Survey of control techniques to alleviate repercussions of shock-wave and boundary-layer interactions – Advances in Aerodynamics

It is evident that the consequences associated with SBLIs are numerous and most often severe for both internal and external aerodynamics. Since the repercussions are primarily dependent on the shock strength, boundary layer, and separation characteristics, the manipulation of the shock structure and the flow field by some suitable control techniques are essential. In the course of our discussion so far, different types of geometries are investigated. However, in the following sections, the discussions will be mostly oriented to the SBLIs in aircraft intakes.

An intake is an essential part of an aircraft engine where the incoming flow is decelerated for an efficient combustion process. Specifically, in ramjets and scramjets, the incoming flow is slowed by the intentionally generated series of shock waves [41, 84, 85]. However, due to the interaction of shock waves with the boundary layer, the detrimental consequences are evident. The repercussions are high viscous dissipation, enormous pressure loss, unsteady shock oscillations, etc. [1, 86]. These undesired phenomena need to be controlled to alleviate or reduce the colossal losses and to increase the stability of the interaction. Considering these aspects into account, the control techniques address and mitigate the issues related to wave drag and viscous losses. Normally, the control techniques are employed to reduce either the wave drag or the viscous drag as the physical mechanisms involved with these control techniques (for reducing the wave drag or the viscous drag) are different from each other.

According to Delery (1985), control mechanisms can be classified into two groups depending on the location of their deployment [38]. First, the control can be employed to improve the nature of the incoming boundary layer (such as mass injectors, vortex generators). Second, a control can be at the interaction region itself (such as boundary layer bleed or deployment of the porous cavity).

The SBLI control techniques are classified as the shock controls (cavity covered with porous surfaces, surface bumps, etc.) and the boundary layer controls (vortex generators). Essentially, the shock control splits or smears shock a strong shock into multiple shocks to reduce the total pressure loss and the boundary layer control method reduces the loss related to viscosity.

According to the energy requirement, the control technique may be classified into active control or passive control. In an active control, an additional source of energy is necessary, whereas, in a passive control, the control utilizes the energy from the main flow itself.

Active control techniques

With the advancements of modern intricate techniques, the active controls nowadays are gaining prominence for improving aerodynamic performance. From the perceptive of shock/boundary-layer interaction control in an intake, several works have been conducted on the tangential blowing or suction, as witnessed in the extensive review work of Delery [38] and Viswanath [39]. The tangential blowing inside the boundary layer increases the kinetic energy of the boundary layer fluids. Depending on the deployment of the tangential blowing upstream or at the interaction point, U-type or D-type mass injectors are utilized, respectively [87,88,89]. The deployment of an array of micro-jets, located upstream of the interaction region, significantly reduces the mean pressure across the separation and reduces the unsteadiness in the flow field [90, 91]. The reductions of separation bubble size with the increase in microjets pressure ratio are clearly shown in the study conducted by Ali et al. (2012) in Mach 2 flow, as shown in Fig. 5 [91]. It can be seen that the separation shock moves upstream and reduces its intensity with the increase in microjet strength. Besides, the air-jet vortex generator was also found to be effective in decreasing the separation length [92]. The experiment was conducted at flow Mach number 1.3. Essentially, the streamwise vortices generated by the air-jet vortex exchange the energy between the low and high momentum fluid and thereby energizes the fluid inside the boundary layer.

Schlieren flow visualization describing the effect of control with the variation of Microjet Pressure Ratio [91]

Another type of widely used active control (can be active or passive, depending on the energy requirement) is suction/surface bleed, where the low momentum fluid inside the boundary layer is sucked to improve the state of the incoming boundary layer [93, 94]. The deployment of the suction geometry relative to the interaction location governed the suction effectiveness. Fukuda et al. (1977) demonstrated that the suction location upstream or downstream of the interaction region is more effective than the interaction region itself at flow Mach number 2.5 [94]. However, this finding contradicts the assessment of Seebaugh and Childs (1970), where the suction geometry placed at the interaction location is found to be superior (at a freestream Mach number of 2.82 and 3.78). The discrepancies in these results may be due to the differences in the suction geometries [95]. Further, Weiss and Olivier (2014) investigated the influence of suction cavity pressure at Mach range 1.45 to 1.85 by introducing the normal suction slot on all four walls of a rectangular nozzle [96]. It is interesting to observe that, when the suction slot is not choked, barrier shock is generated at the suction slot’s downstream corner at a cavity pressure level lesser than the static pressure upstream of the shock. However, the cavity pressure cannot influence the interaction if the suction slot is choked. The efficacy of suction and tangential blowing in reducing the recirculation bubble size was investigated and compared by Sriram and Jagadeesh (2014) at Mach 5.96 [97]. They observed that a suction or injection rate at one order lesser than the momentum deficit essentially reduces the separation bubble size, with a maximum reduction of 20% for tangential blowing. The physical mechanism behind the suction and the surface bleed techniques are quite similar with only the exception that suction utilizes an external pump to draw the low momentum fluids. Although the surface bleed techniques are quite simple (in most cases, it does not require additional power to drain low momentum fluid), additional actuation techniques to control the incoming airflow rate and bleed rate (specifically for active control) may require additional energy, which eventually makes them more complex.

Recently, the importance of the plasma jet has gained much more attention due to its superior effectiveness in stabilizing SBLIs. Normally, the plasma actuators are deployed upstream of the interaction to excite the instability in the recirculated zone. Narayanaswamy et al. (2012) utilized an array of pulsed plasma jets at Mach 3 flow to reduce the unsteadiness in the separated region [98]. Further, the localized arc filament plasma actuators were used at Mach 2.3 by Webb et al. (2013), where they noticed that the actuators are effective in reducing the low-frequency unsteadiness [99]. The actuators essentially induce additional heat inside the boundary layer to modify its nature. Greene et al. (2015) demonstrated at Mach 3 flow condition that the implementation of the pulsed plasma jet suppressed the separation, particularly, the recirculation length from the separation point to the compression corner was reduced significantly [100]. Besides, magnetically driven surface discharge can effectively manipulate the boundary layer in controlling the SBLI at Mach 2.6 [101]. It was observed that the plasma actuation at a low current (below 80 mA) in the downstream side significantly alters the separation bubble structure, whereas, plasma actuation at a higher current (above 80 mA) in the downstream side significantly improves the separation bubble structure and wall pressure profile. When magnetic field strengths of more than 1 Tesla are applied, increasing the actuation current reduces the size and strength of the recirculation. Moreover, separation might be completely eliminated with magnetic field intensities of 3 Tesla and currents larger than 80 mA. The Magnetohydrodynamic (MHD) flow control technique also gained prominence in weakening the shock intensity by altering the boundary layer profile over the ramp surface. Essentially, MHD flow control can result in significant changes in flow behavior near the ramp, as well as there is a significant reduction in total pressure loss across the oblique shock [102]. It is also interesting to observe that the magnetic field causes the velocity of the plasma column to move at a much faster rate than that in the absence of the magnetic field.

Since, active control techniques involve complex mechanisms that sometimes become difficult to implement, particularly, in a hypersonic flight, where the temperature generated inside the intake is too high. This led the researchers to focus on passive control techniques more extensively.

Passive control techniques

In high-speed aerodynamics, passive control techniques are widely used for their simplicity and effectiveness. As already discussed, depending on the nature of the operation, the control techniques can be classified as shock control or boundary layer control. A shock control extends the interaction regions to reduce the strength of the primary shock; boundary layer control energizes the boundary layer to offer more resistance to separation. Some of the well-established control methods used to manipulate the SBLIs are boundary layer bleed, vortex generators, streamwise slots, porous cavity, surface bump, splitter, etc. [103,104,105,106,107,108,109,110,111]. Boundary layer bleed is the widely used control technique (can be active or passive, depending on the energy requirement), where the low momentum fluids near the wall are bled and the high momentum fluids outside the boundary occupy their location. The high momentum fluids eventually impose more resistance to separation [103]. Schulte et al. (2001) investigated the efficacy of the boundary layer bleed in hypersonic flow [104]. It was seen that the boundary layer bleed dramatically reduces the separation bubble size up to 50%. Figure 6 describes a clear reduction in separation bubble size over the ramp controlled with surface bleed when compared with the uncontrolled case. They have described the optimum location of the bleed system, where the maximum pressure recovery can be achieved. It is seen that the streamwise slots are responsible for the bigger lambda-shock structures, which essentially improve the total pressure losses. Besides, the slots shed the streamwise vortices into the flow field, which may be effective in delaying the separation [107, 112]. Later, it was shown that the use of a surface bump on the ramp surface effectively controls the interaction by replacing a bigger separation bubble with two smaller bubbles [109, 113,114,115]. Further, the separation region induced by SBLI, at the Mach number range 4 to 6, was reduced with the introduction of splitters in the internal convergent portion of the inlet, which degenerates a single strong interaction into multiple weak interactions [110]. The modification of the cowl surface also has a prominent impact on SBLI. At Mach 4.03 flow, Senthilkumar and Murganandam (2020) through their numerical investigation found that concavity on the cowl surface significantly controls the SBLI [116]. Later, to lessen SBLI-induced separation, the combined effects of boundary layer bleed and boundary layer suction have recently been computationally explored. Essentially, a secondary recirculation jet is computationally investigated in order to control the flow. To limit the size of the separation bubble, the suction slot was found to be best when deployed on the upstream side of the flow separation zone in the supersonic flow [117].

Schlieren flow visualizations for the uncontrolled ramp surface (reference case) and controlled ramp surface with surface bleed (bleed case) [104]

Among the passive control techniques, few researchers have looked at the use of a porous surface over a shallow cavity as a shock control system. The primary goal of a passive cavity is to reduce the overall pressure loss caused by the extended interaction area due to the smeared shock system. Raghunathan (1988) studied the effects of SBLIs and their controls using passive control techniques at transonic flow [108]. The research looked at the benefits of using a porous surface deployed over a cavity at the shock impingement point for monitoring SBLIs. As a result, recirculation occurs spontaneously across the shock from the high-pressure zone (on the downstream side) to the low-pressure zone (on the upstream side), as shown in Fig. 7. Consequently, the strong shock is broken or smeared into lambda-shock. Later, a systematic analysis of passive control with a perforated plate supports its effectiveness over supercritical airfoils [119]. Rallo et al. (1992) described that the deployment of a porous cavity is responsible for the shock strength reduction which in turn weakens the strength of the interaction [120]. It is important to note that the comment on SBLIs strength reduction can directly be made by observing a drop in the maximum wall static pressure at the interaction region. In addition, their investigation on the influences of surface porosity (introduced on a flat surface) on SBLIs at hypersonic flow indicated that the reattachment pressure decreases considerably when the surface porosity increases. Following that, in the experimental study, at Mach number range 1.56 to 1.65, McCormick [121] used both micro-vortex generators (MVGs) and a shallow cavity to manipulate the SBLIs. Due to the influence of a porous wall with a shallow cavity underneath (with evenly spaced pores) on a flat plate, the generated lambda-shock greatly decreased the shock intensity, as shown in Fig. 8 [121]. However, on the upstream side of the shock, degradation of the boundary layer (boundary layer thickening due to fluid injection) is observed.

Control of SBLI using porous cavity [118]

Wall static pressure distributions [121]

Hanna (1995) investigated SBLI control in a hypersonic flow using a porous surface [122]. Because of upstream pressure transmission across the subsonic portion of the boundary layer, the interaction area for viscous flow is larger than for inviscid flow, according to the research. In addition, constant and variable porosity-controlled models were compared in the study. Surprisingly, the peak pressure comparably decreases in both cases. Bur et al. (1997) explored the effect of a porous cavity on a transonic airfoil when it is working in off-design conditions [123]. The porous cavity was shown to break a single shock into a lambda-shock system. The results showed a substantial decrease in wave drag; however, there was a significant increase in viscous drag. As a result, the net drag reduction was negligible. Nonetheless, under ideal flow conditions, a shallow cavity with a porous wall was proved to be very efficient in regulating SBLIs over a plain surface at a supersonic level of maximum Mach number of 1.35 [124]. The use of a porous surface effectively converts normal shock to lambda shock. As a result of this, the overall pressure loss is reduced, which decreases the typical shock strength. However, the boundary layer degrades downstream of the interaction region, as shown in Fig. 9. Therefore, the ultimate result is a balance of higher viscous losses and lower shock losses. The Schlieren view of the shock structure with and without the porous cavity is presented in Fig. 10. This clearly shows that the porous cavity effectively reduces the shock strength.

Control of Shock/Boundary-Layer Interaction using porous cavity [124]

Schlieren flow visualization (a) without and (b) with the porous cavity [124]

Notice that, these studies concentrated the SBLIs over a flat plate using the externally generated shock wave alone. However, in hypersonic intakes, the influence of expansion waves created at the convex corner of the ramp on the SBLIs cannot be neglected. Thus, Mahapatra and Jagadeesh (2008) investigated cowl and ram-induced multiple shocks inside a generic scramjet intake (Mach 8), where typical features of the shock structure at different contraction ratios were observed [125]. To numerically investigate the impact of throat area and incoming Mach number fluctuations, James and Kim [126] considered a hypersonic mixed compression intake at freestream Mach number 2, 3, and 5. They showed that an increase in the area ratio leads to a possible decrement in the mass flow rate. Besides, at higher Mach numbers, total pressure recovery is less [126]. Zhang et al. (2014) used a computational model to analyze the influence of expansion waves on SBLIs in a hypersonic intake with an upstream Mach number of 3.5 [127]. Expansion waves have both positive and negative effects. The shock-shock-expansion waves and shock-expansion wave-shock interactions are particularly beneficial in controlling the SBLI when the cowl shock occurs near the convex corner of the ramp. As a result, the separation bubble size is decreased. Recently, the influence of a porous cavity on SBLIs, mounted inside a Mach 2.2 supersonic intake, has been studied for various intake contraction ratios by Gunasekaran et al. [128]. The study revealed that the performance of a porous surface with a cavity is strongly dependent on the contraction ratio of a supersonic inlet. However, when the porous cavity is introduced, the boundary layer worsens, and the shape factor downstream of the interactions increases, as compared to the boundary layer in uncontrolled intake [121]. Therefore, to overcome the shortcomings of the shock control methods, the boundary layer control methods such as the micro-vortex generators (MVGs) were extensively studied.

The use of micro-vortex generators (MVGs) to reduce shock-induced separation has gained popularity in recent years due to their interesting result [105]. Furthermore, vortices of mixed size shed in a streamwise manner can effectively control the size of the recirculation field, according to the findings. These vortices can be shed by inserting traditional or sub-boundary layer vortex generators into the flow [129]. Through their research in transonic flow, Inger and Siebersma (1989) showed that placing vortex generators upstream of the contact point energizes the incoming boundary layer wave [130]. As a result, the pressure gradient around the contact zone rises, ultimately creating a strong shock. To suppress or delay the separation of the boundary layer, traditional vortex generators (VGs) of a height equal to the thickness of the undisturbed boundary layer have been used [131, 132]. However, sub-boundary layer vortex generators (SBVGs) with a height less than the undisturbed boundary layer thickness are more efficient in separation control than traditional vortex generators because they have less drag [133, 134]. MVGs are indeed very effective instruments for separation control; however, lowering the MVG’s height below 10% of the boundary layer thickness is not effective in managing shear layer separation [135]. It was concluded that, micro-vortex generators (MVGs) shed adequate strength vortices in reducing separation time, unlike their conventional counterparts, where the vortex strength is too high and disrupts the flow [136]. MVGs’ distribution, in addition to their height, plays an important role in flow management. Due to their capacity to inhibit shock-induced separation, the array of MVGs has recently received a lot of research motivation [137, 138]. The flow control performance of the array of MVGs is also affected by their deployment position. The ability of the array of MVGs to monitor flow is also affected by the location where they are deployed. As the MVGs are deployed at or upstream of the interaction field, the boundary layer regains its momentum due to the augmented mixing, slowing the onset of flow separation and reducing the size of the separation bubble [138, 139]. It is easy to observe that, each control strategy has its advantages and disadvantages. For example, the favorable and adverse effects of two major passive control techniques, namely, shock control (using porous cavity) and boundary layer control (using MVGs) are shown in Fig. 11. Besides, the key technologies in regulating the SBLI with their benefits and drawbacks are summarized in Table 1.

Benefits and drawbacks associated with major passive control techniques

Essentially, the SBLIs can be regulated by the shedding of the two counter-rotating vortices that originate from the MVGs, as shown in Fig. 12 for Mach 2.5 flow. These mixing-promoting vortices (Fig. 12) are extremely successful at transferring momentum between near-wall and outer boundary layer fluids, resulting in a better boundary layer [140]. Essentially, the low momentum region behind the vortex generator moves upward due to the vortex-induced upwash motion. At the same time, the higher momentum region is entrained towards the near-wall lower momentum region (Fig. 13). In this way, the transfer of momentum is happening between the high-speed freestream and the low-speed near-wall region fluid by the vortex-induced streamwise vortices.

Flow visualization with the microramp of 4 mm height [140]

The momentum variation between the baseline velocity profile and ramp flow [140]

The ramp-type MVGs are superior in the reduction of the recirculation region size since they generate counter-rotating micro-vortices with considerably higher intensity in the near-wall region [141]. Furthermore, highly swept micro ramps are very effective as an SBLI control which induces the pre-compression via weak shock to escape swirling flow in the interaction location [142]. Although larger MVGs have higher control performance than shorter MVGs, they encounter the largest momentum shortage and much more drag. However, the separation delay can be overcome by putting smaller MVGs near areas with unfavorable pressure gradients, which decreases the drag due to viscosity [140,141,142,143]. Taking this into account, Saad et al. (2012) used MVGs to analyze the detailed flow structure in a hypersonic flow on a flat plate (Mach 5) [145]. They concluded that the MVGs induced mixed size vortices help to minimize the upstream interaction region, which in turn suppresses SBLIs. In the recent studies carried out by Kaushik [105] and Gunasekaran et al. [128], the efficacy of the MVGs as well as the porous surface, deployed over a shallow cavity mounted in a Mach 2.2 mixed compression intake, have been investigated experimentally. Since the effects of passive controls in a hypersonic intake may not be the same as in the supersonic intake, therefore, an investigation into the hypersonic intake controlled with passive techniques is essential to identify their effectiveness.

Huang et al. (2020) recently published a systematic review that demonstrated the supremacy of MVGs in the reduction of shock-induced separation for both external and internal flows [146]. Interestingly, the micro-vanes (Fig. 14) were found to be more effective at removing the separation bubble. Micro-vanes, on the other hand, are unstable at high Mach numbers due to their slender design [143], while wedge-shaped micro-vortex (Fig. 15) have a more stable configuration. Furthermore, smaller wedge and vane-shaped MVGs are effective and vigorous in regulating the SBLIs. On the other hand, a triangular-shaped micro-vortex generator can efficiently shed mixed-size vortices, which is desirable from the perspective of improved mixing, according to the literature.

Schlieren flow visualizations, wall pressure measurements, and surface oil-flow visualizations for Vane-type Sub-Boundary Layer Vortex Generator [143]

Schlieren flow visualizations, wall pressure measurements, and surface oil-flow visualizations for Wedge-type Sub-Boundary Layer Vortex Generator [143]

It can be seen from the literature that most of the studies are carried out over the simplified model, i.e., over the flat plate or curved surfaces, where the combined effect of multiple shocks and expansion waves is absent. Since their effects on the boundary layer are non-linear, the investigation of their combined effects is essential for a complete understanding. Considering this in mind, Jana et al. (2020) investigated the effect of a porous cavity in a mixed compression double ramp hypersonic intake (operating at Mach 5.7 and Mach 7.9 flow conditions) [118]. It is observed that the shallow cavity deployed over a porous surface can reduce the shock strength; the shock strength is decreased with the increase in surface porosity. However, the separation bubble size is decreased up to 17% surface porosity, and beyond that limit, the bubble size starts increasing again. They concluded that the higher injection of fluid into the incoming boundary layer promotes the SBLI to such an extent which cannot be reduced by the favorable effect of shock strength reduction. In a subsequent study, Jana et al. (2021) examined the effect of an array of Micro Vortex Generators (MVGs) of varied heights separately deployed at two different locations at Mach 5.7 [144]. The MVGs of height 0.7 mm deployed upstream of the interaction region are the most effective in reducing the separation bubble size. However, shock strength reduction is achieved when the MVGs of 1 mm are placed at the interaction region. They have concluded that, though the passive control is very effective in reducing the SBLI, there is an optimum limit of each passive control technique (porous cavity, MVGs), beyond which the control may promote SBLI.

It is observed from the earlier discussions that SBLI has several detrimental consequences in transonic, supersonic, and hypersonic flow regimes. In a transonic flow situation, the subsonic flow behind the shock wave influences the intensity, shape, and location of the shock and thereby, alters the upstream interaction significantly. In these flow regimes, SBLI can be exclusively observed over transonic aircraft wings and transonic turbine- and compressor-blade cascades. The major control strategies in transonic flow, which can be effectively implemented, are cavity-covered porous surface and surface bump deployed over the surface. In the supersonic regime, SBLI is a common occurrence over supersonic aircraft and in the intake isolator section of ramjet/scramjet engines. There are several passive and active control strategies, such as boundary layer suction and blowing, air-jet vortex generators, micro jets, plasma jet actuators, surface grooves, splitter plate, surface bump, porous cavity, MVGs to improve the interactions, which are discussed in detail. Finally, the hypersonic flow region is characterized by severe interaction of shock wave with boundary layer which leads to very high-temperature rise, higher total pressure loss, huge separation bubble, flow unsteadiness, etc. In these flow situations, surface bumps, porous cavities, splitter plates, and MVGs like passive control techniques are mainly implemented in order to control SBLI in the scramjet inlet and isolator section. Boundary layer bleed is also an important choice in controlling hypersonic interactions. Moreover, the recent investigation on low wavelength wavy patches, placed in the interaction region, has gained prominence in effectively controlling the SBLI in hypersonic flow (Mach 8.5) [147].