🎓Guillaume BOGOPOLSKY thesis defense
Monday 10 February 2025 at 9h30
Salle JCA, Cerfacs, Toulouse
Exploration of methods for the numerical simulation of Hall effect thrusters
ED MEGEP – [Subject to defense authorization]
https://youtube.com/live/8lX7ceLACbg?feature=share
In the last fifteen years, the number of satellites orbiting Earth has shown an exponential increase. The high cost of space launches has driven manufacturers towards more efficient propulsion technologies, such as electric thrusters. Hall thrusters belong to this category, and have shown remarkable efficiency and life time for orbital transfers, orbital station- keeping and space exploration applications. However, their physics is not yet fully understood, such as the abnormal erosion of the ceramic material at the thruster’s exit plane, and the various instabilities that appear across a range of frequencies, including rotating spokes, breathing modes, sheath instabilities and electron drift instabilities. Notably, they prevent a simple modeling of the thrusters and thus a simple scaling of the technology in order to reach higher thrust and optimize efficiency, which has to be done empirically.
Given these challenges, computer simulations have emerged as a crucial tool for advancing our understanding and predicting the behavior of Hall thrusters. To this end, CERFACS, in collaboration with Safran Spacecraft Propulsion and the Laboratoire de Physique des Plasmas, has been developing the AVIP code since 2016. AVIP is a massively parallelized simulation tool that employs an unstructured grid and supports both Lagrangian (Particle-in-Cell, or PIC) and Eulerian (fluid) approaches, offering flexibility in modeling different aspects of the thruster’s plasma dynamics. Initial results using AVIP enabled detailed studies of the azimuthal instabilities characteristic of Hall thrusters.
The primary objective of this thesis was to investigate and finalize a version of AVIP that can perform high-fidelity simulations of Hall thrusters with a balance between computational cost and precision, leveraging both PIC and fluid models. We have explicited the implementation details of the required elements for such as solver in a parallel unstructured framework based on academical and industrial solver AVBP, and detailed the numerical and optimization challenges that arise. Then, its capacity to perform plasma computation is evaluated on a 1D benchmark setup and a 2D setup representative of a Hall thruster and showing traces of the Electron Cyclotron Drift Instability. Then, a Penning discharge is computed in the context of an international benchmark using the PIC method in order to verify our ability to capture spoke instabilities and setup a reference case for future fluid simulations.
Jury
Khaled Hassouni | Professor – LSPM, CNRS | Reviewer |
Francesco Taccogna | Researcher– ISTP, CNR | Reviewer |
Anne Bourdon | Professor– LPP, CNRS | Examiner |
Laurent Garrigues | Professor– LAPLACE, CNRS | Examiner |
Bénédicte Cuenot | Engineer – Safran Aircraft Engines | Thesis supervisor |
Benjamin Laurent | Engineer– Safran Spacecraft Propulsion | Invited member |
Nicolas Barléon | Researcher– CERFACS | Invited member |