Development of a Model Order Reduced Arbitrary Lagrangian Eulerian (MORALE) formulation for structures subjected to dynamic moving loads Publication type: Original Article Journal: Preprint submitted to IJNME

Atul Anantheswar 1, Jannick Kehls 2, Ines Wollny 1, Tim Brepols 2, Stefanie Reese 2,3, Michael Kaliske 1, * · 2025

Read Development of a Model Order Reduced Arbitrary Lagrangian Eulerian (MORALE) formulation for structures subjected to dynamic moving loads Publication type: Original Article Journal: Preprint submitted to IJNME by Atul Anantheswar 1, Jannick Kehls 2, Ines Wollny 1, Tim Brepols 2, Stefanie Reese 2,3, Michael Kaliske 1, * (2025) 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-2509-20069.txt.

Introduction

Roads flex under rolling wheels, and simulating that flex fast enough for a digital twin demands shortcuts. Atul Anantheswar and five co-authors merge two: a moving-frame formulation that follows the load, and model order reduction that shrinks the system. Their combined MORALE method handles both elastic and viscoelastic materials. Pavement structures under moving loads are the test case.

Speed and efficiency gains are significant, the authors report. A digital twin must answer what-if questions while the real road is in service. Full simulations cannot keep that pace. Reduced models that respect the moving load's physics can. The work targets roadway infrastructure, where twins could schedule maintenance before failures surface. The formulation is the contribution; pavements are its proving ground.

Catalog record

CREATORAtul Anantheswar 1, Jannick Kehls 2, Ines Wollny 1, Tim Brepols 2, Stefanie Reese 2,3, Michael Kaliske 1, *
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE40 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-20069

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Development of a Model Order Reduced Arbitrary Lagrangian Eulerian (MORALE) formulation for structures subjected to dynamic moving loads Publication type: Original Article Journal: Preprint submitted to IJNME Atul Anantheswar 1, Jannick Kehls 2, Ines Wollny 1, Tim Brepols 2, Stefanie Reese 2,3, Michael Kaliske 1, * Abstract In recent developments, it has been demonstrated that the Arbitrary Lagrangian Eulerian (ALE) formulation can be utilized to improve computational efficiency, when simulating the response of structures subjected to moving loads. It is also well established in literature, that Model Order Reduction (MOR) techniques significantly enhance calculation speed. This contribution details the combination of both these tools into a novel Model Order Reduced Arbitrary Lagrangian Eulerian (MORALE) formulation. Both hyperelastic and viscoelastic material models are considered. Simulations of pavement structures subjected to moving loads are then carried out, which show a significant enhancement in computational speed and efficiency. Such an efficient and fast simulation framework is of vital importance in technologies such as digital twins of roadway infrastructure (like pavements), as it enables engineers to quickly run what-if analyses and make informed decisions about the management of the structure under consideration. keywords Dynamic ALE formulation, Moving loads, Inelastic materials, Transient simulations, Model order reduction † † startpage: 1 † † corresponding: *Corresponding author, Anantheswar et al. \presentaddress Technische Universität Dresden, 01062 Dresden, Germany 1 Introduction The contribution highlighted in this manuscript is part of ongoing research in the collaborative research centre SFB/TRR 339. The aim of this project is to establish a digital twin of the road [ 1 , 2 ] . Such a digital twin of the road necessitates the employment of fast and efficient simulation strategies. One approach enabling fast simulations that can predict the response of pavement structures is the Arbitrary Lagrangian Eulerian (ALE) formulation. ALE approaches are most commonly used in the field of computational fluid dynamics [ 3 , 4 , 5 , 6 ] . They are also applied as mesh refinement techniques to counter issues that arise due to extreme mesh distortions [ 7 , 8 , 9 , 10 , 11 , 12 ] . In recent years, the ALE approach has been used to model the steady state rotation of tires using a fixed finite element mesh [ 13 , 14 ] . Subsequent developments have enabled the simulation of the response of pavement structures to steady rolling (constant velocity) wheel loads [ 15 , 16 ] . Further research in this direction has allowed the simulation of the transient response of pavements subjected to dynamic wheel loads [ 17 , 18 ] , along with the consideration of inelastic material behaviour [ 19 ] . Moreover, another technique to lower the computation time of the simulations is model order reduction (MOR). For finite element simulations, projection-based MOR has been widely used [ 20 , 21 ] . The main idea of projection-based MOR is to reduce the size of the system of equations by projecting the high-dimensional system onto a low-dimensional subspace. This is achieved by using a set of basis functions that span the low-dimensional subspace. The reduced system of equations can then be solved more efficiently than the original system. The reduced order model (ROM) can be constructed using various techniques, such as proper orthogonal decomposition (POD) [ 21 ] , balanced truncation [ 22 ] , and modal basis method [ 23 ] . In this work, we focus on the POD method for constructing the ROM. It has been shown that the POD delivers accurate approximations for nonlinear problems in solid mechanics [ 24 ] and has been extended for various applications such as component-based MOR [ 25 ] , adaptively sub-structured elastoplastic simulations [ 26 ] , and reduced-order modelling of gradient-extended damage [ 27 ] . Furthermore, the POD serves as a basis for most hyper-reduction techniques such as the discrete empirical interpolation method [ 28 , 29 , 30 ] or energy-conserving sampling and weighting [ 31 , 32 ] . The POD is a data-driven approach that uses snapshots of the solution to construct a low-dimensional basis that captures the dominant features of the solution. Recent developments showcase the coupling of ALE and MOR in the simulation of rolling tires in contact with the road [ 33 , 34 , 35 ] . In these works, the behaviour of the tire is the main focus, and the combination of ALE and MOR techniques has been demonstrated to improve computational efficiency. However, in the present work, the combination of ALE and MOR frameworks to simulate the transient response of pavement structures subjected to moving loads is detailed. This manuscript is structured as follows. In Section 2 , the reasoning behind the adoption of the ALE approach to simulate the response of pavement structures is provided with a kinematics background. Then, the balance laws which are the basis for an implementation into a finite element framework are provided in Section 3 , specifically keeping the ALE reference configuration in mind. Details on the advection of the solution field as well as material point quantities are given in Section 4 . Next, Section 5 describes the application of model order reduction techniques to the assembled system of equations which includes contributions from the ALE implementation. Further, numerical studies are carried out and the results are presented in Section 6 , highlighting the benefits of the combined MORALE formulation. Finally, the conclusions drawn from the studies are summarized in Section 7 . 2 Kinematics in the ALE formulation Consider the deformation of a body from the initial configuration to the current configuration as shown in Figure 1 . Conventional methods commonly assume that the reference configuration is identical to the initial configuration and both share the same stationary reference frame \boldsymbol{e}_{\text{i}}^{\text{Lagrange}} . However, for the ALE approach, a reference frame that moves with time \boldsymbol{e}_{\text{i}}^{\text{ALE}} is introduced. For the particular case of structures subjected to moving loads, the ALE reference frame can be moved with the same velocity as that of the moving load. This moving reference frame then offers a unique perspective, where the load appears stationary but the material of the structure (pavements, gantry crane girders etc.) appears to flow beneath the load [ 15 ] . This comes with the advantage that only the relevant region (domain of interest) around the load needs to be analysed. Hence, the computational costs associated with discretizing and simulating regions far away from the load are avoided. When it comes to structures like pavements subjected to moving vehicle loads, this irrelevant domain can be significantly large, as the entire length of the structure in the path of the moving load would need to be analysed. It is also worth mentioning that the deformation from the ALE reference configuration can be further decomposed to accommodate an inelastic intermediate configuration, thereby enabling the simulation of inelastic structures [ 19 ] . Figure 1 : ALE kinematics [ 19 ] . From Figure 1 , the current position \boldsymbol{x} of a material point can be written in terms of the position vectors ( \boldsymbol{X} , \boldsymbol{\chi} and \boldsymbol{\chi}_{\text{0}}^{\text{ALE}} ) and displacements ( \boldsymbol{u} , \boldsymbol{u}_{\text{rig}} and \hat{\boldsymbol{u}} ) in various reference frames

Read the complete text