Structure-preserving dynamical low-rank approximation for parametric elastic guided waves
Dimitri Goutaudier · 2026
Read Structure-preserving dynamical low-rank approximation for parametric elastic guided waves by Dimitri Goutaudier (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-30469.txt.
Introduction
Dimitri Goutaudier speeds up simulations of elastic guided waves used to find hidden damage. Guided waves reveal corrosion and delamination early. Full simulations are costly in many-query settings. Reduced order models cut that cost by projecting onto small spaces. Guided waves are transport-dominated and dispersive, defeating static subspaces. Their Hamiltonian structure also demands care to conserve energy.
Dynamical low-rank approximation lets the reduced basis move with the solution. Goutaudier builds a parametric framework on that idea. An offline stage constructs a time-dependent symplectic reduced basis. For a simplified parameter class, the basis evolution has a closed form. That yields an energy-preserving reduced propagator with no online integration. Tests use a two-dimensional elasticity problem with damage. Compression reaches ranks of only ten to thirty. Reconstruction errors stay between a thousandth and a hundredth. Speedups run to two or three orders of magnitude.
Catalog record
| CREATOR | Dimitri Goutaudier |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 63 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2606.30469
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Structure-preserving dynamical low-rank approximation for parametric elastic guided waves Thanks: Submitted to the editors DATE. Dimitri Goutaudier Abstract Elastic guided waves are widely used in Structural Health Monitoring (SHM). In many-query settings, the computational cost of high-fidelity simulations motivates the use of projection-based reduced order modeling (ROM). However, the transport-dominated and dispersive nature of guided waves challenges static linear subspaces. In addition, preserving the Hamiltonian structure of the equations for energy conservation necessitates dedicated projection techniques. While the Dynamical Low Rank Approximation (DLRA) has proven effective for other wave equations, its application to elastic guided waves in SHM has remained unexplored. In this work, we introduce a structure-preserving parametric ROM framework that leverages the DLRA in an off-line/on-line strategy. During the off-line stage, a time-dependent symplectic reduced basis is constructed from training simulations. For a simplified class of parameter dependencies, we derive a closed-form solution of the nonlinear basis evolution equation. This analytical result yields a closed-form, energy-preserving reduced propagator during wave propagation, eliminating on-line time integration after the loading phase. We validate our approach on a 2D elasticity problem featuring dispersive guided waves interacting with a damage. The results demonstrate high compression ratios (rank ∼ 10 − 30 \sim 10-30 ), low full field reconstruction errors ( ∼ 10 − 3 − 10 − 2 \sim 10^{-3}-10^{-2} ), speedups of two to three orders of magnitude, and excellent long-time energy conservation. keywords wave propagation, Hamiltonian system, symplectic reduced order modeling, dynamic reduced basis, Structural Health Monitoring Funding. This work was supported by the French National Research Agency (ANR) under the “Jeunes Chercheurs et Jeunes Chercheuses” program (Grant ANR-25-CE51-1125-01). 1 Introduction The goal of projection-based reduced order modeling (ROM) is to accelerate simulations [ 3 , 17 ] for multi-query or real time applications, such as uncertainty quantification [ 8 ] or parametric studies [ 14 ] . This is achieved by projecting high-dimensional equations, typically arising from the discretization of time-dependent partial differential equations, onto a low-dimensional subspace. The efficiency of a ROM strategy critically depends on the choice of the reduced subspace and the projector, which must be tailored to the mathematical structure of the underlying physical system. Indeed, physical systems exhibit a range of dynamical behaviors, such as dissipative or conservative, linear or nonlinear, transport-dominated, stochastic, etc., each posing specific challenges. In this study, we focus on Hamiltonian wave equation problems, involving energy conservation and moving features with high gradients. Such systems are central in various Structural Health Monitoring (SHM) applications, where elastic guided waves are used to detect, localize and characterize damages at early stages (e.g., corrosion in metallic parts [ 22 ] , delamination in composite materials [ 10 ] ). However, the interplay between energy preservation, moving wave fronts and dispersion (frequency-dependent wave speeds), renders conventional ROM techniques ineffective [ 25 ] . For instance, the well-established POD-Galerkin method fails for these problems at two fundamental levels. First, a static linear subspace cannot efficiently track translating features over an arbitrary time window. Several methods have been proposed to address this difficulty, such as high-order manifolds [ 1 ] , registration methods [ 30 ] , the shifted Proper Orthogonal Decomposition [ 28 ] , the time-incremental Proper Generalized Decomposition [ 12 , 13 ] , or dynamic reduced bases [ 16 , 24 , 21 ] . Second, the orthogonal projection breaks the Hamiltonian structure, leading to energy drift and loss of physical fidelity. Structure-preserving model order reduction techniques address this challenge with symplectic Galerkin projection frameworks, involving a symplectic basis and an (oblique) symplectic projector. Such a reduced basis can be obtained by approximating the solution of the so-called Proper Symplectic Decomposition (PSD) problem [ 27 ] , which mirrors the traditional POD problem. In particular, the complex SVD method provides the solution of the PSD problem restricted to the set of ortho-symplectic bases. Successful applications of symplectic model order reduction have been reported for mechanical engineering applications [ 5 , 26 ] . However, a PSD basis remains static, which limits its applicability for the transport-dominated dynamics concerning this work. The Dynamical Low Rank Approximation (DLRA) offers a promising framework for Hamiltonian wave equations, as it dynamically adapts the reduced basis to track moving features [ 18 ] . This method seeks a low rank approximation of a large time-dependent matrix satisfying a matrix differential equation. For instance, such situation arises in this study when simultaneously solving the elastodynamic equations at multiple query parameters. The DLRA proceeds by projecting the high-dimensional flow equations onto the tangent space of matrices of fixed rank. This yields differential equations for the factors of an SVD-like decomposition of the time-dependent low-rank approximation. Over the past decades, efficient time marching schemes have been developed [ 20 , 7 , 23 ] to solve these equations with desirable accuracy and robustness properties (e.g., against small singular values if the rank is over-approximated by the user). For a small enough time step, the time-discrete DLRA solution applies instantaneously the truncated SVD of the true solution [ 11 ] , without ever computing an expensive SVD. Symplectic formulations have also been developed to preserve the Hamiltonian structure of wave equations. To name a few, the DLRA has been successfully applied to the Schrödinger wave equation [ 6 ] , the shallow water equation [ 24 ] or the 2D stochastic wave equation [ 21 ] . To the best of the author’s knowledge, however, the DLRA has not yet been applied to SHM-related elastic guided wave problems. In addition, despite the impressive above-mentioned progresses, its computational cost remains hardly actionable for the real-time constraints of SHM applications. In this work, we attempt to reconcile the efficiency of the off-line/on-line logic, well established for projection-based ROM techniques applied to linear dissipative systems, with Hamiltonian wave equations. The key idea is to defer the expensive computation of the time-dependent basis to an off-line phase. More precisely, a symplectic DLRA is applied to simultaneously solve the elastodynamic equations for a discrete set of training parameters \mathcal{D}^{\text{training}}\subset\mathbb{R}^{P} , where P is the number of parameters of interest. In this study, the latter correspond to the diameter of the transducer exciting the structure and to the central frequency of the loading (hence P=2 ). This yields a global-in-parameter but time-dependent basis of low rank r (typically 10-30 ): U(t;\mathcal{D}^{\text{training}}) (1) satisfying three key properties at all times: it is orthonormal, symplectic, and dynamically orthogonal (meaning its time derivative is orthogonal to its range.). The on-line phase then consists in solving a cheap Hamiltonian reduced flow at a discrete query parameter set \mathcal{D}^{\text{query}}\subset\mathbb{R}^{P} . Finally, the quantities of interest are efficiently obtained with the pre-computed basis. For instance, the high-dimensional full wave field Y\in\mathbb{R}^{N\times m} , with N the huge number of spatial grid points and m the number of query parameters, is reconstructed at a user-selected time T as: Y(T;\mathcal{D}^{\text{query}})\approx U(T;\mathcal{D}^{\text{training}})Y_{r}(T;\mathcal{D}^{\text{query}}), (2) where Y_{r}(T;\mathcal{D}^{\text{query}})\in\mathbb{R}^{r\times m} is the low-dimensional matrix solution of a reduced Hamiltonian flow evaluated at the query parameters. We validate this approach on a 2D plane strain elasticity problem with dispersive guided waves, demonstrating high compression ratios, important speedups, and negligible energy fluctuation. The paper is organized as follows. In section 2 ,