CONCHA: COmplex flow simulatioN Codes based on High-order and - PDF document

CONCHA CONCHA: COmplex flow simulatioN Codes based on High-order and Adaptive methods Research unit: INRIA Futurs Theme: Num Localization: Laboratoire de Math´ ematiques Appliqu´ ees (LMA), Universit´ e de Pau et des Pays de l’Adour (UPPA), UMR 1 5142 CNRS 2 – UPPA Keywords: CFD, complex fluids, turbulent flows, combustion, numerical analysis, high-order methods, discontinuous and stabilized finite element methods, adaptivity, multigrid, DWR-method, numerical sensitivity analysis. Contents 1 Introduction 4 2 Members 4 3 Overall Objectives 5 4 Scientific Foundation 7 4.1 Goals: accuracy and efficiency . . . . . . . . . . . . . . . . . . . . . . 7 4.2 Difficulties related to numerical simulations of reacting flows . . . . . . 7 4.2.1 Physical coupling . . . . . . . . . . . . . . . . . . . . . . . . . 7 4.2.2 Reaction mechanisms . . . . . . . . . . . . . . . . . . . . . . . 8 4.2.3 All-Mach regimes . . . . . . . . . . . . . . . . . . . . . . . . 8 4.2.4 Turbulence . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 4.3 Numerical tools: High-order discretization methods . . . . . . . . . . . 9 4.3.1 Motivation for discontinuous finite elements . . . . . . . . . . . 9 4.3.2 Overview on discontinuous Galerkin and other stabilized finite elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 4.3.3 Challenges related to DGFEM . . . . . . . . . . . . . . . . . . 12 4.3.4 Approximation of solutions with shocks . . . . . . . . . . . . . 13 4.3.5 All-Mach approach . . . . . . . . . . . . . . . . . . . . . . . . 14 4.4 Numerical tools: Adaptivity . . . . . . . . . . . . . . . . . . . . . . . 15 1 Unit´ e mixte de recherche 2 Centre national de recherche scientifique 1

CONCHA 4.4.1 Mesh and order adaptation . . . . . . . . . . . . . . . . . . . . 15 4.4.2 Automatic model selection . . . . . . . . . . . . . . . . . . . . 15 4.4.3 DWR-method (dual-weighted residual) . . . . . . . . . . . . . 16 4.4.4 Parameter identification and numerical sensitivities . . . . . . . 16 4.5 Further topics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 5 Application Domains 17 5.1 Targeted issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 5.1.1 Working principle of combustion chambers . . . . . . . . . . . 18 5.1.2 Safety issue: accidental boring of a combustion chamber . . . . 20 5.1.3 Engine efficiency: improved cooling of a combustion chamber wall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 5.1.4 Generating compact propulsive systems: mixing processes and combustion in micro-devices . . . . . . . . . . . . . . . . . . . 23 5.2 Test problems related to the targeted issues . . . . . . . . . . . . . . . . 24 5.2.1 Highly underexpanded supersonic jets . . . . . . . . . . . . . . 24 5.2.2 Subsonic jet in cross flow . . . . . . . . . . . . . . . . . . . . 24 5.2.3 DNS of mixing in micro-channels . . . . . . . . . . . . . . . . 26 6 Software 27 6.1 Aims and Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 6.2 Development strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 6.2.1 Validation and comparison with other software . . . . . . . . . 28 6.2.2 Comparison with experiments . . . . . . . . . . . . . . . . . . 28 6.2.3 Distributed hierarchical software development . . . . . . . . . 28 7 Expected Results 28 7.1 Development and analysis of algorithms . . . . . . . . . . . . . . . . . 28 7.2 Software development . . . . . . . . . . . . . . . . . . . . . . . . . . 29 7.3 Validation of algorithms and numerical approaches . . . . . . . . . . . 29 8 Positioning with respect to other research projects 29 8.1 Positioning within INRIA . . . . . . . . . . . . . . . . . . . . . . . . . 29 8.2 Positioning on the national and international level . . . . . . . . . . . . 30 8.2.1 Reactive flow simulations . . . . . . . . . . . . . . . . . . . . 30 8.2.2 External collaborators . . . . . . . . . . . . . . . . . . . . . . 31 9 Industrial partners 31 9.1 Turbomeca . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 9.2 Airbus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 10 Dissemination 32 2

CONCHA 10.1 Education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 10.2 Scientific community . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 10.3 Participation in conferences, workshops . . . . . . . . . . . . . . . . . 33 11 Agenda and internal organistaion 33 11.1 Short term (2-years) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 11.2 Long term (5-years) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 11.3 Organisation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 A Appendix: Short curriculae of members 35 A.1 Roland Becker . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 A.2 Pascal Bruel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 A.3 Daniela Capatina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 A.4 Robert Luce . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 A.5 Eric Schall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 A.6 David Trujillo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 3

CONCHA 1 Introduction 1 Introduction CONCHA is concerned with the numerical simulation of complex flow problems, with special emphasis on aeronautics. Our particular interest is in problems related to propul- sion engines, which lead to the study of physical phenomena such as high-speed flow, chemical reactions, combustion, and turbulence. The objective of this project is to develop simulation codes which are able to handle the difficulties implied by the physics of our applicational domain: strong nonlinearities, large spectrum of time and space scales, stiff couplings, model uncertainties, and the im- portant size of the discrete systems to be solved; see Section 4.2 for details. Therefore, our special interest lies in robustness with respect to the different physical parameters and good efficiency, defined to be the computational work for a given accuracy. Our tools are high-order and adaptive methods, which are being developed by the math- ematical community since the last two decades; see Section 4.3 for details. Of special interest is the confrontation with specific problems (see Section 5.2 for the definition of test problems) from the considered domain of applications. The medium-term goal of the project is to evaluate the potential of these modern techniques in view of their possible integration into industrial codes. The long-term objective of this project is to provide reliable software for optimization problems related to complex flow problems. Here, our special focus is on optimal de- sign and tools for the coordinating simulation and experimentation, including parameter estimation and the computation of numerical sensitivities, see Appendix 4.4.4. The project relies on an interdisciplinary collaboration between numerical analysts and specialists in fluid dynamics. A particular advantage of the composition of the members is the confrontation of computational results with experiments, made possible by the presence of experimental facilities common to UPPA and Turbomeca. 2 Members Team members from UPPA • BECKER Roland, Professeur, LMA (Team Leader) • BRUEL Pascal, CR1 CNRS, HDR, LMA • CAPATINA Daniela, Maˆ ıtre de Conf´ erences, LMA • LUCE Robert, Maˆ ıtre de Conf´ erences, LMA 4