Michael BRUN - Academia.edu (original) (raw)

Papers by Michael BRUN

Finite Elements in Analysis and Design, 2020

21-25 june 2015International audienceno abstrac

During an earthquake, a bridge crane may be subjected to multiple impacts between crane... more During an earthquake, a bridge crane may be subjected to multiple impacts between crane wheels and rail. In order to model such phenomena, a time-history dynamic analysis with a multi-scale approach is performed. The high frequency aspect of the impacts between wheels and rails is taken into account by a Lagrange explicit event-capturing algorithm based on a velocity-impulse formulation to resolve contacts and impacts. An implicit temporal scheme is used for the rest of the structure. The numerical coupling between the implicit and the explicit schemes is achieved with a heterogeneous asynchronous time-integrator.

Computers and Geotechnics, 2021

Abstract This paper presents a benchmark for three-dimensional explicit asynchronous absorbing la... more Abstract This paper presents a benchmark for three-dimensional explicit asynchronous absorbing layers for modeling unbounded domains through a standard displacement-based finite element method: Three-dimensional Perfectly Matched Layers (PML) is compared to recently proposed Absorbing Layers with Increasing Damping (ALID) based on Rayleigh and Kosloff damping formulations. The coupled problem, including the interior subdomain and the absorbing layers, is implemented in the framework of Heterogeneous Asynchronous Time Integrator (HATI), enabling the absorbing layers to be handed with an explicit Central Difference scheme, with fine time steps, independently from the time stepping procedure adopted in the interior subdomain. Simple 3D Lamb’s tests are considered by using different kinds of absorbing layers. The superiority of the PML in terms of the accuracy and computation time is highlighted. Finally, realistic 3D applications are investigated, such as the Lamb’s test and the study of the screening effect provided by a horizontal wave barrier in mitigating the ground surface vibration generated by an excited plate.

Proceedings of the 5th International Conference on Computational Methods in Structural Dynamics and Earthquake Engineering (COMPDYN 2015), 2015

In the framework of the Finite Element Method, Perfectly Matched Layer (or PML) is recognized as ... more In the framework of the Finite Element Method, Perfectly Matched Layer (or PML) is recognized as a very effective tool for reproducing unbounded domains (Basu et al. [13]). Nonetheless, the computation time required by the PML may be large, especially when an explicit time integration scheme is adopted for dealing with the wave propagation problem both in the domain of interest and in the PML medium (Basu et al. [13]). The paper proposed to investigate subdomain strategies enabling to choose the appropriate time integration scheme in the PML with its own time step, independently on the choice of the explicit time scheme in the domain of interest associated with a fine time step satisfying the stability criterion (CFL condition). The investigated subdomain strategy, proposed by Gravouil and Combescure [14], is based on the Schur dual approach, characterized by the introduction of Lagrange multipliers at the interface between subdomains for ensuring the velocity continuity.

Ce resume etendu a pour but de presenter une methode de couplage pour les problemes d'in-tera... more Ce resume etendu a pour but de presenter une methode de couplage pour les problemes d'in-teraction fluide-structure, dont le but est d'etre conservatif et d'utiliser des methodes de calcul usuelles pour chaque sous-domaine fluide et structure. Ainsi le couplage est base sur une formulation monoli-thique de la conservation de l'energie et est resolu par un algorithme de co-simulation. Le but final est de coupler deux codes commerciaux existants. La methode proposee est validee sur le cas test du piston 1D, en mono et multi-echelle en temps.

A previously developed interface energy-conserving coupling technique allows to carry out accurat... more A previously developed interface energy-conserving coupling technique allows to carry out accurate and stable FSI simulations. In order to prevent the time-step size requirements of one domain from being inherited by the other domain, one must be able to integrate each domain with a different time-step. The purpose of this study is to implement a method allowing to integrate each domain with separate timesteps while respecting the conservation properties of the initially proposed coupling technique Mots clefs : Interaction fluide-structure, conservation de l’energie, couplage multi-echelle en temps

Engineering Structures, 2022

Journal of the Mechanics and Physics of Solids, 2020

Numerous laboratory experiments have demonstrated the dependence of the friction coefficient on t... more Numerous laboratory experiments have demonstrated the dependence of the friction coefficient on the interfacial slip rate and the contact history, a behavior generically called rate and state friction. Although numerical models have been widely used for analyzing rate and state friction, in general they consider infinite elastic domains surrounding the sliding interface and rely on boundary integral formulations. Much less work has been dedicated to modeling finite size systems to account for interactions with boundaries. This paper investigates rate and state frictional interfaces in the context of finite size systems with the finite element method in explicit dynamics. It is shown that due to the highly non-linear nature of rate and state friction and its sensitivity to numerical noise, the time integration step to achieve an accurate steady state solution is orders of magnitude smaller compared to the stable time step required in boundary integral formulations. We provide evidence that the noise, which is source of instability in the finite element solution, originates from internal discretization nodes. We then investigate the long term behavior of the sliding interface for two different friction laws: a velocity weakening law, for which the friction monotonously decreases with increasing sliding velocity, and a velocity weakening-strengthening law, for which the friction coefficient first decreases but then increases above a critical velocity. We show that for both friction laws at finite times, that is before wave reflections from the boundaries come back to the sliding interface, a temporary steady state sliding is reached, with a well-defined stress drop at the interface. This stress drop gives rise to a stress concentration and leads to an analogy between friction and fracture. However, at longer times, that is after multiple wave reflections, the stress drop is essentially zero, resulting in losing the analogy with fracture mechanics. Finally, the simulations reveal that velocity weakening is unstable at long time scales, as it results in an acceleration of the sliding blocks. On the other hand, velocity weakening-strengthening reaches a steady state sliding configuration.

IOP Conference Series: Materials Science and Engineering, 2018

Co-simulation strategies using both Abaqus/Explicit and Abaqus/Implicit are investigated for anal... more Co-simulation strategies using both Abaqus/Explicit and Abaqus/Implicit are investigated for analysing wave barrier in 2D unbounded soil domain. The Co-simulation is based on the coupling GC method, allowing for coupling different finite element codes with different time integrators and timescales depending on the partitions of the domain. Absorbing Layers using Increasing Damping (ALID), based on Rayleigh damping, are considered at the boundary to model the semi-infinite medium. The proposed absorbing region is called Hybrid (different time integrators) Asynchronous (different time steps) Absorbing Layers using Increasing Damping (HA-ALID). First, HA-ALID turns out to be more accurate than nonreflective conditions available in Abaqus/Explicit. Second, the time steps for the solid barrier and the HA-ALID are not restrained by the CFL condition imposed by the explicit partition for stability reasons. Third, the critical time step in the domain of interest remains unaffected by the choice of damping characteristics in the HA-ALID, contrary to the case of a full explicit computation. Because of its good accuracy and ability to be realized using only FE Abaqus package, without a third-party software component, this strategy provides a wide range of applications in soil-structure interaction problems.

Computers and Geotechnics, 2019

This paper presents a novel approach for modeling infinite media, called Hybrid (different time i... more This paper presents a novel approach for modeling infinite media, called Hybrid (different time integrators) Asynchronous (different time steps) Kosloff Absorbing Layers with Increasing Damping (HA-Kosloff ALID). By using strong forms of wave propagation in Kosloff media, its design equation is derived as well as optimal conditions between physical and absorbing domains. Explicit/Implicit co-simulation is adopted to reduce computation time. Examples of semi-infinite bar and Lamb's test are implemented to illustrate the efficiency of our approach in terms of accuracy and CPU time, in comparison to Rayleigh ALID and PML. It turns out to be efficient and convenient for modeling unbounded domain.

Computational Mechanics, 2017

In computational structural dynamics, particularly in the presence of nonsmooth behavior, the cho... more In computational structural dynamics, particularly in the presence of nonsmooth behavior, the choice of the time-step and the time integrator has a critical impact on the feasibility of the simulation. Furthermore, in some cases, as in the case of a bridge crane under seismic loading, multiple timescales coexist in the same problem. In that case, the use of multi-time scale methods is suitable. Here, we propose a new explicit-implicit heterogeneous asynchronous time integrator (HATI) for nonsmooth transient dynamics with frictionless unilateral contacts and impacts. Furthermore, we present a new explicit time integrator for contact/impact problems where the contact constraints are enforced using a Lagrange multiplier method. In other words, the aim of this paper consists in using an explicit time integrator with a fine time scale in the contact area for reproducing high frequency phenomena, while an implicit time integrator is adopted in the other parts in order to reproduce much low frequency phenomena and to optimize the CPU time. In a first step, the explicit time integrator is tested on a one-dimensional example and compared to Moreau-Jean's event-capturing schemes. The explicit algorithm is found to be very accurate and the scheme has generally a higher order of convergence than Moreau-Jean's schemes and provides also an excellent energy behavior. Then, the two time scales explicit-implicit B Fatima-Ezzahra Fekak

Optics Express, 2015

We characterize the nonlinear optical response of low loss Si 0.6 Ge 0.4 / Si waveguides in the m... more We characterize the nonlinear optical response of low loss Si 0.6 Ge 0.4 / Si waveguides in the mid-infrared between 3.3 μm and 4 μm using femtosecond optical pulses. We estimate the three and four-photon absorption coefficients as well as the Kerr nonlinear refractive index from the experimental measurements. The effect of multiphoton absorption on the optical nonlinear Kerr response is evaluated and the nonlinear figure of merit estimated providing some guidelines for designing nonlinear optical devices in the mid-IR. Finally, we compare the impact of free-carrier absorption at mid-infrared wavelengths versus near-infrared wavelengths for these ultra-short pulses.

Computers and Geotechnics, 2016

Abstract Many geotechnical structures are built on sandy soils. This complex medium exhibits a no... more Abstract Many geotechnical structures are built on sandy soils. This complex medium exhibits a non-associated rate-independent behaviour. Owing to this pattern, local and global tangent operators become non-symmetric. Consequently, it has been proved that instabilities and failure may develop before reaching the classic failure limit given by a plasticity limit. A proper analysis with the second-order work criterion allows for a good description of these instabilities and can be used as a good failure criterion. In this paper, we review the main results obtained in the last few decades when this criterion was applied to homogeneous problems. Then, a numerical integration of this quantity and a method for its normalization are proposed. The results of this integration lead to the definition of a safety factor for a global structure even under non-homogeneous conditions. Finally, an application to the design of a nailed wall is proposed. In this framework, a constitutive model that gathers the main basic features of soil behaviour was developed. This model allows a given soil to be described with only one set of parameters, for example from a loose to a dense state or from a normally consolidated to an over-consolidated state. This feature is useful for taking into account initial states or for observing a change in the main behaviour due to a large change in the confining loading conditions.

This paper reports on the integration and evaluation of robust, thick Carbon-PDMS nanocomposite e... more This paper reports on the integration and evaluation of robust, thick Carbon-PDMS nanocomposite electrodes (C-PDMS), integrated in PDMS microfluidic systems to carry out electric field induced fusion of biological cells. Such material preserves PDMS processing properties and sustains a large range of electric field intensities and frequencies without any carbon release. Polystyrene particles, yeasts and HEK-293 cells were manipulated in devices consisting of C-PDMS walls bounded to a glass plate. Different collective behaviors were observed such as alignment in chains parallel to the electric field and in circulating bands tilted with respect to the electric field direction. An AC electrofusion protocol was also tested. INTRODUCTION Thick electrodes are well suited to apply uniform lateral electric fields on a cross section of the channel, which favors successful electrical cell fusion [1,3]. Beyond this purpose, such electrodes are also useful for other electrokinetic functions (ce...

European Journal of Environmental and Civil Engineering, 2012

ABSTRACT This study performs a pushover procedure called Modified-UMRHA for seismic analysis of a... more ABSTRACT This study performs a pushover procedure called Modified-UMRHA for seismic analysis of a reinforced concrete structure, based on the UMRHA (uncoupled modal response history analysis) procedure proposed by Chopra and Goel. The objective is to investigate three approaches for setting up nonlinear single degree of freedom systems drawn from modal pushover curves. These curves are obtained using a non-adaptive load pattern equal to the consistent mass matrix times the modal shape. These approaches are applied on the 14-storey Grenoble City Hall studied in the framework of the ARVISE French research program. The results obtained from the Modified-UMRHA using different proposed nonlinear single degree of freedom systems are compared with those from a rigorous nonlinear response history analysis.Cette étude présente une procédure de pushover nommée Modified-UMRHA pour l’analyse sismique des structures en béton armé, basée sur la procédure UMRHA proposée par Chopra et Goel. L’objectif est d’examiner trois approches se distinguant par la définition de systèmes non linéaires à un degré de liberté à partir des courbes de modal pushover. Ces courbes sont obtenues en utilisant un chargement non adaptatif égal à la matrice de masse consistante multipliée par la déformée modale. Ces approches sont appliquées sur un bâtiment de 14 étages, l’Hôtel de Ville de Grenoble étudié dans le cadre du programme de recherche ARVISE. Les résultats obtenus de la procédure Modified-UMRHA en utilisant différents systèmes non linéaires à un degré de liberté, sont comparés avec ceux issus de l’analyse non linéaire dynamique rigoureuse.

This paper presents an experimental investigation conducted on masonry walls subjected to an outo... more This paper presents an experimental investigation conducted on masonry walls subjected to an outof-plane loading or normal pressure. This study has multiple objectives. It allows, first to quantify the bearing capacity in the case of uniform pressure in quasi-static loading case, and thus to highlight the associated modes of rupture, and secondly to estimate the improvements in terms of global behaviour when the structure is reinforced by Carbon Fiber Reinforced Polymer (CFRP) layers. A gain higher than 140% is observed for the bearing capacity. Finally, numerical simulation with the discrete element method is conducted and compared to the experiments. Good agreement between the numerical and the experimental results is obtained.

European Journal of Environmental and Civil engineering, 2010

... Laboratoire de Génie Civil et d&#x27;Ingénierie Environnementale (LGCIE) 34, avenue d... more ... Laboratoire de Génie Civil et d&#x27;Ingénierie Environnementale (LGCIE) 34, avenue des Arts, F-69621 Villeurbanne {jean-marie.reynouard, michael.brun ... available into the CASTEM code) was adopted for describing the nonlinear behaviour of the fiber beam elements (Guedes et al ...