Nonlinear Dyn (2021) 104:13–34

https://doi.org/10.1007/s11071-021-06303-8

FEATURE ARTICLE

A review of the mechanical inerter: historical context,

physical realisations and nonlinear applications
David J. Wagg

Received: 27 November 2020 / Accepted: 10 February 2021 / Published online: 24 February 2021
© The Author(s) 2021

Abstract In this paper, a review of the nonlinear 1 Introduction

aspects of the mechanical inerter will be presented.
The historical context goes back to the development During the last 20 years, mechanical inerter devices
of isolators and absorbers in the first half of the twen- have been the subject of substantial research interest in
tieth century. Both mechanical and fluid-based non- both academia and industry. These devices have been a
linear inerter devices were developed in the mid- and major innovation in the research field of passive vibra-
late twentieth century. However, interest in the inerter tion control. They have also been primarily described
really accelerated in the early 2000s following the work in the literature in terms of linear vibration phenom-
of Smith [87], who coined the term ‘inerter’ in the ena. However, as this field of research begins to mature,
context of a force–current analogy between electri- there is a growing recognition that nonlinearity plays
cal and mechanical networks. Following the histori- a significant role in these devices. For example, most
cal context, both fluid and mechanical inerter devices of the mechanisms used to realise mechanical inerter
will be reviewed. Then, the application of nonlin- devices in practice are nonlinear, such as gears, ball-
ear inerter-based isolators and absorbers is discussed. screw mechanisms and fluid flow. Furthermore, a num-
These include different types of nonlinear energy sinks, ber of research studies have been undertaken that are
nonlinear inerter isolators and geometrically nonlin- applying the inerter as nonlinear vibration mitigation
ear inerter devices, many relying on concepts such as methods.
quasi-zero-stiffness springs. Finally, rocking structures This paper presents a review of the nonlinear
with inerters attached are considered, before conclu- dynamics aspects of the mechanical inerter—although
sions and some future directions for research are pre- to cover the historical context, a limited amount of
sented. linear dynamics is discussed as well. Inerters are
often considered as vibration control devices, and this
Keywords Nonlinear · Inerter · Mechanical review will cover only passive devices. Semi-active and
active control applications are not considered in detail,
although they will be mentioned in the context of future
developments—see also [2,14,36,38,59,99,122] and
references therein for a selection of topics in these
research fields.
D. J. Wagg (B)
The paper is structured as follows. In Sect. 2, the
Department of Mechanical Engineering, University of
Sheffield, S1 3JD Sheffield, UK historical context and background of the mechanical
e-mail: david.wagg@sheffield.ac.uk inerter are presented. Then, in Sect. 3 physical reali-

123
14 D. J. Wagg

sations of mechanical inerter devices are considered in and stiffness k (the primary system is assumed to have
detail. Section 4 considers nonlinear applications of the zero, or close to zero damping). The response of the pri-
mechanical inerter. Finally, in Sect. 5, conclusions and mary system (without the TMD but with a small amount
future directions for research are presented. of damping) to a sinusoidal excitation is shown as a
solid line in Fig. 1b, and a large displacement resonance
peak can be seen. The response of the same system after
2 History and background to the mechanical the TMD, (also called a tuned-vibration-absorber and
inerter dynamic-vibration-absorber,) has been added is shown
as a dashed line in Fig. 1b. It is clear that the idea
In order to put the inerter into historical context, the proposed by Frahm leads to a dramatic reduction in
first step is to consider the state-of-the-art methods in maximum amplitude. Despite being over 100 years old,
passive vibration control prior to the advent of inerter- remarkably this idea was until very recently the state
based techniques. of the art in almost all relevant areas of engineering
The term ‘inerter’ was first introduced by Smith practice.
[87] using a force–current analogy between mechanical The idea was both popularised and given a rig-
and electrical networks—see also [88]. In this context, orous design process by J.P. Den Hartog [20] and
the inerter is considered to represent the equivalent of then J. E. Brock [11] since when it has been used
the capacitor in electrical networks. As a result, in the extensively across all engineering (with some further
mechanical domain, it has the property that the force refinements—see for example Liu and Liu [60] and
generated is proportional to the relative acceleration references therein). Probably the most famous exam-
between its end points (also called terminals, ports or ple is in the Taipei 101 in Taiwan. This is a 509m
nodes). The constant of proportionality for the inerter high skyscraper, which between 2004 and 2008 was
is called inertance and is measured in kilograms. the tallest building in the world. Taipei suffers from
Long before this definition, engineers were seeking typhoon storms and earthquakes, so the building was
methods to reduce unwanted vibrations, particularly fitted with a tuned-mass-damper using a mass of 660
from resonances. We start by considering one of the tonnes, as shown in Fig. 1c, d. The mass is suspended
earliest proposed solutions to this problem—see also on cables and swings when the building is shaken by
Titurus [100] and Kuhnert et al. [51] for additional his- wind or earthquake. This swing motion is tuned to give
torical perspectives of the inerter. the same cancellation effect of the largest vibrations
based on Frahm’s idea (see review by Gutierrez and
Adeli [32] for a list of TMDs in tall buildings).
2.1 The tuned-mass-damper (TMD) The tuned-mass-damper in the Taipei 101 has been
shown to work amazingly well, but the fact that a 660-
In October 1909, Hermann Frahm filed a patent on a tonne pendulum is required highlights one of the major
new device for ‘damping vibrations of bodies’ [27]. drawbacks with the tuned-mass-damper idea. As the
The idea was simple, and went on to radically improve structure gets larger (or a greater damping effect is
many engineering applications where unwanted vibra- needed), then the mass required also becomes larger,
tions occurred. It was based on the observation that the which has several disadvantages, not least the cost and
resonance of an oscillating system could be reduced by large space required inside the structure. Apart from
deliberately attaching a smaller oscillator to the sys- having to use very large masses in large structures, the
tem. The key insight was that if the resonance fre- conventional TMD also suffers from two other impor-
quency of the smaller device was designed in a par- tant limitations. The first is that the sharp nature of the
ticular way, then the two systems interfered with each resonance peak (i.e. the solid line in Fig. 1b), means
so that the largest amplitudes of vibration were dramat- that small amounts of tuning error (for example from
ically reduced. parameter changes over time) result in a rapid loss of
The concept is shown in Fig. 1a, where the smaller performance. The second is that for systems with mul-
oscillator—called the tuned-mass-damper (TMD)— tiple resonances (which applies to very many real appli-
with mass m a , damping ca and stiffness ka is shown cations), the TMD can only suppress vibrations of one
attached to the primary (or host) system, with mass M resonance peak. In fact, this is less of a problem in large

123
A review of the mechanical inerter 15

(a) (b)

(c) (d)

Fig. 1 The tuned-mass-damper (or tuned-vibration-absorber) ences therein). Start with the given parameters of k = 1000N/m,
showing a a schematic diagram of the primary system, M, k, M = 100kg, m a = 8 kg and c = 10 kg/s (i.e. dampingclose to
with the absorber, m a , ka , ca , attached. b A simulation of the kμ
zero). Then, let μ = m a /M, so that ka = (1+μ) 3μ
2, ζ = (1+μ)3
(damped) primary system without the absorber attached (solid
from which, ca = 2ζ m a ω p , see for example Liu and Liu [60].
line) and tuned-mass-damper (dashed line) subjected to sinu-
c A photograph of the 660-tonne mass from the Taipei 101 tuned-
√ where F = 1N. The frequency ratio
soidal excitation F sin(t)
mass-damper is shown. d The mass is suspended on cables, across
is /ω p where ω p = k/M. The TMD was designed using the
four storeys at the top of the building, and the mass acts like a
‘fixed point’ method of Brock [11] (see also Den Hartog [20], and
pendulum version of the TMD. A review of TMDs with a list of
note that these are the ‘fixed point’ design rules that have been
applications to buildings is reported in Gutierrez and Adeli [32].
adapted for the design of inerter devices—see [37] and refer-
Photograph credits: Guillaume Paumier

buildings, where typically 80% of the transverse vibra- 2.2 The dynamic antiresonant vibration isolator
tion is from a single resonance, but is a major problem (DAVI)
in other applications.
Two of these problems can be mitigated to a large During the early part of the twentieth century, the devel-
extent by inerter-based devices. Firstly, an inerter- oping aerospace industry was concerned with issues
based device creates an inertial force that is signif- related to control and stability, during which time sev-
icantly greater than its own mass. At the civil engi- eral inerter like devices were developed—see for exam-
neering scale, Sugimura et al. [92] described a system ple the literature reviews given in [51,100]. Several of
where the inertial force created was nearly 10000 times these novel devices will be discussed in this review, the
greater than the mass of the device. In addition, because first of which is a mechanical device called the dynamic
an inerter is a relative motion device, it has an effect on antiresonant vibration isolator (DAVI), first patented in
all the resonances in the system, although it is typically 1967 by Flannelly [25]. The antiresonance in the DAVI
still just one that is targeted in the design process. This was exploited in the aerospace industry for applications
and other effects are further described in Sect. 4. including isolating the fuselage of a helicopter against

123
16 D. J. Wagg

(a) (b)

(c)

Fig. 2 The dynamic antiresonant vibration isolator (DAVI) with the DAVI (dashed line) subjected to sinusoidal excitation
showing a a schematic diagram of the primary system (M, k √ where F = 1N. The frequency ratio is /ω p where
F sin(t)
and c) with the DAVI rod (m, J , a and b ) attached. Note that ω p = k/M, and transmissibility is |X/Y |. The parameters are
the vertical displacement of the mass, m, is given by x0 and the k = 80 N/m, M = 50 kg, m = 10 kg, J = 22.5 kg m2 , a = 1m,
angular rotation of the DAVI rod is denoted by ψ. b A simulation b = 1.5m and c = 5 kg/s (i.e. damping close to zero). c The
of the primary system without the DAVI attached (solid line), and prototype DAVI Alpha, reproduced from [46]

the vibration caused by its rotors, as described by Des- equations of motion in terms of just x and y. There-
jardins and Hooper [21]. fore, to eliminate x0 and ψ the following relationships
It should be noted that the DAVI is a vibration isola- are used x0 = αy − (α − 1)x, where α = b /a , and
tor whereas the tuned-mass-damper, shown in Fig. 1, is
a vibration absorber. An example of the DAVI concept y−x ÿ − ẍ
is shown schematically in Fig. 2a, where the objective sin(ψ) = ≈ψ  ψ̈ ≈ (1)
a a
is to isolate the mass M from the support input y.
Although the DAVI system shown in Fig. 2a con-
tains geometric nonlinearities, to the authors knowl- which restricts the subsequent analysis to small angle
edge there has not been any research on the nonlin- ranges for ψ. Note also that a and b must be chosen
ear version of this system. Instead, we present the lin- so that α > 1.
earised version which is related to the original deriva- Considering the equilibrium of mass M and the
tions that can be found in Anderson and Smith [3] and DAVI rod gives
Jones [46], see also [7,10,21,23,62,119].
In order to derive a transmissibility relationship M ẍ + c(ẋ − ẏ) + k(x − y) + F p = 0, and
(max(x)/max(y) as frequency is varied) for the DAVI (2)
 
system shown in Fig. 2a, it is required to derive the Fp = 1
−m ẍ0 (b − a ) − J ψ̈ ,
a

123
A review of the mechanical inerter 17

where F p is the force on the pivot point attached to M. where ωa is the frequency where the antiresonance
Substituting for x0 and ψ̈ using the expressions above occurs (the zero of (5)), and ωr is the resonance fre-
leads to an equation of motion of quency of the isolated system (the positive pole of (5)).
The damped transmissibility function becomes
 
M + m(α − 1)2 + J2 ẍ
 a  X k − b2 2 + ic
+c ẋ + kx = mα(α − 1) + J
ÿ + c ẏ + ky,  =
a2 Y k − (M + b1 )2 + ic
  
X (1 − μ2 ω̂2 )2 + (2ζ ω̂)2
(M + b1 )ẍ + c ẋ + kx = b2 ÿ + c ẏ + ky,    = (7)
Y (1 − (1 + μ1 )ω̂2 )2 + (2ζ ω̂)2
(3)
where c is the viscous damping coefficient, μ1 = b1 /M
where b1 = m(α−1)2 + J2 and b2 = mα(α−1)+ J2
are and μ2 = b2 /M are the inertance to mass ratios,
a a ω̂ = /ω p is the frequency ratio, ζ = c/2Mω p is
the inertance values for the DAVI in units of kilograms. √
the damping ratio, and ω p = k/M is the undamped
It can be seen that by adjusting the parameters of the
natural frequency of the primary system.
DAVI, namely m, J , a and b , the level of inertance
The response of the primary system without the
generated can be defined. This is an important property
DAVI is shown as the solid line in Fig. 2b. The cross-
of an inerter device that allows the vibration mitigation
over frequency ratio is the value where the transmissi-
strategy to be relatively easily designed.
bility equals one (for nonzero frequency ratio values).
Making the idealised assumption that the bar of the
To the left of the cross-over frequency is the amplifi-
DAVI has no mass such that J = m2b leads to inertance
cation region (meaning |X/Y | > 1) and to the right
values of b1 = m((α −1)2 +α 2 ) and b2 = mα(2α −1)
of the cross-over frequency is the attenuation region
and a governing equation given by
(meaning |X/Y | < 1). This can be compared to the
[M + m((α − 1)2 + α 2 )]ẍ + c ẋ + kx primary system with DAVI, computed using Eq. (7),
and shown as the dashed line in Fig. 2b. Note that, now
= mα(2α − 1) ÿ + c ẏ + ky. (4)
the DAVI has been added, the transmissibility plot has
This is the governing equation used to compute the both a resonance and an antiresonance peak.
results shown in Fig. 2b. The DAVI response (dashed line) in Fig. 2b does
Results for the isolator are expressed in terms of three important things: (i) reduces the height of the
the transmissibility, max(x) divided by max(y) for resonance peak, (ii) moves the cross-over frequency to
each frequency value across the range considered. Now the left, which reduces the amplification region, and
assuming that the input y and response x are both sinu- (iii) creates an antiresonance, where the amplitudes
soidal, then Eq. (3) leads to an undamped (i.e. by setting of response are dramatically reduced. In terms of the
c = 0) transmissibility relationship of level of reduction at the antiresonance, it can be seen
for the example in Fig. 2b that there are approximately
two orders of magnitude between resonance peak and
X k − b2 2 antiresonance (in the idealised case, when c = 0 trans-
= (5)
Y k − (M + b1 )2 missibility is zero at the antiresonance). If the operat-
ing point of the primary system can be moved close
where X and Y are the displacement amplitudes of to the antiresonance, then large reductions in vibration
sinusoidal x and y signals, respectively, and  is the transmission can be achieved, and this is a common
frequency of the sinusoidal support motion. For this approach in applications—see Jones [46] for a detailed
undamped DAVI system, there are two important fre- design methodology.
quency values Although there have been more recent applications
of the DAVI (e.g. Liu et al. [62]), the concept was
  not used extensively following the initial development.
k k However, the DAVI did help as a design model for other
ωa = and ωr = (6)
b2 M + b1 vibration isolation problems. In particular, the ability to

123
18 D. J. Wagg

introduce an antiresonance was important as a design this can be achieved is by reducing the stiffness, and (if
possibility for passive vibration isolation. As a result, possible) increasing the damping.
the DAVI-type model has subsequently been used for An example is shown in Fig. 3b) where the response
other vibration isolation devices. For example, it has of the primary system, with parameters M, c, k, is plot-
been used to develop certain types of automotive engine ted as the solid line. The dot-dashed line in Fig. 3b)
mounts that are a type of fluid inerter. This, and other shows the response of a linear isolator, where the stiff-
mechanical inerter devices, will be discussed in the next ness, k I is less than the primary stiffness, k. In addi-
section. tion, the damping of the linear isolator is greater than
More recently, inerter isolation devices with spring, the primary system, c I > c, and it can be seen that the
damper and inerter in parallel have been studied height of the resonance peak is reduced. Furthermore,
extensively—see [37] for an analysis of the paral- the cross-over frequency has moved to the left, there-
lel inerter isolator case. It is interesting to note that fore increasing the attenuation region, when compared
the transmissibility expressions found for the DAVI in to the primary system curve (solid line).
Equation (7) are equivalent to those for the parallel However, reducing the stiffness is often undesirable
inerter isolator case if it is assumed that μ1 ≈ μ2 . This (or impossible) in many practical applications, as the
type of parallel inerter isolator can be realised in prac- primary mass needs to be supported without excessive
tice using a flywheel inerter, as will be described in static deflection. Therefore, it is often preferable to seek
more detail in Sect. 3.2. alternatives where the stiffness of the primary system
does not need to be reduced. From Equation (6) (assum-
ing the DAVI-type model is appropriate), it can be seen
3 Mechanical inerter devices that the natural frequency of the system, ωr can be
reduced by increasing the inertance, without needing
3.1 Fluid inerter devices to change either the mass or linear stiffness values.
The hydramount was designed to improve the per-
3.1.1 The hydramount formance of a linear mount by introducing inertance
using hydraulic fluid that is forced between two cham-
An alternative to generating inertial forces using a lever bers via a helical pipe (also called an annulus or ‘inertia
mechanism, such as the DAVI, is to use fluid flow within track’), as shown schematically in Fig. 3a. As the fluid
a chamber or pipe. This concept was developed exten- rotates around the helical pipe, H , it creates a ‘fluid
sively in the twentieth century for aerospace and auto- flywheel’ effect that can be designed to give an antires-
motive applications, and led to a wide range of devices onance in a similar way to the DAVI. In fact, Flower
generally referred to as ‘shock absorbers’ and ‘isolation [26], proposed a design process that used the DAVI
mounts’ amongst other terminology. A more detailed lever arm model, very similar to that shown in Fig. 2a,
historical review of fluid-based devices is given by Titu- in order to approximate the effect of a rotating fluid
rus [100]. One example, relevant to this review was a inside the mount, and this approach has been used to
fluid inerter that was incorporated into an automotive compute the hydramount response (dashed line) shown
engine mount, called the ‘hydraulic mount’ or some- in Fig. 3b. As a result, the hydramount can create an
times ‘hydramount’—see Flower [26] and references antiresonance, and this can be designed to be very close
therein. to the operating frequency range of the system such that
Vibration isolation mounts of this type had long the isolation effect is maximised.
been based on a rubber element that acted as a com- Note that away from the antiresonance, for exam-
bined ‘spring’ and ‘damper’—see for example Rivin ple at frequency ratios above 2, the hydramount (and
[77]. Although rubber used in this way has nonlinear DAVI in the previous example) are worse than the linear
restoring force and damping properties, many designs case. This demonstrates why the frequency of operation
assume a linear model, typically like the parallel spring needs to be close to the antiresonance in order to work
k and damper c of the system shown in Fig. 3a without effectively. More recent studies of the hydramount are
the hydramount. As a result, the design objective of given by Singh [86], Golnaraghi and Nakhaie [30] and
such a linear isolator is to reduce the resonance peak, Soltani et al. [91].
and move the cross-over frequency to the left. One way

123
A review of the mechanical inerter 19

(a) (b)

(d)
(c)

Fig. 3 Fluid inerter concept, showing a a schematic diagram of of helical fluid inerter. d A comparison of a helical fluid inerter
the hydramount system, where engine mass, M, is to be isolated test results from Smith and Wagg [90] and a linear hysteretic
from chassis input y(t). The rubber mount is assumed to be a model based on the force in the fluid inerter being approximated
linear (i.e. k and c) with the fluid inerter consisting of hydraulic as Fh = bh z̈ + ch ż, where bh is the inertance, ch is the additional
fluid that is passed between an upper, U , and lower, L, cham- viscous damping within the fluid inerter and z = x − y is the
ber via a helical pipe H . b a simulation of the primary system relative displacement across the terminals of the inerter device
(M, k, c solid line), linear isolator (M, k I , c I dot-dash line) and (terminals are shown as a and√b in (c)). When sinusoidal inputs
hydramount (dashed line, computed using a DAVI model) sub- are assumed, Fh = ch ż ±bh  2 Z 2 − ż 2 , where Z is the max-
jected to sinusoidal excitation F sin(t) √ where F = 1N. The imum amplitude of the sine wave with frequency . This forms
frequency ratio is /ω p where ω p = k/M, and transmissi- an ellipse, which is shown as a dotted line in (d). The identified
bility is |X/Y |. The parameters are k = 80N/m, k I = 60N/m, parameters for the model are close to the previously estimated
M = 50kg, m = 10kg, J = 22.5kgm2 , a = 1, b = 1.5, values from Smith and Wagg [90], and were given by bh = 98.4
c = 5kg/s and c I = 10kg/s. c Schematic cross-sectional diagram kg and ch = 1628 kg/s

3.1.2 The helical fluid inerter Rivin [77] and Titurus [100]. The first known applica-
tion of a fluid inerter device is the ‘mass pump’ devel-
Helical tubes of fluid had been proposed as useful oped by Kawamata [47–49] starting in the 1970s. How-
components in dampers a considerable time before ever, interest in helical fluid-filled tubes used as inerters
the development of the hydramount—see for example only really gained significant momentum following the
O’Connor [71]. A more detailed historical description work of Swift et al. [94].
and comprehensive literature reviews can be found in

123
20 D. J. Wagg

In contrast to a hydramount, a helical fluid inerter is inerter shown in Fig. 3c are modelled using
not typically integrated into the mount. The most com-
mon practical realisation is a fluid (usually hydraulic f total = f iner tial + f damping + f f riction (10)
oil) filled cylinder with a helical tube wrapped around
the outside, as shown schematically in Fig. 3c. The
where it is noted that entry and exit losses between the
radius of the main fluid chamber is given by r2 , and
cylinder and the helix tend to be neglected as they’re
the piston through rod which pushes the fluid inside
(usually) small compared to the other effects—see for
the main chamber has radius r1 . The distance r4 is the
example discussions in [15,63,82,83,94]. In terms of
helix radius from the centre of the longitudinal axis of
capturing the physical behaviour, both damping and
the cylinder and r3 is the inner radius of the helical pipe.
frictional effects will be nonlinear.
The cross-sectional area of the cylinder is A1 =
For example, an expression using nonlinear fluid
π(r22 − r12 ) and the cross-sectional area of the helix
damping combined with a Coulomb-type friction model
is A2 = πr32 . The principal of conservation of mass
leads to a force expression of
is normally applied to derive an expression equating a
linear (relative) displacement in the cylinder, z = x − y
to an angular displacement of a fluid element in the f total = bhel z̈ + cd ż β + f 0 sgn(z), with z = x − y
helix, θ . Taking the mass of the fluid in the helix as (11)
m hel ≈ ρ f L h A2 , where ρ f is the mass density of the
fluid at reference temperature and L h is the length of
where cd is the nonlinear damping coefficient, f 0 is the
the helix, then the moment of inertia about the axis of
static friction coefficient, and β = 1.75 is the nonlinear
the piston is defined as J = m hel r42 . Making a series of
damping exponent. This derivation (see for example
assumptions about the ideal nature of the device (see
De Domenico et al. [15]) assumes turbulent flow and a
for example Swift et al. [94]) leads to the idealised
smooth pipe, so that the nonlinear damping coefficient
relationship for the equivalence of kinetic energy in
can be approximated as
the inerter
1.75
L h A1 A1
1 1 cd = 0.664μ0.25
f ρf
0.75
(12)
bhel ż 2 = J θ̇ 2 (8) r31.25 A2
2 2
where μ f is the dynamic viscosity of the fluid. Spe-
where θ is the rotation angle of the fluid in the helix, cific other examples can be found in [63,64,82,83,94],
ż is the relative velocity between the end points of the where other choices for β such as β = 2 are discussed.
inerter, and bhel is the inertance of the fluid in the helix. As discussed for the DAVI and hydramount exam-
These definitions can be used to derive the following ples, inerters are used in combination with other ele-
expression [28] ments, such as masses, springs and dampers, in order
to create vibration absorbers or isolators. This presents
m hel A1 2 two problems for fluid inerters; (i) the tuning rules for
bhel = , (9) isolators and absorbers (e.g. [37]) are linear, and so
(1 + (h/(2πr4 ))2 ) A2
do not translate to nonlinear systems, and (ii) there is
a strong coupling between the inertance and damping
where h is the pitch of the helix. As a result, bhel can be (i.e. see Equations (9) and (12)) making it very difficult
designed using the geometry of the cylinder and helix to design and specify separate inertance and damping
using Equation (9). values.
Fluid inerters have a significant level of inherent One approach used by De Domenico et al. [15] for
damping due to the fluid dynamic effects. This leads earthquake excitation was to apply statistical lineari-
to nonlinear relationships in terms of the velocity, in sation (for an introduction to this topic, see Roberts
addition to which, the friction effects of the piston are and Spanos [78]) which then resulted in a constrained
significant and also nonlinear in nature. In general, the optimisation problem to find the optimal values of bhel
forces between terminals a and b of the helical fluid and cd (friction was neglected, on the basis that for

123
A review of the mechanical inerter 21

(a) (b)

(c) (d)

Fig. 4 Mechanical flywheel inerters, showing a a schematic gears to amplify the flywheel effect. c A flywheel-driven ‘ball-
diagram of a pivoted flywheel inerter, where mass, M, is to be screw’ inerter, with flywheel parameters J , m. d A viscous mass
isolated from input y(t). b A rack-and-pinion, geared flywheel damper [41], which consists of a ball-screw inerter in combina-
inerter called the ‘gyro-mass’ inerter from Saitoh [79], that uses tion with viscous oil damping

high force and frequency inputs, like earthquake exci- and Pei [101], and then friction as well by Zhang et al.
tation, the Stribeck effect reduced the significance of [124]. These studies also included some comparison
the friction force). An alternative approach is to con- between mem- and nonlinear models for the inerter, an
sider different geometries, such as that considered by idea that was also considered by Biolek et al.[8,9] in
Liu et al. [64], where a meander tube was shown to the context of higher-order electrical elements.
give much lower damping values for similar inertance In the next section, we consider a final class of inert-
values to the helix case. ers to be discussed in this review, those that use rota-
Results from an experimental test with a helical tional flywheel effects to create inertance.
fluid inerter are shown in Fig. 3d. It can be seen from
Fig. 3d that the experimental hysteresis loop is strongly
affected by friction, in the regions where velocity 3.2 Mechanical flywheel inerters
changes sign. In comparison, the elliptical linear hys-
teresis model represents a limited approximation, and 3.2.1 Pivoted flywheel and Rack-and-pinion inerters
in these types of inerters, the nonlinear effects are there-
fore highly significant. The flywheel is an ancient technology that has been
It has been noted by some authors that both the used in applications throughout human history as
damping and inertance behaviour of helical fluid inert- described by White Jr [116]. One of the main benefits of
ers have strong similarities with memory element mod- a flywheel is its simplicity. For example, the lever arm
els (also called mem-models). Zhang et al. [123] pro- design of the DAVI, shown in Fig. 2 has an asymmetry
posed the mem-inerter element that is able to cap- which leads to two inertance values, given in Eq. (3).
ture the inerter hysteresis effect. This idea was further This asymmetry is removed if the lever and mass are
extended, using data from experimental tests, to include replaced with a flywheel, to create a pivoted flywheel
memory effects in both inertance and damping by Wagg inerter, as shown schematically in Fig. 4a. Note that to
fully eliminate the asymmetry, the pivots would need

123
22 D. J. Wagg

  
to be equally spaced from the centre of the flywheel, X (1 − μ p ω̂2 )2 + (2ζ ω̂)2
 = (17)
however the design with one pivot offset and one close Y  (1 − (1 + μ p )ω̂2 )2 + (2ζ ω̂)2
to the centre has been found easier to implement from a
practical perspective—see for example John and Wagg where μ p = b p /M and all other parameters are the
[45]—and it is also a sufficiently close approximation same as previously defined in Eq. (7).
for small angles of rotation. One of the earliest inerter devices of this type is the
For the system in Fig. 4a, the flywheel is assumed to so-called gyro-mass which was patented in 1997 by
produce a couple equal to Fa , where F is the force on Okumura [72]. This device used a rack-and-pinion gear
each of the pivots (which are assumed to be massless). system in order to amplify the effect of the flywheel, and
The couple can be directly related to the torque, T , and a related device is shown as an example in Fig. 4b from
the angular acceleration via Saitoh [79]. Another early example of the rack-and-
pinion inerter concept was discussed by Smith [87] and
T = Fa = J φ̈ (13) physical realisations of rack-and-pinion devices that
have been tested experimentally are given by Smith
where J is the moment of inertia of the flywheel, φ is and Wang [89], Papageorgiou et al. [73], Saitoh [79]
the angle of rotation and an overdot represents differen- and Madhamshetty and Manimala [66]. When gears
tiation with respect to time, t. Using the same approxi- are used, the inertance relationship will become
mation for small angles as in Eq. (1) (i.e. with φ instead
J
of ψ) gives bp ∝ , (18)
a2
F = b p ( ÿ − ẍ), (14)
where the proportionality is related to gear ratios, as
for example described by [67,79,87,97].
where the inertance, b p , is defined as

J 3.2.2 Ball-screw inerters

bp = . (15)
a2
In Fig. 4c, a schematic diagram of an inerter system is
Note that, unlike the DAVI, there is now only a single shown where the flywheel rotation occurs in the hori-
inertance value, b p , given by Eq. (15). Furthermore, b p zontal plane. The flywheel rotation, θ , is driven by the
has only two parameters, the moment of inertia J = relative motion of the device in the vertical plane. These
mr 2 types of inerters are generally referred to as ball-screw
2 , and the lever arm distance a as given in Equation inerters, and a derivation and some historical context
(15)
are given by Rivin [77]. For the system in Fig. 4c, the
Using the above relationships, the equation of
torque, T , is now related to the vertical force, F, gen-
motion for the pivoted flywheel inerter system shown
erated by the flywheel by assuming it acts like a nut, so
in Fig. 4a is
that
M ẍ + k(x − y) + c(ẋ − ẏ) + b p (ẍ − ÿ) = 0, (16) T
T = Frm tan(α ± γ )  F =
rm tan(α ± γ )
where c is the viscous damping parameter. Equation J θ̈
= (19)
16 is the simplest model for an inerter isolator system, rm tan(α ± γ )
and as a result it has been widely used to approximate a where rm is the mean radius of the thread (meaning the
range of devices in the literature—see for example the radius to the centre of the contact region), θ is the rota-
discussion in Hu et al. [37] and other references therein. tion of the flywheel, α is the helix angle of the thread,
As a result of there being just a single inertance value, and γ = arctan(μ f ric ) is the friction angle, where
the damped transmissibility function for Equation (16) μ f ric is the friction coefficient. The plus or minus in
becomes front of γ defines the two cases of the flywheel moving
X k − b p 2 + ic up and down (note we are neglecting the mass of the
= 
Y k − (M + b p )2 + ic screw rod and flywheel housing, plus the gravitational

123
A review of the mechanical inerter 23

contribution from the mass of the flywheel, all of which et al. [73] described the friction and backlash effects
should be small compared to the inertance). that can occur. Both proposed backlash models, and
Then, from the geometry of the helix, it can be shown a method for identifying the parameters of the system
that θ = rmy−xtan α such that the vertical force within the (see also Brzeski and Perlikowski [12] for a related dis-
nut (i.e. flywheel and threaded rod) becomes cussion). As the system had little viscous damping, the
authors were able to use a model similar to Eq. (10)
J ( ÿ − ẍ) with f damping ≈ 0 assumed. Another study that was
F= (20)
rm2 tan(α ± γ ) tan α close to the dry-friction ball-screw case was carried out
by Gonzalez-Buelga et al. [31] who used a commer-
which is a similar relationship to Eq. (14), and we write cially manufactured Penske inerter device with most
it as of the damping oil drained out of it. As a result, the
authors were able to use a similar modelling approach,
J by assuming the device is dominated by inertance and
F = bs ( ÿ − ẍ), with bs = .
rm2 tan(α ± γ ) tan α friction forces. These studies relate strongly to automo-
(21) tive examples which are another large domain of appli-
cation for ball-screw inerters (i.e. Penske manufactures
where bs is the inertance of the ball-screw. The equa- inerter devices for this market). For further discussion
tion of motion for the system in Fig. 4c is the same as of this topic, with reference to the nonlinear effects, see
Eq. (16) with bs instead of b p . for example Wang et al. [109], Sun et al. [93] and Shen
In the idealised case where friction is assumed to be et al. [82,83].
zero (i.e. μ f ric = 0), then we can use the relationship There are some other mechanical inerter devices that
between the pitch of the thread, p, and the helix angle have been proposed that include gears. Two such exam-
given by tan α = 2πrp m to simplify Eq. (21) so that the ples are the rotational inerter, based on a realisation
inertance becomes using an epicyclic gearbox, and the gear-pump inerter
which combines fluid flow and gears—see [88] and
2
4π 2 J 2π κ references therein for a discussion of both these sys-
bs =  bs = M for μ = 0 (22)
p2 p tems. However, for the purposes of this current review,
we now consider the topic of inerter applications that
where κ is the radius of gyration from the relationship make use of nonlinearity.
J = Mκ 2 —see for example the derivation in Smith
[88].
Although is possible in theory to have a dry-friction
ball-screw device, in practice they need to be lubricated. 4 Nonlinear applications of the inerter
In many designs, the fluid provides both lubrication and
viscous damping as well. This has been a particular The nonlinearities described in the mechanical inerter
area of development for civil engineering applications. devices in Sect. 3 are significant at the scale of the
The combination of a ball-screw device with viscous devices, but become less significant when used in a
fluid damping was proposed by Arakaki [4,5] to create larger scale system in combination with other (nom-
the rotary damping tube. This concept was refined by inally) linear elements such as springs and dampers.
subsequent researchers such as Sugimura et al. [92] and For example, when the viscous mass damper, shown
Ikago et al. [41] and is now known as the viscous mass in Fig. 4d, is combined with other elements it can be
damper. The example shown in Fig. 4d is reproduced tuned to give a vibration absorber effect and there-
from Ikago et al. [41]. In the civil engineering domain, fore becomes a tuned-viscous-mass-damper (TVMD)
forces are very large. For example, the viscous mass (see Ikago et al.[41], and note that this device is
damper used in Sugimura et al. [92] had a mass of 560kg also sometimes called the parallel-connected-viscous-
and was able to create an inertance of 5400 tonnes, inerter–damper (PVID)). The absorber tuning can be
whilst the viscous damping was 7300kNs/m. done using a linear tuning approach adapted from the
In terms of the nonlinearities that can occur in ball- tuned-mass-damper described in Sect. 2.1 (see also
screw inerters, Wang and Su [107] and Papageorgiou Fig. 2.1).

123
24 D. J. Wagg

Also using linear theory, and varying the arrange- the (non-grounded) tuned-mass-damper–inerter can be
ment of elements has led to other inerter-based absorbers. reconfigured with a nonlinear spring to give a nonlinear
Most notably, the tuned-inerter–damper (TID) was energy sink-inerter of the TMDI type (NESI-TMDI).
proposed by Lazar et al. [53], and the tuned-mass- Similarly, Javidialesaadi and Wierschem [42] showed
damper–inerter (TMDI) proposed by Marian and Gia- that the grounded tuned-mass-damper–inerter can be
ralis [69]. A wider analysis of other configurations reconfigured with a nonlinear spring to give a nonlinear
can be found in Hu et al. [37] and Krenk [50]. energy sink inerter of the grounded TMDI type (NESI-
These three systems (i.e. TVMD, TID and TMDI) gTMDI).
and variants have been studied extensively for a Considering the equilibrium of mass M in the NESI-
range of applications including vehicle suspensions TID system shown in Fig. 5a gives equations of motion
and steering systems [24,38,52,59,73,82,83,89,93, of
102,109,120], train suspension systems [57,104–106],
and civil engineering systems—see for example [15,
16,18,29,33,34,39,53,54,74,84,85,96,103,121] and M ẍ + c ẋ + kx + cd (ẋ − ẋd ) + f d (x − xd ) = F(t),
references therein. It is also worthy of note that the
TVMD, TID and TMDI have been used to largely mit- and b ẍd + cd (ẋd − ẋ) + f d (xd − x) = 0.
igate the limitations of the TMD described at the end (23)
of Sect. 2.1, although only a small number of concepts
have been deployed in real engineering applications.
The majority of these studies assume the systems Linear tuning rules are no longer applicable to these
are linear. However, the relevance of these applications types of NESI systems. Therefore to obtain optimum
should become apparent as we resume the review of parameters values for the nonlinear device, optimisa-
nonlinear inerter applications. tion methods can be used as an alternative method.
An example of the response of the NESI-TID when
subjected to a sine wave excitation force is shown in
4.1 Nonlinear energy sink inerter devices Fig. 5b. The NESI-TID system response (dot-dash line)
is compared with the linear primary system response
Several studies have been carried out to investigate the (blue solid line, similar to the primary system in Fig. 1),
potential benefits of using nonlinearity to create non- and a linear TID system response (dashed line). It
linear vibration absorbers. One way to do this is to should be noted that the parameters for Fig. 5 (b)
use nonlinear springs instead of linear springs in the have not been optimised directly. Instead, for the pur-
inerter-based devices. The resulting oscillator systems pose of illustrating the concept, we have adapted the
are closely aligned to the concept of a nonlinear energy optimised parameters computed by Javidialesaadi and
sink (NES, see the recent review by Ding and Chen Wierschem [42] for a related grounded TMDI-type sys-
[22]), which in simple terms can be considered to be tem (e.g. a NESI-gTMDI) subject to transient input sig-
analogous to the tuned-mass-damped system shown in nals. Despite this limitation, it can be seen in Fig. 5b
Fig. 1a where the spring, ka is replaced with a nonlinear that close to resonance the NESI-TID (dot-dash line)
spring. has the smallest displacement amplitude, when com-
Devices that fall into the category of being nonlin- pared to the uncontrolled primary system (blue solid
ear energy sink inerter devices are often abbreviated line), and a linear TID system response (dashed line).
by NESI. This can include devices with different types In practice, the exact conditions for optimisation
of layout and different types of nonlinear spring. In are dependent on the application being considered, and
order to distinguish between different devices, we use discussions regarding this type of optimisation can be
an additional classification. For example, in terms of found, for example in Javidialesaadi and Wierschem
system layout, using a cubic nonlinear spring in the [42] or Wang et al. [110]. An alternative approach to
tuned-inerter–damper device (see Zhang et al. [126]) direct optimisation is to carry out a harmonic balance
results in the system shown in Fig. 5a. We call this analysis as a preliminary step before determining the
a nonlinear energy sink-inerter of the TID type, or optimum parameters, see for example Zhang et al. [126]
NESI-TID. Likewise, Zhang et al. [125] showed that or Wang et al. [112]. A comparative study of a negative

123
A review of the mechanical inerter 25

(a) (b)

Fig. 5 Nonlinear energy sink inerter (NESI) example, show- c = 5 kg/s, b = 12.15 kg and F(t) = sin(t) N. For the lin-
ing a a schematic diagram of a nonlinear energy sink inerter of ear TID, the following parameters were used cd = 7.7 kg/s,
the TID type (NESI-TID), where mass, M, is excited by force and f d = 182(x − xd ) N. For the NESI-TID, the spring forces
F(t), and the TID spring has a nonlinear restoring force func- was given by f d = 107 (x − xd )3 N, and √ cd = 10.0 kg/s. The
tion, f d (ẑ), where ẑ = x − xd , b a simulation of the NESI-TID frequency ratio is /ω p where ω p = k/M. Note that these
system (dot-dash line) compared with the linear primary sys- parameters are adapted from the optimised parameters computed
tem (blue solid line), and a linear TID system (dashed line). The by Javidialesaadi and Wierschem [42], who computed optimal
following parameters were used; M = 10 kg, k = 8500 N/m parameters for transient input signals rather than sine waves

stiffness damper and inerter damper was also carried effect of adding a linear inerter to the system, whilst
out by Xiang et al. [117]. keeping the linear spring, is shown as the dot-dashed
There is still the problem of how exactly the required line in the transmissibility plot of Fig. 6d. This has had
nonlinear stiffnesses can be achieved in practice. To the desired effects, and in addition has introduced an
overcome this issue, several authors have recently antiresonance (similar to the examples of Figs. 2 and
studied the idea of combining quasi-zero stiffness 3), and in this case with an inertance to mass ration
mechanisms with inerters, see for example Wang et of μ = 0.4. Increasing μ is one way to continue to
al.[110,112,113] and Yang et al. [118]. Note also that improve the isolation, but what about the situation when
for civil engineering structures related negative stiff- μ is already at the maximum possible value? Are there
ness concepts have been considered, as discussed by ways to improve the situation then?
Luo et al. [65]. An example of this type of quasi-zero One possibility is to use a nonlinear spring in com-
stiffness mechanism will be discussed in detail in the bination with the inerter, as was done in the previous
next section. subsection. The example shown in Fig. 6c is a quasi-
zero nonlinear spring function based on the design
method proposed by Shaw et al. [80], which can also be
4.2 Nonlinear inerter isolators
realised experimentally—see for example Alabuzhev
[1], Shaw et al. [81] and Yang et al. [118]. The equa-
Consider the vibration isolation example shown
tion of motion for the nonlinear quasi-zero and inerter
schematically in Fig. 6a. Here, the requirement is to
isolator in Fig. 6a can be written as
isolate mass, M, from input y(t). The response of the
unisolated system with a linear spring and no inertance
(similar to the primary system in Figs. 2 and 3) when
excited with a sine wave is shown as the thick blue line (m + b)z̈ + cż + f (z) = −m ÿ, (24)
in the transmissibility plot of Fig. 6d. There is a sig-
nificant resonance peak that ideally should be reduced,
along with the amplification region that can be reduced where z is the relative displacement z = x − y. Fol-
by moving the cross-over frequency to the left. The lowing the design method proposed by Shaw et al. [80]

123
26 D. J. Wagg

(a) (b)

(d)
(c)

Fig. 6 Nonlinear inerter isolator example, showing a a with M = 7.2 kg, linear spring k = 80 N/m, c = 1.0Ns/m sub-
schematic diagram of a nonlinear inerter isolator, where mass, ject to a forcing of y = 0.04 sin(t) m and b = 0 (thick blue
M, is to be isolated from input y(t), and the spring has a non- line); the linear spring plus inerter case with the same parameters
linear restoring force function, f (z), where z = x − y, b the except inertance b = 2.9 kg so that μ = b/M = 0.4 (dot-dash
geometrically nonlinear inerter isolator system proposed in de line), and; the quasi-zero plus inerter case with the same param-
Haro Moraes et al. [19] and Wang et al. [115], c an example eters as the previous case except a nonlinear quasi-zero spring
of a quasi-zero force function taken from Shaw et al. [80] where function designed using the method in Shaw et al. [80] with
Fs = Mg is the static force due to gravity g, and ks is the equiva- ks = 52N/m and ẑr = 0.577 is used (thin solid line). Note that
lent linear stiffness, and d a comparison of linear- and nonlinear- the thin dashed line represents the part of the solution branch that
inerter isolators showing; the unisolated linear primary system is unstable

we define the nonlinear stiffness function as the nonlinear stiffness values are computed from

f (z) = k1 z + k3 z 3 + k5 z 5 , (25) Fs k1 5ẑr4 − 1

k3 = (1 − ) ,
z s3 ks 5ẑr4 − 3ẑr2
which is the solid curve shown in Fig. 6c. The stiffness Fs k1 1 − 3ẑr2
and k5 = (1 − ) 4 (26)
values, k1 and ks (see Fig. 6c) are predefined and then 5
zs ks 5ẑr − 3ẑr2

123
A review of the mechanical inerter 27

where ẑr = zr /z s and ±ẑr defines the low stiffness 4.3 Rocking structures and inerters
range of the mount—see Shaw et al. [80] for further
details. In applications such as earthquake engineering, it is
The result of using this nonlinear quasi-zero and possible to have gravity-based structures that can rock
inerter isolator when excited with a sine wave is shown when excited by a ground input motion. An exam-
as the thin solid (and dashed) black line in the transmis- ple is shown in Fig. 7a where, following the classi-
sibility plot of Fig. 6d. It can be seen, that for the same cal analysis of Housner [35], a rectangular block of
inertance values, using a quasi-zero spring instead of a dimensions 2H × 2B and mass, M, is able to rotate
linear spring has further improved the isolation effects about points O and O  when excited by the hori-
when compared to the linear spring plus inerter case zontal ground input acceleration, ag . In the classical
(dot dash line). Notice that the antiresonance is now approach by Housner [35], inerters are not consid-
closer to the position of the original resonance peak ered, but more recently the advantages of using rota-
(although the transmissibility is slightly higher), and by tional inertia for earthquake engineering applications
further tuning the quasi-zero spring properties, it is pos- have been studied by Makris and Kampas [67], Thiers-
sible to locate the antiresonance exactly at the original Moggia and Málaga-Chuquitaype [97,98] and Málaga-
resonance position, thereby maximising the isolation Chuquitaype et al.[68]. The inerter(s) can be config-
benefit if operation is at resonance. ured in a variety of locations, and the example shown
One of the potential drawbacks in using nonlinear schematically in Fig. 7a is chosen for simplicity in order
spring functions is that the dynamic behaviour of the to illustrate the concept.
system is more complex. For example in Fig. 6d, there In order to derive equations of motion for the block
is a small section of dashed line, which represents the in Fig. 7a, we use the fact that the effective force on the
unstable solution branch for the quasi-zero and inerter block at point CG due to a ground acceleration ag is
isolator. Here, there will be saddle-node bifurcations, f eff = −mag R cos(α − θ ). Furthermore, the inerter is
leading to jumps in the displacements as frequency considered to be grounded at the left-hand end, and so
is increased or decreased. Other undesirable, complex the force across the inerter is f I = bḧ C G where ḧ C G
nonlinear behaviours may also be possible, and so care- is the horizontal acceleration of point CG. To com-
ful design is required to avoid any unwanted effects. pute this note that the tangential acceleration of point
Another method that can be used to introduce non- CG when the block is rotating about point O is R θ̈ ,
linear behaviour in practice, is geometrically nonlin- so the horizontal component of this is R θ̈ cos(α − θ ).
ear arrangements of the device elements. For example, As a result, f I = bθ̈ R cos(α − θ ) and the moment
in Fig. 6b another type of the geometrically nonlin- of this about the point O is bθ̈ R 2 cos2 (α − θ ). Now
ear inerter isolator system proposed in de Haro Moraes considering the moment equilibrium of the block in
et al. [19] and Wang et al. [115] is shown, where the Fig. 7a around points O and O  , we obtain the follow-
inerters are mounted horizontally, whilst the spring and ing expressions governing their motion (see for exam-
damper are vertical. ple Thiers-Moggia and Málaga-Chuquitaype [97])
A wider group of systems exhibiting this type of
(Jb + b R 2 cos2 (α − θ ))θ̈ + Mg R sin(α − θ )
geometrically nonlinear arrangement of elements has
recently been studied by Yang et al. [118]. In this study, = −Mag R cos(α − θ ), θ > 0
the authors showed how the arrangements could be used (Jb + b R 2 cos2 (−α − θ ))θ̈ + Mg R sin(−α − θ )
to design specific transmissibility curves, by combin- = −Mag R cos(−α − θ ), θ < 0 (27)
ing the geometrically nonlinear effect with a QZS-type
which can be combined into a single equation
spring system. There are multiple different other con-
figurations that have been considered, and the inter- (Jb + b R 2 cos2 (α sgn(θ ) − θ ))θ̈ + Mg R
ested reader can find recent examples in Zhang et al. sin(α sgn(θ ) − θ ) = −Mag R cos(α sgn(θ ) − θ ),
[127] and Yang et al. [118], where it is noted that some
(28)
systems can give the effect of combining isolation with
absorption. where sgn denotes the signum function, and Jb =
(4/3)M R 2 is the moment of inertia of the block around
the rotation points.

123
28 D. J. Wagg

(a) (b)

(d)
(c)

Fig. 7 Rocking block with inerter, showing a a schematic dia- ing the angle of rotation (b) and angular velocity (c) of a block
gram of rocking block system, where the block with mass, M, is subjected to the sinusoidal impulse shown in (d). Parameter val-
to be isolated from ground motion acceleration input ag , and ues are μ = b/M = 0.5, R = 1.0m, α = 0.0873 radians,
the block can rotate about points O and O  . The inerter is  = 10.85rads/s, A g = 1.5m/s2 . The acceleration input is below
assumed to be grounded at the left-hand
√ end, and has an iner- the level for overturning. See also Thiers-Moggia and Málaga-
tance of b. The parameters R = B 2 + H 2 and α are used Chuquitaype [97] for further details and related results
to define the slenderness of the block. b–d Simulation show-

The block is assumed not to slide in the horizontal type of situation is shown in Fig. 7d, where a single sine
direction at points O and O  , but impacts can occur wave is used as a horizontal acceleration input for the
when the block reaches the vertical position, when θ = rocking block with inerter system. The response to this
0. The impact process is modelled using a coefficient of input is shown in Fig. 7b, c, where the rotation angle
restitution, such that θ+ = r θ− , where θ− is the angular θ is shown in (b) and the angular velocity θ̇ is shown
velocity just before impact, and θ+ is the velocity just in (c). In each of Fig. 7b, c, the solid line is the case
after impact. where the block is simulated with no inerter (i.e. b = 0)
Combining Eq. (28) and the coefficient of restitu- and the dashed line shows the case where the inerter is
tion rule gives a nonlinear model for the rocking block included. Clearly, there is a benefit in having the inerter
with inerter system. For earthquake engineering appli- included for the parameters selected for this example,
cations, the primary interest relates to transient loads as both θ and θ̇ are reduced in overall amplitude.
rather than steady-state response such as transmissibil- However, there are several further complexities of
ity’s described in previous sections. An example of this these types of systems. Firstly real earthquake inputs

123
A review of the mechanical inerter 29

are more complex than the simple signal shown in dynamics typically makes it difficult to determine the
Fig. 7d, and the amplitude can be large enough for optimum parameter values required to minimise the
the block to overturn. Determining the optimal iner- unwanted vibration effects. As a result, optimisation is
tance value is non-trivial, and the possibility of using often used in order to design parameter values for the
clutched inerters can be advantageous in certain situa- nonlinear inerter device.
tions. Clutched inerters are nonlinear mechanisms that Finally, in this review, rocking structures with inert-
enable the inertance to be designed in a semi-active ers attached were considered. These types of appli-
way—see [55,56,58,67,68,97,98,108] and references cations arise in earthquake engineering where ground
therein for further details. accelerations can cause blocks, and similar structural
Other applications in earthquake engineering gen- elements, to tip and rock back and forth. In this situa-
erally assume a linear inerter behaviour, but increas- tion, it is the transient response that is of most interest,
ingly they are being considered in combination with and attaching inerters has been shown to be effective
nonlinear friction damping and/or nonlinear material in limiting rocking behaviour.
properties, such as those used recently to model base As with all reviews, there are limitations, and this
isolation systems as recently shown, for example, by review has only considered a selection of passive inerter
De Domenico and Ricciardi [17] and Zhao et al. [130]. devices. There are many other devices, particularly
semi-active and active control applications that we have
5 Conclusions and future directions for research considered to be outside the scope of this review.

In this paper, a review of the mechanical inerter, with

a particular focus on the nonlinear dynamic behaviour 5.1 Future directions for research
has been presented. Although inerters are often mod-
elled as a linear dynamic phenomenon, the physical It could be argued that the inerter is the most exciting
devices that have been manufactured are typically non- development in the field of structural control since the
linear in nature. That said, linear models are often a patent of the tuned-mass-damper by Frahm in 1909.
reasonable approximation for a wide range of appli- Just like Frahm’s idea, part of the appeal of the inerter
cations, although recent experimental tests (e.g. Piet- is the simplicity of its governing equations, and subse-
rosanti et al. [75]) indicate the nonlinear nature of the quent tuning rules that allow engineers to design pas-
response. sive control systems. The nonlinear dynamics of the
The historical context goes back to the development real manufactured inerter devices is something that has
of isolators and absorbers in the first half of the twen- yet to be fully recognised and studied in depth. This
tieth century. Both mechanical and fluid-based nonlin- is partly because the devices are often used in much
ear inerter devices were developed in the mid- and late larger scale systems where their nonlinear behaviour is
twentieth century. less noticeable. A secondary reason is that there are far
However, interest in the inerter really accelerated fewer studies that have experiments and/or real engi-
in the early 2000s following the work of Smith [87], neering applications, compared to those without. Other
who coined the term ‘inerter’ in the context of a force– areas for future development are:
current analogy between electrical and mechanical net- – There is considerably more scope in using non-
works. In particular work on ball-screw and rack-and- linear inerter models within the design of inerter-
pinion inerters developed strongly in this period, along based devices such as nonlinear energy sinks. A
with their use in the inerter-based devices such as, recent example studied by Chen et al. [13] is
the tuned-viscous-mass-damper, tuned-inerter–damper applied to the problem of eliminating unwanted
and tuned-mass-damper–inerter. resonances from a composite plate . This type of
Also important was the application of nonlinear multi-resonance analysis has significant potential
inerter-based isolators and absorbers. These included for future research.
different types of nonlinear energy sink inerters, non- – Seeking forms of device that can be physically
linear inerter isolators, and geometrically nonlinear implemented and that exploit the benefits of nonlin-
inerter devices, many relying on quasi-zero-stiffness ear dynamics is an area of great interest. For exam-
springs. In these devices, the nonlinear nature of the ple, this has been recently investigated for an energy

123
30 D. J. Wagg

harvesting application by Liu et al. [61], and for References

another application using nonlinear viscous damp-
ing by Huang et al. [40]. Building novel devices and 1. Alabuzhev, P.M.: Vibration Protection and Measuring Sys-
tems with Quasi-Zero Stiffness. CRC Press, Boca Raton
device configurations is a related area of interest. (1989)
– Earthquake engineering offers some of the most 2. Alujević, N., Čakmak, D., Wolf, H., Jokić, M.: Passive and
challenging problems in dynamics due to the active vibration isolation systems using inerter. J. Sound
extreme nature of the loadings involved. More gen- Vib. 418, 163–183 (2018)
3. Anderson, R.C., Smith, M.F.: A study of the kaman
erally nonstationary stochastic inputs offer a related
dynamic antiresonant vibration isolator. Technical report.
challenge, which has already been studied for the Kaman Aerospace Corp., Bloomfield CT (1966)
linear case [6,43,70,95]. Often the examples con- 4. Arakaki, T., Kuroda, H., Arima, F., Inoue, Y., Baba, K.:
sidered in earthquake engineering have to be quite Development of seismic devices applied to ball screw: part
1 basic performance test of rd-series. AIJ J. Technol. Des.
idealised when compared to the real application. As 5(8), 239–244 (1999). In Japanese
a result, developing methods for transient responses 5. Arakaki, T., Kuroda, H., Arima, F., Inoue, Y., Baba, K.:
such as earthquakes in the presence of nonlinear Development of seismic devices applied to ball screw: part
effects such as those discussed in this paper is an 2 performance test and evaluation of rd-series. AIJ J. Tech-
nol. Des. 5(9), 265–270 (1999)
area for future development—see for example Radu 6. Baduidana, M., Kenfack-Jiotsa, A.: Optimal design of
et al. [76] and Ji et al. [44]. inerter-based isolators minimizing the compliance and
– Lastly, although semi-active and active control mobility transfer function versus harmonic and ran-
applications were not considered as part of this dom ground acceleration excitation. J. Vib. Control
1077546320940175 (2020)
review, it should be mentioned that they offer 7. Bartlett, F.D., Flannelly, W.G.: Application of antireso-
some of the most interesting areas of development nance theory to helicopters. J. Am. Helicopter Soc. 19(1),
of future nonlinear inerter applications—see for 11–15 (1974)
example the following recent papers [111,114,128, 8. Biolek, D., Biolek, Z., Biolkova, V., Kolka, Z.: Nonlinear
inerter in the light of Chua’s table of higher-order electrical
129] and references therein. elements. In: 2016 IEEE Asia Pacific Conference on Cir-
cuits and Systems (APCCAS), pp. 617–620. IEEE (2016)
Acknowledgements The author would like to thank Hakan 9. Biolek, Z., Biolek, D., Biolková, V., Kolka, Z.: Taxicab
Dogan, Predaricka Deastra and Matthew Tipuric who gave com- geometry in table of higher-order elements. Nonlinear Dyn.
ments on a draft version of the manuscript. 98(1), 623–636 (2019)
10. Braun, D.: Development of antiresonance force isolators
Compliance with ethical standards for helicopter vibration reduction. J. Am. Helicopter Soc.
27(4), 37–44 (1982)
Conflict of interest The author declares that he has no conflict 11. Brock, J.E.: A note on the damped vibration absorber.
of interest. Trans. ASME J. Appl. Mech. 13(4), 284 (1946)
12. Brzeski, P., Perlikowski, P.: Effects of play and inerter non-
Informed consent No informed consent, or other ethical per- linearities on the performance of tuned mass damper. Non-
mission was required as part of this work. linear Dyn. 88(2), 1027–1041 (2017)
13. Chen, H.Y., Mao, X.Y., Ding, H., Chen, L.Q.: Elimina-
Open Access This article is licensed under a Creative Com- tion of multimode resonances of composite plate by inertial
mons Attribution 4.0 International License, which permits use, nonlinear energy sinks. Mech. Syst. Signal Process. 135,
sharing, adaptation, distribution and reproduction in any medium 106383 (2020)
or format, as long as you give appropriate credit to the original 14. Chen, M.Z.Q., Hu, Y., Li, C., Chen, G.: Application of
author(s) and the source, provide a link to the Creative Com- semi-active inerter in semi-active suspensions via force
mons licence, and indicate if changes were made. The images or tracking. J. Vib. Acoust. 138(4), 041014 (2016)
other third party material in this article are included in the article’s 15. De Domenico, D., Deastra, P., Ricciardi, G., Sims, N.D.,
Creative Commons licence, unless indicated otherwise in a credit Wagg, D.J.: Novel fluid inerter based tuned mass dampers
line to the material. If material is not included in the article’s Cre- for optimised structural control of base-isolated buildings.
ative Commons licence and your intended use is not permitted by J. Franklin Inst. 356, 7626–7649 (2019)
statutory regulation or exceeds the permitted use, you will need 16. De Domenico, D., Ricciardi, G.: An enhanced base iso-
to obtain permission directly from the copyright holder. To view lation system equipped with optimal tuned mass damper
a copy of this licence, visit http://creativecommons.org/licenses/ inerter (TMDI). Earthq. Eng. Struct. Dyn. 47(5), 1169–
by/4.0/. 1192 (2018)
17. De Domenico, D., Ricciardi, G.: Optimal design and seis-
mic performance of tuned mass damper inerter (TMDI) for

123
A review of the mechanical inerter 31

structures with nonlinear base isolation systems. Earthq. 37. Hu, Y., Chen, M.Z.Q., Shu, Z., Huang, L.: Analysis and
Eng. Struct. Dyn. 47(12), 2539–2560 (2018) optimisation for inerter-based isolators via fixed-point the-
18. De Domenico, D., Ricciardi, G., Zhang, R.: Optimal design ory and algebraic solution. J. Sound Vib. 346, 17–36 (2015)
and seismic performance of tuned fluid inerter applied 38. Hu, Y., Chen, M.Z.Q., Sun, Y.: Comfort-oriented vehicle
to structures with friction pendulum isolators. Soil Dy. suspension design with skyhook inerter configuration. J.
Earthq. Eng. 132, 106099 (2020) Sound Vib. 405, 34–47 (2017)
19. de Haro Moraes, F., Silveira, M., Goncalves, P.J.P.: On the 39. Hu, Y., Wang, J., Chen, M.Z., Li, Z., Sun, Y.: Load miti-
dynamics of a vibration isolator with geometrically non- gation for a barge-type floating offshore wind turbine via
linear inerter. Nonlinear Dyn. 93(3), 1325–1340 (2018) inerter-based passive structural control. Eng. Struct. 177,
20. Den Hartog, J.P.: Mechanical Vibrations. McGraw-Hill, 198–209 (2018)
New York (1934) 40. Huang, Z., Hua, X., Chen, Z., Niu, H.: Optimal design
21. Desjardins, R.A., Hooper, W.E.: Antiresonant rotor isola- of TVMD with linear and nonlinear viscous damping for
tion for vibration reduction. J. Am. Helicopter Soc. 25(3), SDOF systems subjected to harmonic excitation. Struct.
46–55 (1980) Control Health Monit. 26(10), e2413 (2019)
22. Ding, H., Chen, L.Q.: Designs, analysis, and applications 41. Ikago, K., Saito, K., Inoue, N.: Seismic control of single-
of nonlinear energy sinks. Nonlinear Dyn. 100(4), 3061– degree-of-freedom structure using tuned viscous mass
3107 (2020) damper. Earthq. Eng. Struct. Dyn. 41(3), 453–474 (2012)
23. Dylejko, P.G., MacGillivray, I.R.: On the concept of a trans- 42. Javidialesaadi, A., Wierschem, N.E.: An inerter-enhanced
mission absorber to suppress internal resonance. J. Sound nonlinear energy sink. Mech. Syst. Signal Process. 129,
Vib. 333(10), 2719–2734 (2014) 449–454 (2019)
24. Evangelou, S., Limebeer, D.J., Sharp, R.S., Smith, M.C.: 43. Javidialesaadi, A., Wierschem, N.E.: Response of a SDOF
Steering compensation for high-performance motorcycles. system with an inerter-based tuned mass damper subjected
In: 43rd IEEE Conference on Decision and Control, vol. 1, to non-stationary random excitation. In: Dynamics of Civil
pp. 749–754. IEEE (2004) Structures, vol. 2, pp. 201–203. Springer (2020)
25. Flannelly, W.G.: Dynamic antiresonant vibration isolator. 44. Ji, X., Cheng, Y., Hutt, C.M.: Seismic response of a tuned
US Patent 3,322,379 (1967) viscous mass damper (TVMD) coupled wall system. Eng.
26. Flower, W.C.: Understanding hydraulic mounts for Struct. 225, 111252 (2020)
improved vehicle noise, vibration and ride qualities. SAE 45. John, E.D.A., Wagg, D.J.: Design and testing of a fric-
Tech. Pap. 1, 850975 (1985) tionless mechanical inerter device using living-hinges. J.
27. Frahm, H.: Device for damping vibrations of bodies. US Franklin Inst. 356, 7650–7668 (2019)
Patent US 989958 (1909) 46. Jones, R.: An analytical and model test research study on
28. Gartner, B.J., Smith, M.C.: Force-controlling hydraulic the kaman dynamic antiresonant vibration isolator (DAVI).
device. US Patent 8,881,876 (2014) Technical report. Kaman Aerospace Corp., Bloomfield, CT
29. Giaralis, A., Taflanidis, A.A.: Optimal tuned mass-damper- (1968)
inerter (TMDI) design for seismically excited MDOF struc- 47. Kawamata, S.: Development of a vibration control system
tures with model uncertainties based on reliability criteria. of structures by means of mass pumps. Technical report.
Struct. Control Health Monit. 25(2), e2082 (2018) Institute of Industrial Science, University of Tokyo, Tokyo
30. Golnaraghi, M.F., Nakhaie Jazar, G.: Development and (1973)
analysis of a simplified nonlinear model of a hydraulic 48. Kawamata, S.: Control of structural vibration by inertia
engine mount. J. Vib. Control 7(4), 495–526 (2001) pump damper: part I theoretical model and response to
31. Gonzalez-Buelga, A., Lazar, I.F., Jiang, J.Z., Neild, S.A., harmonic excitation. In: Summaries of Technical Papers of
Inman, D.J.: Assessing the effect of nonlinearities on the Annual Meeting Architectural Institute of Japan (1986)
performance of a tuned inerter damper. Struct. Control 49. Kawamata, S.: Accelerated liquid mass damper and princi-
Health Monit. 24(3), e1879 (2017) ples of structural vibration control. In: Structural Mechan-
32. Gutierrez Soto, M., Adeli, H.: Tuned mass dampers. Arch. ics in Reactor Technology (1987)
Comput. Methods Eng. 20, 419–431 (2013) 50. Krenk, S., Høgsberg, J.: Tuned resonant mass or inerter-
33. Hashimoto, T., Fujita, K., Tsuji, M., Takewaki, I.: Innova- based absorbers: unified calibration with quasi-dynamic
tive base-isolated building with large mass-ratio TMD at flexibility and inertia correction. Proc. R. Soc. A
basement for greater earthquake resilience. Future Cities 472(2185), 20150718 (2016)
Environ. 1(1), 9 (2015) 51. Kuhnert, W.M., Gonçalves, P.J.P., Ledezma-Ramirez, D.F.,
34. Hessabi, R.M., Mercan, O.: Investigations of the applica- Brennan, M.J.: Inerter-like devices used for vibration iso-
tion of gyro-mass dampers with various types of supple- lation: a historical perspective. J. Franklin Inst. (2020)
mental dampers for vibration control of building structures. 52. Kuznetsov, A., Mammadov, M., Sultan, I., Hajilarov, E.:
Eng. Struct. 126, 174–186 (2016) Optimization of improved suspension system with inerter
35. Housner, G.W.: The behavior of inverted pendulum struc- device of the quarter-car model in vibration analysis. Arch.
tures during earthquakes. Bull. Seismol. Soc. Am. 53(2), Appl. Mech. 81(10), 1427–1437 (2011)
403–417 (1963) 53. Lazar, I.F., Neild, S.A., Wagg, D.J.: Using an inerter-based
36. Hu, Y., Chen, M.Z., Xu, S., Liu, Y.: Semiactive inerter and device for structural vibration suppression. Earthq. Eng.
its application in adaptive tuned vibration absorbers. IEEE Struct. Dyn. 43(8), 1129–1147 (2014). https://doi.org/10.
Trans. Control Syst. Technol. 25(1), 294–300 (2017) 1002/eqe.2390

123
32 D. J. Wagg

54. Lazar, I.F., Neild, S.A., Wagg, D.J.: Vibration suppression 73. Papageorgiou, C., Houghton, N.E., Smith, M.C.: Experi-
of cables using tuned inerter dampers. Eng. Struct. 122, mental testing and analysis of inerter devices. J. Dyn. Syst.
62–71 (2016) Meas. Contr. 131(1), 011001 (2009)
55. Lazarek, M., Brzeski, P., Perlikowski, P.: Design and iden- 74. Pietrosanti, D., De Angelis, M., Basili, M.: Optimal design
tification of parameters of tuned mass damper with inerter and performance evaluation of systems with tuned mass
which enables changes of inertance. Mech. Mach. Theory damper inerter (TMDI). Earthq. Eng. Struct. Dyn. 46(8),
119, 161–173 (2018) 1367–1388 (2017)
56. Lazarek, M., Brzeski, P., Perlikowski, P.: Design and mod- 75. Pietrosanti, D., De Angelis, M., Giaralis, A.: Experimen-
eling of the cvt for adjustable inerter. J. Franklin Inst. tal study and numerical modeling of nonlinear dynamic
356(14), 7611–7625 (2019) response of SDOF system equipped with tuned mass
57. Lewis, T.D., Jiang, J.Z., Neild, S.A., Gong, C., Iwnicki, damper inerter (TMDI) tested on shaking table under har-
S.D.: Using an inerter-based suspension to improve both monic excitation. Int. J. Mech. Sci. 105762 (2020)
passenger comfort and track wear in railway vehicles. Vehi- 76. Radu, A., Lazar, I.F., Neild, S.A.: Performance-based seis-
cle Syst. Dyn. 58(3), 472–493 (2020) mic design of tuned inerter dampers. Struct. Control Health
58. Li, L., Liang, Q., Qin, H.: Equivalent linearization methods Monit. 26(5), e2346 (2019)
for a control system with clutching inerter damper. Appl. 77. Rivin, E.I.: Passive Vibration Isolation. ASME Press, New
Sci. 9(4), 688 (2019) York (2003)
59. Li, P., Lam, J., Cheung, K.C.: Control of vehicle suspension 78. Roberts, J.B., Spanos, P.D.: Random Vibration and Statis-
using an adaptive inerter. Proc. Inst. Mech. Eng. Part D J. tical Linearization. Courier Corporation, New York (2003)
Automob. Eng. 229(14), 1934–1943 (2015) 79. Saitoh, M.: On the performance of gyro-mass devices for
60. Liu, K., Liu, J.: The damped dynamic vibration absorbers: displacement mitigation in base isolation systems. Struct.
revisited and new result. J. Sound Vib. 284(3–5), 1181– Control Health Monit. 19(2), 246–259 (2012)
1189 (2005) 80. Shaw, A.D., Neild, S.A., Wagg, D.J.: Dynamic analysis
61. Liu, M., Tai, W.C., Zuo, L.: Enhancing the performance of of high static low dynamic stiffness vibration isolation
backpack energy harvester using nonlinear inerter-based mounts. J. Sound Vib. 332, 1437–1455 (2013)
two degrees of freedom design. Smart Mater. Struct. 29(2), 81. Shaw, A.D., Neild, S.A., Wagg, D.J., Weaver, P.M., Car-
025007 (2020) rella, A.: A nonlinear spring mechanism incorporating a
62. Liu, N., Li, C., Yin, C., Dong, X., Hua, H.: Application bistable composite plate for vibration isolation. J. Sound
of a dynamic antiresonant vibration isolator to minimize Vib. 332(24), 6265–6275 (2013)
the vibration transmission in underwater vehicles. J. Vib. 82. Shen, Y., Chen, L., Liu, Y., Zhang, X.: Modeling and opti-
Control 1077546317711538 (2017) mization of vehicle suspension employing a nonlinear fluid
63. Liu, X., Jiang, J.Z., Titurus, B., Harrison, A.: Model identi- inerter. Shock Vib. 2016 (2016)
fication methodology for fluid-based inerters. Mech. Syst. 83. Shen, Y., Chen, L., Liu, Y., Zhang, X.: Influence of fluid
Signal Process. 106, 479–494 (2018) inerter nonlinearities on vehicle suspension performance.
64. Liu, X., Jiang, J.Z., Titurus, B., Harrison, A.J.L., McBryde, Adv. Mech. Eng. 9(11), 1687814017737257 (2017)
D.: Testing and modelling of the damping effects for fluid- 84. Siami, A., Cigada, A., Karimi, H.R., Zappa, E.: Vibration
based inerters. Proc. Eng. 199, 435–440 (2017) protection of a famous statue against ambient and earth-
65. Luo, H., Ikago, K., Chong, C., Keivan, A., Phillips, B.M.: quake excitation using a tuned inerter-damper. Machines
Performance of low-frequency structures incorporated with 5(4), 33 (2017)
rate-independent linear damping. Eng. Struct. 181, 324– 85. Siami, A., Karimi, H.R., Cigada, A., Zappa, E., Sabbioni,
335 (2019) E.: Parameter optimization of an inerter-based isolator
66. Madhamshetty, K., Manimala, J.M.: Low-rate characteri- for passive vibration control of Michelangelo’s Rondanini
zation of a mechanical inerter. Machines 6(3), 32 (2018) pietà. Mech. Syst. Signal Process. 98, 667–683 (2018)
67. Makris, N., Kampas, G.: Seismic protection of struc- 86. Singh, R.: Dynamic design of automotive systems: Engine
tures with supplemental rotational inertia. J. Eng. Mech. mounts and structural joints. Sadhana 25(3), 319–330
142(11), 04016089 (2016) (2000)
68. Málaga-Chuquitaype, C., Menendez-Vicente, C., Thiers- 87. Smith, M.C.: Synthesis of mechanical networks: the inerter.
Moggia, R.: Experimental and numerical assessment of the IEEE Trans. Autom. Control 47(10), 1648–1662 (2002)
seismic response of steel structures with clutched inerters. 88. Smith, M.C.: The inerter: a retrospective. Ann. Rev. Control
Soil Dyn. Earthq. Eng. 121, 200–211 (2019) Robot. Autonomous Syst. 3, 361–391 (2020)
69. Marian, L., Giaralis, A.: Optimal design of a novel tuned 89. Smith, M.C., Wang, F.C.: Performance benefits in passive
mass-damper-inerter (TMDI) passive vibration control vehicle suspensions employing inerters. Veh. Syst. Dyn.
configuration for stochastically support-excited structural 42(4), 235–257 (2004)
systems. Probab. Eng. Mech. 38, 156–164 (2014) 90. Smith, N., Wagg, D.J.: A fluid inerter with variable iner-
70. Masri, S.F., Caffrey, J.P.: Transient response of a SDOF tance properties. In: Proceedings of the 6th European Con-
system with an inerter to nonstationary stochastic excita- ference on Structural Control, pp. 1–8 (2016). https://doi.
tion. J. Appl. Mech. 84(4) (2017) org/10.15131/shef.data.4206096.v1
71. O’connor, B.E.: Inertia controlled flutter damper. US Patent 91. Soltani, P., Pinna, C., Wagg, D.J., Whear, R.: Ageing sim-
2,742,113 (1956) ulation of a hydraulic engine mount: a data-informed finite
72. Okumura, A.: The gyro-mass inerter Japan patent koukai. element approach. Proc. Inst. Mech. Eng. Part D J. Auto-
h09-177875 (1997) mob. Eng. 233(10), 2432–2442 (2018)

123
A review of the mechanical inerter 33

92. Sugimura, Y., Goto, W., Tanizawa, H., Saito, K., Proc. Inst. Mech. Eng. Part C J. Mech. Eng. Sci. 228(15),
Nimomiya, T.: Response control effect of steel building 2684–2689 (2014)
structure using tuned viscous mass damper. In: Proceedings 110. Wang, X., He, T., Shen, Y., Shan, Y., Liu, X.: Parame-
of the 15th World Conference on Earthquake Engineering, ters optimization and performance evaluation for the novel
pp. 24–28 (2012) inerter-based dynamic vibration absorbers with negative
93. Sun, X.Q., Chen, L., Wang, S.H., Zhang, X.L., Yang, X.F.: stiffness. J. Sound Vib. 463, 114941 (2019)
Performance investigation of vehicle suspension system 111. Wang, Y., Ding, H., Chen, L.Q.: Averaging analysis
with nonlinear ball-screw inerter. Int. J. Automotive Tech- on a semi-active inerter-based suspension system with
nol. 17(3), 399–408 (2016) relative-acceleration-relative-velocity control. J. Vib. Con-
94. Swift, S.J., Smith, M.C., Glover, A.R., Papageorgiou, C., trol 26(13–14), 1199–1215 (2020)
Gartner, B., Houghton, N.E.: Design and modelling of a 112. Wang, Y., Li, H.X., Cheng, C., Ding, H., Chen, L.Q.:
fluid inerter. Int. J. Control 86(11), 2035–2051 (2013) Dynamic performance analysis of a mixed-connected
95. Taflanidis, A.A., Giaralis, A., Patsialis, D.: Multi-objective inerter-based quasi-zero stiffness vibration isolator. Struct.
optimal design of inerter-based vibration absorbers for Control Health Monit. 27(10), e2604 (2020)
earthquake protection of multi-storey building structures. 113. Wang, Y., Li, H.X., Cheng, C., Ding, H., Chen, L.Q.: A
J. Franklin Inst. 356(14), 7754–7784 (2019) nonlinear stiffness and nonlinear inertial vibration isolator.
96. Takewaki, I., Murakami, S., Yoshitomi, S., Tsuji, M.: Fun- J. Vib. Control 1077546320940924 (2020)
damental mechanism of earthquake response reduction in 114. Wang, Y., Meng, H., Zhang, B., Wang, R.: Analyti-
building structures with inertial dampers. Struct. Control cal research on the dynamic performance of semi-active
Health Monit. 19(6), 590–608 (2012) inerter-based vibration isolator with acceleration-velocity-
97. Thiers-Moggia, R., Málaga-Chuquitaype, C.: Seismic pro- based control strategy. Struct. Control Health Monit. 26(4),
tection of rocking structures with inerters. Earthq. Eng. e2336 (2019)
Struct. Dyn. 48(5), 528–547 (2019) 115. Wang, Y., Wang, R., Meng, H., Zhang, B.: An investigation
98. Thiers-Moggia, R., Málaga-Chuquitaype, C.: Dynamic of the dynamic performance of lateral inerter-based vibra-
response of post-tensioned rocking structures with inert- tion isolator with geometrical nonlinearity. Arch. Appl.
ers. Int. J. Mech. Sci. 187, 105927 (2020) Mech. 89(9), 1953–1972 (2019)
99. Tipuric, M., Wagg, D., Sims, N.: Magnetorheological 116. White Jr., L.: Theophilus redivivus. Technol. Culture 5(2),
bypass valve design for a semi-active inerter. In: Active 224–233 (1964)
and Passive Smart Structures and Integrated Systems XII, 117. Xiang, S.H.I., Songye, Z.H.U.: A comparative study of
vol. 10967, p. 109671L. International Society for Optics vibration isolation performance using negative stiffness
and Photonics (2019) and inerter dampers. J. Franklin Inst. 356(14), 7922–7946
100. Titurus, B.: Generalized liquid-based damping device for (2019)
passive vibration control. AIAA J. 56(10), 4134–4145 118. Yang, J., Jiang, J.Z., Neild, S.A.: Dynamic analysis and
(2018) performance evaluation of nonlinear inerter-based vibra-
101. Wagg, D., Pei, J.S.: Modeling a helical fluid inerter system tion isolators. Nonlinear Dyn. 99(3), 1823–1839 (2020)
with time-invariant mem-models. Struct. Control Health 119. Yilmaz, C., Kikuchi, N.: Analysis and design of passive
Monit. (2020) band-stop filter-type vibration isolators for low-frequency
102. Wang, F.C., Chan, H.A.: Vehicle suspensions with a mecha- applications. J. Sound Vib. 291(3–5), 1004–1028 (2006)
tronic network strut. Veh. Syst. Dyn. 49(5), 811–830 (2011) 120. Zeng, Y.C., Ding, H., Du, R.H., Chen, L.Q.: A suspension
103. Wang, F.C., Hong, M.F., Chen, C.W.: Building suspensions system with quasi-zero stiffness characteristics and inerter
with inerters. Proc. Inst. Mech. Eng. Part C J. Mech. Eng. nonlinear energy sink. J Vib. Control 1077546320972904
Sci. 224(8), 1605–1616 (2010) (2020)
104. Wang, F.C., Hsieh, M.R., Chen, H.J.: Stability and perfor- 121. Zhang, R., Zhao, Z., Pan, C., Ikago, K., Xue, S.: Damping
mance analysis of a full-train system with inerters. Veh. enhancement principle of inerter system. Struct. Control
Syst. Dyn. 50(4), 545–571 (2012) Health Monit. 27(5), e2523 (2020)
105. Wang, F.C., Liao, M.K.: The lateral stability of train sus- 122. Zhang, X.J., Ahmadian, M., Guo, K.H.: On the benefits of
pension systems employing inerters. Veh. Syst. Dyn. 48(5), semi-active suspensions with inerters. Shock Vib. 19(3),
619–643 (2010) 257–272 (2012)
106. Wang, F.C., Liao, M.K., Liao, B.H., Su, W.J., Chan, H.A.: 123. Zhang, X.l., Gao, Q., Nie, J., : The mem-inerter: a new
The performance improvements of train suspension sys- mechanical element with memory. Adv. Mech. Eng. 10(6),
tems with mechanical networks employing inerters. Veh. 1687814018778428 (2018)
Syst. Dyn. 47(7), 805–830 (2009) 124. Zhang, X.L., Geng, C., Nie, J.M., Gao, Q.: The missing
107. Wang, F.C., Su, W.J.: Impact of inerter nonlinearities on mem-inerter and extended mem-dashpot found. Nonlinear
vehicle suspension control. Veh. Syst. Dyn. 46(7), 575– Dyn. 101(2), 835–856 (2020)
595 (2008) 125. Zhang, Y.W., Lu, Y.N., Zhang, W., Teng, Y.Y., Yang, H.X.,
108. Wang, M., Sun, F.: Displacement reduction effect and sim- Yang, T.Z., Chen, L.Q.: Nonlinear energy sink with inerter.
plified evaluation method for SDOF systems using a clutch- Mech. Syst. Signal Process. 125, 52–64 (2019)
ing inerter damper. Earthq. Eng. Struct. Dyn. 47(7), 1651– 126. Zhang, Z., Lu, Z.Q., Ding, H., Chen, L.Q.: An inertial non-
1672 (2018) linear energy sink. J. Sound Vib. 450, 199–213 (2019)
109. Wang, R., Meng, X., Shi, D., Zhang, X., Chen, Y., Chen, L.:
Design and test of vehicle suspension system with inerters.

123
34 D. J. Wagg

127. Zhang, Z., Zhang, Y.W., Ding, H.: Vibration control com- 130. Zhao, Z., Zhang, R., Jiang, Y., Pan, C.: Seismic response
bining nonlinear isolation and nonlinear absorption. Non- mitigation of structures with a friction pendulum inerter
linear Dyn. 1–19 (2020) system. Eng. Struct. 193, 110–120 (2019)
128. Zhao, G., Raze, G., Paknejad, A., Deraemaeker, A., Ker-
schen, G., Collette, C.: Active nonlinear energy sink using
force feedback under transient regime. Nonlinear Dyn. 1– Publisher’s Note Springer Nature remains neutral with regard
18 (2020) to jurisdictional claims in published maps and institutional affil-
129. Zhao, G., Raze, G., Paknejad, A., Deraemaeker, A., Ker- iations.
schen, G., Collette, C.: Active nonlinear inerter damper for
vibration mitigation of duffing oscillators. J. Sound Vib.
473, 115236 (2020)

123