Hence, the DNS approach is desirable for this kind of problem. Such a disagreement is mainly because of deficiencies in the models. It can be concluded that turbulence modelling of SBLI has in general, given a poor agreement with experiment, especially when transition exists and when large separation regions occur. (1999) using a full Reynolds stress model. for example Loyau, Batten & Leschziner (1998) using a nonlinear eddy-viscosity model and Batten et al. Channel flow with the Delery (1983) bump geometry has been studied in some detail by RANS. Further 3D studies are needed for strong interactions where the flow exhibits significant three-dimensionality and unsteady behaviour. Numerical studies of an incident oblique shock-wave interacting with a two-dimensional laminar boundary-layer have been carried out by Katzer (1989) and Wasistho (1998). A deflection angle of β = 18° was chosen to generate a small, but more than incipient flow separation, and a database was produced for model assessment. Adams (2000) carried out a direct simulation of turbulent boundary-layer flow over a compression corner at Mach number 3 and Reynolds number Re θ =1685 (based on the inflow momentum thickness). Although DNS is limited to low Reynolds number and simple geometries, it offers a complete reference for the given flow, which is difficult to obtain from experiment, and is invaluable for understanding flow physics and assessing turbulence models. The Reynolds-Averaged Navier-Stokes (RANS) approach has been widely used and direct numerical simulation (DNS), with the advantages of resolving all scales of fluid motions has also been adopted for the study of several problems. With advances in computer technology and the development of suitable numerical algorithms, computation of SBLI has become feasible. Squire (1988) into the curvature effect, using models of different radius and distinguishing between the shock-induced separation and the bump trailing-edge separation (due to the adverse-pressure-gradient). Various techniques were used in the experiments in order to establish the details of both the mean flow and the turbulence. The interactions were significant with a notable lambda-shock pattern and extensive shock-induced flow separation. Experimental investigations of shock/turbulent-boundary-layer interaction with non-zero pressure gradients have been carried out by Delery (1983) using a curved bump geometry, and by Liu & Squire (1988) using a circular-arc bump geometry. However, for many practical flows, the interaction takes place at transonic speed on a curved surface, where the developing boundary-layer is turbulent with non-zero pressure gradients. A review by Green (1970) summarized three major interaction scenarios: a sharp compression corner generating an outgoing oblique shock-wave, the reflection of an incident oblique shock at a plane wall, and a weak normal shock-wave interacting with a spatially-developing boundary-layer, in which no curvature effect was considered. Since then, considerable progresses have been made towards understanding the complex interaction mechanisms. Pioneering research into SBLI was carried out by Liepmann (1946), who did the earliest experiment on laminar and turbulent boundary-layers interacting with a normal shock-wave. In many practical situations the incoming boundary-layer is transitional or turbulent. Shock/boundary-layer interaction (SBLI) phenomena have important applications in a wide range of practical problems, for example transonic airfoils/wings, supersonic engine intakes, diffusers of centrifugal compressors, and turbine-machinery cascades. Sandham, in Engineering Turbulence Modelling and Experime1 INTRODUCTION Read moreĭNS OF TURBULENT FLOW OVER A BUMP WITH SHOCK/BOUNDARY-LAYER INTERACTIONS Sweepback delays the onset of separation and relieves the adverse effect of shock waves due to the reduced Mach number component normal to the isobars on the wing. The small aspect ratio relieves the streamtube constriction at local sonic velocities, and slows up the development of local supersonic velocities on the surface. Small aspect ratios and sweep-back both have a powerful influence in preventing the shock-induced separation. In order to prevent shock-induced separation, it is necessary to move the borderline AB upward and to the right, so that a greater range of flight conditions may be obtained. The line BB′ marks the border of leading-edge separated flow, from the shock-induced separated flow. B is extended to the left in order to indicate the onset of separation effects when the critical separation is of some low-speed type. Onset boundary and subdivision of regime for separation effects Īt point A, which represents zero lift, the shock is at the rear section, and at point B the shock moves to the leading edge.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |