Bienvenue aux Mines Paristech
Bienvenue à MINES ParisTech
Newsletter International
Website
Théorie & Pratique
Vous êtes

webTV

Lecture

Séminaire PSL - Écosystèmes de médias I Session 4 Partie 4

Lecture

Séminaire PSL - Écosystèmes de médias I Session 4 Partie 2

Lecture

Séminaire PSL - Écosystèmes de médias I Session 4 Partie 3

Lecture

Séminaire PSL - Écosystèmes de médias I Session 4 Partie 1

Lecture

Lancement de la Chaire DEEP unissant HUTCHINSON, l'ESPCI et MINES ParisTech

+ Toutes les vidéos

Partager

SUJET POURVU - Apprentissage artificiel de modèles réduits physiques pour la prévision rapide de la durée de vie d'aubes de turbine avec quantification d'incertitudes

SUJET POURVU - Apprentissage artificiel de modèles réduits physiques pour la prévision rapide de la durée de vie d'aubes de turbine avec quantification d'incertitudes

SUBJECT OF THESIS PROVIDED - MACHINE LEARNING OF PHYSICAL REDUCED ORDER MODELS FOR FAST LIFETIME COMPUTATION OF TURBINE BLADES WITH UNCERTAINTY QUANTIFICATION

Proposition de thèse

Spécialité

Mécanique

Ecole doctorale

SMI - Sciences des Métiers de l'Ingénieur

Directeur de thèse

RYCKELYNCK David

Unité de recherche

Centre des Matériaux

ContactDavid RYCKELINCK
Date de validité

01/10/2018

Site Webhttp://www.mat.mines-paristech.fr/Recrutements/Theses/
Mots-clés

Apprentissage automatique, Science des données, Approximation de base réduite, Plasticité, Contact, Multiphysique

Machine learning, Data science, Reduced basis approximation, plasticity, contact, multiphysics

Résumé

The Centre des Materiaux located at present in Evry (35km south of Paris) is a laboratory associated with the CNRS, employing around 200 people including 30 researchers, 50 ITA, 85 PhD students and 11 Post-Doctoral researchers.
Research concerns materials processing and surface modification, the
microstructural characterization and experimental study of the behaviour of
materials. These studies are carried out in close contractual collaboration with industrial partners.

Safran Group is a leader in the design, manufacturing and maintenance of aircraft engines and equipment. As in the vast majority of industrial domains, the numerical simulation is a tool used in many stages of Safran's activities. The complexity of the models leads to computation times of several hours (or even days) for a single run, although optimization and uncertainty quantifications require many runs: new strategies must be derived.

In this thesis, we are interested in the lifetime computation of high-pressure turbine blades. The computation chain contains a coupled aerothermal fluid-solid computation and an ElastoViscoPlastic (EVP) cyclic computation of which the lifetime computation is a post-treatment.

The Centre des Materiaux located at present in Evry (35km south of Paris) is a laboratory associated with the CNRS, employing around 200 people including 30 researchers, 50 ITA, 85 PhD students and 11 Post-Doctoral researchers.
Research concerns materials processing and surface modification, the
microstructural characterization and experimental study of the behaviour of
materials. These studies are carried out in close contractual collaboration with industrial partners.

Safran Group is a leader in the design, manufacturing and maintenance of aircraft engines and equipment. As in the vast majority of industrial domains, the numerical simulation is a tool used in many stages of Safran's activities. The complexity of the models leads to computation times of several hours (or even days) for a single run, although optimization and uncertainty quantifications require many runs: new strategies must be derived.

In this thesis, we are interested in the lifetime computation of high-pressure turbine blades. The computation chain contains a coupled aerothermal fluid-solid computation and an ElastoViscoPlastic (EVP) cyclic computation of which the lifetime computation is a post-treatment.



Contexte

The student will keep in mind the opportunity to optimize each step by making them goal-oriented (dedicated to lifetime computations. Theoretical questions on the unicity of the stabilized cycle, or the conditions for the reduced models to converge towards the stabilized cycle related of the full-order model can be addressed. Finally, the constructed methodology will be applied to an uncertainty quantification study on the lifetime computation of high-turbines blades, for which loading cycles are not accurately known despite having an important influence on the lifetime.
The thesis will start in October 2018 and be part time at Safran and Ecole des Mines ParisTech.

Encadrement

Directeur de thèse : David RYCKELINCK - Centre des Matériaux

Profil candidat

Ingénieur et/ou Master recherche - Bon niveau de culture générale et scientifique. Bon niveau de pratique du français et de l'anglais (niveau B2 ou équivalent minimum). Bonnes capacités d'analyse, de synthèse, d'innovation et de communication. Qualités d'adaptabilité et de créativité. Capacités pédagogiques. Motivation pour l'activité de recherche. Projet professionnel cohérent.

Pour postuler : Envoyer votre dossier à recrutement_these@mat.mines-paristech.fr comportant
• un curriculum vitae détaillé
• une lettre de motivation/projet personnel
• des relevés de notes L3, M1, M2
• 2 lettres de recommandation
• Une attestation de niveau d'anglais

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.

Applicants should supply the following :

- a detailed resume
- a covering letter explaining the applicantÂ'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

to be sent to recrutement_these@mat.mines-paristech.fr

Objectif

The objective of the thesis is to increase the speed of computation of the EVP cyclic part, by constructing a dictionary of reduced order models using machine learning tools. First, a thorough bibliographic study on physical reduced order models (including the hyperreduction method) and machine learning (including clustering algorithms and neural networks) is carried out. Then, the student will propose a first hyperreduced model using the code Zset, on a mesh of industrial complexity, and the HPC solver based on a FETI domain decomposition algorithm. Then, a clustering algorithm is applied to a collection of numerous low-fidelity computations in order to identify a dictionary of possible models for life time prediction. In the exploitation
stage, an efficient criterion based on classification methods taken from the data science literature is derived to decide in real time which prediction model is the best for fast and accurate prediction.

Références

/

Type financement

Convention CIFRE

Retour à la liste des propositions

actualité

Vladislav Yastrebov, Médaille de bronze CNRS 2018

Formation Vladislav Yastrebov, Médaille de bronze CNRS 2018 " La médaille de bronze récompense un premier travail…
> En savoir +

Plenesys, spin-off de l'École, Grand prix i-Lab 2018

Formation Plenesys, spin-off de l'École, Grand prix i-Lab… Belle reconnaissance pour la recherche conduite au sein du…
> En savoir +

Marguerite Champion galope vers la victoire

Formation Marguerite Champion galope vers la victoire Marguerite Champion, discrète élève ingénieure de 1 re année,…
> En savoir +

Le Rapport d'activité 2017 est en ligne

Formation Le Rapport d'activité 2017 est en ligne « L'ambition de MINES ParisTech : former des…
> En savoir +

Pluie de récompenses pour le CGS

Formation Pluie de récompenses pour le CGS La FNEGE, qui célèbre cette année ses 50 ans, organise la…
> En savoir +

+ Toutes les actualités

agenda

Du 10 septembre au 27 novembre 2018 Mooc Conversion thermodynamique de la…

Le 11 septembre 2018 10 e anniversaire de la Chaire "Modélisation…

Du 18 septembre au 14 décembre 2018 Musique aux Mines 2018

Du 22 octobre au 3 décembre 2018 Mooc Conversion thermodynamique de la…

+ Tous les événements

contact

Régine MOLINS
Direction de l'Enseignement
Service du Doctorat
> envoyer un mail

Plan du site
MINES
ParisTech

60, Boulevard Saint-Michel
75272 PARIS Cedex 06
Tél. : +33 1 40 51 90 00

Implantations
Formation
Mentions légales | efil.fr | ©2012 MINES ParisTech | +33 1 40 51 90 00 |