Experimental investigation of blade-shaped riblets for drag reduction on UAV applications

This study presents an experimental investigation of blade-shaped riblets for drag reduction in unmanned aerial vehicle (UAV) applications. UAVs have gained significant attention since they can perform various missions, including surveillance, reconnaissance, and package delivery. However, their aerodynamic performance, specifically the high drag associated with their exposed surfaces, remains a key challenge for enhancing their efficiency and extending their flight endurance. To address this issue, riblet geometries are proposed as a potential solution, which can reduce the turbulent skin friction drag by up to 8%. The experimental investigation involves wind tunnel testing of blade-shaped riblets, with various spacing-to-height (s/h) ratios and constant groove cross-sectional area (Ag). The riblets are designed for application on the wing, empennage, and fuselage surfaces of a UAV. The investigations are performed on a flat plate for various flow conditions, including different freestream velocities, to evaluate the drag reduction effectiveness of the riblet configuration. The drag force is measured using a force balance system and flow visualization techniques are employed to assess the position where the boundary layer has transitioned to fully turbulent. The results demonstrate the drag-reducing effect of blade-shaped and trapezoidal riblets and the different performances observed for the various s/h ratios. The cases with s/h=1 result in the smallest drag coefficients, while the cases with s/h=2 have significantly increased drag values, compared to the smooth flat plate, due to the increased wetted surface area. These findings highlight the potential of riblets as an effective drag-reduction technique for UAV applications, enabling increased endurance and/or enhanced payload capacity.


Introduction
Due to the wide range of operational capabilities, as well as the inherent advantages compared to conventional manned alternatives, the use of unmanned aerial vehicles (UAVs) has been steadily increasing around the globe.The current size of the UAV market is estimated at 15.4 billion USD and with a compound annual growth rate of about 14%, it is forecast to almost double and reach about 29.66 billion USD by 2028 [1].Although a recent design trend exists for unconventional UAV configurations (e.g.flying wing, blended-wing-body), which can have a significant performance improvement, including reduced drag and increased payload [2], most fixed-wing UAVs are still based on the traditional tube-and-wing approach.It is hence evident, the importance of investigating alternative technologies and strategies that can provide performance enhancement for current designs.The use of passive flow control methods is especially interesting since they can be relatively costefficient and simple to implement.Among the most prominent such methods are the natural laminar flow airfoils, wiglets, and riblets, each at a different technology readiness level (TRL) [2].
The relative application simplicity of riblets on external surfaces of aircraft and UAVs makes them an ideal flow control method for both novel and traditional already-designed configurations [3,4].In recent years they have gained a significant new interest [5], due to their passive ability to reduce the friction drag component of turbulent boundary layer flows.As a flow control method, the riblets are small streamwise-aligned micro-grooved geometries (in the order of 100-1,000μm) that affect the near wall region of the boundary layer with three distinct mechanisms: (a) by elevating upwards the crossflow motion in their crests and displacing streamwise vortices and streaks away from the wall, (b) by weakening the near-wall turbulence regeneration cycle and (c) by dampening the spanwise flow fluctuations [6].It is though possible to also have a negative effect if poorly designed or placed, as they can amplify the Tollmien-Schlichting waves and promote premature boundary layer transition, hence increasing the friction drag.
Extensive research on riblets was initially performed in the late 1980s and early 1990s with experiments on flat plates, using mainly V-shaped riblets.These initial studies have shown that a possible 8-10% maximum drag reduction is possible [7].Other studies focused on different riblet geometries, such as the U-, Blade-and Trapezoidal-shaped ones, which also can result in noticeable drag reduction if designed correctly.Among these traditional shapes, the Blade riblets have been shown to provide optimal drag-reducing performance, though they are prone to damage and are more difficult to fabricate compared to V-shaped riblets [8].The correlation between the Reynolds number and the drag-reducing effect of riblets is performed through the use of non-dimensional wall units  + =  •  * /, of various length parameters .The first such parameter  + , was proposed by Walsh and Linderman [9], corresponding to the non-dimensional riblet groove spacing.It was shown that the optimal riblet performance was achieved for a value close to  + ≈ 15, though this value can vary for different riblet geometries and sizes.A much better correlation for optimal performance can be achieved through the non-dimensional square root of the riblet groove cross-section   + , as shown by Garcia-Mayoral and Jimenez [10], using the experiments performed by Bechert et al. [11] with various trapezoidal, blade, triangular and U-shaped riblets.The optimal   + is found to be equal to 10.5, with a variation of less than 10% between different geometries [10].In the present study, two different sets of riblets are experimentally investigated on a wind tunnel, for zero-pressure gradient flat plate conditions, in the framework of the RADAERO project.The project aims to investigate the properties of aerodynamic drag reducing (riblets) and radar cross section signature reducing (MMA) materials for UAV applications with CFRP serving as the main base material.The first riblet set corresponds to three different blade-shaped geometries, designed and fabricated within the RADERO project framework, with significant variation in the groove spacing-toheight (/ℎ) ratio.The second set is comprised of commercial trapezoidal-shaped riblets, produced by the 3M company, and are of smaller dimensions compared to the first set, serving as reference models.The riblet performance is evaluated in different freestream flow conditions, which can correspond to different flight phases of the UAV.Finally, the experimental results of the first riblet set are compared against previously performed LES simulations [12].

Experimental facilities and equipment
The experiments performed in the present work were conducted in a closed-circuit, low-speed wind tunnel facility, at the Laboratory of Fluid Mechanics and Turbomachinery (LFMT) of the Aristotle University of Thessaloniki.The experimental facility has a test section of 1810mm × 600mm × 600mm (length × width × height), depicted in Figure 2, the airflow is accelerated by the 45kW, 1m diameter rotor that can drive the air at freestream velocities between 0.2m/s and 50m/s, with a frequency inverter controlling the wind tunnel blower speed.To achieve good flow quality in the test section, the wind tunnel chamber contains two anti-turbulence screens, which ensure a uniform flow.The turbulence intensity within the test section was also measured and previously reported to be less than 1% (~0.7%).The wind tunnel can accommodate various test models, including airfoils/wings and flat plates.The drag and lift forces are measured with an external force balance.The force balance is in-house developed, consists of an aluminum frame, and linear traversers that can adjust the angle of attack of the examined specimen with high accuracy (less than 0.1 ο ), and is equipped with four Omega LCEB loadcells used for the force measurement.Concurrently with the force measurements, all other crucial flow parameters (i.e.freestream velocity, pressure, and temperature) are monitored and logged.The flow velocity is measured using a pitot-static tube attached to a piezoelectric differential manometer (TSI PVM620), while the temperature is measured with the use of a K-type thermocouple (CENTER 308).The differential manometer and the thermocouple have been calibrated and certified, using highly accurate standard instruments.Additionally, all loadcell measurements were readjusted with the use of M1 class calibrated test weights, and their corresponding fitting curves were drawn.All measurements made are adjusted using each instrument's correction factor, obtained during the calibration/certification procedure.The uncertainty of each measuring component is presented in Table 1.For every test specimen, a range of freestream velocities is examined between 5m/s and 15m/s, at 2.5m/s intervals, along with the maximum velocity of 16.4m/s.For every freestream velocity, the measurements are repeated 10 times, each measurement's drag coefficient is calculated using that measurement's velocity and temperature, and finally, the test's drag coefficient is calculated as the average value obtained from the 10 repetitions (Equation 1). ) 10 ⁄ (1) Since the wind tunnel facility operates at low freestream velocities, the transition of the flat plate boundary layer from laminar to turbulent was enhanced, aiming to achieve a fully turbulent flow over the region where the riblet specimens are measured.This was performed with a dedicated roughness strip applied near the leading edge of the flat plate, with its roughness height being determined in advance, using the simplified method of Braslow and Knox [13].
Figure 2. The wind tunnel facility, with installed models in the test section.

Validation of experimental setup.
The force balance's measurement capabilities have been validated with two different validation cases, each employing a different methodology, to ensure maximum validity.Firstly, a blade segment from a small horizontal wind turbine was investigated [14], and the lift force was measured using the in-house developed force balance, as well as surface pressure measurements (Figure 3).For the latter, a blade model is equipped with pressure taps, which are connected to a pressure measuring instrument (NetScanner™ System Intelligent Pressure Scanner module Model 9016), and the lift coefficient is calculated by integrating the pressure values along the airfoil's profile.The flow conditions examined correspond to a Reynolds number of 200,000, based on the airfoil's chord, with a freestream velocity of 9.74m/s and for a range of AoA between -2° and 10° (at 2° intervals).Secondly, the lift and drag forces of a NACA 0012 model were measured using the in-house developed force balance and were compared with published experimental data (Figure 4) from NASA [15].The flow conditions are selected to achieve a Reynolds number of 180,000, based on the airfoil's chord, identical to that of the NASA experiments.
The experimental results of the wind turbine blade, presented in Figure 3, are deemed very sufficient, ensuring that the force balance is operating correctly and accurately.The minor discrepancy in the results of the two methods can be attributed to the manufacturing differences between the two blade models, as well as the lack of measurement capability of the skin friction drag component interference in lift production for the pressure taps case.For the second case, the comparison of the force balance results with those from NASA shows a maximum divergence of 4.2% regarding the lift coefficient and 5.3% regarding the drag coefficient (Figure 4).These values are deemed completely satisfactory, given the fact that both the geometry of the wing models and the flow phenomena cannot be reproduced with absolute accuracy between two different experimental facilities.

Riblet specimens
For this work, two types of riblet specimens were examined, two sets of commercial riblets manufactured by the 3M company and three sets of riblets manufactured by Nanotypos [16] in the RADAERO project framework.These riblets are representative of the ones that will be applied on the surfaces of a small-scale fixed-wing UAV platform, during the project's later stages, to evaluate their performance during test flights.More specifically, the 3M riblet specimens are trapezoidal-shaped, with different spacing and height values.In contrast, the Nanotypos riblet specimens feature a blade shape (Figure 5) with a constant groove cross-sectional area and spacing-to-height ratios between 0.5 and 2. The specific dimensions of the specimens are given in Table 2.For the riblet specimen fabrication, an initial model of the microgroove geometry is manufactured, using stereolithography (SLA), which in turn is used to produce the negative riblet mold, from PDMS silicone elastomer.Using the PDMS mold the final riblet specimens are manufactured on top of a flexible PET film, using a UV curing resin, developed specifically for this task within the project.

Results
The results of the experimental measurements of all the blade-shaped riblet specimens (Figure 6) follow a similar trend as the freestream velocity increases and hence the corresponding local Reynolds number.More specifically, the drag coefficient is reduced with velocity increase, though the B1 specimen differs significantly from the others.This can be attributed to its very large riblet spacing (s/h=2), which is essentially too large for the riblets to operate properly, with the three mechanisms described earlier.The B2 and B3 specimen results are much closer together, though the B2 riblets with s/h=1 demonstrate the best overall performance, with the minimum drag value.The B3 specimen also produces an increase in the friction drag, compared to the smooth wall case, which is attributed to the overly dense riblet structure, which can be likened as an upward virtual translation of the wall surface.None of the blade-shaped riblet specimens produce any drag reduction compared to the smooth wall, since their dimensions, and more specifically their groove cross-sectional area, are larger than the optimal ones for the local conditions in the wind tunnel.This behavior is expected as there are practical limitations to the freestream velocity and local Reynolds number that can be achieved in the measurement area, as well as in the manufacturing process to achieve the desired spacing-to-height ratios.The B2 specimen demonstrates an almost identical value with the smooth wall, hinting that these riblets produce a drag-reducing effect that can overcome the increase in wetted area due to the riblet structure.The trapezoidal-shaped riblets (Figure 7) show a drag-reducing performance, which is more evident as the local Reynolds number is increased.The T2 specimen shows the best overall performance with a maximum drag reduction of 7.5% at a freestream velocity of 16.4m/s.In Figure 8, the experimental results of the blade-shaped specimens are compared against the computational results of LES simulations, published by Bliamis et al. [12].These results correspond to a freestream velocity of 15m/s and have the same trend, with B2 being the optimal geometry.The LES simulations predict lower drag values for the B2 and B3 specimens by about 8.1% and 2.3% respectively, while the B1 drag is greatly overpredicted.

Conclusions
In this work, the experimental investigations of blade-shaped and trapezoidal riblets are presented, for UAV applications.The former riblet specimens are designed and manufactured in the framework of the RADAERO project, while the latter are commercial ones, manufactured by the 3M company.The blade-shaped riblet specimens feature a constant groove cross-sectional area (Ag), and different spacing-to-height ratios (s/h) ranging from 0.5 to 2. The experiments were performed on a flat plate, equipped with a boundary layer transition enhancement strip, the drag force was measured using an inhouse developed force balance and the freestream flow velocities ranged from 5m/s up to 16.4m/s.The T1 and T2 trapezoidal riblets demonstrate a drag-reducing effect, with a maximum drag reduction of 6.2% and 7.5% respectively, at a flow velocity of 16.4 m/s.The blade-shaped riblets behave differently, with the B2 specimen having almost no effect on the drag, and the B1 and B3 specimens producing a drag increase.Moreover, the B1 specimen's riblet spacing is essentially too large and the riblet drag-reducing mechanisms fail to operate.The experimental results of the blade-shaped riblets are compared against LES simulations, with the results being qualitatively similar.The LES simulations predict very close drag values for the B2 and B3 specimens, though they largely overpredict the wall shear stresses acting on the B1 specimen.

Figure 3 .
Figure 3. Lift coefficient of a SHWT blade segment using the force balance and pressure taps.

Figure 4 .
Figure 4. Lift and drag coefficients of a NACA 0012 model from the force balance and NASA published experimental results [15].

Figure 5 .
Figure 5.The resultant blade-shaped riblet specimen and a close-up SEM photograph.

Figure 6 .
Figure 6.Drag coefficient values of the blade-shaped riblet specimens.

Table 1 .
Uncertainty of the measuring components

Table 2 .
The dimensions of the riblets specimens examined.