Search

A generic external-compression inlet model was tested to investigate the effect of duct height on terminal shock/boundary-layer interactions, with special sidewall arrangements to eliminate corner separations. Experimental results showed that the separation length significantly decreased by 41% as the duct height was reduced from 60 mm [Formula: see text] to 40 mm [Formula: see text]. To obtain the influence of the duct height in a wider range of values, Reynolds-averaged Navier–Stokes simulations were further performed. The scales of the [Formula: see text]-shock structure and separation region decrease as the duct height decreases. Shrinkage of the scale of the λ-shock structure results from a decrease in separation length. The downstream shift in the separation point and the upstream shift in the reattachment point both contribute to this decrease, but the upstream shift in the reattachment point is a dominant reason. The underlying mechanisms of the variation in the reattachment position as the duct height decreases include two aspects. The first is the stronger pressure gradient in the wall-normal direction immediately behind the shock system, which produces a stronger radial pressure gradient above the bumplike separation bubble, quickly turning the streamlines toward the bottom wall and enclosing the bubble. The second results from the postshock acceleration of the irrotational flow above the separated layer as the duct height decreases. An accelerating mainstream with a favorable pressure gradient above the separated layer increases the mixing strength and momentum exchange between the low-speed layer near the wall and high-speed flow above the separated layer.

A generic external-compression inlet test model is designed and fabricated to investigate the flow characteristics of terminal shock/boundary-layer interaction for inlet applications with and without upstream lateral confinement. When the leading edge of the sidewall is positioned 150 mm upstream of the duct entrance, the bifurcation height of the [Formula: see text] shock, the centerline separation length, and the pressure rise ratio at [Formula: see text] are close to reported values in the literature, although the corner separation herein is minor as compared with previous experiments with thick sidewall boundary layers. However, when the leading edge of the sidewall is aligned with the duct entrance, an obvious increase in the bifurcation point height can be observed, and an increase in centerline separation can be inferred from schlieren photographs. The significantly intensified separation is verified directly by surface oil flow visualization. In addition, the focus previously present in the corner region is replaced by a reattachment node located near the junction region of the bottom plate and sidewall leading edge. The intensification of the separation is attributed to a local high-pressure region caused by the impingement of the postshock subsonic flow onto the sidewalls.

The recognition of supersonic inlet flow pattern has become a research hotspot in recent years. In this paper, the dual external pressure supersonic inlet is taken as the research object. To explore the flow characteristics of the inlet, time-mean processing on the inlet pressure signal collected by sensors is conducted first, and the features of the inlet pressure data in time domain and frequency domain are extracted, respectively. As feature selection (FS) plays an important role in classification tasks and has been recently studied as a multi-objective optimization problem, two objectives of FS are considered and an improved non-dominated sorting genetic algorithm NSGA2 with hybrid mutation operators using support vector machines (SVM) as classifiers is proposed, aiming to simultaneously select feature subsets and optimize SVMs hyper-parameters. In addition, a way to deal with variation transgression is proposed to make the mutation operator of the single-objective evolution fit well in the multi-objective evolution algorithm. Experimental results on 31 sensor datasets demonstrate that our proposed algorithm can achieve competitive classification accuracy while obtaining a smaller size of feature subset compared with particle swarm optimization algorithm and some multi-objective optimization algorithms using single-objective evolution mutation operators.

The variable geometry supersonic inlet tends to decrease the throat area to reduce the Mach number upstream of the terminal shock, so as to reduce the flow loss. However, excessive Internal Contraction Ratio (ICR) exposes the inlet to a greater risk of unstart, which inevitably results in a process of increasing the throat area to aid the inlet restart. In the above throat regulation process, the inlet undergoes the start, unstart, and restart states in turn. In order to reveal the flow structure and mechanism of this process, a two-dimensional unsteady numerical simulation combined with a dynamic mesh technique were employed. The shock-on-lip Mach number of the studied inlet is 4.0 and the flight angle of attack is + 6°. Analysis was focused on the state with a freestream Mach number of 3.0. The results clearly show that the flow response hysteresis appears, and restart is only realized when the throat area is obviously increased as compared to that of unstart due to the historical unstart flow structure. In addition, three typical flow fields were analyzed, and it is found that the separation ahead of the inlet was the key factor affecting the hysteresis. Finally, unstart and restart boundaries of the inlet were discussed, and the factors influencing its deviation from the typical boundaries of dual-solution area were analyzed. The newly predicted unstart and restart boundaries are much closer to the CFD results.

Herein, a curved isolator with sidewall contraction of a hypersonic scramjet is extracted and simplified. The flow characteristics in the isolator with an entrance Mach number of 3.46 are studied. The results show that the lateral contraction of the sidewall generates a strong shock and its subsequent reflected shock, which induce glancing shock wave/boundary layer interaction (GSWBLI), rolling up large-scale vortices adjacent to the symmetry plane. As the downstream backpressure increases, the shock train propagates upstream and interacts with the glancing shock and vortices inevitably, forming a typical shock train/glancing shock/boundary layer interaction (STGSBLI) phenomenon. Different from the shock train in a straight isolator, it can be divided into two distinct parts, i.e., the center part and the side part. For the center part, it behaves in a quasi-two-dimensional manner, which is primarily associated with STGSBLI, forcing the low-momentum subsonic flow to principally accumulate near the symmetry plane and weakening the spanwise pressure gradient in the center part. For the side part, it locates behind the glancing and reflected shocks and is much shorter than the center part, resulting in a streamwise extension of the supersonic flow region near the sidewall. As the shock train moves upstream, the quasi-two-dimensional region enlarges and extends to the sidewall. It is demonstrated that the aforementioned two parts of the shock train can also be discovered in curved sidewall-contraction isolators with different centerlines. Therefore, the STGSBLI is the dominant and universal physical phenomenon in isolators with sidewall contraction.

Görtler vortices (GVs) in dual-incident shock-wave/turbulent-boundary-layer interactions (dual-ISWTBLIs) are experimentally investigated in a Mach 2.48 flow. A double-wedge shock generator with two deflection angles of 8° and 5° is used to produce two incident shock waves (ISWs). Flow structures of the experiments with three different shock-wave distances were visualized by the ice-cluster-based planar laser scattering technique at two orthogonal planes ( x– y and x– z planes). The images in the x– y plane present three types of flow patterns of dual-ISWTBLIs corresponding to the first type with a triangle-like separation, the second type with a quadrilateral-like separation, and the third type with two isolated interactions induced by the two ISWs. The images in the x– z plane indicate that the GVs exist in the first type of dual-ISWTBLI originating in the vicinity of the apex of the separation region and cover nearly the whole spanwise range of the reattachment region. By comparison, the GVs intermittently occur in the limited spanwise range of the reattachment region in the second type of dual-ISWTBLI. No GVs are observed in the third type of dual-ISWTBLI because no visible separation is induced under the experimental conditions considered in this situation. In addition, based on the wall-pressure distribution in the former two types of dual-ISWTBLIs, this paper proposes a method to estimate the mean-flow streamline curvature in the reattachment region, thereby obtaining the criteria for the existence of GVs, according to which reasonable explanations for the different distributions of GVs in the two types of dual-ISWTBLIs are provided.