Download as pdf or txt
Download as pdf or txt
You are on page 1of 21

machines

Article
Vibration Characteristics of a Dual-Rotor System
with Non-Concentricity
Shengliang Hou 1 , Lei Hou 1,2, *, Shiwei Dun 3 , Yufeng Cai 3 , Yang Yang 2 and Yushu Chen 1

1 School of Astronautics, Harbin Institute of Technology, Harbin 150001, China; 17B918014@stu.hit.edu.cn (S.H.);
yschen@hit.edu.cn (Y.C.)
2 Applied Mechanics and Structure Safety Key Laboratory of Sichuan Province, School of Mechanics and
Engineering, Southwest Jiaotong University, Chengdu 610031, China; 181042yy@163.com
3 Factory of Xiang Yang Hang Tai Power Machinery, Xiangyang 441000, China; dsw39397850@163.com (S.D.);
caiyufeng51888@163.com (Y.C.)
* Correspondence: houlei@hit.edu.cn

Abstract: A finite element model of an aero-engine dual-rotor system with intermediate bearing
supported by six bearings is set up. Three modes of non-concentricity caused by the assembly
process are defined, namely parallel non-concentricity, front deflection angle non-concentricity and
rear deflection angle non-concentricity. The influence of the non-concentricity on the vibration
characteristics of the dual-rotor system is investigated in detail. The results show that the parallel
non-concentricity and the front deflection angle non-concentricity have a significant influence on
the bending vibration modals of the high-pressure rotor and the low-pressure rotor, but have little
influence on the local vibration modals of the rotors. With the increase in the magnitude of the non-
concentricity, the natural frequencies of the bending modals decrease continuously, and the mode
shapes of bending modals and that of local modals may be interchanged, leading to the emergence of

 bending modals in advance. Therefore, the key parameters to be controlled in the assembly process
Citation: Hou, S.; Hou, L.; Dun, S.;
are the parallel non-concentricity and the front deflection angle non-concentricity. In order to prevent
Cai, Y.; Yang, Y.; Chen, Y. Vibration the bending modal of the dual-rotor system from appearing in advance, it is necessary to control
Characteristics of a Dual-Rotor the parallel non-concentricity within 2 mm and the front deflection angle non-concentricity amount
System with Non-Concentricity. within 0.18◦ .
Machines 2021, 9, 251. https://
doi.org/10.3390/machines9110251 Keywords: dual-rotor system; non-concentricity in the assembly process; finite element model;
vibration characteristics
Academic Editor: Davide Astolfi

Received: 27 August 2021


Accepted: 20 October 2021 1. Introduction
Published: 26 October 2021
Due to the manufacture and installation errors of the support components and the de-
formation of elastic support structure, the concentricity of high-pressure and low-pressure
Publisher’s Note: MDPI stays neutral
rotors, which may cause excessive vibration, is hardly guaranteed during the assembly
with regard to jurisdictional claims in
process of the dual-rotor system. The extreme example will cause the damage of the
published maps and institutional affil-
turboshaft, couplings and other components. Thus, it will affect the flight safety of the
iations.
aircraft. Whether the multi rotor “series” rotor system formed by coupling or the multi
rotor “parallel” rotor system formed by coupling of intermediate bearings, there exist many
types of misalignment and non-concentricity, no matter what the formation of the multi
rotors is [1].
Copyright: © 2021 by the authors.
For the type of series-multi rotors represented by power turbine, the research subjects
Licensee MDPI, Basel, Switzerland.
mainly focus on the misalignment of the coupling and the supporting. Xia et al. [2]
This article is an open access article
introduced the misalignment mechanism of the coupling in detail, pointing out that the
distributed under the terms and
torque of misalignment will be generated in the misalignment of coupling, which will
conditions of the Creative Commons
lead to the transmission of the torque of misalignment to the rotor system. Then, the
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
rotor system will produce more complex vibrational phenomena. The research on the
4.0/).
misalignment of all kinds of couplings showed that the misalignment of the gear coupling

Machines 2021, 9, 251. https://doi.org/10.3390/machines9110251 https://www.mdpi.com/journal/machines


Machines 2021, 9, 251 2 of 21

will lead to a negative value of the angular meshing stiffness, resulting in the instability
of the rotor system [3]. Several harmonic frequencies of torsional vibration excited by the
misalignment of coupling, which is mainly composed of 1X, 2X and 3X components, are
observed [4]. However, the misalignment fault cannot be determined by the doubling
frequency component only. In the locked state, the phase difference in the same direction
perpendicular to the axis on both sides of the coupling is also an important distinguishing
condition for judging the misalignment type of the gear coupling [5]. The multiple doubling
frequencies of the rotor system is caused by the misalignment moment of hook coupling
angle [6]. Misalignment of the flange coupling can lead to an odd number of doubling
frequencies [7]. Supporting misalignment mainly refers to the misalignment caused by
bearing manufacturing, installation, and other reasons, among which the research on the
misalignment of sliding bearing is relatively sufficient. Huber et al. [8] proposed that if
the journal and the bearing bush axis are not aligned, the bearing edge load would be
concentrated and the operation will be unstable. According to the characteristics of long
necked sliding bearing, Arumugam [9] obtained that the misalignment of bearing can
reduce the oil film thickness, and increase the friction and damping of system. Park [10]
put forward that the change of misalignment angle has a direct impact on the film thickness
and pressure distribution.
For the parallel multi rotor system, which is represented by the dual-rotor system,
the research work mainly focuses on the misalignment of the coupling and the non-
concentricity between the rotors. Han et al. [11] put forward six modes for the advanced
turbofan engine rotor system structural features, namely, the rotor system supporting
non-concentricity of the fan section, the sleeve gear coupling misalignment of the low-
pressure rotor system, the supporting non-concentricity of the high-pressure rotor system,
the supporting non-concentricity of a high-pressure rotor system, the non-concentricity
of inner and outer rotors, and so on. They summarized that the rotor system would
produce lateral and axial vibration caused by the sleeve gear coupling misalignment
and the supporting non-concentricity of the roller bearing. There would be frequency
conversion and frequency doubling components, but no proportional relationship with the
value of misalignment. A basic dual-rotor system contains gas generator rotor and power
turbine rotor. Both are connected by coupling which will cause the misalignment of the
dual-rotors in the form of deflection angle by the coupling misalignment. For instance, the
rigid gear sleeve coupling is a typical component that causes the misalignment of the rotor
system. Ma et al. [12] conducted modeling and calculation of the sleeve gear coupling.
Finding that the stiffness of the coupling would change, and stability of the coupling would
be lost when the coupling is misaligned.
The non-concentricity may affect the dynamic characteristics of the rotor system and the
fatigue failure of the bearing. Wu et al. [13] transformed the non-concentricity of supporting
into the clearance model of bearing, and the non-concentricity of supporting was one of the
important reasons leading to the fatigue failure of rolling bearing. Based on the flexible rotor
system with multiple supports, Zhang et al. [14] obtained that the dynamic effect of the non-
concentricity of the bearing varies with the change of the structural characteristics of the rotor,
it may bring additional excitation frequency to the system. Xu et al. [15] studied the influence
of the high-pressure rotor supporting non-concentricity on the dynamic characteristics
of the double rotor system, which indicated that the non-concentricity would create a
new resonance region in the amplitude frequency response of the system. Li et al. [16]
established a dynamic model of a dual-rotor with intermediate support, which indicated
that the vibration characteristics of the low-pressure rotor would be transmitted to high-
pressure rotors. Lu et al. [17] established a dynamic model for a dual-rotor-bearing system
with flexible coupling misalignment faults in the low-pressure rotor system. The results
show that the vibration characteristics of misalignment faults in the low-pressure rotor can
be transmitted to the high-pressure rotor and can be used to diagnose the misalignment
faults. By numerical calculation, Feng et al. [18] showed that the non-concentricity of
the support bearing may lead to the occurrence of 2X frequency components in high and
coupling misalignment. It was found that the conditions leading to the instability of
periodic solution are the period doubling bifurcation and the secondary Hopf bifurcation.
Song [21] studied the vibration mechanism of the two rotors non-concentricity, and
pointed out that the non-concentricity fault of the high-pressure rotor is the dominant
Machines 2021, 9, 251 3 of 21
fault feature of the non-concentricity. Abdou [22] studied the influence of rotor
misalignment on stability of finite width radial plain bearing, it is concluded here that
increasing the value of misalignment
low-pressure rotors. Liudegree
et al. [19]yields to an
established theincrease in the mechanical
non-concentricity value of model critical
of the multi span flexible rotor system under the
bearing stability limit for a given value of misalignment direction angle. The critical non-concentricity excitation, which
indicated that the non-concentricity would bring additional unbalanced excitation to the
stability number increased
rotor system.asLithe et al.misalignment
[20] established a direction
dynamic model angle decreased
for an and/or
offset-disc rotor system the
steady state eccentricity ratio increased.
with a mechanical gear coupling, Santo
which[23]takes studied the effect
into consideration of rotor–tower
the nonlinear restoring
interaction, tilt angle,forceand
of rotor
yaw support and the effect
misalignment onofthecoupling
aeromisalignment.
elasticity ofItawas large found that the
horizontal
conditions leading to the instability of periodic solution are the period doubling bifurcation
axis wind turbine with composite blades. Sinha [24] proposed a method to estimate the
and the secondary Hopf bifurcation. Song [21] studied the vibration mechanism of the
unbalance and misalignment of a flexible and
two rotors non-concentricity, rotating
pointedmachine
out that thefrom a single operation.
non-concentricity Chen
fault of the high-
pressure rotor is the dominant fault feature of the
et al. [25] proposed a method to suppress the vibration of the multi-stage rotor, a non-concentricity. Abdou [22] studied
the influence of rotor misalignment on stability of finite width radial plain bearing, it is
simplified four-stage high-pressure rotor system was assembled according to the optimal
concluded here that increasing the value of misalignment degree yields to an increase in the
assembly angles calculated in the
value of critical simulations.
bearing stability limit for a given value of misalignment direction angle. The
In this paper, acritical stability number
dual-rotor systemincreased as the misalignment
of an aero-engine direction
is taken as theangle decreasedobject
research and/orto
the steady state eccentricity ratio increased. Santo [23] studied the effect of rotor–tower
study the influence of the non-concentricity induced by the assembly process on the
interaction, tilt angle, and yaw misalignment on the aero elasticity of a large horizontal
vibration characteristics
axis wind ofturbine
the system.
with compositeThree kinds
blades. of [24]
Sinha non-concentricity
proposed a method modes to estimate are
the unbalance and misalignment of a flexible rotating
investigated, namely parallel non-concentricity, front angle non-concentricity and rear machine from a single operation.
Chen et al. [25] proposed a method to suppress the vibration of the multi-stage rotor, a
angle non-concentricity. Based on the three-dimensional solid finite element model, the
simplified four-stage high-pressure rotor system was assembled according to the optimal
influence of the non-concentricity magnitude
assembly angles calculated on the vibration characteristics of the dual-
in the simulations.
rotor system under theInthree this paper,
kinds a dual-rotor system of an aero-engine
of non-concentricity modes are is taken as the research
investigated in object
detail.
to study the influence of the non-concentricity induced by the assembly process on the
The reasonable control suggestions for the non-concentricity in the assembly process are
vibration characteristics of the system. Three kinds of non-concentricity modes are inves-
given without changing tigated,the order
namely of the
parallel vibration modals
non-concentricity, of the
front angle system.
non-concentricity and rear angle
non-concentricity. Based on the three-dimensional solid finite element model, the influence
of the non-concentricity magnitude on the vibration characteristics of the dual-rotor system
2. FE Modeling and Experimental Set-Up of the Dual-Rotor System
under the three kinds of non-concentricity modes are investigated in detail. The reasonable
2.1. The Model of the Dual-Rotor
control suggestionsSystem
for the non-concentricity in the assembly process are given without
changing the order of the vibration modals of the system.
In this paper, the dual-rotor system of an aero engine is taken as the research object,
as shown in Figure2.1.FEItModeling
consistsand of Experimental
a high-pressureSet-Up rotor,
of the Dual-Rotor System rotor, coupling,
a low-pressure
2.1. The Model of the Dual-Rotor System
bearing and other structures. It includes six supporting points, in which the No. 1, No. 2
In this paper, the dual-rotor system of an aero engine is taken as the research object,
and No. 6 bearingsasbelong to low-pressure rotor bearings; the No. 1 and No. 2 bearings
shown in Figure 1. It consists of a high-pressure rotor, a low-pressure rotor, coupling,
are fixed between the outer
bearing and casing and theIt includes
other structures. low-pressure rotor;points,
six supporting the No. 6 bearing
in which the No. 1,isNo.
fixed
2
and No. 6 bearings belong to low-pressure rotor bearings; the No.
be-tween the inner casing and the low-pressure rotor; the No. 3 and No. 5 bearings belong 1 and No. 2 bearings are
fixed between the outer casing and the low-pressure rotor; the No. 6 bearing is fixed be-
to the high-pressure rotor and are fixed between the inner casing and the high-pressure
tween the inner casing and the low-pressure rotor; the No. 3 and No. 5 bearings belong to
rotor; and the No. 4thebearing is the
high-pressure intermediate
rotor bearing,
and are fixed between the which is fixed
inner casing and thebetween the rotor;
high-pressure high-
and the No.rotors.
pressure and low-pressure 4 bearing is the intermediate bearing, which is fixed between the high-pressure
and low-pressure rotors.

Figure 1. Supporting structure of dual-rotor system.


Figure 1. Supporting structure of dual-rotor system.
The actual structure is simplified, the bolt connection and coupling connection are
simplified into adhesive structure; the position of tenon, chamfer and comb disc are
alent; the blade is treated by the method of equivalent mass center, and the so
Machines 2021, 9, x FOR PEER REVIEW 4 of 21
element model is established [26]. Among them, solid element solid185 is use
high-pressure rotor, low-pressure rotor and gearbox, and the support structure
fied as linear spring element combin14. According to the empirical calculation
Machines 2021, 9, 251 The actual structure is simplified, the bolt connection and coupling connection
4 of 21 are
the linearization results of the stiffness parameters of the support structure are
simplified into adhesive structure; the position of tenon, chamfer and comb disc are equiv-
The parameter results are added to the finite element linear element structure,
alent; the blade is treated by the method of equivalent mass center, and the solid finite
a
in Figure
element model2.is established [26]. Among them, solid element solid185 is used for the
equivalent; the blade is treated by the method of equivalent mass center, and the solid finite
high-pressure
element model rotor, low-pressure
is established [26]. rotor
Amongandthem,
gearbox,
solid and the support
element structure
solid185 is used foristhe
simpli-
fied as linear spring element combin14. According to the empirical calculation
high-pressure rotor, low-pressure rotor and gearbox, and the support structure is simplified formula,
theaslinearization
linear springresults
elementofcombin14.
the stiffness parameters
According to theof the support
empirical structure
calculation are obtained.
formula, the
Thelinearization
parameterresults
resultsofare
the added
stiffnesstoparameters
the finite of
element linear
the support elementarestructure,
structure obtained. as
Theshown
parameter
in Figure 2. results are added to the finite element linear element structure, as shown in
Figure 2.

Figure 2. Finite element model of dual-rotor system.

The model includes high-pressure rotor, low-pressure rotor, fulcrum beari


mediate bearing, gearbox and other structures, totaling 582,814 elements and
Figure
Figure 2. Finite
2. Finite elementmodel
element model of
of dual-rotor
dual-rotor system.
system.low-pressure compressor disk, low-press
nodes. The low-pressure rotor includes
pressor
The modelshaft, low-pressure
includes turbine
high-pressure disk and low-pressure
rotor, low-pressure turbineinterme-
rotor, fulcrum bearing, shaft, totalin
Thebearing,
diate modelgearbox
includes high-pressure
and other structures,rotor, low-pressure
totaling 582,814 rotor,and
elements fulcrum bearing,
9,000,000 nodes. inter-
elements and 83,046 nodes. The high-pressure rotor includes high-pressure com
mediate bearing, gearbox and other structures, totaling 582,814 elements
The low-pressure rotor includes low-pressure compressor disk, low-pressure compressor and 9,000,000
disk, high-pressure
nodes. compressor shaft and high-pressure disk,compressor shaft. T
shaft,The low-pressure
low-pressure rotor
turbine diskincludes low-pressure
and low-pressure compressor
turbine shaft, totaling low-pressure
48,964 elements com-
47,328
pressor units
shaft,nodes.
and 83,046 and 76,731
low-pressure nodes in
turbine disk
The high-pressure the
rotor and structure of
low-pressure
includes pressure
high-pressureturbine turbine disk
shaft, disk,
compressor totalingand high-
high-48,964
turbine
pressure
elements and shaft,
compressor
83,046486,103 units
shaft and
nodes. and 8,840,227
Thehigh-pressure
high-pressure nodes
compressor
rotor in thehigh-pressure
shaft.
includes structure
There of compressor
are 47,328gearbox
units and
and 76,731 nodes in the
ries.
disk, high-pressure structure ofshaft
compressor pressure
andturbine disk and high-pressure
high-pressure compressor turbine
shaft. shaft,
There are
486,103
47,328 units
units and and 8,840,227
76,731 nodes nodes in the
in the structure
structure ofofpressure
gearbox turbine
and accessories.
disk and high-pressure
2.2.
turbine
2.2. Experimental
shaft, 486,103
Experimental Setof up
units
Set Up the of
and the Dual-Rotor
System System
8,840,227
Dual-Rotor nodes in the structure of gearbox and accesso-
ries. Taking
Taking a dual-rotor
a dual-rotor aero-engine
aero-engine as the experimental
as the experimental object,
object, two motors aretwo
usedmotors
to ar
drive
drivethethe
high- and low-pressure
high- rotors torotors
and low-pressure rotate. Its
to structure is shown
rotate. Its in Figure
structure 3.
is shown in Figur
2.2. Experimental Set up of the Dual-Rotor System
Taking a dual-rotor aero-engine as the experimental object, two motors are used to
drive the high- and low-pressure rotors to rotate. Its structure is shown in Figure 3.

Figure
Figure 3. Aero
3. Aero Engine
Engine Test Bench.
Test Bench.

Figure 3.The
Aero Engine Test
acceleration Bench.
sensors are set on the aero-engine. The natural frequency and mode
The acceleration sensors are set on the aero-engine. The natural frequency a
of the dual-rotor system are measured by the hammer method. The testing principle of
itsof theacceleration
The dual-rotor
vibration system
sensors
characteristics areset
isare
shownmeasured
on by theacceleration
hammer
the aero-engine.
in Figure 4. The method.
The natural The
frequency
sensors are testing princ
and mode
arranged
vibration
of the characteristics
dual-rotor is shown
system are measured in Figure
by the hammer4.method.
The acceleration
The testing sensors
principleare arra
of its
vibration characteristics is shown in Figure 4. The acceleration sensors are arranged on
Machines 2021, 9, x FOR PEER REVIEW 5 of 21
Machines 2021, 9, 251 5 of 21

the dual-rotor system of the aero-engine. Three measuring points are arranged
Machines 2021, 9, x FOR PEER REVIEW 5 of 21 on the low
on the dual-rotor system of the aero-engine. Three measuring points are arranged on the
pressure compressor, two measuring points on the high-pressure compressor, and one
low-pressure compressor, two measuring points on the high-pressure compressor, and one
measuring pointisisarranged
measuring point arranged at the
at the rearrear support
support of theoflow-pressure
the low-pressure rotor.
rotor. The The natura
natural
frequency
the
frequency and
dual-rotor
and modal
system
modal tests
oftests
the areare completed
aero-engine. Three
completed by determining
bymeasuring points
determining the
the are hammer
arranged
hammer hitting
on the
hitting low-onpoint
point the on the
pressure
rotor compressor,
structure
rotor structure ofof two
the
the measuring points on the high-pressure compressor, and one
engine.
engine.
measuring point is arranged at the rear support of the low-pressure rotor. The natural
frequency and modal tests are completed by determining the hammer hitting point on the
rotor structure of the engine.

Figure 4.4.Schematic
Figure Schematicdiagram
diagramof dual-rotor modal
of dual-rotor test for
modal aero-engine
test whole system.
for aero-engine whole system.
Figure 4. Schematic diagram of dual-rotor modal test for aero-engine whole system.
An LMS system was used in the testing process, and the test equipment used is shown
AnAnLMS
in Figure
LMS system was used in the testing process, and the test equipment used is
5. system was used in the testing process, and the test equipment used is
shown
shown in in Figure
Figure 5. 5.

(a) (b)

(a) (b)

(c) (d)
Figure
Figure5.5.LMS
LMSmodal testing
modal and
testing analysis
and equipment:
analysis (a) LMS
equipment: datadata
(a) LMS acquisition device;
acquisition (b) LMS
device; (b) LMS
modal analysis system; (c) KISTLER acceleration sensor; (d) Force hammer.
modal analysis system; (c) KISTLER acceleration sensor; (d) Force hammer.

(c) (d)
Figure 5. LMS modal testing and analysis equipment: (a) LMS data acquisition device; (b) LMS
modal analysis system; (c) KISTLER acceleration sensor; (d) Force hammer.
Machines 2021, 9, 251 6 of 21
Machines 2021, 9, x FOR PEER REVIEW 6 of 21

The measurement points of the acceleration sensor on the aero-engine rotor system
The measurement points of the acceleration sensor on the aero-engine rotor system
are shown in Figure 6.
are shown in Figure 6.

(a) (b)

(a) (b)

(c) (d)
Figure 6. Sensor(c) measuring point arrangement for dual-rotor (d) system of aero-engin
pointsSensor
Figure of the
Figure6.6.Sensor
low-pressure
measuring
measuringpoint
compressor;
pointarrangement
arrangementfor
(b) Measuring
fordual-rotor
dual-rotorsystem
points of(a)the
systemofofaero-engine:
aero-engine:(a)
high-pressure
Measuring
Measuring
Measuring
points of the points
low-pressure of the rear
compressor; supporting
(b) Measuring of the
points of low-pressure
the high-pressure turbine;
compressor;
points of the low-pressure compressor; (b) Measuring points of the high-pressure compressor; (d)
(c) Measur
Measuring points of the rear supporting of the low-pressure turbine; (d) Measuring points of the
(c)force hammer.
Measuring points of the rear supporting of the low-pressure turbine; (d) Measuring points of the
force hammer.
force hammer.
LMS
LMSLMS
systemsystem
system isisused istoestablish
usedto used toa aestablish
establish simplified abeam
simplifiedbeam simplified
element beam
elementmodel.
model.Theelement
Thehigh-
high-and
andmodel.
low-
low- The
pressure
pressurerotors
pressure are
rotors
rotors simplified into
are simplified
are simplified linear
into linear beambeam elements
intoelements
linearand and
beam connected
elements
connected by two
by two andinterme-
connected b
intermediate
diate bearings.
bearings. The The
model model structure
structure is shown
is shown in Figure
in Figure 7. 7.
diate bearings. The model structure is shown in Figure 7.

Figure 7. Beam element model established by LMS (L1, L2, L3 and L4 are low-pressure rotor meas-
uring points, H1 is high-pressure rotor measuring points, yellow line is low-pressure rotor, red line
Figure 7. Beam element model established by LMS (L1, L2, L3 and L4 are low-pressure rotor
Figure 7. Beam
is high-pressure rotor). element model established by LMS (L1, L2, L3 and L4 are low-pres
measuring points, H1 is high-pressure rotor measuring points, yellow line is low-pressure rotor, red
uring
line points, H1
is high-pressure is high-pressure rotor measuring points, yellow line is low-pressu
rotor).
is Taking the second-order modal shape as an example, the comparison between the
high-pressure rotor).
modal shapes obtained by testing and the simulation results is shown in Figure 8.

Taking the second-order modal shape as an example, the compariso


modal shapes obtained by testing and the simulation results is shown in F
Machines 2021, 9, 251 7 of 21

Machines 2021, 9, x FOR PEER REVIEW 7 of 21


Machines 2021, 9, x FOR PEER REVIEW 7 of 21
Taking the second-order modal shape as an example, the comparison between the
modal shapes obtained by testing and the simulation results is shown in Figure 8.

(a) (b)
(a) (b)
Figure 8. The second-order modal shape comparisons between experiments and FEM analyses: (a)
Figure
Figure 8.8.The
Thesecond-order
The second-order
second-order modal
modal shape
modal shape
of shape comparisons
FEM analyses; between
(b) The
comparisons experiments
second-order
between modal
experiments and
andshape
FEM FEM
of analyses: (a)
LMS.
analyses:
The second-order
(a) The modal
second-order modalshape
shapeofofFEM
FEManalyses; (b)The
analyses; (b) Thesecond-order
second-order modal
modal shape
shape of LMS.
of LMS.
Table 1 shows the comparison of the FEM analysis of the model. The relative errors
Table
of the FEM
Table 11 shows
shows the
analyses comparison
theand the first of
comparison thethe
three
of FEM
natural
FEM analysis of the
frequencies
analysis ofmodel.
ofthe The relative
themodel.
experimental errors
resultserrors
The relative are
ofobtained.
the FEM Theanalyses and
minimum the first
relative three natural
error of frequencies
the first of the
order natural experimental
frequency results are
is 6.9%,results
and the
of the FEM analyses and the first three natural frequencies of the experimental are
obtained. The minimum relative error of the first order natural frequency is 6.9%, and
maximum relative error of the third order is 16.7%. Through comparison,
obtained. The minimum relative error of the first order natural frequency is 6.9%, and the it is found that
the maximum relative error of the third order is 16.7%. Through comparison, it is found
the relative
maximum error between
relative error of experiments
theexperiments
third orderandisFEM analysis is small, which proves the accu-
that the relative error between and16.7%. Through
FEM analysis comparison,
is small, it is found
which proves the that
racy
the of the
relative model.
accuracy of error between experiments and FEM analysis is small, which proves the accu-
the model.
racy of the model.
Table 1. Natural frequencies comparisons between experiments and FEM analyses.
Table 1. Natural frequencies comparisons between experiments and FEM analyses.
Table 1. Natural
Order frequencies comparisons
Experiment between experiments
(Hz) and
Calculate FEM analyses. Error
(Hz)
Order Experiment (Hz) Calculate (Hz) Error
First
Order 123.1
Experiment 114.6 (Hz) 6.9%
First 123.1 (Hz) Calculate
114.6 6.9% Error
Second
Second 193.4
193.4 210.1 210.1 8.6% 8.6%
First 123.1 114.6 6.9%
Third
Third 307.6
307.6 256.2 256.2 16.7% 16.7%
Second 193.4 210.1 8.6%
Third 307.6 256.2 16.7%
3.3.The
TheDefinition
Definitionofof Three
Three Modes
Modes of Non-Concentricity
of Non-Concentricity
3.1.
3.1.The
TheCause
CauseofofMisalignment
Misalignment andand
Non-Concentricity
Non-Concentricity
3. The Definition of Three Modes of Non-Concentricity
There
Thereareare
many types
many of non-concentricity
types of a dual-rotor
of non-concentricity system.system.
of a dual-rotor The fault
Thetreefault
of the
tree of
3.1. The Cause
possible causes ofisMisalignment
shown in and9.Non-Concentricity
Figure
the possible causes is shown in Figure 9.
There are many types of non-concentricity of a dual-rotor system. The fault tree of
the possible causes is shown in Figure 9.

Figure9.9.Non-concentricity
Figure Non-concentricity fault
fault treetree of dual-rotor
of dual-rotor systems.
systems.

The
Themisalignment
misalignment in in
rotor systems
rotor cancan
systems be classified intointo
be classified two two types: parallel
types: misalign-
parallel misalign-
Figure
ment 9. Non-concentricity
and angle faultThe
misalignment. treecauses
of dual-rotor
of the systems.
parallel misalignment are manufacturing
ment and angle misalignment. The causes of the parallel misalignment are manufacturing
error,
error,installation
installation error and
error andstatic deformation
static deformation of the elastic
of the supports
elastic on both
supports sidessides
on both of theof the
The
rotor, whilemisalignment
the causes of in
the rotor
angle systems can be
misalignment classified
are into two
manufacturing types:
error, parallel
installation misalign-
error,
rotor, while the causes of the angle misalignment are manufacturing error, installation
ment and
and static angle misalignment.
deformation of the oneThesidecauses
elastic of the
support.
error, and static deformation of the one side elastic support.
parallel misalignment are manufacturing
error,From the perspective
installation error and
From the perspective
of the dual-rotor system of
static
of the deformation
dual-rotor system
structure described
thestructure
elastic in previous
supports
described oninbothsection,
sides of
previous the
sec-
the No.
rotor, 3 support
while the positionof
causes of the
the high-pressure
angle rotor is anare
misalignment elastic support. The No.
manufacturing 4 bearing
error, installation
tion, the No. 3 support
is an intermediate
position of with
bearing. Coupled
the high-pressure rotorit is an elastic support. The No. 4
error, and static deformation of the one the
sideNo. 5 bearing,
elastic support. can also be regarded as an
bearing is an intermediate bearing. Coupled with the No. 5 bearing, it can also be regarded
From
as an the support.
elastic perspective of the dual-rotor
In addition system structure
to the manufacturing errors ofdescribed in previousdue
support structures, sec-
tion, thedeformation
to the No. 3 support position
of the elasticof the high-pressure
supports, rotor is of
the elastic support anthe
elastic
frontsupport.
and rear The No. 4
support
bearing is an intermediate bearing. Coupled with the No. 5 bearing, it can also be
points of the high-pressure rotor is prone to static deformation, which leads to the decen- regarded
astraction
an elastic support.
of the In addition
front and to the
rear support manufacturing
points errors ofrotor.
of the high-pressure supportThestructures, due
concentricity
Machines 2021, 9, x FOR PEER REVIEW 8 of 2
rors at two supports, it will result in parallel misalignment.

3.2. Three Modes of the Non-Concentricity


of theThe
axis of the two rotorsin
non-concentricity cannot be accurately
dual-rotor guaranteed,
systems can and the
be categorized problem
into three of the con
Machines 2021, 9, 251 8 ofmodes.
21 Th
first mode is that the high-pressure rotor axis produces parallel decentractionanwith
centricity of the two rotors is prone to occur in the assembly process. If there is errortha
one support of
low-pressure the dual-rotor
rotor axis to formsystem,
parallelit non-concentricity,
will result in angleasmisalignment.
shown in FigureIf there aresec
10, the er
rors at two supports, it will result in parallel misalignment.
ond mode is that the high-pressure rotor produces decentraction with the low-pressur
elastic support. In addition to the manufacturing errors of support structures, due to the
rotor axis to
deformation form
of the front
elastic deflection
supports, anglesupport
the elastic non-concentricity,
of the front and asrear
shown in Figure
support points 11. Th
3.2. Three Modes of the Non-Concentricity
third
of mode is just rotor
the high-pressure opposite to the
is prone second
to static mode, as shown
deformation, in Figure
which leads to the 12.
decentraction
The
frontnon-concentricity
of the The main
and rear reason
support in dual-rotor
forpoints
parallel systems can
non-concentricity
of the high-pressure be categorized
ofThe
rotor. intoof
theconcentricity
dual-rotor isthree
that
the modes. Th
the elasti
axis
first
of the mode
supports at is
two rotorsthethat theand
front
cannot behigh-pressure
rear support
accurately rotor
of theaxis
guaranteed, andproduces
high-pressure
the problem parallel
rotor decentraction
have
of the of with th
static de-formatio
concentricity
low-pressure
the two same
at the rotor
rotors time
is prone axis
or the to
to occurform parallel
in thecomponents
support non-concentricity,
assembly process. as
have Ifmanufacturing shown
there is an errorand in Figure
at one support the
10,
installation sec
error
of
ond
at the dual-rotor
themode
sameistime.
thatsystem,
Astheshownit willinresult
high-pressureFigureinrotor
angle misalignment.
produces
10, Axis 1 is the axisIfof
there
the are
decentraction errors
with theatlow-pressur
high-pressure tworotor an
supports,
rotor axis it will resultfront
to form in parallel misalignment.
deflection angle non-concentricity, as shown in Figure 11. Th
Axis 2 is the axis of the low-pressure rotor. δ refers to the decentraction formed by th
third mode
parallel is just oppositeoftothethedual-rotor.
second mode, as shown in Figure 12.
3.2. Three non-concentricity
Modes of the Non-Concentricity
The main reason for parallel non-concentricity of the dual-rotor is that the elasti
The non-concentricity in dual-rotor systems can be categorized into three modes. The
supports at the front and rear support of the high-pressure rotor have static de-formatio
first mode is that the high-pressure rotor axis produces parallel decentraction with the
at the samerotor
low-pressure timeaxis
or to
theform
support components
parallel have as
non-concentricity, manufacturing
shown in Figureand installation
10, the second error
mode is that the high-pressure rotor produces decentraction with the low-pressure rotorrotor an
at the same time. As shown in Figure 10, Axis 1 is the axis of the high-pressure
Axis
axis to 2form
is the
frontaxis of the angle
deflection low-pressure rotor. δasrefers
non-concentricity, showntointhe decentraction
Figure 11. The thirdformed
mode by th
isparallel non-concentricity
just opposite of the as
to the second mode, dual-rotor.
shown in Figure 12.

Figure 10. Parallel non-concentricity of dual-rotor.

The main reason for the front deflection angle of the dual-rotor is that the rear sup
port of the high-pressure rotor has static deformation or the support component hav
manufacturing and installation errors. As shown in Figure 11, Axis 1 is the axis of th
high-pressure rotor, Axis 2 is the axis of the low-pressure rotor, and θ is the angle formed
Figure
by Parallel
the10.10.
Figure front non-concentricity
Parallel of dual-rotor.
non-concentricity
deflection of dual-rotor. of the dual-rotor.
angle non-concentricity

The main reason for the front deflection angle of the dual-rotor is that the rear sup
port of the high-pressure rotor has static deformation or the support component hav
manufacturing and installation errors. As shown in Figure 11, Axis 1 is the axis of th
high-pressure rotor, Axis 2 is the axis of the low-pressure rotor, and θ is the angle forme
by the front deflection angle non-concentricity of the dual-rotor.

Machines 2021, 9, x FOR PEER REVIEW 9 of 2

Figure 11.Front
Figure11. Frontdeflection angle
deflection non-concentricity
angle of dual-rotor.
non-concentricity of dual-rotor.

The main reason for the front deflection angle of the dual-rotor is that the elastic sup
ports at the front support of the high-pressure rotor have static deformation or the suppor
component have manufacturing and installation errors. As shown in Figure 12, Axis 1 i
the axis of the high-pressure rotor, Axis 2 is the axis of the low-pressure rotor, and θ is th
Figureformed
angle 11. Frontby
deflection
the rearangle non-concentricity
deflection of dual-rotor. of the dual-rotor.
angle non-concentricity

The main reason for the front deflection angle of the dual-rotor is that the elastic sup
Figure
ports 12.
Figure Rear
at12.the deflection
front
Rear angle
support
deflection non-concentricity
of
anglethe of dual-rotor.
high-pressure
non-concentricity rotor
of have static deformation or the suppor
dual-rotor.
component
The mainhave reasonmanufacturing and installationoferrors.
for parallel non-concentricity As shown
the dual-rotor in Figure
is that 12, Axis 1 i
the elastic
3.3. Realization
the axis at
supports of the of
thefront the Non-Concentricity
high-pressure rotor,of
and rear support Axis in the Dual-Rotor
the 2high-pressure System
is the axis ofrotor Model
the low-pressure rotor, and θ is th
have static de-formation
angle
at formed
the In thetime
same by
finite the
the rear
orelement deflection
model,
support theangle non-concentricity
high-pressure
components have of the
rotor is translated
manufacturing and dual-rotor.
upward
installation to produc
errors
at the same
parallel time. As shown
decentraction, in Figure
forming 10, Axisof
a model 1 is the axisnon-concentricity.
parallel of the high-pressureBy rotor and
controlling th
Axis 2 is the axis
displacement ofof
thethehigh-pressure
low-pressure rotor.
rotor, δtherefers to thenon-concentricity
parallel decentraction formed by the
of two axes is real
parallel
ized. Innon-concentricity
order to make the of the dual-rotor.
front deflection angle non-concentricity, the rear support of th
The main reason for the front deflection angle of the dual-rotor is that the rear support
high-pressure rotor is shifted upward to form the front deflection angle model of the dual
of the high-pressure rotor has static deformation or the support component have manufac-
rotor. The front deflection angle non-concentricity is shifted upward by controlling th
rear support of the high-pressure rotor. The front support of the high-pressure rotor i
deflected upward to make the rear deflection angle, forming a model of the rear deflectio
angle non-concentricity.
Machines 2021, 9, 251 9 of 21

turing and installation errors. As shown in Figure 11, Axis 1 is the axis of the high-pressure
rotor, Axis 2 is the axis of the low-pressure rotor, and θ is the angle formed by the front
deflection angle non-concentricity of the dual-rotor.
The main reason for the front deflection angle of the dual-rotor is that the elastic
supports at the front support of the high-pressure rotor have static deformation or the
support component have manufacturing and installation errors. As shown in Figure 12,
Axis 1 is the axis of the high-pressure rotor, Axis 2 is the axis of the low-pressure rotor, and
θ is the angle formed by the rear deflection angle non-concentricity of the dual-rotor.

3.3. Realization of the Non-Concentricity in the Dual-Rotor System Model


In the finite element model, the high-pressure rotor is translated upward to produce
parallel decentraction, forming a model of parallel non-concentricity. By controlling the
displacement of the high-pressure rotor, the parallel non-concentricity of two axes is
realized. In order to make the front deflection angle non-concentricity, the rear support of
the high-pressure rotor is shifted upward to form the front deflection angle model of the
dual-rotor. The front deflection angle non-concentricity is shifted upward by controlling
the rear support of the high-pressure rotor. The front support of the high-pressure rotor is
deflected upward to make the rear deflection angle, forming a model of the rear deflection
angle non-concentricity.

4. Influence of Non-Concentricity on Natural Frequency and Mode Shape of the


Dual-Rotor System
4.1. Natural Frequency and Mode Shape of the Dual-Rotor System without Non-Concentricity
For the established finite element model of the dual-rotor system, the natural fre-
quencies and mode shapes are calculated. The first five modes and corresponding natural
frequencies are obtained, which are listed in Table 2. The first modal is the local modal of
the high-pressure turbine; the second modal is the local modal of stage 1 disc of the low-
pressure turbofan rotor; the third modal is the overall bending modal of the high-pressure
rotor; the fourth modal is the local modal of stage 2 and stage 3 discs of the low-pressure
turbofan rotor; the fifth modal is the overall bending modal of the low-pressure rotor.

Table 2. The first five modals.

Order Natural Frequency (Hz) Mode Shape


1st 114.6 Local vibration mode of HP Turbine
2nd 210.1 Local vibration mode of LP Compressor
3rd 256.2 Bending vibration mode of HP Rotor
4th 284.7 Local vibration mode of LP Compressor
5th 327.8 Bending vibration mode of LP Rotor

4.2. The Influence of the Parallel Non-Concentricity on the Natural Frequency and Modal Shape of
the Dual-Rotor System
In this subsection, δ is selected in the range of 0 mm ~ 3 mm. Based on same rule,
the first five natural frequencies and modal shapes of the dual-rotor with parallel non-
concentricity are calculated. The change curves of the first five natural frequencies are
given out in Figure 13. Curves with different colors correspond to different orders of
natural frequency In Figure 13, the horizontal axis represents the different values of parallel
non-concentricity of the dual-rotor system. The vertical axis represents the corresponding
frequency of the dual-rotor system with set value parallel non-concentricity. With the
increase in δ, the 1st, 2nd and 4th orders of natural frequency do not change. While the 3rd
and 5th orders decrease with the increase in the δ. Thus, at points A and B, the frequency
orders of 2 and 3 and 4 and 5 interchange.
Machines 2021, 9, x FOR PEER REVIEW
Machines 2021, 9, 251 10 of 21

Figure 13. Variation trend of natural frequency of parallel non-concentricity in the range of 0 mm~3 mm.
Figure 13. Variation trend of natural frequency of parallel non-concentricity in the r
mm.More in detail, when δ increases from 0 to 3 mm, the 1st, 2nd, and 4th order natural
frequencies change little, while the 3rd and 5th order natural frequencies change obviously.
With the increase
More in indetail,
δ, the 3rd and 5th
when order naturalfrom
δ increases frequencies
0 to decrease.
3 mm, the When δ increases
1st, 2nd, and 4th
to 2.1 mm, the 3rd order natural frequency will be less than the 2nd order natural frequency.
frequencies
Thusm change little,
the corresponding 2nd andwhile themodals
3rd order 3rd and 5th order
interchange. When natural frequencies
δ continues to
ously. to
increase With the the
2.25 mm, increase in δ,
5th natural the 3rdwill
frequency and 5ththan
be less order natural
the 4th naturalfrequencies
frequency, dec
and the corresponding 4th and 5th modals will be interchanged.
increases to 2.1 mm, the 3rd order natural frequency will be less than the 2nd
In order to further explore the change of the 3rd order bending modal and 5th order
bending modalThusm
frequency. under thethe corresponding
parallel 2nd
non-concentricity, theand 3rd5thorder
3rd and order modals interchang
natural frequen-
tinues
cies of theto increase
system to 2.25Following
are extracted, mm, the 5th natural
Equation frequency
(1), the change rates canwill be less than t
be calculated.
Selecting 0.5 mm, 1 mm, 1.5 mm, 2 mm, 2.5 mm and 3 mm, the
frequency, and the corresponding 4th and 5th modals will be interchanged change rates of the 3rd
order are obtained. The calculation results are shown in Table 3.
In order to further explore the change of the 3rd order bending modal
bending modal under the parallel ω = ( f i − fnon-concentricity,
0 )/ f 0 , the 3rd and(1)5th ord
quencies
where of thethe
f 0 represents system
naturalare extracted,
frequency Following
of the system Equation
when there (1), the
is no parallel change
offset. fi ra
culated.the
represents Selecting 0.5 mm,
natural frequency 1 mm,
under 1.5 mm,
the different 2 non-concentricity.
parallel mm, 2.5 mm and 3 mm, the c
ω represents
the
thechange rate of natural
3rd order frequency.The calculation results are shown in Table 3.
are obtained.
ω
Table 3. The 3rd order natural frequency error under different=parallel
( f i −non-concentricity
f 0 ) / f 0 , of the 3rd order.
Decentraction (mm) Natural Frequency (Hz) Error
where f0 represents the natural frequency of the system when there is no p
0 256.17 0
represents0.5the natural frequency under the different parallel
253.11 1.2% non-concentr
1
sents the change rate of natural 245.45
frequency. 4.2%
1.5 230.28 10.1%
According
2 to the analysis 210.92
of the results listed in17.7% Table 3, the chang
2.5 191.61 25.2%
change rate of the natural frequency of the bending modal of the 3rd order
3 176.84 31%
the increase in the parallel non-concentricity. When the parallel non-concen
3 mm, the change
According rate ofofthe
to the analysis thenatural frequency
results listed reaches
in Table 3, theamount
the change maximum
and valu
change rate of the natural frequency of the bending modal of the 3rd order increase with
the increase
Table in the
3. The 3rdparallel
ordernon-concentricity.
natural frequencyWhen the parallel
error non-concentricity
under different parallel reaches
non-concen
3order.
mm, the change rate of the natural frequency reaches the maximum value of 31%.
The 3rd order modals with different parallel non-concentricity are extracted. The
comparison of mode shapes is shown in Figure 14. The results show that with the increase
Decentraction (mm) Natural Frequency (Hz)
in parallel non-concentricity, the 3rd order mode of the dual-rotor changes its shape from
0 low-pressure turbofan to the bending vibration256.17
local vibration mode of the mode of the low-
0.5 253.11
1 245.45
1.5 230.28
FOR PEER REVIEW
Machines 2021, 9, 251 11 of 21 11 of 21

increase in parallelpressure rotor. The deformation


non-concentricity, while theof bending
the high-pressure
level ofrotor
the decreases with the
low-pressure increase in
rotor
parallel non-concentricity, while the bending level of the low-pressure rotor increases with
increases with the increase in parallel non-concentricity. The maximum deformation po-
the increase in parallel non-concentricity. The maximum deformation position is gradually
sition is gradually transferred from the high-pressure rotor to the low-pressure rotor, and
transferred from the high-pressure rotor to the low-pressure rotor, and the deformation of
the deformation ofthethetransformed
transformed modal
modal is larger
is larger and obvious.
and more more obvious.

(a) (b)

(c) (d)

(e) (f)
Figure
Figure 14.3rd
14. The The 3rd modals
order order modals forin
for contrast contrast
differentin different
parallel parallel non-concentricity:
non-concentricity: (a)δ δ= =1.00.5
(a) δ = 0.5 mm; (b) mm;mm;
(c) δ =
(b) δ(d)
1.5 mm; = 1.0 mm;
δ = 2.0 mm(c)(e)δ δ==1.5
2.5mm; (d)δδ==3.0
mm; (f) 2.0 mm (e) δ = 2.5 mm; (f) δ = 3.0 mm.
mm.

Similarly, we chose 0.5 mm, 1 mm, 1.5 mm, 2 mm, 2.5 mm, and 3 mm as the repre-
Similarly, we chose 0.5 mm, 1 mm, 1.5 mm, 2 mm, 2.5 mm, and 3 mm as the repre-
sentative parameters to extract the bending vibration mode and its corresponding nat-
sentative parametersural frequencythe
to extract of bending vibration
the 5th order. moderates
The change and of
itsnatural
corresponding natural
frequencies with different
frequency of the 5th order. Thevalues
decentration change
arerates of natural
calculated frequencies
and listed with
in Table 4. different
With decen-
the increase in parallel
tration values are calculated and listed in Table 4. With the increase in parallel non-con-
centricity, the change rate of natural frequency increases gradually. When the parallel
non-concentricity reaches 3 mm, the change rate of natural frequency reaches the maxi-
mum value of 14.5%.
FOR PEER REVIEW 12 of 21

Machines 2021, 9, 251 12 of 21

Table 4. The 5th order natural frequency error under different parallel non-concentricity of the 3rd
order. non-concentricity, the change rate of natural frequency increases gradually. When the
parallel non-concentricity reaches 3 mm, the change rate of natural frequency reaches the
Decentraction (mm) Natural Frequency (Hz) Error
maximum value of 14.5%.
0 327.84 0
0.5 Table 4. The 5th order natural frequency error under different parallel non-concentricity
322.48 1.6%of the 3rd order.
1 Decentraction (mm) 312.30
Natural Frequency (Hz) 4.7%
Error
1.5 0 296.95 327.84 9.4% 0
2 0.5 286.19 322.48 12.7%
1.6%
1 312.30 4.7%
2.5 1.5
282.74 296.95
13.8%
9.4%
3 2 280.40 286.19 14.5%
12.7%
2.5 282.74 13.8%
3 280.40 14.5%
The comparison of 5th modal shapes with different decentration is shown in Figure
15. It is found that the 5th modal of the dual-rotor system changes from the bending modal
The comparison of 5th modal shapes with different decentration is shown in Figure 15.
of the low-pressure rotor to the bending modal of the high-pressure rotor with the increase
It is found that the 5th modal of the dual-rotor system changes from the bending modal of
in the parallel non-concentricity. Thistokind
the low-pressure rotor of bending
the bending modal
modal of is similar to
the high-pressure thewith
rotor bending
the increase in
modal of the 3rd order high-pressure
the parallel rotor, This
non-concentricity. but there
kind ofare somemodal
bending differences.
is similarIttoisthe
different
bending modal
from the bending ofmodal the 3rdof the high-pressure
order dual-rotor caused
rotor, butby theare
there bending of the high-pressure
some differences. It is different from the
bending
rotor, it is the reverse modalofofthe
bending thehigh
dual-rotor caused by the bending
and low-pressure rotors. of the high-pressure
Additionally, withrotor,
the it is the
increase in the parallel non-concentricity, the reverse bending of the two rotors becomes in the
reverse bending of the high and low-pressure rotors. Additionally, with the increase
parallel non-concentricity, the reverse bending of the two rotors becomes more and more
more and more severe. Compared with the bending modal of the 3rd order high-pressure
severe. Compared with the bending modal of the 3rd order high-pressure rotor, this kind
rotor, this kind of bending
of bendingmodal
modal has
has aamore
moreserious
seriousandand destructive
destructive effecteffect
on theon
rotor.the rotor.

(a) (b)

(c) (d)
Figure 15. Cont.
FOR PEER REVIEW
Machines 2021, 9, 251 13 of 21 13 of 21

(e) (f)
Figure
Figure 15. 5th
15. The Theorder
5th modals
order modals for in
for contrast contrast
differentinparallel
different parallel non-concentricity:
non-concentricity: (a)δδ==1.0
(a) δ = 0.5 mm; (b) 0.5mm;
mm;(c) δ =
1.5 (b)
mm;δ(d)
= 1.0
δ = mm; (c) (e)
2.0 mm; δ =δ 1.5 mm;
= 2.5 mm;(d)
(f) δδ == 3.0
2.0mm.
mm; (e) δ = 2.5 mm; (f) δ = 3.0 mm.

4.3. The Influence of the Front Deflection Angle Non-Concentricity on the Natural Frequency and
4.3. The Influence of the
ModalFront
ShapeDeflection Angle Non-Concentricity
of the Dual-Rotor System on the Natural Frequency
and Modal Shape of the Dual-Rotor System
Following same analysis method, the angle non-concentricity is selected within the
Following same analysis
range of 0–0.26 ◦ to investigate
method, the angle
thenon-concentricity is selected
effects on the dual-rotor system.within the
The corresponding
range of 0°–0.26° torelationship
investigatebetween the angle
the effects on non-concentricity
the dual-rotor and the parallel
system. non-concentricity can be
The corresponding
relationship betweengotthe
by Equation (2).
angle non-concentricity and the parallel non-concentricity can
δ = Lsinθ (2)
be got by Equation (2).
L is the distance from the front supporting of the high-pressure to the intermediate bearing.
δ = L sinθ
Taking the angle of 0.02◦ as the spacing, the value corresponding relationship (2) between
θ and δ is found out. The results are shown in Table 5. Selecting the angle values of 0.05◦ ,
L is the distance from the front supporting of the high-pressure to the intermediate
0.1◦ , 0.15◦ , 0.2◦ , 0.25◦ and 0.3◦ , the natural frequency curves of each order with respect
bearing. of angle value are shown in Figure 16. In Figure 16, the horizontal axis represents the
Taking the angle angle ofof0.02° as the
the front spacing,
deflection anglethenon-concentricity,
value corresponding relationship
and the vertical be-
axis represents the
tween θ and δ is found out. The frequency.
corresponding results areAshown in Table
is the modal 5. Selecting
interchange point ofthe angle
orders values
2 and 3; B isof
the modal
interchange point of orders 4 and 5.
0.05°, 0.1°, 0.15°, 0.2°, 0.25° and 0.3°, the natural frequency curves of each order with re-
spect of angle value areInshown the case of the front deflection angle, non-concentricity of the dual-rotor system
in Figure 16. In Figure 16, the horizontal axis represents
exists, and the natural frequencies of the 1st, 2nd, and 4th orders have not changed signifi-
the angle of the front deflection angle non-concentricity, and the vertical axis represents
cantly, while the natural frequencies of the 3rd and 5th orders have changed significantly.
the corresponding frequency.The ruleAisis thethe
that modal
naturalinterchange
frequencies of point of orders
the 3rd and 5th 2orders
and 3; B is the
appear to decrease,
modal interchange as point of orders
the angle of the4front
and deflection
5. angle non-concentricity increases. When it increases
to 0.18◦ , the natural frequency of the 3rd order will be smaller than that of the 2nd order.
The between
Table 5. The relationship θ and
corresponding δ .and 3rd mode shapes will be interchanged. At this time, δ is
2nd
approximately 2.1 mm, which agrees with the parallel non-concentricity of the dual-rotor.
θ (°) When the δangle (mm) δ (mm)
value continues to increase θ natural
to 0.22◦ , the 5th order (°) frequency will
0 be smaller than0 the 4th order natural frequency,
0.16 and the 4th and 5th
1.874 order modals will be
interchanged. The 2nd and 3rd order modal interchange position is met earlier than the 4th
0.02 0.234 0.18 2.108
and 5th order modal interchange position. Calculating the natural frequency change rates
0.04 of the bending 0.468vibration mode of the 3rd 0.20 2.342rotor with different
order of the high-pressure
0.06 0.703
front deflection angle non-concentricity, 0.22
the calculation results2.576
are shown in Table 6. As
0.08 the angle value0.937 of the front declination angle
0.24 non-concentricity 2.811increases, the change rate
of the natural frequency gradually increases accordingly. When the front deflection angle
0.10 1.171 0.26 3.045
reaches 0.3◦ , the natural frequency change rate reaches a maximum of 33.8%.
0.12 1.405 0.28 3.279
0.14 1.640 0.30 3.513

In the case of the front deflection angle, non-concentricity of the dual-rotor system
exists, and the natural frequencies of the 1st, 2nd, and 4th orders have not changed signif-
icantly, while the natural frequencies of the 3rd and 5th orders have changed significantly.
Machines 2021, 9, 251 14 of 21

Table 5. The relationship between θ and δ.

θ (◦ ) δ (mm) δ (mm) θ (◦ )
0 0 0.16 1.874
0.02 0.234 0.18 2.108
0.04 0.468 0.20 2.342
0.06 0.703 0.22 2.576
0.08 0.937 0.24 2.811
0.10 1.171 0.26 3.045
Machines 2021, 9, x FOR PEER REVIEW
0.12 1.405 0.28 3.279
0.14 1.640 0.30 3.513

Figure
Figure 16.16. Variation
Variation trendtrend of natural
of natural frequency
frequency of frontangle
of front deflection deflection angle non-concentric
non-concentricity in the
◦.
range
range of of 0°–0.3°.
0–0.3

Table 6.The rule israte


The change that thenatural
of the natural frequencies
frequency of the
3rd and 5th orders appear to d
of the 3rd order.
as the angle of the◦ front deflection angle non-concentricity increases. When it inc
Decentraction ( ) Natural Frequency (Hz) Error
0.18°, the natural frequency of the 3rd order will be smaller than that of the 2nd or
1.9%At this time, δ is
0 256.17 0
corresponding0.05 2nd and 3rd mode shapes 251.32 will be interchanged.
imately 2.10.1 mm, which agrees with 243 the parallel non-concentricity
5.1% of the dual-roto
0.15 224.60 12.3%
the angle value
0.2
continues to increase 202.06
to 0.22°, the 5th order21.1%
natural frequency
smaller than 0.25 the 4th order natural frequency,
187.14 and the 4th and
27% 5th order modal
0.3 169.64 33.8%
interchanged. The 2nd and 3rd order modal interchange position is met earlier
4th and 5th order modal interchange position. Calculating the natural frequency
With
rates of each set angle value
the bending of the front
vibration mode deflection
of theangle non-concentricity,
3rd order the 3rd order rotor
of the high-pressure
modal is extracted. As shown in Figure 17, the modals with different angle values are
ferent front deflection angle non-concentricity, the calculation results are shown
compared and analyzed. When the front deflection angle non-concentricity increases, the
6. As shape
modal the angle
changevalue
fromofthe
the front declination
bending angle non-concentricity
of the high-pressure rotor to the bending increases,
of the th
rate of the natural frequency gradually increases accordingly. When
low-pressure rotor. The maximum deformation position is shifted from the high-pressure the front d
rotor
angleto reaches
the low-pressure rotor.
0.3°, the Whenfrequency
natural the angle increases,
change the
rateamount
reachesof deformation
a maximum of of 33.8
the low-pressure rotor gradually increases.
Similarly, 0.05◦ , 0.1◦ , 0.15◦ , 0.2◦ , 0.25◦ , and 0.3◦ are selected as representatives to
Table 6. The change rate of the natural frequency of the 3rd order.
extract the 5th order modals and their corresponding natural frequencies. The change rates
are shown in Table 7. With(°)
Decentraction the increase in theNatural
front deflection angle non-concentricity,
Frequency (Hz) the Error
change rate of the natural frequency gradually increases. When the front deflection angle
non-concentricity 0reaches 3 mm, the natural frequency change 256.17rate reaches the maximum 0
of 14.7%. 0.05 251.32 1.9%
0.1 243 5.1%
0.15 224.60 12.3%
0.2 202.06 21.1%
0.25 187.14 27%
, 9, x FOR PEER REVIEW 15 of 21
Machines 2021, 9, 251 15 of 21

(a) (b)

(c) (d)

(e) (f)
Figure
Figure 17. 17. order
The 3rd The 3rd order
modal for modal
contrastfor contrast front
in different in different front
deflection deflection
angle angle non-concentricity:
non-concentricity: (a) θ = 0.05◦ ; (b) θ(a)
= 0.1◦ ;
θ =◦ 0.05°; (b) θ◦ = 0.1°; (c) θ◦ = 0.15°; (d)

(c) θ = 0.15 ; (d) θ = 0.2 ; (e) θ = 0.25 ; (f) θ = 0.3 . θ = 0.2°; (e) θ = 0.25°; (f) θ = 0.3°.

Similarly, 0.05°,
Table 0.1°,
7. The0.15°,
change0.2°, 0.25°,
rate of and 0.3°
the natural are selected
frequency of the 5th as representatives to ex-
order.
tract the 5th order modals and their◦ corresponding natural frequencies. The change rates
Decentraction ( ) Natural Frequency (Hz) Error
are shown in Table 7. With the increase in the front deflection angle non-concentricity, the
0 327.84 0
change rate of the natural frequency
0.05
gradually increases. When the front deflection
319.54
angle
2.53%
non-concentricity reaches 3 mm, 0.1 the natural frequency307.82 change rate reaches the maximum
6.11%
of 14.7%. 0.15 293.03 10.6%
0.2 285.52 12.9%
0.25 281.98 14%
Table 7. The change rate of the natural
0.3
frequency of the 5th order.
279.55 14.7%
Decentraction (°) Natural Frequency (Hz) Error
0 327.84 0
0.05 319.54 2.53%
0.1 307.82 6.11%
0.15 293.03 10.6%
0.2 285.52 12.9%
0.25 281.98 14%
Machines 2021, 9, 251 0.3 279.55 14.7% 16 of 21

The comparison of the 5th order vibration modes is shown in Figure 18. It finds that
the change pattern of The
vibration modeofisthe
comparison similar withvibration
5th order the parallel
modes non-concentricity.
is shown in FigureAs18.the
It finds that
parallel non-concentricity
the changeincreases,
pattern of the 5th order
vibration mode vibration
is similarmode of the
with the dual-rotor
parallel system As the
non-concentricity.
changes from theparallel non-concentricity
bending modal of the increases, the 5throtor
low-pressure ordertovibration mode of
the bending the dual-rotor
modal of the system
changes from the bending modal of the low-pressure rotor
high-pressure rotor, which is different from the bending modal of the 3rd order. to the bending modal of the
high-pressure rotor, which is different from the bending modal of the 3rd order.

(a) (b)

(c) (d)

(e) (f)
Figure Figure
18. The 18.5th The
order5th order
modal formodal
contrast for
incontrast
different in different
front front
deflection deflection
angle (a) θ = 0.05◦ ; (b) (a)
angle non-concentricity:
non-concentricity: θ = 0.1◦ ;
(c) θ = 0.15 ◦
θ = 0.05°;
; (d) θ (b)
= 0.2 ◦
θ =; (e)
0.1°;
θ =(c) ◦
θ =; (f)
0.25 0.15°; ◦
(d) θ
θ = 0.3 . = 0.2°; (e) θ = 0.25°; (f) θ = 0.3°.

4.4. The Influence of the Rear Deflection Angle Non-Concentricity on the Natural Frequency and
Modal Shape of the Dual-Rotor System
Machines 2021, 9, x FOR PEER REVIEW 18 of 22
Machines 2021, 9, 251 17 of 21

Machines 2021, 9, x FOR PEER REVIEW 17 of 2


Comparing the first five order natural frequencies with different rear deflection angle
non-concentricity
4.4. The Influence of in
the the
Rearrange of 0°–0.3°,
Deflection the comparisononresults
Angle Non-Concentricity are shown
the Natural inand
Frequency Figure 19.
Modal Shape of the Dual-Rotor System
The natural frequencies
Comparing of five
the first the first
orderfive modals
natural did not change
frequencies significantly.
with different rear deflection ang
Comparing the first five order natural frequencies with different rear deflection angle
non-concentricity
non-concentricity in the
in the range
range of 0◦of 0°–0.3°,
–0.3 the comparison
◦ , the comparison resultsresults are in
are shown shown
Figurein19.
Figure 1
Thenatural
The natural frequencies
frequencies of first
of the the first five modals
five modals did
did not not change
change significantly.
significantly.

Figure 19. Variation trend of natural frequency of rear deflection angle non-concentricity in the
range of 0°~0.3°.
Figure
Figure 19.19. Variation
Variation trendtrend of natural
of natural frequency
frequency of rear angle
of rear deflection deflection angle non-concentricity
non-concentricity in the range in th
5.ofrange
Influence
0~0.3◦of of Non-Concentricity on Critical Speed and Suggestions
. 0°~0.3°.
5.1. Influence of Non-Concentricity on Critical Speed
5. Influence of Non-Concentricity on Critical Speed and Suggestions
5. Influence
Based on of
theNon-Concentricity
finite element model on Critical Speed andsystem,
of the dual-rotor Suggestions
the critical speeds are
5.1. Influence of Non-Concentricity on Critical Speed
calculated.
5.1. Influence The ofCampbell diagram
Non-Concentricity of
onthe dual-rotor
Critical Speed system is shown in Figure 20. In order
Based on the finite element model of the dual-rotor system, the critical speeds are
to investigate
BasedThe
calculated. onthe theeffects
Campbell of the non-concentricity,
finitediagram
element ofmodel basedison
of the dual-rotor
the dual-rotor system the two
system,
shown inthe non-concentricity
critical
Figure speeds ar
20. In
models
calculated.
order of the dual-rotor
The Campbell
to investigate system,
the effectsdiagram the Campbell
of the dual-rotor
of the non-concentricity, diagrams
system
based of the
on theistwonon-concentricity
shown in Figure 20. model
non-concentricity In orde
are
todrawn
models and
of the
investigate the
thecritical
dual-rotor speeds
system,
effects of the ofnon-concentricity,
the two
theCampbell non-concentricity
diagrams of the onmodels
non-concentricity
based the two aremodel
calculated.
are The
non-concentricit
laws
drawn of parallel
and the non-concentricity
critical speeds of the two and front deflection
non-concentricity models
models of the dual-rotor system, the Campbell diagrams of the non-concentricity modangle
are non-concentricity
calculated. The laws on the
critical
of parallel
are drawn speedand aretheobtained.
critical The
non-concentricity and
speedscomparison
front deflection
of the two ofnon-concentricity
parallel non-concentricity
angle non-concentricity on
models are with
the values ofTh
critical
calculated. 1
mm
speedand 2 mm
are obtained. of the
The dual-rotor
comparison ofsystem
parallel is shown in
non-concentricity Figure
laws of parallel non-concentricity and front deflection angle non-concentricity on the cri 21.
with values Additionally,
of 1 mm the
and 2 mm of the dual-rotor system is shown in Figure 21. Additionally, the comparison of
comparison
ical speed of 0.1°and 0.18° front deflection angle of the dual-rotor system is shown in
0.1 ◦ and 0.18◦are
frontobtained.
deflection The comparison
angle of parallel
of the dual-rotor systemnon-concentricity
is shown in Figure 22. with values
Through of 1 mm
Figure
and
the
22. Through
2 mm of the
comparison,
the comparison,
it isdual-rotor
found that system
it is
is shown
both parallel
found that both
in Figure 21.
non concentricity and
parallel non
Additionally, concentricity and
the comparison o
angle non concentricity
angle
0.1°and
could non
reduceconcentricity
0.18°
the front could
deflection
second-order reduce
angle
critical the
speed second-order
ofand
thethedual-rotor critical
third-ordersystem speed and
is shown
critical speed. the in
third-order
However, Figure 2
critical
Through
both speed. However,
the effect
have little comparison, both have little
it is found
on the first-order effect
thatspeed.
critical on the first-order critical speed.
both parallel non concentricity and angle no
concentricity could reduce the second-order critical speed and the third-order critica
speed. However, both have little effect on the first-order critical speed.

Figure 20. Campbell diagram of the dual-rotor system.


Figure 20. Campbell diagram of the dual-rotor system.

Figure 20. Campbell diagram of the dual-rotor system.


Machines 2021, 9, x FOR PEER REVIEW 18 of 21
Machines 2021, 9, x FOR PEER REVIEW 18 of 21
Machines 2021, 9, 251 18 of 21

(a) (b)
(a) (b)
Figure 21. Campbell diagram of parallel non-concentricity: (a) δ = 1 mm; (b) δ = 2 mm.
Figure
Figure 21.
21. Campbell
Campbelldiagram
diagram of
of parallel
parallel non-concentricity:
non-concentricity: (a)
(a) δδ ==11mm;
mm;(b)
(b)δδ==22mm.
mm.

(a) (b)
(a) (b)
Figure
Figure 22.
22. Campbell
Campbell diagram
diagram of
of front
front deflection
deflection angle
angle non-concentricity:
non-concentricity: (a)
(a) θ
θ= 0.1◦ ; (b)
= 0.1°; 0.18◦ .
(b) θθ == 0.18°.
Figure 22. Campbell diagram of front deflection angle non-concentricity: (a) θ = 0.1°; (b) θ = 0.18°.
5.2.
5.2.Suggestions
SuggestionsforforEccentric
EccentricControl
ControlofofDual-Rotor
Dual-RotorSystems
Systems
5.2. Suggestions for Eccentric Control of Dual-Rotor Systems
Based
Based onon the study
study of
ofthree
threedifferent
differenttypes
typesofof non-concentricity,
non-concentricity, it isitworth
is worth noting
noting that
Based on the study of three different types of non-concentricity, it is worth noting
that the rule
the rule of vibration
of vibration characteristics
characteristics of theofparallel
the parallel non-concentricity
non-concentricity and the and thedeflection
front front de-
that the rule of vibration characteristics of the parallel non-concentricity and the front de-
angle non-concentricity
flection are similar.
angle non-concentricity When When
are similar. the bending modalmodal
the bending of theofhigh-pressure
the high-pressure rotor
flection angle non-concentricity are similar. When the bending modal of the high-pressure
Machines 2021, 9, x FOR PEER REVIEW 19 of 21
changes into the bending modal of the low-pressure rotor, the parallel non-concentricity
rotor changes into the bending modal of the low-pressure rotor, the parallel non-concen-
rotor changes into the bending modal of the low-pressure rotor, the parallel non-concen- ◦
of the of
tricity dual-rotor is 2 mm,
the dual-rotor is 2 and
mm,the
andfront declination
the front angleangle
declination non-concentricity
non-concentricity is 0.18 . The
is 0.18°.
tricity of the dual-rotor is 2 mm, and the front declination angle non-concentricity is 0.18°.
corresponding
The correspondingmodal comparison
modal is given
comparison in Figure
is given 23. 23.
in Figure
The corresponding modal comparison is given in Figure 23.
Comparing the front deflection angle non-concentricity and the rear deflection angle
Comparing the front deflection angle non-concentricity and the rear deflection angle
non-concentricity, it is found that the rear deflection angle non-concentricity has less in-
non-concentricity, it is found that the rear deflection angle non-concentricity has less in-
fluence on the vibration characteristics. The main reason for this conclusion is that the
fluence on the vibration characteristics. The main reason for this conclusion is that the
blade quality and moment of inertia of the high-pressure turbine are larger than those of
blade quality and moment of inertia of the high-pressure turbine are larger than those of
the turbofan, which leads to the vibration characteristics of the dual-rotor system being
the turbofan, which leads to the vibration characteristics of the dual-rotor system being
prone to change when the turbine position varies. At the same time, the intermediate bear-
prone to change when the turbine position varies. At the same time, the intermediate bear-
ing of the dual-rotor system is close to the rear support of the high-pressure impeller. The
ing of the dual-rotor system is close to the rear support of the high-pressure impeller. The
front deflection angle non-concentricity is more severely affected by the intermediate
front deflection angle non-concentricity is more severely affected by the intermediate
bearing than the rear deflection angle non-concentricity. In short, the main reason is that
bearing than the rear deflection angle non-concentricity. In short, the main reason is that
the position of the intermediate bearing, the mass of the blade and the moment of inertia
the position of the intermediate bearing, the mass of the blade and the moment of inertia
affect the natural frequency and modal of the dual-rotor.
affect the natural frequency and modal of the dual-rotor.
(a) (b)
Figure 23.
Figure 23. Comparison
Comparisonofofthe
the3rd
3rd modal
modal between
between parallel
parallel non-concentricity
non-concentricity andand the front
the front deflection
deflection angleangle non-concen-
non-concentricity:
tricity: (a) δ = 2 mm; (b) θ

(a) δ = 2 mm; (b) θ = 0.18 . = 0.18°.

In summary, during the assembly process of the dual-rotor system of the engine, the
parallel non-concentricity and the front deflection angle non-concentricity of the dual-ro-
tor should be controlled through research and comparison. Because the 2nd and 3rd order
modal interchanges earlier than the 4th and 5th order modal interchanges, the non-con-
Machines 2021, 9, 251 19 of 21

Comparing the front deflection angle non-concentricity and the rear deflection angle
non-concentricity, it is found that the rear deflection angle non-concentricity has less
influence on the vibration characteristics. The main reason for this conclusion is that the
blade quality and moment of inertia of the high-pressure turbine are larger than those of
the turbofan, which leads to the vibration characteristics of the dual-rotor system being
prone to change when the turbine position varies. At the same time, the intermediate
bearing of the dual-rotor system is close to the rear support of the high-pressure impeller.
The front deflection angle non-concentricity is more severely affected by the intermediate
bearing than the rear deflection angle non-concentricity. In short, the main reason is that
the position of the intermediate bearing, the mass of the blade and the moment of inertia
affect the natural frequency and modal of the dual-rotor.
In summary, during the assembly process of the dual-rotor system of the engine, the
parallel non-concentricity and the front deflection angle non-concentricity of the dual-
rotor should be controlled through research and comparison. Because the 2nd and 3rd
order modal interchanges earlier than the 4th and 5th order modal interchanges, the non-
concentricity should be controlled as minutely as possible to prevent the occurrence of the
2nd modal and 3rd order modal interchange, in order to avoid a bending modal appearing
in advance. The parallel non-concentricity should be controlled within 2 mm, and the
front deflection angle non-concentricity should be controlled within 0.18◦ . Therefore, it is
ensured that the vibration characteristics of the dual-rotor system of this aero engine do
not change significantly.

6. Conclusions
In this paper, a three-dimensional finite element solid model of an aero-engine dual-
rotor system has been established. Three different modes of parallel non-concentricity,
front deflection angle non-concentricity and rear deflection angle non-concentricity have
been simulated on the solid finite element model. The influence of three different heart
modes on the vibration characteristics of the system has discussed in detail. The main
conclusions are as follows.
(1) Parallel non-concentricity and front deflection angle non-concentricity have a signifi-
cant impact on the vibration characteristics of the dual-rotor system, and the influence
law is similar, while the back-deflection angle non-concentricity has little effect on
the vibration characteristics of the dual-rotor system. Therefore, these two types
disagreement requires focus control.
(2) With the increase in the parallel non-concentricity and the front deflection angle
non-concentricity, the natural frequency of the bending mode of the dual-rotor system
will gradually decrease, while the natural frequency of the local mode will not change
much. When the parallel non-concentricity reaches 2 mm or when the front deflection
angle non-concentricity reaches 0.18◦ , the 2nd order mode and the 3rd order mode
are interchanged. When the parallel non-concentricity reaches 2.25 mm or the front
deflection angle non-concentricity reaches 0.22◦ , the 4th order mode and the 5 order
mode will be interchanged. The phenomenon of modal interchange will make the
bending mode of the engine appear in advance, causing the engine to deviate from
the design working condition.
(3) The bending modes of the high-pressure and low-pressure rotors will mutually
transform in terms of the modal changes of the parallel non-concentricity and the
front deflection angle non-concentricity. Moreover, the new bending method deforms
more and causes more damage to the rotor.
(4) Parallel non-concentricity and front deflection angle non-concentricity need to be
controlled because the 2nd order and 3rd order modes are interchanged earlier than
the 4th order and 5th order modes are interchanged. In order to prevent the occurrence
of mode switching, it is recommended that the parallel non-concentricity be controlled
within 2.0 mm, and the front deflection angle non-concentricity should be controlled
within 0.18◦ of the non-concentricity.
Machines 2021, 9, 251 20 of 21

Author Contributions: Conceptualization, L.H.; methodology, S.H. and L.H.; investigation, S.H.;
data analysis, S.D. and Y.C. (Yufeng Cai); writing—original draft preparation, S.H.; writing—review
and editing, L.H. and Y.Y.; supervision, Y.C. (Yushu Chen); funding acquisition, L.H. All authors
have read and agreed to the published version of the manuscript.
Funding: This research was funded by National Major Science and Technology Projects of China,
grant number 2017-IV-0008-0045, National Natural Science Foundation of China, grant numbers
11972129 and 12172307, and the Opening Project of Applied Mechanics and Structure Safety Key
Laboratory of Sichuan Province, grant number SZDKF-201903.
Institutional Review Board Statement: The study was conducted according to the guidelines of the
Declaration of Helsinki, and approved by the Institutional Review Board.
Data Availability Statement: All data included in this study are available upon request by contact
with the corresponding author.
Acknowledgments: The authors are very grateful for help of editors and reviewers.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Cheng, Y.S.; Zhang, H.B. Review and Prospect of dynamics of complete aeroengine systems. Acta Aeronaut. Et Astronuatica Sin.
2011, 32, 1371–1391.
2. Xia, S.B.; Zhang, X.J.; Liu, Z.S.; Xu, S.C. A survey of research on misalignment of rotary machinery. J. Vib. Meas. Diagn. 1998, 18, 158–161.
3. Zhao, G.; Liu, Z.S.; Ye, J.H.; Chen, F. Meshing force model of misaligned gear coupling and its influence on a rotor system.
J. Harbin Eng. Univ. 2009, 30, 33–39.
4. He, C.B.; Gu, Y.J.; Yang, K. Properties analysis on coupled lateral and torsional vibrations of the rotor connected by the gear
coupling. J. Mech. Strength 2005, 27, 725–729.
5. Han, J.; Shi, L.D. Study on kinematic mechanism of misalignment fault of rotor system connected by gear coupling. J. Vib. Eng.
2004, 17, 416–520.
6. Xu, M.; Marangoni, R.D. Vibration analysis of a motor-flexible coupling-rotor system subject to misalign and unbalance. Part I:
Theoretical model and analysis. J. Sound Vib. 1994, 176, 663–679.
7. Lu, Z.; Li, H.C. Vibration characteristics and the test of misaligned rotors. J. Shanghai Sci. Technol. Univ. 1987, 9, 49–54.
8. Huber, M.; Strzelecki, S.; Steinhilper, W. Theoretical and experimental determination of the performances of misaligned and
statically loaded cylindrical journal bearings. Inst. Eng. Aust. 2000, 24, 31–37.
9. Arumugam, P.; Swarnamani, S.; Prabhu, B.S. Effects of journal misalignment on the performance characteristics of three-lobe
bearings. Wear 1997, 206, 122–129. [CrossRef]
10. Park, T.J. Effect of roller profile and misalignment in EHL of finite line contacts. ASME Bienn. Conf. Eng. Syst. Des. Anal. 2010, 1,
385–401.
11. Han, Q.K.; Wang, M.L.; Zhao, G.; Feng, G.Q. A review of rotor systems with misalignment. J. Dyn. Control 2016, 14, 1–13.
12. Liu, S.G.; Ma, Y.H.; Zhang, D.Y.; Hong, J. Studies on dynamic characteristics of the joint in the aero-engine rotor system. Mech.
Syst. Signal Process. 2012, 29, 120–136.
13. Wu, Y.X.; Zhang, Y.S.; Chen, G. Dynamic modeling and analysis of aeroengine with supporting non-concentricity. Chem. Eur. J.
2018, 44, 1–8.
14. Zhang, Z.B.; Ma, Y.H.; Li, J.; Hong, J. Research on the dynamic model of aero-engine rotors with bearing misalignment. Aeroengine
2014, 31, 208–214.
15. Xu, M.P.; Hou, L.; Li, H.L.; Chen, Y.S. Dynamic characteristics of a dual-rotor system with outer-rotor misalignment. J. Vib. Shock
2019, 38, 1–6.
16. Li, Q.K.; Liao, M.F.; Jiang, Y.F. The vibration features analysis of twin spool rotor with misalignment fault. Mech. Sci. Technol.
Aerosp. Eng. 2014, 33, 1916–1920.
17. Lu, Z.Y.; Wang, X.D.; Hou, L.; Chen, Y.S.; Li, H.L. Dynamic response analysis for the aero-engine dual-rotor-bearing system with
flexible coupling misalignment faults. J. Vibroeng. 2018, 20, 2012–2026. [CrossRef]
18. Feng, G.Q.; Zhou, B.Z.; Lin, L.J.; Han, Q.K. Misalignment analysis for support bearing in an inner-and-outer dual-rotor system.
J. Vib. Shock 2012, 31, 142–147.
19. Li, H.L.; Chen, Y.S.; Hou, L.; Zhang, Z.Y. Periodic response analysis of a misaligned rotor system by harmonic balance method
with alternating frequency/time domain technique. Sci. China Technol. Sci. 2016, 59, 1717–1729. [CrossRef]
20. Liu, Y.Q.; Xiao, S.; Hong, J.; Ma, Y.H. Excitation characteristic and dynamic response of misalignment of flexible rotor system with
three supportings. Acta Aeronaut. Et Astronuatica Sin. 2017, 38, 184–193.
21. Song, B.B. Study on Vibration Mechanism and Application of Aeroengine Rotor Non-Concentricity; Nanjing University of Aeronautics
and Astronautics: Nanjing, China, 2017.
Machines 2021, 9, 251 21 of 21

22. Abdou, M.K.; Saber, E. Effect of Rotor Misalignment on Stability of Journal Bearings with Finite Width. Alex. Eng. J. 2020, 59,
3407–3417. [CrossRef]
23. Santo, G.; Peeters, M.; van Paepegem, W.; Degroote, J. Effect of Rotor-Tower Interaction, Tilt Angle, and Yaw Misalignment on the
Aeroelasticity of a Large Horizontal Axis Wind Turbine with Composite Blades. Wind. Energy 2020, 23, 1578–1595. [CrossRef]
24. Sinha, J.K.; Leesb, A.W.; Friswellc, M.I. Estimating Unbalance and Misalignment of Flexible Rotating Machine from Single
Run-down. J. Sound Vib. 2004, 272, 967–989. [CrossRef]
25. Chen, Y.; Cui, J.; Sun, X. A Vibration Suppression Method for the Multistage Rotor of an Aero-Engine Based on Assembly
Optimization. Machines 2021, 9, 189. [CrossRef]
26. Sun, C.Z.; Chen, Y.S.; Hou, L. Nonlinear dynamical behaviors of a complicated dual–rotor aero-engine with rub-impact. Arch.
Appl. Mech. 2018, 88, 1305–1324. [CrossRef]

You might also like