Airiau Christophe

RTRA projects [1, 2]

The institution dedicated to Sciences and technology for Aeronautics and Space (RTRA – STAE) has funded the CARPE project (2013 – 2017) and 3C2T (2018 – 2020) supervised by IMFT (C. Airiau) and related to the studies of sensitivities and flow control. 

The first project concerned the robust control of wake instability of a thick flat plate in the laminar and turbulent regime. It was carried out in collaboration with the LAAS, CERFACS, ISAE and ONERA research centers and contained theoretical, numerical and experimental works. The vorticity given from URANS simulation are shown in the figure, in the active control configuration for different time instants (Carini et al, Phys. Fluids, 2017). An original stability and sensitivity analysis of the experimental wake obtained in a ONERA wind tunnel, has been performed by reconstruction of the velocity field and with a adjoint stability approach. 

 Figure :wake active control. Turbulent streamwise velocity field longitudinale (URANS) for (a) t=380, (b) t=440, (c) t=560, (d) t=680 (streamlines show at final state asymetry). The control law stabilizes the global unstable modes.

The second project is related to the control of the transitional and turbulent boundary layer in the low speed and hypersonc regime. The CERFACS and ONERA research centers were the project partners. The goal was to create a new numerical platform based on the new high order JAGUAR code developed by CERFACS able to solve accurately aerodynamics and flow control of complex configurations. The sensivity and flow control of reference cases as airfoil and flate plate boundary layers flows have been performed with the help of algorithmic differentiation. The Tapenade tools has been used to do that, in collaboration with INRIA and the Adjoint MPI library. These original study on unsteady flows were really challenging, from numerical and computer science view points.

Turbulent jet and uncertainty quantification [3, 4, 5]

The uncertainty quantification on a under-expanded supersonic turbulent jet at outlet Mach number of 1.15 has been developed in the framework of a Marie Curie AeroTranet 2 projet ( add years) in collaboration with CERFACS and Greenwich university, by RANS simulations. The influence of two input physical parameters (stagnation pression and laminar to turbulent viscosity ratio) onto the shock cells have been analysed and compared to experimental results. The high global sensitivity (variance) of the nozzle lips has been experienced. Approaches details can be found in (Granados-Ortiz et al., Phys. Fluids 2017).

This work has been completed (2020-2021) by coupling the RANS simulations to a Linear Stability approach (Parabolized Stability Equations) in order to get uncertainty of the two main parameters to the stability characteristics of the turbulent jet (“Addressing the Impact of Non-intrinsic Uncertainty Sources in the Stability Analysis of a High-Speed Jet”, submitted, june 2021)

Related publications

1. Global stability and control of the confined turbulent flow past a thick flat plate M. Carini,  C. Airiau,  A. Debien,  O. Leon,  and J. O. Pralits, Physics fluids,  Physics of Fluids 29, 024102 (2017); doi
2. Adjoint computations by algorithmic differentiation of a parallel solver for time-dependent PDEs, J.I.Cardesa, L.Hascoët, C.Airiau, Journal of Computational Science, vol. 45, 101155, doi, 2020
3. Identication of temporal and spatial signatures of broadband shock-associated noise mechanism,C. Pérez-Arroyo, G. Daviller, G. Puigt, C. Airiau, S. Moreau (2018) Shock Waves,  January 2019, Volume 29, Issue 1, pp 117–134. Link
4. On the Influence of Uncertainty in Computational Simulations of  High-Speed Jet Flow from an Aircraft Exhaust.  F. Granados-Ortiz,  C. Perez Arroyo,  G.  Puigt, C.-H. Lai,    C.  Airiau.   Computers and Fluids, déc. 2018.   doi . Vol . 180, pp. 139-158, February 2019
5. Addressing the Impact of Non-intrinsic Uncertainty Sources in the Stability Analysis of a High-Speed Jet, F. Granados-Ortiz, C. Airiau, J. Ortega-Casanova (submitted, june 2021)

In 2015 A. Bottaro got a chaire of excellence at IMFT, funding by the IDEX UNITI program of the University of Toulouse. The global project refered as “Bioskins” provided a PhD thesis with a co supervision  A. Bottaro and C. Airiau  on the modelling and simulation of flow around heterogenous walls, non rigid and non smooth.

The natural skins for instance  are  build from micro-fibers and can be rough, porous, elastic,  or equipped with complexe geometrical at different scales. These bio-inspired wall research assumes that the nature is optimized to reduce drag and to enhance global aerodynamic  performance of fishes and birds or other natural shapes. Natural canopy flow is another example of natural flexible structures . A better understanding of the physics and an accurate theroretical and numerical modelling of that physics  should lead to reduce the energetic consumption of future vehicles. The scientic issues are related to fluid-structure interaction, flow control, the flexible porous media modelling, hydodynamic stability, multi-scale numerical simulation, model reduction and surrogates.

Canopy flow modelling [1]

Figure : sketch of the canopy momami modes (a). streamwise velocity fiel in fibrous media given with VANS approach (b) and by direct numerical simulation  for Re = 1000 (c).

Fibrous interface modelling [2]

Porous media modelling in non-Newtonian fluid [3]

Pseudo-plastic fluids exhibit a non-linear stress-strain relationship which can provoke large, localized viscosity gradients. For the flow of such fluids in porous media the consequence is a strong variability of the effective permeability with porosity, angle of the macroscopic pressure gradient, and rheological parameters of the fluid. Such a variability is investigated on the basis of adjoint homogenization theory for a Carreau fluid in an idealized porous medium geometry, highlighting differences with respect to the Newtonian case. It is shown in particular that the more we depart from Newtonian conditions, the more the (often used) hypothesis of an effective viscosity in Darcy’s law is a poor approximation, for the effective permeability tensor becomes strongly anisotropic. The pictures below shows the cell numerically solved with FREEFEM for a Carreau fluid.

 

Figure :(a) Second invariant of the rate of strain tensor, γ̇(0) , and (b) viscosity µ (0) . The macroscopic pressure gradient within the unit cell is inclined at α = 30 degrees with respect to the direction of the x-axis.  

1. N. Luminari, C. Airiau, A. Bottaro Drag-model sensitivity of Kelvin-Helmholtz waves in canopy flows,  physics of fluids,Vol. 28, (12), 2016, doi   124103
2. Effects of porosity and inertia on the apparent permeability tensor in fibrous media N. Luminari, C. Airiau, A. Bottaro (2018), International Journal of Multiphase Flow  Volume 106, September 2018, Pages 60-74 doi
3. Flow of shear-thinning fluids through porous media, C. Airiau, A. Bottaro,  Advances in Water Resources, Volume 143, September 2020, 103658, doi

Propellers – wing interactions

Period : 2020 – 2022

Participants : C. Airiau (IMFT), J.C. Jouhaud (Cerfacs), S. Subramanya (ISAE-Sup Aero), …

Future aircraft of small and middle size will use electrical energy to generate thrust. Many recent studies are dedicated to the distributed propulsion meaning that thrust will be provided by a large numbers (upt to 17) of engines. The present activity is related to first developped tools able to pre design a propeller and a wing taking into account of the propellers wakes and wing interactions, and eventually propellers-propellers interactions. First, the old single-propeller / wing interaction is currently implemented in a new python tool for rapid design.

In parallel numerical simulations (RANS) on various configurations are tested based on a rotordisk approach interacting with a wing. The idea is to generate a parametrized problem and response surface to get a surrogates model, later implemented in the new tool.

Work is currently in progress and started in mid-2020.

Propulsion with ionic wind

Period : 2020 – 2022 

Participants: F.  Plouraboué, C. Airiau, F. Fabre

Fundings :  ANR

This work is supervised by F. Plouraboué and in collaboration with D. Fabre. The idea is to design a ionic wind just upstream a wing to generate a local flow able to generate thrust without destroying lift.
Many issues related to ionic wing generation have to be solved. My present contribution will be related to the modelling of the aerodynamics with distributed thrust along the wing surface and to perform some optimization with respect to the main ionic wing parameters.