• Home
• Program
• Contact

16:15   Instability, Transition and Control of Turbulent Flows 5 16:15 15 mins #237 TURBULENT DRAG REDUCTION FOR A WALL WITH A BUMP Jacopo Banchetti, Maurizio Quadrio Abstract: This work investigates the effects of turbulent skin-friction drag reduction techniques applied over non-planar walls, with a view to understanding the relationship between skin-friction drag reduction and changes in the total drag. The existing proofs of concept for skin-friction drag reduction are mostly limited to (i) low-Reynolds number, and (ii) very simple geometries, such as flat plates. Recently, [1] showed that limitation (i) does not preclude spanwise forcing techniques from obtaining large amount of drag reduction, even at flight Re. In order address limitation (ii), [2] recently assessed via a RANS-based calculation the effects of riblets on the total aerodynamic drag of an airplane in transonic flight. They observed a total drag reduction higher than expected due to secondary effect on the pressure drag. The present contribution intends to confirm this result, by employing more reliable prediction tools (DNS instead of RANS), and by considering a skin-friction reduction technique capable to provide larger and clearer effects than riblets. DNS simulations are carried out for a turbulent flow in a channel at Re b = 3173, where one wall possesses a small-height bump. The computational domain has dimensions of (Lx , Ly , Lz ) = (24.56h, πh, 2h) in the streamwise, spanwise and wall-normal directions respectively, being h the half-width of the channel in the flat sections. Around 9 × 10^7 grid points are used. The considered skin-friction reduction technique is the streamwise-travelling wave (STW) of spanwise velocity [3]. Figure 1 shows isosurfaces of the intermediate eigenvalue λ2 of the velocity gradient tensor and shows the local increase of turbulent activity due to the bump. To investigate the effect of STW on total drag, we plot in fig. 2 and 3 the streamwise distribution of the friction coefficient (C f (x) = 2τ w /ρU b ) and the pressure coefficient (C p (x) = 2p/ρU b ). In both reference and controlled case, the bump induces flow separation, where friction becomes negative: the separation bubble is extended by STW. Figure also plots the local friction reduction (R(x) = (C f − C f,stw )/C f ), which far from the bump takes the expected value of 45% for the indefinite channel flow. Pressure in the reference case increases before the bump, then a minimum is reached at its tip, followed by a recompression. The effect of STW is non-trivial. The positive pressure peak before the bump is noticeably reduced, and with it, the pressure drag associated to the anterior part of the bump. Also, the pressure minimum near the bump top is decreased by STW. Overall, the changes in the pressure distribution, once translated into drag changes by accounting for the geometry of the bump, result in an additional 10% of pressure drag reduction. Qualitative changes are also observed in several turbulent statistics, which will be presented at the conference. [1] D. Gatti and M. Quadrio. Reynolds-number dependence of turbulent skin-friction drag reduction induced by spanwise forcing. J. Fluid Mech.,802:553–58, 2016. [2] B. Mele, R. Tognaccini, and P. Catalano. Performance assessment of a transonic wing-body configuration with riblets installed. J. Aircr.,53(1):129–140, 2016. [3] M. Quadrio, P. Ricco, and C. Viotti. Streamwise-traveling waves of spanwise wall velocity for turbulent drag reduction. J. Fluid Mech., 627:161–178, 2009. 16:30 15 mins #480 Statistical characterization of viscoelastic flows in extended domains Alessia Ferraro, Tobias M. Schneider Abstract: The introduction of a tiny amount of elastic polymers to a Newtonian fluid can efficiently reduce the friction losses in the turbulent regime and thereby the pressure drop necessary to sustain a given flow rate. While at the onset the drag reducing effect increases with the polymer concentration and the Reynolds number, it is eventually bounded by the universal maximum drag reduction asymptote (MDR) that is shown to be independent of the rheological properties of the solution. Despite the numerous industrial applications, the physical origin of this phenomenon remains incompletely understood. Many numerical observations have shown that polymers do not decrease the wall drag homogeneously in space and time across the domain revealing the spatio-temporal nature of the phenomenon. Regions of highly dissipative turbulence coexist with regions characterized by lower turbulent intensity, they interact with each other, evolve in time and are advected by the mean flow. Whether this state corresponds to the MDR asymptote or it is a step in the reverse transition prior the relaminarization, understanding the local interaction between polymers and turbulence is key for understanding polymer drag reduction.\\ In order to characterize this spatio-temporal intermittency we compute the statistical properties of viscoelastic flows in extended domains for different flow parameters. Simulations of viscoelastic pressure driven channelflows, modeled via FENE-P equations, are carried out in computational domains that are large enough not be affected by finite size effects. The aim is to quantify the emergence and the evolution of local low-drag structures in terms of frequency, characteristic size, temporal permanence as well as their correlation with the polymers dynamics. 16:45 15 mins #586 NUMERICAL INVESTIGATION OF TURBULENCE DEVELOPMENT IN THE CHANNEL WITH A SMALL CONE ANGLES Alexander I. Reshmin, Sergey Kh. Teplovodskii, Vladimir Trifonov Abstract: We numerically investigate the phenomenon of laminar-turbulent transition in diffusers, of different cone angles and expansion ratios of the channel. The calculation was conducted by using a Lushchik-Pavel'ev-Yakubenko model of shear turbulence [1]. At the entrance of the channel, either a developed turbulent pipe flow or a flow with a uniform velocity profile with different turbulent energies was applied. The calculations were performed for channel expansion ratios of 1.65, 2.5, 3.4 and 6.8, with the cone angle varying from 0.01° to 4° and the Reynolds numbers changed from 1000 to 10,000 at the channel entrance. The results of the calculations were compared with previously obtained experimental data [2] for a diffuser with cone angle of ~ 0.6°. Experimental data are in good agreement with the calculated data. Calculations showed that with an increase of the diffuser length, the turbulent flow disappears and smoothly transforms into laminar flow. The critical Reynolds number Re* vs the diffuser cone angle θ is obtained (Figure 1). This kind of diffusers can be used in industry, when designing various devices. As an example, such diffusers with sufficiently low speeds inside can be used in the new type of heat exchangers. 17:00 15 mins #592 Streamwise-constant large-scale structures in Couette and Poiseuille flows Simon Illingworth Abstract: This work will consider the linear amplification mechanisms leading to streamwise-constant large-scale structures in laminar and turbulent channel flows. A key feature of the analysis is that the Orr--Sommerfeld and Squire operators are each considered separately. Physically this corresponds to considering two separate processes: (i) the response of wall-normal velocity fluctuations to external forcing; and (ii) the response of streamwise velocity fluctuations to wall-normal velocity fluctuations. In this way we exploit the fact that, for streamwise-constant fluctuations, the dynamics governing the wall-normal velocity are independent of the mean velocity profile (and therefore the mean shear). The analysis is performed for both plane Couette flow and plane Poiseuille flow; and for each we consider linear amplification mechanisms about both the laminar and turbulent mean velocity profiles. The work reveals two things. First, that the most amplified structures---with a spanwise spacing of approximately 4h irrespective of the details of the mean flow---are to a great extent encoded in the Orr-Sommerfeld operator alone. Second---and consistent with numerical and experimental observations---that Couette flow is significantly more efficient than Poiseuille flow (with the efficiency suitably defined) in leveraging wall-normal velocity fluctuations to produce large-scale streamwise streaks. 17:15 15 mins #446 SENSITIVITY ANALYSIS OF ANALYTICAL MODELS FOR THE PREDICTION OF TRAILING-EDGE NOISE Gerardo Zampino, Andrea Ferrero, Renzo Arina Abstract: Airfoil self noise occurs when an airfoil is placed in a disturbed or uniform and steady fluid flow. As in most aeroacoustic noise generation situations, noise is generated by flow unsteadiness. In the case of airfoil self noise, it is the interaction of flow unsteadiness, in form of fluid turbulence, with the airfoil surface leading to broadband noise. Though less painful than tonal noise, broadband noise is a matter of great interest in technologies that utilize airfoils and airfoil-like shapes. It may become the major contribution either because tonal noise is hidden due to its low level, or because tonal noise escapes the range of human hearing. Moreover broadband noise is the only remaining contribution for surfaces in non accelerated motion, such as structures in the wind. There are mainly two broadband noise-generating mechanisms, namely the interaction with upstream turbulence, the turbulence-interaction noise, involving the breakdown of oncoming vortices on the leading edge of the airfoil, and the boundary-layer turbulence scattering at the trailing edge, referred as trailing-edge (TE) noise. Turbulent eddies are formed within the boundary layer and it is the interaction of these eddies with the TE that generates broadband aerodynamic noise. In acoustic terms, the edge presents itself as a sharp impedance discontinuity, scattering acoustic waves generated by fluid turbulence and creates an intensified radiated acoustic field. The focus of the present work in on TE noise generation and its prediction. Calculating TE noise is made difficult due to the complexity of the noise source, which is a turbulent flow. Numerical approaches based on direct simulations or hybrid techniques are aimed at understanding the tiniest details of the sound generating mechanism. This is achieved at a price of time-consuming computations. On the other side, analytical methods, based on acoustic analogy and wave scattering theory, are not devoted to reproducing the details of noise generating mechanisms, but rather to provide acceptable order of magnitude of the generated sound with fast and inexpensive calculations. They provide the right trends and scaling laws which are needed for the assesment of low-noise airfoil design.The TE noise radiated by aNACA0012 airfoil, at zero incidence, is studied. The mean turbulent flow past the airfoil is calculated solving the Reynolds Averaged Navier-Stokes equations with the Spalart-Allmaras turbulence model. The boundary-layer thickness at the TE, an empirical wall pressure spectral density and the correlation length, calculated introducing the Corcos hypothesis, as function of the frequency, are the inputs for the analytical models describing the acoustic radiation. The most popular models, proposed in the literature, are compared. Namely the models of Amiet [1], Goody [2], Rozenberg [7], Kamruzzaman [5] and Lee [3]. The predicted pressure spectra are compared with the experimental data [6, 4] in Figure 1. A sensitivity analysis will be carried out in order to understand how the output of each acoustic model is influenced by the boundarylayer parameters predicted by the RANS simulations. In particular, a Monte Carlo method will be adopted to estimate the uncertainty propagation for the different analytical models. 17:30 15 mins #487 EXPERIMENTAL INVESTIGATION OF LAMINAR-TURBULENT TRANSITION IN SUPERSONIC BOUNDARY LAYER ON SWEPT WINGS Nikolai Semionov, Alexander Kosinov, Vasiliy Kocharin, Yury Yermolaev, Alexander Semenov, Boris Smorodsky, Aleksey Yatskikh, Alexandra Panina, Gleb Kolosov Abstract: The paper is devoted to an experimental study of laminar-turbulent transition and instability disturbances evolution in a three-dimensional supersonic boundary layer on swept wing. The turbulence origination process in 3D boundary layers is one of the poorly studied fundamental problems in fluid mechanics. Simultaneously, those studies are of interest from the practical point of view as they concern the flows around aircraft swept wings. The latter is related to the fact that a number of instabilities can arise in 3D boundary layer on a real swept wing; such instabilities include the TollmienSchlichting instability, which leads to the transition in 2D case; cross flow instability expressed in the form of stationary and travelling disturbances, etc. Excitation and evolution of all instability disturbances and their relative role in transition strongly depend on the environmental conditions. Transition Reynolds numbers obtained in different wind tunnels may differ significantly. The influence of flow parameters, such as the Mach number, unit Reynolds number, swept angle, angle of attack and external disturbances on the transitions to turbulence onset are under the consideration. Experimental results obtained during last ten years are presented and some problems with determination of transition position in supersonic boundary layer on swept wings are discussed. The experiments are conducted at the Institute of Theoretical and Applied Mechanics of the Siberian Division of the Russian Academy of Sciences in the T-325 low nose supersonic wind tunnel with test-section dimensions 0.20.20.6 m at Mach numbers M=2  4. Several models of swept wings with supersonic or subsonic leading edge and different curvature are used in experiments. To determinate a transition location hot-wire sensor is used. The disturbances are measured by constant temperature hot-wire anemometer. Vortex external perturbations are produced by a wire of various diameters, stretched in front of nozzle inserts. Evolution of natural disturbances in supersonic boundary layer of swept wing is investigated at different Mach numbers, external vortex disturbances, unit Reynold number. Curves of natural disturbances growth are measured. Maxima in the distributions correspond to the location of laminar-turbulent transition. Characteristic zones of disturbances evolution are determined. It is found that for Mach numbers 2 and 2.5 measurements can be performed in the region of linear stage of disturbances evolution, and the experimental data can be compared with the results of the calculations on linear stability theory. In cases where this comparison was made, good agreement between the experimental and theoretical data was obtained. Obtained strong effect of mentioned above flow parameters on transitional Reynolds numbers. For example it is obtained that with the increasing of the unit Reynolds number the position of laminar-turbulent transition is shifted downstream. The transition position moves almost 1.6 times at Mach number 2 and varies approximately 1.3 times at M = 2.5. Some of these results are described in more detail in [1-4]. This work was financially supported by the Russian Science Foundation (Grant No. 17-19-01289) References [1] Yu.G. Yermolaev, A.D. Kosinov, N. V. Semionov. Experimental study of nonlinear processes in a swept-wing boundary layer at the Mach number M=2. Journal of Applied Mechanics and Technical Physics 55(5):764-772, 2014. [2] N.V. Semionov, A.D. Kosinov, Yu.G. Yermolaev. Experimental study of turbulence beginning of supersonic boundary layer on swept wing at Mach numbers 2 – 4. Journal of Physics: Conference Series (JPCS) 318: No.032018, 2011 [3] Y.G. Ermolaev, A.D. Kosinov, A.N. Semenov, N.V.Semionov, A.A. Yatskikh. Effect of unit Reynolds number on the laminar-turbulent transition on a swept wing in supersonic flow. Thermophysics and Aeromechanics 25(5): 659-665, 2018. [4] Semionov N.V., Yermolaev Yu.G., Kocharin V.L., Kosinov A.D., Semenov A.N., Smorodsky B.V., Yatskikh A.A. An effect of small angle of attack on disturbances evolution in swept wing boundary layer at Mach number M=2 AIP Conference Proceedings. -Vol.2027, No.1. -S.l., 2018. -030156(6) p. DOI: 10.1063/1.5065250 17:45 15 mins #124 ON DRAG REDUCTION AND WAKE ASYMMETRY OF 3D BLUFF BODIES WITH LOCAL BASE BLOWING Manuel Lorite-Díez, Luc Pastur, José Ignacio Jiménez-González, Olivier Cadot, Carlos Martínez-Bazán Abstract: In this experimental study, we investigate the effect, on both the drag and the flow dynamics, of a steady perimetric base blowing at the turbulent wake behind a three-dimensional square-back model. At the Reynolds numbers under consider- ation, the system has successively undergone a pitchfork and a Hopf bifurcation. As a consequence, the mean wake flow is horizontally asymmetric, with two possible reflectional symmetry-breaking base flows, while the instantaneous wake flow sustains periodic vortex shedding [1]. Flow unsteadiness triggers random switchings between the two horizontally mirror-deflected flows, resulting in a bistable stochastic dynamics between the two states, as reported in [2]. [1] M. Grandemange O. Cadot, and M. Gohlke. Reflectional symmetry breaking of the separated flow over three-dimensional bluff bodies. Physical review E 86 035302, 2012. [2] M. Grandemange, M. Gohlke and O. Cadot, Turbulent wake past a three-dimensional blunt body. part 1. global modes and bi-stability. Journal of Fluid Mechanics 722: 51–84, 2013. 18:00 15 mins #465 Bifurcations in a shear-driven cavity Yacine Bengana, Laurette Laurette Tuckerman Abstract: A complete numerical investigation is carried out of the two-dimensional shear-driven flow in an open cavity. Two successive Hopf bifurcations take place to flows with two or three pairs of organized structures, which we call LC_2 and LC_3. These states are connected via secondary bifurcations to a branch of quasiperiodic states QP, which mediate the transfer of stability from LC_2 to LC_3.