PhD position in Plasma Instability Identification through Machine Learning
Aix-Marseille Université

PhD position in Plasma Instability Identification through Machine Learning

France 01 Jun 2021

ABOUT THE INSTITUTION

Aix-Marseille Université is a research-intensive university of international standing, yet deeply rooted in its territory. Its reputation enjoys wide recognition and it was granted, in 2016, the natio
Visit institution page

OPPORTUNITY DETAILS

State University
Area
Host Country
Deadline
01 Jun 2021
Study level
Opportunity type
PhD
Specialities
Eligible Countries
This opportunity is destined for all countries
Eligible Region
All Regions

RESEARCH FIELD(S)

Nuclear fusion, modelling and simulation, artificial Intelligence, statistical methods

JOB LOCATION

PIIM (Physique des Interactions Ioniques et Moléculaires), Avenue Escadrille Normandie Niemen, 13013 Marseille, in close collaboration with ITER-IO.

JOB DESCRIPTION

Aix-Marseille Université’s 3-year PhD contract

Abstract

Understanding, predicting and controlling plasma instabilities is of major importance on the route towards achieving steady-state burning plasma scenarios where energetic particles must remain well confined in order to transfer their energy to the thermal plasma through Coulomb collisions. In this respect, accurate identification of the various plasma instabilities that might tend to de-confine particles and their energy is required in order to develop possible control techniques. Advanced simulations using experimental data allow off-line identification of the instabilities but are insufficient for real-time control applications. The proposed PhD topic will focus on the identification of plasma instabilities using state-of-the-art Artificial Intelligence (AI) techniques aiming to identify instabilities in real time. For this purpose, the PhD candidate will address three missions. First, building a database using synthetic experimental data for various calculated instabilities such as MHD and energetic particle driven modes. Second, training a multi-class classification model based on supervised algorithms. Finally, applying the model to test cases and experimental data to investigate the portability of the AI model between machines.

QUALIFICATIONS/SKILLS/EDUCATION & RESERACH REQUIREMENTS/DUTIES

Master’s degree in one of the following areas: physics, applied mathematics, data science. Candidates must have a strong interest in numerical modelling and simulations applied to plasma physics. Expertise in Python and knowledge in artificial intelligence tools are required.

APPLICATION DEADLINE

June 1st 2021

REQUESTED DOCUMENTS OF APPLICATION

CONTACT TO APPLY

Please send your application and any information request at the two following addresses: isfin-direction@univ-amu.fr and david.zarzoso-fernandez@univ-amu.fr

Choose your study destination


Choose the country you wish to travel to study for free, work or volunteer

Please find also