Analysis and Control of Vortex-Induced

Vibrations in a Cantilever Beam Using

Piezo-Stack Actuator

Devarakonda Vinay Kumar, Sonnathi Harika, and Ashesh Saha

Abstract A mathematical model is derived to explain the flow velocity-response

amplitude relations of a cylindrical cantilever beam in a fluid flow observed from
experiments. Thereafter, these vibrations are controlled using a piezo-stack actuator.
A Van der Pol damping term is considered in the mathematical model which intro-
duces negative damping in the system beyond a critical flow velocity. This causes
the instability of the system. The partial differential equation (PDE) governing the
vortex-induced vibrations along with the boundary conditions is reduced to a set of
ordinary differential equations (ODE) using the method of modal decomposition.
The actuating moment applied by the piezo-stack on the beam is also incorporated in
the mathematical model. Linear stability analysis shows that the stability of steady
state changes via a Hopf bifurcation. Amplitude and phase equations are obtained
using the method of averaging. A detailed parametric study is conducted to deter-
mine the functional relations of the coefficients of the Van der Pol damping on the
flow velocity so as to predict the experimentally observed response amplitudes with
minimum errors. Control signal from a simple proportional-derivative (PD) controller
is found to be effective in controlling the flow-induced vibrations.

Keywords Flow-induced vibrations · Continuous system · Piezo-stack actuator ·

Van der Pol oscillator

Abbreviations

V Beam transverse displacement

E Modulus of elasticity of beam material
I Area moment of inertia
ρ Density of beam material
A Cross-sectional area

D. V. Kumar · S. Harika · A. Saha (B)

National Institute of Technology Calicut, Kattangal, Calicut, Kerala 673601, India
e-mail: ashesh@nitc.ac.in

© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023 1021
X. Li et al. (eds.), Emerging Trends in Mechanical and Industrial Engineering,
Lecture Notes in Mechanical Engineering,
https://doi.org/10.1007/978-981-19-6945-4_78
1022 D. V. Kumar et al.

Mr Actuator moment
K∗ Piezoelectric constant
Va Actuator voltage
r1 Initial position of piezo-stack
r2 Final position of piezo-stack
t Time
U Normalized flow velocity
v Dimensionless transverse displacement
γa Dimensionless actuator voltage
τ Dimensionless time
α Dimensionless Van der Pol parameter
β Dimensionless Van der Pol parameter
u(·) Unit step function
&i (x) Orthonormal mode shapes
n Number of mode shapes
pi Natural frequency for ith mode
Kp Proportional gain
Kd Derivative gain
d Diameter of beam cross section

1 Introduction

Flow-induced vibrations (FIV) refer to the motion induced in a structure due to

interaction with a flowing fluid. FIV of a cylinder is caused due to the periodic forcing
on the cylinder induced by the alternate shedding of vortices. Accordingly, the FIV of
a cylinder is referred to as vortex-induced vibrations (VIV) [1]. FIV can be witnessed
in many practical applications such as biomedical equipment, turbine blades, bridge
cables, transmission wires, fuel rods used in nuclear reactors, and thermometers used
in the cooling system of fast breeder reactor. The flow-induced vibrations cause wear
of the turbine blades and cracks in shrouds, nozzles, valves, and casings [2]. Sodium
leakage accident in MONJU is a consequence of underestimating the ill-effects of
FIV [3]. Naturally, vibrations of this kind should be minimized whenever possible as
they can damage multiple components if ignored for a long time, potentially causing
the shutdown of the plant.
Several experimental studies on the VIV of cantilevered cylinders are available
in literature. The most notable one is by Fujarra et al. [4] who obtained the rela-
tions between the amplitude of flow-induced vibrations and the flow velocity under
different operating conditions. The major objective of this paper is to develop a math-
ematical model which can predict the flow velocity-response amplitude relations
reported by Fujarra et al. [4].
A lot of studies were performed to analyze the flow-induced vibrations of cylin-
drical cantilever beams. The actual problem is notoriously complex requiring the
Analysis and Control of Vortex-Induced Vibrations in a Cantilever Beam … 1023

application of equations from fluid dynamics to determine the generation of flow

field around the cylinder along with the beam elasticity equations to obtain the vibra-
tions of the cantilevered cylinder. Many phenomenological models are developed to
simplify the fluid–structure problem which can predict the self-excited oscillations of
the structure without any computational complexity. In one such class of simplified
low-order models, Van der Pol equation is used to describe the near wake dynamics
responsible for the fluctuating forces on the cylinder due to vortex shedding [5,
6]. This wake oscillator is coupled with the equation of motion of the structure to
describe, both qualitatively and quantitatively, several phenomena observed exper-
imentally in VIV. Further simplification can be achieved by adding a Van der Pol
term in the cantilever beam equation [7].
The degree of simplicity in a phenomenological model varies with the primary
objective of a problem. Barrón and Sen [7] directly used the Van der Pol term in
the equation governing the vibration of the beam as a mechanism for self-excited
oscillations of the beam. They were not concerned about explaining the experimen-
tally observed phenomena. Their main objective was to analyze the synchronization
phenomenon between coupled self-excited oscillators. We extended their analysis on
a single oscillator model to quantitatively describe the flow velocity-response ampli-
tude relations observed from experiments. Van der Pol terms included in the beam
equation introduce negative damping in the system beyond a critical flow velocity
resulting in the vibrations of the beam. With proper choice of the parameters in
Van der Pol terms, the results obtained from our theoretical analysis are found to
be matching closely with the experimentally obtained flow velocity-response ampli-
tude relations. A further extension of this model to describe other experimentally
observed phenomena will be considered in future.
A significant advancement in the field of energy harvesting can be noticed in
recent years. In one of the methods, the energy from the flow-induced vibrations of
cantilever beam is harvested using piezoelectric materials. Acciani et al. [8] examined
the effectiveness of various piezo materials in harvesting flow-induced vibrations.
Asghar et al. [9] carried out a vibration analysis using CFD and FEA on a vertical axis
wind turbine to compute the vibration stresses. The method of modal decomposition
analysis and most critical modes for blade failure are detailed in their paper. The
latest study by Seyed-Aghazadeh et al. [10] proposed a novel design to improve the
power density of energy harvester by using an L-elastic structured beam. In our work,
we are proposing a much simpler method of analyzing the flow-induced vibrations
in place of abovementioned complicated and expensive computational methods.
Piezoelectric actuators are commonly used for vibration suppression as well as
in precision positioning, shape control devices, etc. [11]. There have been some
recent advancements in nanotechnology which are enabled by piezoelectric actuators
[11]. In this paper, we are using a piezo-stack actuator to control the flow-induced
vibrations of the cantilever beam.
The rest of the paper is organized as follows. The mathematical model of the
proposed system is discussed in Sect. 2. Linear stability analysis is performed in
Sect. 3 to obtain the critical flow velocity corresponding to the Hopf bifurcation point.
The amplitude and phase equations are then derived using an analytical approach
1024 D. V. Kumar et al.

known as the averaging method in Sect. 3. All the results are discussed in Sect. 4
followed by conclusions in Sect. 5.

2 Mathematical Modeling

The schematic representation of a cylindrical cantilever beam with a piezo-stack

actuator is shown in Fig. 1. The flow-induced vibrations in the beam are generated
by a Van der Pol oscillator attached to the beam shown schematically by a dashpot
in Fig. 1. The force from the Van der Pol oscillator acts throughout the surface of the
cantilever beam. Accordingly, the equation governing the motion of the cantilever
beam can be written as

∂4V ∂2V ! ∗ "

∗ 2 ∂V ∂ 2 Mr
EI + ρ A − α − β V = , (1)
∂r 4 ∂t 2 ∂t ∂r 2

where E denotes the Young’s modulus of beam material, I = π d 4 /64 denotes the
area moment of inertia of beam cross section, ρ is the density of the beam material,
A is the cross-sectional area, V (r, t) is the transverse deflection of the beam, α ∗ and
β ∗ are the coefficients in Van der Pol oscillator, Mr is the moment applied by the
actuator on the beam [12, 13] which is given by the following expression:

Mr = Fc∗ h ∗ [u(r − r1 ) − u(r − r2 )] = K ∗ Va [u(r − r1 ) − u(r − r2 )], (2)

where Fc∗ is the control force, h ∗ is the distance of the centerline of the piezo-stack
actuator from the surface of the beam, u(·) is the unit step function, Va is the actuator
voltage, and K ∗ is a piezoelectric constant.
The governing equation in dimensionless form is obtained as

∂ 4v ∂ 2v ∂v 2 ∂v ∂2
+ − α + βv = γa [u(x − x1 ) − u(x − x2 )], (3)
∂x4 ∂τ 2 ∂τ ∂τ ∂x2
where the dimensionless quantities used are

Fig. 1 Schematic diagram

of the fixed cantilever beam
attached with a piezoelectric
stack actuator and a Van der
Pol oscillator
Analysis and Control of Vortex-Induced Vibrations in a Cantilever Beam … 1025
#
r V EI R0
x = , v = , τ = )0 t, )0 = , r0 = ,
L L ρ AL 4 L
L K ∗ Va α ∗ )0 L 4 β ∗ )0 L 6
γa = ,α = ,β = .
EI EI EI
The boundary conditions in dimensionless form are
$ $ $
∂v $$ ∂ 2 v $$ ∂ 3 v $$
v|x=0 = 0, = 0, = 0, = 0.
∂ x $x=0 ∂ x 2 $x=1 ∂ x 3 $x=1

2.1 Modal Decomposition

The partial differential equation (Eq. (3)) is reduced to an ordinary differential equa-
tion using the method of modal decomposition [14]. The modal solution is first
assumed as
n
%
v(x, τ ) = &i (x)vi (τ ), (4)
i=1

where n is the number of mode shapes to be considered in the analysis and &i (x)
is the ith orthonormal mode shape of a cantilever beam [15] given by the following
expression

sin pi − sinh pi
&i (x) = [sin pi x − sinh pi x] + (cos pi x − cosh pi x). (5)
cos pi + cosh pi

Natural frequencies pi are obtained from the frequency equation

1 + cos pi cosh pi = 0. (6)

Natural frequencies in dimensional form can be obtained as ωi = pi2 )0 . The

orthonormality conditions [16] for the mode shapes &i (x) are given by the following
expressions,

&1
&i (x)& j (x) = δi j , (7a)
0

&1 &1
d 4& j d 4 &i
&i (x) dx = & j (x) d x = pi4 δi j . (7b)
dx4 dx4
0 0
1026 D. V. Kumar et al.

In the modal decomposition method, the modal solution (4) is substituted in the
governing Eq. (3). Then, the equation is multiplied with &m and integrated over
the domain [0,1]. Subsequently, orthonormality conditions (7a), (7b) are applied to
reduce the partial differential Eq. (3) governing the system vibrations to an ordinary
differential equation for mth mode. These steps are not elaborated due to lack of
space. We straightforwardly write the ordinary differential equation for the analysis
of 1st mode of vibration as

d 2 v1 dv1 ' ′ (
2 dv1 4 ′
− α + βC v
1111 1 + p v
1 1 = −γa ψ (x
1 1 ) − ψ1 2 ,
(x ) (8)
dτ 2 dτ dτ

)1 $
dψ1 $
where C1111 = &14 d x, and ψ1′ (xi ) = d x $x=x
. Considering the case of a simple
0 i

uncontrolled system without any piezoelectric stack actuator, the above equation
becomes

d 2 v1 dv1 dv1
−α + βC1111 v12 + p14 v1 = 0, (9)
dτ 2 dτ dτ

2.2 Governing Equation with Control Force Derived from PD

Algorithm

A simple proportional-derivative (PD) control algorithm is considered for the control

vortex-induced vibrations in the cantilever beam. The voltage to be supplied to the
piezo-stack actuator derived using PD control algorithm is
n
% n
% ! "
γa (t) = −K p & j (1) · v j (t) − K d & j (1) · D v j (t) , (10)
j=1 j=1

where K p is the proportional gain, and K d is the derivative gain. Substitution of

Eq. (10) into Eq. (8) yields
! "
v1′′ + β C1111 v12 − α − m 1 · K d v1′ + ( p14 − m 1 · K p )v1 = 0, (11)
* +
where m 1 = &1 (1) &1′ (x1 ) − &1′ (x2 ) . The primes in Eq. (11) denote derivative with
respect to the dimensionless time ‘τ ’.
Analysis and Control of Vortex-Induced Vibrations in a Cantilever Beam … 1027

3 Linear Stability Analysis

The linearized form of Eq. (11) is obtained by simply removing the nonlinear term
(βC1111 v12 v1′ ) as

v1′′ − (α + m 1 · K d )v1′ + ( p14 − m 1 · K p )v1 = 0. (12)

The characteristic equation is obtained be substituting v1 = esτ in Eq. (11) as

! "
s 2 − (α + m 1 · K d )s + p14 − m 1 · K p = 0. (13)

At the Hopf bifurcation point, a pair of complex conjugate roots crosses the
imaginary axis.√Accordingly, the condition at Hopf point is obtained by substituting
s = ±iω (i = −1) in Eq. (13) as
,
ω= p14 − m 1 · K p , (14a)

(α + m 1 · K d ) = 0. (14b)

The real part of the roots becomes positive for (α + m 1 · K d ) > 0. This corre-
sponds to negative damping in linearized Eq. (12). Accordingly, the response of the
linearized system (12) will exponentially increase. Damping becomes positive for
(α + m 1 · K d ) < 0 causing the response to decay exponential toward the equilib-
rium solution v1 = 0. Clearly, the stability boundary is given by Eq. (14b). It will
be shown in Sect. 4 that both the parameters α and β in the Van der Pol oscillator
are functions of flow velocity. Hence, a critical flow velocity corresponding to Hopf
bifurcation point can be obtained from Eq. (14b). Vibrations in the cantilever beam
are induced once this critical flow velocity is exceeded.

4 Method of Averaging

In this section, the approximate analytical solution of the governing Eq. (11a) is
obtained using the method of averaging [17]. To apply the method of averaging,
Eq. (11) is written in the following form
′′
- ′
.
v1 + ω2 v1 − ϵ f v1 , v1 = 0, (15)

where
- ′
. ′ ′ ! "
ϵ f v1 , v1 = αv1 − βC1111 v12 v1 + m 1 K d , ω2 = p14 − m 1 .K p , 0 < ϵ ≪ 1.
1028 D. V. Kumar et al.

The slowly time varying solution is assumed as

v1 = A(τ ) sin(ωτ + ϕ(τ )), (16a)

′
v1 = A(τ )ω cos(ωτ + ϕ(τ )), (16b)

where amplitude ( A) and phase (ϕ) are slowly varying functions of time (τ ). Substitu-
tion of Eqs. (16a, 16b) into Eq. (15) leads to the following equations for the evolution
of amplitude and phase
ϵ - ′
.
A′ (τ ) = f v1 , v1 cos(ωτ + ϕ(τ )), (17a)
ω
ϵ - ′
.
ϕ ′ (τ ) = − f v1 , v1 sin(ωτ + ϕ(τ )). (17b)
ωA
Substituting ωτ + ϕ(τ )) = ψ(τ ) in Eqs. (17a, 17b) yields
ϵ
A′ (τ ) = cos(ψ) f (A sin(ψ), A cos(ψ)), (18a)
ω
ϵ
ϕ ′ (τ ) = − sin(ψ) f (Asin(ψ), Acos(ψ)). (18b)
ωA
By averaging Eqs. (14a, 14b) over one period gives,

&2/
′ ϵ
A (τ ) = f (Asin(ψ), Acos(ψ))cosψdψ, (19a)
2/ω
0

&2/
′ ϵ
ϕ (τ ) = − f (Asin(ψ), Acos(ψ))sinψdψ. (19b)
2/Aω
0

In our case,
! "
ϵ f v1 , v1′ = ϵ f (x = A sin ψ, A cos ψ)
* +
= αv1′ − βC1111 v12 v1′ + c1 K d v1′ (v1 =A sin ψ, v′ =Acosψ ) .
1

It is to be noted that A and ϕ are assumed to be constants while evaluating the

above integrals. The final expressions for the amplitude and phase equations obtained
using ‘MAPLE’ software are

α + c1 K d βC1111 3
A′ = A− A , (20a)
2 8
Analysis and Control of Vortex-Induced Vibrations in a Cantilever Beam … 1029

ϕ ′ (τ ) = 0. (20b)

The fixed points obtained by setting A′ = 0 are the steady-state A = 0 and the
amplitude of limit cycle
#
4(α + m 1 K d )
A= . (21)
βC1111

As explained in Sect. 3, flow-induced vibrations are obtained only when

(α + m 1 · K d ) > 0. Accordingly, real values for the limit cycle amplitude are
obtained when β > 0, as C1111 > 0 always.

5 Results and Discussion

All the results are discussed in this section. The analysis of the uncontrolled system
is presented in Sect. 5.1 followed by the analysis of the controlled system in Sect. 5.2.

5.1 Parametric Study of the Uncontrolled System

According to the literature [18], Van der Pol equation displays a large degree of
physical coherence with FIV generated through fluid–structure interactions. Exper-
imental data picked up from the work of Fujarra et al. [4] is plotted in Fig. 2 which
shows that the amplitude of FIV is a function of flow velocity. The experimental
results in Fig. 2 are normalized so as to match the limit cycle amplitudes A obtained
from Eq. (21). Accordingly, Van der Pol coefficients α and β should be functions of
flow velocity.

Fig. 2 Graph of amplitude

response in the transverse
direction (A) versus flow
velocity (U) obtained using
strain gage (open circle) and
optical transducer (solid
circles) by Fujarra et al. [4]
1030 D. V. Kumar et al.

Fig. 3 Comparison of
amplitude of FIV obtained
from Eq. (19a) (solid line)
with experimental results
(solid circles)

To obtain the functional forms of α and β, a detailed parametric analysis is

performed. Various polynomial functions for α and β are chosen, and the exper-
imental results shown in solid circles in Fig. 2 are curve fitted with the amplitude A
in Eq. (21). From all the diagrams, the one which qualitatively matches the experi-
mental plot in the literature (solid circles in Fig. 2) with minimal error is considered.
This gives us α = 4.297 U − 0.134 ; β = (2.778e(+03))U 4 − 2.184e(+03)U 3 +
6.302e(+02)U 2 − 78.683U + 3.611)·105 . Bifurcation diagram obtained with these
expressions for α and β is plotted along with the experimental results in Fig. 3. The
root mean square error between the experimental and theoretical results in Fig. 3 is
1.75943686104E-03. In the subsequent analysis, these functional forms of α and β
are used.
Once the expressions for α and β are obtained, the result obtained from the
averaging method is compared with the result obtained by numerically simulating
the governing Eq. (11) for the flow velocity U = 0.15 as shown in Fig. 4. In the time-
displacement diagram in Fig. 4a, the displacement (&1 (1)ν) of the tip of the cantilever
beam is plotted, whereas the phase-plane diagram in Fig. 4b is obtained by plotting
the displacement (&1 (1)ν) and velocity (&1 (1)ν ′ ) of the tip of the cantilever beam.
It is observed from Fig. 4 that the amplitude of oscillation predicted by averaging
method matches very closely with those obtained from numerical simulation.

5.2 Analysis of the Controlled System

The reliability of the averaging method in predicting the dynamics of the controlled
system is first verified by comparing the averaging method results with those obtained
using the numerical simulation of the governing Eq. (11). The comparisons are shown
in Figs. 5 and 6. The time-displacement response and phase-plane diagrams in Fig. 5
are plotted for the control parameters K d = 0, K p = 3, and those in Fig. 6 are
Analysis and Control of Vortex-Induced Vibrations in a Cantilever Beam … 1031

Fig. 4 Comparisons of a time-displacement response and b phase-plane diagram of the uncon-

trolled system obtained using numerical simulation of Eq. (11) and analytical averaging method.
Parameters: U = 0.15

plotted for the control parameters K d = 0.4, K p = 3. It is observed from Figs. 5

and 6 that the averaging method results match well with the numerical simulation
results. The match becomes better for smaller amplitude of oscillations obtained for
higher values of derivative gain K d .
The comparisons of phase-plane diagrams obtained by numerically simulating
Eq. (11) are shown in Fig. 7. The uncontrolled system (K d = K p = 0) is compared
with the controlled system for K d = 0, K p = 3 in Fig. 7a, whereas Fig. 7b shows

Fig. 5 Comparisons of a time-displacement response and b phase-plane diagram of the uncon-

trolled system obtained using numerical simulation of Eq. (11) and averaging method. Parameters:
U = 0.15, K d = 0, K p = 3, x1 = 0, x2 = 0.1
1032 D. V. Kumar et al.

Fig. 6 Comparisons of a time-displacement response and b phase-plane diagram of the uncon-

trolled system obtained using numerical simulation of Eq. (11) and analytical averaging method.
Parameters: U = 0.15, K d = 0.4, K p = 3, x1 = 0, x2 = 0.1

the comparison for two different values of derivative gains (K d = 0 and K d = 0.4),
keeping the proportional gain the same at K p = 3. In all the cases discussed in
this paper, the piezo-stack actuator is considered to be attached on the surface of
the cantilever beam at the fixed end between x1 = 0 to x2 = 0.1. It is observed
from Fig. 7a that the displacement amplitude remains unchanged if a control force
is applied without any derivative gain (K d = 0). The independence of K p in the
expression for the amplitude of limit cycles obtained using the averaging method
(given by Eq. (21)) further verifies this observation. The frequency of oscillations
(refer Eq. (14a)) is increased for positive proportional gain K p causing an increase
in the velocity amplitude as shown in Fig. 7a. The ratio of the velocity and the
displacement amplitudes is approximately ω, the frequency of oscillations (refer
Eqs. (16a) and (16b)). Addition of a small amount of derivative gain in the control
force significantly reduces the amplitude of oscillation as shown in Fig. 7b.
The evolution of time-displacement response for different values of derivative
gains K d is shown in Fig. 8a. It is observed from Fig. 8a that the frequency of oscil-
lations for all the cases remains the same. This is because of the same proportional
gain K p = 3 considered for all the cases. The amplitude of oscillations gradually
decreases with the increase of derivative gain until a complete stability of steady
state is achieved for K d = 0.6. This is evident from both the time-displacement and
phase-plane diagrams in Fig. 8.
The efficacy of the control force is further verified by plotting the bifurcation
diagrams for different values of K d in Fig. 9. In the bifurcation diagrams, the displace-
ment amplitude A of flow-induced vibrations obtained from Eq. (21) is plotted with
respect to the flow velocity U. It is observed from Fig. 9 that the critical flow velocity
corresponding to the Hopf bifurcation point increases with the increase of K d . The
Analysis and Control of Vortex-Induced Vibrations in a Cantilever Beam … 1033

Fig. 7 Comparison of phase-plane diagrams obtained from the numerical simulation of Eq. (11).
Parameters: U = 0.15, x1 = 0, x2 = 0.1, a K d = 0, b K p = 3

Fig. 8 a Time-displacement response showing the evolution of amplitude of oscillations and b

the phase-plane diagrams for different values of derivative gains. Parameters: U = 0.15, K p = 3,
x1 = 0, x2 = 0.1

exact value of the critical flow velocity for a particular set of control gains can be
obtained from Eq. (14b) derived from linear stability analysis.

6 Conclusions

The major objective of this paper is to derive a mathematical model capable of

predicting the vortex-induced vibrations (VIV) in a cantilever beam observed from
1034 D. V. Kumar et al.

Fig. 9 Bifurcation diagram

corresponding to various K d
values. Parameters: x1 = 0,
x2 = 0.1, K p = 3

experiments. Subsequently, PD control of flow-induced vibrations using a piezo-

stack actuator attached on the surface of the cantilever beam is explored. The Euler–
Bernoulli beam theory is used to construct a mathematical model of the cantilever
beam system. In the governing equation, the Van der Pol term is viewed as an approx-
imation of flow-induced vibrations. In accordance with the method of modal decom-
position, the governing partial differential equation along with the boundary condi-
tions is transformed into a set of ordinary differential equations. Linear stability anal-
ysis predicts the loss of stability of steady states via Hopf bifurcation. An expression
is derived from the linear stability analysis using which the critical flow velocity can
be obtained. The method of averaging is used to derive analytical results for ampli-
tude and phase equations. The analytical results so obtained are found to match well
with the numerically simulated results, particularly for smaller amplitude of oscilla-
tions. A detailed parametric study is conducted to obtain functional form of the Van
der Pol coefficients in terms of flow velocity so that the experimental results could
be predicted with negligible error.
PD control algorithm is derived to control the VIV in the cantilever. The propor-
tional gain only changes the frequency of oscillations. To reduce the displacement
amplitude, derivative gain is needed. Bifurcation diagrams, wherein the displace-
ment amplitudes of the VIV are plotted for different values of derivative gains, show
the critical flow velocity corresponding to the Hopf bifurcation point increases with
the increase of the derivative gain. The displacement amplitude decreases with the
increase of derivative gain, and complete quenching is achieved once a critical value
is exceeded.
The model developed in this paper is capable of quantitatively predicting the
experimentally obtained flow velocity-response amplitude relations only. To describe
other experimentally observed phenomena in VIV, an extension of the present model
should be developed taking into account the fluctuating forces on the cylinder due to
vortex shedding. This will be explored in future.
Analysis and Control of Vortex-Induced Vibrations in a Cantilever Beam … 1035

References

1. Ali U, Islam M, Janajreh I, Fatt Y Mahbub Alam Md (2021) Flow-Induced vibrations of single
and multiple heated circular cylinders: a review. Energies 14(24):8496
2. Chirag T (2017) A review on fluid structure interaction in hydraulic turbines: a focus on
hydrodynamic damping. Eng Fail Anal 77:1–22
3. Aoi H, Zhao Y, Hayase T (2002) Improvement of the damaged thermowell of the monju
fast-breeder reactor based on vibration analysis. SIMULATION: Trans Soc Model Simul Int
78(8):504–511
4. Fujarra ALC, Pesce CP, Flemming F, Williamson CHK (2001) Vortex—induced vibration of
a flexible cantilever. J Fluids Struct 1:651–658
5. Facchinetti ML, de Langre E, Biolley F (2004) Coupling of structure and wake oscillators in
vortex-induced vibrations 19:123–140
6. Vigneshvar V, Vishwanth SR, Venkateshwaran S, Balaram B (2019) A comparative study of
wake oscillator models for flow induced vibrations. AIP Conf Proc 2134:080006
7. Barrón MA, Sen M (2009) Synchronization of coupled self-excited elastic beams. J Sound Vib
324:209–220
8. Acciani G, Adamo F, Modugno FD, Gelao G (2015) Modeling and simulation of cantilever
beam for wind energy harvesting. J Vibroengineering 18(2):1167–1174
9. Asghar U, Aziz I, Sher F (2017) Modelling and simulation of flow-induced vibrations in
vertical axis wind turbine blade. In: 14th international Bhurban conference on applied sciences
and technology, pp 635–645
10. Syed-Aghazadeh B, Samandari H, Dulac S (2020) Flow-induced vibration of inherently
nonlinear structures with applications in energy harvesting. Phys Fluids 32(7):1–6, 071701
11. Al-Ashtari KW (2011) The deflection control of a thin cantilever beam by using a piezoelectric
actuator/sensor. In: 1st Regional conference of engineering science NUCEJ 11, 43–51
12. Saha A (2018) Time-delayed control of friction-induced vibrations in a continuous system. Int
J Mech Prod Eng Res Dev 111–118
13. Li W, Yang Z, Li K, Wang W (2021) Hybrid feedback PID-FxLMS algorithm for active vibration
control of cantilever beam with piezoelectric stack actuator. J Sound Vib 509:116243
14. Astrom KJ, Hagglund T (1995) PID controllers: theory, design, and tuning, 2nd edn. Instrument
Society of America, USA
15. Kelly SG (2000) Fundamentals of mechanical vibrations, 2nd edn. McGraw-Hill Higher
Education, Singapore
16. Hongliang H, Zhenqiang L, Xiangyan Z (2017) The self-excited vibrations of an axially
retracting cantilever beam using the Galerkin method with fitted polynomial basis functions. J
Mech Sci Technol 32(1):29–36
17. Sanders JA, Verhulst F, Murdock J (2007) Averaging methods in nonlinear dynamical systems.
Applied mathematical sciences. Halliday Lithograph, USA
18. Sloss JM, Bruch JC Jr, Adali S, Sadek IS (2001) Piezoelectric patch control using an integral
equation approach. Thin-Walled Struct 39(1):45–63