aerospace

Article
Effect of Aerodynamic Damping Approximations on
Aeroelastic Eigensensitivities
Christoph Kaiser* and David Quero

German Aerospace Center (DLR), Institute of Aeroelasticity, Bunsenstraße 10, 37073 Göttingen, Germany;
[email protected]
* Correspondence: [email protected]

Abstract: Aeroelastic sensitivities for the flutter solution are a crucial component of the multi-
disciplinary optimization methods employed in modern aircraft design. This paper derives the
aeroelastic sensitivities for different aerodynamic damping approximations—the p-k method, the
g method and the generalized aeroelastic analysis method—highlighting the influence of the em-
ployed aerodynamic approximation on the required derivatives. The derived formulation for the
determination of aeroelastic sensitivities by means of a direct method is verified for the case of a
two-degree-of-freedom typical section model, where analytical aeroelastic sensitivities can be analyti-
cally obtained. For this and for an additional model, namely the AGARD 445.6 weakened wing, the
significant effect of the aerodynamic damping approximation on the aeroelastic sensitivities is shown.

Keywords: aeroelasticity; aerodynamic damping; flutter sensitivities; eigensensitivities

1. Introduction



The design of new aircraft entails a demand for high efficiency, which eventually is


achieved by extensive multi-disciplinary design optimization (MDO) while considering
Citation: Kaiser, C.; Quero, D. Effect
constraints on stability and safety. Therefore, the MDO must ensure the aircraft to be free
of Aerodynamic Damping
of flutter conditions [1] within the flight envelope and the aeroelastic stability margin as
Approximations on Aeroelastic
required by certification. Aeroelastic stability is typically expressed by the eigenvalues of
Eigensensitivities. Aerospace 2022, 9,
the aeroelastic eigenproblem formulated in the Laplace domain. Hence, the eigensensi-
127. https://doi.org/10.3390/
tivities of the aeroelastic eigenproblem facilitate the design process for compliance with
aerospace9030127
flutter constraints as they describe the change in stability by changing design parameters.
Academic Editor: Rosario Pecora This applies especially for gradient-based MDO, where sensitivities are required to fulfill
Received: 31 January 2022
imposed constraints and obtain the next iterated design [2,3].
Accepted: 27 February 2022
This paper follows up on the work of Cardani and Mantegazza [4], as well as Bindolino
Published: 1 March 2022
and Mantegazza [5], who have thoroughly described the computation of eigenvalue and
eigenvector derivatives for flutter eigenproblems. Murthy and Haftka [6] have given
Publisher’s Note: MDPI stays neutral
a detailed description on solving for the eigensensitivities using the direct and adjoint
with regard to jurisdictional claims in
methods. Additionally, the commercial solver MSC Nastran provides the computation of
published maps and institutional affil-
eigenvalue sensitivities [7], based on [8], employing the doublet lattice method (DLM) [9].
iations.
Extending the achievements made so far, this work addresses the influence of aero-
dynamic damping approximations on the aeroelastic eigensensitivities. Unsteady aerody-
namic methods providing the air loads for the aeroelastic eigenproblem by the aerodynamic
Copyright: © 2022 by the authors.
transfer function matrix typically account for purely harmonic structural oscillations. There-
Licensee MDPI, Basel, Switzerland. fore, the present damping of the oscillations as described by the real part of the eigenvalues
This article is an open access article is not captured in the aerodynamic loads. This shortcoming affects the solution of the
distributed under the terms and aeroelastic eigenproblem, leading to several aeroelastic solution methods. The well-known
conditions of the Creative Commons p-k method provides the flutter solution for purely harmonic air loads by iteratively match-
Attribution (CC BY) license (https:// ing the frequencies of the oscillations and the air loads [10]. The g method developed by
creativecommons.org/licenses/by/ Chen [11] extends the p-k method for small values of the real part of the eigenvalues by pro-
4.0/). viding a first-order approximation of the analytic continuation of the aerodynamic transfer

Aerospace 2022, 9, 127. https://doi.org/10.3390/aerospace9030127 https://www.mdpi.com/journal/aerospace

Aerospace 2022, 9, 127 2 of 21

function matrix. Edwards et al. [12] showed that, for linear potential flow, the complete
analytic continuation is valid for the aeroelastic analysis and termed it the generalized
aeroelastic analysis method (GAAM).
Existing work regarding the computation of aeroelastic eigensensitivities in the context
of gradient-based MDO has typically focused on the precision of the derivatives involved,
obtained either analytically by automatic differentiation or numerically by finite differences
or complex step formulas [13], but using the common p-k approximation when computing
the aerodynamic damping term. This work focuses on the effect that different levels of
physical approximation used for the determination of the aerodynamic damping term may
have on the aeroelastic eigensensitivites, as this influence may overtake the approximation
error introduced by the numerical method used for the computation of the derivatives.
Thus, the present work focuses on differentiating the nonlinear aeroelastic eigenproblem,
identifying the involved sensitivities of the structural dynamics and different aerodynamic
damping approximations. To this aim, the aerodynamic sensitivities for the p-k method,
the g method and GAAM are derived, which allows us to accurately obtain the aeroelastic
eigensensitivities for each method and display the difference in the resulting eigenvalue sen-
sitivities. This is demonstrated for a typical section model as well as for the AGARD 445.6
wing. Furthermore, the derived aeroelastic sensitivities are verified for the typical section
model employing Theodorsen’s aerodynamic airfoil theory by obtaining the analytical
derivatives.
In Section 2.1, the nonlinear aeroelastic eigenproblem is introduced, as well as the
employed solution method in Section 2.2, and the investigated aerodynamic damping
approximations are presented in Section 2.3. Sections 3.1 and 3.2 describe the aeroelastic
sensitivities and their solution by the direct method in physical and modal coordinates
employing GAAM. The aerodynamic sensitivities of the p-k and g methods are derived
in Sections 3.3 and 3.4. Section 3.5 presents the sensitivity of the modal matrix for em-
ploying modal coordinates. The verification for the typical section model for the three
approximations is shown in Sections 4.1 and 4.2, showing the eigenvalue sensitivities for
the AGARD 445.6 wing with varying sweep obtained with the p-k and g method. For
the verification, the derivatives of the generalized Theodorsen aerodynamic airfoil theory
are presented together with the derivatives of the generalized Theodorsen function in
Appendices A and B.

2. Aeroelastic Stability
2.1. Aeroelastic Eigenvalue Problem
The aeroelastic governing equations couple the structural dynamics with the aero-
dynamic forces by second-order differential equations resulting from Newton’s second
law. For obtaining linear stability, complex harmonic motion x̂est around a steady state is
assumed. This transforms the second-order differential equations into the Laplace domain
with the Laplace variable s = σ + iω as a system of nonlinear, algebraic equations

(s2 M + sD + K − A(s)) x̂ = G (s) x̂ = 0 , (1)

where the unsteady aerodynamic forces are described in a time-linearized manner by the
aerodynamic transfer function matrix A(s) multiplied by the amplitude of motion x̂. The
complex-valued matrix A(s) especially depends on steady-state flow parameters such
as Mach number and dynamic pressure or Reynolds number according to the employed
aerodynamic method. The mass matrix M, the damping matrix D and the stiffness matrix
K are real symmetric matrices describing the structural properties. Equation (1) is a
complex-valued, nonlinear eigenproblem with eigenvalues s and nonzero eigenvectors x̂.
Consequently, the real part of the eigenvalues indicates growing, σ > 0, or decaying, σ < 0,
oscillations. Hence, the instability onset is given for σ = 0 denoted as the flutter onset.
The aeroelastic governing equations and, thus, Equation (1) are typically applied in
modal coordinates in order to reduce the system size by modal truncation, where only
a small number of structural mode shapes are considered. Therefore, the amplitudes of
Aerospace 2022, 9, 127 3 of 21

the physical displacements x̂ are transformed into the modal participation factors q̂ by the
modal matrix
h i
Φ = · · · ϕj · · · ,

which consists of structural mode shapes ϕ obtained from linear modal analysis of the
structural model; see Section 3.5. The transformed Equation (1) in modal coordinates
results in

(s2 M̃ + s D̃ + K̃ − Ã(s))q̂ = Φ T G (s)Φq̂ = G̃ (s)q̂ = 0 (2)

with the modal mass matrix M̃, modal damping matrix D̃, modal stiffness matrix K̃ and
the generalized aerodynamic force matrix Ã(s). It is noted that, without modal truncation,
the solutions of Equation (2) are identical to those of Equation (1), with the eigenvectors
related by x̂ = Φq̂. Furthermore, the modal approach eases the use of different spatial
discretizations for the structural and aerodynamic disciplines by employing the common
set of modal coordinates.

2.2. Solutions of the Aeroelastic Eigenproblem

A general method for obtaining the solutions of the nonlinear aeroelastic eigenproblem
is by finding the roots of Equations (1) and (2) when considered as nonlinear functions
of the vector of eigenpairs s = [s x̂ T ] T and s̃ = [s q̂ T ] T , respectively. In order to equal
the number of unknowns, this approach requires one additional equation that normalizes
the eigenvectors. The employed equations for the eigenvector normalization are x̂ T W x̂ −
1 = 0 and q̂ T W̃ q̂ − 1 = 0, where the weighting matrices W and W̃ are free to choose
symmetric matrices including the identity matrix; see also Section 3.2. The evaluation of the
aerodynamic transfer function matrix A(s) is directly integrated in the function evaluation,
ensuring the matched values of the aerodynamic transfer function matrix with the iterated
eigenvalues s. In this work, the root finding is performed by a quasi-Newton method with
a Krylov approximation for the inverse of the Jacobian [14]. The quasi-Newton approach
requires good initial estimates in order to properly converge to the roots. Therefore, the
structural eigenvectors and eigenfrequencies are used as starting solutions for the wind-off
condition A(s) = 0. Then, by stepwise increasing A(s), e.g., by a velocity or density sweep,
and solving the nonlinear eigenproblem at each step, the starting solutions are marched up
to the final solution. Other starting points may be required to find static divergence [12] or
the roots of fluid modes [15].

2.3. Aerodynamic Damping Approximations

The aerodynamic transfer function matrix A(s) of Equations (1) and (2) serves as a
general description of the linearized unsteady aerodynamic forces. Depending on the
employed aerodynamic method, no explicit expression can be determined and it has to be
determined numerically. Typically, aerodynamic methods make use of the nondimensional
frequency, which is the reduced complex-valued frequency

L
s∗ = s = σ∗ + iω ∗ , (3)
V
where L is the characteristic length, V the freestream velocity and ω ∗ the reduced frequency.
In the general case, termed the generalized aeroelastic analysis method (GAAM) by Ed-
wards and Wieseman [12], A(s∗ ) is provided for the both the real and imaginary part of s∗
and, therefore, the aerodynamic damping is exactly incorporated. Recently, a modification
to the extensively used doublet lattice method (DLM) [9] allowed for the application of
GAAM to general nonplanar configurations in subsonic compressible flow [16]. However,
if the evaluation of the aerodynamic transfer function matrix is limited to harmonic motion
that is the evaluation at only the imaginary part of s∗
Aerospace 2022, 9, 127 4 of 21

Ak (ω ∗ ) = A(iω ∗ ) , (4)

this leads to the well-known p-k method [10]. In this case, the evaluation of the aerodynamic
transfer function matrix disregards the damping coefficient σ∗ and, thus, it is a constant
approximation for σ∗ 6= 0. Another approximation is the g method by Chen [11], which
extends the p-k method by a first-order approximation for σ∗ at σ∗ = 0, resulting in

∂A(iω ∗ ) ∗ ∂A(iω ∗ ) ∗ ∂Ak (ω ∗ ) ∗

A g (s∗ ) = A(iω ∗ ) + σ = A ( iω ∗
) − i σ = A k ( ω ∗
) − i σ . (5)
∂σ∗ ∂ω ∗ ∂ω ∗
Equation (5) shows that the g method uses the first-order derivative of Ak (ω ∗ ), which, in
general, is available either analytically, by finite difference approximations or interpolation.
This first-order approximation for σ∗ is found from the assumption that A(s∗ ) is analytic
and, thus, complex-differentiable at s∗ ([17] 1.9(ii)). Hence, the Cauchy–Riemann equations
apply, relating the partial derivatives of A(s∗ ) with respect to the real and imaginary part
of s∗ by

∂A(s∗ ) ∂A(s∗ )
∗
= −i .
∂σ ∂ω ∗
Employing this property for σ∗ = 0 results in the first-order term for σ∗ in Equation (5)
and, subsequently, A g (s∗ ) is analytic at σ∗ = 0.
Recently, the p-L method, which recovers a true aerodynamic damping representation
from the values at the imaginary axis, has been presented [18]. Even though the aeroelastic
eigensensitivities produced by the p-L method could be embedded in the framework
presented in this work, they can also be obtained by using its corresponding generalized
state-space form. This formulation shall be presented in a separate work.
Overall, the p-k method, the g method and GAAM yield the same aerodynamic
forces for σ∗ = 0, which implies that the three methods return the same flutter onset but
their solutions are different for σ∗ 6= 0. Furthermore, the methods differ in their analytic
properties: the p-k method is not analytic, the g method is analytic at reduced frequencies
with σ∗ = 0, and GAAM is considered to be analytic at typical values of s∗ .

3. Aeroelastic Sensitivities
3.1. Derivatives of the Aeroelastic Eigenproblem
The aeroelastic sensitivities for the eigenvalues of the aeroelastic eigenproblem in
physical coordinates, Equation (1), are obtained by differentiating Equation (1) with respect
to a design parameter β. In terms of the dynamic aeroelastic matrix G (s), the differentiated
equation results in

dG (s) dx̂ ∂G (s) ∂G (s) ds dx̂

x̂ + G (s) = x̂ + x̂ + G (s) = 0, (6)
dβ dβ ∂β ∂s dβ dβ

where the sought eigenvalue derivative ds/dβ appears by the chain rule with the par-
tial derivative ∂G (s)/∂s evaluated at s. Moreover, Equation (6) contains the eigenvector
derivative dx̂/dβ as well as the partial derivative ∂G (s)/∂β at s. Solving for ds/dβ in
Equation (6) must factor in the also unknown eigenvector derivative dx̂/dβ, leading to the
direct method described in Section 3.2.
According to Equation (1), the partial derivative ∂G (s)/∂β aggregates the sensitivities
of the aeroelastic model matrices to
∂G (s) dM dD dK ∂A(s)
= s2 +s + − (7)
∂β dβ dβ dβ ∂β

with the partial derivative of the aerodynamic transfer function matrix ∂A(s)/∂β evaluated
at s. By categorizing the design parameters in structural and geometrical parameters,
the partial derivative ∂A(s)/∂β only applies for geometrical design parameters affecting
Aerospace 2022, 9, 127 5 of 21

the aerodynamic shape. The partial derivative ∂G (s)/∂s comprises the derivative of the
polynomial part for the structural dynamics and the partial derivative of the aerodynamic
transfer function matrix, resulting in

∂G (s) ∂A(s)
= 2sM + D − . (8)
∂s ∂s
In the general case, GAAM, where the aerodynamic transfer function matrix is given
in terms of the reduced complex-valued frequency s∗ , the partial derivatives must account
for the chained function of s∗ ; see Equation (3). Hence, the partial derivatives of the
aerodynamic transfer function matrix in Equations (7) and (8) expand to

∂A(s) ∂A(s∗ ) ∂s∗ L ∂A(s∗ )

= ∗
= , (9)
∂s ∂s ∂s V ∂s∗
∂A(s) ∂A(s∗ ) ∂A(s∗ ) ∂s∗ (s) ∂A(s∗ ) ∂A(s∗ ) ∂(sL/V )
= + = + , (10)
∂β ∂β ∂s∗ ∂β ∂β ∂s∗ ∂β

where the latter term ∂(sL/V )/∂β is zero unless β refers to one of the characteristic values
L, V. Importantly, it is noted that the partial derivatives with respect to s or s∗ demand
the aerodynamic transfer function matrix to be analytic, i.e., complex-differentiable, at
the evaluated complex-valued frequencies. Consequently, the derived equations must be
reformulated for the application with the p-k and g methods; see Sections 3.3 and 3.4.
Employing modal coordinates for the aeroelastic eigenproblem instead of physical
coordinates leads to differentiating Equation (2), which can be expanded to

dΦ T dG (s) dΦ dq̂
G (s)Φq̂ + Φ T Φq̂ + Φ T G (s) q̂ + Φ T G (s)Φ =0 (11)
dβ dβ dβ dβ

by the product rule, revealing the derivative of the modal matrix dΦ/dβ and the modal
eigenvector derivative dq̂/dβ. The derivative dG (s)/dβ entails the eigenvalue derivative
ds/dβ as well as the partial derivatives ∂G (s)/∂β and ∂G (s)/∂s; cf. Equation (6). The
modal matrix derivative comprising structural eigenvector derivatives is derived from the
structural model as described in Section 3.5.

3.2. Direct Method with Eigenvector Normalization

The direct method solves Equations (6) and (11) for the eigenvalue and eigenvector
derivatives by appending the eigenvector normalization in order to complete the set of
equations [4–6]. In this work, the eigenvectors are normalized with a symmetric weighting
matrix W by
x̂ T W x̂ = 1 , (12)
which is in line with the solution approach for the eigenproblem; see Section 2.2. It is
acknowledged that this equation is a degenerate quadratic form for the complex-valued
eigenvectors, which can yield a zero norm for nonzero vectors [6]. However, this condition
did not occur in any of the investigations carried out.
Differentiating Equation (12) with respect to β states the additional condition that the
eigenvector normalization constraint remains fulfilled, resulting in the additional equation

dx̂ T dW dx̂
W x̂ + x̂ T x̂ + x̂ T W = 0. (13)
dβ dβ dβ

Together with Equation (6), this renders the linear system of equations for the eigenvalue
and eigenvector derivatives

− ∂G∂β(s) x̂
" #" ds # " #
∂G (s)
∂s x̂ G (s) dβ
= , (14)
dx̂
0 2x̂ T W dβ − x̂ T dW
dβ x̂
Aerospace 2022, 9, 127 6 of 21

where the bilinear forms in Equation (13) are summarized for a symmetric weighting matrix
W. The right-hand side of this system of equations gathers the derivatives of the aeroelastic
model matrices, which allows us to solve the system of equations for multiple right-hand
sides and, thus, multiple design parameters. Alternatively, as discussed in [5], the adjoint
system of equations can be solved for an indicator vector as the right-hand side selecting
one component of the solution vector, such as the eigenvalue derivative. Subsequently, the
eigenvalue derivatives result efficiently from the vector product of this solution with an
arbitrary number of right-hand sides of Equation (14). For the eigenvalue derivative, this
reads as
  
" ds # " ∂G(s) # T " # T  −1 1
ds − ∂β x̂   ∂G∂s(s) x̂ G (s)  0
 
= 1 0 · · · 0 dβ
 
dx̂ = T dW T  ..  .
dβ − x̂ dβ x̂ 

dβ 0 2x̂ W  . 
0

For the aeroelastic eigenproblem in modal coordinates, the linear system of equations
is obtained according to Equation (14), where the terms are replaced by their modal
counterparts and the modal eigenvectors are normalized by the symmetric weighting
matrix W̃, resulting in
" #" ds #  ∂G̃(s) 
∂ G̃ (s) − ∂β q̂
q̂ G̃ ( s ) dβ
dq̂ =
∂s  . (15)
T W̃
0 2q̂ W̃ dβ −q̂ T ddβ q̂

The terms containing the modal matrix derivative—see Equation (11)—are included in the
right-hand side by ∂ G̃ (s)/∂β.

3.3. Aeroelastic Sensitivities for the p-k Method

As described in Section 2.3, the aerodynamic transfer function matrix of the p-k method
is not analytic. Therefore, the partial derivative with respect to the Laplace variable s does
not exist, which is a requirement for the aeroelastic derivatives determined in Section 3.1.
Consequently, the aeroelastic sensitivities are derived for nonanalytic aerodynamic transfer
function matrices by splitting the argument s into its real and imaginary part. Considering
the dynamic aeroelastic matrix G (s) as a function of ω and σ renders the differentiation of
Equation (1) to

∂G (s) ∂G (s) dσ ∂G (s) dω dx̂

x̂ + x̂ + x̂ + G (s) = 0, (16)
∂β ∂σ dβ ∂ω dβ dβ

where s still serves as the argument summarizing σ and ω. In contrast to Equation (6), the
partial derivatives of G (s) to σ and ω appear together with the derivatives dσ/dβ and
dω/dβ, which correspond to the real and imaginary part of the eigenvalue derivative. The
partial derivatives of G (s) are obtained as

∂G (s) ∂A(s)
= 2sM + D − , (17)
∂σ ∂σ
∂G (s) ∂A(s)
= i2sM + iD − , (18)
∂ω ∂ω
where the partial derivatives of A(s) with respect to σ and ω are found. This allows us to
evaluate the nonanalytic formulations of the aerodynamic transfer function matrix, where
the partial derivatives ∂A(s)/∂σ and ∂A(s)/∂ω are differently provided.
For the p-k method, the partial derivative ∂A(s)/∂σ is zero because of the constant
approximation for σ∗ 6= 0, and the partial derivative ∂A(s)/∂ω is the partial derivative of
Ak (ω ∗ ), including the partial derivative of the reduced frequency ω ∗ . The same applies to
the partial derivative with respect to β, which is obtained analogously to Equation (10).
Aerospace 2022, 9, 127 7 of 21

In summary, the partial derivatives in the case of the p-k method result in

∂A(s) L ∂Ak (ω ∗ )
= = 0, (19)
∂σ V ∂σ∗
∂A(s) L ∂Ak (ω ∗ )
= , (20)
∂ω V ∂ω ∗
∂A(s) ∂Ak (ω ∗ ) ∂Ak (ω ∗ ) ∂(ωL/V )
= + . (21)
∂β ∂β ∂ω ∗ ∂β

In order to solve Equation (16) for the eigensensitivities, the eigenvector normalization
from Equation (12) is required as a second equation. Subsequently, both equations are split
into their real and imaginary parts, resulting in four equations for the unknown eigensen-
sitivities dσ/dβ and dω/dβ, as well as the real and imaginary counterparts of dx̂/dβ,
namely <(dx̂/dβ) and =(dx̂/dβ). By expanding the complex-valued multiplications with
the eigenvector derivative into real and imaginary components, the four equations establish
the direct method for the real and imaginary parts of the eigensensitivities:
  
 dσ  ∂G (s)
 
−< x̂
  
∂G (s) ∂G (s)
< x̂ < ∂ω x̂ <( G (s)) −=( G (s))  dβ    ∂β  
  ∂σ   dω   ∂G (s)
−= ∂β x̂  
  
= ∂G ( s ) ∂G ( s )  dβ  
x̂ = x̂ =( G ( s )) <( G ( s ))
dx̂  = 
 ∂σ ∂ω     . (22)
 T

T

< dβ  −< x̂ T dW x̂ 
 0 0 2< x̂ W
−2= x̂ W
   
 dβ 

T T dx̂
0 0 2= x̂ W 2< x̂ W = dβ T dW
−= x̂ dβ x̂

Taken together, Equation (22) is equivalent to Equation (14) if the aerodynamic transfer
function matrix is analytic. However, Equation (22) enables the separate evaluation of
the partial derivatives with respect to the real and imaginary part of s, as required in the
nonanalytic case. In addition, the linear system of equations for the eigensensitivities
employing modal coordinates is obtained by substituting the terms in Equation (22) with
their modal counterparts analogous to Equation (15).

3.4. Aeroelastic Sensitivities for the g Method

For the g method, similar to the p-k method, the aeroelastic derivatives of Section 3.1
are not applicable, since the g method is only analytic at σ = 0. Therefore, it is necessary to
separately provide the partial derivatives of the aerodynamic transfer function matrix and to
employ the direct method stated in Equation (22) in order to solve for the eigensensitivities.
This requires the partial derivatives of the aerodynamic transfer function matrix
appearing in Equations (17), (18) and (7). They are obtained by differentiating the aero-
dynamic transfer function matrix of the g method, A g (s∗ ), defined in Equation (5), and
by accounting for the chained partial derivative of the reduced quantities σ∗ , ω ∗ to the
dimensional quantities σ, ω, respectively.
The partial derivative of A g (s∗ ) to σ∗ results in the linear factor −i∂Ak (ω ∗ )/∂ω ∗ for
σ in A g (s∗ ). For the partial derivative with respect to ω ∗ , this factor yields a second-order
∗

partial derivative of Ak (ω ∗ ) to ω ∗ , which, together with the first-order partial derivative,

renders the partial derivative of A g (s∗ ) to ω ∗ . Similarly, the partial derivative with respect
to β includes the mixed partial derivative of Ak (ω ∗ ) to ω ∗ and β. Moreover, by the chain
rule, it includes the derivatives ∂A g (s∗ )/∂σ∗ and ∂A g (s∗ )/∂ω ∗ , together with the partial
derivatives of σ∗ and ω ∗ to β, which only apply if β refers to L or V.
In summary, the partial derivatives in the case of the g method result in
Aerospace 2022, 9, 127 8 of 21

∂A(s) L ∂A g (s∗ ) L ∂Ak (ω ∗ )

= = − i , (23)
∂σ V ∂σ∗ V ∂ω ∗
L ∂A g (s∗ ) L ∂Ak (ω ∗ ) ∂2 A k ( ω ∗ ) ∗
 
∂A(s)
= = −i σ , (24)
∂ω V ∂ω ∗ V ∂ω ∗ ∂ω ∗ 2
∂A(s) ∂A g (s∗ ) ∂Ak (ω ∗ ) ∂2 A k ( ω ∗ ) ∗
= = −i σ
∂β ∂β ∂β ∂β∂ω ∗
(25)
∂Ak (ω ∗ ) ∂2 Ak (ω ∗ ) ∗ ∂(ωL/V ) ∂2 Ak (ω ∗ ) ∂(σL/V )
 
+ ∗
− i 2
σ −i
∂ω ∂ω ∗ ∂β ∂ω ∗ 2 ∂β

3.5. Sensitivities of the Modal Matrix

The sensitivities of the modal matrix require determination of the derivatives of the
structural mode shapes in wind-off condition and neglecting any structural damping terms
with respect to a design parameter β. This is equivalent to finding the sensitivities of the
structural eigenproblem providing the mode shapes.
The structural mode shapes ϕ are the eigenvectors of the generalized eigenproblem

Kϕ = λMϕ (26)

resulting from the structural dynamics

√ equations for harmonic motion. The structural
eigenfrequencies are given by λ and the structural eigenvectors are mass-normalized by

ϕT Mϕ = 1 , (27)

which renders the modal mass matrix M̃ = Φ T MΦ as the identity matrix.

Differentiating Equations (26) and (27) with respect to β allow us to solve for the
structural eigenvalue and eigenvector sensitivities with the direct method:

dK dM dλ dϕ
( −λ )ϕ − Mϕ + (K − λM ) = 0, (28)
dβ dβ dβ dβ
dϕ dM
2ϕT M + ϕT ϕ = 0. (29)
dβ dβ

The latter equation requires the mass matrix to be symmetric in order to combine both
bilinear forms containing the eigenvector derivative. Moreover, it enforces the condition
imposed by Equation (27) with a changing parameter β. This constitutes a linear system of
equations for the eigenvalue and eigenvector sensitivities
" dλ #
−( dK dM
" #
dβ − λ dβ )ϕ

− Mϕ K − λM dβ
dϕ = , (30)
0 2ϕT M dβ
−ϕT dM
dβ ϕ

which is then solved for each eigenvector and corresponding eigenvalue included in the
modal matrix. Subsequently, the derivative of the modal matrix is given by

dΦ h dϕj
i
= ··· dβ ··· . (31)
dβ

4. Results
4.1. Verification of Aeroelastic Sensitivities
The verification of the analytically derived aeroelastic sensitivities is conducted for
the two-degree-of-freedom (DOF) typical section model by analyzing the convergence
error to finite difference approximations. The investigated typical section model employs
Theodorsen’s aerodynamic airfoil theory [19] with the generalized Theodorsen function
developed by Edwards [20,21]. This allows us to derive the analytical partial derivatives
of the aerodynamic transfer function matrix presented in Section 3.1 and, moreover, in
Aerospace 2022, 9, 127 9 of 21

Sections 3.3 and 3.4 for the p-k and g methods. The analytical derivatives of the aerody-
namic transfer function matrix and the generalized Theodorsen function are derived in
Appendices A and B. Additionally, two degrees of freedom facilitate the employment of
the modal approach without modal truncation in order to verify the derivatives devel-
oped for the aeroelastic eigenproblem in modal coordinates—see Sections 3.1 and 3.5—by
comparison to the solutions obtained in physical coordinates.
The 2-DOF typical section model constitutes a two-dimensional airfoil section with a
plunge and pitch degrees of freedom, denoted with subscripts h and α, in incompressible
flow; see Figure 1. The air loads are described by the lift force and the aerodynamic moment
around the elastic axis, where the section model is elastically restrained with a vertical
stiffness k h and a torsional stiffness k α . Therefore, the nonlinear aeroelastic eigenproblem
for the typical section model results in
        
m Sα k 0 ∗ x̂h 0
s2 + h − A(s ) = , (32)
Sα Iα 0 kα x̂α 0

where the mass matrix includes the total mass m, the first moment of mass Sα around
the elastic axis and the moment of inertia Iα . The aerodynamic transfer function matrix
A(s∗ ) provides the negative lift force and the aerodynamic moment in the Laplace domain
from the amplitudes of motion x̂h , x̂α . Based on Theodorsen’s aerodynamic airfoil the-
ory [19], commonly described in the aeroelastic literature [22,23], the unsteady aerodynamic
forces are extended for growing and decaying oscillations by employing the generalized
Theodorsen function C (s∗ ) [20]. Expressed in dimensional form with reference to the unit
span of the section, the aerodynamic transfer function matrix is given by

 
∗ 2 −1
∗2 eb
A(s ) = ρV π s
eb −( 81 + e2 )b2
   
− 2C ( s ∗) − 1 − 2C ( s ∗ )( 1 − e ) b
2
+ s∗    
2C (s∗ )( 12 + e)b ( 21 − e) 2C (s∗ )( 12 + e) − 1 b2
!
−2C (s∗ )b

0
+ , (33)
0 2C (s∗ )( 21 + e)b2

which includes the dynamic pressure 12 ρV 2 and the area per unit span 2b. The reference
length L for the reduced frequencies is the half chord length b. The elastic axis is positioned
relative to the mid-point by the nondimensional parameter e. The generalized Theodorsen
function C (s∗ ) is provided in Equation (A5) in Appendix B.

Figure 1. Sketch of the 2-DOF typical section model with elastically restrained plunge and pitch
degree of freedom illustrated by an airfoil.

The aerodynamic transfer function matrix for the p-k method Ak (ω ∗ ) is found by
evaluation of Equation (33) with σ∗ = 0, neglecting the real part of the reduced complex-
Aerospace 2022, 9, 127 10 of 21

valued frequency s∗ ; see Equation (4). For the g method, in Equation (5), the additionally
required partial derivative ∂Ak (ω ∗ )/∂ω ∗ is obtained by applying the analytic property of
A(s∗ ) at σ∗ = 0, relating

∂Ak (ω ∗ ) ∂A(iω ∗ ) ∂A(iω ∗ )

∗
= ∗
=i ,
∂ω ∂ω ∂s∗
where ∂A(s∗ )/∂s∗ , given in Equation (A1) in Appendix A, requires the derivative of the
generalized Theodorsen function derived in Appendix B.
The employed parameter values for the section model are listed in Table 1.

Table 1. Parameter values for the typical section model. The structural parameters m, Sα , Iα , k h and
k α are given per unit span, ρ is the air density, e the relative position of the elastic axis and b the half
chord length.

m (kg/m) Sα (kg) Iα (kg·m) kh (N/m2 ) kα (N) b (m) e ρ (kg/m3 )

292.4823 73.1206 113.482 9.1396 × 105 4.1965 × 105 1 −0.15 1.225

In Figure 2, the eigenvalue solutions for the section model obtained by the p-k method,
the g method and GAAM are shown over a velocity sweep from 0 m/s to 300 m/s. Overall,
the three methods yield similar results. From Figure 2b,c, the flutter onset is found for the
second eigenvalue s2 , where the real part σ2 is zero at the freestream velocity V = 212.2 m/s.
As expected, the flutter onset is identically identified by all three methods. However, for
other velocities where σ 6= 0, the eigenvalue solutions exhibit slight variations. For the
p-k method, the differences are more pronounced, while the g method deviates little from
GAAM. In fact, for high velocities where σ sufficiently increases from zero, small deviations
for the g method are observed. These results confirm the main distinction between the
aerodynamic damping approximations in which the real part of the Laplace variable is
included to different degrees.

Figure 2. Eigenvalue solutions of the nonlinear aeroelastic eigenproblem for the typical section model
employing the p-k method, the g method and GAAM for freesteam velocities from 0 m/s to 300 m/s
with the flutter onset velocity at 212.2 m/s. (a) Angular frequencies. (b) Damping coefficients.
(c) Eigenvalues in the complex plane.
Aerospace 2022, 9, 127 11 of 21

At each velocity, the derivatives of the eigenvalue solutions with respect to a design
parameter are obtained by employing the direct method; see Section 3.2. The design
parameter for the verification of the derived equations is selected to be the half chord
length b. Hence, the quantities of interest are how the eigenvalues change with changing b.
Varying the half chord length implies a change in shape, requiring the partial derivative of
the aerodynamic transfer function matrix ∂A(s)/∂β. Moreover, since the half chord length
is the reference length, the additional terms regarding the partial derivative of the reduced
frequency with respect to L are included. Therefore, the design parameter b is well suited to
verify the derived eigenvalue derivatives. However, the structural derivatives with respect
to b are neglected in the following verification and the structural parameters are considered
to be independent of b.
In order to obtain the eigenvalue derivatives, GAAM requires the partial derivatives
∂A(s)/∂s∗ and ∂A(s∗ )/∂b, given by Equations (A1) and (A2) in Appendix A. The partial
derivative ∂Ak (ω ∗ )/∂b, for the p-k and g methods, is the derivative ∂A(iω ∗ )/∂b, where
the real part σ∗ is neglected. The g method additionally requires the second-order partial
derivatives ∂2 Ak (ω ∗ )/∂ω ∗ 2 and ∂2 Ak (ω ∗ )/∂ω ∗ ∂b, as discussed in Section 3.4. These are
again obtained from the analytic property of A(s∗ ), resulting in

∂2 A k ( ω ∗ ) ∂2 A(iω ∗ ) ∂2 A(iω ∗ )
= =− ,
∂ω ∗2 ∂ω ∗2 ∂s∗ 2
∂2 A k ( ω ∗ ) ∂2 A(iω ∗ ) ∂2 A(iω ∗ )
∗
= ∗
=i ,
∂ω ∂b ∂ω ∂b ∂s∗ ∂b
where the second-order partial derivatives of A(s∗ ) are evaluated at σ∗ = 0.
For the verification, the half chord length of the section model is changed by step
sizes ∆b ranging from 1 × 10−6 m to ∆bmax = 0.05 m. Subsequently, for a given velocity,
the eigenvalue solutions for each modified section model are obtained by solving the
nonlinear eigenproblem. Figure 3 displays the change in the eigenvalues for changing
half chord length by ±∆b at a freestream velocity V = 209.6 m/s. The velocity is close to
the flutter onset and, thus, the eigenvalue s2 is close to the imaginary axis where σ = 0.
To illustrate the results, at this velocity, the second eigenvalue would become less stable
with increasing half chord length and eventually become unstable. Additionally, the
predicted linear changes in the eigenvalues by the eigenvalue derivatives are shown as
lines attached to the eigenvalues of the unchanged section model. The depicted lines display
the scaled eigenvalue derivatives by multiplication with ±∆bmax . From visual judgement,
the resulting tangent lines show very good agreement with the slopes of the nonlinear
curves at the solution of the unchanged model. In addition to the visual comparison, the
numerical values of the obtained eigenvalue derivatives are given in Table 2.
By comparing the result between the p-k method, g method and GAAM in Figure 3,
the difference in the eigenvalue solution for the unchanged section model is noted, as is
also shown in Figure 2. This difference is more pronounced at s1 with greater distance
from the imaginary axis. Hence, the nonlinearly changing eigenvalues exhibit differences,
which are the greatest for the p-k method, since the unchanged eigenvalue solution already
shows the greatest offset. The g method shows very little difference to GAAM for the
nonlinearly and linearly obtained second eigenvalues s2 . For the first eigenvalue, a small
offset in the unchanged eigenvalue is observed, with little differences in the nonlinearly
obtained eigenvalues for greater values of +∆b. Although the unchanged eigenvalues for
the g method and GAAM are very close, a small but visible difference in the visualized
eigenvalue derivatives at the first eigenvalue is observed, as reported in Table 2, reflecting
the difference in the underlying aerodynamic damping approximation. This is one of
the main results of the current work, pointing out the effect of the aerodynamic damping
approximation on the aeroelastic eigensensitivities, even though the predicted eigensolution
to the flutter equation is very similar.
Aerospace 2022, 9, 127 12 of 21

Figure 3. Eigenvalue solutions for the typical section model employing the p-k method, the g method
and GAAM at V = 209.6 m/s for changing half chord length b displayed in the complex plane. The
eigenvalue derivatives scaled by ∆bmax are depicted as tangent lines at the unchanged eigenvalue
solution (center ×).

Table 2. Eigenvalue derivatives with respect to the half chord length b for the typical section model
employing the p-k method, the g method and GAAM at V = 209.6 m/s.

ds1 /db (rad/(m·s)) ds2 /db (rad/(m·s))

p-k method −44.180995 − 9.676179i 31.725084 − 13.803641i
g method −54.545970 − 0.113813i 45.695638 − 15.883591i
GAAM −54.064094 + 0.513874i 45.905266 − 16.045078i

Finally, the derived analytical eigenvalue derivatives are verified by the analysis of
the relative error to forward finite difference approximations. Considering the eigenvalues
as a function of β renders the finite difference approximations using nonlinearly obtained
eigenvalues evaluated at β + ∆β. Subsequently, the relative error is given by

ds/dβ − (s( β + ∆β) − s( β))/∆β O(∆β)

ηds/dβ = = , (34)
ds/dβ ds/dβ

where the latter equality results from Taylor’s theorem, leading to a first-order approxi-
mation error if the evaluated derivatives are the analytical derivatives. Consequently, the
relative error converges linearly with decreasing ∆b for the analytical derivatives.
Figure 4 displays the relative errors of Equation (34) for the eigenvalue derivatives at
two freestream velocities V = 209.6 m/s and V = 241.2 m/s. For the three investigated
aerodynamic damping approximations, the relative errors separated in real and imaginary
parts exhibit an exponential decline for the decreasing step size ∆b. Examining the slopes
shows a first-order reduction in the relative error, which is in line with the stated condition
for analytical derivatives in Equation (34). For very small step sizes ∆b, the relative error
deviates from the first-order reduction because of the increasing numerical error in the
finite difference approximations due to numerical cancellation. From these results, it
is concluded that the obtained eigenvalue derivatives are in agreement with the finite
difference approximations, verifying the derived equations for the eigenvalue sensitivities.
To this end, it remains to verify the eigenvalue sensitivities employing modal coor-
dinates. For this, it is sufficient to compare the eigenvalue derivatives obtained in modal
coordinates without modal truncation to the verified solutions obtained in physical coor-
dinates. Figure 5 shows this comparison for the design parameter b employing GAAM.
Aerospace 2022, 9, 127 13 of 21

In Figure 5a, the eigenvalue solutions are displayed for a velocity sweep from 0 m/s to
300 m/s. Figure 5b,c show the real and imaginary parts of the eigenvalue derivatives at
each velocity. The eigenvalue derivatives are obtained by the direct method in physical
coordinates—see Equation (14)—and in modal coordinates—see Equation (15). No differ-
ences are observed in Figure 5, verifying the equality of the solutions of the direct method
in physical and modal coordinates without modal truncation.

Figure 4. Relative errors of the eigenvalue derivatives for the typical section model over ∆b dis-
playing linear convergence in the real (solid) and imaginary parts (dotted). (a) V = 209.6 m/s.
(b) V = 241.2 m/s.

Figure 5. Comparison of eigenvalue solutions and eigenvalue derivatives in physical and modal coor-
dinates for GAAM. (a) Eigenvalues in the complex plane. (b) Real part of the eigenvalue derivatives
over freestream velocity. (c) Imaginary part of the eigenvalue derivatives over freestream velocity.
Aerospace 2022, 9, 127 14 of 21

4.2. Aeroelastic Sensitivities of the AGARD 445.6 Wing

In this section, the derived aeroelastic sensitivities are applied to the AGARD 445.6
weakened wing model for varying wing sweep. The AGARD 445.6 weakened wing model
is a widely known aeroelastic test case [24,25]. It is a swept wing clamped at the wing
root with a quarter chord sweep of Λ = 45◦ , a root chord length of c = 0.5578 m, a span
of 0.762 m and a taper ratio of 0.66. The model comprises a finite element model (FEM)
with 200 shell elements and a doublet lattice model (DLM) with 200 boxes, which both are
obtained from [26] for a Mach number of 0.9 and air density ρ = 0.31 kg/m3 .
The commercial solver MSC Nastran [7] is employed in order to extract the mass
matrix M and stiffness matrix K as well as the modal matrix Φ by performing a linear
modal analysis. In this model, the first four structural mode shapes are considered, which
are the first wing bending and first wing torsion, followed by the second wing bending and
second wing torsion. The structural eigenfrequencies are listed in Table 3 together with
the parameter values. Furthermore, the aerodynamic transfer function matrices Ak (ω ∗ )
employing DLM are extracted from MSC Nastran, sampled at 17 reduced frequencies
ω ∗ ranging from 0.001 to 5, clustered densely in the region up to 1. An interpolation
method based on piecewise cubic functions provides the evaluation of Ak (ω ∗ ) at additional
frequencies, which are required in obtaining the eigenvalue solutions. Moreover, the
interpolation method enables the efficient evaluation of the derivatives of Ak (ω ∗ ) with
respect to ω ∗ , which are, therefore, continuous up to the second order. The aerodynamic
transfer function matrices include the spline matrix, which is required for mapping the
structural displacement onto the aerodynamic grid as well as the aerodynamic forces onto
the structural grid using infinite plate spline interpolation [27]. Taken together, the extracted
matrices allow us to solve the nonlinear aeroelastic eigenproblem in modal coordinates;
cf. Equation (2).

Table 3. Parameter values and structural eigenfrequencies for the AGARD 445.6 weakened model.
The reference length is the half chord length at the wing root, L = c/2.

Mach Number ρ (kg/m3 ) Λ (◦ ) c (m) L (m) Eigenfrequencies (rad/s)

0.9 0.31 45 0.5578 0.2789 59.4, 294.4, 310.7, 597.5

In Figure 6, the eigenvalue solutions for the AGARD 445.6 weakened model obtained
by the p-k method and the g method are shown over a velocity sweep from 0 m/s to
350 m/s. The flutter onset is found by both methods for the first eigenvalue s1 at the
freestream velocity V = 193.9 m/s, as shown in Figure 6b,c. For smaller freestream
velocities, the p-k method and the g method yield very similar solutions. The solutions start
to differ for higher freestream velocities, where the real part of the eigenvalues σ increases.
This is most pronounced for the first and second eigenvalues, s1 and s2 , which exhibit the
most damping and, thus, significant deviations between both methods are observed for
freestream velocities greater than the flutter onset velocity.
The sweep angle of the AGARD 445.6 weakened model is varied by applying a shear
transformation for a step size ∆Λ without affecting the other parameters, such as root
chord length, span and taper ratio. The sweep angle is changed for step sizes ranging from
4.67 × 10−8 ◦ to ∆Λmax = 4.67◦ . For each step size, MSC Nastran is employed for acquiring
M, K, Φ and Ak (ω ∗ ), as described previously. Subsequently, the nonlinear aeroelastic
eigenproblem is solved for each sweep angle in order to obtain the nonlinearly changing
eigenvalues for varying sweep. Furthermore, the extracted mass matrix M and stiffness
matrix K are used for forward finite difference approximations of the model sensitivities
dM/dβ and dK/dβ. From these sensitivities, the sensitivity of the modal matrix dΦ/dβ is
obtained following Section 3.5. Additionally, the sensitivity of the aerodynamic transfer
function matrix dAk (ω ∗ )/dβ is determined by forward finite difference approximations for
each sampled frequency. With these defined model sensitivities, the eigenvalue sensitivities
are solved by employing the direct method for separated real and imaginary parts, as stated
Aerospace 2022, 9, 127 15 of 21

in Equation (22) in model coordinates. Hence, the linear system of equations includes the
additional terms regarding the derivative of the modal matrix; see Equation (11).

Figure 6. Eigenvalue solutions of the nonlinear aeroelastic eigenproblem for the AGARD 445.6
weakened model employing the p-k method and the g method for freestream velocities from 0 m/s to
350 m/s with the flutter onset velocity at 193.9 m/s. (a) Angular frequencies. (b) Damping coefficients.
(c) Eigenvalues in the complex plane.

Figure 7 shows the nonlinearly changing eigenvalues as well as the predicted linear
change from the eigenvalue sensitivities for varying sweep at the flutter onset velocity.
Similar to Section 4.1, the lines representing the eigenvalue sensitivities are defined by the
multiplication of the eigenvalue sensitivities with ±∆Λmax . The eigenvalue sensitivities
are shown for the model sensitivities obtained for the step size ∆Λ = 5.9 × 10−5 ◦ , resulting
from a convergence study, shown in Figure 8. For the flutter onset velocity, the p-k and g
methods yield the same eigenvalue s1 for the unchanged model, as is confirmed in Figure 7a.
However, the nonlinearly obtained eigenvalues for changing sweep differ for both methods,
which is correctly reflected by the observed difference in the linearly predicted eigenvalues.
Hence, the eigenvalue sensitivities are different for the p-k method and the g method,
although the eigenvalue solutions are the same and, moreover, have a real part of zero.
For the second eigenvalue s2 —see Figure 7b—the difference between both methods
results primarily from the offset between the eigenvalue solutions for the unchanged
model. As a result, different eigenvalue sensitivities are obtained—see also Figure 8, where
numerical values are compared. The difference in the eigenvalue solutions of the third and
fourth eigenvalue shown in Figure 7c,d is less pronounced. This is in line with the previous
observation that the first and second eigenvalues are more affected by the investigated
aerodynamic damping approximations.
In Figure 9, the eigenvalue solutions for changing sweep are shown for the freestream
velocity V = 230.7 m/s, above the flutter onset. In this case, the difference between
both methods results from the offset for the first and second eigenvalue, s1 and s2 , of
the unchanged model. Moreover, the eigenvalue sensitivities as well as the nonlinearly
obtained eigenvalues reveal the difference in the change in the aeroelastic stability resulting
from the different aerodynamic damping approximations.
Aerospace 2022, 9, 127 16 of 21

Figure 7. Eigenvalue solutions for the AGARD 445.6 weakened model employing the p-k method
and the g method at V = 193.9 m/s for changing sweep angle Λ displayed in the complex plane. The
eigenvalue derivatives obtained at ∆Λ = 5.9 × 10−5 ◦ and scaled by ∆Λmax are depicted as tangent
lines at the unchanged eigenvalue solution (center ×). (a) First eigenvalues s1 . (b) Second eigenvalues
s2 . (c) Third eigenvalues s3 . (d) Fourth eigenvalues s4 .

Figure 8. Convergence of the eigenvalue sensitivities for the AGARD 445.6 weakened model
employing the p-k method and the g method at V = 193.9 m/s over the step size ∆Λ for the
model sensitivities.
Aerospace 2022, 9, 127 17 of 21

Figure 9. Eigenvalue solutions for the AGARD 445.6 weakened model employing the p-k method
and the g method at V = 230.7 m/s for changing sweep angle Λ displayed in the complex plane. The
eigenvalue derivatives obtained at ∆Λ = 5.9 × 10−5 ◦ and scaled by ∆Λmax are depicted as tangent
lines at the unchanged eigenvalue solution (center ×). (a) First eigenvalues s1 . (b) Second eigenvalues
s2 . (c) Third eigenvalues s3 . (d) Fourth eigenvalues s4 .

5. Conclusions
In this work, the aeroelastic eigensensitivities are derived from the nonlinear aeroelas-
tic eigenproblem accounting for different aerodynamic damping approximations, which
are the p-k method, the g method and the generalized aeroelastic analysis method (GAAM).
Moreover, the solution method for the aeroelastic eigensensitivities employing the direct
method is presented in the context of different aerodynamic damping approximations. The
nonlinear aeroelastic eigenproblem is considered in both physical and modal coordinates,
leading to additional terms regarding the modal matrix sensitivities in the latter case.
The derived aeroelastic eigensensitivities are verified for the typical section model
employing the three presented aerodynamic damping approximations applied to the
generalized Theodorsen aerodynamic airfoil theory. The verification is performed by
analytical derivatives of the involved sensitivities regarding a shape design parameter. This
includes the first and second analytical derivatives of the generalized Theodorsen function.
Additionally, the verification includes the approach in modal coordinates.
The influence of the aerodynamic damping approximations on the resulting eigenvalue
sensitivities is discussed by means of the typical section model as well as the AGARD 445.6
wing model with varying sweep. The presented results display differences in the obtained
eigenvalue sensitivities for the three investigated aerodynamic damping approximations,
with the p-k method showing the most pronounced deviations. The differences in the eigen-
value sensitivities can be attributed in part to the offset in the underlying eigensolutions,
resulting from increasing absolute values of the damping coefficients. However, the results
demonstrate that, for very close or equal eigensolutions, the aerodynamic damping approx-
imations also affect the eigenvalue sensitivities. Moreover, this applies to eigensolutions
with zero damping coefficients.
Aerospace 2022, 9, 127 18 of 21

For the application of flutter constraints in gradient-based MDO, the presented results
indicate the preference of GAAM and the g method over the p-k method. Although the three
methods identify the identical flutter onset where flutter constraints become active, GAAM
and the g method are able to provide more accurate sensitivities because of the improved
aerodynamic damping. Future work will address the application of the derived approach
for the recently presented analytic continuation of DLM [16] as well as the application for
high-fidelity CFD methods.

Author Contributions: Conceptualization, C.K. and D.Q.; methodology, C.K. and D.Q.; software,
C.K.; validation, C.K. and D.Q.; writing—original draft preparation, C.K.; writing—review and
editing, D.Q. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the German Aerospace Center (DLR) in the research project
oLAF (Optimally Load-Adaptive Aircraft). There is no external funding involved.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: The data presented in this study are available on request from the
corresponding author.
Conflicts of Interest: The authors declare no conflicts of interest. The funders had no role in the
design of the study; in the collection, analyses, or interpretation of data; in the writing of the
manuscript, or in the decision to publish the results.

Abbreviations
The following abbreviations are used in this manuscript:

CFD Computational Fluid Dynamics

DLM Doublet Lattice Method
DOF Degrees of Freedom
FEM Finite Element Method
GAAM Generalized Aeroelastic Analysis Method
MDO Multi-Disciplinary Design Optimization

Appendix A. Derivatives of the Aerodynamic Transfer Function Matrix for the 2-DOF
Typical Section Model
In this section, the partial derivatives of the aerodynamic transfer function matrix of
the 2-DOF typical section model investigated in Section 4.1 are presented. The aerodynamic
transfer function matrix A(s∗ ), as stated in Equation (33), is differentiated with respect to
s∗ and b up to the second order.
First, the partial derivative with respect to s∗ is obtained by mainly applying the prod-
uct rule for terms including s∗ C (s∗ ). Consequently, the first derivative of the generalized
Theodorsen function is required, which is derived in Equation (A6) in Appendix B. Hence,
the partial derivative results in

∂A(s∗ )
 
2 ∗ −1 eb
= ρV π 2s
∂s∗ eb −( 81 + e2 )b2
   
−2C (s∗ ) −1 − 2C (s∗ )( 21 − e) b
+   
2C (s∗ )( 12 + e)b ( 21 − e) 2C (s∗ )( 12 + e) − 1 b2
dC (s∗ ) −( 21 − e)b
 
−1
+ 2s∗
ds∗ ( 21 + e)b ( 12 − e)( 21 + e)b2
!
dC (s∗ ) 0

−b
+2 . (A1)
ds∗ 0 ( 21 + e)b2
Aerospace 2022, 9, 127 19 of 21

The partial derivative with respect to b is given by

∂A(s∗ )
 
2 ∗2 0 e
= ρV π s
∂b e −( 18 + e2 )2b
−1 − 2C (s∗ )( 21 − e) 
" #
∗
0
+s
2C (s∗ )( 12 + e) ( 12 − e) 2C (s∗ )( 21 + e) − 1 2b
!
−2C (s∗ )

0
+ (A2)
0 2C (s∗ )( 21 + e)2b

In order to obtain the eigensensitivities by the g method, the second-order partial

derivatives are required. Differentiating ∂A(s∗ )/∂s∗ with respect to s∗ yields the second-
order partial derivative where the second derivative for the generalized Theodorsen func-
tion appears; see Equation (A7) in Appendix B. The second-order partial derivative re-
sults in

∂2 A ( s ∗ )
 
2 −1 eb
= ρV π 2
∂s∗ 2 eb −( 81 + e2 )b2
dC (s∗ ) 2 ∗ 
−( 21 − e)b
 
∗ d C (s ) −1
+ 4 + 2s 1
ds ∗ ds ∗ 2 ( 2 + e)b ( 21 − e)( 12 + e)b2
!
d2 C (s∗ ) 0

−b
+2 . (A3)
ds∗ 2 0 ( 12 + e)b2

Finally, the mixed partial derivative is obtained by differentiating ∂A(s∗ )/∂s∗ with
respect to b, yielding

∂2 A ( s ∗ )
 
2 ∗ 0 e
= ρV π 2s
∂s∗ ∂b e −( 18 + e2 )2b
−1 − 2C (s∗ )( 12 − e) 
" #
0
+
2C (s∗ )( 12 + e) ( 21 − e) 2C (s∗ )( 12 + e) − 1 2b
dC (s∗ ) −( 21 − e)
 
0
+ 2s∗
ds∗ ( 12 + e) ( 12 − e)( 21 + e)2b
!
dC (s∗ ) 0

−1
+2 . (A4)
ds∗ 0 ( 12 + e)2b

Appendix B. Derivatives of the Generalized Theodorsen Function

The generalized Theodorsen function developed by Edwards [20,21] extends the
classical Theodorsen function [19] for complex-valued reduced frequencies to account for
growing and decaying airfoil oscillations. It is given by

K1 ( s ∗ )
C (s∗ ) = , (A5)
K0 ( s ) + K1 ( s ∗ )
∗

where K0 (s∗ ) and K1 (s∗ ) are the modified Bessel functions of the second kind for integer
order zero and one, respectively. C (s∗ ) is typically evaluated on the principal branch that
is for π < arg s∗ ≤ π by numerical implementations of the modified Bessel functions
of the second kind. Evaluating the generalized Theodorsen function at iω ∗ recovers the
classical Theodorsen function, which is typically expressed in terms of Hankel functions for
Aerospace 2022, 9, 127 20 of 21

the real-valued argument ω ∗ . Hence, the generalized Theodorsen function is the analytic
continuation of the classical Theodorsen function for σ∗ 6= 0.
The derivatives of the generalized Theodorsen function are obtained by the help of the re-
currence relations for the modified Bessel functions of the second kind ([17] Equation (10.29.1)),
which give, in particular, the derivatives

dK0 (s∗ ) 1
= − (K−1 (s∗ ) + K1 (s∗ )) = −K1 (s∗ ) ,
ds∗ 2
dK1 (s∗ ) 1
∗
= − (K0 (s∗ ) + K2 (s∗ )) ,
ds 2
dK2 (s∗ ) 1
= − (K1 (s∗ ) + K3 (s∗ )) ,
ds∗ 2
where the latter derivatives include the modified Bessel functions of the second kind for
integer order two and three.
By the quotient rule and after some cancellation in the nominator, the first derivative
of the generalized Theodorsen function results in

dK1 (s∗ ) ∗
∗
∗ dK0 (s )
dC (s∗ ) ds∗ K0 ( s ) − K1 ( s ) ds∗
=
ds∗ (K0 (s∗ ) + K1 (s∗ ))2
(A6)
2K1 (s∗ )2 − K0 (s∗ )2 − K0 (s∗ )K2 (s∗ )
= .
2(K0 (s∗ ) + K1 (s∗ ))2

Subsequently, by differentiating dC (s∗ )/ds∗ , the second derivative of the generalized

Theodorsen function can be expressed as

d2 C ( s ∗ ) (K0 K1 + 2K0 K3 − K1 K2 )(K0 + K1 ) + 2K12 − K02 − K0 K2 (4K1 + 2K0 + 2K2 )

= , (A7)
ds∗ 2 4 ( K0 + K1 ) 3
where the arguments for the modified Bessel functions Kν (s∗ ) are omitted for improved
readability.

References
1. Jonsson, E.; Riso, C.; Lupp, C.A.; Cesnik C.E.; Martins J.R.; Epureanu, B.I. Flutter and post-flutter constraints in aircraft design
optimization. Prog. Aerosp. Sci. 2019, 109, 100537.
2. Kenway, G.K.; Martins, J.R. Multipoint high-fidelity aerostructural optimization of a transport aircraft configuration. J. Aircraft
2014, 51, 144–160.
3. Abu-Zurayk, M.; Merle, A.; Ilic, C.; Keye, S.; Goertz, S.; Schulze, M.; Klimmek, T.; Kaiser, C.; Quero, D.; Häßy, J.; et al. Sensitivity-
based multifidelity multidisciplinary optimization of a powered aircraft subject to a comprehensive set of loads. In Proceedings
of the AIAA Aviation 2020 Forum, Virtual Event, 15–19 June 2020.
4. Cardani C.; Mantegazza P. Calculation of eigenvalue and eigenvector derivatives for algebraic flutter and divergence eigenprob-
lems. AIAA J. 1979, 17, 408–412.
5. Bindolino G.; Mantegazza P. Aeroelastic derivatives as a sensitivity analysis of nonlinear equations. AIAA J. 1987, 25, 1145–1146.
6. Murthy, D.V.; Haftka, R.T. Derivatives of Eigenvalues and Eigenvectors of a general Complex Matrix. Int. J. Numer. Methods Eng.
1988, 26, 293–311.
7. MSC Software Cooperation. MSC Nastran Design Sensitivity and Optimization User’s Guide; Version 2019; Technical Report; MSC
Software Cooperation: Newport Beach, CA, USA, 2019.
8. Neill, D.J.; Johnson, E.H.; Canfield, R. ASTROS—A Multidisciplinary Automated Structural Design Tool. J. Aircraft 1990, 27,
1021–1027.
9. Rodden, W.P.; Taylor, P.F.; McIntosh, S.C., Jr. Further Refinement of the Subsonic Doublet-Lattice Method. J. Aircraft 1998, 35, 5.
10. Hassig, H.J. An Approximate True Damping Solution of the Flutter Equation by Determinant Iteration. J. Aircraft 1971, 8, 885–889.
11. Chen, P.C. Damping Perturbation Method for Flutter Solution: The g-method. AIAA J. 2000, 38, 1519–1524.
12. Edwards, J.W.; Wieseman, C.D. Flutter and Divergence Analysis Using the Generalized Aeroelastic Analysis Method. J. Aircraft
2008, 45, 906–915.
13. Jonsson, E.; Kenway, G.; Martins J.R.; Kennedy, G.J. Development of Flutter Constraints for High-fidelity Aerostructural
Optimization. In Proceedings of the AIAA Aviation 2017 Forum, Denver, CO, USA, 5–9 June 2017.
Aerospace 2022, 9, 127 21 of 21

14. Knoll, D.A.; Keyes, D.E. Jacobian-free Newton-Krylov methods: A survey of approaches and applications. J. Comput. Phys. 2003,
193, 357–397.
15. Nitzsche, J.; Ringel, L.M.; Kaiser, C.; Hennings, H. Fluid-Mode Flutter in Plane Transonic Flows. In Proceedings of the International
Forum on Aeroelasticity and Structural Dynamics, IFASD 2019, Savannah, GA, USA, 9–13 June 2019.
16. Quero, D. Modified Doublet Lattice Method for Its Analytical Continuation in the Complex Plane. J. Aircraft 2022. Advance
online publication. https://doi.org/10.2514/1.C036453
17. Olver, F.W.J.; Daalhuis, A.B.O.; Lozier, D.W.; Schneider, B.I.; Boisvert, R.F.; Clark, C.W.; Miller, B.R.; Saunders, B.V.; Cohl,
H.S.; McClain, M.A. (Eds.) NIST Digital Library of Mathematical Functions. Release 1.1.4 of 2022-01-15. Available online:
https://dlmf.nist.gov/ (accessed on 31 January 2022).
18. Quero, D.; Vuillemin, P.; Poussot-Vassal C. A generalized eigenvalue solution to the flutter stability problem with true damping:
The p-L method. J. Fluids Struct. 2021, 103, 103266.
19. Theodorsen T. General Theory of Aerodynamic Instability and the Mechanism of Flutter; NACA Report 496; National Advisory
Committee for Aeronautics, NACA, Washington DC, USA, 1949.
20. Edwards, J.W. Unsteady Aerodynamic Modeling and Active Aeroelastic Control. Doctoral Dissertation, Department of Aeronau-
tics and Astronautics, Stanford University, Stanford, CA, USA, 1977.
21. Edwards, J.W.; Ashley, H.; Breakwell, J.V. Unsteady Aerodynamic Modelling for Arbitrary Motions. AIAA J. 1979, 17, 365–374.
22. Bisplinghoff, R.L.; Ashley, H.; Halfman, R.L. Aeroelasticity; Dover Publications, Inc.: Mineola, NY, USA, 1996.
23. Fung, Y.C. An Introduction to the Theory of Aeroelasticity; Dover Publications, Inc.: Mineola, NY, USA, 1993.
24. Allen, C.B.; Jones, D.; Taylor, N.V.; Badcock, K.J.; Woodgate, M.A.; Rampurawala, A.M.; Cooper, J.E.; Vio, G.A. A comparison of
linear and non-linear flutter prediction methods: A summary of PUMA DARP aeroelastic results. In Proceedings of the Royal
Aeronautical Society Aerodynamics Conference, London, UK, 14–15 September 2004.
25. Beaubien, R.J.; Nitzsche, F.; Feszty, D. Time and frequency domain flutter solutions for the AGARD 445.6 wing. In Proceedings of
the International Forum on Aeroelasticity and Structural Dynamics, IFASD 2005, Munich, Germany, 28 June–1 July 2005.
26. Test Case AGARD 445.6 Weakened Model. Available online: http://www.cfd4aircraft.com/models/agard/agard.php (accessed
on 31 January 2022).
27. Harder, R.L.; Desmarais, R.N. Interpolation using surface splines. J. Aircraft 1972, 9, 189–191.