Experimental study of the development of natural perturbations of a swept wing boundary layer under attached flow and control of them with distributed suction

. This work is of a fundamental nature and is devoted to the study of the influence of distributed suction through a finely perforated hydrodynamically smooth surface on natural and acoustically enhanced boundary layer perturbations under conditions of predominantly non-separated flow. The issue of the minimum necessary suction for suppression of perturbations is considered. The results obtained give grounds to assert that the efficiency of distributed suction in terms of suppressing the growth of disturbances is due not only to the reattachment of a separated flow, but, in general, to a change in the stability parameters of the flow.


Introduction
Boundary layer suction is an effective way to control the laminar-turbulent transition.The effect of gas suction on the flow manifests itself in the form of a decrease in the thickness of the boundary layer, and, consequently, a change in its susceptibility to disturbances.Distributed suction is the most efficient in terms of flow laminarization.For example, in [1], the influence of distributed suction through a transverse slot on the development of nonlinear disturbances in the boundary layer is considered.In the same work, conditions were determined under which gas suction as a means of flow laminarization is effective.With the development of electron beam processing technologies, it became possible to obtain finely perforated metal surfaces on an industrial scale.This is of particular interest to aircraft engineers, since such surfaces can be hydrodynamically smooth, i.e. the roughness is so small that the flow does not react to them.Distributed suction through such a surface can be part of the CLFC (combined laminar flow control) of the aircraft.Such systems were implemented on the Boeing 787-9 experimental aircraft (Fig. 1) in the area of the tail unit and the leading edge of the wing.The system uses the natural differential pressure distribution along the flow profile to suction the boundary layer when moving at cruising speed and blow it out from the surface during takeoff -landing to protect against insects.
In [3], the effect of distributed suction through a hydrodynamically smooth permeable surface on the development of natural and acoustically enhanced disturbances in the boundary layer of a sliding wing under separation conditions was studied.The authors managed to achieve the suppression of both natural and enhanced with the help of acoustic exposure.It is known that, under conditions of separation, disturbances actively grow, and the elimination of separation as a result of the effect of distributed suction on the mean flow could make the main contribution to the result obtained.This work is of a fundamental nature and is devoted to the study of the impact of distributed suction through a finely perforated hydrodynamically smooth surface on intrinsic and acoustically amplified boundary layer perturbations under conditions of predominantly nonseparated flow.The issue of the minimum necessary suction for suppression of perturbations is being solved.

Experimental setup
The experiment was carried out in the test section of the T-324 low-turbulence wind tunnel at ITAM SB RAS.The oncoming flow velocity was  ∞ = 15 m/s.The degree of free flow turbulence did not exceed Tu = 0.04%  ∞ .The model (Fig. 2) was a section of a sliding wing with a chord c = 807 mm and span z = 950 mm.At a distance of 0.5c along the chord from the leading edge, a permeable hydrodynamically smooth insert was mounted flush with the surface, through which distributed suction was carried out.The surface permeability was 17% and was achieved due to round pores with a diameter of 0.17 mm located in a checkerboard pattern.The distributed suction power was varied and, due to the use of a finely perforated surface, was determined experimentally by directly measuring the gas velocity over the surface in the absence of the main flow.After averaging the obtained data, the dimensionless suction coefficient was calculated by the formula

E3S Web of
The leading edge of the surface is located at the coordinate x = 416 mm, which corresponds to the region of an unfavorable pressure gradient (Fig. 3).It is assumed that perturbations increase in this region.The work also uses the amplification of natural disturbances by a sound wave, which has the main frequency of the disturbance.The generation and amplification of disturbances in the boundary layer on a swept wing is a proven method used, for example, in [4].

Results
Figure 4 shows the average velocity profiles at several points located on a straight line along the flow direction.The x = 416 coordinate corresponds to the point located in front of the suction area.Globally, the flow is not separated, however, at the point x = 556, a small local separation bubble is observed, followed by flow reattachment at x = 616.The velocity profile at the end of the measurement region x = 646 has the form of a turbulent one.The profiles of velocity pulsations at the corresponding points (Fig. 5) demonstrate the growth of natural disturbances, up to turbulence at the end of the measurement region.The spectral analysis of the perturbation at the point x = 370 shows (Fig. 6 (a)) that the perturbation consists of high-frequency pulsations, the fundamental frequency of which is 200 Hz.At this frequency, an acoustic field with an intensity of 90 dB was used to amplify the natural oscillations.The result was an increase in self-disturbances at the linear stage of development with the appearance of harmonics at multiple frequencies in Fig. 6 (b).The profiles of the mean velocity after the activation of the acoustic field (Fig. 7) show that the intensification of the disturbance affected the mean flow.An early increase in the amplitude of disturbances and turbulence contributed to the elimination of the local separation bubble at x = 556, compared to natural disturbances.The flow is fully attached.The early increase in the velocity fluctuation amplitude can be illustrated by comparing Figs. 8 and 9. From the graphs, it is clear that in the case of natural development, an active increase in the amplitude of velocity pulsations begins from x = 566 mm (Fig. 8 (a)).However,   = 0.029 is already enough to reduce the velocity fluctuation amplitude at the end of the measurement area by a factor of 4, and to delay the moment of active growth of the perturbation velocity fluctuation amplitude by 60 mm downstream as shown in Fig. 8(b).When using an external acoustic field at the fundamental frequency of the perturbation, the increase in the amplitude of velocity pulsations starts from x = 486 mm (Fig. 9 (a)).Empirically, it was found that for stable suppression of the development of disturbances in the measurement area, it is sufficient to use a distributed suction with a dimensionless coefficient   = 0.043 (Fig. 9 (b)).Fig. 10 shows the effect of distributed suction on average velocity profiles.The activation of suction reduces the thickness of the boundary layer, and the velocity profiles become more stable.At   = 0.043, the perturbation is not completely eliminated, as can be seen from the graph of velocity pulsations at the end of the measurement area (fig.11  (a)).However, there were no trends in the increase in the intensity of disturbances in the measurement region.The perturbation spectrogram in fig.11(b) also shows the residual presence of disturbances caused by the 200 Hz acoustic field.

Conclusion
In the work, the influence of distributed suction through a hydrodynamically smooth permeable surface on the development of natural and acoustically enhanced disturbances in the boundary layer of a swept wing was studied.Under these conditions, natural perturbations actively grow and pass through all stages of evolution up to a turbulent spot at the end of the measurement region.The amplitude of velocity fluctuations at the end of the measurement area was 12.2%  ∞ .The use of a relatively small suction with a coefficient   = 0.029 made it possible to reduce the amplitude of velocity pulsations at the end of the measurement area by 4 times and shift the moment of disturbance growth by 60 mm downstream.Amplification of the disturbance with the help of an acoustic field with a frequency of 200 Hz led to an early active growth of the disturbance (starting from x = 300 mm) and elimination of the local separation.The peak amplitude of the perturbation in this case was about 18%  ∞ , the final one was also about 12.2%  ∞ .Measurements of the amplitude of velocity pulsations at the end of the measurement area make it clear that under these conditions, for stable suppression of the growth of a disturbance, it is sufficient to use suction with a coefficient   = 0.043 With such parameters, the initial and final amplitudes of the disturbances do not differ and are about 0.25%  ∞ .Thus, there is a decrease in the amplitude of velocity pulsations by 48 times.
The results obtained give grounds to state that the efficiency of distributed suction in terms of suppressing the growth of disturbances is due not only to the reattachment of a separated flow, as could be assumed from the results obtained in [3], but in general, to a change in the flow stability parameters.

Fig. 1 .
Fig. 1.The use of distributed suction in the layout of the Boeing 787-9.Diagram of the device [2] (a), photo (b).

Fig. 3 .
Fig. 3. Average velocity distribution outside the boundary layer along the wing chord at z = 20.

Fig. 6 .
Fig. 6.Disturbance spectrogram at the point x = 546 mm, in the case of natural disturbances (a) and amplified by an acoustic signal with a frequency of 200 Hz and an intensity of 90 dB (b).

Fig. 7 .
Fig. 7. Average velocity profiles along the flow direction under the influence of an acoustic field with a frequency of 200 Hz.

Fig. 8 .
Fig. 8. Evolution of the maximum amplitude of velocity pulsations along the flow in the case of disturbances developing naturally (a), when suction is activated   = 0.029 (b).

Fig. 9 .
Fig. 9. Evolution of the maximum amplitude of velocity pulsations along the flow in the case of disturbances amplified by acoustics (a), when suction is activated   = 0.043 (b).

Fig. 10 .
Fig. 10.Velocity pulsation profile (a) and spectral pattern (b) under the influence of an acoustic field with a frequency of 200 Hz and suction activation   = 0.043 at the end of the measurement area x = 460.