CAGIRE brings together since May 2016 researchers coming from different horizons and backgrounds (turbulence modelling, applied mathematics, experiment) who elaborated since 2011 a common vision of what should be the simulation tool of fluid dynamics of tomorrow. If not entirely application-driven, this project is based on the will for developing tools that could be useful in a way or another to the companies (big but also small !) that are active in the competitive fields of aviation/automotive propulsion and energy production. The targeted flows are (mostly) wall bounded and turbulent. As a consequence, they feature a multiplicity of time and scale fluctuations that renders their simulation extremely challenging. The team's motto is agility or equivalently a clever use of adaptativity in the developed simulation suite based on I) The capability of being run on any present or future new supercomputer in a way that fully benefits from the hardware evolution while limiting the painful and time consuming phase of machine adaptation by the recourse to an efficient runtime, ii) A high level of accuracy and robustness permitting the use on a large range of flow configurations: ranging from the generic lab scale geometry to that of a practical interest, iii) The capability of adapting on the fly the physical modelling by recourse to dynamic hybridization of the most relevant models of a given class with a focus on the turbulence modeling.

The post-doc position is part of the project ASTURIES, supported by the E2S (Energy and Environment Solutions, click here to see a presentation) consortium of the university of Pau. The project is a collaboration between the Applied Mathematics Department of the university of Pau/CNRS/INRIA, where the post-doc will be working, CEA an IFPEN.

In the context of turbulent flows relevant to the many industrial sectors, the project aims at developing an innovative CFD methodology. The next generation of industrial CFD tools will be based on the only approach compatible with admissible CPU costs in a foreseeable future, hybrid RANS/LES. However, state-of-the-art hybrid RANS/LES methods (such as DES, SAS, PANS, among many others) suffer from a severe limitation: their results are strongly user-dependant, since the local level of description of the turbulent flow is determined by the mesh designed by the user. In order to lift this technological barrier, an adaptive, or agile methodology will be developed: the scale of description of turbulence will be locally and automatically adapted during the computation based on local physical criteria independent of the grid step, and the mesh will be automatically refined in accordance.

The objective of this post-doc is to carry out tasks 3.2 and 4 of the ASTURIES project:

Task 3.2: Several challenging configurations will be used in the present project in order to evaluate the performance of the developed agile strategy. In order to evaluate the new strategy, reference data are necessary, obtained using state-of-the-art models and numerical methods. The purpose is two-fold: comparing the accuracy of the solutions; making sure that the CPU cost of the new strategy remains competitive. The aim of the new method is to introduce agility, i.e., a simultaneous automatic model and grid refinement, but in order to be applicable to industrial configurations in the future, the CPU cost of the method must remain comparable to the CPU cost of methods available today. A typical state-of-the-art hybrid RANS-LES model and a typical state-of-the-art numerical method (second-order accurate finite-volume method) will be used. Therefore, the configurations selected for validation will be computed using the open-source solver Code Saturne Archambeau et al. 2004, using the standard HTLES methodology (based on a user-designed mesh, Manceau 2019; Duffal et al. 2021). The precise list of test cases remains evolving and will be discussed by the partners during the course of the project. However, the following configurations are already envisaged: Periodic hill; Backward facing-step, against experimental data (Driver and Seegmiller 1985) or DNS (Lamballais 2014); Jet in cross-flow configuration, which is investigated experimentally in Task 3.

Task 4: The new adaptive hybrid RANS/LES methodology will be developed. In order to validate the different approaches developped during the project, the challenging configurations described above will be computed with this new method. The comparison of the results provided by the agile simulation on some test cases, with the corresponding CPU cost, will serve as a demonstrator of the success of the project. This will pave the way towards the dissemination of this method to our usual industrial partners in the field of aeronautic and automotive propulsion and energy production, and beyond, to all the scientific and industrial CFD communities

- A good knowledge of programming is necessary, of fortran would be appreciated

- Languages : english

• Subsidized meals
• Partial reimbursement of public transport costs
• Possibility of teleworking and flexible organization of working hours
• Professional equipment available (videoconferencing, loan of computer equipment, etc.)
• Social, cultural and sports events and activities
• Access to vocational training
• Social security coverage

2653€ / month (before taxs)