and ROM solves use spatial filtering to eliminate potential spurious oscillations that may develop. We test the proposed method on inviscid compressible flow problems and demonstrate speedups up to an order of magnitude.

math.NA↗

arXiv2023

Accelerated solutions of convection-dominated partial differential equations using implicit feature tracking and empirical quadrature

This work introduces an empirical quadrature-based hyperreduction procedure and greedy training algorithm to effectively reduce the computational cost of solving convection-dominated problems with limited training. The proposed approach circumvents the slowly decaying $n$-width limitation of linear model reduction techniques applied to convection-dominated problems by using a nonlinear approximation manifold systematically defined by composing a low-dimensional affine space with bijections of the underlying domain. The reduced-order model is defined as the solution of a residual minimization problem over the nonlinear manifold. An online-efficient method is obtained by using empirical quadrature to approximate the optimality system such that it can be solved with mesh-independent operations. The proposed reduced-order model is trained using a greedy procedure to systematically sample the parameter domain. The effectiveness of the proposed approach is demonstrated on two shock-dominated computational fluid dynamics benchmarks.

math.NA↗

arXiv2023

High-order implicit shock tracking boundary conditions for flows with parametrized shocks

High-order implicit shock tracking (fitting) is a class of high-order, optimization-based numerical methods to approximate solutions of conservation laws with non-smooth features by aligning elements of the computational mesh with non-smooth features. This ensures the non-smooth features are perfectly represented by inter-element jumps and high-order basis functions approximate smooth regions of the solution without nonlinear stabilization, which leads to accurate approximations on traditionally coarse meshes. In this work, we introduce a robust implicit shock tracking framework specialized for problems with parameter-dependent lead shocks (i.e., shocks separating a farfield condition from the downstream flow), which commonly arise in high-speed aerodynamics and astrophysics applications. After a shock-aligned mesh is produced at one parameter configuration, all elements upstream of the lead shock are removed and the nodes on the lead shock are positioned for new parameter configurations using the implicit shock tracking solver. The proposed framework can be used for most many-query applications involving parametrized lead shocks such as optimization, uncertainty quantification, parameter sweeps, "what-if" scenarios, or parameter-based continuation. We demonstrate the robustness and flexibility of the framework using a one-dimensional space-time Riemann problem, and two- and three-dimensional supersonic and hypersonic benchmark problems.

math.NA↗

arXiv2022

A globally convergent method to accelerate large-scale optimization using on-the-fly model hyperreduction: application to shape optimization

We present a numerical method to efficiently solve optimization problems governed by large-scale nonlinear systems of equations, including discretized partial differential equations, using projection-based reduced-order models accelerated with hyperreduction (empirical quadrature) and embedded in a trust-region framework that guarantees global convergence. The proposed framework constructs a hyperreduced model on-the-fly during the solution of the optimization problem, which completely avoids an offline training phase. This ensures all snapshot information is collected along the optimization trajectory, which avoids wasting samples in remote regions of the parameters space that are never visited, and inherently avoids the curse of dimensionality of sampling in a high-dimensional parameter space. At each iteration of the proposed algorithm, a reduced basis and empirical quadrature weights are constructed precisely to ensure the global convergence criteria of the trust-region method are satisfied, ensuring global convergence to a local minimum of the original (unreduced) problem. Numerical experiments are performed on two fluid shape optimization problems to verify the global convergence of the method and demonstrate its computational efficiency; speedups over 18x (accounting for all computational cost, even cost that is traditionally considered "offline" such as snapshot collection and data compression) relative to standard optimization approaches that do not leverage model reduction are shown.

math.OC↗

arXiv2022

A robust, high-order implicit shock tracking method for simulation of complex, high-speed flows

High-order implicit shock tracking is a new class of numerical methods to approximate solutions of conservation laws with non-smooth features. These methods align elements of the computational mesh with non-smooth features to represent them perfectly, allowing high-order basis functions to approximate smooth regions of the solution without the need for nonlinear stabilization, which leads to accurate approximations on traditionally coarse meshes. The hallmark of these methods is the underlying optimization formulation whose solution is a feature-aligned mesh and the corresponding high-order approximation to the flow; the key challenge is robustly solving the central optimization problem. In this work, we develop a robust optimization solver for high-order implicit shock tracking methods so they can be reliably used to simulate complex, high-speed, compressible flows in multiple dimensions. The proposed method integrates practical robustness measures into a sequential quadratic programming method, including dimension- and order-independent simplex element collapses, mesh smoothing, and element-wise solution re-initialization, which prove to be necessary to reliably track complex discontinuity surfaces, such as curved and reflecting shocks, shock formation, and shock-shock interaction. A series of nine numerical experiments -- including two- and three-dimensional compressible flows with complex discontinuity surfaces -- are used to demonstrate: 1) the robustness of the solver, 2) the meshes produced are high-quality and track continuous, non-smooth features in addition to discontinuities, 3) the method achieves the optimal convergence rate of the underlying discretization even for flows containing discontinuities, and 4) the method produces highly accurate solutions on extremely coarse meshes relative to approaches based on shock capturing.

math.NA↗

arXiv2021

Accurate quantification of blood flow wall shear stress using simulation-based imaging: a synthetic, comparative study

Simulation-based imaging (SBI) is a blood flow imaging technique that optimally fits a computational fluid dynamics (CFD) simulation to low-resolution, noisy magnetic resonance (MR) flow data to produce a high-resolution velocity field. In this work, we study the effectivity of SBI in predicting wall shear stress (WSS) relative to standard magnetic resonance imaging (MRI) postprocessing techniques using two synthetic numerical experiments: flow through an idealized, two-dimensional stenotic vessel and a model of an adult aorta. In particular, we study the sensitivity of these two approaches with respect to the Reynolds number of the underlying flow, the resolution of the MRI data, and the noise in the MRI data. We found that the SBI WSS reconstruction: 1) is insensitive to Reynolds number over the range considered (Re $\leq$ 1000), 2) improves as the amount of MRI data increases and provides accurate reconstructions with as little as three MRI voxels per diameter, and 3) degrades linearly as the noise in the data increases with a slope determined by the resolution of the MRI data. We also consider the sensitivity of SBI to the resolution of the CFD mesh and found there is flexibility in the mesh used for SBI, although the WSS reconstruction becomes more sensitive to other parameters, particularly the resolution of the MRI data, for coarser meshes. This indicates a fundamental trade-off between scan time (i.e., MRI data quality and resolution) and reconstruction time using SBI, which is inherently different than the trade-off between scan time and reconstruction quality observed in standard MRI postprocessing techniques.

math.NA↗

arXiv2021

Model reduction of convection-dominated partial differential equations via optimization-based implicit feature tracking

This work introduces a new approach to reduce the computational cost of solving partial differential equations (PDEs) with convection-dominated solutions: model reduction with implicit feature tracking. Traditional model reduction techniques use an affine subspace to reduce the dimensionality of the solution manifold and, as a result, yield limited reduction and require extensive training due to the slowly decaying Kolmogorov $n$-width of convection-dominated problems. The proposed approach circumvents the slowly decaying $n$-width limitation by using a nonlinear approximation manifold systematically defined by composing a low-dimensional affine space with a space of bijections of the underlying domain. Central to the implicit feature tracking approach is a residual minimization problem over the reduced nonlinear manifold that simultaneously determines the reduced coordinates in the affine space and the domain mapping that minimize the residual of the unreduced PDE discretization. The nonlinear trial manifold is constructed by using the proposed residual minimization formulation to determine domain mappings that cause parametrized features to align in a reference domain for a set of training parameters. Because the feature is stationary in the reference domain, i.e., the convective nature of solution removed, the snapshots are effectively compressed to define an affine subspace. The space of domain mappings, originally constructed using high-order finite elements, are also compressed in a way that ensures the boundaries of the original domain are maintained. Several numerical experiments are provided, including transonic and supersonic, inviscid, compressible flows, to demonstrate the potential of the method to yield accurate approximations to convection-dominated problems with limited training.

math.NA↗

arXiv2021

Physics-informed graph neural Galerkin networks: A unified framework for solving PDE-governed forward and inverse problems

Despite the great promise of the physics-informed neural networks (PINNs) in solving forward and inverse problems, several technical challenges are present as roadblocks for more complex and realistic applications. First, most existing PINNs are based on point-wise formulation with fully-connected networks to learn continuous functions, which suffer from poor scalability and hard boundary enforcement. Second, the infinite search space over-complicates the non-convex optimization for network training. Third, although the convolutional neural network (CNN)-based discrete learning can significantly improve training efficiency, CNNs struggle to handle irregular geometries with unstructured meshes. To properly address these challenges, we present a novel discrete PINN framework based on graph convolutional network (GCN) and variational structure of PDE to solve forward and inverse partial differential equations (PDEs) in a unified manner. The use of a piecewise polynomial basis can reduce the dimension of search space and facilitate training and convergence. Without the need of tuning penalty parameters in classic PINNs, the proposed method can strictly impose boundary conditions and assimilate sparse data in both forward and inverse settings. The flexibility of GCNs is leveraged for irregular geometries with unstructured meshes. The effectiveness and merit of the proposed method are demonstrated over a variety of forward and inverse computational mechanics problems governed by both linear and nonlinear PDEs.

cs.CE↗