Original Article

Proc IMechE Part G:

J Aerospace Engineering
Transonic flutter characteristics 2021, Vol. 235(6) 661­–674
! IMechE 2020
of an airfoil with morphing devices
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/0954410020953046
journals.sagepub.com/home/pig
Shun He , Shijun Guo and Wenhao Li

Abstract
An investigation into transonic flutter characteristic of an airfoil conceived with the morphing leading and trailing edges
has been carried out. Computational fluid dynamics (CFD) is used to calculate the unsteady aerodynamic force in
transonic flow. An aerodynamic reduced order model (ROM) based on autoregressive model with exogenous input
(ARX) is used in the numerical simulation. The flutter solution is determined by eigenvalue analysis at specific Mach
number. The approach is validated by comparing the transonic flutter characteristics of the Isogai wing with relevant
literatures before applied to a morphing airfoil. The study reveals that by employing the morphing trailing edge, the
shock wave forms and shifts to the trailing edge at a lower Mach number, and aerodynamic force stabilization happens
earlier. Meanwhile, the minimum flutter speed increases and transonic dip occurs at a lower Mach number. It is also
noted that leading edge morphing has negligible effect on the appearance of the shock wave and transonic flutter. The
mechanism of improving the transonic flutter characteristics by morphing technology is discussed by correlating shock
wave location on airfoil surface, unsteady aerodynamics with flutter solution.

Keywords
Morphing wing, transonic dip, flutter speed, shock wave
Date received: 8 October 2019; accepted: 29 July 2020

Introduction
adoption in the real structure.6 Hu et al.7 paid more
Morphing wing technology has been developed to attention to the nonlinear aeroelastic characteristics
improve aerodynamic efficiency and flight perfor- of a folding wing with cubic connection stiffness in
mance by changing the wing shape adaptively the quasi-steady condition and during the morphing
during flight.1 It is regarded as one of the potential process, and variable motion types, such as chaos
and feasible approach contributing to next generation motion, were observed in the numerical calculation.
green transport aircraft.2 However, the coupling Flutter and divergent speeds significantly changed
between the aerodynamics and structures of a morph- during the transition between take-off, climb, cruise
ing wing may have more significant influence to the and loiter phases of a fully morphing wing studied by
aeroelastic characteristics especially in transonic €
Unlüsoy and Yaman.8 In particular, unsteady aero-
regime. The drop of transonic flutter speed, the so- dynamic analysis of a morphing wing conducted by
called ‘transonic dip’,3,4 is a major concern for a con- Kan et al.9 and Xiang et al.10 showed that the stall
ventional transport aircraft. For a morphing wing, can be delayed by implementing a flexible periodical
this aeroelastic phenomenon raises a new challenge trailing-edge deflection. These results suggest that
to be addressed. applying morphing technology on a conventional
In recent years, many efforts have been made to wing may have a favourable effect on the flutter char-
assess the aeroelastic stability for different types of acteristic. Flutter suppression by active control of
morphing concepts. The flutter mechanism of a morphing wing was also studied. Based on a low-
span-morphing wing was analysed by Huang et al.,5 fidelity aeroelastic model and cantilever uniform
revealing the rigid-body modes especially pitch
motion of the aircraft have a significant effect on flut-
Centre for Aeronautics, Cranfield University, Cranfield, Bedford, UK
ter characteristics. The flutter and divergence charac-
Corresponding author:
teristics of a large civil aircraft with morphing Shijun Guo, Centre for Aeronautics, Cranfield University, Cranfield,
winglets and adaptive flap tabs were systemically Bedford MK43 0AL, UK.
examined to assess the robustness and safety of Email: [email protected]
2662 Proc
Proc IMechE
IMechE Part
Part G: G: J Aerospace
J Aerospace Engineering
Engineering 0(0)
235(6)

morphing wings, the feasibility of active flutter sup- mechanism of improving the transonic flutter charac-
pression by using morphing device was studied by teristics by morphing technology is investigated by
Ajaj and Friswell.11 Investigation was also extended altering the angle of attack of the airfoil due to the
into the dynamic behaviour and stability of an axially aerodynamic similarity of them.
morphing wing in supersonic airflow12 to demonstrate In this paper, the governing equations of motion
that a proposed morphing law is effective for flutter and technical methods are formulated for flutter anal-
suppression. ysis of the morphing wing in the next section. The
From a practical point of view so far, morphing validation study of the methods using the classical
technology is more suitable for small unmanned Isogai wing is performed. Subsequently, the numeri-
vehicles at low flight speed.13 Hence most of the cal results and discussion of the transonic flutter char-
study on aeroelastic characteristics of morphing acteristic of the morphing wing are presented.
wing has been focused in subsonic air flow. Few inves- Conclusions are drawn in the last section.
tigation has been found into the effect of morphing
wing on the transonic flutter despite the transonic
Aeroelastic airfoil model with morphing
flutter is a critical concern for large commercial
aircraft.14,15 leading and trailing edges
Generally speaking, transonic flutter is a nonlinear
The governing equation of a wing section motion
aeroelastic phenomenon due to shock wave acting on
the wing. In transonic regime, the steady flow param- A typical wing section modelled as an aeroelastic
eters vary with spatial position in the flow field airfoil system with plunging (h) and pitching (a)
around the wing that generates unsteady aerodynamic motion is illustrated in Figure 1. The elastic axis
forces induced by the wing motion. To simplify the of the airfoil (E point) is located at a distance of
analysis, it is assumed that the parameters of flow ab after the mid-chord point; the gravity centre
field and the shock wave motion vary in a linear fash- (G point) is located at xa b after the elastic axis,
ion with the wing motion of small perturbation in where b is the semi-chord length, m is the mass per
transonic steady air flow. This is usually called unit span, Sa ¼ xa b is the first moment of inertia
dynamically linear but statically nonlinear about the elastic axis, and Ia ¼ mr2a is the moment
aerodynamics.16 of inertia about the elastic axis. The bending stiffness
Wing transonic flutter analysis is a particularly dif- and torsion stiffness of coefficient Kh ¼ mx2h and
ficult task but also an attractive academic topic for Ka ¼ Ia x2a are modelled by springs attached to the
aeroelasticians due to the complexity in aerodynamics elastic axis.
modelling. The coupled CFD/CSD (Computational The governing equations of motion of such a 2-D
Structural Dynamics) approach, or time marching system were derived from the Lagrange equations
approach based on CFD, has long been used to according to Dowell et al.,16 written as
obtain the wing transonic flutter characteristic since (
the 1990s.17,18 However, this approach is time- mh€ þ Sa €a þ Kh h ¼ �L
consuming and requires much computation cost (1)
Sa h€ þ Ia €a þ Ka a ¼ Mea
even to perform only a single solution of wing tran-
sonic flutter. To overcome the disadvantage of the
coupled CFD/CSD time marching method, ROM where L ¼ 1=2qV2 cl and Mea ¼ 1=2qV2 cm are
approaches,19–21 based on CFD technique are devel- the aerodynamic lift and moment about the elastic
oped to calculate the wing flutter speed with the axis, respectively. cl is the lift coefficient, cm is
assumption of dynamically linear aerodynamics.16 the aerodynamic moment coefficient, and q is the
Amongst these approaches, the system identification air density.
method19,22 is a robust and effective technique to By introducing non-dimensional time s ¼ xa t and
build the transonic aerodynamic ROM, which is mass ratio l ¼ m=pqb2 , equation (1) can be
used in the present study. rewritten as
A morphing wing model with the leading edge and
trailing edge devices from previous study23,24 was
employed in this paper. Since the transonic flutter is
more sensitive to airfoil profile than structural prop-
erty, the mass and stiffness distribution was kept con-
V
stant for different morphing configurations in the
present study. This setting enables us to address the E G
effect of the aerodynamic shape of a morphing wing
on the transonic flutter characteristics. The results
indicate that the morphing trailing edge device can
improve the transonic flutter characteristic, but the
morphing leading edge has negligible influence. The Figure 1. A typical two-dimensional aeroelastic airfoil system.
He et
He et al.
al. 6633

8
>
> � �2 U2 na past outputs and nb inputs, the only task is to
>
< h00 þ xa a00 þ x h
h ¼ ð�cl Þ identify the constant coefficient matrices Ai and Bi .
xa pl
2 (2) In the present study, a so-called ‘3211’ signal devel-
>
> U
> xa h00 þ r2a a00 þ r2a a ¼
: ð2cm Þ oped by Cowan et al.19 is utilized as the input of the
pl
CFD solve due to its broad frequency spectra. Then the
where ðÞ00 ¼ d2 ðÞ=ds2 , U ¼ V=bxa is the non- least squares method is adopted to fit the time history of
dimensional air speed. Subsequently, the governing the output of the CFD solver, i.e., f a , to carry out the
equation can be written in matrix form, unknown coefficient matrices in equation (5).
One challenging problem with using the ARX
U2 model to build aerodynamic ROM is how to determine
Mn00 þ Kn ¼ f (3) the order of the model. In theory, the order of aero-
pl a
dynamic ROM could vary at different flow conditions,
� � for instance, Mach number, angle of attack even airfoil
1 xa
where M¼ is the mass matrix, profile. In the present study, identifying the model
xa r2
" #a order is treated as a minimization problem at specific
K¼ ðxh =xa Þ2 0 is the stiffness matrix. For this flow condition, which can be written in a general form
0 1
� �T min :J
aeroelastic system, n ¼ h=b a and f a ¼ na ¼ 1;2; . . . ; 12
� �T nb ¼ 1;2; . . . ; 12
� cl 2cm serve as the generalized displacements
and the generalized aerodynamic forces, respectively.
The generalized aerodynamic forces corresponding to
the generalized displacements in transonic air flow where J ¼ wcl;CFD � cl;ROM þ ð1 � wÞcm;CFD � cm;ROM
can be obtained from full CFD simulation or an aero- where w is the weight factor, chosen from 0.2 to 0.4
dynamic ROM. herein.
By� defining the structural state vector Thus, the Genetic Algorithm (GA) can be imple-
�
0 T mented to search the most appropriate order of ARX.
xs ¼ n n , the structural motion equation of
the aeroelastic system can be written as And with well-determined orders and corresponding
coefficient matrix, the discrete-time ARX model can
8
< U2 be transformed into the continuous-time form
x0s ¼ As xs þ Bs f a (4) through Tustin approximation, which can be
: pl
n ¼ Cs xs þ Ds f a described in state-space form as
�
� � x0a ¼ Aa xa þ Ba n
0 0 I (7)
where ð Þ ¼ dðÞ=ds, As ¼ , f a ¼ C a x a þ Da n
�M �1 K 0
� � � �
0 I
Bs ¼ , C s ¼ , and Ds ¼ ½0�. where the aerodynamic state vector is
M�1 0
� �T
Aerodynamic ROM and flutter equation f a ðs � DsÞ . . . f a ðs � naDsÞ
xa ¼
nðs � DsÞ . . . nðs � ðnb � 1ÞDsÞ
Amongst the numerous methods, the system identifi-
cation method is an effective and efficient technique
and Aa , Ba , Ca , Da are the coefficient matrices for the
to establish aerodynamic ROM. Following the sug-
state-space model of aerodynamic
� ROM.
�T
gestion from Cowan et al.19 and Zhang and Ye,22 the
Then, introducing x ¼ xTs xTa and coupling
ARX model is used to establish the ROM of transon-
the structural motion equation (4) with the aerody-
ic aerodynamics. The time domain equation for Multi
namic ROM equation (7), the governing equation for
Input and Multi Output (MIMO) ARX model can be
the aeroelastic system in state-space form can be
described as
obtained
X
na nX
b�1 2 3
f a ðkÞ ¼ Ai f a ðk � iÞ þ Bi nðk � iÞ (5) U2 U2
0
x ¼ Ax ¼ 4 A s þ Bs Da Cs B s C a 5x (8)
i¼1 i¼0 pl pl
Ba Cs Aa
One advantage of ARX model is that the system
response at any time step f a ðkÞ is a linear combina- Then, the critical non-dimensional flutter speed U
tion of past inputs nðk � iÞ and outputs f a ðk � iÞ, so and flutter frequency ratio x=xa can be obtained by
that this model is easy to establish the ROM mathe- conventional stability analysis, i.e., solving the eigen-
matically. With an assumed model order consisting of value of A in equation (8) at different air speeds.
4
664 Proc
Proc IMechE
IMechE Part
Part G: G: J Aerospace
J Aerospace Engineering
Engineering 0(0)
235(6)

The morphing wing mechanism

(a)
The original morphing technology was developed in
the DARPA Smart Wing project.25,26 A new morph-
ing mechanism called Eccentric Beam Actuation
Mechanism (EBAM) for both leading and trailing
edges has been designed, analysed and tested in pre-
vious research23,24,27–29 as shown in Figure 2(a). By
rotating the EBAM at one end using an actuator, the (b)
wing shape is forced to deform depending on the
EBAM rotation angle as illustrated in Figure 2(b)
with the curvature of the chord line shown in
Figure 2(c) to fulfill a specified morphing shape. (c)
The EBAM is connected to the skin through discs
and stringers, which provide a pass-way to transfer
the actuating force to the skin.
Taking the morphing trailing edge for example, the
bending angle hðxÞ varies from 0 at the rear spar to
hTe at the trailing edge, as displayed in Figure 2(b). In Figure 2. A morphing wing: (a) leading and trailing edge
the same way, hLe is defined as the deflection angle at device, (b) airfoil shape, (c) deformed chord line.
the leading edge. Note that hLe and hTe are used to
measure the deformation of morphing wing in the
following sections. As presented in Figure 2(c), the simulations. For spatial discretization, the second-
horizontal and vertical displacements of the ith order upwind scheme is implemented to interpolate
point along the morphing wing chord-line can be the convection terms. In terms of temporal discretiza-
obtained from the geometric relationship described tion, a technique called bounded second order implic-
in equation (9). The position of the deformed upper it time integration is employed for real-time
and lower skin follows the original thickness distribu- advancement.
tion along the morphed chord-line. As for the upper A Radial Basis Functions (RBF) interpolation for
and lower skins, the displacements are taken at the large mesh deformation30 is implemented to enhance
point with the same x-coordinate on chord-line in the the capability of mesh deformation in ANSYS Fluent
current study. via user-defined function (UDF). The RBF interpola-
tion sðxÞ, representing the displacement of the CFD
8 � � � �
>
> hTe hTe mesh,30,31 can be expressed by a sum of base functions
>
< uiþ1 ¼ ui þ jxiþ1 � xi jsin xi tan xi
� Lm � Lm
X
nbd
>
> hTe sðxÞ ¼ ai /ðx � xbi Þ (10)
>
: viþ1 ¼ vi þ jxiþ1 � xi jsin L xi
m i¼1

(9)
where xbi is the centre for RBF, describing the dis-
where ui and vi are the horizontal and vertical dis- placement of boundary nodes, and nbd is the number
placement of the ith point with x-coordinate of xi , of boundary node, | | is the norm biasing.
and Lm is the length of morphing device. Note that qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u1 ¼ 0 and v1 ¼ 0. jx � xi j ¼ ðx � xi Þ2 þ ðy � yi Þ2 =R (11)

Isogai wing model where R is the support radius, and 10c is adopted
ANSYS Fluent was used to model the aerodynamics herein. / is the basis function, and Wendland’s C2
based on the fluid governing equations, in which the function is implemented in the current investigation
Euler and Navier-Stokes equations were solved, (
respectively, by using the pressure-based coupled
/ðjxjÞ ¼ ð1 � jxjÞ4 ð4jxj þ 1Þ; jxj < 1 (12)
algorithm in the current paper. In Fluent, a control- 0; jxj � 1
volume-based technique is employed to convert the
general scalar transport equation to an algebraic
When the motion of nodes on � the boundary,
�Ti.e.,
equation, which is solved by using a point implicit
(Gauss-Seidel) linear equation solver in conjunction d b , is specified, coefficients a ¼ a1 . . . anbd can
with an algebraic multigrid (AMG) method. To deal be obtained by the inverse of equation (10)
with viscous flow problems, a classical one-equation
turbulence model, the S-A model, is used in current a ¼ d b M �1
b;b (13)
He et al.
He et al. 5
665

where M b;b is a nbd � nbd matrix containing the eval- feasibility of the present methods. It is also found
uations of the basis function /bi ;bj ¼ /ðxbi � xbj Þ. The that the so-called S shape flutter boundary observed
displacement of all remaining nodes can be deter- in the Euler computational results disappears for
mined by equation (10). Navier-Stoke solutions.
The Isogai Wing model is of NACA 64A010 airfoil
with the following parameters3,4
Transonic flutter of a morphing wing
xh
a ¼ �2:0; xa ¼ 1:8; ra ¼ 1:8655; ¼ 1:0;
xa Flutter boundary of Isogai wing with
l ¼ 60: morphing devices
The study was then extended to the Isogai wing
For a specified Mach number, ‘3211’ signal, as model with morphing leading edge (LE) and
shown in Figure 3(a), is used as the inputs of CFD trailing edge (TE) to investigate the transonic flutter
solver, and the corresponding coefficients of lift and characteristics. To focus on the effect of the aerody-
moment can be obtained. Subsequently, the aerody- namic profile variation of the morphing wing,
namic ROM is established for this Mach number. the mass and stiffness distribution of the morphing
Typical time history of the aerodynamic coefficients wing keeps in the same values as the original
at Mach 0.8 from CFD simulation and the aerody- Isogai wing.
namic ROM is shown in Figure 3(b). The interface location of the LE and TE with the
Based on the aerodynamic ROM, the root locus wing box is set at 0:3c and 0:6c chord of the morphing
for different Mach numbers are used to determinate wing, respectively. Figure 5(a) shows the deformed
the critical flutter speed and flutter frequency. After airfoil profile of morphing TE and LE with different
depicting these results versus Mach numbers, the
deflection angles forming a smooth aerodynamic sur-
resulting flutter boundary in transonic regime is
face. The aerodynamic pressure distributions were
shown in Figure 4.
obtained at different Mach numbers based on Euler
For the Euler solutions as shown in Figure 4(a)
equations for the original and the morphing TE as
and (b), the results are in good agreement with
shown in Figure 5(b) and (c). Smooth pressure distri-
those obtained by transonic flutter solution in the fre-
quency domain32 and time marching solutions.17,33,34 bution was also obtained for the morphing LE case
The results show that the flutter speeds in transonic despite some fluctuation is observed near the LE as
regime are lower than in the subsonic regime and shown in Figure 5(d).
there are multiple values of flutter speed occurring The effect of morphing TE and LE on the transon-
between Mach 0.85 and 0.9, which forms the so- ic characteristic of the wing can be considered via the
called S shape flutter boundary.35 aerodynamic ROM as well. By setting the morphing
Only a few studies of Isogai wing model in viscous TE or LE at a specific deflection angle with zero mean
flow are available in the existing literature. angle of attack, the generalized aerodynamic forces
Aerodynamic ROM method are utilized to obtain due to ‘3211’ signal for the deformed airfoil shape
the flutter solution at different Mach numbers as can be obtained from CFD solver. The aerodynamic
shown in Figure 4(c) and (d). Compared with the ROM can then be established at different Mach num-
existing time marching solutions,33,34 reasonably bers. Subsequently, the process for searching the crit-
good agreements are achieved demonstrating the ical flutter condition over a range of flight speed was

Figure 3. Comparison of aerodynamic force coefficient from CFD solver and ARX ROM for Isogai wing model at Mach 0.8 (Euler
solution): (a) ‘3211’ signal, (b) lift and aerodynamic moment coefficients.
6666 Proc IMechE
Proc Part
IMechE G: G:
Part J Aerospace Engineering
J Aerospace 235(6)
Engineering 0(0)

Figure 4. Comparison of flutter boundary of Isogai wing model: flutter speed (a) and flutter frequency (b) versus Mach number from
inviscid solution; flutter speed (c) and flutter frequency (d) versus Mach number from viscous solution.

Figure 5. (a) Airfoil profile for morphing TE and LE and pressure distributions for (b) wing without morphing, (c) morphing TE
hte ¼ 5� , (d) morphing LE hle ¼ 5� for Euler solution.

executed to determine the transonic flutter character- angles based on Euler and Navier-Stokes equations.
istics of the morphing wing. For the Euler results, for 5� morphing TE, the flutter
Figure 6 shows the transonic flutter boundary of the speed drops with respect to the original Isogai wing
wing with morphing LE and TE at different deflection between Mach 0.725 and 0.81, whilst flutter speed
He et al.
He et al. 7
667

Figure 6. Comparison of flutter boundary of Isogai wing with the morphing wing based on: (a) Euler equation, (b) Navier-Stokes
equations.

increases above Mach 0.82. Conversely, the morphing is defined as the freestream Mach number at
LE at 2� and 5� has a beneficial effect on the which shock waves on both upper and lower
flutter characteristic in a small Mach number range surfaces reach the TE of an airfoil.36,37 And these
below 0.85, but exhibits an adverse effect above phenomena have been also observed in our previous
Mach number 0.85. It is evident that the effect of numerical cases.38
morphing LE on transonic flutter is much less than In order to give physical-based explanations of
the TE. morphing wing on transonic flutter, the relationship
From Figure 6(a), it is apparent that the minimum between the shock wave locations on airfoil surface,
non-dimensional flutter speed is about 5.45 for the aerodynamic coefficient and the flutter characteristic
morphing TE model with deflection angle of 5� in is addressed.
the transonic region. Theoretically the transonic dip To evaluate the unsteady characteristic of the tran-
can be further increased to 5.53 with a proper morph- sonic aerodynamics, the aerodynamic influence coef-
ing law, i.e. deploying morphing TE from Mach 0.81, ficient20 is employed in the present study. The input of
increasing by 36.9% compared to original Isogai CFD, i.e. the pitching motion of the aeroelastic air-
wing. The results indicate that employing a morphing foil, is taken as a sinusoid function
TE is feasible and effective to improve the transonic
flutter characteristic. aðtÞ ¼ a0 sinðxtÞ (14)
Similar effects of morphing LE and TE on the flut-
ter speed are observed for Navier-Stokes solutions where a0 denotes the amplitude and x is the frequen-
as shown in Figure 6(b). Ideally, the minimum cy. The corresponding output of CFD, taking aero-
flutter speed in transonic regime can be increased dynamic moment coefficient as an example, is
from 4.82 of the original model to 8.41 with 5� periodic which can be written as
morphing TE. And the effect of morphing LE on
bottom of flutter boundary is much less than that of cm ðtÞ ¼ ðcm Þc cosðxtÞ þ ðcm Þs sinðxtÞ
morphing TE. ¼ ðcm Þcs sinðxt þ uÞ (15)
From linear aerodynamics,3,4 the flutter speed gen-
erally decreases in transonic regime below Mach 1,
and increases significantly above Mach 1. The current As for the coefficients ðcm Þc and ðcm Þs can be
results, as shown in Figure 6, present the same trend obtained by fitting the time history of aerodynamic
for both original and morphing wings. To the moment coefficient to the above equation. The aero-
authors’ knowledge, the exact Mach number, in dynamic influence coefficient can be expressed as
which the flutter speed starts to increase in the tran-
sonic regime, has not yet been fully understood. For ðcm Þcs iu ðcm Þcs  
cma ¼ e ¼ cosðuÞ þ isinðuÞ
that reason, it is difficult to give a general conclusion a0 a0
on flutter characteristics over transonic wings. ¼ Rðcma Þ þ iIðcma Þ (16)
According Bendiksen,36,37 the Mach number freeze
phenomenon, or transonic stabilization law, plays an Then the aerodynamic influence coefficient of dif-
important role in the wing transonic flutter. Flutter ferent reduced frequencies (k ¼ xb=V) at different
speed may increase significantly when the freestream Mach numbers are calculated based on Euler equa-
Mach number reaches the freeze Mach number. It tions, as displayed in Figure 7.
8
668 Proc
Proc IMechE
IMechE Part
Part G: G: J Aerospace
J Aerospace Engineering
Engineering 0(0)
235(6)

Figure 7. Aerodynamic influence coefficient due to pitching oscillation around a ¼ �2 as a function of Mach number and reduced
�
frequency (k ¼ xb=V) for: (a) NACA 64A010 airfoil, (b) NACA 64A010 with morphing trailing edge deflection 5 .

From Figure 5(b), according to the Euler solution, characteristics, the analysis was conducted at differ-
it can be observed that the freeze Mach number is ent AoA in this section. Special attention was paid to
around 0.9 for the original NACA 64A010 airfoil the relationship of the appearance of the strong shock
according to the location of shock wave. When wave on the upper surface of the airfoil and the tran-
Mach number is greater than freeze Mach number, sonic flutter boundary. It should be noted that Euler
the unsteady aerodynamic moment coefficient equations are applied in this section.
become stable as clearly observed in Figure 7(a). Similar to a morphing wing, the AoA effect on the
Meanwhile, a strong stabilization of the flutter wing transonic characteristic can be considered via
boundary occurs as shown in Figure 6(a), which coin- the establishment of aerodynamic ROMs. The AoA
cides with the observation in Ref. 36. of the Isogai wing model was set to a specific value in
Following the Mach number freeze phenomenon for CFD solver when calculating aerodynamic coeffi-
the undeformed airfoil, the shock wave for the morphed cients due to ‘3211’ signals. Then the wing transonic
�
airfoil with a TE deflection 5 initially appears near flutter characteristics at this AoA was obtained by
Mach 0.75, and became stronger on the upper surface eigenvalue analysis.
at Mach 0.775 and subsequently moved backward Since Euler equations are adopted, the AoA is
towards TE at Mach 0.85 as shown in Figure 5(c). restricted to a small range from 0� to 4� in the current
Two interesting things are noted. Firstly, when a study. The resulting pressure coefficient distributions
strong shock wave forms on the upper surface of the at different AoAs are presented in Figure 8, which
airfoil, the flutter speed stops decreasing and starts to shows the Mach number when the shock wave on
increase with increasing Mach number. Secondly, when the upper surface approaches the TE. At AoA ¼ 3� ,
the shock wave reaches the TE, another freeze region for instance, a local supersonic flow appears on the
emerges at a lower Mach number around 0.85 as shown upper surface at Mach number near 0.7, and a strong
in Figure 7(b). Consequently, the unsteady aerodynam- shock wave comes apparent at Mach 0.8, and moves
ic force over the airfoil stabilizes and can lead to an backward to TE around Mach 0.84.
increase of the flutter speed, which is consistent with Figure 9 shows the flutter boundary of the original
the flutter boundary shown in Figure 6(a). Isogai wing model at different AoAs. It is noted that
In summary, a possible explanation in improving the bottom of the flutter boundary moves up as AoA
the transonic flutter characteristic is that employing increases. For small AoAs between 1� and 2� , the
the morphing TE makes shock wave and Mach AoA effect on flutter speed at the transonic dip is
number freeze region occur at an early Mach number. unapparent. This might be caused by a strong shock
wave occurring on the upper surface around Mach
Appearance of shock wave and transonic 0.8. When the shock wave moves close to TE, how-
ever, the Mach number remains around Mach 0.9
flutter characteristic similar to the case of AoA ¼ 0� . When the AoA is
From previous results, it can be seen that the shock further increased, remarkable change of transonic
wave location plays a vital role in the transonic flutter flutter boundary takes place. Taking AoA ¼ 3� for
characteristics. A more general approach to alter the example, the downward trend of flutter speed starts
appearance of a shock wave is to vary the angle of to reverse at Mach 0.775, which is consistent with the
attack (AoA) of the airfoil. To study the effect of appearance of a strong shock wave on the upper sur-
morphing devices on the transonic flutter face. When the shock wave moves close to the TE, the
He et al.
He et al. 9
669

Figure 8. Pressure distribution across transonic region at angle of attack of (a) 1� , (b) 2� , (c) 3� , (d) 4� .

Another approach to evaluating the effect of the

shock wave on the flutter boundary can be made by
altering the AoA at the same Mach number. As the
AoA increases at Mach 0.8, the flutter speed decreases
slightly at small AoAs and then recovers, as shown in
Figure 10(a). When the AoA is larger than 2� , a
strong shock wave emerges on the upper surface as
shown in Figure 10(b). Meanwhile, the downwards
trend of flutter speed stops and starts to grow
gradually.
As shown in Figure 11, the scenario at Mach 0.84
is slightly different from Mach 0.8 when a strong
shock wave has already been established at
AoA¼0� , and the shock wave approaches to TE at
AoA¼3� . The flutter speed increases with the AoA
from 0� as shown in Figure 11(a). This is consistent
with the trend of flutter speed over AoA at Mach 0.8
after a strong shock emerges. When the shock wave
Figure 9. Flutter boundary of Isogai wing model at different moves to TE at AoA¼3� , significant improvement of
angle of attack.
flutter speed is observed.
It is noted that the flutter boundary varied with the
flutter speed increases significantly around Mach 0.84 AoA and formed an S shape as shown in Figures 10
showing a big difference from the AoA¼0� case. (a) and 11(a). This was also observed from the con-
Yates et al.39 also noted that the growth of the super- ventional flutter boundary corresponding to Mach
sonic area on the wing surface causes the aerodynamic number in Isogai wing model section. When the
center of the wing move backward, and the flutter speed AoA is increased at a fixed Mach number, the flow
increases with the AoA. Moreover, the numerical sim- velocity on the upper surface of the airfoil is acceler-
ulation by Edwards et al.40 indicates that the flutter ated. A similar effect on the flow field exists
speed rises when the shock wave forms on the upper by increasing the Mach number at a fixed AoA.
surface. The results obtained in the current study are When the AoA is between 0� and 1� in Mach 0.84,
consistent with the above observations. multiple flutter speeds take place as shown in
10
670 Proc
Proc IMechE
IMechE Part
Part G: G: J Aerospace
J Aerospace Engineering
Engineering 0(0)
235(6)

Figure 10. Effect of AoA on the flutter speed for Isogai wing model at Mach number 0.8: (a) The flutter speed at different AoA,
(b) Steady pressure coefficient distribution at different AoA.

Figure 11. Effect of AoA on the Isogai wing flutter speed at Mach number 0.84: (a) flutter speed at different AoA, (b) steady pressure
coefficient distribution at different AoA, real part of aeroelastic eigenvalue versus non-dimensional air speed at (c) AoA ¼ 0:5� ,
(d) AoA ¼ 2� , (e) AoA ¼ 3� , (f) AoA ¼ 3:5� .
He et
He et al.
al. 11
671

Figure 12. Aeroelastic response of morphing wing with different morphing speeds: (a) plunging motion and (b) pitching motion at
U ¼ 4.5 of Mach 0.85, (c) plunging motion and (d) pitching motion at U ¼ 4.6 of Mach 0.875. (bTE ¼ 5� starts at s ¼ 100).

Figure 11(c). After AoA is increased slightly above aerodynamic configuration is being altered during
1:5� , only a single flutter speed is detected due to this process, which may lead to unexpected influences
the change of aerodynamic damping caused by on the aeroelastic behavior. For that reason, a time
the varying AoA. Thus, an isolated stable region domain analysis considering TE morphing process is
is observed in the high speed area at low AoAs in performed to assess these effects, which is conducted
Figure 11(a). When the AoA is further increased by the time marching method based on CFD.
to greater than a specific AoA 3� , a jump of the In the following simulations, the morphing in the
flutter speed is exhibited. This is caused by the TE starts at s ¼ 100 with a constant morphing speed
transition of ‘hump mode’ as shown in Figure 11(e) dbTE � �
ds varying from 0:05 to 0:25 , and the final deflec-
and (f). �
tion angle (bTE ) is set to 5 . The resulting aeroelastic
In summary, the above results reveal that by vary- responses of the system are presented in Figure 12. It
ing Mach number, AoA or morphing TE deflection is observed that the morphing TE has a stable effect
angle, the appearance of a strong shock wave on the on the aeroelastic system, since the divergent motion
upper surface of the airfoil leads to a reverse of the turns to convergent when the morphing TE is
downward trend of flutter speed. When the shock employed. In addition to this, it was also found that
wave on the upper surface reaches to the TE, the flut- the morphing speed has negligible influence on the
ter speed starts to increase significantly. Thus, the stability of aerodynamic system, and only affects the
mechanism of improving the transonic flutter charac-
transition period of the motion shifting from diver-
teristics by morphing TE is due to the shock wave
gence to convergence.
emerging and moving to the TE at an early stage
Also observed in Figure 12, the static aeroelastic
with lower Mach number. It is also note that the
deflections of pitching and plunging motions con-
morphing LE has negligible effect on the shock
verge to non-zero positions when the TE morphing
wave and the flutter boundary in the transonic dip
is deployed. In particular, a negative mean AoA
region.
occurs because of the negative pitching aerodynamic
moment induced by employing the morphing TE.
Time domain transonic flutter analysis of a From previous section, it is known that the AoA
morphing wing can influence the flutter characteristic of the aeroelas-
Due to the fact that morphing change requires a tic system. Thus, the flutter speed obtained from the
period of time to be implemented in structure, the time marching method in this section is a result of
672
12 Proc IMechE
Proc Part
IMechE G: G:
Part J Aerospace Engineering
J Aerospace 235(6)
Engineering 0(0)

characteristic corresponding to AoA and Mach

number for Euler solutions.
The relationship between shock wave locations on
transonic flutter characteristics for the present model
shows that the transonic flutter boundary depends on
the forming and location of the shock wave on the
upper surface of the airfoil. Corresponding to Mach
number, the flutter speed will revise the downward
trend and increase significantly following a strong
shock wave taking place and moving close to the TE.
It is observed that the morphing TE provides an
effective way to improve the transonic characteristics.
For the presente wing model, the morphing TE leads
to an increase of the transonic flutter speed by 36.9%
based on Euler equation or 74.5% based on Navier-
Stokes equation. The study reveals the a possible
mechanism of the improvement of transonic flutter
Figure 13. Flutter boundary of Isogai wing with the morphing characteristics of morphing wing is that it makes
trailing edge from time marching method. shock wave and Mach number freeze region occur
at a lower Mach number. However, the morphing
both mean AoA and morphing TE. Note the results LE has negligible effect on the transonic flutter char-
in this section are different from Figure 6 where only acteristics because of its weak influence to the shock
deformed shape is considered and zero mean AoA is wave taking place near the TE.
assumed. Time domain analysis show that that the morphing
The flutter speed boundary obtained by time speed has negligible influence on the stability of aero-
marching approach is displayed in Figure 13, and dynamic airfoil, and the minimum flutter speed of
the flutter speeds for the original Isogai wing and flutter boundary is improved by 16.8% with respect
to the original wing. The effective Mach region for
morphing wing with constant mean AoA are pre-
employing morphing TE to improve transonic flutter
sented. It is observed that the minimum flutter
characteristic narrows, which is caused by the change
speed of flutter boundary is improved by 16.8% to
of mean AoA.
4.72 with respect to the original Isogai wing. During
the current analysis, it was noted that the mean AoA
� Declaration of conflicting interests
induced by morphing TE is about �2 from Mach 0.8
to 0.85. So the flutter boundary of the morphed wing The author(s) declared no potential conflicts of interest with
�
at mean AoA ¼ �2 is calculated, and the minimum respect to the research, authorship, and/or publication of
flutter speed agrees well with that of time marching this article.
analysis, as shown in Figure 12. Comparing to the
Funding
results in Figure 6, the minimum flutter speed
improvement in this section is less, and the effective The author(s) disclosed receipt of the following financial
Mach region to increase flutter speed narrows, which support for the research, authorship, and/or publication
is caused by the change of mean AoA induced by of this article: This work is supported by the National
morphing TE. Natural Science Foundation of China (Grant 11972079)
and Innovate UK.

Conclusion ORCID iDs

The transonic flutter characteristic of a 2D airfoil Shun He https://orcid.org/0000-0002-1780-2945
Shijun Guo https://orcid.org/0000-0002-5400-6243
system with morphing trailing and leading edges is
studied. The Fluent software based on Euler and
Navier-Stokes equations is employed with an RBF Reference
interpolation model to improve the capability of 1. Li D, Zhao S, Ronch AD, et al. A review of modelling
mesh deformation in the CFD model via UDF. and analysis of morphing. Prog Aerosp Sci 2018; 100:
An ARX model is implemented to establish aerody- 46–62.
2. Burdette DA and Martins JRRA. Design of a transonic
namic ROM with the GA used to identify the model
wing with an adaptive morphing trailing edge via aero-
order. The critical flutter speed and frequency can be structural optimization. Aerosp Sci Technol 2018; 81:
obtained by solving the eigenvalue equations at spe- 192–203.
cific flow conditions. The flutter boundary can be sub- 3. Isogai K. On the transonic-dip mechanism of flutter of
sequently obtained with a type of S shape flutter a sweptback wing. AIAA J 1979; 17: 793–795.
He et al.
He et al. 13
673

4. Isogai K. Numerical study of transonic flutter of a two- variable camber wing. In: 53rd AIAA/ASME/ASCE/
dimensional airfoil. Tokyo, Japan: National Aerospace AHS/ASC Structures, structural dynamics and materials
Laboratory, 1980. conference, Honolulu, Hawaii, 23–26 April 2012.
5. Huang C, Yang C, Wu Z, et al. Variations of flutter 24. Li D, Guo S and Xiang J. Modeling and nonlinear
mechanism of a span-morphing wing involving rigid- aeroelastic analysis of a wing with morphing trailing
body motions. Chinese J Aeronaut 2018; 31: 490–497. edge. Proc IMechE, Part G: J Aerospace Engineering
6. Pecora R, Amoroso F, Dimino I, et al. Aeroelastic sta- 2013; 227: 619–631.
bility analysis of a large civil aircraft equipped with 25. Kudva JN, Martin CA and Scherer LB. Overview of the
morphing winglets and adaptive flap tabs. Act Passiv DARPA/AFRL/NASA smart wing program. In: Smart
Smart Struct Integr Syst XII 2018; 10595: 1–13. structures and materials 1999: industrial and commercial
7. Hu W, Yang Z, Gu Y, et al. The nonlinear aeroelastic applications of smart structures technologies. 1999, pp.
characteristics of a folding wing with cubic stiffness. 230–237. SPIE Smart Structures and Materials.
J Sound Vib 2017; 400: 22–39. 26. Bartley-Cho JD, Wang DP, Martin CA, et al.
€
8. Unlüsoy L and Yaman Y. Aeroelastic behaviour of Development of high-rate, adaptive trailing edge con-
UAV wings due to morphing. Aircraft Eng Aerosp trol surface for the smart wing phase 2 wind tunnel
Tech 2017; 89: 30–38. model. J Intell Mater Syst Struct 2004; 15: 279–291.
9. Kan Z, Li D, Xiang J, et al. Delaying stall of morphing 27. Perera M and Guo S. Optimal design of a seamless
wing by periodic trailing-edge deflection. Chinese J aeroelastic wing structure. In: 50th AIAA/ASME/
Aeronaut 2020; 33: 493–500. ASCE/AHS/ASC structures, structural dynamics, and
10. Xiang J, Liu K, Li D, et al. Unsteady aerodynamic materials conference, 17th AIAA/ASME/AHS adaptive
characteristics of a morphing wing. Aircraft Eng structures conference, 11th AIAA, Palm Springs,
Aerosp Tech 2019; 91: 1–9. California, 4–7 May 2009, p. 2195.
11. Ajaj RM and Friswell MI. Flutter of compliant span 28. Di Matteo N, Guo S, Ahmed S, et al. Design and anal-
morphing wings. Smart Mater Struct 2018; 27: 1–16. ysis of a morphing flap structure for high lift wing. In:
12. Li W and Jin D. Flutter suppression and stability anal- 51st AIAA/ASME/ASCE/AHS/ASC Structures, struc-
ysis for a variable-span wing via morphing technology. tural dynamics, and materials conference, 18th AIAA/
J Sound Vib 2018; 412: 410–423. ASME/AHS adaptive structures conference, 12th
13. Ajaj RM and Jankee GK. The transformer aircraft : a AIAA, Orlando, Florida, 12–15 April 2010, p. 3096.
multimission unmanned aerial vehicle capable of sym- 29. Perera M and Guo S. Structural and dynamic analysis
metric and asymmetric span morphing. Aerosp Sci of a seamless aeroelastic wing. In: 51st AIAA/ASME/
Technol 2018; 76: 512–522. ASCE/AHS/ASC Structures, structural dynamics, and
14. Opgenoord MM, Drela M and Willcox KE. Influence materials conference, 18th AIAA/ASME/AHS adaptive
of transonic flutter on the conceptual design of next- structures conference, 12th AIAA, Orlando, Florida,
generation transport aircraft. AIAA J 2019; 57: 12–15 April 2010, p. 2878.
1973–1987. 30. De Boer A, der Schoot MS and Bijl H. Mesh deforma-
15. Lv B, Lu Z, Guo T, et al. Investigation of winglet on the tion based on radial basis function interpolation.
transonic flutter characteristics for a wind tunnel test Comput Struct 2007; 85: 784–795.
model CHNT-1. Aerosp Sci Technol 2019; 86: 430–437. 31. Rendall TCS and Allen CB. Unified fluid–structure
16. Dowell EH, Clark R, Cox D, et al. A modern course in interpolation and mesh motion using radial basis func-
aeroelasticity. 4th ed. Dordrecht: Kluwer Academic tions. Int J Numer Meth Eng 2008; 74: 1519–1559.
Publishers, 2004. 32. He S, Yang Z and Gu Y. Transonic limit cycle oscilla-
17. Alonso J and Jameson A. Fully-implicit time-marching tion analysis using aerodynamic describing functions
aeroelastic solutions. In: 32nd Aerospace science meeting and superposition principle. AIAA J 2014; 52:
and exhibit, 1994. DOI: 10.2514/6.1994-56. 1393–1403.
18. Rausch RD, Batina JT and Yang HTY. Three- 33. Prananta BB, Hounjet MHL and Zwaan RJ. Two-
dimensional time-marching aeroelastic analyses using dimensional transonic aeroelastic analysis using thin-
an unstructured-grid Euler method. AIAA J 1993; 31: layer Navier–Stokes method. J Fluids Struct 1998; 12:
1626–1633. 655–676.
19. Cowan TJ, Arena AS and Gupta KK. Accelerating 34. Yang S, Zhang Z, Liu F, et al. Time-domain aeroelastic
computational fluid dynamics based aeroelastic predic- simulation by a coupled Euler and integral boundary-
tions using system identification. J Aircr 2001; 38: layer method. In: 22nd AIAA applied aerodynamic con-
81–87. ference, Providence, RI, 2004.
20. He S, Yang Z and Gu Y. Nonlinear dynamics of an 35. Badcock KJ, Timme S, Marques S, et al. Transonic
aeroelastic airfoil with free-play in transonic flow. aeroelastic simulation for instability searches and
Nonlinear Dyn 2017; 87: 2099–2125. uncertainty analysis. Prog Aerosp Sci 2011; 47:
21. Wang X, Kou J and Zhang W. Unsteady aerodynamic 392–423.
modeling based on fuzzy scalar radial basis function 36. Bendiksen OO. Transonic stabilization laws for
neural networks. Proc IMechE, Part G: J Aerospace unsteady aerodynamics and flutter. In: 53rd AIAA/
Engineering 2019; 233: 5107–5121. ASME/ASCE/AHS/ASC Structures, structural dynam-
22. Zhang W and Ye Z. Reduced-order-model-based flutter ics and materials conference, 20th AIAA/ASME/AHS
analysis at high angle of attack. J Aircr 2007; 44: adaptive structures conference, 14th AIAA, Honolulu,
2086–2089. Hawaii, 23–26 April 2012, p. 1718.
23. Di Matteo N, Guo S and Morishima R. Optimization 37. Bendiksen OO. Transonic flutter characteristics of
of leading edge and flap with actuation system for a advanced fighter wings. In: 56th AIAA/ASCE/AHS/
14
674 Proc
Proc IMechE
IMechE Part
Part G: G: J Aerospace
J Aerospace Engineering
Engineering 0(0)
235(6)

ASC Structures, structural dynamics, and materials con- 39. Yates EC, Wynne EC and Farmer MG. Effects of angle
ference, Kissimmee, Florida, 5–9 January 2015, p. 438. of attack on transonic flutter of a supercritical wing.
38. Yang Z, He S and Gu Y. Transonic limit cycle oscilla- J Aircr 1983; 20: 841–847.
tion behavior of an aeroelastic airfoil with free-play. 40. Edwards JW, Bennett RM, Whitlow W, et al. Time-
J Fluids Struct 2016; 66: 1–18. marching transonic flutter solutions including angle-
of-attack effects. J Aircr 1983; 20: 899–906.