Vendredi 7 Décembre 2013
11 h 00 salle de réunion du CEAT
Senior Lecturer
School of Engineering and Information Technology, The University of New South Wales at ADFA, Canberra ACT 2600 Australia.
Invité par Bernd Noack
In this seminar, the method of control volume analysis - the standard engineering tool for the analysis of flow systems - is connected to the maximum entropy (MaxEnt) method of Jaynes. Firstly, the generic concept of entropy H is established and, by application to an equilibrium system, is used to define the thermodynamic entropy S. The principles of control volume analysis are then enunciated, giving integral (Reynolds transport theorem) and differential forms of the conservation equations. By consideration of the entropy balance, the concept of entropy production is then introduced. Equations for the entropy production are derived for simple, integral and infinitesimal flow systems. Some technical aspects of entropy balance calculations are examined, and several definitions of "steady state" are discussed. Application of a Reynolds decomposition then reveals an "entropy production closure problem" in turbulent systems, analogous to that for the local fluid velocity. Finally, direct MaxEnt analyses are used to predict the steady state of infinitesimal and integral flow systems. These analyses are cast in terms of a "minimum potential" principle, analogous to the Planck potential concept; these reduce, in different circumstances, to maximum or minimum entropy production (MaxEP or MinEP) principles.
Lundi 3 Décembre 2012
16 h 00 Campus Universitaire - ENSIP 3ème étage du Bâtiment Mécanique
PHD student
LEMTA, UMR 7563 CNRS-Nancy Université
Invité par Serguei Martemianov
Les piles à membrane échangeuse de protons (PEMFC) sont présentées comme un système de conversion d'énergie chimique en énergie électrique prometteur. Leur commercialisation est tempérée par des problèmes de coûts et de durabilité liée à la gestion de l'eau et de la chaleur au sein de la pile.
L'objectif de ce travail est d'examiner expérimentalement les transferts d'eau et de chaleur dans une PEMFC. Pour étudier ces transferts, une pile innovante a été fabriquée au laboratoire permettant de mesurer, de façon simultanée, les températures internes, les flux de chaleur et les flux d'eau dans une cellule de pile à combustible.
La mesure de température s'effectue à l'aide de 4 fils de platine insérés entre l'électrode et la couche microporeuse, à l'anode et à la cathode. La température est déduite de la mesure de la résistance électrique du fil de platine via une courbe d'étalonnage établie pour chaque fil.
La mesure des flux de chaleur se fait à l'aide de fluxmètres placés sur les plaques d'alimentation, régulée par un système de thermorégulation. Les flux d'eau sont, quant à eux, mesurés à l'aide d'un bilan matière entrée-sortie à l'aide notamment de balances de mesure en sortie de pile.
Les mesures simultanées des flux de chaleur, et des températures sont basées sur une analyse du transfert de chaleur dans les couches poreuses de la pile qui requière notamment une étude sur la localisation des points de condensation de l'eau produite par la pile. A travers l'analyse des flux de chaleur en fonction de la température des électrodes, il a été démontré que la condensation de l'eau produite se produit à l'interface entre la couche poreuse et les plaques d'alimentation.
De plus, les variations de température aux électrodes suggèrent que les flux d'eau sont dirigés vers la température la plus basse. Cela permet d'obtenir un éclairage intéressant sur le couplage entre le transfert de chaleur et le transport de l'eau dans une PEMFC. En effet cela suggère que l'eau produite par la réaction électrochimique traverse les couches poreuses sous forme vapeur à l'aide du gradient de pression de vapeur saturante obtenu par la différence de température à travers la couche poreuse.
Ces mesures permettent aussi d'estimer in-situ de la conductivité thermique effective des couches poreuses, paramètre clé pour l'analyse des transferts thermiques au sein de la pile. Une valeur de 0.25 W.m-1.K-1 a été obtenue en accord avec les mesures données de la littérature.
Jeudi 29 Novembre 2012
14 h 00 salle de réunion du bâtiment H2 (SP2MI)
Lecturer
University of Sydney
Invité par Sébastien JARNY
The objective of this work was to increase our understanding of gravity-driven geophysical flows by developping a new platform to simulate avalanches of fluid in the laboratory.
To simulate flow avalanches in the laboratory, we created a unique experimental setup consisting of a metallic frame supporting a reservoir, an inclined aluminum plane, and a horizontal run-out zone. At 6-m long, 1.8-m wide, and 3.5-m high, the structure is probably the largest laboratory setup of its kind in the world. In a dam- break experiment, up to 120 liters of fluid can be released from the reservoir down the 4-m long inclined plane. We precisely control initial and boundary conditions.
To measure the free-surface profile, a novel imaging system consisting of a high-speed digital camera coupled to a synchronized micro-mirror projector was developed. The camera records how regular patterns projected onto the surface are deformed when the free surface moves. We developed algorithms to post- process the image data, determine the spreading rate, and generate whole-field 3-dimensional shape measurements of the free-surface profile. We compute the phase of the projected pattern, unwrap the phase, and then apply a calibration matrix to extract the flow thickness from the unwrapped phase.
56 different flow configurations, with a wide range of inclinations, were finally tested with Newtonian and viscoplastic fluids. For each test, the evolution of the free surface was recorded in 3 dimensions. Different flow regimes were observed, which depend on: the plane inclination, the setup geometry, the volume, and characteristics of the fluid. Partial agreements were found between theoretical models and our results.
Short bio: Steve has joined, three years ago, the Fluid and Environmental Group at the School of Civil Engineering after holding a Postdoc position in the Complex Fluids Laboratory at the University of British Columbia. His research has focused on the design, the development and the validation of a PIV system to provide an accurate measurement of the velocity field inside transient free-surface flow of viscoplastic materials. He is now leading the Wind Engineering group at the University of Sydney where he develops new tomographic-PV techniques. His Ph.D. thesis entitled `Measurements of Time-Dependent free-surface Viscoplastic flows Down Steep Slopes' was completed under the supervision of Pr. C. Ancey at the Environmental Hydraulic Laboratory at EPFL. A novel fringe projection system was developed to provid accurate instantaneous whole-field 3D shape measurements of the free-surface profile and the spreading rate of non-Newtonian fluids at high repetition rate. Before joining the Environmental Hydraulic Laboratory Steve had been working for more than four years in industry as an aero-thermodynamics engineer. There he had been associated with multidisciplinary projects with state-of-the art technology and competences. At first he joined for one year the Research and Technology Department of SNCF (Société Nationale des Chemins de Fer, French National Railway Company) in Paris, France as an internship for my master thesis and then as a consultant on the implementation of solutions to problems related to the complex and hazardous side-wind effect of the new TGV line Paris-Marseille. Some of the key duties were the implementation of a network of weather stations, the development of new algorithms to post-treat wind data and to detect and predict in real time the apparition of dangerous wind conditions that are used to adapt the kinematics of trains to avoid accidents. Later, he joined the Tunnel Ventilation Group at Electrowatt Infra AG in Zürich as a Projet Engineer, where he was a project leader for the development and improvement of ventilation systems on several existing tunnels or tunnel projects in particular in case of accidents leading to hazardous situations for people caught inside the tunnels. He was in charge of the development of numerical tools to simulate smoke, toxic gases and temperature propagation in case of fire in any underground facility with different ventilation configurations and he additionally supervised all the wind-tunnel tests realized for the new Bangkok metro line. Since graduating, Steve has been involved on both numerical and experimental projects. He particularly enjoys developing new measurement techniques.
Vendredi 23 Novembre 2012
14 h 00 salle de réunion du CEAT
Head of flow control research group
TU Braunschweig
Invité par Bernd Noack
The Collaborative Research Center (CRC) 880 "Fundamentals of high-lift for future commercial aircraft" aims at providing the scientific basis for a new kind of civil, low-noise aircraft. The new airplanes are to be driven by new, efficient high-lift systems that can not only reduce noise, but also allow taking off and landing on short runways, which would better integrate the aircraft as a transport vehicle in peri-urban areas. The new CRC intends to increase the efficiency of the active high-lift system, particularly by reducing the power required for blowing out air over the high-lift flaps. Through the innovation of a shape-variable aerofoil, the exploration of the synergies between suction and blowing of a pneumatically-autonomous, active high-lift system and the exploitation of the potentials of flow control, lift coefficients up to twice the values common today are believed to be possible.
The efficiency of active blowing over high-lift flaps is being investigated by the flow control group at TU Braunschweig. The aerodynamical benefits of the proposed improvements, which include formadaptive leading edges, spanwise-segmented blowing, combined suction and blowing using a dynamically actuated lip are being examined. The potentials of closed-loop flow control are explored through the use of low-order models. The objectives will be approached using numerical simulations and experimental measurements in the large watertunnel of TU Braunschweig.
Short bio: Richard Semaan is the manager of the Collaborative Research Centers SFB 880, and the head of flow control research group at TU Braunschweig. He earned his Masters in 2005 in Engineering Mathematics and Turbulence from Chalmers University of Technology, Sweden. He obtained his PhD in 2010 from University of Wyoming researching swirling jet flows. After two years as a post-doctoral researcher at TU Braunschweig investigating transition flows and pressure correction methods, he founded the flow control research group.
Jeudi 22 Novembre 2012
14 h 00 salle de réunion du bâtiment H2
CR CNRS
Institut PPRIME
Invité par Nicolas Benard
Les plasmas froids possèdent un faible niveau d'ionisation et une basse montée en température du gaz comparativement aux plasmas chauds. Ils sont fortement hors équilibre, et l'énergie produite peut contribuée à certains processus chimiques et électrodynamiques. Grace à ces propriétés, les plasmas froids sont actuellement utilisés pour de nombreuses applications telles que la combustion, le contrôle d'écoulement aérodynamique, la synthèse de matériaux, le traitement de surface ou pour des applications liées à de nouvelles thérapies médicinales.
Une grande partie de ces applications exigent la maitrise des plasmas à pression atmosphérique. Parmi ces plasmas, les décharges impulsionnelles nanosecondes répétitives (NRP) sont actuellement un centre d'intérêt des différentes communautés. Ils s'agit de décharges produites par des impulsions de haute tension dont les durées sont de l'ordre de quelques dizaines de nanosecondes et qui sont appliquées à des hautes cadences de répétition (typiquement supérieures a 1 kHz). Les décharges NRP sont à l'étude du fait de leur efficacité énergétique, et des nombreux effets intrinsèques à ces décharges tels que le chauffage ultra-rapide ou encore la génération de rayons x et d'électrons à haute énergie.
Aux pressions supérieures à 1 atm, la physique des plasmas est peu connue mais pourrait être exploitée pour les applications en combustion, synthèse de matériaux et de produits chimiques. En particulier, les propriétés de transport des fluides en phase supercritique sont une combinaison de celles de gaz et de liquides, auquel s'ajoutent des phénomènes physiques uniques tels que des fluctuations de densité. Il est donc attendu que les plasmas générés dans des fluides supercritiques conduisent à des effets uniques, comme par exemple une forte diminution du champ de claquage vers le point critique.
L'objectif de ce séminaire est de présenter les travaux menés au sein de l'EM2C (Ecole Centrale Paris) et de l'Université de Tokyo, travaux relatifs à l'utilisation de décharges plasmas NRP et à très haute pression pour la combustion ou la microfluidique. Enfin, quelques pistes de réflexion sur les travaux pouvant être conduit au sein de l'Institut Pprime seront abordées.
Mercredi 24 Octobre 2012 (14 h 30 - 17 h 00)
Jeudi 25 Octobre 2012 (9 h 00 - 12 h 00)
Salle des actes de l'ENSIP
Organisé par Hélène BAILLIET
Le programme de ces journées peut être téléchargé ici.
Lundi 22 Octobre 2012
14 h 00 salle de réunion "Branche Fluides"
Professor
Department of Aerospace Engineering Indian Institute of Science
Invité par Jean Paul BONNET
Turbulent free shear flows grow by entraining fluid from surrounding nonturbulent regions. Visualizations had suggested that the process is large scale and is accomplished by engulfment of large packets of surrounding fluid which is then broken down. Over the past decade it has become clear that engulfment has a minor role: most of the entrainment is accomplished by small scale processes that occur at the boundaries of the turbulent flow. This was first revealed in our DNS of an incompressible round jet. Later DNS at TU-Munich, of compressible mixing layers, confirmed the importance of this interface process while also showing a retarding effect of compressibility. The basic features have now been confirmed by experiments using PIV and PTV.
Short bio: Dr. Joseph Mathew is a Professor in the Dept. of Aerospace Engg. at the Indian Institute of Science, Bangalore. He obtained his B.Tech. in Mechanical Engg. in 1984 from IIT Madras, and Ph.D. in 1990 from the Massachusetts Institute of Technology in the area of Fluid mechanics and Wave propagation. After a post-doctoral appointments at ICOMP at NASA Lewis (Glenn) Research Center, and at the National Aerospace Laboratories, Bangalore, he joined IISc. in 1995. His research interests are in turbulent flows, its mechanisms and computation, especially DNS and LES.
Lundi 15 Octobre 2012
11 h 00 Salle de Com Batiment H2
Associate Professor Director
Applied Physics Research Group Department of Mechanical and Aerospace Engineering - University of Florida
Invité par Nicolas Benard
Plasma actuators have become the key enabler for boundary layer flow control especially when the device needs to be surface compliant. While the effects of plasma actuators are striking for low speeds, their efficacy becomes very limited for a wide range of flow speeds due to the lack of understanding of the multiphysics controlling adjustable authority of such devices. The research by the Applied Physics Research Group (APRG) at the University of Florida aims towards an integrated effort of theoretical prediction and experimental validation for developing a class of novel high performance actuators. This research advances the state-of-the-art of plasma actuators by developing sophisticated physics based modeling tools and fostering innovation of a class of plasma based devices. Gaining insight of plasma actuators will help robust, scalable actuation of transitional and turbulent flows.
Short bio: Subrata Roy is an Associate Professor of the Mechanical and Aerospace Engineering department of the University of Florida, Gainesville. He serves as the Director of the Applied Physics Research Group at the University of Florida that hosts postdoctoral researchers, graduate and undergraduate students studying the plasma physics and its engineering applications. Dr. Roy is a Fellow of the American Society of Mechanical Engineers and an Associate Fellow of the American Institute for Aeronautics and Astronautics.
Mercredi 10 Octobre 2012
14 h 00 salle de réunion du CEAT
PHD student
TU Berlin - Department of Mathematics
Invité par Bernd Noack
I. Direct Discretization of the Input to Output Behavior and II. Generalized Riccati Equations
For the simulation of flow control problems one has to call on reduced models and on efficient numerical algorithms to keep the computations feasible. The two approaches I will present address both issues but indedependently of each other. The direct discretization of the input to output behavior delivers a reduced model of the control problem by restricting the model to finite dimensional input and output spaces. On the other hand, the approach via generalized Riccati equations does no reduction but a reformulation of the model that makes the numerical computation of the optimal solution feasible. Both approaches were developed for linearizations of Navier-Stokes equations. However, the ultimate goal is to use them in iterative schemes for the solution of the corresponding nonlinear problems.
The slides can be downloaded here.
Jeudi 6 Septembre 2012
14 h 00 salle de réunion du bâtiment B17 (Campus)
Professeur
Laboratoire d'Acoustique de l'Université du Maine
Invité par Yves Gervais
La méthode RIFF (Résolution Inverse Filtrée Fenêtrée) est une technique d'identification des sources vibratoires qui peuvent être de différentes natures. Elle a, en particulier, fait ses preuves dans l'identification d'excitations mécaniques par forces et/ou moments, excitations acoustiques, etc. Le principe de cette méthode est basé sur la vérification des équilibres d'efforts locaux que décrit l'équation du mouvement de la structure à laquelle s'ajoute un fenêtrage spatial et un filtrage en nombre d'onde pour éliminer les amplifications des incertitudes de mesure sur le résultat. L'intérêt principal de cette méthode réside dans son aspect local, en ce sens que les conditions aux limites peuvent être inconnues et que le champ de déplacement doit être mesuré uniquement dans les zones où l'on souhaite identifier l'excitation. Contrairement aux problèmes inverses que l'on trouve dans la littérature, on a ici une approche où certaines données du problème direct peuvent être ignorées, ce qui réduit le besoin de modéliser complètement la structure. La présentation sera d'abord constituée d'une description des différentes étapes qui constituent la méthode RIFF, accompagnée d'exemples de validations. Certaines applications seront ensuite montrées, telles que la caractérisation de sources solidiennes, la caractérisation de matériaux, la détection de défaut.
Dans une deuxième partie, on montrera comment la méthode RIFF peut être un excellent moyen d'extraire la composante acoustique d'un écoulement turbulent (couche limite ou écoulement décollé) excitant une structure, grâce aux aspects de filtrages passe-bas dans le domaine des nombres d'onde induits par la structure et par la méthode RIFF. L'exposé se basera sur les observations des résultats obtenus lors d'une campagne expérimentale au LMFA (Laboratoire de Mécanique des Fluides de l'Ecole Centrale de Lyon) réalisée en 2007 et des simulations numériques obtenues plus récemment dans le cadre de la thèse de Damien Lecoq au LAUM.
Lundi 3 Septembre 2012
14 h 00 salle de réunion du CEAT
Postdoctoral Researcher
Department of Scientific Computing - Florida State University
Invité par Bernd Noack
In this study the proper orthogonal decomposition (POD) is coupled with the discrete empirical interpolation method (DEIM) to construct an efficient reduced-order inverse problem of the four - dimensional variational (4D-Var) data assimilation applied to a finite element shallow water equations model. To the best of our knowledge this is the first time when POD accelerated by DEIM is used for numerical solving of an inverse problem. Previous studies on nonlinear forward parameterized PDEs models showed that the POD/DEIM technique reduces the computational complexity of the reduced order model due to its dependence on the nonlinear full dimensional model to regain the full model reduction capability expected from the POD reduced order model. In fluid dynamics, DEIM was applied to a shallow water equations POD reduced model on a β - plane (see Stefanescu and Navon 2012) using an alternating direction fully implicit finite difference scheme (ADI SWE) and the Euler explicit finite difference scheme. It was shown that the method decreases the computational time of POD reduced model directly proportional to the number of mesh points with errors close to those of the POD reduced order model errors.
Different approaches of POD/DEIM implementation of the reduced 4D-Var data assimilation problem are compared, including a dual-weighed method for snapshots selection coupled with a trust-region POD/DEIM approach. Based on a posteriori error analysis we incorporate an adaptive technique to increase the inverse problem solution accuracy updating the POD/DEIM bases as the cost function minimization progresses.
By employing the discrete empirical interpolation method we will reduce the computational complexity of the Trust-Region POD-4D-Var finite element shallow water equations model by decreasing the CPU time required to calculate the solutions of the forward and adjoint models and indirectly by reducing the condition number of the Hessian of the cost functional, thus accelerating convergence of the minimization process.
Mardi 10 Juillet 2012
14 h 00 salle de réunion du CEAT
PHD student
University of Strasbourg
Invité par Bernd Noack
In multiparticle flows, the regimes of transition to turbulence of sedimenting or rising particles bring about significant changes in the rates of dispersion. To understand the underlying physics, it is necessary to study the dynamics of an individual particle. Recent results pertaining to spherical bodies serve as a reference in simulations of the sedimentation of spherical particles and are used as benchmarks for testing numerical codes. Flat bodies, like disks, behave significantly differently, however, but their exhaustive numerical investigation has been impossible because of high computing costs.
Our numerical method generalizes the spectral -- spectral element discretization combined with the implicit treatment of particle motion equation used in Ref. [1] for the simulation of the free movement of a spherical particle under the action of gravity, buoyancy and hydrodynamic force to the case of non-spherical axisymmetric bodies. In this case, a body-fitted mesh must follow the rotation of the body but, at the same time, the far wake remains essentially vertical and is best captured by a vertical cylindrical domain. To satisfy both constraints, the computational domain is decomposed into a relatively small spherical sub-domain rotating with the body connected to the remaining cylindrical sub-domain by a spherical function expansion at the common interface. The use of spherical functions minimizes the computing costs of the three-dimensional rotations at the interface. As in the previous version, the fully implicit fluid-solid coupling makes the computing costs independent of the body inertia. A direct pressure solver makes the CPU time spent for the pressure computation negligible compared to that needed for solving the velocity equations. The accuracy and reliability of the method is tested by reproducing the known transition scenario for a spherical body with a spherical sub-domain following the body rotation.
Our accurate and computationally efficient method allowed us to carry out a very complete two-parameter investigation of the bifurcating states characterizing the transition scenario of infinitely thin disks.
[1] Jenny, M. & Dusek, J. 2004 Efficient numerical method for the direct numerical simulation of the flow past a single light moving spherical body in transitional regimes. Journal of Computational Physics 194, 215-232.
Short bio: The author is a final year PHD student working under the supervision of Jan Dusek at the University of Strasbourg. His current research is devoted to the study of sedimentation of non-spherical particles in transitional regimes.
Mardi 26 Juin 2012
15 h 30 salle de réunion du CEAT
Associate Professor
Syracuse University, USA
Invité par Bernd Noack
The isolation of individual coherent noise sources in subsonic jets is achieved by wavelet transform of the far-field acoustic field. The precise determination of the differences in detection times by several microphones points is related, by triangulation, to the individual source location. Non-linear filtering of the far-field separates the coherent noise from its background. This information will be cross-correlated with the history of suitable indicators calculated from near-field velocity fields. We will be using 2-D TR-PIV data at 10kHz (Spectral Energies Inc.) from the Mach 0.6 jet at Syracuse University, and the 3-D LES data for a Mach 0.9 jet.
Short bio: Ingénieur Physicien, Université de Liège, 1975 ; PhD, Aerospace Engineering, Cornell University, 1981 ; Post-doc, Chemical Engineering, Ohio State University, 1980-82 ; Faculty, Mechanical Engineering, Syracuse University, since 1982 ; Chercheur Associé, CEAT-Université de Poitiers, 1996 ; Awards: 2004 L.F. Moody Award of ASME/FED; teaching awards ; Research interests: turbulence theory, continuous wavelets, signal processing.
Mardi 26 Juin 2012
14 h 00 salle de réunion du CEAT
Researcher
Syracuse University, USA
Invité par Bernd Noack
This work seeks to explore novel methods for reducing aeroacoustically generated noise through the introduction of localized perturbations around the periphery of a Mach 0.6 jet. Simultaneous time-resolved measurements of both the near-field and far-field pressure are acquired alongside simultaneous time-resolved measurements of the instantaneous radial and axial components of the velocity field along the jet centerline. We assess the effectiveness of both open loop and proportional closed loop flow control as they compare to an uncontrolled jet, through the examination of resultant modifications to the low dimensional characteristics and spectral content of the jet. We also investigate the modifications to the cross-correlations between different regions both in the near- and far-field. The chosen control apparatus is a piezoelectrically driven synthetic jet actuation system. Results suggest that the shear layer is very receptive to relatively subtle excitation introduced at the jet lip, which also translates to the far-field region.
Short bio: B.S., Mechanical Engineering, Rensselaer Polytechnic Institute, 2007 ; M.S., Mechanical and Aerospace Engineering, Syracuse University, 2009 ; Ph.D., Mechanical and Aerospace Engineering, Syracuse University, 2012 ; Invited Researcher, Institut P' CNRS, Summer 2012.
Lundi 25 Juin 2012
14 h 00 salle de réunion Branche Fluides
Professor
Chalmers University, Sweden
Invité par Bernd Noack
The talk will present research by the author on Large Eddy Simulation (LES) and related hybrid approaches (such as Detached-Eddy Simulation (DES) and Partially-Averaged Navier-Stokes (PANS)) in ground vehicle aerodynamics. The applications range from flows around cars and trains in strong crosswinds to fundamental research for understanding of flow mechanisms in flow control. The crosswind stability of ground vehicles being very important in vehicle's development requires often novel numerical approaches due to difficulties to perform realistic experimental tests. Several examples of crosswind-like effects on ground vehicles ranging from meeting trains to vehicles exiting tunnels in strong crosswinds will be presented.
Recent results of the LES for active and passive flow control applied on generic vehicles such as Ahmed body will be presented. Application of LES for several different flow control strategies will be demonstrated. Examples of LES of flow control include passive flow control using vortex generators and other add on devices. The LES of active control strategies that change the wake of vehicles such as Moving Surface Boundary-layer Control and periodic blowing and suction will be presented. The final example is a simulations of the constant blowing or suction with the aim to control development of longitudinal vortex that develops on the A-pillar of a generic vehicle.
Short bio: Professor Krajnovic was a pioneer in using Large Eddy Simulation (LES) technique for studies of vehicle aerodynamics flows. He has written widely on application of unsteady simulations such as LES and hybrid approaches (such as Detached Eddy Simulation (DES) and Partially-Averaged Navier Stokes (PANS)) in aerodynamics of road vehicles and trains. His current interests include: prediction of aerodynamics flows, bluff-body aerodynamics, development of hybrid numerical approaches (such as PANS), vehicle aerodynamic shape optimization and flow control for vehicles. He is Professor of Computational Fluid Dynamics at Chalmers University of Technology in Gothenburg, Sweden. His research group "Vehicle Aerodynamic Laboratory" (www.tfd.chalmers.se/~sinisa) at Chalmers is devoted to fundamental and applied numerical research in vehicle aerodynamics.
Jeudi 21 Juin 2012
10 h 00 salle de réunion du B17
Associate Professor
Syracuse University, USA
Invité par Vincent Valeau
Mardi 19 Juin 2012
14 h 00 salle de réunion CEAT
Tafila Technical University
Jordan
Invité par Bernd Noack
Vendredi 8 Juin 2012
10 h 30 salle de réunion CEAT
Laboratorio de Fluidodinamica
Universitad de Buenos Aires
Invité par Bernd Noack
DBD plasma actuators have some very interesting features (direction and timing) which allow the introduction of disturbances that can be harder to produce with other tools. In few words it is a cheap way to produce a streamwise and wall bounded perturbation with a very good control its timing.
In a first part the talk will focus on the introduction of features usually found in turbulent boundary layer in laminar boundary layer. It will be shown how streamwise vorticity and streamwise velocity streaks produced by solid obstacles can be placed in a situation where a self sustaining process occurs. Then DBD plasma actuators are used to emulate these obstacles using the force produced.
In a second part we will focus on strong disturbances upstream a separation on a backward facing step and a free shear layer. Playing on the actuation frequency and duty cycle of the actuation we will show that vortex pairing can be forced, leading to an enhanced mixing downstream. Additionaly it will be shown how DBD plasma actuator can be easilly used in order to gather knowledge on these flow stability.
Short bio:Thomas Duriez is an ESPCI engineer and a Paris Diderot PhD in fluids dynamics obtained under the supervision of J.E. Wesfreid and J.-L. Aider at the PMMH Laboratory in Paris. Focused on experimental fluids dynamics he has been working on bluff body flows, bounded and unbounded shear layer flows and their control. Lately he has been a post-doctoral fellow at the Laboratorio de FluidoDinamica de la Facultad de Ingeneria in Buenos Aires, Argentina.
Lundi 4 Juin 2012
14 h 00 salle de réunion Branche Fluides
Dept of Scientific Computing
Florida State University
Invité par Bernd Noack
Mercredi 4 Avril 2012
14 h 00 salle de réunion Branche Fluides
Aerospace Computing Laboratory
Stanford University
Invité par Eric Lamballais
During the last fifteen years computational fluid dynamics (CFD) has been on a plateau with numerous production CFD codes with a well validated capability to solve the Reynolds-Averaged Navier-Stokes (RANS) equations for complex configurations over a wide range of Mach numbers. Advances in computer hardware will enable Large-Eddy Simulation (LES) for industrial problems of interest within the foreseeable future. In order to enable LES for complex configurations, there is a need for methods using unstructured meshes.
Given the current computational capabilities, it is recognized that developing convergence acceleration and hybrid RANS-LES methodologies is crucial in order for LES to become a viable option in a very broad range of applications, including high Reynolds number massively detached flows and noise prediction. There is also general consensus within the scientific community about the necessity to direct research efforts toward high-order methods rather than the more common, and excessively dissipative, low-order upwind biased schemes (Fujii, 2005, Prog. Aerosp. Sci. 41; Georgiadis, et al., 2010, AIAA J. 48:8). Hence it is necessary to combine high-order unstructured numerical schemes with advanced SGS modeling techniques in order for LES to become a valuable and reliable tool for fundamental flow physics and industrial applications.
High-order numerical schemes for solving the compressible Navier-Stokes equations on unstructured grids have been widely studied during the last decade. By far the most mature and widely used of these schemes are based on the Discontinuous Galerkin (DG) method. Recently, however, several alternative high-order methods have been proposed, including Spectral Difference (SD) and Flux Reconstruction (FR) type schemes, which potentially offer increased efficiency compared with DG methods. Within the framework of the SD and FR methods, the Aerospace Computing Laboratory (ACL) at Stanford University have focused extensive research in the past years and successfully addressed viscous compressible flows with shocks, implicit LES of turbulent channel flow and flow around circular cylinders, as well as, transitional turbulent flows. Other relevant applications include the implementation of adaptive refinement and implicit time integration techniques and moving and deformable meshes.
More recently, the implementation of explicit filtering structural SGS models for LES has been addressed.
The object of the present talk is to give a short overview on the actual status of the development of the high-order SD scheme and its capabilities in a relatively broad range of applications.
Mardi 27 Mars 2012
14 h 00 salle de réunion CEAT
Professor of Mathematics
Department of Mathematics & Statistics
McMaster University, Hamilton, Canada
Invité par Bernd Noack
I will survey wavelet-based techniques that have been developed over the last 20 years for the analysis and computation of turbulent flows. The analysis techniques allow the measurement of local spectral and singular properties associated with particular flow structures, such as vortices. Wavelet filtering, which uses the compression and denoising properties of wavelets, forms the basis of a group of techniques for dynamically adaptive computation of turbulent flows and fluid-structure interaction. These techniques range from wavelet direct numerical simulation, to a local form of large eddy simulation called SCALES. Examples will include experimental mixing layer data (from Poitiers!), vortex interaction, fluid-structure interaction, two-dimensional turbulence and shallow water turbulence.
Lundi 26 Mars 2012
14 h 00 salle de réunion CEAT
Research Scientist
DLR German Aerospace Center
Institute of Aerodynamics and Flow Technology
Invité par Peter Jordan
Tip vortices generated by a helicopter rotor are highly dynamic velocity fields which remain close to the rotor plane for several revolutions. The passing blades interact with the tip vortices, resulting in a phenomenon known as blade-vortex interaction (BVI). The interaction is accompanied by significant airloads and noise generation. One of the governing parameters to reduce the induced BVI noise is the blade position with respect to a wake. The position can be changed actively by using a constantly moving flap located close to the blade tip. For the measurements of the spatial position of the blade tip as well as of the whole blade due to the motion of the flap, a stereo pattern recognition (SPR) technique can be applied. The main idea of this technique is based on the 3d object reconstruction by known positions of markers attached to the region of the interest and captured by cameras. For the case reported here, the markers were distributed over the entire length of the blade and the blade tip and were illuminated during the measurements by stroboscopic light. The stereo pattern recognition is proved to be a reliable technique for detection of spatial coordinates. The results obtained with the SPR can be easily used for further evaluation of blade motion parameters in flap, lead-lag and torsion.
Mardi 20 Mars 2012
14 h 00 salle de réunion Branche Fluides
Directeur de Recherche
Laboratoire de Physique et Mécanique des Milieux Hétérogènes - UMR CNRS 7636
Invité par Bernd R. Noack
Pour l'étude des instabilités hydrodynamiques et de la transition à la turbulence, nous nous concentrons sur le rôle des structures cohérentes présentes dans les écoulements. Celles ci, avec vorticité transversale et longitudinale, prennent la forme de rouleaux, de stries et de hairpins. Ces structures jouent un rôle important dans la dynamique des écoulements dit ouverts, comme dans les tubes, les canaux ou les parois ainsi que pour le contrôle des écoulements.
Nous présenterons des expériences mettant en jeu un mécanisme tridimensionnel et nonlinéaire, d'auto-entretien de la turbulence (Self-Sustaining Process), qui semble générique dans les écoulements cisaillés. Ces travaux ont été réalisés au laboratoire PMMH de l'ESPCI, avec J.L. Aider, T. Duriez, S. Goujon-Durand et G. Lemoult.
Short bio: José Eduardo Wesfreid, Directeur de Recherche au CNRS, est au laboratoire de Physique et Mécanique des Milieux Hétérogènes (PMMH) de l'ESPCI. Il s'intéresse aux études expérimentales sur les instabilités hydrodynamiques, la transition à la turbulence et le contrôle des écoulements.
Mardi 6 Mars 2012
14 h 00 salle de réunion CEAT
PhD student
Duke University, Mechanical Engineering
Invité par Bernd Noack
It has often been remarked that turbulence is the last unsolved problem in Newtonian physics. The aggregate cost to our society resulting from our incomplete understanding of turbulence is quite staggering. Consider, for example, the environmental and economic costs associated with sub-optimal performance of virtually every fluid-thermal system such as the internal combustion engine and heat exchangers. The turbulence problem might appear puzzling to the outsider. After all, the equations governing turbulent flow, the Navier-Stokes equations, have been known for over a century and are deterministic and relatively compact. Turbulence remains a problem today because the only truly reliable tool for its approximation - direct numerical simulation - is computationally prohibitive. In this talk, I will summarize a new projection-based approach to model order reduction of the Navier-Stokes equations.
Short bio: Maciej (Mike) Balajewicz is a Ph.D. candidate in Mechanical Engineering in Duke University under the supervision or Professor Earl Dowell. He completed his M.S. and B.Eng. in Aerospace Engineering in Carleton University in Canada. His research interests include Applied Mathematics, Fluid-Structure Interactions and Fluid Dynamics.
Lundi 5 Mars 2012
10 h 00 salle de réunion CEAT
Department of Mathematics & Statistics
McMaster University, Canada
Invité par Bernd R. Noack
Short bio: Dr. Protas is an applied mathematician interested in problems combining theoretical and computational fluid dynamics with the theory of optimization and control. He holds a Ph.D. awarded jointly by the Warsaw University of Technology and Universite Pierre et Marie Curie (PARIS VI) in 2000. After post-doctoral experience at the University of California, San Diego, he joined the Department of Mathematics and Statistics at McMaster University in Canada as SHARCNET Chair in Scientific Computing, and has since been on the faculty there. Dr. Protas is a receipient of Ontario's Early Researcher Award and serves on the Editorial Board of the International Journal of Computer Mathematics (Section B).
Lundi 5 Mars 2012
14 h 00 salle de réunion CEAT
Department of Mathematics & Statistics
McMaster University, Canada
Invité par Bernd R. Noack
Short bio: Dr. Protas is an applied mathematician interested in problems combining theoretical and computational fluid dynamics with the theory of optimization and control. He holds a Ph.D. awarded jointly by the Warsaw University of Technology and Universite Pierre et Marie Curie (PARIS VI) in 2000. After post-doctoral experience at the University of California, San Diego, he joined the Department of Mathematics and Statistics at McMaster University in Canada as SHARCNET Chair in Scientific Computing, and has since been on the faculty there. Dr. Protas is a receipient of Ontario's Early Researcher Award and serves on the Editorial Board of the International Journal of Computer Mathematics (Section B).
Mercredi 22 Février 2012
14 h 00 salle de réunion Branche Fluides
Department of Mathematics & Statistics
McMaster University, Canada
Invité par Bernd R. Noack
In the presentation we will review recent results and discuss some emerging research directions concerning application of the modern methods of PDE-constrained optimization to address two classes of fundamental problems in turbulence research. The first class of problems is related to the applicability of some statistical theories, such as the Kraichnan-Leith-Batchelor (KLB) theory of the forced homogeneous isotropic two-dimensional turbulence, whereas the second class of problems concerns sharpness of certain mathematical estimates, such as the bounds on the maximum enstrophy growth in 3D flows. The latter issue is intimately related to the question of spontaneous singularity formation, known as the ``blow-up'' problem. We demonstrate how new insights concerning these problems can be obtained by formulating them as variational PDE optimization problems which can be solved computationally using adjoint-based gradient descent (or ascent) methods. In offering a systematic approach to finding flow solutions closest to the predictions of a given theory, the proposed paradigm provides a bridge between theory and computation. In the presentation we will show a number of new results concerning 2D Navier-Stokes flows characterized by the maximal growth of palinstrophy, and will discuss their relation to the available theoretical bounds obtained with rigorous methods of mathematical analysis. [joint work with D. Ayala]
Short bio: Dr. Protas is an applied mathematician interested in problems combining theoretical and computational fluid dynamics with the theory of optimization and control. He holds a Ph.D. awarded jointly by the Warsaw University of Technology and Universite Pierre et Marie Curie (PARIS VI) in 2000. After post-doctoral experience at the University of California, San Diego, he joined the Department of Mathematics and Statistics at McMaster University in Canada as SHARCNET Chair in Scientific Computing, and has since been on the faculty there. Dr. Protas is a receipient of Ontario's Early Researcher Award and serves on the Editorial Board of the International Journal of Computer Mathematics (Section B).
Lundi 9 Janvier 2012
14 h 00 salle de réunion Branche Fluides
Professor of Mechanical Engineering
University of Michoacan, Mexico
Invité par Gérard Touchard
L'interaction entre les contraintes mécaniques dans les matériaux solides et la polarisation des charges électriques au sein de ces matériaux a été mise en évidence depuis le début du XIX siècle par les phénomènes piézoélectriques.
Les applications industrielles des dispositifs piézoélectriques ont vu une évolution exponentielle dans les dernières décennies ; toutefois, l'étude de l'amorçage et de la propagation de fissures dans des matériaux solides soumis à des contraintes mécaniques a été développée principalement à partir de modèles mécaniques-énergétiques : le facteur d'intensité de contrainte, l'intégrale J, Loi de Paris,...
Etant donné que l'amorçage et la propagation de fissures constituent un phénomène essentiellement énergétique (énergie de déformation élastique et plastique, dissipation de chaleur, relaxation des contraintes,...), quelques études ont été dirigées pour comprendre la relation entre les phénomènes thermiques et l'amorçage et la propagation de fissures ; toutefois, très peu d'études ont abordé le couplage entre la variation des propriétés électriques des matériaux solides dans la zone de fissure et la dynamique de propagation de celle-ci.
L'objectif de ce séminaire est d'apporter quelques réflexions sur les possibilités de mettre en place l'étude sur l'amorçage et la propagation de fissures dans des matériaux solides, en associant le phénomène mécanique aux propriétés électriques dans le matériau à l'échelle macroscopique et de micro-plasticité.
Short bio: Doctor Gonzalo M. Dominguez-Almaraz is Professor at the University of Michoacan (UMSNH), Mexico. His main research interests are actually in ultrasonic fatigue tests (gigacycle fatigue tests) on metallic alloys: steels, cast irons, aluminums,... and high strength plastics and the interaction with the environment: corrosion, humidity, natural and artificial pitting holes, temperature, stress concentration and fretting fatigue under ultrasonic testing. Applications range from fatigue life evaluation on mechanical elements and systems attaining the very high cycle fatigue regime (>108 cycles), present in most of modern industries: aeronautical, energy production, automotive, oil, high speed trains... to the analysis of fracture surface to determine the failure origin and the interaction with the manufacturing, environmental, machining, and loading parameters of testing materials. He has developed an ultrasonic fatigue machine to carry out these tests.
Laurent Cordier Dernière modification le 31/01/2011