The Shock and Vibration Digest

http://svd.sagepub.com

Active Structural Vibration Control: A Review

Rabih Alkhatib and M. F. Golnaraghi
The Shock and Vibration Digest 2003; 35; 367
DOI: 10.1177/05831024030355002

The online version of this article can be found at:

http://svd.sagepub.com/cgi/content/abstract/35/5/367

Published by:

http://www.sagepublications.com

Additional services and information for The Shock and Vibration Digest can be found at:

Email Alerts: http://svd.sagepub.com/cgi/alerts

Subscriptions: http://svd.sagepub.com/subscriptions

Reprints: http://www.sagepub.com/journalsReprints.nav

Permissions: http://www.sagepub.com/journalsPermissions.nav

Citations (this article cites 117 articles hosted on the

SAGE Journals Online and HighWire Press platforms):
http://svd.sagepub.com/cgi/content/refs/35/5/367

Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008

© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Articles
Active Structural Vibration Control: A Review
Rabih Alkhatib and M. F. Golnaraghi
ABSTRACT—In this paper we review essential aspects in-
volved in the design of an active vibration control system.
We present a generic procedure to the design process and
give selective examples from the literature on relevant ma-
terial. Together with examples of their applications, various
topics are briefly introduced, such as structure modeling,
model reduction, feedback control, feedforward control, con-
trollability and observability, spillover, eigenstructure assign-
ment (pole placement), coordinate coupling control, robust
control, optimal control, state observers (estimators), intelli-
gent structure and controller, adaptive control, active con-
trol effects on the system, time delay, actuator–structure
interaction, and optimal placement of actuators.
KEYWORDS: structures, vibration, model reduction, active
control, control algorithms, actuator dynamics, time delay.
1. Introduction
The control of mechanical and structural vibration has sig-
nificant applications in manufacturing, infrastructure engi-
neering, and consumer products. In the machine tool indus-
try, mechanical vibration degrades both the fabrication rate
and quality of end products. In civil engineering constructs,
structural vibration degrades human comfort. In automo-
tive and aerospace fields, vibration reduces component life,
and the associated acoustics noise annoys passengers.
Various methods have been applied to vibration control
in the engineering field. Traditionally, passive isolators and
dampers are used to attenuate mechanical vibrations. For
example, installing rubber mounting between the machines
and foundations (Harris, 1997; Nakra, 1998) or adding pas-
sive dampers to structures (Soong and Dargush, 1997) are
common practices in vibration isolation and attenuation.
Recent advances in digital signal processing (DSP) and
sensors and actuators technology have prompted interest in
active vibration control (Inman, 1989; Meirovitch, 1990;
Soong, 1990; Preumont, 2002, Clark et al., 1998).
Active vibration control is a vast research area that incor-
porates interdisciplinary technologies. For example, a typ-
ical active vibration control system is an integration of
mechanical and electronic components in synergistic com-
bination with computer/microprocessor control. The major
components of any active vibration control system are the
Dr Rabih Alkhatib, Postdoctoral Fellow, and Professor M. F. Golnaraghi
(Canada Research Chair, Intelligent Mechatronics and Materials Systems),
Department of Mechanical Engineering, University of Waterloo, 200 Uni-
versity Avenue, Waterloo, Ontario, Canada N2L 3G1.
The Shock and Vibration Digest, Vol. 35, No. 5, September 2003 367–383
©2003 Sage Publications
Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008
© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
mechanical structure influenced by disturbance (creating
unwanted vibration), sensors (to perceive the vibration),
controllers (to intelligently make use of the signals from
the sensors and to generate the appropriate control signals),
and actuators (which counteract the influence of the distur-
bance on the structure). Destructive interference from the
forces generated by the actuators reduces and/or cancels
the effects of the disturbance on the structure.
In the past few decades, various methods of active vibra-
tion control have been developed. The different algorithms
can be classified under two general categories: feedback
and feedforward active vibration control. Variations of the
two general methods exist, each with advantages, disad-
vantages, and limitations.
This paper is intended as a guide for the issues involved
in designing an active control system. The paper further
provides a review of active vibration control techniques,
with examples from mechanical and civil engineering appli-
cations. Central topics are briefly introduced and relevant
references in the literature are provided.
2. Development of Active Vibration Control
System
The process of designing an active control system for the
attenuation of vibration in machines and structures gener-
ally involves many steps. A typical scenario is as follows.
1. Analyze the vibratory system, e.g. machine or struc-
ture, to be controlled.
2. Obtain an idealized mathematical model of the system
to be controlled utilizing tools such as finite element
analysis or experimental modal analysis.
3. Reduce and simplify the model if necessary so that it is
amenable.
4. Analyze the resulting model; determine its properties,
dynamic characteristics, and types of disturbance and
system response.
5. Quantify sensors and actuators requirements and decide
on their types and where they will be placed.
6. Analyze the impact of the sensors and actuators on the
overall dynamic characteristics of the system.
7. Specify performance criteria and stability tradeoffs.
8. Decide on the type of control algorithm to be employed
and design a controller to meet the specifications.
9. Simulate the resulting controlled system on a computer
and assess the potential to meet requirements
368 The Shock and Vibration Digest / September 2003

Disturbance Performance
System
specification objectives

Sensors / Actuators
selection and placemen

Controllability
Identification Model
Observability

Model reduction

Controller
Sensors / Actuators
continuous
dynamics
design

Digital
implementation

iterate until
performance
objectives
are met
Closed loop
system

Evaluation

Figure 1. The various steps of an active system design (Preumont, 2002).

10. If the controller does not meet the requirements, adjust
the specifications or modify the type of controller sought.
11. Choose hardware and software and integrate the com-
ponents on a pilot plant.
12. Formulate experiments and perform system identifica-
tion and model updating.
13. Implement controller and carry out system test to eval-
uate the overall performance of the system.
14. Repeat some or all of the above steps if necessary.
The various steps in the design of a controlled structure are
depicted in Figure 1 (Preumont, 2002). What follows is a
survey of some of the current literature on modeling and
active vibration control of structure in the context of the
above points.
3. Structure Modeling
The first step in designing a control system is to build a
mathematical model of the system and disturbances. Prin-
ciples of mechanics and physical laws are employed to
derive a representative, idealized model (Meirovitch, 1997).
Based on the mathematical model, a controller can be
designed to provide desirable performance. Mechanical
and structural systems are often modeled as either lumped
or distributed parameter systems. Lumped systems are math-

Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008

© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 Alkhatib and Golnaraghi / ACTIVE STRUCTURAL VIBRATION CONTROL: A REVIEW
ematically represented by ordinary differential equations
while distributed parameter systems are governed by par-
tial differential equations. Usually systems have compli-
cated shapes and structural patterns, which make the devel-
opment and the solution of descriptive partial differential
equations burdensome, if not impossible. Alternatively,
various discretization techniques, such as finite element
(FE), modal analysis, and lumped parameters, allow us to
approximate the partial differential equations by a finite set
of ordinary differential equations. Several authors have used
FE analysis to study vibration of flexible plate-like struc-
tures in general (Arabyan et al., 1999; Chen, 1999; Hanagan
and Murray, 1997). The structure is divided into grids, the
distributed mass is lumped at the nodes, and applying the
appropriate boundary conditions gives the following gen-
eral form
&& + Dq& +Kq = Nu + f
Mq
y = Pq +Rq& (1)
where M, D, and K are the lumped mass, damping and
stiffness matrices, respectively, q = [ q 1 , q 2 , ..., q n ] T , q& and
&&q are node displacement, velocity, and acceleration vectors
of n dimension, f is the disturbance vector, y is the output
of the system measured by the sensors, and the matrices P
and R give the location of and electronic gains associated
with sensors used to measure the state variables.
Since the damping in a structure is rather poorly known
and is expected to be quite small, it is usually assumed that
D = 0. When the effects of the damping are needed in the
model, Rayleigh proportional damping is usually introduced
by structural dynamists:
D = α 1 M + α 2 K. (2)
This is a convenient damping model for analysis and simu-
lation but not necessarily an accurate model since it tends
to overestimate the damping of high-frequency modes
(Preumont, 2002). Proportional damping allows simulta-
neous diagonalization of the mass, stiffness, and damping
matrices through coordinate transformation. Therefore, the
system coupled equations of motion can be transformed to
a system of uncoupled equations that can be easily solved.
This model is also referred to as modal damping (Inman,
1989).
The mode shapes φk and mode frequencies ωk are calcu-
lated from the generalized eigenproblem associated with
equations (Maia et al., 1997):
(K - ω2k M) φk = 0. (3)
Modal coordinates z are obtained by the transformation
q = Φp (4)
where Φ= [ φ1 , φ2 , ..., φn ]. Thus
Φ T MΦ = I (5)
which implies
Φ T KΦ = Λ = diag[ ω12 , ω22 , ..., ωn2 ] (6)
Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008
© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
369
and
Φ T D Φ = Ξ = diag[ 2ξ1 ω1 , 2ξ 2 ω2 , ..., 2ξ n ωn ]. (7)
The system in modal coordinates:
&& + Ξp& + Λp = Φ T Nu + Φ T f .
p (8)
From the control standpoint, plant (i.e. dynamic system)
models are generally expressed in the state-variable or
first-order format. For example, with z as the state variable
vector, u as the input vector, and y as the measurement
vector, the classical first-order form is
z& = Az + Bu
y = Cz . (9)
In control applications, the second-order model is frequently
converted to a first-order model by using the following equa-
tions:
ìqü
z =í ý
îq& þ
é 0 I ù
A =ê
ë-M -1 K -M -1 D ú
û
ìM -1 Nü
B=í ý
î 0 þ
C = [P R] (10)
The integration of structural modeling and control is very
critical in the development of active vibration control sys-
tems. The designers and engineers should consider the
incorporation of active control devices at an early stage of
the mechanical and structural design process. Sophisticated
tools are needed for modeling mechatronic systems com-
prising a variety of technologies and energy domains
(Karnopp et al., 2000).
A large part of the control theory deals with continuous
systems characterized by analog signals, which use differ-
ential equations as a basis for modeling. Nonetheless, the
nature of the digital computer makes it convenient to model
systems in terms of the discrete-time domain for which dif-
ference equations provide the appropriate mathematical
basis. A discrete-time model can be readily obtained from
the continuous-time state-space equation by a zero-order
hold on the input. A zero-order hold takes a continuous sig-
nal and turns it into a stepwise signal in which the signal is
sampled and held constant for a certain interval of time,
called the sampling period. Let t = k ∆t, where D t is the
sample period and k is an integer, then z ( k) = z ( k ∆t),
u( k) = u( k ∆t), and y( k) = y( k ∆t). The corresponding dis-
crete-time state-space model is of the form (Ogata, 1995)
z ( k +1) = Ψ( k) z ( k) + Θ( k) u( k)
y( k) = Cz ( k) (11)
where Y is the discrete-time state transition matrix defined
as
370 The Shock and Vibration Digest / September 2003

d(s)
r(s) e(s) y(s)
- h(s) g(s)

Figure 2. Principle of feedback control.

Ψ( k) = e A∆t (12) δq T f = δq 1T LT f = δq 1T f$ (17)

and where f$ = LT f .

Θ( k) = (ò0
∆t
)
e Aτ dτ B = (e A∆t - I)A -1 B.
Finally, the reduced equation of motion reads
$ && +Kq
$ = f$.
Mq 1 1 (18)

4. Model Reduction More sophisticated methods have been developed and can
be found in Lee and Tsuha (1994), Friswell et al. (1995,
Finite element modeling of flexible structures produces far 1996, 1998), Bouhaddi and Fillod (1996), Dyka et al. (1996),
too many degrees of freedom to be tractable for practical Qu (1998) and Mykleburst and Skallerud (2002). The reduced
control design. Large-order FE stress models are tradition- model will still require further reduction for use in control
ally reduced to smaller order for use in dynamic analysis. system design and implementation. In control design, the
The employment of reduction techniques, such as the Guyan
model is converted to state-space form and reduced by one
reduction (Guyan, 1965), removes some of the insignifi-
of the reduction methods presently employed by the control
cant physical coordinates. The degrees of freedom kept in
community (Moore, 1981; Pernebo and Silverman, 1982;
the reduced model, denoted q 1 in what follows, are called
Yae and Inman, 1993). These methods in essence approxi-
primary or master coordinates; those to be eliminated,
mate a large dynamic system with a fewer number of state
denoted q 2 , are frequently referred to as secondary or slave
coordinates. To begin with, consider the undamped forced variables while making minimal change on the input–out-
vibration of the structure. The governing equation of motion put characteristics.
in partitioned form is
5. Vibration Control Strategies
é M 11 M 12 ùìq
&& 1 ü éK 11 K 12 ùìq 1 ü ìf1 ü
êM úí ý+ê í ý= í ý. (14)
ë 21 && 2 þ ëK 21
M 22 ûîq K 22 ú
ûîq 2 þ îf 2 þ Vibrations can be depicted in a number of ways, with the
most common descriptions being in terms of modal modes
and in terms of wave motion. Active vibration control can
The slaves q 2 are completely determined by the masters q 1
be designed in terms of either modal or wave characteris-
ìq 1 ü ì I ü tics, with each method having advantages and disadvan-
q = í ý= í ýq 1 = Lq 1 (15) tages (Fuller et al., 1997). In broad terms, modal control
îq 2 þ î(1- β)K 22 K 12 +βM 22 M 12 þ
-1 -1
aims to control the global behavior (i.e. the modes of vibra-
tion) of the structure while wave control aims to control the
where β= 0 for static reduction (Guyan reduction) and β=1 flow of vibration energy through the structure. The two
for dynamic reduction. types of control strategy are realized using feedback and
The reduced mass and stiffness matrices are obtained by feedforward controllers, respectively.
substituting the above transformation into the kinetic and
strain energy 5.1. Feedback Control
1 1 1 $& The scheme of feedback control is depicted in Figure 2.
T = q& T Mq& = q& 1T LT MLq& 1 = q& 1T Mq
2 2 2 1
The error signal, e = r - y, is generated from the compari-
son of the output y of the system with the reference input r.
1 T 1 1 $ The error signal is passed into a compensator h(s) and
U= q Kq = q 1T LT KLq 1 = q 1T Kq 1 (16)
2 2 2 applied to the system g(s). In designing the controller, we
seek to determine the appropriate compensator h(s) to
$ = LT ML and K$ = LT KL.
where M induce the sought-after performance without affecting the
The reduced load is obtained by the equation of virtual closed-loop system stability. Furthermore, application of
work feedback in the control of lightly damped structures can be

Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008

© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 Alkhatib and Golnaraghi / ACTIVE STRUCTURAL VIBRATION CONTROL: A REVIEW
Disturbance
Plant
Actuator
Input
Adaptive
Filter
Reference
Signal
Figure 3. Principle of feedforward control.
separated into two distinct types: active damping and model-
based feedback.
The objective of active damping is to reduce the reso-
nant peaks of the closed-loop transfer function
y( s) g( s) h( s)
f ( s) = = .
r( s) 1+ g( s) h( s)
In this case f (s) is very close to g(s), except near the reso-
nance peaks where the amplitude is reduced. Active damping
can generally be achieved with moderate actuator efforts.
In addition, the method can be realized without a model of
the structure and with guaranteed stability, granted that the
actuators and sensors are collocated (i.e. the actuators exert
their forces in the same location in which the sensors observe
the motion of the structure, and have negligible dynamics).
In practice, it can be very difficult to achieve a perfect sen-
sor–actuator collocation, and in actual systems, actuators
and sensors always have significant dynamics.
Another highly involved objective of active control is to
keep a control variable (position, velocity, or acceleration)
to a desired value in spite of external disturbances d(s) in
some frequency range. It can be seen from
y( s) 1 itative properties of dynamic systems (Chen, 1984; Antsaklis
= (20)
d( s) 1+ g( s) h( s)
that large values of g(s)h(s) are required to reduce the effect
of external disturbances in the frequency range where the
disturbance has considerable effect. From equation (19),
g(s)h(s) >> 1 entails that the closed-loop transfer function
f(s) is close to 1, which means that the output y(s) tracks
the input r(s) accurately. In general, a more intricate approach
involving a mathematical model of the system is needed to
achieve such an objective. At best, a mathematical model
can only be a reduced order approximation of the actual
system g(s). Therefore, the controller bandwidth, ωc , and
effectiveness are restricted by the accuracy of the model.
Unmodeled structural dynamics (residual modes) outside
ωc may destabilize the system.
5.2. Feedforward Control
Feedforward control methods can significantly improve
performance over simple feedback control whenever a sig-
nal correlated to the disturbance is available. For example,
Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008
© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
371
in rotating machinery, vibration is typically associated with
the spinning frequency of the rotor. A tachometer signal
can be used in implementing feedforward adaptive filtering
for disturbance rejection. The original development of the
technique is tied to noise control but eventually found its
way to vibration control (Fuller et al., 1997). The scheme
of feedforward control is shown in Figure 3.
In the most ideal situation, feedforward control can com-
pletely eliminate the effect of the measured disturbance on
Error the system. An adaptive filter manipulates the signal that is
Signal correlated to the primary disturbance and the output is
applied to the system by the actuator. The filter coefficients
are adapted in such a way that an error signal at one or sev-
eral critical points is minimized. The idea is to generate a
secondary disturbance, which destructively interferes with
the effect of the primary disturbance at the location of the
error sensor. Unfortunately, there is no guarantee that the
global response is also reduced at other locations, which
represents a drawback to the method. If the structural response
is not dominated by a single mode, the actuator may reduce
the vibration at the point of error measurements but introduce
higher levels of vibration in other parts of the structure.
(19) Thus, the method is considered to be a local technique, in
contrast to feedback which is global.
Contrary to active damping which can only reduce the
vibration near the resonance, feedforward control can be
effective for any frequency. The method works without the
need to model the system, but the adaptation process involves
measurements of the impulse response. The approach has
many applications in the cancellation of narrowband distur-
bances, but wideband applications have also been reported
(Vipperman et al., 1993). The major restriction to the
application of feedforward adaptive filtering is the accessi-
bility of a reference signal correlated to the disturbance.
The major aspects of feedback and feedforward control are
summarized in Table 1.
6. Controllability and Observability
Controllability and observability are two fundamental qual-
and Michel, 1997; Kuo and Golnaraghi, 2003). Generally
speaking, controllability corresponds to the ability of steer-
ing the state vector from one vector value to any other vec-
tor value in finite time. A system is said to be controllable
if every state vector z(t) can be transformed to a desired
state in finite time by the application of unconstrained con-
trol inputs u(t). Evidently then, an uncontrollable system is
one where some elements of the state vector z(t) cannot be
affected by the control input. Observability analyzes the
deducibility of the initial state vector of the system from
knowledge of the input and the corresponding output over
time. A system is said to be observable at time t0 if for a
state z(t0) at that time, there is a finite time t1 > t0 such that
knowledge of the input u(t) and output y(t) from t0 to t1 are
sufficient to determine the state z(t0). Following this, an
unobservable system is one where the values of some ele-
ments in the state vector at time t0 may not be determined
from the examination of the system output regardless of the
number of observations taken. Both properties can be deter-
mined by the ranks of the system matrices. A basic result in
372 The Shock and Vibration Digest / September 2003

Table 1. Comparison of control strategies.

Type of Control Advantage Disadvantage

Feedback

Active damping • Simple to implement and requires fewer computation • Effective only near resonance
• Does not require accurate model of the plant
• Guaranteed stability when actuators and sensors are
collocated

Model Based • Global method

• Limited bandwidth
(LQG, H¥ , …) • Requires accurate model of the plant
• Requires low delay for wide bandwidth
• Attenuates all disturbance within the control band-
• Spillover
width
Feedforward
Adaptive filtering of • No model is necessary • Reference/error signal is required
reference (x-filtered • Robust to inaccuracies in plant estimate and to • Local method and may amplify vibration
LMS) change in plant transfer functions somewhere else
• More effective for narrowband disturbance • Large amount of real-time computations

control theory is that a system in state-space form is con-
trollable if and only if the matrix
[B AB A 2 B L A 2 n-1 B]
has rank 2n. Similarly, a state-space system is observable if
and only if
é C ù
ê CA ú
ê ú
ê CA 2 ú
ê ú
ê M ú
ê
ëCA 2 n-1 ú
û x r are either uncontrollable (Arc =0, B r =0) or unobservable
has rank 2n.
Hughes and Skelton (1980) developed simple controllabil-
ity and observability conditions of linear matrix second-order
systems. The controllability can be used to measure the abil-
ity of a particular actuator configuration to control all the
states of the system; conversely, observability can be
employed to measure the ability of a particular sensor con-
figuration to supply all the information necessary to esti-
mate all the states of the system. For example, Gawronski
(1997a) has addressed the problem of actuator and sensor
placement using the system notion of modal controllability
and observability. Other related work can be found in Triller
and Kammer (1994), Gawronski and Lim (1993a), Wang
and Wang (2001), Chen et al. (2001), Gorain and Bose
(2002), Slater and Inman (1997), Yang et al. (1994) and
Gawronski (1997b).
7. Spillover
The mathematical model of structure and the control-
ler design are not independent aspects of vibration control.
Flexible structures are distributed parameter systems that
have an infinite number of degrees of freedom. Controller
design often requires a mathematical model of the struc-
ture, which is at best a reduced-order model with only a
finite number of coordinates. A feedback controller based
on a reduced model can destabilize the residual modes
(21) (unmodeled dynamics) (Meirovitch, 1987). The nature of
instability can be seen in Figure 4, where C(Ac, B c, C c)
denotes the controller based on a reduced model (Ac, B c,
C c). The residual modes are described by (Ar, B r, C r). The
controller is designed to reduce the vibration in (Ac, B c,
C c). The feedback excites the unmodeled state x r via the
term B r u(t) (called control spillover), and the sensor sig-
(22) nals are contaminated by the residual modes via the term
C r x r (called observation spillover). The controller will not
destabilize the closed-loop system if the unmodeled states
(Acr =0, C r =0). Methods to reduce the effects of spillover
are investigated in Balas (1978), Mei and Mace (2002) and
Kim and Inman (2001).
8. Eigenstructure Assignment (Pole Placement)
This is a simple method of designing a feedback control
system. The objective of the method is to ensure that the
eigenvalues of the closed-loop system matrix are closer to
those specified by the designer than those of the open-loop
system. For example, the closed-loop eigenvalues may be
moved further into the left half of the complex plane to
improve stability robustness. The method works in the fol-
lowing way.
We assume the mechanical system (before control) is
governed by equations (1). If u is the feedback control
signal of the output feedback form u =-Gy, then the
closed-loop model:
&& +( D + NGR) q& +(K + NGP) q = 0.
Mq (23)
The stability of the system can be assessed by solving the
eigenvalue problem
Mλ2 +( D + NGR) λ+(K + NGP) = 0. (24)

Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008

© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
 Alkhatib and Golnaraghi / ACTIVE STRUCTURAL VIBRATION CONTROL: A REVIEW 373

Figure 4. Control and observation spillover.

The closed-loop system is stable as long as all the

eigenvalues have negative real parts. Hence, the designer
seeks to find a feedback gain matrix G that satisfies the
above condition. A special class of controller is obtained
when actuators and sensors are collocated. The conditions
NGP = NG 1 N T and NGR = NG 2 N T imply collocated sen-
sors and actuators which simplifies the controller design
and theoretically guarantees stability regardless of the mod-
eling error in M, D, and K as long as M > 0, D ³0, K > 0,
G 1 > 0 and G 2 > 0. The asymptotic stability of this special
class of feedback controller can be verified by considering
the following Lyapunov function:
Figure 5. Asymptotic state observer.
1 T 1
V= q& Mq& + q T (K + NG 1 N T ) q. (25)
2 2 (1997), Juang and Maghami (1990), Junkins and Kim (1990),
Slater and Zhang (1990), Inman (2001) and Chu (2002).
By differentiating the preceding equation with respect to
Hanagan and Murray (1997) used velocity feedback to
time and substituting the closed-loop equation (23), V& can
be written as actively control floor vibration. Such a technique is known
as active vibration damping or a skyhook damper.
V& =-q& T ( D + NG 2 N T ) q& . (26)
9. Coordinate Coupling Control
From equations (25) and (26), the Lyapunov function V is
positive definite and V& is negative semidefinite for all posi- Coordinate coupling control (CCC) is an energy-based
tive definite G 1 and G 2. Therefore, the closed-loop system method that is either linear or nonlinear. Initially, Golnaraghi
is asymptotically stable in the Lyapunov sense even with (1991) proposed a nonlinear CCC to eliminate the transient
arbitrary finite variation in M, D, and K (as long as their vibration of an oscillatory system. Later, this technique was
definiteness properties are maintained). However, such sta- extended to accommodate the steady-state vibrations as well
bility robustness falls short in the presence of real hardware (Khajepour and Golnaraghi, 1997; Oueini et al., 1997, 1998,
because the actuators and sensors cannot be perfectly aligned 1999a, 1999b; Ashour and Nayfeh, 2002; Pai and Schulz,
and their dynamics cannot be ignored (see Section 17). 2000; Pai et al., 1998, 2000; Tuer et al., 1991; Duquette et
The implementation of the pole assignment method to al., 1991; Salemi et al., 1997; Fariborzi et al., 1997). This
vibration control is investigated in Hanagan and Murray method provides a bridge to transfer the vibrations from a

Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008

© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
374 The Shock and Vibration Digest / September 2003
Figure 6. Coordinate coupling.
low or undamped oscillatory system (plant) to a damped
system (controller). In effect, this technique introduces a
single-degree-of-freedom or multi-degree-of-freedom lin-
ear or nonlinear absorber to suppress vibrations of a plant.
The key advantage of this technique is that it can be imple-
mented in a virtual environment by replacing the physical
absorber with an equivalent computer model. As a result,
the coupling terms can be selected such that the energy
transfer from the plant to the controller is maximized. The
system in Golnaraghi (1991) is depicted in Figure 6 and
consists of a spring–mass–dashpot mechanism attached to
the end of a cantilever beam. The overall system is charac-
terized by two degrees of freedom with dynamic quadratic
nonlinearities. The linear spring is tuned such that the lin-
ear natural frequencies of the two modes of oscillation are
in the ratio of 2:1. Golnaraghi has shown that under this
condition a state of internal resonance is established result-
ing in a maximum transfer of energy from the beam to the
slider. Control is achieved by the slider absorbing the vibra-
tion of the beam. This passive method is developed to an
active technique (Khajepour and Golnaraghi, 1997; Oueini et
al., 1997, 1998, 1999a, 1999b; Ashour and Nayfeh, 2002;
Pai and Schulz, 2000; Pai et al., 1998, 2000; Tuer et al.,
1991; Duquette et al., 1991; Salemi et al., 1997; Fariborzi
et al., 1997). Tuer et al. (1991) and Duqutte et al. (1991)
replaced the slider mechanism with a rotating moment arm.
The equations of such a system are coupled with quadratic
nonlinearities. Numerical and experimental investigations
show better results than those in Golnaraghi (1991) due to
a more dominant nonlinear effect resulting from the pro-
posed configuration. Pai et al. (2000) used higher-order
internal resonances (3:1, 4:1) to implement CCC. Salemi et
al. (1997) and Fariborzi et al. (1997) used linear coupling
for the plant and controller. The controller was embedded
inside a computer. The method was implemented with great
success on a cantilever beam for both free and force vibra-
tion. The experiments showed that vibration attenuation as
high as 98% was easily attainable in a relatively short time.
Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008
© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
10. Robust Control
Robust control concentrates on the tradeoffs between per-
formance and stability in the presence of uncertainty in the
system model as well as the exogenous inputs to which it is
subjected. The uncertainty has its source in the modeling
errors, which can be divided into four types: parameter
errors, errors in model order, neglected disturbances, and
neglected nonlinearities. Thus, the H¥ controller is devel-
oped to address uncertainty by systematic means. A simple
explanation of H¥ theory proceeds as follows. A general
block diagram of a control system is shown in Figure 7,
where P(s) is the system on which the active control is
implemented and G(s) is the generalized plant which con-
sists of: the plant; actuators that generate inputs to the
plant; sensors measuring certain signals; analog-to-digital
and digital-to-analog converters; frequency shaped costs;
external noise sources or disturbances; and performance
weightings used in the design and synthesis of the con-
troller.
There are two vector inputs to and two vector outputs
from the generalized plant. The exogenous input vector
signal d(s) contains all disturbances, sensor noise, and com-
mand reference signals. The error output vector signal z(s)
contains all of the signals in the cost functional that are
important in the design of the control system for perfor-
mance. The vector signal y(s) contains all of the measured
output signals from the plant P(s) required for implementa-
tion of the control system. The vector signal u(s) contains
all of the control inputs to the plant P(s) required to imple-
ment the controller. Furthermore, a frequency shaped filter
W(s), coupled to selected outputs and inputs of the plant
model, is also included in the model. The filter can be used
to weight the outputs and the inputs to the control system.
The outputs of this frequency shaped filter define the error
outputs used to evaluate the system performance and gen-
erate the cost that will be used in the design process. The
transfer function matrix of the generalized plant can be par-
titioned according to the input–output variables:
 Alkhatib and Golnaraghi / ACTIVE STRUCTURAL VIBRATION CONTROL: A REVIEW
Figure 7. Block diagram for robust control.
éG zd G zu ù
G =ê . (27)
ëG yd G yu ú
û only in controlling the steady-state vibrational response
The measured output vector y(s) is coupled to the control
input u(s) through some form of static or dynamic compen-
sator K(s). In general, the cost function is constructed from
some norm associated with the exogenous input d(s) and
the error signal output z(s). The transfer function matrix
from d(s) to z(s), H zd, is written as
H zd = G zd +G zu K(I - G yu K) -1 G yd .
This transfer function matrix contains measures of nominal
performance and stability robustness. The objective of H¥
control is to design an admissible controller u(s) = K(s)y(s)
such that the infinity norm of the transfer function matrix
H zd, i.e. H zd ¥ , is minimum. The H¥ norm can be viewed
as a measure of the worst-case response of the system over
an entire class of model error or input disturbance.
H¥ theory has been applied to the active vibration con-
trol of a number of civil and mechanical engineering struc-
tures (Seto, 1998). Application of the H¥ controller in the
vibration control of flexible structures can be found in
Bayard and Chiang (1998), Zhang et al. (2001), Li and
Yam (2001a), Moreira et al. (2001), Chang et al. (2002),
Seto and Kar (2000), Wang et al. (2001) and Kar et al.
(2000). Rijanto et al. (2000) utilized an H¥ controller to
design an active mass damper (AMD) for the attenuation of
floor vibrations.
Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008
© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
375
11. Optimal Control
The ultimate aim of feedback vibration control is to reduce
the motion of the mechanical system to the greatest possi-
ble extent. One of the commonly used methods of modern
control theory is called optimal control. The method calcu-
lates the feedback gain by the minimization of a cost func-
tion or performance index, which is proportional to the
required measure of the system’s response. In other words,
the control u(t) is designed to minimize a cost function or
performance index, denoted by J = J( z , t, u(t)), given the
initial conditions z(t 0 ) and z&(t 0 ) subject to the constraint
that
z& = Az + Bu
y = Cz (29)
is satisfied with appropriate initial conditions. One such
cost function appropriate to a vibration control is
t
1 f
J = ò ( z T Qz + u T Ru) dt (30)
2 t0
where Q and R are positive definite symmetric weighting
matrices. The scalar quantity z T Qz is quadratically depend-
ent on the outputs of the system under control, and u T Ru is
quadratically dependent on the effort being expanded by
the control system. Q and R should be selected such that a
small reduction in the output is not obtained at the expense
of physically unreasonable input levels. If we are interested
and controlling the structure over a long time interval, the
solution to the optimal problem becomes
u(t) =-Gx(t) (31)
where G is a constant gain matrix defined as
G = R -1 B T S (32)
(28) and S is the solution to the Riccati equation
Q - SBR -1 B T S + A T S + SA = 0 (33)
which is a nonlinear algebraic equation in the constant
matrix S.
There is a large amount of research in the area of optimal
control and the implementation of the method is addressed
in Ram and Inman (1999), Zhang (2002), Friswell and Inman
(1999), Hori and Seto (2000), Seto et al. (1999, 2001),
Hwang et al. (1997), Ren et al. (1997) and Cai and Huang
(2002).
12. State Observers (Estimators)
It is not always possible to determine the entire state
variables from operating directly on the output since there
are too many degrees of freedom in a real structural system
and only very limited measurements can be collected from
the response of the structure. The state vector z(t) can be
estimated independently of the control problem, and the
resulting estimate z$(t) can be used in its place in equation
376 The Shock and Vibration Digest / September 2003
Figure 8. Closed-loop control of intelligent structures.
to arrive at an optimal solution. The state observer or esti-
mator is a method developed to estimate the state variables
from a partial number of observations or measurements.
The dynamic equation of the state observer shown in Fig-
ure 5 is given by
z&$ = Az$ + Bu + L( y - y$)
y$ = Cz$.
The state observer has the same characteristics as the struc-
ture and is fed the same signals as the mechanical system
being controlled, u(t). The output y$(t) is constantly com-
pared with the output of the mechanical system y(t), and
their difference is introduced, through a gain matrix L, into
the observer. The designer should choose the observer gain
matrix L such that the estimated state z$(t) converges to the
true state z(t). If the state estimate error is defined as
e (t) = z (t) - z$(t)
then the governing equation of e(t) is
e&(t) = ( A - LC ) e (t).
If L is chosen such that all the eigenvalues ( A - LC ) of
have negative real parts, then the solution for equation that
governs the state estimate error will converge to zero as
time tends to infinity. Thus, the estimated state converges
to the true state as can be seen from equation . The reader
can refer to Roh and Park (1999), Loh and Lin (1997) and
Yoshimura et al. (1993) for examples on the use of the
state estimator in active vibration control.
13. Intelligent Structure and Controller
Recent progress in the areas of intelligent control, artifi-
cial intelligence, machine learning, microprocessor tech-
nology, and actuator/sensor technology has provided a strong
groundwork for the development and realization of intel-
ligent structures. An intelligent structure would have the
capability to:
Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008
© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
• recognize the present dynamic state of its own structure
and monitor/deduce critical structural properties and
evaluate the functional performance of the structure;
• identify functional descriptions of external and internal
disturbances;
• detect changes in structural properties and changes in
external and internal disturbances;
• predict/infer possible future changes in structural prop-
(34) erties and changes in external and internal disturbances;
• make intelligent decisions regarding compensations for
disturbances and adequately generate actuation forces;
• identify potential future failures and identify the cause
of current failures and failed components;
• remedy failures and/or inform the appropriate opera-
tors;
• learn from past performance to improve future actions.
The above capabilities require the structure to be embed-
ded with a closed-loop controller equipped with intelligent
(35) sensors, intelligent actuators, and an intelligent control algo-
rithm. Figure 8 depicts the closed-loop control of intelligent
structures. Two main methodologies related to intelligent
(36) control have been developed: (1) artificial neural networks
and (2) fuzzy logic. Artificial neural networks were devel-
oped as a methodology for emulating the biology of the
human brain, resulting in systems that learn by experience.
Fuzzy logic was developed as a means of processing impre-
cise and vague linguistic information. The application of
fuzzy logic and neural network to control is found in
Norgaard et al. (2000) and Passino and Yurkovich (1998).
There is an ever growing literature related to the applica-
tion of intelligent control to active vibration control (Cohen et
al., 2002; Visioli, 2001; Faravelli and Rossi, 2002; Forrai et
al., 2000; Schurter and Roschke, 2001; Jha and Rower, 2002;
Kwak and Sciulli, 1996; Casciati et al., 1999; de Abreu and
Ribeiro, 2002; Li and Yam, 2001b; Battaini et al., 1998; Rao
and Prahlad, 1997; Yoshimura et al., 1999; Aldawod et al.,
2001; Teng et al., 2000; Gwo-Shiang, 1996; Valoor et al.,
2001; Ahn et al., 1996; Zhang et al., 2002; Al-Nassar et al.,
2000; Bani-Hani et al., 1999a, 1999b; Ma and Sinha, 1996;
Cheong and Cho, 1997; Kidner and Brennan, 2001).
 Alkhatib and Golnaraghi / ACTIVE STRUCTURAL VIBRATION CONTROL: A REVIEW
Figure 9. A model reference adaptive control system.
14. Adaptive Control
Figure 10. Active system (single degree of freedom).
Adaptive control is frequently used to control systems whose
parameters are unknown, uncertain, or slowly varying. An
adaptive controller maintains desired performance under
changing conditions by incorporating a mechanism for adjust-
ing its parameters. The design of an adaptive controller
involves several steps:
• selection of a controller structure with adjustable pa-
rameters;
• selection of an adaptation law for adjusting those pa-
rameters;
• selection of a performance index;
• real-time evaluation of the performance with respect to
some desired behavior;
• real-time plant identification and model updating;
• real-time adjustment of the controller parameters to
bring the performance closer to the desired behavior.
An adaptive controller essentially consists of a real-time
system identification technique integrated with a control
algorithm. The idea is to update the parameters of the plant
model based on input and output data, use the updated
model to compute a new set of controller parameters, and
then compute the next control output. The adaptive control
techniques are in general viewed as either direct or indirect
methods. In the direct method, the controller parameters
are adjusted directly based on the error between the mea-
sured and desired outputs. The identification scheme is for-
mulated such that the identified parameters are the control
parameters without the need for explicit system identifica-
tion or model updating. Since the control parameters are
directly solved for, this decreases the computation burden.
In the indirect method, the computations are divided into
two separate and consecutive phases. In each time step, the
parameters of the plant model are first estimated in real
time. Next, controller parameters are modified based on the
most recent updated plant parameters. In this way, the con-
trol system can track changes in the plant and disturbance
characteristics every time step. Parameters estimation can
be understood simply as the process of finding a set of
parameters that fits the input–output data from the plant.
For linear plants, many techniques are available for esti-
mating the unknown parameters. The most popular tech-
nique is the least-squares method and its extensions.
Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008
© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
377
Generally, a model reference adaptive control system
can be schematically represented by Figure 9. The output
yp of the plant to be controlled is required to follow the out-
put ym of the reference model. The parameters of the con-
troller are modified through an adaptation mechanism is
such a way as to reduce the error between yp and ym. Many
authors have investigated the implementation of adaptive
control theory to vibration attenuation (Elliott and Billet,
1993; Vipperman et al., 1993; Sun et al., 1998; Tokhi and
Hossain, 1996; Shaw, 1993; Pan and Hansen, 1998;
Sommerfeldt and Nashif, 1994; Pan and Ming, 1998; Kim
and Park, 1998; Elliott, 1998).
15. Active Control Effects on the System
In this section we show how active control influences
the response of the system. Basically, an active control sys-
tem uses an actuating device to provide a means of shaping
or controlling the response of the system. Consider the lin-
ear mechanical system with a single degree of freedom
shown in Figure 10. The control force is applied by an
actuator with negligible dynamics. With reference to Fig-
ure 10, the equation of motion of the system is
mx&&+ cx& + kx = f a + f (37)
where m, c, and k are the mass, damping coefficient, and
stiffness of the system, respectively, fa is the force applied
by the actuator, and f is the excitation force.
The steady-state response of the system without the con-
trol action is found from
X ( s) 1
= 2 . (38)
F( s) ms + cs + k
The system response is established from the mass, damping
coefficient, and stiffness of the system. The feedback
control produced by the active controller can be envi-
sioned as a modification of the structural characteristics.
For example, the controller force can be expressed as
f a =-g a x&&- g v x& - g d x, and the equation of motion of the
controlled system becomes
(m + g a )&&x +( c+ g v ) x& +( k + g d ) x = f . (39)
378 The Shock and Vibration Digest / September 2003
Depending on the type of signal utilized in the feedback
loop, the active control adds/subtracts mass, damping, and
stiffness to/from the mechanical system. The transfer func-
tion of the system with feedback becomes
X ( s) 1
= (40)
F( s) (m + g a ) s 2 +( c+ g v ) s +( k + g d )
or
X ( s) 1 becoming unstable if equations are not satisfied. The influ-
= 2
F( s) ms $ + k$
$ + cs
where m$ = m + g a is the modified mass, c$ = c+ g v is the
modified damping, and k$ = k + g d is the modified stiffness.
By means of the Routh–Hurwitz criterion, the stability of
the closed-loop system is guaranteed provided m, $ c$, and k$
are all positive. In theory, three gains in the feedback con-
troller can independently modify the mass, damping, and
stiffness of the system and set them anywhere in the range
of zero to something arbitrarily large. When dealing with
actual implementation, a number of physical constraints
restrict the range over which properties of the mechanical
system can be modified by feedback control.
16. Time Delay
One of the limits to the performance and application of
active vibration control is the (unavoidable) time delay in
controllers and actuators. Time delay exists in active con-
trol systems inevitably. The entire control process involves
measuring vibration data, conditioning and filtering the data,
computing the control forces, transmitting data and signals
to actuators, and applying control forces to the structure.
This process results in a time delay when applying the
required control forces to the structure. Time delay intro-
duces phase shift, which deteriorates the controller perfor-
mance or even causes instability in the system if the actua-
tors add energy to the structures when it is not required.
For a fixed time delay τ in the control force fa(t), the trans-
fer function Fa(s) for the feedback part, equation (40), is
modified as
Fa ( s) = e-τs [-g a s 2 - g v s - g d ]X ( s).
If this delay is small, its frequency can be expressed as
e- τs @ 1- τs.
The closed-loop transfer function with delay is
X( s )
= electromagnetic gain of the PMA coil, and G2 (N/V) is the
F( s )
1
= .
[( τga )s 3 + ( m+ ga - τgv )s 2 + ( c + gv - τgd )s + ( k + gd )]
Using the Hurwitz stability criterion (Kuo and Golnaraghi,
2003), the parameters of the system must satisfy the fol-
lowing relations
( τg a ) > 0
Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008
© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
[(m + g a - τg v )( c+ g v - τg d ) - ( τg a )( k + g d )] > 0 (45)
or
( τg a ) > 0
( g v g d ) τ 2 +(mc+mg v + g a c+ g a g v )
(46)
> (mg d + 2g a g d + g v c+ g v2 + kg a ) τ.
With time delay present, there is a possibility of the system
(41)
ence of time delays on the efficiency of the active system is
investigated in Andrade et al. (1995), Ali et al. (1998) and
Hu and Wang (1998). A state-of-the-art review on the effects
of time delay on the performance and stability of active
control systems is presented in Agrawal and Yang (1997)
and some methods to deal with the problem are offered in
Cai and Huang (2002), Chu et al. (2002) and Agrawal and
Yang (2000).
17. Actuator–Structure Interaction
The behavior of any active structure is the result of the
integration of the behavior of the structural subsystem with
that of the controller, the actuators, and the sensors; the
only reasonable approach to design this is to design the
system as a whole. To add a controller to an already exist-
ing structure or one designed without taking into account
the presence of the former may lead to performances far
from the expected (Dyke et al., 1995). Structural dynamics,
control engineering, transducer design, and electronics
must merge from the beginning with the interdisciplinary
approach often referred to as mechatronics. In this section,
we discuss the effects of including the actuator dynamics
on the overall system response. The proof-mass actuator
(PMA) is an example of devices which are employed in the
active vibration control of a flexible structure (Gawronski
and Lim, 1993b; Garcia et al., 1995). The PMA consists of
a reaction mass ma that is attached to a current-carrying
coil moving in a magnetic field created by a permanent
magnet. The moving part is attached to the permanent mag-
net by a spring ka and damper ca (Figure 11). The transfer
(42) function relating the force output of the PMA to the volt-
age input is (Preumont, 2002)
Fa ( s) s2
= G1 G2 2 (47)
(43) V ( s) s + 2ξ a ωa s + ωa2
G1 G2 k a G G ca
where ω2a = , 2ξ a ωa = 1 2 , G1 (N/A) is the
ma ma
(44) power amplifier gain.
If we disregard the actuator dynamics and assume an
ideal force generator (Figure 12), the equation of motion of
the system is
mx&&+ cx& + kx = f a . (48)
On the other hand, including the actuator dynamics (Fig-
ure 13) modifies the equation of motion to
 Alkhatib and Golnaraghi / ACTIVE STRUCTURAL VIBRATION CONTROL: A REVIEW 379

Figure 13. Structures with actuator dynamics.

and Inman (1990) and Inman (1990) have investigated the

interaction between a PMA and a structure. The actuator is
assumed to have second-order dynamics. Actuator satura-
tion and its effects on performance and stability have been
Figure 11. Sketch of a proof-mass. examined in Agrawal et al. (1997), Nguyen and Jabbari
(2000) and Lindner et al. (1994a, 1994b).

18. Optimal Placement of Actuators

In active vibration control, actuator placement on the struc-

ture is a very significant issue. It has a direct effect on the
control efficiency and cost. For example, there can be many
modes within the frequency band of interest, and this could
require many actuators. Also, large flexible structures (Park
et al., 2002; Jha et al., 2002; Sun et al., 1994) require many
actuators for active vibration control. In these cases, the
problem of optimizing the locations of the actuators becomes
extremely significant to maximize the system controllabil-
ity and therefore increase its efficiency. In such cases, the
problem of actuator placement becomes more significant
than the control law itself. An arbitrary choice of actuator
Figure 12. Structure with ideal force generator. location can seriously degrade the system performance. It
can also limit the actuator’s range of practical application.
If the actuator is placed at the wrong location, the system
ém +m p 0 ùì x&& ü éc+ c a -c a ùì x& ü will require a greater control force. In this case, the system
ê 0 í ý+ê úí ý
ë ma ú
ûîx&& a þ ë -c a c a ûîx& a þ is said to have a low degree of controllability. To see this
(49) more clearly, consider the case in which the actuator is
ék + k a -k a ùì x ü ì f a ü placed on a nodal line for a certain mode of interest. In this
+ê í ý= í ý.
ë -k a ka úûîx a þ î- f a þ case, suppressing this mode becomes impossible, and the
system is said to be uncontrollable. A vast amount of
If we consider velocity feedback research has been devoted to the optimization of actua-
tors/sensors numbers and locations (Arbel, 1981; Kang et
f a =-g v x& (50) al., 1998; Yam and Li, 2001; Shih et al., 1998; Abdullah,
1998; De Fonseca et al., 1999; Yam and Yan, 2002; Gao et
the feedback modifies the equations of motion and yields al., 2000; Gawronski, 1997a; Zimmerman, 1993; Hanagan
et al., 2000; Kulasekere et al., 2000). Hanagan et al. (2000)
ém +m p 0 ùì x&& ü éc+ c a + g v -c a ùì x& ü developed a method to optimize the locations of actua-
ê 0 í ý+ê úí ý
ë ma ú
ûîx&& a þ ë -c a + g v c a ûîx& a þ tors/sensors for active vibration control of floors.
(51)
ék + k a -k a ùì x ü ì0ü
+ê í ý= í ý . 19. Example of the Implementation of
ë -k a ka úûîx a þ î0þ a Decentralized Control Strategy

Such a system has the potential to become unstable since Some researchers (Serrand and Elliott, 2000; Elliott et
the damping matrix is no longer symmetric. Zimmerman al., 2001) have investigated the application of decentralized

Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008

© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
380 The Shock and Vibration Digest / September 2003
Figure 14. Two-degrees-of-freedom active system.
velocity feedback control to vibration isolation of a system,
as depicted in Figure 14. Each actuator is operated inde-
pendently by simply feeding back a signal proportional to
the corresponding equipment vibration at the same loca-
tion. Using direct velocity feedback, the amplitude of the
heave and pitching modes are reduced up to 40 and 26 dB,
respectively. The stability of the controller is found to be
robust to changes in the dynamics of the base structure.
The method is extended and applied to a four-channel actua-
tor with decentralized controller (Kim et al., 2001). Experi-
mental implementation shows up to 14 dB reduction in the
kinetic energy of the equipment. However, instability is
encountered for high gain values. A method based on H¥
robust control theory has been studied in Gallet and
Bellizzi (1997) as a means to achieve decoupled local con-
trol in the same system. The design of the controller associ-
ated with each mount took into account the effects of the
other controllers acting as model uncertainties. Comparison
between the performance of coupled and decoupled con-
trollers are made and found to be similar.
20. Conclusion
The overview presented in this paper may be of interest to
researchers and engineers engaged in the study and devel-
opment of the rapidly evolving field of active vibration
control. A synopsis of the skeleton of an active vibration
control system is introduced.
We have briefly explained concepts such as structure
modeling, model reduction, feedback control, feedforward
control, controllability and observability, spillover, eigen-
structure assignment (pole placement), coordinate coupling
control, robust control, optimal control, state observers
(estimators), intelligent structure and controller, adaptive
control, active control effects on the system, time delay,
Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008
© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
actuator–structure interaction, and optimal placement of
actuators. The literature is rich with theoretical studies and
practical application of active control to the attenuation of
mechanical and structural vibration. In this paper we have
made reference to some papers relevant to each subject.
References
Abdullah, M. M., 1998, “Optimal Location and Gains of Feedback Con-
trollers as Discrete Locations,” AIAA Journal, Vol. 36, No. 11,
2109–2116.
Agrawal, A. K., and Yang, J. N., 1997, “Effect of Fixed Time Delay on
Stability and Performance of Actively Controlled Civil Engi-
neering Structures,” Journal of Earthquake Engineering and Struc-
tural Dynamics, Vol. 26, 1169–1185.
Agrawal, A. K., and Yang, J. N., 2000, “Compensation of Time-Delay
for Control of Civil Engineering Structures,” Journal of Earth-
quake Engineering and Structural Dynamics, Vol. 29, 37–62.
Agrawal, A. K., Yang, J. N., Schmitendorf, W. E., and Jabbari, F., 1997,
“Stability of Actively Controlled Structure with Actuator Satura-
tion,” Journal of Structural Engineering, Vol. 123, No.4, 505–512.
Ahn, K. G., Pahk, H. J., Jung, M. Y., and Cho, D. W., 1996, “A Hybrid-
Type Active Vibration Isolation System Using Neural Networks,”
Journal of Sound and Vibration, Vol. 192, No. 4, 793–805.
Aldawod, M., Samali, B., Naghdy, F., and Kwok, K. C. S., 2001, “Active
Control of Wind Response of Tall Building Using a Fuzzy Control-
ler,” Engineering Structures, Vol. 23, No. 11, 1512–1522.
Ali, M. S., Hou, Z. K., and Noori, M. N., 1998, “Stability and Perfor-
mance of Feedback Control Systems with Time Delays,” Com-
puters and Structures, Vol. 66, No. 2/3, 241–248.
Al-Nassar, Y. N., Siddiqui, M., and Al-Garni, A. Z., 2000, “Artificial
Neural Networks in Vibration Control of Rotor-Bearing Systems,”
Simulation Practice and Theory, Vol. 7, No. 8, 729–740.
Andrade, R. A., Lopez-Almansa, F., and Rodellar, J., 1995, “Influence of
Time Delays on the Efficiency of Active Mass Dampers,” Smart
Materials and Structures, Vol. 4, A1–A8.
Antsaklis, P. J., and Michel, A. N., 1997, Linear Systems, McGraw-Hill,
New York.
Arabyan, A., Chemishkian, S., and Meroyan E., 1999, “Limits of Vibra-
tion Suppression in Flexible Structures”, Journal of Dynamics and
Control, Vol. 9, 223–246.
 Alkhatib and Golnaraghi / ACTIVE STRUCTURAL VIBRATION CONTROL: A REVIEW
Arbel, A., 1981, “Controllability Measures and Actuator Placement in
Oscillatory Systems,” International Journal of Control, Vol. 33,
565–574.
Ashour, O. N., and Nayfeh, A. H., 2002, “Adaptive Control of Flexible
Structures Using a Nonlinear Vibration Absorber,” Nonlinear
Dynamics, Vol. 28, No. 3, 309–322.
Balas, M. J., 1978, “Feedback Control of Flexible Systems,” IEEE Trans-
actions on Automatic Control, Vol. 23, No. 4, 673–679.
Bani-Hani, K., Ghaboussi, J., and Schneider, S. P., 1999a, “Experimental
Study of Identification and Control of Structures Using Neural Net-
work. Part 1: Identification,” Earthquake Engineering and Struc-
tural Dynamics, Vol. 28, No. 9, 995–1018.
Bani-Hani, K., Ghaboussi, J., and Schneider, S. P., 1999b, “Experimental
Study of Identification and Control of Structures Using Neural Net-
work. Part 2: Control,” Earthquake Engineering and Structural
Dynamics, Vol. 28, No. 9, 1019–1039.
Battaini, M., Casciati, F., and Faravelli, L., 1998, “Fuzzy Control of
Structural Vibration: An Active Mass System Driven by a Fuzzy
Controller,” Earthquake Engineering and Structural Dynamics,
Vol. 27, No. 11, 1267–1276.
Bayard, D. S., and Chiang, R. Y., 1998, “Identification, Uncertainty
Characterization and Robust Control Synthesis Applied to Large
Flexible Structures Control,” International Journal of Robust and
Nonlinear Control, Vol. 8, 97–112.
Bouhaddi, N., and Fillod, R., 1996, “Model Reduction by a Simplified
Variant of Dynamic Condensation,” Journal of Sound and Vibra-
tion, Vol. 191, No. 2, 233–250.
Cai, G., and Huang, J., 2002, “Optimal Control Method with Time Delay
in Control,” Journal of Sound and Vibration, Vol. 251, No. 3,
383–394.
Casciati, F., Faravelli, L., and Torelli, G., 1999, “A Fuzzy Chip Control-
ler for Nonlinear Vibrations,” Nonlinear Dynamics, Vol. 20, No. 1,
85–98.
Chang, W., Gopinathan, S. V., Varadan, V. V., and Varadan, V. K.,
2002, “Design of Robust Vibration Controller for a Smart Panel
Using Finite Element Model,” Journal of Vibration and Acoustics,
Vol. 124, 265–276.
Chen, C.-T., 1984, Linear System Theory and Design, Saunders HBJ.
Chen, Y., 1999, “Finite Element Analysis for Walking Vibration Prob-
lems for Composite Precast Building Floors Using ADINA:
Modeling, Simulation, and Comparison,” Journal of Computer and
Structure, Vol. 72, 109–126.
Chen, Y. D., Chen, S. H., and Liu, Z. S., 2001, “Quantitative Measures of
Modal Controllability and Observability in Vibration Control of
Defective and Near-Defective Systems,” Journal of Sound and
Vibration, Vol. 248, No. 3, 413–426.
Cheong, M.-S., and Cho, D.-W., 1997, “A New Vibration Isolation Of A
Hydraulic System Using A Heterosynaptic Neural Network,”
Mechanical Systems and Signal Processing, Vol. 11, No. 6, 887–901.
Chu, E. K., 2002, “Pole Assignment for Second-Order System,” Journal
of Mechanical Systems and Signal Processing, Vol. 16, No. 1, 39–59.
Chu, S. Y., Soong, T. T., Lin, C. C., and Chen, Y. Z., 2002, “Time-Delay
Effect and Compensation on Direct Output Feedback Controlled
Mass Damper Systems,” Journal of Earthquake Engineering and
Structural Dynamics, Vol. 31, 121–137.
Clark, R. L., Saunders, W. R., and Gibbs, G. P., 1998, Adaptive Struc-
tures: Dynamics and Control, Wiley, New York.
Cohen, K., Weller, T., and Ben-Asher, J. Z., 2002, “Active Control of
Flexible Structures Using a Fuzzy Logic Algorithm,” Smart Mate-
rials and Structures, Vol. 11, 541–552.
de Abreu, G. L., and Ribeiro, J. F., 2002, “A Self-Organizing Fuzzy
Logic Controller for the Active Control of Flexible Structures
using Piezoelectric Actuators,” Applied Soft Computing Journal,
Vol. 1, No. 4, 271–283.
De Fonseca, P., Sas, P., and Van Brussel, H., 1999, “Comparative Study
of Methods for Optimising Sensor and Actuator Location in Active
Control Applications,” Journal of Sound and Vibration, Vol. 221,
No. 4, 651–679.
Duquette, A. P., Tuer, K. L., and Golnaraghi, M. F., 1991, “Vibration
Control of a Flexible Beam Using a Rotational Internal Resonance
Controller, Part II: Experiment,” Journal of Sound and Vibration,
Vol. 167, No. 1, 63–75.
Dyka, C. T., Ingel, R. P., and Flippen, L.D., 1996, “A New Approach to
Dynamic Condensation for FEM,” Computers and Structures, Vol.
61, No. 4, 763–773.
Dyke, J. S., Spencer Jr., B. F., Quast, P., and Sain, M. K., 1995, “Role of
Control-Structure Interaction in Protective System Design,” Jour-
nal of Engineering Mechanics ASCE, Vol. 121, No. 2, 322–338.
Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008
© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
381
Elliott, S. J., 1998, “Filtered Reference and Filtered Error LMS Algo-
rithm for Adaptive Feedforward Control,” Mechanical Systems and
Signals Processing, Vol. 12, No. 6, 769–781.
Elliott, S. J., and Billet, L., 1993, “Adaptive Control of Flexural Waves
Propagating in a Beam,” Journal of Sound and Vibration, Vol. 163,
No. 2, 295–310.
Elliott, S.J., Serrand, M., and Gardonio, P., 2001, “Feedback Stability
Limits for Active Isolation Systems with Reactive and Inertial Actu-
ators,” Journal of Vibration and Acoustics, Vol. 123, 250–261.
Faravelli, L., and Rossi, R., 2002 “Adaptive Fuzzy Control: Theory Ver-
sus Implementation,” Journal of Structural Control, Vol. 9, 59–73.
Fariborzi, F., Golnaraghi, M.F., and Heppler, G.R, 1997, “Experimental
Control of Free and Forced Structural Vibration using A Linear
Coupling Strategy,” Smart Materials and Structures, Vol. 6, 540–548.
Forrai, A., Hashimoto, S., Funato, H., and Kamiyama, K., 2000, “Fuzzy
Logic Based Vibration Suppression Control of Flexible Structures,”
in AMC 2000, Nagoya, 378–383.
Friswell, M. I., and Inman, D. J., 1999, “The Relationship Between Posi-
tive Position Feedback and Output Feedback Controllers,” Smart
Materials and Structures, Vol. 8, 285–291.
Friswell, M. I., Garvey, S. D., and Penny, J. E. T., 1995, “Model Reduc-
tion Using Dynamic and Iterated IRS Techniques,” Journal of
Sound and Vibration, Vol. 186, No. 2, 311–323.
Friswell, M. I., Garvey, S. D., and Penny, J. E. T., 1996, “The Applica-
tion of the IRS and Balanced Realization Method to Obtain
Reduced Models of Structures with Local Non-Linearities,” Jour-
nal of Sound and Vibration, Vol. 196, No. 4, 453–468.
Friswell, M. I., Garvey, S. D., and Penny, J. E. T., 1998, “The Conver-
gence of the Iterated IRS Method,” Journal of Sound and Vibra-
tion, Vol. 211, No. 1, 123–132.
Fuller, C. R., Elliott, S. J., and Nelson, P. A., 1997, Active Control of
Vibration. Academic Press, New York.
Gallet, A., and Bellizzi, S., 1997, “Decoupling Actuators with H Robust
Control for Actively Isolating Vibrating Machines,” Journal of
Mechanical Systems and Signal Processing, Vol. 11, No. 6,
869–886.
Gao, F., Shen, Y., and Li, L., 2000, “The Optimal Design of Piezoelectric
Actuators for Plate Vibroacoustic Control Using Genetic Algorithms
with Immune Diversity,” Smart Materials and Structures, Vol. 9,
485–491.
Garcia, E., Webb, C. S., and Duke, M. J., 1995, “Passive and Active Con-
trol of Complex Flexible Structure Using Reaction Mass Actua-
tors,” Journal of Vibration and Acoustics, Vol. 117, 116–122.
Gawronski, W., 1997a, “Actuator and Sensor Placement for Structural
Testing and Control,” Journal of Sound and Vibration, Vol. 208,
No. 1, 101–109.
Gawronski, W., 1997b, “Almost-Balanced Structure Dynamics,” Journal
of Sound and Vibration, Vol. 202, No. 5, 669–687.
Gawronski, W., and Lim, K. B., 1993a, “Controllability and Observability
of Flexible Structure with Proof-Mass Actuator,” Journal of Guid-
ance, Control, and Dynamics, Vol. 17, No. 5, 899–902.
Gawronski, W., and Lim, K. B., 1993b, “Controllability and Observability
of Flexible Structure with Proof-Mass Actuators,” Journal of Guid-
ance, Control, and Dynamics, Vol. 16, No. 5, 899–902.
Golnaraghi, M. F., 1991, “Regulation of Flexible Structures via Non-linear
Coupling,” Journal of Dynamics and Control, Vol. 1, 405–428.
Gorain, G. C., and Bose, S. K., 2002, “Exact Controllability and Bound-
ary Stabilization of Flexural Vibration of an Internally Damped
Flexible Space Structure,” Applied Mathematics and Computation,
Vol. 126, 341–360.
Guyan, R. J., 1965, “Reduction of Stiffness and Mass Matrices,” AIAA
Journal, Vol. 3, No. 2, 380.
Gwo-Shiang, L., 1996, “System Identification and Control of Smart Struc-
tures Using Neural Networks,” Acta Astronautica, Vol. 38, No.
4–8, 269–276.
Hanagan, L. M., and Murray, T. M., 1997, “Active Control Approach for
Reducing Floor Vibrations,” Journal of Structural Engineering,
ASCE, Vol. 123, No. 11, 1497–1505.
Hanagan, L.M., Kulasekere, E.C., Walgama, K.S., and Premaratne, K.,
2000, “Optimal Placement of Actuators and Sensors for Floor Vibra-
tion Control,” Journal of Structural Engineering, ASCE, Vol. 126,
No. 12, 1380–1387.
Harris, C. M., 1997, Shock and Vibration Handbook, 4th Ed., McGraw-Hill,
New York.
Hori, N., and Seto, K., 2000, “Vibration Control of Flexible Space Struc-
ture Based on Reduced Order Modeling Method and Filtered LQ
Control Theory,” JSME International Journal – Series C, Vol. 43,
No. 3, 697–718.
382 The Shock and Vibration Digest / September 2003
Hu, H. Y., and Wang, Z. H., 1998, “Stability Analysis of Damped SDOF
Systems with Two Time Delays in the State Feedback,” Journal of
Sound and Vibration, Vol. 214, No. 2, 213–225.
Hughes, P. C., and Skelton, R. E., 1980, “Controllability and Observability
of Linear Matrix-Second-Order Systems,” Journal of Applied
Mechanics, Vol. 47, 415–420.
Hwang, J. K., Choi, C., Song, C. K., and Lee, J. M., 1997, “Robust LQG
Control of an All-Clamped Thin Plate with Piezoelectric Actua-
tors/Sensors,” IEEE/ASME Transactions on Mechatronics, Vol. 2,
No. 3, 205–212.
Inman, D. J., 1989, Vibration with Control, Measurement, and Stability,
Prentice Hall, Englewood Cliffs, NJ.
Inman, D. J., 1990, “Control/Structure Interaction: Effects of Actuators
Dynamics,” in Mechanics and Control of Large Flexible Structure,
Progress in Astronautics and Aeronautics, Vol. 129, Chap. 20,
507–533.
Inman, J. D., 2001, “Active Modal Control for Smart Structure,” Philo-
sophical Transactions: Mathematical, Physical and Engineering
Sciences, Vol. 359, 205–219.
Jha, R., and Rower, J., 2002, “Experimental Investigation of Active Vibra-
tion Control Using Neural Networks and Piezoelectric Actuators,”
Smart Materials and Structures, Vol. 11, 115–121.
Jha, A., Park, G., and Inman, D., 2002, “Vibration Testing and Analysis
of Inflatable Structures,” in International Conference on Structural
Dynamics and Modeling, 3–5 June, Madeira Island, Portugal.
Juang, J., and Maghami, P. G., 1990, “Robust Eigensystem Assignment
for Second-Order Dynamic Systems,” in Mechanics and Control of
Large Flexible Structure, Progress in Astronautics and Aeronau-
tics, Vol. 129, Chap. 14, 373–387.
Junkins, J. L., and Kim, Y., 1990, “Minimum Sensitivity Design Method
for Output Feedback,” in Mechanics and Control of Large Flexible
Structure, Progress in Astronautics and Aeronautics, Vol. 129,
Chap. 15, 389–409.
Kang, Y. K., Park, H. C., and Arawal, B., 1998, “Optimization of
Piezoceramic Sensor/Actuator Placement for Vibration Control Of
Laminated Plates,” AIAA Journal, Vol. 36, No. 9, 1763–1765.
Kar, I. N., Seto, K., and Doi, F., 2000, “Multimode Vibration Control of
a Flexible Structure Using H-Based Robust Control,” IEEE/ASME
Transactions on Mechatronics, Vol. 5, No. 1, 23–31.
Karnopp, D. C., Margolis, D. L., Rosenberg, R. C., 2000, System Dynam-
ics: Modeling and Simulation of Mechatronics, 3rd Ed., Wiley,
New York.
Khajepour, A., and Golnaraghi, M. F., 1997, “Experimental Control of
Flexible Structure Using Nonlinear Modal Coupling: Forced and
Free Vibration,” Journal of Intelligent Material Systems and Struc-
tures, Vol. 8, 697–710.
Kidner, M. R. F., and Brennan, M. J., 2001, “Real-Time Control of Both
Stiffness and Damping in an Active Vibration Neutralizer,” Smart
Materials and Structures, Vol. 10, 758–769.
Kim, H.-S., and Park, Y., 1998, “Delayed-X LMS Algorithm: An Effi-
cient ANC Algorithm Utilizing Robustness Of Cancellation Path
Model,” Journal of Sound and Vibration, Vol. 212, No. 5, 875–887.
Kim, M.-H., and Inman, D. J., 2001, “Spillover Reduction in the Vibra-
tion Control of Flexible Structures Using Sliding Mode Observer,”
Journal of Vibration and Control, Vol. 7, 1087–1105.
Kim, S.-M., Elliott, S.J., and Bernnan, M. J., 2001, “Decentralized Con-
trol for Multichannel Active Vibration Isolation,” IEEE Transac-
tions on Control Systems Technology, Vol. 9, No. 1, 93–100.
Kulasekere, E.C., Hanagan, L.M., Walgama, K.S., and Premaratne, K.,
2000, “MIMO Floor Vibration Controller Design,” in Proceedings
of the American Control Conference (AACC), 508–512.
Kuo, B. C., and Golnaraghi, F., 2003, Automatic Control Systems, 8th
Ed., Wiley, New York.
Kwak, M. K., and Sciulli, D., 1996, “Fuzzy-Logic Based Vibration Sup-
pression Control Experiments on Active Structures,” Journal of
Sound and Vibration, Vol. 191, No. 1, 15–28.
Lee, A. Y., and Tsuha, W. S., 1994, “Model Reduction Methodology for
Articulated, Multiflexible Body Structures,” Journal of Guidance,
Control, and Dynamics, Vol. 17, No. 1, 69–75.
Li, Y. Y., and Yam, L. H., 2001a, “A Robust Design Method of Active
Controller for Uncertain Vibration Systems,” Journal of Vibration
and Control, Vol. 7, 453–466.
Li, Y.Y., and Yam, L.H., 2001b, “Robust Vibration Control of Uncertain
Systems Using Variable Parameter Feedback and Model-Based
Fuzzy Strategies,” Computers and Structures, Vol. 79, No. 11,
1109–1119.
Lindner, D. K., Zvonar, G. A., and Borojevic, D., 1994a, “Performance
and Control of Proof-Mass Actuators Accounting for Stroke Satu-
Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008
© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
ration,” Journal of Guidance, Control, and Dynamics, Vol. 17, No.
5, 1103–1108.
Lindner, D. K., Zvonar, G. A., and Borojevic, D., 1994b, “Nonlinear
Control of Proof-Mass Actuator,” Journal of Guidance, Control,
and Dynamics, Vol. 20, No. 3, 464–470.
Loh, C.-H., Lin, P.-Y., 1997, “Kalman Filter Approach For the Control of
Seismic-Induced Building Vibration Using Active Mass Damper
Systems,” The Structural Design of Tall Buildings, Vol. 6, No. 3,
209–224.
Ma, R. P., and Sinha, A., 1996, “A Neural Network Based Active Vibra-
tion Absorber With State Feedback Control,” Journal of Sound and
Vibration, Vol. 190, No. 1, 121–128.
Maia, S., He, L., Lin, S., and To, U., 1997, Theoretical and Experimental
Modal Analysis, Wiley, New York.
Mei, C., and Mace, B. R., 2002, “Reduction of Control Spillover in
Active Vibration Control of Distributed Structures Using Multi-
optimal Schemes,” Journal of Sound and Vibration, Vol. 251, No.
1, 184–192.
Meirovitch, L., 1987, “Some Problems Associated with the Control of
Distributed Structures,” Journal of Optimization Theory and Appli-
cations, Vol. 54, 1–20.
Meirovitch, L., 1990, Dynamics and Control of Structure, Wiley, New
York.
Meirovitch, L., 1997, Principle and Techniques of Vibration, Prentice
Hall, Englewood Cliffs, NJ.
Moore, B., 1981, “Principal Component Analysis in Linear Systems: Con-
trollability, Obvervability, and Model Reduction,” IEEE Transac-
tions on Automatic Control, Vol. 26, No. 1, 17–31.
Moreira, J. O. F., Arruda, J. R. D., and Inman, J. D., 2001, “Design of a
Reduced-Order H Controller for Smart Structure Satellite Applica-
tions,” Philosophical Transactions: Mathematical, Physical and
Engineering Sciences, Vol. 359, 2251–2269.
Myklebust, L. I., and Skallerud, B., 2002, “Model Reduction Methods for
Flexible Structures,” in Proceedings 15th Nordic Seminar on Com-
putational Mechanics, 18–19 October, Aalborg, Denmark, 171–175.
Nakra, B. C., 1998, “Vibration Control in Machines and Structures Using
Viscoelastic Damping,” Journal of Sound and Vibration, Vol. 211,
No. 3, 449–465.
Nguyen, T., and Jabbari, F., 2000, “Output Feedback Controllers for Dis-
turbance Attenuation with Actuator Amplitude and Rate Satura-
tion,” Automatica, Vol. 36, No. 9, 1339–1346.
Norgaard, M., Ravn, O., Poulsen, N. K., and Hansen, L. K., 2000, Neural
Networks for Modeling and Control of Dynamic Systems, Springer,
Berlin.
Ogata, K., 1995, Discrete-Time Control Systems, 2nd Ed., Prentice Hall,
Englewood Cliffs, NJ.
Oueini, S. S., Nayfeh, A. H., and Golnaraghi, M. F., 1997, “A Theoreti-
cal and Experimental Implementation of a Control Method Based
on Saturation,” Nonlinear Dynamics, Vol. 13, 189–202.
Oueini, S. S., Nayfeh, A. H., and Pratt, J. R., 1998, “A Nonlinear Vibra-
tion Absorber for Flexible Structures,” Nonlinear Dynamics, Vol.
15, No. 3, 259–282.
Oueini, S. S., Nayfeh, A. H., and Pratt, J. R., 1999a, “A Review of
Development and Implementation of an Active Nonlinear Vibra-
tion Absorber,” Archive of Applied Mechanics, Vol. 69, No. 8,
585–620.
Oueini, S. S., Chin, C.-M., and Nayfeh, A. H., 1999b, “Dynamics of a
Cubic Nonlinear Vibration Absorber,” Nonlinear Dynamics, Vol.
20, No. 3, 283–295.
Pai, P. F., and Schulz, M. J., 2000, “A Refined Nonlinear Vibration
Absorber,” International Journal of Mechanical Sciences, Vol. 42,
No. 3, 537–560.
Pai, P. F., Wen, B., Naser, A. S., and Schulz, M. J., 1998, “Structural
Vibration Control Using PZT Patches and Non-Linear Phenom-
ena,” Journal of Sound and Vibration, Vol. 215, No. 2, 273–296.
Pai, P. F., Rommel, B. A. S., and Schulz, M. J., 2000, “Non-Linear
Vibration Absorbers Using Higher Order Internal Resonances,”
Journal of Sound and Vibration, Vol. 234, No. 5, 799–817.
Pan, X., and Hansen, C. H., 1998, “Active Vibration Control of Waves in
Simple Structure with Multiple Error Sensors,” Journal of the
Acoustic Society of America, Vol. 103, No. 3, 1673–1676.
Pan, J., and Ming, R., 1998, “Experimental Determination of the Total
Vibratory Power Transmission in an Elastic Beam,” Journal of the
Acoustic Society of America, Vol. 104, No. 2, 898–906.
Park, G., Ruggiero, E., and Inman, D. J., 2002, “Dynamic Testing of an
Inflatable Structure using Smart Materials,” Smart Materials and
Structures, Vol. 11, No.1, 147–166.
 Alkhatib and Golnaraghi / ACTIVE STRUCTURAL VIBRATION CONTROL: A REVIEW
Passino, K. M., and Yurkovich, S., 1998, Fuzzy Control, Addison-Wesley,
Reading, MA.
Pernebo, L., and Silverman, L., 1982, “Model Reduction via Balanced
State-Space Representation,” IEEE Transactions on Automatic
Control, Vol. 27, No. 2, 382–387.
Preumont, A., 2002, Vibration Control of Active Structures: An Introduc-
tion, 2nd Ed., Kluwer Academic, Dordrecht.
Qu, Z. Q., 1998, “A Multi-Step Method for Matrices Condensation of
Finite Element Methods,” Journal of Sound and Vibration, Vol.
214, No. 5, 965–971.
Ram, Y. M., and Inman, D. J., 1999, “Optimal Control for Vibrating Sys-
tems,” Mechanical Systems and Signal Processing, Vol. 13, No. 6,
879–892.
Rao, M.V.C., and Prahlad, V., 1997, “A Tunable Fuzzy Logic Controller
for Vehicle-Active Suspension Systems,” Fuzzy Sets and Systems,
Vol. 85, No. 1, 11–21.
Ren, M., Seto, K., and Doi, F., 1997, “Feedback Structure-Borne Sound
Control of a Flexible Plate with an Electromagnetic Actuator: The
Phase Lag Problem,” Journal of Sound and Vibration, Vol. 205,
No. 1, 57–80.
Rijanto, E., Okamoto, M., Kuang, W., and Tagawa, Y., 2000, “A Floor
Vibration Attenuation Device Using Robust H Controller,” in
AMC 2000, 363–366.
Roh, H.-S., and Park, Y., 1999, “Stochastic Optimal Preview Control of
an Active Vehicle Suspension,” Journal of Sound and Vibration,
Vol. 220, No. 2, 313–330.
Salemi, P., Golnaraghi, M.F., and Heppler, G.R., 1997, “Active Control
of Forced and Unforced Structural Vibration,” Journal of Sound
and Vibration, Vol. 208, No. 1, 15–32.
Schurter, K. C., and Roschke, P. N., 2001, “Neuro-Fuzzy Control of
Structures using Acceleration Feedback,” Smart Materials and
Structures, Vol. 10, 770–779.
Serrand, M., and Elliott, S.J., 2000, “Multichannel Feedback Control for
the Isolation of Base-Excited Vibration,” Journal of Sound and
Vibration, Vol. 234, No. 4, 681–704.
Seto, K., 1998, “Trends on Active Vibration and Noise Control” Interna-
tional Journal of the Japanese Society of Precision Engineering,
Vol. 32, No. 3, 155–159.
Seto, K., Doi, F., and Ren, M., 1999, “Vibration Control of Bridge Towers
Using Lumped Modeling Approach,” Journal of Vibration and
Acoustics, Vol. 121, 95–100.
Seto, K., and Kar, I. N., 2000, “A Comparative Study on H-infinity
Based Vibration Controller of a Flexible Structure System,” in Pro-
ceedings of the American Control Conference, Chicago, Illinois,
513–518.
Seto, K., Ren, M., and Doi, F., 2001, “Modeling and Feedback Structural
Acoustics Control of a Flexible Plate,” Journal of Vibration and
Acoustics, Vol. 123, 18–23.
Shaw, J., 1993, “Adaptive Control for Sound and Vibration Attenuation:
A Comparative Study,” Journal of Sound and Vibration, Vol. 235,
No. 4, 671–684.
Shih, Y.-T., Lee, A. C., and Chin, J.-H., 1998, “Sensor and Actuator
Placement for Modal Identification,” Mechanical Systems and Sig-
nal Processing, Vol. 12, No. 5, 641–659.
Slater, G. L., and Zhang, Q., 1990, “Controller Design by Eigenspace
Assignment,” in Mechanics and Control of Large Flexible Struc-
ture, Progress in Astronautics and Aeronautics, Vol. 129, Chap.
17, 435–462.
Slater, J. C., and Inman, D. J., 1997, “On the Effect of Weak Non-Linearities
on Linear Controllability and Observability Norms: An Invariant
Manifold Approach,” Journal of Sound and Vibration, Vol. 199,
No. 3, 417–429.
Sommerfeldt, S. D., and Nashif, P., 1994, “An Adaptive Filtered-x Algo-
rithm for Energy-Based Active Control,” Journal of the Acoustic
Society of America, Vol. 96, No. 1, 300–306.
Soong, T. T., 1990, Active Structural Control: Theory and Practice,
Wiley, New York.
Soong, T. T., and Dargush, G. F., 1997, Passive Energy Dissipation in
Structural Engineering, Wiley, New York.
Sun, F. P., Liang, C., and Rogers, C. A., 1994, “Structural Modal Analy-
sis Using Collocated Piezoelectric Actuator/Sensors – an Electrome-
chanical Approach,” Proceedings of the SPIE, Vol. 2190, 238–249.
Downloaded from http://svd.sagepub.com at PENNSYLVANIA STATE UNIV on April 16, 2008
© 2003 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
383
Sun, L., Krodkiewski, J. M., and Cen, Y., 1998, “Self-Tuning Adaptive
Control of Forced Vibration in Rotor Systems Using an Active
Journal Bearing,” Journal of Sound and Vibration, Vol. 213, No. 1,
1–14.
Teng, T.-L., Peng, C.-P., and Chuang, C., 2000, “A Study on the Appli-
cation of Fuzzy Theory to Structural Active Control,” Computer
Methods in Applied Mechanics and Engineering, Vol. 189, No. 2,
439–448.
Tokhi, M. O., and Hossain, M. A., 1996, “A Unified Adaptive Active
Control Mechanism for Noise Cancellation and Vibration Suppres-
sion,” Mechanical Systems and Signals Processing, Vol. 10, No. 6,
667–682.
Triller, M. J., and Kammer, D. C., 1994, “Controllability and Observability
Measure for Craig–Bampton Substructure Representations,” Journal
of Guidance, Control, and Dynamics, Vol. 17, No. 6, 1198–1204.
Tuer, K. L., Duquette, A. P., and Golnaraghi, M. F., 1991, “Vibration
Control of a Flexible Beam Using a Rotational Internal Resonance
Controller, Part I: Theoretical Development and Analysis,” Journal
of Sound and Vibration, Vol. 167, No. 1, 41–62.
Valoor, M. T., Chandrashekhara, K., and Agarwal, S., 2001, “Self-Adaptive
Vibration Control of Smart Composite Beams Using Recurrent Neu-
ral Architecture,” International Journal of Solids and Structures,
Vol. 38, No. 44/45, 7857–7874.
Vipperman, J. S., Burdisso, R. A., and Fuller, C. R., 1993, “Active Con-
trol of Broadband Structural Vibration Using the LMS Adaptive
Algorithm,” Journal of Sound and Vibration, Vol. 166, No. 2,
283–299.
Visioli, A., 2001, “Tuning PID Controllers with Fuzzy Logic,” IEE Pro-
ceedings. Control Theory and Applications, Vol. 148, No. 1, 1–8.
Wang, Q., and Wang, C. M., 2001, “A Controllability Index for Optimal
Design of Piezoelectric Actuators in Vibration Control of Beam
Structures,” Journal of Sound and Vibration, Vol. 242, No. 3,
507–518.
Wang, S., Yeh, H. Y., and Roschke, P. N., 2001, “Robust Control of
Structural
System with Parametric and Unstructured Uncertainties,” Journal
of Vibration and Control, Vol. 7, 753–772.
Yae, K. H., and Inman, J., 1993, “Control-Oriented Order Reduction of
Finite Element Model,” Journal of Dynamic Systems, Measure-
ment, and Control, Vol. 115, 708.
Yam, L. H., and Li, Y. Y., 2001, “Sensitivity Analysis of Sensor Loca-
tion for Vibration Control and Damage Detection of Thin Plate
Systems,” Journal of Sound and Vibration, Vol. 240, No. 4, 623–636.
Yam, L. H., and Yan, Y. J., 2002, “Optimal Design of Number and Loca-
tions of Actuators in Active Vibration Control of Space Truss,”
Smart Materials and Structures, Vol. 11, 496–503.
Yang, Q. J., Zhang, P. Q., Li, C. Q., and Wu, X. P., 1994, “System The-
ory Approach to Multi-Input Multi-Output Modal Parameters Iden-
tification Methods,” Mechanical Systems and Signal Processing,
Vol. 8, No. 2, 159–174.
Yoshimura, T., Edokoro, K., and Ananthanarayana, N., 1993, “An Active
Suspension Model For Rail/Vehicle Systems With Preview and
Stochastic Optimal Control,” Journal of Sound and Vibration, Vol.
166, No. 3, 507–519.
Yoshimura, T., Nakaminami, K., Kurimoto, M., and Hino, J., 1999, “Active
Suspension of Passenger Cars Using Linear and Fuzzy-Logic Con-
trols,” Control Engineering Practice, Vol. 7, No. 1, 41–47.
Zhang, J. F., 2002, “Optimal Control for Mechanical Vibration Systems
Based on Second-Order Matrix Equations,” Mechanical Systems
and Signal Processing, Vol. 16, No. 1, 61–67.
Zhang, X., Shao, C., Li, S., Xu, D., and Erdman, A. G., 2001, “Robust
H8 Vibration Control for Flexible Linkage Mechanism Systems
with Piezoelectric Sensors and Actuators,” Journal of Sound and
Vibration, Vol. 243, No. 1, 145–155.
Zhang, C.L., Mei, D.Q., and Chen, Z.C., 2002, “Active Vibration Isola-
tion of a Micro-Manufacturing Platform Based on a Neural Net-
work,” Journal of Materials Processing Technology, Vol. 129, No.
1–3, 634–639.
Zimmerman, D. C., 1993, “A Darwinian Approach to the Actuator Num-
ber and Placement Problem with Non-Negligible Actuator Mass,”
Mechanical Systems and Signal Processing, Vol. 7, No. 4, 363–374.
Zimmerman, D. V., and Inman, D. J., 1990, “On the Nature of the Inter-
action Between Structures and Proof-Mass Actuators,” Journal of
Guidance and Control, Vol. 13, No. 1, 82–88.