Advertisement

Journal of Visualization

, Volume 21, Issue 4, pp 543–556 | Cite as

Visualization of a finite wall-mounted cylinder wake controlled by a horizontal or inclined hole

  • Hiroka Rinoshika
  • Akira Rinoshika
  • Shun Fujimoto
Regular Paper
First Online:
Received:
Revised:
Accepted:
  • 141 Downloads

Abstract

A passive flow control technique, which is to drill a horizontal hole going from the front surface to the rear surface or an inclined hole going from the front surface to the top surface inside a circular cylinder of an aspect ratio H/D = 1, is proposed to control the rear recirculation region and the vortices near the free end surface. Here, both the diameter D and height H of cylinder are 70 mm. PIV measurement was performed with Reynolds number of 8570 in a water tunnel. Furthermore, in order to consider the effect of the hole position of the front surface, the cylinder models having different hole height of front surface from wall were tested. It is found that the rear recirculation zones of the horizontal hole cylinders are smaller than that of the standard cylinder. The Reynolds shear stresses and the turbulent kinetic energy are evidently reduced by the flow issued from the horizontal hole. Meanwhile, the instantaneous large-scale vortical structures of the rear recirculation zone are broken down into several small-scale vortices with decreasing the height of the horizontal hole. Although the rear separation zone of the inclined hole cylinder increases, the recirculation region near the free end surface decreases. The areas of large Reynolds shear stress and high turbulent kinetic energy increase in the rear recirculation zone. However, with increasing the height of the inclined hole, the Reynolds shear stress decreases.

Graphical Abstract

Keywords

Flow controlling hole Flow separation region Passive flow control PIV Low aspect ratio cylinder Vortex Wake 

1 Introduction

The wake flow of a finite-height cylinder exhibits a strongly three-dimensional complex vortex flow, and the aspect ratio of the cylinder affects the wake flow structures. It is different from two-dimensional wake structure (Rinoshika and Zhou 2005, 2009) since the free end of the cylinder and the connection between the cylinder and wall cause three-dimensional flow (Sumner et al. 2004; Pattenden et al. 2005; Wang and Zhou 2009; Gonçalves et al. 2015). Some engineering applications, like buildings, automobile, chimney stacks, offshore structures, heat exchanger and so on, meet such complex flow. Until now researchers have well studied the three-dimensional wake structures around a finite circular cylinder. They have found that wake structures are mainly composed of Kármán vortex originated from the sides of the cylinder, the necklace vortex or horse shoe vortex generated near the ground of the cylinder-wall connection (Tanaka and Murata 1999; Sumner et al. 2004) and a pair of counter-rotating vortices or trailing vortices shaded from the free end (Kawamura et al. 1984a; Johnston and Wilson 1996; Adaramola et al. 2006). Lee (1997) reported that Kármán vortex exhibits a structure like a cell that is reduced with the aspect ratio. However, the wake of a low aspect ratio cylinder is composed of a horse shoe vortex (Krajnović 2011; Rostamy et al. 2012) and tip vortices (Okamoto and Yagita 1973; Kawamura et al. 1984b; Roh and Park 2003), in which mutual vortex shedding (Kárman vortex street) disappears. At very low aspect ratio, “arch-type vortex” structure can be seen observed in the near-wake region (Lee 1997) because vortices formed from the free end surface connect with the vortices generated from the both sides. Sumner (2013) and Porteous et al. (2014) described the detailed reviews on wake flow of finite-height cylinder.

However, little fundamental study on controlling vortex-induced vibration of low aspect ratio cylinder wake is found. It attracts interest in many engineering fields, such as suppressing noise and drag in designing airplane and automobile and increasing drag in offshore structures. Most of the studies on the noise caused by the flow around the cylinder have paid attention to aeolian tone of infinite-height cylinders in consistent crossflow (Ali et al. 2013). Several flow control methods used in “infinite” wake structure, like blowing, suction, surface roughness elements and splitter plate, have been well investigated (Choi et al. 2008; Rashidi et al. 2016). Splitter plates mounted behind the circular cylinder may effectively reduce the transverse oscillations generated by the vortex in the case of undersea cables and oscillating cylinders (Hu and Koterayama 1994). However, less research aims at controlling three-dimensional vortical structures of a low aspect ratio cylinder wake, which becomes the purpose of the present paper. In general, one of the flow control methods is to reduce the occurrence of vortices or the recirculation zone behind the cylinder. Although both passive and active control methods have been carried out to control the flow structures around a bluff body, the passive control method can be applied more easily and control the flow structures without consuming external energy (New et al. 2015; Ozkana et al. 2017). Therefore, the present study focused on the passive flow control of a low aspect ratio cylinder wakes (Sumner et al. 2004; Pattenden et al. 2005). Recently, Rinoshika et al. (2017) proposed a passive control method for a low aspect ratio cylinder, in which a hole is drilled from the free end surface to the side of the rear surface inside cylinder for generating both suction and blowing flows, which effectively reduced the rear separation region of cylinder.

To control the rear recirculation zone of cylinder and the vortices near the free end surface, the present paper suggested another two passive flow control methods: (1) a horizontal hole going from the front side surface to the rear surface is drilled inside cylinder; (2) an inclined hole going from the front side surface to the free end surface is drilled inside cylinder. Then, the hole and no-hole wake structures of the low aspect ratio cylinder mounted on a flat plate are measured by PIV in a circulation water tunnel. Finally, the mean velocity components, mean streamlines, the instantaneous flow structures and vorticity, Reynolds stresses and turbulent kinetic energy are compared between the hole cylinder and standard cylinder.

2 Experimental details

Figure 1 shows a finite standard circular cylinder model of an aspect ratio H/D = 1 (with diameter D and height H of 70 mm) being placed on a flat plate. The streamwise, transverse and spanwise directions are, respectively, indicated by the x, y, and z axes. The following two methods of drilling hole inside cylinder are used to control the flow fields: (1) a hole that has a diameter d = 10 mm (d/D = 0.14) is horizontally drilled from the front side surface to the side of the rear surface (Fig. 1b), referred to as horizontal hole (HH); (2) a hole that has a diameter d = 10 mm (d/D = 0.14) is inclinedly drilled from the front side surface to the free end surface (Fig. 1c), called the inclined hole (IH). At the same time, the variation of the hole position is also considered. Three kinds of the HH models change the hole heights with h = 20, 35 and 50 mm. Three kinds of the IH models, which change the central positions of the hole on the front side surface with h = 20, 35 and 50 mm (h/D = 0.29, 0.5, 0.71) and fix the central position of the hole on the free end surface at L = 30 mm, are used. The cylinder models used in this study are summarized in Table 1. All circular cylinder models are made of acrylic. In order to measure the flow in the hole, all holes inside acrylic cylinder are polished.
Fig. 1

Circular cylinder models

Table 1

Models of horizontal hole and inclined hole cylinders

Hole height on front side (mm)

Horizontal hole

Inclined hole

20

HH20

IH20

35

HH35

IH35

50

HH50

IH50

The water tunnel is adopted and its turbulence intensity is lower than 5%. A constant free stream velocity of U = 0.16 m/s, which is corresponding to Reynolds number of Re (≡ UV/D) = 8570, is performed in this experiment. Polystyrene particles with an averaged diameter of 68 µm were employed as the PIV tracer. A laser light sheet having a thickness of 1.0 mm was adapted to illuminate the flow field behind the cylinder model. The digital images were taken by a high-speed camera at the flame rates of 250 fps (frames per second).

Figure 2 shows the setup of the high-speed PIV measurements in the (x, z)-plane. The measurement area of PIV is about 200 mm × 200 mm with a resolution of 1024 × 1024 pixels. PIV software is used to analyze 7000 digital images (the acquisition duration is 28 s) with the PIV interrogation window size of 24 × 24 pixels and 50% overlap. The time interval of two successive images is set at 4 ms and the shutter speed of each frame is 1 ms. The PIV measurement of the uncertainty of velocity is presumed at ± 1.5% (the reliability is 95%).
Fig. 2

Experimental setup

The profiles of mean streamwise velocity and turbulent intensity in the boundary layer on the flat plate were also measured by PIV. The profiles of time-averaged streamwise velocity (\((\bar u)\)) and turbulence intensity (urms and wrms) normalized by free stream velocity U at the streamwise locations of x = 250 mm (x/D = 3.6) (all data acquired without cylinder) are plotted in Fig. 3. At the location of the cylinder (without a cylinder), the boundary layer thickness δ is 24.6 mm, and this boundary layer provided a thickness-to-diameter or -height ratio of δ/D = 0.35. It is well known fact that the boundary layer thickness influences the flow structure behind a finite-length cylinder (Hearst et al. 2016). A significant upwash flow due to the base vortex can be found at δ/D ≈ 1.02 (Wang et al. 2006), and the boundary layer thickness may make the upwash flow stronger.
Fig. 3

Profiles of mean streamwise velocity and turbulence intensity at the location of x = 250 mm (in the absence of the cylinder) in the flat plate boundary layer

In this study, the hole position of the front side surface for HH20 or IH20 is located in the boundary layer.

3 Results and discussion

3.1 Time-averaged flow structures

The time-averaged streamlines and contours of time-averaged streamwise velocity \((\bar u/U)\) around a standard cylinder in the (x, z)-plane of y/D = 0, computed by the measured instantaneous velocity, are plotted in Fig. 4a. Where the white line indicates \((\bar u/U)\) = 0. It is clearly observed that a mean large vortex shedding from side of the rear surface and edge of the free end surface is located at height of about 0.71D from the flat plate and a mean small vortex on the free end surface is located at about 0.43D from the leading edge. The free end surface of the cylinder generates a strong downwash flow at downstream of the cylinder. The mean streamlines of circular cylinder are very similar to that of a wall-mounted cube (Hearst et al. 2016). The distance from the cylinder to reattachment position on the flat plate at the downstream of separation is usually defined as the recirculation length (Pattenden et al. 2005). In the present study, the boundary streamline of large rear separation region may be identified by a streamline reattached on the flat plate at x/D = 1.32. Pattenden et al. (2005) found a value of x/D = 1.1 for this reattachment position, determined by the surface flow visualization images. This difference may be due to the different Reynolds number and measurement method of Pattenden et al. (2005).
Fig. 4

The time-averaged streamlines and contours of time-averaged streamwise velocity \((\bar u/U)\) in the (x, z)-plane at y/D = 0. Here the white line indicates \((\bar u/U)\) = 0

The time-averaged streamlines and contours of time-averaged streamwise velocity \((\bar u/U)\) around the various HH cylinders having different height of holes are plotted in Fig. 4b–d. Compared with the standard cylinder, the rear separation zone is evidently decreased by supplying flow energy from the flow of the holes with h = 35 and 50 mm (Fig. 4b, c). Although the position of streamline reattached point on the flat plate has no evident difference, it is evident that the height of the vortex becomes lower than that of the standard cylinder due to the effect of the flow issued from the hole. Especially it is obvious for HH35. As shown in Fig. 4d, the length of recirculation zone for HH20 on the flat plate (x/D = 1.6) is longer than that of standard cylinder because the flow issued from the hole blows the bottom of the large vortex. However, the region of negative u-component velocity decreases due to being accelerated by the blowing flow from the hole. The separation region for the HH35 is smallest due to the strongest flow issued from the horizontal hole. The flow velocity of the horizontal hole exhibits the highest value due to the incoming stagnation streamline. In the case of the HH20 and HH50, the velocity of jet flow becomes slightly weaker because of the effect of the free end surface and the flat plate. It may say that the decrease of the separation region could result in the drag reduction of cylinder using the HH cylinders.

Figure 4e–g shows the time-averaged streamlines and contours of time-averaged streamwise velocity \((\bar u/U)\) around the various IH cylinder for the holes of different height. Compared with the standard cylinder, the small separation region on the free end surface is evidently reduced due to the flow energy supplement from blowing flow of the inclined hole. With increasing the hole height h of the IH or decreasing the inclined angle of the IH with the streamwise direction, the recirculation region of the free end surface decreases. It is because the horizontal velocity component of blowing flow from the inclined hole increases. The decrease of inclined angle also results in increase of velocity component in the streamwise direction. It implies that the horizontal velocity component of blowing flow plays an important role for controlling the recirculation region on the free end surface. However, the rear separation lengths of the IH50 and IH35 become larger than that of the standard cylinder. Although the region of negative u-component velocity has no evident variation, larger negative u-component velocity can be observed near the flat plate in the case of IH35. It indicates that the blowing flow from the inclined hole increases the strength of the rear separation region and may result in increase of drag force, which is of an important application in heat exchangers and offshore structures, especially reducing Tsunami energy.

Profiles of time-mean streamwise velocity \((\bar u/U)\) at the locations of x/D = 0.5 and 1.0 are plotted in Fig. 5. As shown in Fig. 5a, \((\bar u/U)\) of various HH cylinders increases behind cylinder at the location of x/D = 0.5, and \((\bar u/U)\) becomes higher near the height of hole due to the blowing flow issued from the hole. Especially, at position of z/D = 0.5, \((\bar u/U)\) of HH35 reaches at 0.2, but that of standard cylinder is about − 0.25. Therefore, the rear recirculation zone of HH35 decreases. It is because the strongest flow issued from the hole controls the rear recirculation zone. The difference between the standard cylinder and HH cylinders becomes smaller at the downstream position of x/D = 1.0 because the influence of the flow issued from the hole gets weaker (Fig. 5b). Compared with the standard cylinder, \((\bar u/U)\) of HH cylinders is slightly larger in the range of 0.6 < z/D < 1.1. However, the velocity profiles of various HH cylinders are relatively self-similar after x/D = 1.0. It implies that the interaction between separation flow and hole flow makes quickly mixing and the mean velocity distribution along z-direction is nonsense to the hole location.
Fig. 5

Profiles of time-averaged streamwise velocity in the (x, z)-plane

By comparing the IH cylinders and standard cylinder in Fig. 5c, d, the profiles are only marginally different regardless of where the hole. Although the inclined hole influences the flow near the free end surface largely, it has less effect on the rear separation region. In the case of the IH cylinder at x/D = 0.5, as shown in Fig. 5c, \((\bar u/U)\) of IH35 increases in the opposite to the streamwise direction and is slightly larger than that of the standard cylinder in the range of 0 < z/D < 0.4 near the flat plate. Especially, at the position of z/D = 0.05, \((\bar u/U)\) of IH35 shows − 0.5, but that of standard cylinder is about − 0.4. It is because the strong upwash made by the inclined hole influences the wake flow, which also results in increase of the rear separation region. However, in the range of 0.4 < z/D < 0.65, \((\bar u/U)\) of various IH cylinders becomes slightly larger than that of the standard cylinder and is smaller than that of various HH cylinders. At the position of x/D = 1.0, as indicated in Fig. 5d, the difference between the IH cylinders and the standard cylinder becomes smaller because the effect of the blowing flow near the free end surface gets weaker.

3.2 Reynolds stresses and turbulent kinetic energy

Figure 6a indicates the distribution of the normalized Reynolds shear stress \((\overline {u^{\prime} w^{\prime}} /U^2 )\) for the standard cylinder wake in the (x, z)-plane. It is evidently observed that a large area of negative Reynolds shear stress and a small area of positive Reynolds shear stress appear near the free end and in the near-wake, respectively. The maximum negative magnitude is − 0.04 around x/D = 1.2 and z/D = 0.7.
Fig. 6

Contours of Reynolds shear stress \((\overline {u^{\prime} w^{\prime}} /U^2 )\) in the (x, z)-plane at y/D = 0

Figure 6b–d plots the contours of the normalized Reynolds shear stress \((\overline {u^{\prime} w^{\prime}} /U^2 )\) around the various HH cylinders. The positive Reynolds shear stress is near the hole and the negative Reynolds shear stress distributes near the boundary of the recirculation region. The maximum magnitude of HH35 and HH50 is about 0.01, and that of HH20 is about 0.005. As decreasing the height of the hole, the area of the positive Reynolds shear stress increases. It results from the velocity of the jet flow and jet position acting on the separation area behind the cylinder. Compared with the standard cylinder, the maximum negative magnitude of Reynolds shear stress behind the cylinder decreases and the region of the negative Reynolds shear stress reduces. It suggests that the flow issued from the hole may suppress the Reynolds shear stress.

The contours of the normalized Reynolds shear stress \((\overline {u^{\prime} w^{\prime}} /U^2 )\) around the IH cylinder of different height holes are plotted in Fig. 6e–g. In the case of the IH50 and IH35, the positive Reynolds shear stresses are observed near the outlet of the hole (on the free end surface). However, the negative Reynolds shear stresses also appear near the hole of the IH20. The flow velocity issued from the hole and the angle of the blowing flow on the free end may control the Reynolds shear stress near the free end. Compared with the standard cylinder, the Reynolds shear stress decreases near the free end, but the area of the Reynolds shear stress increases in the rear separation region. As the height of the hole increases, the Reynolds shear stress becomes weaker. It is because the inclined angle or position of the hole determines the vertical velocity component of the blowing flow and effects or changes the flow structures around the IH cylinder. Therefore, the recirculation zone and the Reynolds shear stresses are controlled.

The profiles of Reynolds shear stress \((\overline {u^{\prime} w^{\prime}} /U^2 )\) in the (x, z)-plane at the locations of x/D = 0.5 and 1.0 are presented in Fig. 7. At the position of x/D = 0.5, as shown in Fig. 7a, \((\overline {u^{\prime} w^{\prime}} /U^2 )\) of HH cylinder is larger than that of the standard cylinder in the range of 0.4 < z/D < 0.8 because of the effect of the blowing flow. But in the range of z/D < 0.4, \((\overline {u^{\prime} w^{\prime}} /U^2 )\) of HH cylinder exhibits negative values. At downstream of x/D = 1.0 (Fig. 7b), however, \((\overline {u^{\prime} w^{\prime}} /U^2 )\) of the standard cylinder shows the largest negative magnitude near the height of z/D = 0.65 due to the high velocity gradient appearing near the boundary of the separation region or large-scale vortex.
Fig. 7

Profiles of Reynolds shear stress \((\overline {u^{\prime} w^{\prime}} /U^2 )\) in the (x, z)-plane

On the other hand, \((\overline {u^{\prime} w^{\prime}} /U^2 )\) of HH cylinders are decreased because the flow issued from the hole generates positive Reynolds shear stress in the separation zone and reduces negative magnitude of the Reynolds shear stress. Focusing on the HH35, \((\overline {u^{\prime} w^{\prime}} /U^2 )\) shows the smallest negative magnitudes among hole and non-hole cylinders since the strong flow issued from the horizontal hole destroys the large-scale vortices and generates several relatively small-scale vortices (to be described in Sect. 3.3), which may result in the drag reduction. It also indicates that the flow issued from the horizontal hole may reduce the Reynolds shear stress in the rear separation zone.

Figure 7c shows profiles of Reynolds shear stress \((\overline {u^{\prime} w^{\prime}} /U^2 )\) around the standard cylinder and IH cylinders at the locations of x/D = 0.5. Compared with the standard cylinder, it is evident that the Reynolds shear stress of IH cylinders increases in the range of 0.1 < z/D < 0.6. It comes from the effect of the blowing flow of the hole on the rear recirculation zone. The Reynolds shear stress of IH35 shows the highest negative value among the standard cylinder and the IH cylinders near the height of z/D = 0.5. It implies that the strong blowing flow increases the Reynolds shear stress behind the cylinder. At downstream of x/D = 1.0 (Fig. 7d), however, the Reynolds shear stresses of various IH cylinders become smaller than that of the standard cylinder in the range of 0.5 < z/D < 0.8. It is because that the flow issued from the inclined hole accelerates the main flow near the top surface (Fig. 4) and suppresses the Reynolds shear stress. At the position of z/D = 0.65, the Reynolds shear stress of IH35 exhibits the smallest negative magnitude of all the models because the inclined hole flow causes high flow velocity (Fig. 5c).

The turbulent features and vortical structures are closely related to the turbulent kinetic energy. In the present study, a 2D turbulent kinetic energy TKE analogy, used by Lim and Lee (2003) and Oruc (2012), is computed from the Reynolds normal stresses (\(\overline {u^\prime } /U^2 ,\overline {w^{\prime 2} } /U^2\)) in the (x, z)-plane by
$${\text{TKE}} = 0.75({\overline {u^\prime } /U^2 + \overline {w^{\prime 2} } /U^2 })$$
Figure 8 shows the distributions of the turbulent kinetic energy (TKE) for the with and without hole cylinders in the (x, z)-plane. As indicated in Fig. 8a–d, it is evident that the maximum magnitude of TKE in the rear separation region for the HH cylinders decreases compared to the standard cylinder (TKE = 0.07). The maximum magnitudes of TKE behind the HH20 and HH50 are about 0.05 and that of HH35 exhibits 0.06. It implies that the HH controls the flow of the rear recirculation zone and reduces the turbulent kinetic energy. The decrease of TKE may lead to the drag reduction (Lim and Lee 2003). The TKE of HH35 shows the highest magnitude among three HH cylinders in the rear separation zone. On the other hand, compared with HH20, HH50 and standard cylinders, TKE of HH35 also appears high magnitude near the free end surface. As indicated in Fig. 4, the flow velocity going through the hole of HH35 exhibits the highest value since the hole of HH35 is best aligned with the incoming stagnation streamline and the streamlines are not deflected substantially before entering the hole. Therefore, the flow issued from the hole of HH35 destroys the large-scale structures into several relatively small-scale vortices (to be described in Sect. 3.3). They induce the downwash and upwash unsteady flows and cause large velocity fluctuation and high TKE near the free end surface.
Fig. 8

Contours of TKE in the (x, z)-plane at y/D = 0

As compared to the standard cylinder, as shown in Fig. 8e–g, the areas of high value of TKE around the IH cylinders increase and appear in the rear separation region. The maximum magnitudes of TKE for IH20 and IH35 are 0.06, and that of IH50 exhibits 0.05. The inclined direction and flow velocity of hole near the free end largely affect the flow structures around the IH cylinders and increase the high TKE region in the rear separation zone. This may result in increase of the drag force acted on the model. Focusing on IH50 in Fig. 8g, the large inclined angle of hole generates large z-component velocity issued from the hole, which produces strong interaction between the main flow and hole flow, and also results in a high TKE region near the free end surface.

3.3 Instantaneous flow fields

To compare the flow structures among with and without hole cylinders, the instantaneous normalized vorticity contours of ωyD/U and corresponding streamlines in the (x, z)-plane are indicated in Fig. 9. The color mappings show the vorticity values and red and blue indicate the highest and lowest concentrations, respectively. Four vortical structures having different size and strength are observed in the rear separation zone and near the free end surface of the standard cylinder (Fig. 9a). In the case of the various HH cylinders, as shown in Fig. 9b–d, several relative small vortices are observed in the rear separation zone due to the blowing flow issued from the horizontal hole. The rear separation region of the HH50 (Fig. 9b) exhibits smaller vortices than the other two HH cylinders since the height of hole locates at the large vortex centre of the standard cylinder (Fig. 9a) approximately and the large vortex is broken down. As decreasing the height of hole, the size of vortex increases. A strong vorticity is found in the wake of the HH35 (Fig. 9c). It is because that the flow issued from the horizontal hole locating at the half height of the cylinder has a high velocity. The hole flow of HH20 becomes weak and locates at the large vortex edge of the non-controlling cylinder (Fig. 9a) because of the effect of the boundary layer of flat plate; therefore, the sizes of vortices increase slightly.
Fig. 9

Instantaneous streamlines and corresponding vorticity \((\omega_y D/U)\) contours in the (x, z)-plane at y/D = 0

The instantaneous normalized vorticity contours and corresponding streamlines of the IH cylinders are displayed in Fig. 9e–g. It is evident that the rear separation region of the IH50 (Fig. 9e) appears a larger vortex, which is larger than the other two IH cylinders. Since the inclined angle of hole is small, the flow issued from the hole accelerates the fluid near the free end surface and forms a large rear separation zone. As increasing the inclined angle of hole, the size of the rear separation region decreases and several relative small vortices appear. It is due to the component of horizontal velocity blowing from the inclined hole decreases. As compared with the standard and HH cylinders, the rear recirculation zones of the IH cylinder increase. However, the separation zone of each IH cylinder on the free end surface becomes smaller.

It is well known that a dominant feature of the local flow field around a low aspect ratio finite cylinder is mainly characterized by an arch vortex in the rear recirculation region (Lee 1997; Pattenden et al. 2005). In the case of the HH cylinders, the flow issued from the horizontal hole flows through the center of arch vortex and breaks down the arch vortex rapidly. Therefore, the rear recirculation zone decreases. The flow of horizontal hole with h = 50 mm (HH50) controls the head of the arch vortex originated from the free end surface, and the flow of horizontal hole with h = 20 mm (HH20) controls the two foot of arch vortex near the flat plate. However, the various IH cylinders only control the arch vortex in recirculation zone near the free end surface (Pattenden et al. 2005) and accelerate the flow of the head of the arch vortex. It may result in the increase of the rear recirculation region.

4 Conclusions

The flow structures of the various hole cylinder wakes are experimentally studied. Making a comparison with the standard cylinder, the main conclusions are obtained as follows:
  1. 1.

    The rear separation region is evidently reduced by the flow issued from the various HH cylinders. However, the rear separation lengths of the various IH cylinders increase inversely.

     
  2. 2.

    The separation region of the IH cylinders near the free end surface is decreased compared to the standard and HH cylinders.

     
  3. 3.

    The TKE and negative value of Reynolds shear stress behind the HH cylinder are suppressed. As the hole height of the HH cylinder decreases, the area of the positive Reynolds shear stress increases.

     
  4. 4.

    The Reynolds shear stress of the IH cylinders near the free end decreases, but the areas of large Reynolds shear stress and high TKE increase in the rear recirculation zone. The Reynolds shear stress decreases as increasing the hole height of the IH cylinder.

     
  5. 5.

    Several relative small vortices are observed in the instantaneous flow structures of the rear separation region in the HH cylinder wakes. As decreasing the hole height of the HH cylinder, the size of vortex increases.

     
  6. 6.

    As increasing the inclined angle of IH cylinder, the size of the instantaneous rear separation region decreases and several relative small vortices appear.

     

Notes

Acknowledgements

The second author (AR) wishes to acknowledge support given to him by Grant-in-Aid for Scientific Research (C) (No. 16K06067) from the Japanese Society for the Promotion of Science and Natural Science Foundation of China (Grant Nos. 11721202 and 11772035).

References

  1. Adaramola MS, Akinlade OJ, Sumner D, Bergstrom DJ, Schenstead AJ (2006) Turbulent wake of a finite circular cylinder of small aspect ratio. J Fluids Struct 22:919–928CrossRefGoogle Scholar
  2. Ali M, Doolan C, Wheatley V (2013) Aeolian tones generated by a square cylinder with a detached flat plate. AIAA J 51(2):291–301CrossRefGoogle Scholar
  3. Choi H, Jeon W, Kim J (2008) Control of flow over a bluff body. Annu Rev Fluid Mech 40:113–139MathSciNetCrossRefzbMATHGoogle Scholar
  4. Gonçalves RT, Franzini GR, Rosetti GF, Meneghini JR, Fujarra ALC (2015) Flow around circular cylinders with very low aspect ratio. J Fluids Struct 54:122–141CrossRefGoogle Scholar
  5. Hearst RJ, Guillaume G, Bharath G (2016) Effect of turbulence on the wake of a wall-mounted cube. J Fluid Mech 804:513–530MathSciNetCrossRefGoogle Scholar
  6. Hu C, Koterayama W (1994) Numerical study on a two-dimensional circular cylinder with a rigid and an elastic splitter plate in uniform flow. Int J Offshore Polar Eng 4:193–199Google Scholar
  7. Johnston CR, Wilson DJ (1996) A vortex pair model for plume downwash into stack wakes. Atmos Environ 31:13–20CrossRefGoogle Scholar
  8. Kawamura T, Hiwada M, Hibino T, Mabuchi T, Kumada M (1984a) Flow around a finite circular cylinder on a flat plate. Bull JSME 27:2142–2150CrossRefGoogle Scholar
  9. Kawamura T, Hiwada M, Hibino T, Mabuchi T, Kumada M (1984b) Flow around a finite circular on flat plate: in the case of cylinder length larger than turbulent boundary layer thickness. Trans JSME 50:332–341 (in Japanese) CrossRefGoogle Scholar
  10. Krajnović S (2011) Flow around a tall finite cylinder explored by large eddy simulation. J Fluid Mech 676:294–317MathSciNetCrossRefzbMATHGoogle Scholar
  11. Lee LW (1997) Wake structure behind a circular cylinder with a free end. In: Proceedings of the Heat Transfer and Fluid Mechanics Institute, pp 241–251Google Scholar
  12. Lim HC, Lee SJ (2003) PIV measurements of near wake behind a U-grooved cylinder. J Fluids Struct 18:119–130CrossRefGoogle Scholar
  13. New TH, Shi SG, Liu YZ (2015) On the flow behavior of confined finite-length wavy cylinders. J Fluids Struct 54:281–296CrossRefGoogle Scholar
  14. Okamoto T, Yagita M (1973) The experimental investigation on the flow past a circular cylinder of finite length placed normal to the plane surface in a uniform stream. Bull JSME 16:805–814CrossRefGoogle Scholar
  15. Oruc V (2012) Passive control of flow structures around a circular cylinder by using screen. J Fluids Struct 33:229–242CrossRefGoogle Scholar
  16. Ozkana GM, Firatb E, Akilli H (2017) Passive flow control in the near wake of a circular cylinder using attached permeable and inclined short plates. Ocean Eng 134:35–49CrossRefGoogle Scholar
  17. Pattenden RJ, Turnock SR, Zhang X (2005) Measurements of the flow over a low-aspect-ratio cylinder mounted on a ground plane. Exp Fluids 39:10–21CrossRefGoogle Scholar
  18. Porteous R, Moreau DJ, Doolan CJ (2014) A review of flow-induced noise from finite wall-mounted cylinders. J Fluids Struct 51:240–254CrossRefGoogle Scholar
  19. Rashidi S, Hayatdavoodi M, Esfahani JA (2016) Vortex shedding suppression and wake control: a review. Ocean Eng 126:57–80CrossRefGoogle Scholar
  20. Rinoshika H, Rinoshika A, Fujimoto S (2017) Passive control on flow structure around a wall-mounted low aspect ratio circular cylinder by using an inclined hole. J Fluid Sci Tech 12(1):JFST0006CrossRefGoogle Scholar
  21. Rinoshika A, Zhou Y (2005) Orthogonal wavelet multi-resolution analysis of a turbulent cylinder wake. J Fluid Mech 524:229–248CrossRefzbMATHGoogle Scholar
  22. Rinoshika A, Zhou Y (2009) Reynolds number effects on wavelet components of self-preserving turbulent structures. Phys Rev E 79(046322):1–11Google Scholar
  23. Roh SC, Park SO (2003) Vortical flow over the free end surface of a finite circular cylinder mounted on a flat plate. Exp Fluids 34:63–67CrossRefGoogle Scholar
  24. Rostamy N, Sumner D, Bergstrom DJ, Bugg JD (2012) Local flow field of a surface-mounted finite circular cylinder. J Fluids Struct 34:105–122CrossRefGoogle Scholar
  25. Sumner D (2013) Flow above the free end of a surface-mounted finite-height circular cylinder: a review. J Fluids Struct 43:41–63CrossRefGoogle Scholar
  26. Sumner D, Heseltine JL, Dansereau OJP (2004) Wake structure of a finite circular cylinder of small aspect ratio. Exp Fluids 37:720–730CrossRefGoogle Scholar
  27. Tanaka S, Murata S (1999) An investigation of the wake structure and aerodynamic characteristics of a finite circular cylinder. JSME Int J B 42:178–187CrossRefGoogle Scholar
  28. Wang HF, Zhou Y (2009) The finite-length square cylinder near wake. J Fluid Mech 638:453–490CrossRefzbMATHGoogle Scholar
  29. Wang H, Zhou Y, Chan CK, Lam KS (2006) Effect of initial conditions on interaction between boundary layer and a wall-mounted finite-length-cylinder wake. Phys Fluids 18:065106CrossRefGoogle Scholar

Copyright information

© The Visualization Society of Japan 2018

Authors and Affiliations

  • Hiroka Rinoshika
    • 1
  • Akira Rinoshika
    • 1
    • 2
  • Shun Fujimoto
    • 1
  1. 1.Department of Mechanical Systems EngineeringYamagata UniversityYonezawa-shiJapan
  2. 2.School of Aeronautic Science and EngineeringBeijing University of Aeronautics and AstronauticsBeijingChina

Personalised recommendations