Offre de postdoc sur la simulation haute fidélité en mécanique des fluides compressibles et réactifs (simulation numérique directe ou aux grandes échelles) et l'adaptation d'un code de calcul aux nouvelles architectures hétérogènes (CPU/GPU).

This project aims to model the dynamics of suspended cabins for urban applications. During the project stability and resolvent analysis will be used to educe the key physical mechanisms of the system and to guide the construction of reduced-order models (ROMs). Ce projet vise à modéliser la dynamique des cabines suspendues pour des applications urbaines. Au cours du projet, l'analyse de stabilité et de résolvant sera utilisée pour déduire les mécanismes physiques clés du système et pour guider la construction de modèles d'ordre réduit (ROM).

Building on recent advances in mean-field-based linearised models for turbulent jet dynamics and aeroacoustics, the 2AT group has developed resolvent-based tools for jet-noise modelling, estimation, and control. The postdoc will collaborate with Airbus and 2AT researchers to extend these tools to fully three-dimensional mean fields. We aim to develop a high-order finite-difference resolvent solver, incorporating stabilising filters and a domain-decomposition strategy suitable for complex geometries. Efficient time-integration methods will be employed to compute resolvent modes. The resulting framework will support the development of next-generation models for turbulent jets and jet noise.

Building on recent advances in modal decomposition for the identification of coherent structures in turbulent shear flows, the 2AT group is developing physics-based data-compression strategies for the mining and analysis of large datasets generated by high-fidelity numerical simulations. The postdoc will focus on developing and benchmarking modal-decomposition techniques—using on-the-fly approaches—applied to data produced by airframe aeroacoustics simulations.

Building on recent advances in mean-field-based linearised models for turbulent jet dynamics and aeroacoustics, the 2AT group has developed resolvent-based tools for jet-noise modelling, estimation, and control. The successful candidate will use these tools—working in collaboration with Airbus and 2AT researchers—to develop a predictive model for sound radiation associated with the dominant resolvent modes, explicitly accounting for decoherence effects arising from two-point turbulence statistics. This framework will initially be validated on axisymmetric jets and subsequently applied to fully three-dimensional cases.

Building on recent advances in mean-field-based linearised modelling for aeroacoustics, and in particularfor modelling sound generation by flow–structure interaction, the 2AT group is developing resolvent-based shape-optimisation tools. The successful candidate will collaborate with Airbus and 2AT researchers to develop an adjoint-based shape-optimisation framework for landing-gear aeroacoustics. The numerical methodology will rely on a resolvent formulation of the Boundary Element Method. Optimised geometries will be tested in the anechoic windtunnels of Institut PPRIME using microphone-array based source localisation.