materials

Article
Research on Low-Frequency Noise Control of Automobiles
Based on Acoustic Metamaterial
Yi Liao 1,† , Haibo Huang 2,† , Guangbao Chang 1 , Deyang Luo 1 , Chuanlai Xu 3,4, *, Yudong Wu 5, * and Jiyou Tang 2

1 SAIC GM WULING Automobile Co., Ltd., Liuzhou 545005, China; yi.liao1@sgmw.com.cn (Y.L.);
guangbao.chang1@sgmw.com.cn (G.C.); deyang.luo@sgmw.com.cn (D.L.)
2 School of Mechanical Engineering, Southwest Jiaotong University, Chengdu 610031, China;
huanghaibo214@my.swjtu.edu.cn (H.H.); jyt@my.swjtu.edu.cn (J.T.)
3 Sichuan Jiuzhou Electric Group Co., Ltd., No. 6 Jiuhua Road, Mianyang 621000, China
4 Sichuan Avionics System Product Lightweight Design and Manufacturing Engineering Laboratory,
Mianyang 621000, China
5 National Laboratory of Rail Transit (in Preparation), Chengdu 610031, China
* Correspondence: xuchuanlai@163.com (C.X.); wu045043@my.swjtu.edu.cn (Y.W.)
† These authors contributed equally to this work.

Abstract: With the transformation of the trend of vehicle electrification, the overall noise level in
the vehicle is gradually reduced. The problem of low-frequency noise in the vehicle, which was
previously ignored, is becoming more and more prominent. To solve the vehicle low-frequency
noise problem, a combination of real-vehicle tests and simulation analysis is carried out. During
the test, the driver and passengers feel that there is a relatively obvious low-frequency roar noise in
the car, which results from the structural radiation noise of the trunk door vibration. Therefore, to
solve this problem, we design an acoustic metamaterial with lightweight and miniaturized features



based on the local resonance principle of phononic crystals. Firstly, the selection of the resonant unit


configuration and the design of the band gap are implemented. Then, the layout planning of the
Citation: Liao, Y.; Huang, H.; Chang, whole vehicle, the layout of the resonance unit and the design of the base frame are implemented.
G.; Luo, D.; Xu, C.; Wu, Y.; Tang, J.
The actual vehicle test results show that: after attaching the designed acoustic metamaterial, the
Research on Low-Frequency Noise
low-frequency noise sound pressure levels in the front and rear of the vehicle were reduced by
Control of Automobiles Based on
2.0 dB (A) and 2.3 dB (A), respectively, meanwhile, the interior noise sound quality was improved.
Acoustic Metamaterial. Materials
2022, 15, 3261. https://doi.org/
The sound pressure level at the driver’s right ear in the car has an abnormal peak of around 35Hz. In
10.3390/ma15093261 addition, the driver and passengers feel that there is a relatively obvious low-frequency roar noise in
the car, and through low-pass filtering of the collected signals, it is confirmed that the peak frequency
Academic Editor: Theodore
is the main cause of the low-frequency roar in the car. The low-frequency steady-state noise of the car
E. Matikas
is generally considered to be the low-frequency vibration of the body panel and the radiation occurs.
Received: 8 March 2022 Through the finite element simulation analysis (Grid Participation Analysis) of the abnormal peak
Accepted: 14 April 2022 frequency, the results show that the low-frequency roar is caused by the low-frequency vibration of
Published: 1 May 2022 the tailgate sheet metal, and the problem peak frequency is not coupled with the acoustic cavity mode.
Publisher’s Note: MDPI stays neutral Facing the problem of the low-frequency roar radiated into the car by the vibration of the tailgate
with regard to jurisdictional claims in sheet metal parts, based on the local resonance band gap theory, we developed a design to suppress
published maps and institutional affil- the 35 Hz vibration of the tailgate sheet metal parts and meet the characteristics of lightweight
iations. and miniaturization. By attaching the acoustic metamaterial to the tailgate and performing CAE
simulation of the whole vehicle, it is determined that the structure can indeed reduce the 35 Hz noise
in the car and the peak value of the tailgate sheet metal vibration.

Copyright: © 2022 by the authors.

Keywords: acoustic metamaterial; local resonance; low-frequency; roar noise control; steady-state noise
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
1. Introduction
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/ In the field of automotive NVH (Noise, Vibration, Harshness), high-frequency noise
4.0/). and vibration control can achieve ideal results through acoustic packaging or attaching

Materials 2022, 15, 3261. https://doi.org/10.3390/ma15093261 https://www.mdpi.com/journal/materials

 1. Introduction
In the field of automotive NVH (Noise, Vibration, Harshness), high-frequency noise
Materials 2022, 15, 3261 2 of 18
and vibration control can achieve ideal results through acoustic packaging or attaching
vibration damping, but controlling the radiation caused by the vibration of automotive
sheet metal parts low-frequency noise has always been a major and difficult problem in
vibration
the field ofdamping,
automotivebutnoise
controlling the radiation
and vibration caused
control [1–3].by the
For vibration ofthe
automobiles, automotive
low-fre-
sheet metal parts low-frequency noise has always been a major and difficult problem
quency noise and vibration problems radiated into the car caused by the vibration of sheet in the
field of automotive noise and vibration control [1–3]. For automobiles, the low-frequency
metal parts will not only seriously affect the ride comfort of the driver and passengers,
noise and vibration problems radiated into the car caused by the vibration of sheet metal
but also have a great impact on the driving safety and service life of the car [4]. In order
parts will not only seriously affect the ride comfort of the driver and passengers, but also
to obtain a more comfortable and safe driving and riding environment, and to improve
have a great impact on the driving safety and service life of the car [4]. In order to obtain a
the market competitiveness of vehicles, it is extremely important to control the low-fre-
more comfortable and safe driving and riding environment, and to improve the market
quency noise radiated into the vehicle by the vibration of the automobile sheet metal parts
competitiveness of vehicles, it is extremely important to control the low-frequency noise
[5]. The generation path of low-frequency noise radiated by automobile (tailgate) sheet
radiated into the vehicle by the vibration of the automobile sheet metal parts [5]. The
metal parts is shown in Figure 1. The car is driving on rough roads, and the displacement
generation path of low-frequency noise radiated by automobile (tailgate) sheet metal parts
excitation of the road is transmitted to the body sheet metal parts through the tire–sus-
is shown in Figure 1. The car is driving on rough roads, and the displacement excitation of
pension system [6]. Excitation generates low-frequency vibration and radiates low-fre-
the road is transmitted to the body sheet metal parts through the tire–suspension system [6].
quency structural noise into the car, causing serious discomfort to passengers and drivers
Excitation generates low-frequency vibration and radiates low-frequency structural noise
[7,8].
into the car, causing serious discomfort to passengers and drivers [7,8].

Figure 1.
Figure Noiseradiation
1. Noise radiationpath
pathof
ofrough
rough road
road structure
structure (tailgate).
(tailgate).

At present,
At present, there
there are
are three
three main
mainways
waystotoreduce
reducethe thevibration
vibrationofofsheet
sheetmetal
metal parts in in
parts re-
sponse to the problem of low-frequency noise radiated by the vibration
response to the problem of low-frequency noise radiated by the vibration of automobileof automobile sheet
metalmetal
sheet partsparts
at home and and
at home abroad [9,10].
abroad First,
[9,10]. using
First, spring-damping
using spring-damping vibration-reducing
vibration-reduc-
materials
ing materialsor rubber vibration-reducing
or rubber vibration-reducingmaterials. WithWith
materials. the development
the developmentof current indus-
of current
trial technology, it is found that the use of spring-damping vibration-reducing materials or
industrial technology, it is found that the use of spring-damping vibration-reducing ma-
rubber damping vibration-reducing materials has a more obvious effect on high-frequency
terials or rubber damping vibration-reducing materials has a more obvious effect on high-
vibration, but it is difficult to achieve an ideal suppression effect of the middle and low-
frequency vibration, but it is difficult to achieve an ideal suppression effect of the middle
frequency noise [11]. Second, applying reinforcing ribs to the sheet metal parts to increase
and low-frequency noise [11]. Second, applying reinforcing ribs to the sheet metal parts
the natural frequency of the low-frequency band of the sheet metal structure by increasing
to increase the natural frequency of the low-frequency band of the sheet metal structure
the rigidity, but the structure of the automobile sheet metal parts is greatly changed which
by increasing the rigidity, but the structure of the automobile sheet metal parts is greatly
will cause the production cost of the model to increase sharply, and seriously affect the
changed which will cause the production cost of the model to increase sharply, and seri-
economic benefits of the enterprise [12]. Third, installing the dynamic vibration absorber,
ously affect the economic benefits of the enterprise [12]. Third, installing the dynamic vi-
but its cost is higher and the mass is higher [13]. In view of the above problems, the local
bration absorber, but its cost is higher and the mass is higher [13]. In view of the above
resonance type vibration-damping acoustic metamaterial with negative dynamic equivalent
problems,
parameter the local resonance
characteristics typeavibration-damping
provides acousticofmetamaterial
new idea for noise control low-frequency with nega-
vibration
tive dynamic equivalent parameter characteristics
radiation of automobile sheet metal parts [14–19]. provides a new idea for noise control
of low-frequency vibration
In recent years, studiesradiation of automobile
on the locally sheet metal
resonant phonon parts
crystal [14–19].
formed by the periodic
In recent years, studies on the locally resonant phonon crystal
arrangement of elastomers and mass scatterers have been studied with great formed by thevigor
periodic
[20].
arrangement of elastomers and mass scatterers have been studied with great
Since the band gap produced by it is not limited by the size of the unit cell, it has a broad vigor [20].
Since the band
application gap for
market produced by it is not
low-frequency limited
vibration bynoise
and the size of the unit
reduction cell, it has aIn
in automobiles. broad
2000,
application market for low-frequency vibration and noise reduction
when Liu Zhengyou et al. studied the three-dimensional phononic crystal formed by thein automobiles. In
cubic lattice structure of lead balls coated with viscoelastic silicone rubber soft material and
embedded in epoxy resin, they found that the wavelength of the phononic crystal forbidden
band is much larger [21]. Due to the size of the lattice, it breaks through the limitation of
the Bragg scattering mechanism and can control elastic waves two orders of magnitude
lower than the resonant unit. Moreover, the scatterers are not strictly periodic or even
Materials 2022, 15, 3261 3 of 18

randomly distributed, the composite structure also has a band gap effect [22]. This puts
forward the local resonance mechanism of the elastic wave band gap, which provides a new
theoretical basis for the follow-up automotive low-frequency noise and vibration control
field [23]. In 2001, BAZ introduced active control to study the transmission characteristics
of elastic waves in periodic beams with two materials alternately arranged, gave the corre-
sponding simplified theoretical analysis model of the spring–mass system, explained the
active adjustment method of passband and forbidden band, and proposed the directional
design method of local resonance type acoustic metamaterial [24]. In 2008, Hong Kong
University of Science and Technology Yang Zhiyu and others proposed a thin-film acous-
tic metamaterial based on the local resonance theory [25–27]. The lightweight structure
with a thickness as low as millimeters successfully achieved outstanding sound insulation
capabilities in the low-frequency range of 500~800 Hz, which provided a new solution
for low-frequency noise reduction. The elastic film needs external tension to propagate
the vibration [28]. As the stiffness of the elastic film is very low and the performance is
unstable, a small change in tension will cause a frequency shift of tens or even hundreds of
hertz. Therefore, no matter from a theoretical or technical point of view, effective control of
film tension is very necessary. In 2015, Ma Fuyin and others at Xi’an Jiaotong University
proposed a square lattice thin-film acoustic sound insulation metamaterial that requires
almost no tension by increasing the thickness of the film [29]. In 2019, Niu Jiamin and others
from Xi’an Jiaotong University proposed that under the asymmetric mode, the overall
sound absorption performance of the structure has been greatly improved, and the sound
absorption bandwidth has been expanded from the original narrow frequency to a wide
frequency [30].
In summary, although local resonance acoustic metamaterials seem to be relatively
new products now, there are a lot of data on the research of its mechanism, and there
are also certain physical samples to verify its vibration suppression effect. However, the
current acoustic metamaterials developed by technical institutions have poor stability and
the vibration suppression frequency is difficult to reduce to less than 100 Hz. The acoustic
metamaterial research is almost limited to simulation design and theoretical research.
Focusing on the problem of the low-frequency roar radiated into the car by the vi-
bration of the tailgate sheet metal of a certain vehicle model, based on the theory of local
resonance bandgap, a design is designed to suppress the 35 Hz vibration of the tailgate
sheet metal with stable and small equivalent stiffness. Through the vehicle finite element
simulation, it is verified that the structure can indeed reduce the 35 Hz noise peak in the
vehicle and reduce the vibration of the tailgate sheet metal parts, thereby providing a new
idea and new solution for the field of vehicle low-frequency noise and vibration control.

2. Basic Theory of Local Resonance Acoustic Metamaterial

2.1. Elastic Wave Equation in Acoustic Metamaterial
An acoustic metamaterial is a kind of man-made periodic composite structure material,
which is developed from crystals in solid mechanics [31,32]. Therefore, the theory of
the energy band of crystals in solid mechanics is also applicable to periodic composite
structures. The content of acoustic metamaterial exploration is the propagation of elastic
waves in the medium, and the band gap is due to the special effects of elastic waves when
they are transmitted in the structure [33]. According to the knowledge of elastic dynamics
theory, under the condition of ensuring an ideal elastic medium, small displacement
and no initial stress, consider any particle in the homogeneous acoustic metamaterial
medium, and use the displacement method used in solving dynamic problems to accurately
establish three types of equations describing the relationship between the mass point force,
displacement, stress and strain in the acoustic metamaterial are as follows:
Differential Equation of Motion:

∂σx ∂τyx ∂τzx ∂2 u

+ + + ρx = ρ 2 (1)
∂x ∂y ∂z αt
Materials 2022, 15, 3261 4 of 18

∂τxy ∂σy ∂τzy ∂2 v

+ + + ρy = ρ 2 (2)
∂x ∂y ∂z αt
∂τxz ∂τyz ∂σz ∂2 w
+ + + ρz = ρ 2 (3)
∂x ∂y ∂z αt
Geometric Equation:
∂u ∂w ∂v
εx = γyz = + (4)
∂x ∂y ∂z
∂v ∂u ∂w
εy = γzx = + (5)
∂y ∂z ∂x
∂w ∂v ∂u
εz = γxy = + (6)
∂x ∂x ∂y
Physical Equation:
σx = λθt + 2µε x τyz = µγy (7)
σy = λθt + 2µε y τzx = µγz (8)
σz = λθt + 2µε z τxy = µγx (9)
The above equation involves the six components of the stress tensor (σx , σy , σz , τyz ,
τzx , τxy ), six components of strain tensor (γxy , γxz , γyz , ε x , ε y , ε z ) and three displacement
components (u, v, w), a total of 15 unknowns of spatial coordinates x, y, z and time t. σx is
the normal stress, τxy is the shear stress, ρ is mass point density, γx is shear strain, ε x is the
normal strain, λ, µ are Lame coefficients, θt = ε x + ε y + ε z .
Not every equation contains the unknowns in the above equation, by using a displace-
ment method often used to solve the equation of elastic dynamics, considering some of
these unknown function hypotheses into the mathematic expression of the “basic unknown
functions”, it can be brought into the rest of the equation, and can obtain the expressed in
the displacement of stress components, the final three displacements as unknown functions
derived the elastic dynamics equation:
→ → →
 
(λ + 2µ)∇ ∇ ∗ µ − µ∇ ∗ ∇ u + ρw2 u = 0 (10)

→
In the equation, u is the displacement vector; λ, µ are the Lame coefficient; ρ is the
density of the medium; ∇ is the Laplace operator; the relation between
p them and the wave
velocity is expressed
p as: Longitudinal wave velocity: C l = ( λ + 2µ )/ρ, Shear wave
velocity: Ct = µ/ρ.

2.2. Acoustic Metamaterial Band Gap Mechanism

When the vibrational elastic wave propagates in the local resonant phonon crystal,
it will be affected by the periodic elastic scatterers, and the elastic wave within a certain
intrinsic frequency range cannot continue to propagate through the elastic scatterers, which
is called the forbidden band of the acoustic metamaterial. Other eigenfrequency ranges
that can pass through these elastic scatterers without obstruction are called passbands.
As shown in Figure 2a, it is a typical phonon crystal structure with local resonance. The
structure is composed of mass scatterers, an elastic coating layer and matrix. The structure
can be simplified to a typical spring–mass system, as shown in Figure 2b. The mass
scatterers are composed of a dense metal material, which provides the mass component
(expressed as m) for the cell structure. The elastic coating layer is composed of a thin film
with certain elasticity, which provides the elastic component of the crystal cell structure
(expressed as k). The matrix is rigid material (expressed as M), which is the carrier of
vibration elastic wave transmission [34–37].
 The structure is composed of mass scatterers, an elastic coating layer and matrix. The
structure can be simplified to a typical spring–mass system, as shown in Figure 2b. The
mass scatterers are composed of a dense metal material, which provides the mass compo-
nent (expressed as m) for the cell structure. The elastic coating layer is composed of a thin
film with certain elasticity, which provides the elastic component of the crystal cell struc-
Materials 2022, 15, 3261 5 of 18
ture (expressed as k). The matrix is rigid material (expressed as M), which is the carrier of
vibration elastic wave transmission [34–37].

(a) (b)
FigureFigure
2. Typical local resonant
2. Typical acoustic
local resonant metamaterial
acoustic crystal cell
metamaterial structure
crystal and simplified
cell structure model. model.
and simplified
(a) Typical acoustic
(a) Typical metamaterial
acoustic metamaterial (b) Simplified
model,model, acoustic
(b) Simplified metamaterial
acoustic modelmodel.
metamaterial .

Suppose
Suppose the displacement
the displacement of matrix M andMmass
of matrix and component
mass component is Xx,and
m is Xmand the x, the matrix
matrix
excitation is F, and the reaction force of the mass component is F
excitation is F, and the reaction force of the mass component is 𝐹1 . By1 Newton’s second law
. By Newton’s second
and Hooke’s
law and Hooke’s law: law:
F − F1 = ( jw)2 MX (11)
𝐹 − 𝐹1 = (𝑗𝑤)2 𝑀𝑋 (11)
F1 = ( jw)2 mx (12)
2
𝐹1 =k((𝑗𝑤)
X − 𝑚𝑥
x ) = F1 (12) (13)

Consider the base body and spring–mass

𝑘(𝑋 − 𝑥) = 𝐹system as a whole: (13)
1

Consider the base body and spring–mass systemmw as 0a2 whole:

me = M + 2 2
(14)
𝑚𝑤w002 − w
𝑚𝑒 = 𝑀 + 2 2
(14)
𝑤X0 −𝑤 1
H (w) = =− (15)
F m e w2
𝑋 1
In the equation: me is the 𝐻(𝑤)
system= equivalent
𝐹
=−
𝑚𝑒 𝑤mass,
2 H (w) is the system (15)
√ displacement
frequency response function, ω is the natural circular frequency, w0 = /m is the natural
k
In of the𝑚internal
the equation:
frequency 𝑒 is the spring equivalent mass, 𝐻(𝑤) is the system displacement
systemoscillator.
It can be seen from the above formula that: with the change of external excitation
frequency response function, ω is the natural circular frequency, 𝑤0 = √𝑘⁄𝑚 is the natu-
frequency, the system shows different characteristics due to its different
Materials 2022, 15, x FOR PEER REVIEW 6 of 21 equivalent mass,
ral frequency
as shown ofin
the internal
Figure 3. spring oscillator.
It can be seen from the above formula that: with the change of external excitation
frequency, the system shows different characteristics due to its different equivalent mass,
as shown in Figure 3.

Figure 3. Relationship between external excitation frequency and displacement frequency response
Figure 3. Relationship between external excitation frequency and displacement frequency
function.
response function.
According to Newton’s second law and Hooke’s law theory and the external excita-
tion frequency and displacement frequency response function, the relationship between
the dynamic equivalent quality of the system and the excitation frequency can be inferred,
as shown in Table 1.

Table 1. The relationship between system dynamic equivalent mass 𝑚𝑒 and excitation frequency
w.
 Materials 2022, 15, 3261 6 of 18

According to Newton’s second law and Hooke’s law theory and the external excitation
frequency and displacement frequency response function, the relationship between the
dynamic equivalent quality of the system and the excitation frequency can be inferred, as
shown in Table 1.

Table 1. The relationship between system dynamic equivalent mass me and excitation frequency w.

Excitation
Dynamic Equivalent Mass System Movement Status
Frequency Range
w=0 me = M + m M and m move synchronously
0 < w < w0 me > 0 M and m move in the same direction
M and m resonate, and the system
w → w0 me → ∞
vibration tends to be static
√ The equivalent mass is negative, M
w0 < w < w0 ( M + m)/M me < 0
and m move in opposite directions
√ Equivalent mass tends to 0,
w → w0 ( M + m)/M me → 0
√ system resonance
w > w0 ( M + m)/M me > 0 M and m move in the same direction
w→∞ me → M m tends to stand still

In conclusion, the band gap is caused by the resonance of the internal mass component.
The band gap range is determined by the natural frequency of the whole structure and
internal mass component. The band gap effect occurs when the system  is in the
√ negative
equivalent mass phase, the excitation frequency is in the range of w0 , w0 ( M + m)/M .
Therefore, the frequency range generated by bandgap can be adjusted directionally by
Materials 2022, 15, x FOR PEER REVIEW
changing the mass or spring of the equivalent simplified model of the cell unit. Based on
7 of 21
the above theory, according to the vibration frequency of auto sheet metal parts (equiv-
alent to excitation frequency), an acoustic metamaterial can be designed to suppress the
specific frequency.
metal parts (equivalent to excitation frequency), an acoustic metamaterial can be designed
to suppress the specific frequency.
3. Analysis of Low-Frequency Roaring in Car
Aiming
3. Analysis at the relatively
of Low-Frequency obvious
Roaring low-frequency roar (20–100 Hz) generated by a certain
in Car
hatch
Aimingcar at when it is running
the relatively obviousatlow-frequency
a constant speed of 30 km/h
roar (20-100 on a rough
Hz) generated by a road
cer- surface, the
LMS equipment
tain hatch car when it isisrunning
used toatcollect the sound
a constant speed ofpressure
30 km/h level at theroad
on a rough driver’s right ear and at
surface,
the middle
the LMS equipment back passenger
is used seat.
to collect theSound pressurelevel
sound pressure sensors
at theare arranged
driver’s rightaccording
ear and to relevant
at thenational
middle back passenger
standards seat. Sound pressure
GB/T18697-2002, andsensors are arranged
the location accordingpoints
of measuring to rel- is shown in
evantFigure
national
4. standards
During the GB/T18697-2002,
signal acquisition and the location
process, theofadjustable
measuringseat points is shown
is back as far as possible
in Figure 4. During the signal acquisition process, the adjustable seat is back as far
to put it in a vertical position. The test is empty, with only the driver and measurement of as pos-
siblepersonnel
to put it inin
a vertical
the car. position.
Before theThetest
testweis empty,
ensure with onlycar
that the theisdriver and measure-
in a closed state, the four-wheel
ment of personnel in the car. Before the test we ensure that the car is in a closed state, the
positioning is normal and in the test environment, the influence of background noise is
four-wheel positioning is normal and in the test environment, the influence of background
less than the allowable error. When the vehicle speed is adjusted to a constant speed of
noise is less than the allowable error. When the vehicle speed is adjusted to a constant
30 km/h, LMS onboard equipment is used to collect the sound pressure signal inside the
speed of 30 km/h, LMS onboard equipment is used to collect the sound pressure signal
vehicle. The The
inside the vehicle. noise testtest
noise results areare
results shown
shownininFigure
Figure 5.
5.

(a) (b)
Figure 4. Position
Figure of sound
4. Position ofpressure sensors. (a)
sound pressure driver’s(a)
sensors. seat , (b) passenger’s
driver’s seat .
seat, (b) passenger’s seat.

55

50

45
B(A)

40
 Materials 2022, 15, 3261 7 of 18
(a) (b)
Figure 4. Position of sound pressure sensors. (a) driver’s seat , (b) passenger’s seat .

55

50

45

SPL dB(A)
40

35

30

25

20
Materials 2022, 15, x FOR PEER REVIEW 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 8 of 21
Frequency[Hz]
(a)

55
50
45
SPL dB(A)

40
35
30
25
20
20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Frequency[Hz]
(b)
Figure Figure
5. Spectrum diagram of
5. Spectrum noise inof
diagram the car. (a)
noise Driver’s
in the seat,Driver’s
car. (a) (b) Passenger’s
seat, (b)seat.
Passenger’s seat.

The testTheresults
testshow that
results the car
show hasthe
that ancar
obvious
has anpeak value at
obvious 35 Hz
peak in the
value atfrequency
35 Hz in the frequency
range of 20–100 Hz. The peak value at 35 Hz in the right ear of the driver
range of 20–100 Hz. The peak value at 35 Hz in the right ear of the driver is 50.39 dBis(A),
50.39 dB (A), and
and the peak value at 35 Hz in the middle of the passenger rear seat is 50.78 dB (A), as
the peak value at 35 Hz in the middle of the passenger rear seat is 50.78 dB (A), as shown
shown in Figure 4 (ordinate of Figure 4a is dB (A) and ordinate of Figure 4b is Pa). In the
in Figure 4 (ordinate of Figure 4a is dB (A) and ordinate of Figure 4b is Pa). In the LMS test
LMS test software, the collected signals were filtered to filter the peak value at 35 Hz, and
software, the collected signals were filtered to filter the peak value at 35 Hz, and the filtered
the filtered noise signals were listened to by earphones, and it was found that the low-
noise
frequency signals
roar were listened
had disappeared. to by earphones,
Therefore, and it wasthat
it can be determined found
the that the low-frequency roar
low-frequency
had disappeared. Therefore, it can be determined that the low-frequency
roar of this vehicle is caused by the peak noise at 35 Hz. As the peak frequency is very roar of this vehicle
low, theis continuous
caused bysteady-state
the peak noise
noiseat of35
lowHz. As the ispeak
frequency frequency
generally is very
considered low, the continuous
structural
steady-state noise of low frequency is generally considered structural noise [38,39].
noise [38,39].

3.1. Analysis
3.1. Analysis of Noise
of Interior Interior Noise Problem
Problem
GenerallyGenerally
speaking,speaking,
there are there are two
two ways ways to
to generate generatenoise:
structural structural noise:
the sound the sound cavity
cavity
mode ofmode of thevehicle
the whole wholeandvehicle and theresonance
the coupled coupled of resonance
the body of theamplify
parts body parts amplify the noise
the noise
to formtoa form
low-frequency roar; the low-frequency
a low-frequency vibration ofvibration
roar; the low-frequency body sheet ofmetal
bodyradiates to
sheet metal radiates to
the car,the
forming a low-frequency roar. In view of the above two approaches, this
car, forming a low-frequency roar. In view of the above two approaches, this paper paper
solves solves
the acoustic cavity mode
the acoustic of the
cavity mode vehicle and
of the analyzes
vehicle andtheanalyzes
joint contribution
the joint of the
contribution of the
panel parts based on the peak noise frequency of 35 Hz in the simulation analysis of
panel parts based on the peak noise frequency of 35 Hz in the simulation analysis of road road
noise. The analysis process is shown in Figure 6.
noise. The analysis process is shown in Figure 6.
Materials 2022, 15, x FOR PEER REVIEW 9 of 21
Materials 2022, 15, 3261 8 of 18

Start

Analysis of interior noise problem

The establishment of finite element model

Simulation analysis of road noise Analysis of joint contribution Modal analysis of vehicle voice

Analysis of joint contribution of

body parts

End

Figure 6. CAE diagnostic analysis flow of low-frequency roaring in the car.

Figure 6. CAE diagnostic analysis flow of low-frequency roaring in the car.
Through CAE simulation analysis, the sound mode of the vehicle has no resonance
with the body parts at 35 Hz. Therefore, the peak noise frequency of 35 Hz is mainly caused
Through CAE simulation analysis, the sound mode of the vehicle has no resonance
by the vibration radiation of the automobile panel.
with the body parts at 35 Hz. Therefore, the peak noise frequency of 35 Hz is mainly
caused by the
3.1.1. The vibration radiation
Establishment of Finiteof the automobile
Element Model panel.
In order to carry out acoustic cavity modal analysis, road noise simulation and body
3.1.1. The Establishment of Finite Element Model
plate joint contribution analysis of the whole vehicle, it is necessary to establish the sound
In finite
cavity order element
to carry model
out acoustic cavity modal coupled
and acoustic–solid analysis,finite
road element
noise simulation
model. and body
plate joint
The TB contribution
(Trim Body) analysis
bodyofisthe wholeconstructed
further vehicle, it isonnecessary
the basistoof establish
the finitetheelement
sound
cavity
modelfinite element modelthat
of body-in-white andhas acoustic–solid
been constructed coupled finite
and elementthrough
modified model. tests, which
The TB
includes all (Trim Body) the
parts except body is further
engine constructed
and chassis system,on many
the basis of theare
of which finite element
replaced by
model of body-in-white
centralized unit mass and that has been constructed
one-dimensional and modified
unit. Determine the through
modelingtests, which and
standards in-
cludes all parts
guidelines except
between thethe engine andofchassis
connections various system, many of(1)which
components: are replaced
Modeling by cen-
of spot welding
tralized
structuralunit mass and
adhesives. Spotone-dimensional
welding structural unit. Determine
adhesives the modeling
are mainly used in standards
key parts ofand the
guidelines between
car body, which canthe connections
greatly improveofthe various components:
stiffness of the car body(1) Modeling
when used of in
spot welding
conjunction
structural
with welding adhesives. Spot welding
spots. Modeling structural adhesives
of RBE3-hexa-RBE3 arewelding
for spot mainly of used in key adhesive.
structural parts of
(2) car
the Modeling of shock-absorbing
body, which can greatly improveexpansion theadhesive.
stiffnessShock-absorbing
of the car body when expansionusedadhesive
in con-
mainly connects
junction the outer
with welding spots. plate of the top
Modeling cover with the beam
of RBE3-hexa-RBE3 of the
for spot top cover,
welding the outer
of structural
plate of the
adhesive. (2) open and closed
Modeling parts with theexpansion
of shock-absorbing support plate or theShock-absorbing
adhesive. inner plate, and expan-mainly
plays
sion a shock-absorbing
adhesive mainly connectsrole. (3)theModeling
outer plate of of
reinforcement
the top cover plate
withandthedamping
beam of plate.
the topA
reinforcement
cover, the outerglueplateplate
of theis generally
open and used closed forparts
sidewith
circumference
the support and rearordoor
plate plates
the inner
with aand
plate, thickness
mainly of 1.5 mm.
plays It mainly plays
a shock-absorbing the(3)
role. role of stiffness
Modeling reinforcement plate
of reinforcement and hasanda
certain damping
damping plate. A dissipation
reinforcement function. (4) Seal
glue plate strip modeling.
is generally used for The
sideseal strip is simulated
circumference and
using
rear BUSH
door plates(used
withtoa thickness
define theofnominal1.5 mm.property
It mainlyvalues
plays the for role
a generalized
of stiffnessspring and
reinforce-
damper structural element) or ELAST (used to define the stiffness
ment and has a certain damping dissipation function. (4) Seal strip modeling. The seal and stress coefficient of
a scalar elastic element) in three translational directions. (5) Limit
strip is simulated using BUSH (used to define the nominal property values for a general- buffer block modeling.
A limit
ized springbuffer
andblock
damperis mainly
structural usedelement)
in the engine
or ELAST room cover
(used toand back
define thedoor, to play
stiffness anda
shockcoefficient
stress absorptionof and support
a scalar role. The
elastic BUSHinelement
element) is used to simulate
three translational and only
directions. three
(5) Limit
translational
buffer stiffnessesAare
block modeling. set.buffer
limit (6) Connection between
block is mainly car in
used door
thelock androom
engine lock. cover
The BUSHand
unit can be used to simulate the model car door lock and lock. (7) Hinge modeling for
Materials 2022,
Materials 2022, 15,
15, xx FOR
FOR PEER
PEER REVIEW
REVIEW 10 of
10 of 21
21

Materials 2022, 15, 3261 back door, to play a shock absorption and support role. The BUSH element is used 9 of to
18

simulate and only three translational stiffnesses are set. (6) Connection between car door
lock and lock. The BUSH unit can be used to simulate the model car door lock and lock.
(7) Hinge
open modeling
and closed for The
parts. openhinge
and closed
needs parts. The modeling,
solid grid hinge needs thesolid
gridgrid
sizemodeling,
can be setthe to
3–5
gridmm,
size and
can the solid
be set to grid requires
3–5 mm, andatthe
least three
solid layers
grid of body
requires elements
at least threein the direction
layers of body
of thickness.
elements If the
in the hinge is
direction of modeled
thickness.using
If theahinge
solid isgrid or a shell
modeled element,
using a solidthe stiffness
grid or a shellof
the hinge the
element, willstiffness
be overestimated.
of the hingeDoors
willgenerally contain two
be overestimated. hinges
Doors connected
generally by RBE2
contain two
hinges
and connected
CBEAM, by RBE2
and the RBE2 and CBEAM,
center of the twoandhinges
the RBE2 center
should be of
setthe twocollinear
to be hinges should
to ensure be
set to can
doors be collinear to ensure
rotate freely arounddoors can rotate
the hinge. freelyshaft
The hinge around the hinge.
is simulated The three
using hingeCBEAM
shaft is
units and releases
simulated rotational
using three CBEAM degrees
unitsofand
freedom at both
releases ends. The
rotational TB finite
degrees elementatmodel
of freedom both
established
ends. The TB is finite
shownelement
in Figure 7. established is shown in Figure 7.
model

Figure 7.
Figure 7. TB
TB finite
finite element
element model.
model.

In
InHypermesh
Hypermesh2017.01
2017.01software,
software,Altair
AltairEngineering
Engineering Inc. (Troy, MI,
Inc.,we USA),
extract thewe extract
inner the
surface
inner
of thesurface of the body’s
body's internal internal
contact contact
with air, with
to form air, to form closed
a completely a completely
acoustic closed
cavityacoustic
model,
cavity
which model,
defines which
the spokedefines the spoke
boundary boundary
as a fully as a When
rigid wall. fully rigid wall.the
dividing When
finitedividing
element
the finite
mesh, element
it should bemesh,
noteditthat
should be noted
the size of thethat the size
acoustic of theand
element acoustic element and
the structural the
element
structural
should beelement should
consistent, and be
theconsistent, and the should
acoustic element acoustichave
element should
at least have at least
six elements six
in each
elements in each wavelength range. As we are focusing on frequency
wavelength range. As we are focusing on frequency within 100 Hz, the minimum wave- within 100 Hz, the
minimum
length λ = c/fwavelength = c/fm,=and
= 344/100 =λ 3.44 344/100 = 3.44 m, length
the maximum and theofmaximum
the soundlength
cavity of theelement
grid sound
cavity grid element is 0.52 m. In this paper, a hexahedral element with a
is 0.52 m. In this paper, a hexahedral element with a side length of 50 mm is used to auto- side length of
50 mm is used to automatically divide the cavity mesh. The generated results
matically divide the cavity mesh. The generated results of the finite element model of the of the finite
element model
body cavity areofshown
the body cavity8.are shown in Figure 8.
in Figure

Figure 8.
Figure 8. Acoustic
Acoustic cavity
cavity finite
finite element
element model.
model.

The body structure and the sound cavity system are coupled through the vibration of
the boundary nodes. In Hypermesh2017.01 software, the coupling between the structure
and the sound cavity is realized by using the ACMODF card to search the structure nodes
Materials 2022,
Materials 2022, 15,
15, xx FOR
FOR PEER
PEER REVIEW
REVIEW 11 of
11 of 21
21

Materials 2022, 15, 3261 The body structure and the sound cavity system are coupled through the vibration 10 of 18

of the boundary nodes. In Hypermesh2017.01 software, the coupling between the struc-
ture and the sound cavity is realized by using the ACMODF card to search the structure
automatically throughthrough
nodes automatically the nodes
theonnodes
the sound
on thecavity
soundboundary. The acoustic–solid
cavity boundary. coupling
The acoustic–solid
model
couplingof the body
model ofsystem is established
the body as shown in
system is established asFigure
shown9in(the left door
Figure and
9 (the leftpart
door of and
the
side
partcircumference are hidden to
of the side circumference aredisplay
hiddenthe
to acoustic cavity
display the model).
acoustic cavity model).

Figure 9.
Figure 9. Sound–solid
Sound–solid coupling
coupling finite
finite element
element model
modelof
ofthe
thebody
bodysystem.
system.

3.1.2.
3.1.2. Simulation
Simulation Results
Resultsof ofRoad
RoadNoise
Noise
Based
Based on
on the
the acoustic–solid
acoustic–solid coupled
coupled finite
finite element
element model,
model, the
the time-domain
time-domain signals
signals
of
ofacceleration
acceleration measured
measured by by engineers
engineers onon the
the actual
actual road
roadsurface
surface at
atthe
thesteering
steeringknuckle
knuckle
of
of the
the front and rear
front and rear suspension
suspensionare areconverted
convertedintointoPSD
PSDfrequency-domain
frequency-domain signal
signal files
files to
to
stimulate the four wheel cores of the vehicle, and the solution response points are the
stimulate the four wheel cores of the vehicle, and the solution response points are the
sound
sound pressure
pressure level
level at
at the
the driver’s
driver’s right
right ear
ear and
and the
the middle
middle sound
sound pressure
pressure level
level of
of the
the
rear passenger’s seat. The solution frequency range is set as 20–100 Hz.
rear passenger’s seat. The solution frequency range is set as 20–100 Hz. The comparison The comparison
between
betweenthe thetest
testdata
dataand
and the
the simulation
simulation data
data of
of road
road noise
noise is
is shown
shown inin Figures
Figures 1010 and
and 11.
11.

Sound Pressure Level Response Comparison

55

50

45
dB(A)
SPL dB(A)

40

35
SPL

30 ------ Simulation
—— Test
25
Test
20
20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Frequency[Hz]
Figure10.
Figure 10.Comparison
Comparisonof
oftest
testresults
resultsand
andsimulation
simulationresults
resultsof
ofdriver’s
driver’sright
rightear
earnoise.
noise.

The results show that the sound pressure level of the driver’s right ear and the middle
of the passenger’s rear seat has an obvious peak value at 35 Hz in the low-frequency band,
which is corresponding to the peak frequency of the test results of the real car and the
overall trend is basically corresponding. The sound pressure level in the right ear of the
real car test driver is 50.39 dB (A), and the simulation data is 49.30 dB (A). The test sound
pressure level in the middle of the rear passenger row is 50.79 dB (A), and the simulation
data is 50.50 dB (A). In the simulation and test, the difference in the sound pressure level
between the driver’s right ear and the middle of the rear passenger’s row is relatively small
at the peak point, indicating that the previous modified finite element model has high
Materials 2022, 15, 3261 11 of 18

accuracy. In view of this kind of steady-state structural noise, the main contribution of the
Materials 2022, 15, x FOR PEER REVIEW
peak
frequency can be effectively determined by analyzing the joint contribution of 12 of
the21
body panel.

Sound Pressure Level Response Comparison

55

50

45
SPL dB(A)

40

35

30 ------ Simulation
25 —— Test
20
20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Frequency[Hz]
Figure11.
Figure 11.Comparison
Comparisonof
oftest
testresults
resultsand
andsimulation
simulationresults
resultsof
ofpassenger’s
passenger’sright
rightear
earnoise.
noise

3.1.3. The
Analysis ofshow
results Joint that
Contribution
the soundof Body Parts
pressure level of the driver’s right ear and the middle
of the
Thepassenger’s
control pathrear
ofseat has an
vehicle obvious
interior peak
noise value at 35excitation
is generally: Hz in thesource–transfer
low-frequency path– band,
which is point.
response corresponding to the peak models,
For mass-produced frequency of very
it is the test results
difficult to of the real
control thecar and the
excitation
source
overall(chassis–suspension system). Modifying
trend is basically corresponding. The soundthe chassis–suspension
pressure level in the system
rightwill
earhave a
of the
great impact
real car on theissmoothness,
test driver 50.39 dB (A), operation stability and
and the simulation datasafety performance
is 49.30 dB (A). The of the
testwhole
sound
vehicle,
pressureand theinmodification
level the middle of cost
theisrear
huge. Therefore,
passenger row foristhe noise
50.79 dB problem of body
(A), and the vibra-
simulation
tion
dataradiation,
is 50.50 dB general
(A). Inenterprises tend and
the simulation to consider
test, theitdifference
from the perspective
in the sound ofpressure
transmissionlevel
path (body
between thesheet metal).
driver’s rightThe
earcarandbody and theofcar
the middle sound
the cavity can be
rear passenger’s rowregarded as a
is relatively
closed
small atcavity. The point,
the peak closedindicating
cavity sheet thatmetal S low-frequency
the previous modified vibration radiates
finite element into the
model has
cavity to form the
high accuracy. In car’s
viewlow-frequency
of this kind of roar. Take a point
steady-state insidenoise,
structural the space O as the
the main origin of
contribution
coordinates, and the rest
of the peak frequency canofbe
theeffectively
points in determined
the space usebyit analyzing
as a reference point.
the joint Assume that
contribution of
point J is a point
the body panel. on the closed cavity sheet metal part S, and its radiation sound pressure
on a point i in the car is:
3.1.3. Analysis of Joint Contribution Pj (r, w)
Pij (r0 , wof) =Body Parts vi (r, w)σS j (16)
The control path of vehicle interior noise Qi is generally: excitation source–transfer
where Pij (r0 , w)point.
path–response is the For mass-produced
radiation sound pressure models, from very Jdifficult
it ispoint to pointtoI,control
Qi is the thevolume
excita-
sound
tion source at point i, Pj (r, w) issystem).
source(chassis–suspension the sound pressurethe
Modifying at point j, vi (r, w) is thesystem
chassis–suspension speed thatwill
the sound source j transmits to point
have a great impact on the smoothness, operation i, r 0 is the distance i
stability and safety performance ofthe
from point to point o, r is the
distance from point
whole vehicle, and thej to modification
point o, w is thecostfrequency. Many points
is huge. Therefore, on the
for the noisebody closedofcavity
problem body
sheet
vibration parts S radiation
metalradiation, noise to point
general enterprises tend i, and the proportion
to consider it fromof the point j is:
perspective of trans-
mission path (body sheet metal). The car body and the car sound cavity can be regarded
as a closed cavity. The closed cavity Pij (r0 , w)
t j (sheet
w) = metal S low-frequency vibration radiates (17) into
P ( r , w )
the cavity to form the car's low-frequency roar. Take a point inside the space O as the
0
origin of coordinates, and the rest of the points in the space use it as a reference point.
P(r0 , w) is the total sound pressure at point i, and t j (w) is the contribution of point j
Assume that point J is a point on the closed cavity sheet metal part S, and its radiation
on the body closed cavity sheet metal part S to the interior noise. When all the point
sound pressure on a point i in the car is:
contributions are plotted together, the contribution of each area to the interior acoustic
radiation can be determined. Through the analysis 𝑃𝑗 (𝑟, 𝑤)of the contribution source of the nodes,
𝑃 (𝑟0 , 𝑤) = 𝑣 (𝑟, 𝑤)𝜎𝑆𝑗 (16)
the main contribution node area𝑖𝑗can be found, 𝑄𝑖so as𝑖 to determine the source of the low-
frequency roar contribution. Analysis results of 35 Hz body panel joint contribution were
where 𝑃 (𝑟 , 𝑤) is the radiation sound pressure from point J to point I, 𝑄𝑖 is the volume
extracted,𝑖𝑗 as0 shown in Figure 12. The results show that the rear door has the largest
sound sourcetoat35point
contribution i, 𝑃𝑗 (𝑟, 𝑤) is the sound pressure at point j, 𝑣𝑖 (𝑟, 𝑤) is the speed that
Hz noise.
the sound source j transmits to point i, 𝑟0 is the distance from point i to point o, r is the
distance from point j to point o, w is the frequency. Many points on the body closed cavity
sheet metal parts S radiation noise to point i, and the proportion of point j is:
𝑃𝑖𝑗 (𝑟0 , 𝑤)
𝑡𝑗 (𝑤) = (17)
𝑃(𝑟0 , 𝑤)
 𝑃(𝑟0 , 𝑤) is the total sound pressure at point i, and 𝑡𝑗 (𝑤) is the contribution of point
j on the body closed cavity sheet metal part S to the interior noise. When all the point
contributions are plotted together, the contribution of each area to the interior acoustic
radiation can be determined. Through the analysis of the contribution source of the nodes,
the main contribution node area can be found, so as to determine the source of the low-
frequency roar contribution. Analysis results of 35 Hz body panel joint contribution were
Materials 2022, 15, 3261 12 of 18
extracted, as shown in Figure 12. The results show that the rear door has the largest con-
tribution to 35 Hz noise.

Figure 12. A 35 Hz cloud graph of plate joint contribution.

Figure 12. A 35 Hz cloud graph of plate joint contribution.
3.1.4. Simulation Analysis of Vehicle Acoustic Cavity Mode
3.1.4. Simulation Analysis of Vehicle Acoustic Cavity Mode
In the process of driving a car, the sheet metal parts of the car body and other sealing
In the process of driving a car, the sheet metal parts of the car body and other sealing
components form a closed cavity, and the acoustic cavity mode of the car reflects the
components form a closed cavity, and the acoustic cavity mode of the car reflects the in-
inherent
herent properties of the of
properties the interior
interior air fluid.
air fluid. When When the
the structure structure
resonates resonates
with the acoustic with the acoustic
cavity mode,
cavity mode, thethe interior
interior noise
noise will will be amplified
be amplified to form the low-frequency
to form the low-frequency roaring sound roaring sound
(20 Hz–100
(20 Hz–100 Hz).
Hz).By comparing
By comparing the result
theof result
free modeof analysis
free mode of theanalysis
sound cavity withsound cavity with
of the
the noise in the car, it can be judged whether the 35 Hz noise in the car is caused by the
the noise in the car, it can be judged whether the 35 Hz noise in the car is caused by the
resonance between the sound cavity and the body panel.
resonance
The modalbetween
solutionthecardsound
was setcavity andcavity
up in the the body panel. model established
finite element
The
earlier, andmodal
the modal solution
solutioncard wasimported
file was set up into
in the cavity finite
Nastran14.0 solver element model established
for solving.
earlier, and the modal
Through calculation, the modalsolution
frequency file
andwas
mode imported
shape of theinto Nastran14.0
acoustic solver for solving.
cavity of the
structure are
Through obtained in Table
calculation, the 2,modal
Figuresfrequency
13 and 14: and mode shape of the acoustic cavity of the
structure are obtained
Table 2. Acoustic cavity modesin Table 2, Figures 13 and 14:
table.

Table 2. Order Time

Acoustic Model Frequency
cavity modes table. Model Vibration Shape
The first-order 49.98 Hz First-order longitudinal
The second-order
Order Time 94.63Model
Hz FrequencyFirst-order lateral
Model Vibration Shape
Materials
Materials2022,
2022,15,
15,xxFOR
FORPEER
PEERREVIEW
REVIEW
The first-order 49.98 Hz First-order14longitudinal
14of
of21
21
The second-order 94.63 Hz First-order lateral

13. Model vibration

Figure Model
Figure shape of the first-order acoustic cavity.
Figure13.
13. Modelvibration
vibrationshape
shapeof
ofthe
thefirst-order
first-orderacoustic
acousticcavity.
cavity.

Figure
Figure
Figure 14. 14.
14.ModelModel
Model vibration
vibration
vibration shapeshape
shape of
ofthe of the second-order
thesecond-order
second-order acoustic acoustic
cavity. cavity.
acousticcavity.

3.2.
3.2. Acoustic
Acoustic Metamaterial
Metamaterial Design
Design and
and Simulation
Simulation Analysis
Analysis
3.2.1.
3.2.1. Establishment of Mathematical Model
Establishment of Mathematical Model of
of Acoustic
Acoustic Metamaterial
Metamaterial
Through
Throughsimulation,
simulation, the
thefront
frontsection
sectionhas
hasdetermined
determinedthat
that the
the peak
peaknoise
noise of
of35
35 Hz
Hz in
in
Materials 2022, 15, 3261 13 of 18

3.2. Acoustic Metamaterial Design and Simulation Analysis

3.2.1. Establishment of Mathematical Model of Acoustic Metamaterial
Through simulation, the front section has determined that the peak noise of 35 Hz in
the car is mainly caused by low-frequency vibration radiation of the tail door. Therefore,
according to the corresponding size structure of the plate of the tail door, an acoustic
metamaterial with specific frequency suppression can be designed. According to the
mathematical model of the tailgate, the installation position of the acoustic metamaterial is
determined corresponding to the position of the red box shown in Figure 15 (the rectangular
shape of the acoustic metamaterial is 60 × 204 × 3 mm, 1 is the inner panel of the tail door,
2 is the reinforced floor of the tail door, 3 is the outer panel of the tail door, 4 is the car tail
door lock, and the red dotted box is the attachment position of the acoustic metamaterial of
the tail door). The acoustic metamaterial is composed of three parts: Acoustic metamaterial
matrix plate 5, adjustable mass block 6 and acoustic metamaterial damping layer 7, as is
shown in Figure 16. The designed acoustic metamaterial is directly attached to the sheet
metal parts of the tailgate through the magnetic damping layer, eliminating the need for
a complicated installation process. The adjustable mass block and the cantilever beam in
Materials 2022, 15, x FOR PEER REVIEW the acoustic metamaterial layer constitute a typical acoustic metamaterial spring–mass
Materials 2022, 15, x FOR PEER REVIEW
15 of 21
15
system, which directionally regulates the band gap frequency of the acoustic metamaterial
by adjusting the weight of the mass block.

Figure 15.
Figure Installation position
15. Installation position of
of acoustic
acoustic metamaterial
metamaterial (red
(red dotted
dotted box).
box).
Figure 15. Installation position of acoustic metamaterial (red dotted box).

(a) (a) (b) (b)

Figure
Figure 16.
16. Mathematical
Mathematical model
model ofof acoustic
acoustic metamaterial.
metamaterial. (a)
(a) The
The designed
designed local resonance
localdesigned
resonance acoustic
acoustic
Figure 16. Mathematical model of acoustic metamaterial. (a) The local resonance acou
metamaterial, (b) the local resonance unit.
metamaterial, (b) the local
metamaterial, (b) the local resonance unit. resonance unit.

3.2.2. Frequency
3.2.2.Design of Acoustic
Frequency DesignMetamaterial Resonance Element
of Acoustic Metamaterial Resonance Element
The mathematical
The mathematical
The model model of acoustic
of acoustic
mathematical ofmetamaterial
metamaterial
model was imported
acoustic wasinto imported
metamaterial was into
Hypermesh2017.01,Hy- into
imported
which was processed
permesh2017.01, whichby was
permesh2017.01,solidprocessed
cutting
which was and placed
byprocessedin the
by mapping
solid cutting state.
and placed
solid cutting in A
thehexahedral
and mapping
placed mesh
state.
in the mapping s
of hexahedral
A 1mm was established
mesh of for each
1mm was cut entity. The
established fortotal
each number
cut of acoustic
entity. The metamaterials
total number of grid
acous-
A hexahedral mesh of 1mm was established for each cut entity. The total number of ac
tic metamaterials grid is 40,324.
tic metamaterials Assign
grid material
is 40,324. Assignattributes
material to attributes
each component, as shown as sho
to each component,
in Table 3. After designing the modal solution card, it is imported into Nastran
in Table 3. After designing the modal solution card, it is imported into Nastran solver for solver
solving the calculation. Based on the principle of local resonance of acoustic metamaterial,
solving the calculation. Based on the principle of local resonance of acoustic metamate
the resonancetheelement has element
resonance to produce haslocal resonance
to produce localwith the sheet
resonance metal.
with The natural
the sheet metal. The nat
Materials 2022, 15, 3261 14 of 18

is 40,324. Assign material attributes to each component, as shown in Table 3. After design-
ing the modal solution card, it is imported into Nastran solver for solving the calculation.
Based on the principle of local resonance of acoustic metamaterial, the resonance element
has to produce local resonance with
q the sheet metal. The natural frequency of the resonant
1 k
element
Materials 2022, 15, x FOR PEER REVIEW is calculated as f = 2π m (k is the equivalent stiffness, m is the equivalent 16 mass).
of 21
By adjusting the thickness of attached mass block 6 (which is equivalent to changing the
value of f by changing m in the calculation formula of natural frequency), the first-order
natural frequency of each resonance element in the acoustic metamaterial is within the
Table 3. Properties of acoustic metamaterial materials.
range of 35 ± 0.5 Hz. The adjusted calculation results are shown in Figure 17. The natural
frequencies of each resonance
Modulus unitofare, respectively, 34.94 Hz, 34.94
Elastic- Hz, 34.95 Hz, 34.95 Hz,
Part Name Density (𝑻/𝒎𝒎𝟑 ) Poisson’s Ratio
35.00 Hz and 35.00 Hz. ity (MPa)
Acoustic metamate-
71,000
Table 3. Properties of acoustic metamaterial materials. 2.7 × 10−9 0.3
rial Matrix Plate 5
AdjustablePartMass Modulus of 3
Name 210,000 Elasticity (MPa)7.9 ×Density
10−9 (T/mm ) Poisson’s
0.3 Ratio
Block 6
Acoustic metamaterial Matrix Plate 5 71,000 2.7 × 10−9 0.3
Acoustic metamate- −9
Adjustable Mass Block 6 3457 210,000 2.1 × 107.9
−9 × 10 0.49 0.3
rial Damping
Acoustic LayerDamping
metamaterial 7 Layer 7 3457 2.1 × 10−9 0.49

(a) (b)

(c) (d)

(e) (f)
Figure
Figure 17.
17. Natural
Natural frequency
frequency ofof acoustic metamaterial resonance
acoustic metamaterial resonance element.
element. (a)
(a) 34.94
34.94Hz,
Hz,(b)
(b)34.94
34.94Hz,
Hz, (c) 34.95
(c) 34.95 Hz, Hz, (d) 34.95
(d) 34.95 Hz, Hz, (e) 35.00
(e) 35.00 Hz,35.00
Hz, (f) (f) 35.00
Hz. Hz.

Results and
4. Results and Discussion
Discussion
After the
the corresponding
corresponding acoustic
acoustic metamaterial
metamaterial is designed,
designed, thethe acoustic
acoustic metamaterial
metamaterial
finite element
finite element model
model isis attached
attached toto the
the design
design part,
part, and
and the whole vehicle
vehicle finite element
element
model with
model with thethe acoustic
acoustic metamaterial
metamaterial attached
attached to
to the
the tailgate
tailgate is
is established,
established, as
as shown
shown inin
Figure 18
Figure 18 (in
(inorder
ordertotosee thethe
see installation position
installation of the
position of acoustic metamaterial,
the acoustic the attached
metamaterial, the at-
sheet metal
tached sheetparts
metalis parts
displayed transparently).
is displayed The response
transparently). points are
The response the sound
points are thepressure
sound
level in the driver’s right ear, the middle pressure level in the rear passenger’s
pressure level in the driver’s right ear, the middle pressure level in the rear passenger’s seat and the
vibration
seat and the acceleration
vibration in the middleinofthe
acceleration themiddle
tailgateofpanel.
the tailgate panel.
Materials 2022, 15, 3261
Materials 2022, 15, x FOR PEER REVIEW 17 of 21
15 of 18
Materials 2022, 15, x FOR PEER REVIEW 17 of 21

Figure 18.
Figure 18. Acoustic
Acousticmetamaterial
metamaterialfinite element
finite model
element attached
model to the vehicle.
attached to the vehicle.
Figure 18. Acoustic metamaterial finite element model attached to the vehicle.
The same
The sameroadroadnoise
noise excitation
excitationsignal and and
signal excitation point point
excitation positions as in Section
positions as in 3Section 3
were The
usedsame
to road noise
conduct road excitation
noise signal and
simulation excitation
analysis and point positions
vibration as inanalysis
response Sectionof3
were used to conduct road noise simulation analysis and vibration response analysis of the
were usedmetal
the plate to conduct
parts ofroad
thenoise simulation
tailgate analysis
before and after and vibration of
the attaching response analysis of
the acoustic–solid
plate
the
metal parts
platefinite
of thethe
metalelement
tailgate
parts ofmodel.
before
tailgate
and after
before and
the
afterare
attaching
theshown
attaching
ofofthe acoustic–solid
the 19–22.
acoustic–solid
coupled
coupled The simulation results in Figures By com-
finite
coupled element
finite model.
element The simulation results are shown in Figures 19–22. By comparing the
paring the data before model.
and after The
thesimulation results are shown
acoustic metamaterial, in Figuresand
the vibration 19–22. Byreduc-
noise com-
data
paring before
the of
tion effect
and
data after and
thebefore
themetamaterial
acoustic
acoustic
after metamaterial,
the acoustic the vibration
metamaterial,
was verified.
and noise
the vibration reduction
and noise reduc- effect of
the
tionacoustic metamaterial
effect of the was verified.
acoustic metamaterial was verified.

Sound Pressure Level Response Comparison

55 Sound Pressure Level Response Comparison
55 49.30
50 49.30
50
45
dB(A)

45
SPLdB(A)

40
40
35 47.30
35 47.30
SPL

30 ------ Original State

30 ------
—— Original
AcousticState
Superstructure
25 —— Acoustic Superstructure
25
20
20 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
20 25 30 35 40 45 50Frequency[Hz]
55 60 65 70 75 80 85 90 95 100
Frequency[Hz]
Figure 19. Comparison of sound pressure levels in the driver’s right ear.
Materials 2022, 15, x FOR PEER REVIEW 18 of 21
Figure 19.
Figure 19.Comparison
Comparisonof of
sound pressure
sound levels
pressure in theindriver’s
levels right ear.
the driver’s right ear.

Sound Pressure Level Response Comparison

55
50.50
50

45
SPL dB(A)

40
48.20
35

30 ------ Original State

25 —— Acoustic Superstructure
20
20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Frequency[Hz]
Figure 20.
Figure 20.Comparison
Comparisonof of
sound pressure
sound levels
pressure in theinmiddle
levels of rearof
the middle passenger row.
rear passenger row.

Comparison of Vibration Acceleration Response：X

160
RMS(mm/s2)

140 139.00
120
108.00
100
------ Original State
 48.20
35

SPL
S
30 ------ Original State
30 ------ Original
25 —— AcousticState
Superstructure
25 —— Acoustic Superstructure
20
20 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Materials 2022, 15, 3261 20 25 30 35 40 45 Frequency[Hz]
50 55 60 65 70 75 80 85 90 95 100 16 of 18
Frequency[Hz]
Figure 20. Comparison of sound pressure levels in the middle of rear passenger row.
Figure 20. Comparison of sound pressure levels in the middle of rear passenger row.
Comparison of Vibration Acceleration Response：X
160 Comparison of Vibration Acceleration Response：X

2) 2)
160
139.00

Accelerations—RMS(mm/s
140
139.00

Accelerations—RMS(mm/s
140
120
120 108.00
100 108.00
100 ------ Original State
80 ------ Original
80 —— AcousticState
Superstructure
60 —— Acoustic Superstructure
60
40
40
20
20
0
0 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
20 25 30 35 40 45 50 55 60 65
Frequency[Hz] 70 75 80 85 90 95 100
Frequency[Hz]
Figure 21.
Figure 21. Comparison
Comparison of of
x-direction vibration
x-direction response
vibration of sheet
response of metal
sheet parts
metalofparts
the tail
of door.
the tail door.
Figure 21. Comparison of x-direction vibration response of sheet metal parts of the tail door.

Comparison of Vibration Acceleration Response：Z

300 Comparison of Vibration Acceleration Response：Z
2] 2]

300
Accelerations—RMS[mm/s

250
Accelerations—RMS[mm/s

250
200 175.00
200 ------ Original State
175.00 ------
150
139.00 —— Original
AcousticState
Superstructure
150
139.00 —— Acoustic Superstructure
100
100
50
50
0
0 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
20 25 30 35 40 45 Frequency[Hz]
50 55 60 65 70 75 80 85 90 95 100
Frequency[Hz]
Figure 22. Comparison of z-direction vibration response of sheet metal parts of tail door.
Figure 22.
Figure 22. Comparison
Comparison of of
z-direction vibration
z-direction response
vibration of sheet
response of metal
sheet parts
metalofparts
tail door.
of tail door.
In the original state without an acoustic metamaterial attached, the sound pressure
In
levelInat the
theoriginal
the driver’s state
originalright without
ear
state an an
and the
without acoustic
middle metamaterial
seat
acoustic the rear attached,
ofmetamaterial theseat
passenger’s
attached,sound
has pressure
the an obvi-pressure
sound
level at
ous peak
level the
at the driver’s
value right
at 35right
driver’s ear
Hz; the and the middle
earamplitude seat of
of the sound
and the middle the rear passenger’s
seat ofpressure
the rear level seat has
seatanhas
of the driver’s
passenger’s obvi-
right
an obvious
ous peak value at 35 Hz; the amplitude of the sound pressure level of the driver’s right
peak value at 35 Hz; the amplitude of the sound pressure level of the driver’s right ear at
35 Hz is 49.30 dB (A), and the amplitude of the sound pressure level of the passenger’s
middle seat at 35 Hz is 50.50 dB (A). Under the condition of attached acoustic metamaterial,
the amplitude of the sound pressure level in the driver’s right ear and the middle position
of the passenger’s rear row is significantly reduced at 35 Hz. The amplitude of sound
pressure level at 35 Hz in the driver’s right ear decreased from 49.30 dB (A) to 47.30 dB (A),
and the improvement reached 2.0 dB (A). The amplitude of sound pressure level at 35 Hz
in the middle of the passenger rear row decreased from 50.50 dB (A) to 48.20 dB (A), and
the improvement reached 2.30 dB (A).
In the original condition without attached acoustic metamaterial, the vibration ac-
celeration of the tailgate panel appears at an obvious peak value of 35 Hz. In the vehicle
coordinate system, the amplitude of vibration acceleration of the tailgate panel at 35 Hz
in the x-direction is 139 m/s2 , and the amplitude of vibration acceleration at 35 Hz in the
z-direction is 175 m/s2 . Under the acoustic metamaterial condition, the amplitude of vibra-
tion acceleration of the tailgate panel at 35 Hz was significantly reduced. The amplitude
of vibration acceleration at the x-direction of 35 Hz of the tailgate panel is reduced from
139 m/s2 to 108 m/s2 , and the improvement reaches 31 m/s2 . The amplitude of vibration
acceleration at z-direction 35 Hz of the tailgate sheet metal parts decreased from 175 m/s2
to 139 m/s2 , and the improvement reached 36 m/s2 .
Materials 2022, 15, 3261 17 of 18

By comparing the data verified by the above simulation, it is found that the local
resonant acoustic metamaterial has an ideal effect on the control of low-frequency noise
radiated from the vibration of sheet metal parts to the vehicle.
Regarding the durability of the device, components of acoustic metamaterial: alu-
minum plate and the damping layer have good durability and have been widely used in
the automotive industry, its durability can meet the needs of mass production applications.
In addition to that, the three-interval method of random vibration fatigue life calculation
based on Gaussian distribution and Miner’s linear cumulative damage criterion shows
that the structure can carry out infinite cycles under the random load, which meets the
engineering requirements.

5. Conclusions
Focusing on the problem of low-frequency noise in a certain car, this paper determines
the frequency point of an abnormal peak value of noise in the car through test and simu-
lation analysis. Through CAE simulation analysis, it is concluded that the low-frequency
noise is caused by the low-frequency vibration of sheet metal parts of the tailgate. To solve
this problem, an acoustic metamaterial is designed based on the band gap theory of local
resonant phonon crystal. Through simulation verification, it is judged that the structure
can indeed reduce the amplitude of interior noise and vibration acceleration, which verifies
the effectiveness of the acoustic metamaterial. It provides a new way of thinking about
the problem that the low-frequency noise of vibration and radiation of sheet metal parts is
difficult to be controlled.

Author Contributions: Writing—original draft preparation, Y.L.; writing—review and editing, H.H.;
investigation, G.C.; software, D.L.; methodology, C.X.; data curation, Y.W.; software, J.T. All authors
have read and agreed to the published version of the manuscript.
Funding: This work was supported by the Chinese National Science Foundation Grant (No. 51905408,
51775451).
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Huang, H.B.; Wu, J.H.; Huang, X.R.; Yang, M.L.; Ding, W.P. The development of a deep neural network and its application to
evaluating the interior sound quality of pure electric vehicles. Mech. Syst. Signal Process. 2019, 120, 98–116. [CrossRef]
2. Rubino, C.; Bonet Aracil, M.; Gisbert-Payá, J.; Liuzzi, S.; Stefanizzi, P.; Zamorano Cantó, M.; Martellotta, F. Composite eco-friendly
sound absorbing materials made of recycled textile waste and biopolymers. Materials 2019, 12, 4020. [CrossRef] [PubMed]
3. Qin, Y.; Tang, X.; Jia, T.; Duan, Z.; Zhang, J.; Li, Y.; Zheng, L. Noise and vibration suppression in hybrid electric vehicles: State of
the art and challenges. Renew. Sustain. Energy Rev. 2020, 124, 109782. [CrossRef]
4. Huang, H.B.; Wu, J.H.; Huang, X.R.; Ding, W.P.; Yang, M.L. A novel interval analysis method to identify and reduce pure electric
vehicle structure-borne noise. J. Sound Vib. 2020, 475, 115258. [CrossRef]
5. Wellmann, T.; Govindswamy, K.; Tomazic, D. Impact of the future fuel economy targets on powertrain, driveline and vehicle
NVH development. SAE Int. J. Veh. Dyn. Stab. NVH 2017, 1, 428–438. [CrossRef]
6. Wu, J.; Ma, F. Application of acoustic metamaterials in low frequency vibration and noise reduction. J. Mech. Eng. 2016, 52, 68–78.
[CrossRef]
7. Liu, Z.; Chan, C.T.; Sheng, P. Three-component elastic wave band-gap material. Phys. Rev. B 2002, 65, 165116. [CrossRef]
8. Petrella, A.; Di Mundo, R.; Notarnicola, M. Recycled expanded polystyrene as lightweight aggregate for environmentally
sustainable cement conglomerates. Materials 2020, 13, 988. [CrossRef]
9. Cummer, S.A.; Christensen, J.; Alù, A. Controlling sound with acoustic metamaterials. Nat. Rev. Mater. 2016, 1, 16001. [CrossRef]
10. Yang, Z.; Mei, J.; Yang, M.; Chan, N.H.; Sheng, P. Membrane-type acoustic metamaterial with negative dynamic mas. Phys. Rev.
Lett. 2008, 101, 204301. [CrossRef]
11. Yang, Z.; Dai, H.M.; Chan, N.H.; Ma, G.C.; Sheng, P. Acoustic metamaterial panels for sound attenuation in the 50–1000 Hz
regime. Appl. Phys. Lett. 2010, 96, 41906. [CrossRef]
12. Huang, H.B.; Wu, J.H.; Huang, X.R.; Yang, M.L.; Ding, W.P. A generalized inverse cascade method to identify and optimize
vehicle interior noise sources. J. Sound Vib. 2020, 467, 115062. [CrossRef]
13. Qiu, C.; Liu, Z. Acoustic directional radiation and enhancement caused by band-edgestates of two-dimensional phononic crystals.
Appl. Phys. Lett. 2006, 89, 063106. [CrossRef]
Materials 2022, 15, 3261 18 of 18

14. Zhang, S.; Yin, L.L.; Fang, N. Focusing Ultrasound with an Acoustic Metamaterial Network. Phys. Rev. Lett. 2009, 102, 194301.
[CrossRef]
15. Deulgaonkar, V.R.; Kallurkar, S.P.; Mattani, A.G. Review and Diagnostics of noise and vibrations in automobiles. Int. J. Mod. Eng.
Res. 2011, 1, 242–246.
16. Huo, G.; Chen, Z.; Jiao, Y.; Zhu, X. Optimization of Dynamic Vibration Absorbers for Suppressing Sound Radiation of Plate
Structures. J. Vib. Acoust. 2021, 143, 021003. [CrossRef]
17. Pochivalov, K.V.; Shilov, A.N.; Lebedeva, T.N.; Ilyasova, A.N.; Golovanov, R.Y.; Basko, A.V.; Kudryavtsev, Y.V. Development of
vibration damping materials based on butyl rubber: A study of the phase equilibrium, rheological, and dynamic properties of
compositions. J. Appl. Polym. Sci. 2020, 138, 50196. [CrossRef]
18. Kafesaki, M.; Sigalas, M.; García, N. Frequency Modulation in the Transmittivity of Wave Guides in Elastic-Wave Band-Gap
Materials. Phys. Rev. Lett. 2000, 85, 4044. [CrossRef]
19. Goyal, D.; Mittal, S.K.; Choudhary, A.; Dang, R.K. Graphene: A two dimensional super material for sensor applications. Mater.
Today Proc. 2021, 43, 203–208.
20. Zhu, X.; Liang, B.; Cheng, J. Acoustic unusual material and acoustic invisibility cloak. Knowl. Mod. Phys. 2012, 24, 40–46.
21. Wen, X.; Wen, J.; Yu, D. Phononic Crystals; National Defense Industry Press: Beijing, China, 2009; pp. 1–180.
22. Chen, Y.; Li, J.; Huang, T.; Zhou, M. Calculation of band gap of layered ternary periodic structure. Noise Vib. Control 2014, 34,
19–22.
23. Kweun, J.M.; Lee, H.J.; Oh, J.H.; Seung, H.M.; Kim, Y.Y. Transmodal Fabry-Pérot resonance: Theory and realization with elastic
metamaterials. Phys. Rev. Lett. 2017, 118, 205901. [CrossRef] [PubMed]
24. Liu, J.; Guo, H.; Wang, T. A review of acoustic metamaterials and phononic crystals. Crystals 2020, 10, 305. [CrossRef]
25. Qian, D.; Shi, Z. Study on the bandgap characteristics of the simplified model of local resonant periodic double-layer plate.
Vibration. Test. Diagn. 2019, 39, 884–891.
26. Chen, J.; Sun, C.T. Reducing vibration of sandwich structures using antiresonance frequencies. Compos. Struct. 2012, 94, 2819–2826.
[CrossRef]
27. Zhang, S.; Hui Wu, J.; Hu, Z. Low-frequency locally resonant band-gaps in phononic crystalplates with periodic spiral resonators.
J. Appl. Phys. 2013, 113, 163511. [CrossRef]
28. Zhou, T.; Cheng, L. A resonant beam damper tailored with acoustic black hole features for broadband vibration reduction. J.
Sound Vib. 2018, 430, 174–184. [CrossRef]
29. Li, S.; Dou, Y.; Chen, T.; Wan, Z. Formation mechanism of meso-lf wide band gap in phonon crystal plate. J. Xi’an Jiaotong Univ.
2008, 52, 159–166.
30. Zhang, X.; Takezawa, A.; Kang, Z. A phase-field based robust topology optimization method for phononic crystals design
considering uncertain diffuse regions. Comput. Mater. Sci. 2019, 160, 159–172. [CrossRef]
31. Huang, H.B.; Li, R.X.; Yang, M.L.; Lim, T.C.; Ding, W.P. Evaluation of vehicle interior sound quality using a continuous restricted
Boltzmann machine-based DBN. Mech. Syst. Signal Process. 2017, 84, 245–267. [CrossRef]
32. Wang, Y.; Guo, H.; Lu, J.; Wang, Y.; Liu, N. Study on sound insulation characteristics of inhomogeneous local resonant phononic
crystals. J. Artif. Cryst. 2007, 46, 1250–1257.
33. Zhang, J.; Yao, H.; Du, J.; Zhao, J.; Dong, Y.; Qi, P. Low-frequency sound insulation characteristics based on local resonant
phononic crystals in engine room. J. Silic. 2016, 44, 1440–1445.
34. Zhang, Y.; Yin, J.; Wen, J.; Yu, D. Low-frequency vibration reduction design based on mass amplification of local resonant phonon
crystals. Vib. Impact 2016, 35, 26–32.
35. Dong, Y.; Du, J.; Yao, H.; Zhao, J.; Wu, J. Study on the band gap characteristics of double-clad locally resonant phonon crystals. J.
Artif. Cryst. 2015, 44, 3676–3680.
36. Oudich, M.; Zhou, X.; Badreddine Assouar, M. General analytical approach for sound transmission loss analysis through a thick
metamaterial plate. J. Appl. Phys. 2014, 116, 193509. [CrossRef]
37. Ma, F.; Wu, J.; Huang, M. Resonant Modal Group Theory of Membrane-Type Acoustical Metamaterials for Low-Frequency Sound
Attenuation. Eur. Phys. J. Appl. Phys. 2015, 71, 30504. [CrossRef]
38. Long, G.; Lim, T. A new robust delayless subband adaptive filtering algorithm with variable step sizes for active control of
broadband noise. Appl. Acoust. 2021, 176, 107858. [CrossRef]
39. Huang, H.; Huang, X.; Ding, W.; Yang, M.; Fan, D.; Pang, J. Uncertainty optimization of pure electric vehicle interior tire/road
noise comfort based on data-driven. Mech. Syst. Signal Process. 2022, 165, 108300. [CrossRef]