Appel à candidatures | Recherche, Emploi

Postdoc onTurbulence wall modelling for Lattice-Boltzmann Methods, Marseille (France)

Du 1 septembre 2021 au 31 août 2023

1-2 years position

 

  1. Lab M2P2, Marseille France

    Contact : 
    Prof. Pierre Sagaut 
    Email : pierre.sagaut@univ-amu.fr
    URL: www.lmm.jussieu.fr/~sagaut

 

The aim of the present project is to develop an efficient solid wall treatment for LBM, including turbulent wall models. Several issues will be addressed, among which: • interpolation methods used within the Immersed-Boundary Method (IBM) or Cartesian meshes framework for solid boundary treatment [1-2] • use of a turbulent wall model in conjunction with immersed solid boundaries [3] • reconstruction of the density distribution functions of LBM at boundaries nodes within this framework (accuracy, mass conservation, ...) • reconstruction of macroscopic quantities on the solid surface for post-processing purposes (eg. calculation of aerodynamic forces applied on an airfoil) • extension to moving/deforming solid boundaries

page1image2638864 page1image2638656

Postdoctoral position
Turbulence wall modelling for Lattice-Boltzmann Methods

Motivations

Unsteady simulation of turbulent flows has become a very valuable tool for both fundamental and applied research, but also for design and optimization in real systems engineering. Therefore, the development of improved physical and numerical models that are able to handle the full complexity of real systems is among the main research topic in Computational Fluid Dynamics (CFD). It should be emphasized that the system complexity originates in the coupling between very different scales (very small and very large ones) and the coupling between physical mechanisms of different nature. The exponential growth of computing power has enabled, in the recent past, to predict the dynamics of complex turbulent flows (in which complexity stems from geometry and multiphysic couplings), but mostly for simplified systems. A main objective is still to develop new methods that will allow to handle the total complexity of full scale real systems, and to diffuse these methods thanks to simulation tools that are available for both academic and industrial CFD communities.

This research project aims is part of the development of the ProLB software for aerodynamics, aeroacoustics and heat transfer simulations (http://www.prolb-cfd.com). This software is co-owned by Airbus, Renault, CS, Aix-Marseille University, Ecole Centrale de Lyon and CNRS. It is based on Lattice Boltzmann Methods (LBM). LBM appeared, under its modern formulation, in the early 1990s. This approach has now reached a maturity level which is sufficient to consider it as a potential alternative to Navier-Stokes- based CFD tools. From the theoretical standpoint, LBM is a method to solve the Boltzmann equation, and therefore to describe flow dynamics at a finer level than continuum mechanics, considering velocity density distribution functions instead of usual macroscopic quantities (velocity, pressure, ...) From a practical standpoint, LBM-based simulation tools have proved to be more efficient than classical Navier-Stokes based solvers for subsonic aerodynamics and aeroacoustics.

Despite their success and their huge potential, LBM methods have received a little attention compared to classical CFD approaches based on Navier-Stokes equations. Therefore, there are many fundamental and practical open questions in the LBM theory, along with a very important industrial demand. The present project is built in this context, with particular emphasis put on the coupling between wall modeling via Immersed boundary Methods and turbulence modelling. It aims at improving turbulence wall models along with solid wall treatment within the LBM framework for the unsteady simulation of turbulent flows (LES, DES, hybrid RANS/LES methods) in complex geometries.

Research program

The aim of the present project is to develop an efficient solid wall treatment for LBM, including turbulent wall models. Several issues will be addressed, among which:

  • interpolation methods used within the Immersed-Boundary Method (IBM) or Cartesian meshes framework for solid boundary treatment [1-2]

  • use of a turbulent wall model in conjunction with immersed solid boundaries [3]

  • reconstruction of the density distribution functions of LBM at boundaries nodes

    within this framework (accuracy, mass conservation, ...)

  • reconstruction of macroscopic quantities on the solid surface for post-processing purposes (eg. calculation of aerodynamic forces applied on an airfoil)

  • extension to moving/deforming solid boundaries

    [1] M. Berger and M. Aftosmis, “Progress towards a Cartesian cut-cell method for viscous compressible flow,” in 50th AIAA Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace Exposition (AIAA, 2012), p. 1301
    [2] F. Capizzano, “Turbulent wall model for immersed boundary methods,”

    AIAA J. 49, 2367–2381 (2011).
    [3] Wilhelm, S., Jacob, J., & Sagaut, P. (2018). An explicit power-law-based wall model for lattice Boltzmann method–Reynolds-averaged numerical simulations of the flow around airfoils. Physics of Fluids, 30(6), 065111.

    The work will be performed in close collaboration with industrial partners (Airbus, Renault, Safran ) and research centres (ONERA, CERFACS).

    Candidate

    The candidate must have a significant background concerning at least one of the following items:

  • Immersed Boundary Methods for CFD

  • Lattice-Boltzmann Methods for CFD

    Informations

    Location: M2P2 Laboratory UMR 7340, Campus of Ecole Centrale Marseille at Château-Gombert, Marseille, France

    Duration: 12 - 24 months

    Net monthly salary: 1930 euros( <3 years after PhD) or 2300 euros (>3 years after PhD)

    Contact : Prof. Pierre Sagaut 
    Email : pierre.sagaut@univ-amu.fr
    URL: www.m2p2.fr