PhD opportunities

Physics-informed Deep Learning for the Understanding of Mesostructure Effects on the Mechanical Failure of Reinforced Polymers

Thesis proposal

Area of expertiseMécanique
Doctoral SchoolISMME - Ingénierie des Systèmes, Matériaux, Mécanique, Énergétique
SupervisorLAIARINANDRASANA Lucien
Co-supervisorOVALLE-RODAS Cristian
Research unitCentre des Matériaux
KeywordsMesostructure , Polymers
AbstractIn order to find efficient, low-density engineering materials resistant to complex load conditions while ensuring thermal stability, semi-crystalline polymers have been frequently used and reinforced by rigid secondary phases. This reinforcement phase (formed by fibres, balls, … etc) provides the material with the required mechanical properties. In order to achieve these properties the reinforcement must meet certain structural characteristics:
- Global and “controllable” parameters: volume fraction of various material phases, geometrical properties.
- Local and “uncontrollable” parameters: spatial distribution, orientation.
Local deviations from these characteristics result in a concentration of the strain locally and a consequent loss of mechanical properties at a global domain.
Besides the well-known damage micro-mechanisms in semi-crystalline polymers, additional processes come into play when dealing with reinforced semi-crystalline polymers. An early damage scenario supported by surface microscopy observations of post-mortem specimens was published by Sato et al. [1]. It was recently updated by using Synchrotron Radiation Computed Tomography in Rolland et al. [2] using in situ tensile tests to observe and describe damage features in a volume located in the constriction of a cylindrical specimen. Using an accurate phase segmentation method, including reinforcements, matrix, and cavities, they identified typical mechanisms such as damage at fibre ends, debonding, fibre failure, and, eventually, matrix damage in 3D. Quantification of new damage markers was reported at different loading steps both in tensile and fatigue tests. Moreover, the influence of humidity, as well as the orientation gradient of fibres in reinforced PolyAmide 6,6, were reported. Yet, no clear damage quantification, combining both fibre and damage data at a sufficiently large scale to explore the influence of the core-shell effect, was conducted. Taking into account this mesostructure gradient (due to fibre orientation) would allow better prediction of fracture as it is expected to compete with the stress triaxiality field.
From the industrial point of view, a fine analysis of mesostructure characteristics of the polymers’ reinforcement is essential throughout the entire life-cycle of products: i) design and optimization of manufacturing processes, ii) control (after manufacturing, or periodic) and also iii) repair. This analysis however is currently not feasible with classical tools. Hence the need of using AI techniques (and specifically deep models) capable of inferring unknown effects of (even hidden) variables in complex processes.
ProfileTypical profile for a thesis at MINES ParisTech: Engineer and / or Master of Science - Good level of general and scientific culture. Good level of knowledge of French (B2 level in french is required) and English. (B2 level in english is required) Good analytical, synthesis, innovation and communication skills. Qualities of adaptability and creativity. Teaching skills. Motivation for research activity. Coherent professional project.

Prerequisite (specific skills for this thesis):
● A successful applicant has accomplished a 2-year master in machine learning and AI techniques. An engineer-level knowledge in material science is welcome but not mandatory.
● The Centre for Mathematical Morphology partners with a German Fraunhofer institute in a french-german doctoral college programme that offers exchange stays to PhD candidates and organizes workshops on a yearly basis. The PhD candidate will be offered the possibility to make a few-months international exchange stay and contribute to this workshop.
Mobility condition : eligible candidates must not have resided in France more than 12 months during the last three years.

Applicants should supply the following :
• a detailed resume
• a copy of the identity card or passport
• a covering letter explaining the applicant’s motivation for the position
• detailed exam results
• two references : the name and contact details of at least two people who could be contacted
• to provide an appreciation of the candidate
• Your notes of M1, M2
• level of English equivalent TOEIC
to be sent to recrutement_these@mat.mines-paristech.fr
FundingFinancement d'un Etablissement d'enseignement supérieur