Sage Journals: Discover world-class research

Abstract

Vortex formation has numerous influences on the aerodynamic characteristics of fixed-wing micro air vehicle wings. Despite the mature understanding of vortices on fixed-wing and flapping micro air vehicle wings, the behavior of vortices over the morphing micro air vehicle wing has not been fully explored. Thus, the current work is conducted to investigate the influence of vortex structure over a series of twist-morphing micro air vehicle wings. Twist morphing micro air vehicle and baseline wings are simulated through fluid–structure interaction analysis. The validation results for each wing exhibited good correlation in the overall lift coefficient distribution trend. The vortex formation results show that vortex formations are significantly altered throughout angle of attack changes. For a given angle of attack cases below the stall angle, each morphing wing exhibited higher intensities of tip vortex structure formations and leading edge vortex–tip vortex interactions compared to the baseline wings. Stronger leading edge vortex–tip vortex interactions improved the low-pressure region over the morphing wing surface and further induce better lift performance. In fact, the morphing wing with higher morphing force induces better lift performance.

Keywords

Lift coefficient Micro air vehicle Morphing wing Twist morphing Vortices

Introduction

In recent years, aviation progress has focused on information gathering missions such as border patrol, environmental monitoring, military operations, and search and rescue.¹ Most of these missions require the rapid deployment of aircraft with stealth mode flight. Hence, small-scale aircraft such as a micro air vehicles (MAVs) or unmanned aerial vehicle are preferred for these operations because of their stealthy characteristics,² lower production costs, lower safety and certification requirements, and lower aerodynamic loads.³ However, LAR aircrafts suffer from low aerodynamic efficiency.⁴ Traditional high-lift devices such as flaps and slats are not efficient on LAR wings because the mechanisms create surface discontinuities and contribute to unnecessary complex airflow.³ Moreover, their conventional hinged mechanisms are too complex, bulky, and heavy to be applied on such a small wing area. Therefore, wing morphing was identified as a promising solution to replace conventional control surfaces and increase the overall aerodynamic performance.⁵ Wing morphing also has a huge potential in reducing flutter phenomenon, which directly improves aircraft comfort, safety, and fatigue problems.^6,7 Furthermore, morphing improves overall MAV wing aerodynamics⁸ and maneuverability.^9–11

Vortex occurrence has numerous influences on the aerodynamic characteristics of fixed-wing MAV wings. The vortex strength on a fixed-wing MAV wing varies throughout the angle of attack (AOA) changes.^12,13 The occurrence of leading edge vortex (LEV) generates higher lift force on a fixed-wing MAV,^14,15 and tends to interact with the wing tip vortex (TV), creating a complex flow couple. During vortex interaction, TV circulation pushes the LEV downwards and maintains its attachment on the wing surface area longer. The vortex attachment results in improved low-pressure distribution (over the wing surface), which enhances lift generation over the fixed-wing MAV.^16–18 The LEV attachment in a flapping MAV wing improves during the down stroke motion of the wing, which enhances the lift performance of the MAV wing type.¹⁹ By contrast, the wing TV influences the fixed MAV wing by creating a low-pressure core region near the wingtip and by interacting with the LEV.^16,17,20,21 The intensity of the low-pressure core region (within the wingtip vortex) highly contributes to the induced drag penalty.^17,22 The wingtip vortex in a flapping MAV wing interacts with the root vortex to generate the wake structure, which in turn affects the wing lift and drag performance.²³ Despite the mature understanding of vortices on fixed-wing and flapping MAV wings, the formation and vortices behavior over the morphing MAV wings have not been fully explored. Thus, the current work is conducted to elucidate the formation of vortices over a morphing MAV wing based on the fluid–structure interaction (FSI) results. A series of simulations works involving morphing MAV wings with twist mobility (and baseline wings, namely, membrane and rigid MAV wings) are simulated through the FSI analysis. A validation of lift coefficient (C_L) on each wing was conducted in the initial analysis to justify the FSI simulation model. A detailed simulation study on vortex formation and its pressure distribution resumed over the twist-morphing (TM) and baseline wing to elucidate the influence of vortices on the wings.

Methodology

FSI frameworks

The FSI problems of morphing wings are solved in 3D, quasi-static, and linear structural models coupled with steady state, incompressible, and turbulent flow domains. The 3D turbulent flow is defined based on Reynolds-averaged Navier–Stokes (RANS) equations and the shear stress turbulence (SST) model. All simulation methods found in this work is set up based on the FSI ANSYS-Workbench framework, coupled with static structural analyses (ANSYS-Mechanical), and the flow solver module (ANSYS-CFX). The details of the FSI simulation method is found in Ismail et al.⁸ The FSI simulation framework is given in Figure 1.

Figure 1.

FSI simulation framework.

MAV wing model

Three levels of the TM wing (TM 5N, TM 3N, and TM 1N) are used to create comparisons with the baseline wing models (membrane wing and rigid wing). The TM wing and baseline wing models are developed based on previous research found at the Universiti Teknologi MARA, Malaysia^8,24 and University of Florida (UF).^25,26 Generally, all wings have almost identical characteristics in terms of planform shape and dimension. The distinctive parts among the wings are the morphing force and flexible membrane skin components. A summary of TM (TM 5N, TM 3N, and TM 1N), membrane, and rigid wings configurations is given in Table 1. Each wing has 1 mm thickness (including the membrane skin component). The following coordinate system is adopted for all wing models: x is the chordwise direction, z is the spanwise direction, and y is the directed normal to the wing upper surface, where the origin is located at the outermost wing leading edge point.

Table 1.

Summary of twist morphing (TM) and baseline wing configurations.

	TM 5N	TM 3N	TM 1N	Membrane wing	Rigid wing
Wingspan, b	150 mm	150 mm	150 mm	150 mm	150 mm
Root chord, c	150 mm	150 mm	150 mm	150 mm	150 mm
Aspect ratio, A	1.25	1.25	1.25	1.25	1.25
Maximum camber at the root	6.7% of c (at x/c = 0.3)	6.7% of c (at x/c = 0.3)	6.7% of c (at x/c = 0.3)	6.7% of c (at x/c = 0.3)	6.7% of c (at x/c = 0.3)
Maximum reflex at the root	1.4% of c (at x/c = 0.86)	1.4% of c (at x/c = 0.86)	1.4% of c (at x/c = 0.86)	1.4% of c (at x/c = 0.86)	1.4% of c (at x/c = 0.86)
Built-in geometric twist	0.55°	0.55°	0.55°	0.55°	0.55°
Force component	F = 5 N	F = 3 N	F = 1 N	Excluded	Excluded
Membrane skin component	Included	Included	Included	Included	Excluded
Symmetrical wing geometry

MAV wing model

The objective function of the morphing force (found on the TM wing) is to produce significant wing deformations that consequently alter the wing twist characteristics on the TM wing. The morphing force imposed at an optimized morphing point location for morphing mobility is shown in Figure 2. The optimized morphing point location is positioned near the wing edge to ensure efficient morphing mobility. The wing deformation (y-direction) and geometric twist (ε) results for all wings are presented in Figures 3 and 4, respectively. The initial results concluded that higher morphing force configuration could produce greater wing deformations and larger positive twist (washed-in) magnitudes.

Figure 2.

Morphing force applied on TM wing.

Figure 3.

The wing deformation (y-direction) results for all wings.

Figure 4.

The geometric twist characteristics on all wings.

Flow domains and mesh generation

The computational flow domain (CFD), which is built around each MAV wing with a symmetrical condition was applied. The 3D CFD is created in the root chord unit (c), as shown in Figure 5. The grid-independent test results show that the optimized grid is achieved at 1,000,000 elements, wherein the first cell above the wing surface is set at y⁺≤ 1. The inlet and outlet are marked by flow vectors (Figure 6). The magnitudes of velocity are set at 9.5 m/s (equivalent to the maximum MAV wing condition at Re = 100,000). Inlet velocity was specified at the inlet, and a zero pressure boundary condition is implemented at the outlet. The AOA varies from −10° to 35°. The symmetrical wall (as shown in Figure 6) and side walls are defined as symmetrical and slip surface boundary conditions, respectively. The wing surface is defined as a no-slip boundary surface and assigned as the boundary interaction for FSI investigation. Automatic wall function is fully employed to solve the flow viscous effect.

Figure 5.

3D CFD domain size.

Figure 6.

The boundary condition applied on CFD domain.

Results

Validation of the C_L performances

Before the vortex formation study was conducted, a validation of the C_L performances of each MAV wing was conducted. Figure 7 presents the C_L performances for all wings based on the simulation and experimental results at U = 9.5 m/s. The experimental results were obtained from a wind tunnel testing provided in Ismail.²⁷ The validation results are presented as the preliminary justification and reliability of the simulation results.

Figure 7.

The simulation (left) and experimental (right) lift coefficient results for all wings.

The simulation slightly under predicted the C_L distribution for every wing and U case. Based on the analysis taken at AOA between −5° and 15° (pre-stall angle), the simulation results is slightly below the experimental C_L result (5%)for every wing case. However, the discrepancy between the simulation and experimental C_L increased (between 7% and 20%) as the AOA increased beyond 15°. Rojratsirikul et al.²⁸ suggested that the discrepancy in C_L magnitude is contributed by the self-induced vibrations on the membrane component because the occurrence of membrane vibration increased with the incremental increase of AOA. Therefore, rather than comparing the C_L magnitude, the following validation work is conducted mostly by comparing the common C_L trend found in the simulation and experimental results.

By comparing the major C_L distribution trend found in the experimental and simulation results, the C_L slopes (taken at AOA = 0° to 15°) for every wing (both experimental and simulation results) were found at 0.033. The experimental and simulation results also exhibited similar C_L distribution trends (taken at AOA = 0° to 15°) in which, the TM 5N wing had produced the highest C_L magnitude among the wings, followed by the TM 3N and TM 1N wing. The membrane and rigid wing produced lower C_L distributions compared with the TM wings.

The experimental and simulation results also showed that the TM 5N wing produced the lowest AOA_stall at an AOA range of 15°–20°. TM 3N and TM 1N were predicted to stall at AOA_stall = 18° and 21° respectively, which was very close to its actual AOA_stall between 20° and 25°. The similarities continue to the baseline wings cases, where both methods agreed in delaying the AOA_stall incidence. The membrane and rigid wings were predicted to stall at AOA_stall = 22° and AOA_stall = 24°, respectively. These results were parallel to the experimental results, in which both baseline wings stalled at an AOA range between 25° and 30°. Based on these AOA_stall characteristics, the overall trend of AOA_stall characteristics for both methods was almost similar. Both results agreed in the overall AOA_stall characteristics. Based on this C_L distribution, the experimental and simulation results exhibited a good correlation in the C_L distribution trend. Despite the slight differences found in C_L magnitude, the simulation satisfactorily predicted the C_L slope, AOA_stall characteristics, and the C_L distribution. The differences in the C_L and AOA_stall magnitudes are contributed by the suspicion of self-induced membrane vibrations,²⁹ time-average solution, and the selected turbulence model.²⁷

Vortex formation

Three-dimensional vortices are visualized based on the vortex core region by using limited Q criterion magnitude at Q = 0.03 as shown in Figure 8.²⁴ The results are viewed from the wing top view angle (y-direction) to capture the details of the LEV and TV formations. Based on the results, the LEV and TV formations are visually recognized through their positions on the wing surface area. The LEV structures normally occur on the wing leading edge area because the LEV is generated by the roll up of the separated shear layer flow produced at the wing leading edge.¹⁶ In contrast, TV structures are normally found near the wing tip area because TV structures are visibly recognized as a circulatory 3D flow motion that trails downstream from the wing tip area. TV structures are produced from the flow that leaks around the wing tips because of finite wing pressure difference.³⁰

Figure 8.

(a) LEV–TV interaction area, (b) the location and measurement of D_TV, L_TV, and D_LEV-TV.

The visualization of LEVs and TV structures cannot be separated into two components because of the limitation in the simulation post processing module. Thus, the TV and LEV–TV interactions area were determined based on approximate location as shown in Figure 8(a). In order to quantify the TV and LEV–TV structures, a digital image measuring software is used to estimate the maximum diameter and length of TV. The software is also used to measure the approximate LEV–TV diameter. The approximate size of maximum TV diameter (D_TV), TV length (L_TV), and LEV–TV diameter (D_LEV-TV) were measured based on chord length (c) wing at certain location as shown in Figure 8(b). In the following section, the discussion on vortices formation is based on the magnitude of D_TV, L_TV, and D_LEV-TV. The physical enlargement of LEV–TV connections and TV structure formations were quantify (based on D_TV, L_TV, and D_LEV-TV magnitude) and relatively compared (between the AOA cases) to indicate the vortices intensities and also its relative influence towards the C_L generation shown in Figure 7 (simulation results). The approximate LEV coverage area over the wing surface is used to indicate the dominancy of the LEV attachment. The following discussion on the study of vortices is concentrated on the comparative TV sizes and the proportional size of the LEV–TV connection found on every MAV wing.

Figure 9 presents the 3D vortex formations on the MAV wings at U = 9.5 m/s (equivalent to Re = 100,000). The results show that each wing had produced a variation of vortex formation in every AOA cases (started from 0° to 20°). At low AOA (AOA = 0°), each wing produced a dominant LEV attachment over the wing surface. LEV domination is justified based on the LEV attachment that covered nearly half (x/c ≈ 0.45) of the wing surfaces. At this AOA stage, each wing managed to produce almost similar LEV dominances on the wing surface. Despite the dominance in the LEV attachment, the existence of TV structures showed some diminutive variations on each MAV wing.

Figure 9.

3D vortex formations over the MAV wings.

Based on the comparative TV sizes (at AOA = 0°), the TM 5N wing relatively produced the largest TV structure (D_TV = 0.09c and L_TV = 0.6c) among the MAV wings, followed by TM 3N (D_TV = 0.07c and L_TV = 0.53c) and TM 1N (D_TV = 0.05c and L_TV = 0.4c) wings. The membrane and rigid wings produced relatively similar TV structure sizes (D_TV = 0.035c–0.04c and L_TV = 0.24c–0.27c). At this AOA stage, the LEV–TV interactions for each MAV wing had slightly varied among the wings. The morphing wings (TM 5N, TM 3N, and TM 1N) managed to produce D_LEV-TV = 0.05c–0.035c, in which about more than 14% larger than the baseline (membrane and rigid) wings produced. In order to relate this vortices formation towards the C_L performance, a detailed study on the low-pressure distribution over the wing upper surface was conducted as shown in Figure 10. Therefore, the low-pressure coefficient found over the wing upper surface is used to indicate or induce better lift distribution. Thus, based on pressure distribution results (Figure 10 at AOA = 0°), the low-pressure region found over the morphing wings is highly concentrated at similar LEV–TV interactions region (1.0 < 2z/b < 0.8). The magnitude of $C min □$ is used and define as on the lowest pressure coefficient found at LEV–TV interactions area (1.0 < 2z/b < 0.8). Based on Figure 10 (at AOA = 0°), TM 5N wing exhibited the lowest $C min □$ magnitude among the wings with $C min □$ = −1.773. The intensity of $C min □$ found on TM 3N and TM 1N wings were approximately 85% (TM 3N) and 98% (TM 1N) higher than the TM 5N wing. The baseline wing produced higher $C min □$ at −0.053. The low $C min □$ magnitude found in every morphing wings have translated into higher C_L magnitude (simulation results shown in Figure 7), in which the C_L magnitude for TM 5N, TM3N, and TM 1N are 0.47, 0.33, and 0.19, respectively. The C_L for baseline wing is almost similar at 0.12–0.13.

Figure 10.

Pressure distribution and magnitude of $C min □$ over the wing upper surface.

As the AOA increased to 5°, the LEV dominance over the wing surface area gradually reduced with the incremental increase of AOA. Despite the lower LEV dominance, the TV structures simultaneously grew larger than the previous AOA case (AOA = 0°). The TV structure formations for every wing increased based on D_TV, L_TV, and D_LEV-TV magnitudes. Again, TM 5N wing produced the largest TV structure (D_TV = 0.1c and L_TV = 0.62c) among the MAV wings, followed by the TM 3N (D_TV = 0.07c and L_TV = 0.6c) and TM 1N (D_TV = 0.06c and L_TV = 0.5c) wings. The LEV–TV interactions for each morphing wing also continues to enlarge with D_LEV-TV = 0.09c–0.04c and maintain at least 14% greater than the baseline wings produced. Due to larger TV and LEV–TV interactions structure, TM 5N wing again produced the lowest $C min □$ (Figure 10) at −2.265. The intensity of $C min □$ found on TM 3N and TM 1N also continues to improve with 50% (TM 3N) and 60% (TM 1N) higher than the TM 5N wing produced. As a result, the C_L magnitude (Figure 7) for TM 5N, TM3N, and TM 1N are significantly improved at 0.71, 0.62, and 0.47, respectively. However, the C_L magnitude for baseline wings are slightly increased at 0.4 due to low D_TV (0.05c), L_TV (0.47–0.42), D_LEV-TV (0.035c), and $C min □$ (−0.35) magnitudes.

The intensity of D_TV, L_TV, and D_LEV-TV increased progressively with the incremental increase of AOA. At AOA = 10°, TM 5N wing induced the largest D_TV magnitude at 0.13c but it started to show a depletion in L_TV length at 0.46c (compared to AOA = 5°case). However, TM 3N (D_TV = 0.12c, L_TV = 0.6c) and TM 1N (D_TV = 0.1c, L_TV = 0.5c) wings continue to increase in D_TV and L_TV intensity. The membrane (D_TV = 0.09c and L_TV = 0.55c) and rigid (D_TV = 0.08c and L_TV = 0.5c) wings also exhibited a slight increment in TV structures. Despite the more obvious TV structures found on the baseline wings, the LEV–TV interactions on morphing wings continue to enlarge at D_LEV-TV = 0.13c–0.1c and maintain at least 10% larger than the baseline wings produced. Thus, the $C min □$ and C_L magnitude exhibited by TM 5N ( $C min □ = - 3.267,$ C_L = 0.98), TM 3N ( $C min □ = - 1.971,$ C_L = 0.83), and TM 1N ( $C min □ = - 1.632,$ C_L = 0.67) wing are better than the baseline wings ( $C min □ = - 0.750,$ C_L = 0.59–0.52) as shown in Figures 7 and 10 (AOA = 10° case).

The intensity of the TV structure formations and LEV–TV connections on every wing varies significantly as the AOA increase at 15°. Surprisingly, the TV structure and LEV–TV interactions on TM 5N wing has drastically started to deteriorate and detached from the wing surface. Thus, D_TV, L_TV, and D_LEV-TV data for TM 5N are not available to be measured at this AOA stage. As a result, the $C min □$ and C_L magnitude for TM 5N wing depleted at $C min □ = - 1.253$ and C_L = 1.0 (shown in Figures 7 and 10). This situation is expected since the stall angle (AOA_stall) for TM 5N wing occurred at relatively low angle 15°.

TM 3N (D_TV = 0.13c, L_TV = 0.42c, and D_LEV-TV = 0.2c), TM 1N (D_TV = 0.12c, L_TV = 0.5c, and D_LEV-TV = 0.14c) and membrane (D_TV = 0.1c, L_TV = 0.54c, and D_LEV-TV = 0.11c) wings also suffer from a slight decrement in L_TV length. Despite the slight L_TV depletion found on morphing and membrane wings, the rigid (D_TV = 0.09c, L_TV = 0.55c, and D_LEV-TV = 0.07c) wing continue to increase in D_TV, L_TV, and D_LEV-TV intensities. However, due to higher D_LEV-TV intensities found on TM 3N ( $C min □ = - 2.251,$ C_L = 1.06) and TM 1N ( $C min □ = - 1.821,$ C_L = 0.96), the morphing wings (except TM 5N) managed to produce better $C min □$ and C_L magnitude than the baseline ( $C min □ = - 1.783,$ C_L = 0.88–0.85) wings.

At AOA = 20°, the vortex deterioration and LEV detachment occurrences continued to intensify for the TM 5N wing. In fact, the TM 3N wing also demonstrate the deterioration of TV and LEV–TV structure which is similar to the stall vortex formations found on TM 5N wing at AOA = 15°. This situation is expected for TM 3N wing since the AOA_stall for the TM 3N wing occurred at ≈ 19° (Figure 7). Thus, the magnitude of D_TV, L_TV, and D_LEV-TV for TM 5N and TM 3N wings are not available to be measured. However, the intensity of the TV structure formations and the LEV–TV connections continued to increase for the TM 1N and baseline wing. TM 1N (D_TV = 0.15c, L_TV = 0.35c, D_LEV-TV = 0.22c), membrane (D_TV = 0.14c, L_TV = 0.45c, D_LEV-TV = 0.17c) and rigid (D_TV = 0.12c, L_TV = 0.5c, D_LEV-TV = 0.13c) wings exhibited enlargement in D_TV and D_LEV-TV magnitude but also suffer from a slight decrement in L_TV length. Despite the slight depletion in L_TV length, TM 1N ( $C min □ = - 2.121,$ C_L = 1.13) wing still managed to produce better C_L than the baseline ( $C min □ = - 2.231 ~ - 1.74,$ C_L = 1.06–1.04) wings. This is due larger low-pressure area found on the TM 1N wing surface (Figure 10). However, stall vortex formations (similar to TM 5N and TM 3N) is expected to occur on the TM 1N and baseline wings once the angle increased to AOA_stall incidences (AOA_stall = 21°(TM 1N), 22° (membrane), and 24° (rigid).

Based on the vortex formation results, it shows that the overall vortex formations on the current MAV wings are significantly altered throughout AOA changes. The intensity of TV structure formations and LEV–TV interactions which are measured through the D_TV, L_TV and D_LEV-TV magnitude increased with the incremental increase of AOA. Larger D_LEV-TV magnitude signifies stronger LEV–TV interactions which subsequently improve the low-pressure region over the wing surface (denoted by $C min □$ ) and induce better lift performance (C_L). However, as the AOA reached the stall angle, the TV structure and LEV–TV interactions drastically deteriorate and detach from the wing surfaces and, thus, creates a depletion on low-pressure creation over the wing surface. LEV dominance over the wing surface area has also reduced gradually with the incremental increase of AOA. However, the LEV formation also drastically deteriorated and detached from the wing surface as the AOA approached wing AOA_stall.

Based on D_TV, L_TV, and D_LEV-TV results, it shows that for a given AOA cases below the stall angle, each morphing wing demonstrated higher intensities of TV structure formations and LEV–TV interactions compared to the baseline wings. Despite the slight depletion in L_TV magnitude (as AOA increase near the wing’s stall angle), each morphing wing still managed to produce larger D_TV and D_LEV-TV magnitude compared to the baseline wings. Theoretically, the twist morphing mobility (washed-in twist) has encouraged higher intensity of TV structure and LEV–TV interactions on morphing wing. In fact, the morphing wings with greater morphing force (5 N or 3 N) induced the higher intensity of TV structure and LEV–TV interactions on the wings. The twist morphing mobility improves the low-pressure region over the wing surface and further induces better lift performance for morphing wings. Despite the better lift performance, the morphing wings also suffered from earlier stall vortex formations than the baseline wings. The twist morphing mobility induced earlier vortices deterioration and detachment on the MAV wing. Therefore, the morphing wing with higher morphing force promotes earlier stall condition on MAV wing.

Conclusion

The vortex formation results show that vortex formations are significantly altered throughout AOA changes. The intensity of TV structure formations and LEV–TV interactions, which are measured through the D_TV, L_TV, and D_LEV-TV magnitude increased with the incremental increase of AOA. The results shows that for a given AOA cases below the stall angle, each morphing wings demonstrated higher intensities of TV structure formations and LEV–TV interactions compared to the baseline wings. Stronger LEV–TV interaction improves the low-pressure region over the morphing wing surfaces and further induces better lift performance. In fact, the morphing wing configuration with higher morphing force produce better lift performance. However, the morphing wings also suffered from earlier stall vortex formations compared to the membrane or rigid wings. The morphing mobility induced earlier vortices deterioration and detachment on the MAV wing.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research,authorship,and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research,authorship,and/or publication of this article: The authors acknowledge the technical and financial support from Universiti Teknologi MARA (Pulau Pinang) and the financial supports from the Government of Malaysia via Malaysia Ministry of Higher Education’s Fundamental Research Grant Scheme (FRGS) (No. 600-RMI/FRGS 5/3 (152/2014)).

References

Durai DT, Viswamurthy SR, Chaplod S, et al. Development of a blended wing configuration MAV. In: Symposium on applied aerodynamics and design of aerospace vehicle (SAROD 2011), Bangalore, India, 2011.

Petricca

Ohlckers

Grinde

. Micro- and nano-air vehicles: State of the art. Int J Aerosp Eng 2011; 2011: 1–17.

Barbarino

Bilgen

Ajaj

. A review of morphing aircraft. J Intell Mater Syst Struct 2011; 22: 823–877.

McCormik

. Aerodynamics, aeronautics and flight mechanics, 2nd ed. New York: Wiley, 1995.

Vasista

Tong

Wong

. Realization of morphing wings: A multidisciplinary challenge. J Aircraft 2012; 49: 11–28.

Thill

Etches

Bond

. Morphing skins. Aeronaut J 2008; 3216: 1–23.

Gomez

Garcia

. Morphing unmanned aerial vehicles. Smart Mater Struct 2011; 20: 1–16.

Ismail

Zulkifl

Abdullah

. Computational aerodynamic analysis on perimeter reinforced (PR)-compliant wing. Chin J Aeronaut 2013; 26: 1093–1105.

Abdulrahim M. Flight performance characteristics of a biologically-inspired morphing aircraft. In: 43rd AIAA aerospace sciences meeting and exhibit, Reno, NV, USA, 2005.

10.

Shang

Combes

Finio

. Artificial insect wings of diverse morphology for flapping-wing micro air vehicles. Bioinsp Biomim 2009; 4: 1–6.

11.

Bachmann

Vaidyanathan

Boria

. Flying insects and robots: A miniature vehicle with extended aerial and terrestrial mobility, Berlin Heidelberg: Springer, 2009, pp. 247–269.

12.

Mueller

DeLaurier

. Aerodynamics of small vehicles. Ann Rev Fluid Mech 2003; 35: 89–111.

13.

Mueller TJ and Torres GE. Aerodynamics of low aspect ratio wings at low reynolds numbers with applications to micro air vehicle design and optimization. Naval Research Laboratory Report, University of Notre Dame, USA, 2001.

14.

Srygley

Thomas

ALR

. Unconventional lift-generating mechanisms in free-flying butterflies. Nature 2002; 420: 660–664.

15.

Sane

. The aerodynamics of insect flight. J Exp Biol 2003; 206: 4191–4208.

16.

Shields

Mohseni

. Effects of sideslip on the aerodynamics of low-aspect-ratio. AIAA J 2012; 50: 85–99.

17.

Taira

Colonius

. Effect of tip vortices in low-Reynolds-number poststall flow control. AIAA J 2009; 47: 749–756.

18.

Ringuette

Milano

Gharib

. Role of the tip vortex in the force generation of low-aspect-ratio normal flat plates. J Fluid Mech 2007; 581: 453–468.

19.

Koekkoek

Muijres

Johansson

. Stroke plane angle controls leading edge vortex in a bat-inspired flapper. Comptes Rendus Mécanique 2012; 340: 95–106.

20.

Colonius

TIM

Taira

. Three-dimensional flows around low-aspect-ratio flat-plate wings at low Reynolds numbers. J Fluid Mech 2009; 623: 187–207.

21.

Albertani

Stanford

Hubner

. Aerodynamic coefficients and deformation measurements on flexible micro air vehicle wings. Exp Mech 2007; 47: 625–635.

22.

Stanford

Ifju

. Membrane micro air vehicles with adaptive aerodynamic twist: Numerical modeling. J Aerosp Eng 2009; 22: 173–184.

23.

Nakata

Liu

Tanaka

. Aerodynamics of a bio-inspired flexible flapping-wing micro air vehicle. Bioinsp Biomim 2011; 6: 1–11.

24.

Ismail

Zulkifli

Abdullah

. Optimization of aerodynamic efficiency for twist morphing MAV wing. Chin J Aeronaut 2014; 27: 475–487.

25.

Stanford BK. Aeroelastic analysis and optimization of membrane micro air vehicle wings. PhD Thesis Dissertation, University of Florida, USA, 2008.

26.

Stanford

Ifju

Albertani

. Fixed membrane wings for micro air vehicles: Experimental characterization, numerical modeling, and tailoring. Prog Aerosp Sci 2008; 44: 258–294.

27.

Ismail NI. Aerodynamic performances of twist morphing MAV wing. PhD Thesis, Universiti Teknologi MARA, Malaysia, 2015.

28.

Rojratsirikul

Genc

Wang

. Flow-induced vibrations of low aspect ratio rectangular membrane wings. J Fluids Struct 2011; 27: 1296–1309.

29.

Bleischwitz

de Kat

Ganapathisubramani

. Aspect-ratio effects on aeromechanics of membrane wings at moderate Reynolds numbers. AIAA J 2015; 53: 780–788.

30.

Sohn

Chang

. Visualization and PIV study of wing-tip vortices for three different tip configurations. Aerosp Sci Technol 2011; 16: 40–46.

Vortex structure on twist-morphing micro air vehicle wings

Abstract

Keywords

Introduction

Methodology

FSI frameworks

MAV wing model

MAV wing model

Flow domains and mesh generation

Results

Validation of the CL performances

Vortex formation

Conclusion

Footnotes

Declaration of conflicting interests

Funding

References

Validation of the C_L performances