David Néron

Virtual testing for the prediction of damping in joints

Purpose – The purpose of this paper is to propose a virtual testing strategy in order to predict damping due to the joints which are present in the ARIANE 5 launcher.Design/methodology/approach – Since engineering finite element codes do not give satisfactory results, either because they are too slow or because they cannot calculate dissipation accurately, a new computational tool is introduced based on the LArge Time INcrement (LATIN) method in its multiscale version.Findings – The capabilities of the new strategy are illustrated on one of the joints of ARIANE 5. The damping predicted virtually is compared to experimental results, and the approach appears promising.Originality/value – The tool which has been developed gives access to calculations which were previously unaffordable with standard computational codes, which may improve the design process of launchers. The code is transferred into ASTRIUM‐ST, where it is being used to build a database of dissipations in the joints of the ARIANE 5 launcher.

Multiscale elastic-viscoplastic computational analysis: A detailed example

The objective of this work is to develop an efficient strategy for quasi-static problems with elastic-viscoplastic constitutive laws. Our approach is based on the multiscale LATIN method for domain decomposition, and particularly on the use of the Proper Generalized Decomposition (PGD) method, which allows a drastic decrease in computation costs. We present the method in its general form applicable to problems with constitutive laws expressed using internal variables; then we discuss the technical features which are necessary in order to deal with elastic-viscoplastic models. We illustrate the method in detail through a onedimensional example using a Chaboche-type elastic-viscoplastic constitutive law.

Accounting for Nonlinear Aspects in Multiphysics Problems: Application to Poroelasticity

Multiphysics phenomena lead to computationally intensive structural analyses. Recently, a new strategy derived from the LATIN method was described and successfully applied to the consolidation of saturated porous soils.

A Model Reduction Technique in Space and Time for Fatigue Simulation

The simulation of mechanical responses of structures subjected to cyclic loadings for a large number of cycles remains a challenge. The goal herein is to develop an innovative computational scheme for fatigue computations involving non-linear mechanical behaviour of materials, described by internal variables. The focus is on the Large Time Increment (LATIN) method coupled with a model reduction technique, the Proper Generalized Decomposition (PGD). Moreover, a multi-time scale approach is proposed for the simulation of fatigue involving large number of cycles. The quantities of interest are calculated only at particular pre-defined cycles called the “nodal cycles” and a suitable interpolation is used to estimate their evolution at the intermediate cycles. The proposed framework is exemplified for a structure subjected to cyclic loading, where damage is considered to be isotropic and micro-defect closure effects are taken into account. The combination of these techniques reduce the numerical cost drastically and allows to create virtual S-N curves for large number of cycles.

Extended-PGD Model Reduction for Nonlinear Solid Mechanics Problems Involving Many Parameters

Reduced models and especially those based on Proper Generalized Decomposition (PGD) are decision-making tools which are about to revolutionize many domains. Unfortunately, their calculation remains problematic for problems involving many parameters, for which one can invoke the “curse of dimensionality”. The paper starts with the state-of-the-art for nonlinear problems involving stochastic parameters. Then, an answer to the challenge of many parameters is given in solid mechanics with the so-called “parameter-multiscale PGD”, which is based on the Saint-Venant principle.

A Data Compression Approach for PGD Reduced-Order Modeling

This work concerns the Proper Generalized Decomposition (PGD) which is used to solve problems defined over the time-space domain and which are possibly nonlinear. The PGD is an a priori model reduction technique which allows to decrease CPU costs drastically by seeking the solution of a problem in a reduced-order basis generated automatically by a dedicated algorithm. The algorithm which is used herein is the LATIN method, a non incremental iterative strategy which enables to generate the approximations of the solution over the entire time-space domain by successive enrichments. The problematics which is addressed in this paper is the construction of the resulting reduced-order models along the iterations, which can represent a large part of the remaining global CPU cost. To make easier the construction of these models, we propose an algebraic framework adapted to PGD which allows to define a “compressed” version of the data. The space of the compressed fields exhibits some very interesting properties that lead to an important increase of the performances of the global strategy.Copyright

A Virtual Testing Approach for Laminated Composites Based on Micromechanics

The chapter deals with a crucial question for the design of composite structures: how can one predict the evolution of damage up to and including final fracture? Virtual testing, whose goal is to drastically reduce the huge number of industrial tests involved in current characterization procedures, constitutes one of today’s main industrial challenges. In this work, one revisits our multiscale modeling answer through its practical aspects. Some complements regarding identification, kinking, and crack initiation are also given. Finally, the current capabilities and limits of this approach are discussed, as well as the computational challenges that are inherent to “Virtual Structural Testing.”

Using X-ray computed tomography for quantification of cell proliferation within a perfusion bioreactor

perfusion bioreactor, and cell proliferation dynamics have been monitored by X-ray μCT. 3D digital model reproducing the medium flow within the stack of biomaterials has been generated to better understand the role of local shear stress on cell proliferation in the engineered bone tissue production. Experiments and numerical simulation allow us to better understand the influence of fluid induced shear stress on cell behavior.

A time-space framework suitable for the LATIN computational strategy for multiphysics problems

Since the last few decades, the simulation of multiphysics phenomena has become one of the major issue in the design of computational methods. Indeed, these problems often lead to computationally intensive analyses and then strategies to keep these problems affordable are of special interest.