Defense of Quentin Marcot (mechanics department)

The aim of this thesis work is to implement a test featuring dual mechanical and thermal heterogeneity, with a view to identifying material behavior laws by inverse method.

In the aerospace industry, certification programs are increasingly relying on modeling and digital simulation to reduce the number of physical tests performed on structures.

These virtual certification processes rely heavily on the predictive capability of behavior laws for materials subjected to complex thermomechanical loading.

Standard material characterization methods are generally performed on standardized samples and rely on static equilibrium (i.e., statically determined) assumptions that are difficult to meet in practice, leading to the need for a large number of tests.

With the emergence of optical techniques for measuring kinematic and thermal fields, growing interest has developed in the use of statically indeterminate approaches, enabling the calibration of behavior models from heterogeneous information, and thus reducing the number of tests required. This new testing paradigm has been named Material Testing 2.0.

Until now, these methods have mainly been applied to tests featuring heterogeneities in terms of deformations and strain rates.

Although recent studies have exploited tests with dual heterogeneity in strain rate and temperature, these have used temperature gradients generated using Gleeble-type thermomechanical testing machines, which limits both the attainable temperature range and the spatial distribution of temperatures.

In this work, a new experimental methodology is developed for introducing a thermal gradient into the test, allowing control over both the temperature range generated and the position of the heat source.

For this proof of concept, a viscoelastic polymer (PMMA) was chosen because of its sensitivity to temperature in relatively low ranges. The Virtual Field Method is used for material characterization.

Numerical work demonstrates the reliability of this approach for the identification of linear viscoelastic behavior, both under isothermal and non-isothermal conditions.

Concerning the experimental part, a thermal gradient generation system is set up.

Then, an analysis methodology is introduced, relying on tools enabling Lagrangian tracking of kinematic and thermal data in the same reference frame.

The measurement biases associated with the introduction of a heterogeneous temperature field into the test are then analyzed.

Finally, an experimental campaign  is conducted based on the methodology developed for PMMA characterization. 

• M.Bertrand LANGRAND DMAS, ONERA, Thesis co-director
• Mr Benoît BLAYSAT, Université Clermont-Auvergne, Rapporteur
• Mr Jean-Luc BOUVARD, CEMEF - Mines Paris | PSL Research University, Rapporteur
• Ms Julie DIANI, Laboratoire de Mécanique des Solides, Ecole Polytechnique, Examiner
• Mr Rian SEGHIR, Nantes Université, Ecole Centrale Nantes, CNRS, GeM, UMR 6183, Examiner
• Mrs Delphine NOTTA-CUVIER, LAMIH UMR CNRS 8201 - Département Mécanique (Department of Mechanics), Director of International Master Transportation & Energy INSA Hauts-de-France, Université Polytechnique Hauts-de-France, Examiner
• Mr Thomas FOUREST DMAS, ONERA, Thesis co-supervisor
• Mr Fabrice PIERRON, MatchID Nv, Thesis co-supervisor