Menu
fr / eng

Vibrations, Acoustic, Structures & Mechanical Shapes

The VAST (Vibroacoustics & Structures) team develops methods aimed at making the prediction and control of dynamic behaviours more reliable: vibrations, shocks and noise. The issues tackled by the VAST team are related to the dynamics, monitoring and diagnosis of systems, structures and materials, with a particular focus on energy dissipation behaviours (damping). Past and recurring research efforts concern viscoelasticity, contact dynamics and structured materials.

Understanding damping in just a few minutes…

 

The activities conducted by the VAST team focus on:

  • Viscoelasticity
  • Nonlinear waves and vibrations in granular media
  • Acoustic dissipation and metamaterials
  • Damping and energy dissipation in various assemblies
  • Measurement of vibration fields using high-speed cameras
  • Real-time digital twins in structural dynamics
  • Scale models in structural dynamics
  • Artificial intelligence in structural dynamics and for the monitoring of dynamic systems.

 

3D vibrations reconstruction with only one camera

 

 

Applications

The team’s research activities focus on transportation technologies (in aeronautics and in the space, automotive, railway and naval industries) as well as civil structures that are subject to dynamic stress and energy production systems. These activities are conducted together with industrial partners (AIRBUS, ARIANE Group, SAFRAN, RENAULT, PSA, PTC (FAYAT Group), JPB SYSTEME, KEYPROD, NEXTER, ALSTOM, etc.) and laboratories (ONERA – national aerospace research centre, LAMCOS INSA Lyon – Contact and Structure Mechanics Laboratory, ROBERVAL – University of Technology of Compiègne (UTC), NAVIER research institute – Ecole des Ponts ParisTech engineering school, etc.)

 

Doctoral candidates

The topics can be found on the list of on-going thesis projects

Head of the team

Jean-Luc Dion

Organisation of specific events

The VAST team has been organising the VISHNO conference for several decades. It was last held in Aix en Provence in 2014 (https://sites.google.com/site/vishno2014/) and in Le Mans in 2016 (together with the French Congress of Acoustics – http://cfa2016.univ-lemans.fr/). About 150 people attended each of those years.

The team also organised the one-day event dedicated to young researchers in vibrations and acoustics called Journée des jeunes chercheurs en vibrations et acoustique (JJCAB) that was held at Supméca in 2017, as well as the one-day event dedicated to field measurements in structural dynamics called Mesure de champs en dynamique des structures in 2018.  https://sites.google.com/view/jmdc2018-at-supmeca

Team members


  Jean-Bapiste CASIMIR

jean-baptiste.casimir@supmeca.fr

+33 1 49 45 29 63

Professeur des Universités

Dynamique des structures,

Méthodes « Meshless »

Viscoélasticité

 


  Jean-Luc DION

jean-luc.dion@supmeca.fr

+33 1 49 45 29 12

Professeur des Universités

Vibrations non linéaires

Traitement du signal                     

Dynamique Multicorps

Page perso. Researchgate

 


  Stéphane JOB

stephane.job@supmeca.fr

+33 1 49 45 29 00

Maître de Conférences

Docteur en acoustique – Expérimentateur

Ondes et vibrations non-linéaires, Milieux granulaires, Réseaux phononiques, Métamatériaux

 


  Stefania LO FEUDO

stefania.lofeudo@supmeca.fr

+33 1 49 45 29 00

Maître de Conférences

Contôle passif des vibrations

Vibrations non linéaires

Mesure de champs par camera rapide

 


  Martin GHIENNE

martin.ghienne@supmeca.fr

+33 1 49 45 29 00

Enseignant – Chercheur       

Dynamique stochastique des structures

Mesure de champs par camera rapide

 


  Benoit NENNIG

benoit.nennig@supmeca.fr

+33 1 49 45 29 00

Maître de Conférences

Acoustique, Matériaux absorbants, Méthode numérique, Ondes guidées

https://cv.archives-ouvertes.fr/benoit-nennig

 


  Nicolas  PEYRET

nicolas.peyret@supmeca.fr

+33 1 49 45 29 00

Maître de Conférences

Amortissement des structures

Calcul non linéaire des structures

Conception

 


  Franck RENAUD

franck.renaud@supmeca.fr

+33 1 49 45 29 00

Maître de Conférences

Dynamique des structures

Viscoélasticité

Dynamique Multicorps

 


  Alain STRICHER

alain.stricher@supmeca.fr

+33 1 49 45 29 00

Professeur agrégé

Tolérancement des structures assemblées

 

 

Doctorants


  Reza BABAJANIVALASHEDI

reza.babajanivalashedi@supmeca.fr

Dynamique et contrôle du ballotement

 


    Sophie Charles

sophie.charles@supmeca.fr

Impact de l’habilité spatiale sur la capacité à concevoir en 3D

 


  adrien.chassaigne@supmeca.fr

Propriété amortissante des collages structuraux

 


  Adrien GOELLER

adrien.goeller@supmeca.fr

Analyse dynamique vibratoire par vidéo rapide

« Contribution à la perception augmentée de scènes dynamiques : schémas temps réels d’assimilation de données pour la mécanique du solide et des structures » Janvier 2018

 


    Kévin Jaboviste

kevin.jaboviste@univ-fcomte.fr

Contribution à l’étude et à la conception des suspensions et amortisseurs vibratoires utilisant des matériaux métalliques et élastomères

 


  Tanguy LOREAU

tanguy.loreau@renault.com

Approche multicorps des calculs de Crash

 


    Anthony Meurdefroid

anthony.meurdefroid@supmeca.fr

Dynamique des structures assemblées – Amortissement non linéaire

 


  Marco ROSATELLO

marco.rosatello@supmeca.fr

Amortissement dans les structures assemblées

 


    Emna SGHAIER

emna.sghaier@supmeca.fr

Dynamique des machines tournantes

 


    Hadrien TOURNAIRE

hadrien.tournaire@gmail.com

Modèles réduits pour la dynamique des structures assemblées

 « Méthodologie pour génération de modèles réduits dynamiques multiphysiques : application aux open rotors » Juillet 2017

 

 

Photographies : Florence Dujarric www.florencedujarric.com

Recent publications

Sequential Harmonic Component Tracking For Underdetermined Blind Source Separation in a Multi-Target Tracking Framework (lire)
Model Validation and Uncertainty Quantification, Volume 3 Springer Nature Switzerland 2024 pp. 93-100
Experimental observation of exceptional points in coupled pendulums (lire)
Journal of Sound and Vibration Elsevier 2024 575 pp. 118239
A semi-implicit homogeneous discretized differentiator based on two projectors (lire)
Mechanics & Industry EDP Sciences 2024 Vol. 2024 pp. 10.1051/meca/2024005
Learning structural stress virtual sensors from on-board instrumentation of a commercial aircraft (lire)
Computers & Structures Elsevier 2023 289 pp. 107155
Unsupervised Complex Semi-Binary Matrix Factorization for Activation Sequence Recovery of Quasi-Stationary Sources (lire)
2023
Learning Through Screens During COVID-19 Crisis: Foresee Tomorrow's Education By Analyzing Yesterday's Setbacks And Barriers (lire)
European Society for Engineering Education (SEFI) European Society for Engineering Education (SEFI) 2023
Airplane turbulence detection with hybrid deep learning model (lire)
Surveillance, Vibrations, Shock and Noise 2023
A metaporoelastic structure that overcomes the sound insulation weaknesses of single and double panel partitions (lire)
Applied Acoustics Elsevier 2023 210 pp. 109409
Dynamic Parameter Identification for Cable-Driven Parallel Robots (lire)
The Sixth International Conference on Cable-Driven Parallel Robots (CableCon 2023) 2023
Video Analysis of Nonlinear Systems with Extended Kalman Filtering for Modal Identification (lire)
Nonlinear Dynamics Springer Verlag 2023
Analytic mode-matching for accurate handling of exceptional points in a lined acoustic waveguide (lire)
Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences Royal Society, The 2022 478
Continuous element method for aeroacoustics' waves in confined ducts (lire)
ADVANCES IN NANO RESEARCH TECHNO-PRESS 2022
Geometry-controlled phase transition in vibrated granular media (lire)
Scientific Reports Nature Publishing Group 2022 12 pp. 14989
Unsupervised manufacturing process identification using non-intrusive sensors (lire)
25ème Congrès Français de la Mécanique 2022
Experimental validation of two semi-implicit homogeneous discretized differentiators on the CRAFT cable-driven parallel robot (lire)
CFM2022, Congrès Français de Mécanique 2022
Modèle effectif hystérétique de milieux granulaires cisaillés (lire)
25ème Congrès Français de Mécanique 2022
Localization of higher order exceptional points from finite element model and their applications to duct acoustics (lire)
INTER-NOISE and NOISE-CON Congress and Conference Proceedings 2022 265 pp. 5151-5159
Localized Energy Absorbers in Hertzian Chains (lire)
Physical Review Applied American Physical Society 2022 18 pp. 014078
Finite contact duration modeling of a Vibro-Impact Nonlinear Energy Sink to protect a civil engineering frame structure against seismic events (lire)
Engineering Structures Elsevier 2022 259 pp. 114137
Investigations on the performance and the robustness of a metabsorber designed for structural vibration mitigation (lire)
Mechanical Systems and Signal Processing Elsevier 2022 170 pp. 108830
Localization of high order exceptional points : applications to acoustic waveguides (lire)
16ème Congrès Français d’Acoustique 2022
Geometry-controlled Dynamics of Vibrated Granular Media (lire)
16ème Congrès Français d'Acoustique, CFA2022 2022
Dynamique non-linéaire d’un modèle réduit de milieux granulaires cisaillés (lire)
16e Congrès Français d'Acoustique 2022
Dynamique non-linéaire d’un modèle réduit de milieux granulaires cisaillés (lire)
16ème Congrès Français d'Acoustique, CFA2022 2022
Voir toutes les publications (HAL)

Our activities are focused on:

Viscoelasticity

Understanding viscoelasticity in just a few minutes…

Polymeric materials possess significant vibration-energy dissipation capabilities. They are used to protect structures, equipment and passengers from vibrations, shock and noise.

 

 

Challenges Related to the Experimental Characterisation of Viscoelastic Materials

In order to characterise these materials their dynamic stiffness and loss angle are typically represented according to the loading frequency. These experimental observations highlight a noticeable evolution of these values depending on the frequency ranges (figure 2).

 

Figure 2: frequency representations

The experimental characterisation of these properties — dynamic mechanical analysis — is a delicate process because the frequency range that testing should cover is often very broad and difficult to access with a single experimental device. Besides, the mechanical properties of viscoelastic materials are highly temperature-dependent. One solution consists in using the time-temperature superposition principle. This method is based on vibration tests covering a relatively narrow frequency range and conducted at various temperatures. Time-temperature superposition then enables to reconstruct the dynamic behaviour of a given material over a broad frequency range and at a given temperature.

There are several measurement biases with this method. The main one stems from the time-temperature superposition hypothesis which is not applicable to all materials. Time-temperature superposition is not relevant anymore for modern materials that have emerged from the industrial world such as multilayer materials, filled polymers, functionalised materials or structural adhesives. Therefore, the temperatures and frequency ranges used for testing must be consistent with the real-life operating conditions of such materials in order to characterise their dynamic properties.

For several decades the VAST team has been developing innovative, high-performance viscoanalysers that enable to characterise the dynamic mechanical properties of these materials directly and under different loading configurations (tension-compression, shear, combined static and dynamic loading, etc.) These new experimental technologies (figure 3) currently allow for the direct characterisation of viscoelastic materials up to 10 kHz (without using time-temperature superposition or modes that are specific to the experimental device).

 

Figure 3: three viscoanalysers developed at Supméca

Therefore, conducting these tests at various preload values and loading amplitudes enables to characterise the nonlinear viscoelasticity of the studied material precisely.

 The quality of the experimental data acquired from these tests is a key issue when predicting the dynamic behaviours of the components and structures that these viscoelastic materials are used in.

 

Challenges Related to the Modelling of Viscoelastic Materials

The main purpose of acquiring experimental data on the dynamic behaviours of viscoelastic materials is to feed prediction models for components, structures or systems.

Viscoelastic mechanical behaviours can be separated in two main categories of models: those based on fractional derivatives and those based on elementary rheological components such as the Generalized Maxwell model.

The first bottleneck, modelling-bottleneck, stems from the identification of the parameters of these models. This point is covered in several publications of the VAST team [1 – 3] and valued by several industrial partners (Hutchinson, BOSCH braking systems, AER, ADERIS).

The second bottleneck tackled by the VAST team concerns the integration of these models into commercial, structure calculation codes [4, 5].

Viscoelasticity has been a particularly active research-area for decades. New fields of study are currently emerging with functionalised materials, metamaterials, additive manufacturing and bio-based or environment-friendly and sustainable-development compatible materials.

 

 

[1] J.-L. Dion and S. Vialard, “Identification or rubber shock absorber mounts,” Mécanique industrielle et matériaux, vol. 50, no. 5, pp. 232–237, 1997.

[2] F. Renaud, J.-L. Dion, G. Chevallier, I. Tawfiq, and R. Lemaire, “A new identification method of viscoelastic behavior: Application to the generalized maxwell model,” Mechanical Systems and Signal Processing, vol. 25, no. 3, pp. 991 – 1010, 2011.

[3] H. Jrad, J. L. Dion, F. Renaud, I. Tawfiq, and M. Haddar, “Experimental characterization, modeling and parametric identification of the non linear dynamic behavior of viscoelastic components,” European Journal of Mechanics – A/Solids, vol. 42, no. 0, pp. 176 – 187, 2013.

[4] S. Thouviot, G. Chevallier, F. Renaud, J.-L. Dion, and R. Lemaire, “Prise en compte des comportements viscoélastiques dans la simulation dynamique des systèmes de freinage,” Mechanics & Industry, vol. 10, no. 05, pp. 385–396, 2009.

[5] H. Festjens, G. Chevallier, F. Renaud, J.-L. Dion, and R. Lemaire, “Effectiveness of multilayer viscoelastic insulators to prevent occurrences of brake squeal: A numerical study,” Applied Acoustics, vol. 73, no. 11, pp. 1121 – 1128, 2012.

Nonlinear Waves and Vibrations in Granular Media

We take an interest in the propagation and attenuation of waves and vibrations in granular media such as sand through experimentation, using laboratory model systems. Granular media are specific in that (i) when dry, they interact according to the strongly nonlinear Hertz potential. Also, more generally speaking (ii) they mobilise singular mechanisms associated with the contacts that occur between particles that have no geometric conformity. For example, this is the case with wet granular media, colloidal suspensions in particular, when interstitial fluid is confined between the particles and induces a complex elasto-hydrodynamic interaction.

 

Propagation d’ondes dans un cristal phononique

Wave propagation in a phononic crystal composed of an alignment of centimetre-scale particles that interact via the nonlinear Hertz potential; an instrumented particle enables to highlight the band gaps of such a network (PRL 2005, PRL 2010).

 

Propagation d’ondes dans un milieu granulaire mouillé

Wave propagation in a wet granular medium (by addition of a low quantity of viscous interstitial fluid); measurement of the dispersion relationship shows that the fluid induces an elasto-hydrodynamic interaction which translates into contact stiffness, an increase in propagation speed and greater dissipation (Kamil Chrzaszcz, 2016 Thesis)

 

Observation directe de la propagation d’une impulsion dans une suspension colloïdale

Direct observation of impulse propagation in a colloidal suspension; demonstration of the role played by particle elasticity, fluid viscosity and topological disorder on attenuation length and transport properties (PNAS 2017).

 

Another advantage (iii) of granular media is to exhibit crystalline symmetry when the particles are identical or, on the contrary (iv) disorder when the grains are polydisperse. Finally, they are known (v) for being highly dissipative by way of mobilising strong frictional effects for instance.

 

The research conducted in the laboratory, primarily of experimental nature, covers several scales (centimetre, millimetre and micrometre scales) and ranges from contact dynamics (dry or wet) to the disordered stacking of vibrated grains (granular dampers), not to mention model structures (phononic crystals composed of periodically arranged spheres).

 

 

Mesure du facteur de perte et de la masse effective d’un dissipateur granulaire

Measurement of a granular damper’s loss factor and effective mass; this system’s loss factor is proportional to the mass of the vibrated grains, the apparent mass of which decreases with acceleration amplitude until it disappears (Marwa Masmoudi, Gran. Matt., 2016 thesis).

Acoustic Dissipation and Metamaterials

Acoustic absorption and attenuation are key issues in several areas of construction, transportation or ventilation systems. These practical aspects are related to acoustic propagation in thermoviscous fluids and subwavelength structuring being taken into consideration.

These mechanisms form the basis of the models that concern porous, poroelastic and metamaterials where the competing interaction between resonances and dissipation is crucial for their functioning (2013-2017 METAUDIBLE Project, National Research Agency (ANR)) 

 

Waveguide attenuation using a metaporous material. Guided wave, metamaterials and homogenisation.

 

 

Taking losses into account remains a challenge when conducting simulations and it enables to explore unusual wave phenomena such as exceptional points (collaboration with the University of Technology of Compiègne (UTC)).

In addition to simulation methods, such as EasterEig project, the team also develops novel materials and the related resources required for experimentation.  

 

Metamaterials used for sound attenuation in an acoustic duct (Collaboration: LAUM Xiong, Aurégan, Bi https://dx.doi.org/10.1121/1.5007851)

 

Example of the microstructure of polymeric foam (SUPMECA SEM micrography)  

Exceptional point on a Riemann surface (merging of two eigenvalues)

                                                  

Damping and Energy Dissipation in Various Assemblies

When designing a structure, predicting its dynamic behaviour is often marred with errors caused by poor assessment of its overall damping capacity. This vibration level greatly depends on the damping induced by the assembling of substructures together. This damping results from an energy loss by friction caused by partial sliding in the contact interfaces. The purpose of this research is to focus on this energy that is dissipated by friction that appears when vibrations occur in the structure which influences its dynamic behaviour.

 

The conducted research work is aimed at:

  • quantifying the described damping by experimentation,
  • evaluating said damping numerically at various scales (contact occurring at the micro-geometric scale, linkage, structure).

Moreover, recent projects are aimed at designing damping linkage technologies for metallic assemblies and building the related prototypes. Modern structures must combine reliability, long service-life, comfort, lightness, low costs, speed and reduced energy consumption, which explains the relevance of this research. Even though they are all intrinsically different, an increase in damping in a given structure enables to improve all of these parameters when observed separately. Reducing vibrations is always the result of a compromise between these parameters. As such, an increase in damping does not enable to improve all of them at once. The goal of these projects is to maximise damping increases in mechanical linkages and contribute to the improvement of structure performance more generally speaking.

 

The research projects are conducted with support from various organisms and industrial partners in the frame of several research programs (the Research & Technology program (R&T) of the National Centre for Space Studies (CNES), Unique Inter-ministry Fund projects (MAIAS, CLIMA), collaboration with the SystemX research institute (IRT), EDF, ASTECH, JPB SYSTEM, AIRBUS, SOPEMEA, AVNIR, SDTOOLS, FCBA).

Measurement of Vibration Fields Using High-Speed Cameras

Vibration analysis is of prime importance for mechanical systems and structures in order to determine their behaviour in operating conditions or subsequently to exceptional events. Therefore, conducting experimental studies on the vibrations occurring in a system (be it mechanical, aeronautical or civil) is now an essential step of the design and monitoring processes. To this end, in addition to standard measurement tools such as piezoelectric accelerometers, laser motion sensors and laser vibrometers, high-speed camera systems appear to be promising because of the contactless recording of large scenes at very high rates, even thousands of frames per second.

Therefore, measurements performed with high-speed cameras enable to define vibration fields for a given structure in both the transient and steady states. Using several cameras to surround the structure enables to measure its behaviour in 3D and detect oscillations from elements that are difficult to reach with standard measurement tools. For example that is the case with the edges of a contact interface within a linkage, or the free surface of a fluid contained in a moving, inflexible container.

 

 

Therefore, our studies focus on exploiting this new technology, with a special interest in the below:

  • image processing (binarization, code optimisation, calculation-time reduction);
  • feature detection (points, edges, surfaces, fluids);
  • marker tracking;
  • data filtering and assimilation;
  • operational modal analysis using multi-view 3D visualisation;
  • reconstruction of the behaviour of the 3D structure based on the models and measurements;
  • real-time vibration analysis.

To be more precise, the developed image-processing algorithms enable to detect features (or points of interest) and track how they move during dynamic testing. Signal-filtering and data-assimilation techniques thus provide the information necessary for experimental modal analysis and to identify the dynamic properties of a given system. Our research is also aimed at developing real-time, vibration-behaviour analysis-methods, while relying on the complementary nature of experimental measurements and numerical models.

 

Grayscale image / Binarized image / Feature detection

 

Experimental measurement campaigns are conducted in the VAST team’s test laboratory as well as in the frame of various research projects (the CLIMA and EUGENE Unique Inter-ministry Fund (FUI) projects) and collaborations with academic institutes and industrial partners (University of Liege, SOPEMEA). 

 

Understanding field measurement with high-speed cameras in just a few minutes…

Data Assimilation Methods Applied to Structural Dynamics

Alongside the development of field measurement methods applied to structural dynamics, the VAST team also conducts research on data assimilation methods aimed at enhancing the connection between a given dynamic system and its digital twin with real-time data feeding.

The developed methods are based on Bayesian stochastic approaches (Kalman filtering) and built with state observers derived from structural dynamics and rigid-body dynamics.

The conducted data-assimilation studies make use of a high number of various sensor technologies that include high-speed cameras, accelerometers, force sensors and more, and their aim is to increase model quality and build upon the experience acquired through simulations and measurements.

 

The development of these techniques concerns various applications (structural dynamics, multibody system dynamics, dynamic failure-mode analysis of a production line) and it involves various industrial partners from the aerospace, automotive and cosmetics industries (AIRBUS, SOPEMEA, PSA, PUIG, PKB, VISIOLASER)

 

Understanding digital twins in just a few minutes…

Scale Models in Structural Dynamics

Analysing the dynamic behaviours of structures, for instance when studying their linear or nonlinear vibrations, as well as coupled multiphysics or vibroacoustics problems, problems related to transient response and short-duration loadings, all lead to building models that require extreme numerical resources. Such models are often difficult to exploit or they prove unsuitable when such analyses are used for iterative optimisation processes, when exploring various configurations for a given system at the pre-design stage in varied and/or broad frequency ranges, but also when designing real-time diagnostic systems or when implementing those in embedded mechatronic systems.

Scale models are one of the means that the VAST team has been developing and using for about twenty years in order to address these problems. These models provide effective and robust simulation tools that are perfectly suited to the type of simulation work that is envisioned and the available numerical resources. The model-scaling techniques that are classically used in structural dynamics, which are based on the exploitation of modal bases and on experimental modal analysis, are applied to nonlinear and multiphysics problems. Current efforts are aimed at complementing these approaches with models developed using meshfree methods. Continuous models of structural elements made from orthotropic materials are currently being developed in the frame of the Dynamic Stiffness Method and applied to the design of smart structures for example.

Frequent industrial partnerships enable to leverage these approaches in order to develop operational tools that meet designers’ requirements as closely as possible. (Renault, Thales, etc.)

A continuous model developed using a meshfree method