F9 - Digital Image Correlation: a tool for
strengthening the experiment/simulation dialog


Organizers: JM. Perié & JC. Passieux

Jean-Noël Périé is a former student of the Ecole Normale Supérieure de Cachan. He passed the Agrégation (french professional teaching qualification) in 1994 and graduated in 1995. He obtained his PhD in 2000 from Ecole Normale Supérieure de Cachan under the supervision of Professor P. Ladevèze on the topic of "Damage meso-modelling of a 2.5D C/C composite". He passed his habilitation thesis to supervise PhDs on "Digital images and mechanics of composite materials and biological tissues". He is now Associate Professor at "Université Paul Sabatier". He teaches mainly in the mechanical engineering department of the "IUT 'A' Paul Sabatier", but he also gives master classes focused on "Photomechanics". He conducts his research at Clement Ader Institute. He is also the head of the team "Identification and Control of Thermic and Mechanical Properties" (5 researchers). His main interests are the experiment/simulation dialog and the identification of constitutive parameters. In particular, he developed identification techniques to retrieve constitutive parameters from full field measurements. He participated several times to CNRS summer school, and he has coauthored chapters in 3 books on this subject. He has also been involved in the development of several measurement techniques such as Digital Image Correlation (DIC), Digital Volume Correlation (DVC), Stereo Digital Image Correlation (SDIC), Multiscale DIC, etc. In addition, He has significant experience in the characterization of composite materials, and more generally in experimental mechanics.




Jean-Charles Passieux has graduated from the Ecole Normale Supérieure de Cachan (ENS Cachan) in 2003. He obtained his PhD in Computational Mechanics from Ecole Normale Supérieure de Cachan under supervision of Professor P. Ladevèze on the topic of domain decomposition methods, parallel computing and reduced order modelling in 2008. He then obtained his "Habilitation à Diriger des Recherches" (HDR, degree required to supervise PhDs) on the development on numerical methods in computational and experimental mechanics and aeronautics. He is Associate Professor at "Institut National des Sciences Appliquées de Toulouse" in the mechanical engineering department and at Clement Ader Institute. He worked on several projects concerning numerical methods in Digital Image Correlation (DIC) , Digital Volume Correlation (DVC), Stereoscopic Digital Image Correlation (S-DIC), identification of mechanical properties, Multiscale DIC, Parallel Computing and High Performance Computing (HPC) in DIC. He also has a solid background in computational mechanics and in the Finite Element modelling of lightweight aeronautical structures, up to the simulation of crack propagation and post-buckling of stiffened panels. He is the head of the first year of the Master in mechanical engineering (80 students) at INSA de Toulouse. He is also the head of the team "Integrity of Structures and Systems" (11 researchers).


Contacts:
jean-noel.perie@iut-tlse3.fr
passieux@insa-toulouse.fr

Digital Image Correlation is now commonly used in experimental mechanics (cf. Forum 2). The widespread subset based approach has proven to be a versatile, convenient and remarkable tool for better understanding the behavior of material and structures in a wide amount of situations (non-conventional testing, multiscale investigations, dynamic loadings, etc.). However, it is not always obvious to take advantage of the DIC kinematical measurements for validating or identifying constitutive models. Many reasons may indeed jeopardize a meaningful dialogue between simulations and experiments. For instance:

(1) Measured and simulated quantities are usually not expressed within the same framework (e.g. different reference frames and kinematical descriptions)

(2) Measurements are always prone to noise, and the noise is closely related to the acquisition and the DIC settings. In practice, one has to face a complex compromise between spatial resolution and displacement resolution

(3) Boundary conditions used in the simulation may not reflect the actual ones.

To circumvent some of these issues, new approaches to DIC have been introduced approximately 10 years ago. They allow taking advantage of an a priori knowledge and to resort conveniently to the simulation toolkit. As a result, measurements can be more naturally connected to simulations, the measurements can be conveniently regularized (e.g. thanks to a reasonable mechanical model) and, ultimately, the simulation can eventually be integrated to the measurement. However, this kind of approach may involve a greater computational effort and some questions are still open. This forum invites presenters who are interested in presenting their experience in the experiment/simulation dialog, new procedures or new tools aiming at bridging the gap between experiments and simulations.




Proposed strategy of simulation/experiment dialog in the context of structural testing performed on large composite panels (Project VERTEX ANR-12-RMNP-0001). J.-E. Pierré, J.-C. Passieux and J.-N. Périé. Finite Element Stereo Digital Image Correlation: framework and mechanical regularization. Experimental Mechanics. 53(7)443-456. 2017



Non-exhaustive list of suggested Topics:

- Validation and Identification from DIC measurements

- Global DIC methods

- Multiscale measurement

- Numerical methods for measurement and identification