materials

Article
Bird-Strike Resistance of Composite Laminates with
Different Materials
Yadong Zhou 1, * , Youchao Sun 1, * and Tianlin Huang 2
1 College of Civil Aviation, Nanjing University of Aeronautics and Astronautics, Nanjing 211100, China
2 Commercial Aircraft Engine Co., Ltd., Aero Engine Cooperation of China, Shanghai 201100, China
* Correspondence: yzhou@nuaa.edu.cn (Y.Z.); sunyc@nuaa.edu.cn (Y.S.)




Received: 6 September 2019; Accepted: 20 December 2019; Published: 27 December 2019


Abstract: To obtain some basic laws for bird-strike resistance of composite materials in aeronautical
application, the high-velocity impact behaviors of composite laminates with different materials were
studied by numerical methods. The smoothed particle hydrodynamics (SPH) and finite element
method (FEM) coupling models were validated from various perspectives, and the numerical
results were comparatively investigated. Results show that the different composite materials have
relatively little effect on projectile deformations during the bird impact. However, the impact-damage
distributions can be significantly different for different composite materials. The strength parameters
and fracture energy parameters play different roles in different damage modes. Lastly, modal
frequency was tentatively used to explain the damage behavior of the composite laminates, for it can
manifest the mass and stiffness characteristics of a dynamic structure. The dynamic properties and
strength properties jointly determine the impact-damage resistance of composite laminates under
bird strike. Future optimization study can be considered from these two aspects.

Keywords: bird strike; composite materials; damage mode; impact energy; modal frequency

1. Introduction
In civil aviation industry, bird-strike accidents are a serious threat to airplane passengers and
crew. A number of bird-strike accidents occur every year, and many of them are due to the collision
with wing leading edges or windshields or the ingestion of birds into aero-engines, which has a higher
probability than with other front-facing components. Bird-strike impacts are the major cause of blade
damage for aircraft engines [1]. Therefore, the bird-induced failure of both fan blades [2–5] and leading
edges [6–11] have caused much research interest. The latest bird-strike example was an Airbus A321 of
Russian Ural Airlines on 15 August 2019, which was struck by birds during take-off, and subsequently
its dual engines failed. The airworthiness regulations of the Federal Aviation Administration (FAA),
European Aviation Safety Agency (EASA), and Civil Aviation Administration of China (CAAC) have
all defined the requirements of anti-bird impact for aircraft structures [12].
To improve the aerodynamic performance and structural efficiency, material systems for
aero-engine fan blades and leading edges are transferring from alloys to composites. Consequently,
it is highly important to assess and optimize the bird-strike resistance of composite materials for
aeronautical applications. The knowledge of high-velocity impact resistance of composite material
systems is highly required and has aroused much attention in recent years. Grimaldi et al. [13] studied
the effect of different through-thickness lay-up configurations on bird-impact energies of laminates.
Hedayati et al. [14] investigated five types of lay-ups to calculate each critical thickness preventing
the bird from perforating. Hedayati et al. [15] showed that the optimum inner plate position was
inserted at the middle of the foam in a sandwich structure under bird strike. Mohagheghian et al. [16]
showed that the order of the glass layers can significantly influence the high-velocity soft impact

Materials 2020, 13, 129; doi:10.3390/ma13010129 www.mdpi.com/journal/materials

Materials 2020, 13, 129 2 of 22

performance. Mohagheghian et al. [17] considered the choice of the polymer interlayer for laminated
aircraft windshields. Kaboglu et al. [18] studied the high-velocity impact properties of various types
(different stacking configurations and thicknesses) of fiber metal laminates, showing perforation
resistance was significantly improved with increasing the number of alternating layers. Guida et al. [6]
hybridized carbon fiber reinforced plastic laminates with shape memory alloy materials and showed
better impact properties of the laminates containing the hybrid layers. Di Caprio et al. [7] considered
the bird-strike crashworthiness of different material systems (different stacking sequences and total
thicknesses), without changing the overall geometrical shape. Liu et al. [19] experimentally studied
the damage of hygrothermally conditioned carbon epoxy composites under high-velocity impact.
Wagner et al. [20] experimentally and numerically assessed different aerospace-grade composite
materials under high-velocity impact. Sun et al. [21] simulated the eccentric impact of square
and rectangular composite laminates embedded with shape-memory-alloy wires. Liu et al. [22]
experimentally observed matrix cracking and fiber breakage at the high velocity of 100 m/s on the rear
surface of the impacted carbon/epoxy composite. These abovementioned works have indicated that the
selection of lamina materials is a key issue and have shown promising results in material modification
of composite laminates for bird-strike resistance. Authors have recently investigated some critical
influencing factors for bird-strike damage of laminated composite materials, such as the total laminate
thickness [23], ply angles [24], rotational speeds [25], and impact locations/configurations [26,27].
However, the influence of different composite materials has not been considered very well.
As Ali et al. [28] pointed out, developing the improved and tough composite materials to lessen
severity of impact damage has been a major challenge for researchers. It is well-known that impact
resistance of different materials will be different, and it should be considered that the total thickness is
frequently a limit for critical aircraft structures. However, some generic laws for anti-bird-strike ability
are still not studied very well. Therefore, in the present article, different lamina materials with a fixed
total thickness are considered to compare the impact properties so as to obtain some generic laws for
bird-strike resistance of composite materials.

2. Numerical Model
At high velocities, the bird projectile behaves like fluidic materials, and its inhomogeneity becomes
negligible and hence can be treated as “soft body” [29,30]. The hydrodynamic responses of fluidic
materials can be accurately modelled by the smoothed particle hydrodynamics (SPH) method, which
has experienced wide use in bird-strike simulations [2,25,29,31–34]. In the present simulation study
with ABAQUS software, the bird projectile was firstly meshed by finite elements (C3D8R), and then
they were converted to SPH particles (PC3D) at the beginning of the bird-impact analysis (time
threshold = 0). A cylinder-shaped projectile was considered, with the length-to-diameter ratio being
2 (length 200 mm and diameter 100 mm, respectively). The mass density of the bird projectile was
950 kg/m3 , and the total mass was 1.49 kg. The Mie–Grüneisen equation of state (EOS) was used to
model the hydrodynamic behavior of the bird projectile. This type of EOS is linear in energy and
assumes a linear relationship between the shock velocity (Us ) and the particle velocity (Up ), which
provides the linear Us − Up Hugoniot form. In the linear Hugoniot form, the relationship Us = c0 + sUp
is determined by the parameters c0 (the sound velocity in the projectile) and s (material property).
For gelatin projectiles [35], c0 = 1483 m/s, and material properties s = 0 and Γ0 = 0 were used in
ABAQUS modelling.
Dealing with the impact damage of composite laminates, the continuum damage mechanics
(CDM) can well describe the intra-laminar damage [36] and have been widely adopted [7,33,37,38]
in the framework of finite element method (FEM). The square laminate was modelled by CDM
in present study. The in-plane size was 500 mm × 500 mm, and the total thickness was 3.6 mm
(0.15 mm × 24 mm). In the global coordinate, the area centroid of the plate corresponds to the center
of the projectile’s front surface. The boundary condition was the encastre case on the four edges;
that is, the perimeter of the plate was clamped, and thus all degrees of freedom (DOFs) of the four
Materials 2020, 13, 129 3 of 22

edges were constrained. Three different composite material systems, generally used for aeronautical
applications, were investigated: (1) T700/M21 carbon/epoxy material, (2) M91/IM7 carbon/epoxy
material, (3) lamina properties from reference [33], which was the carbon fiber reinforced plastics
composites. The material properties of the first two materials (T700/M21 and M91/IM7) were taken
and simplified from the reference [36]. Table 1 gives the material properties of the three laminas,
containing four sections: Mass density (ρ), elasticity (E, G, ν), strength (X, Y, S), and fracture energy
(Gc). Mechanical behavior of composite materials under high-velocity impact shows brittle features,
and the plasticity can be neglected. The orthotropic elasticity and Hashin’s damage criteria were
applied in the explicit dynamic analysis, which can model the initiation of four different types of
impact damage: Fiber/matrix tension/compression modes.

Table 1. Material properties of the considered laminas.

Parameter Unit T700/M21 M91/IM7 Reference [33]

ρ kg/m3 1600 1570 1400
E11 GPa 130 170 130.05
E22 GPa 7.7 8.8 11.55
G12 GPa 4.8 5.5 6
G13 GPa 4.8 5.5 6
G23 GPa 4.8 5.5 6
ν12 0.33 0.228 0.312
XT MPa 2080 2700 1022.7
XC MPa 1100 1590 613.5
YT MPa 60 105 54
YC MPa 180 252 170
S12 MPa 110 105 63
S13 MPa 110 105 63
GT1c kJ/m2 0.5 0.5 11.48
GT2c kJ/m2 2.1 2.1 4.13
GC1c
kJ/m2 0.5 0.5 0.35
GC2c kJ/m2 2.1 2.1 3.23

Contact was defined between the impacted laminate and the bird projectile. The contact includes
surface pairs of all exterior faces. In the impact contact process, the particles behave as spheres.
The frictionless tangential behavior and hard contact of normal behavior were defined for the
contact properties.
The laminate’s layups of the three plates are [0/45/0/−45/−45/0/45/0]3 . The stacking sequence
in the square brackets gives the basic eight layups, and the total through-thickness layups are three
copies of it, i.e., 24 plies in total. Figure 1 shows the numerical model of the bird projectile and the
square laminate. For the impacted plate, 2500 linear quadrilateral elements of type S4R were used, and
the total number of nodes was 2601. For a structural level simulation, typical element sizes of 5 mm
were performed [20]. A preliminary study was conducted, which validated that the present meshing
density (10 mm) was reasonable for comparison purpose, so that the computation efficiency could also
be ensured. For the bird projectile, 2220 linear hexahedral elements of type C3D8R were used, and
the total number of nodes was 2688. These nodes were converted into SPH particles (PC3D) during
the explicit dynamic analysis. The initial velocity was 200 m/s along the z direction, i.e., vertical to
the plate.
Materials 2020, 13, 129 4 of 22
Materials 2020, 13, x FOR PEER REVIEW 4 of 22

Z
Y

Figure 1.1. Numerical

Figure Numericalmodel
modelofof
thethe
bird projectile
bird andand
projectile the square laminate.
the square FiniteFinite
laminate. element methodmethod
element (FEM)
meshes are indicated by the dark cyan, smoothed particle hydrodynamics (SPH) particles
(FEM) meshes are indicated by the dark cyan, smoothed particle hydrodynamics (SPH) particles are are indicated
by the greyish
indicated white.
by the greyish white.
3. Results and Discussion
3. Results and Discussion
A total time period of 10 ms was simulated. Figure 2 depicts the deformation series of the bird
A total time period of 10 ms was simulated. Figure 2 depicts the deformation series of the bird
projectile impacting the plate with material 1, with the contour denoting the displacement along
projectile impacting the plate with material 1, with the contour denoting the displacement along z-
z-direction. It can be seen that the deformation series agree well with the experimental results in
direction. It can be seen that the deformation series agree well with the experimental results in
previous literature for high-velocity soft-body impacts. The SPH particles firstly contacted the target
previous literature for high-velocity soft-body impacts. The SPH particles firstly contacted the target
and then spread radially over the surface. Results in Figure 2 can also indicate the rationality of the
and then spread radially over the surface. Results in Figure 2 can also indicate the rationality of the
particle density of the bird projectile. Figure 3 depicts the deflection histories at the plate centers
particle density of the bird projectile. Figure 3 depicts the deflection histories at the plate centers with
with the three different materials. From the time histories, it can be observed that the three plates
the three different materials. From the time histories, it can be observed that the three plates
experienced the maximum deflections at approximately 0.6 ms, and then the plates began to rebound.
experienced the maximum deflections at approximately 0.6 ms, and then the plates began to rebound.
Figure 4 depicts the global magnitude of the displacements for both the bird particles and the plate
Figure 4 depicts the global magnitude of the displacements for both the bird particles and the plate
with the material system 3 at three different time steps: (a) 0.3 ms; (b) 1.0 ms; (c) 1.7 ms, and both the
with the material system 3 at three different time steps: (a) 0.3 ms; (b) 1.0 ms; (c) 1.7 ms, and both the
front and side views are presented to further observe the impact deformations. The impact contact
front and side views are presented to further observe the impact deformations. The impact contact
began at 0 ms. It can be seen that the projectile firstly contacted the plate, and then started to spread
began at 0 ms. It can be seen that the projectile firstly contacted the plate, and then started to spread
onto the plate surface, of which the deformation processes are similar to the previous experimental
onto the plate surface, of which the deformation processes are similar to the previous experimental
observation of soft-body impact on flat plate with other materials [7,39]. At 1.7 ms approximately, the
observation of soft-body impact on flat plate with other materials [7,39]. At 1.7 ms approximately, the
plate began to return the neutral plane (the original undeformed plane), and the dispersed debris of
plate began to return the neutral plane (the original undeformed plane), and the dispersed debris of
the projectile began to depart the plate surface. At this point in time, the deformation of the plate was
the projectile began to depart the plate surface. At this point in time, the deformation of the plate was
relatively small. Because of the lack of any 90◦ ply in the stacking sequence, the laminate was not quasi
relatively small. Because of the lack of any 90° ply in the stacking sequence, the laminate was not
isotropic but still maintained the symmetry. In order to highlight the effect of not having any ply at 90◦ ,
quasi isotropic but still maintained the symmetry. In order to highlight the effect of not having any
the rotational deformations in yz and xz directions are studied, as shown in Figure 5. The rotational
ply at 90°, the rotational deformations in yz and xz directions are studied, as shown in Figure 5. The
deformations extended from the center to the boundaries during the impact. It can be observed that
rotational deformations extended from the center to the boundaries during the impact. It can be
the rotational deformations exhibit a symmetrical feature in each direction, which can be ascribed to
observed that the rotational deformations exhibit a symmetrical feature in each direction, which can
the axial symmetry of the stacking sequence.
be ascribed to the axial symmetry of the stacking sequence.
Materials 2020, 13, 129 5 of 22
Materials 2020, 13, x FOR PEER REVIEW 5 of 22
Materials 2020, 13, x FOR PEER REVIEW 5 of 22

U, U3 U, U3
+1.591e+00 −1.590e+01
U, U3
+1.458e+00 U, U3
−1.961e+01
+1.326e+00
+1.591e+00 −2.332e+01
−1.590e+01
+1.193e+00
+1.458e+00 −2.703e+01
−1.961e+01
+1.061e+00
+1.326e+00 −3.074e+01
−2.332e+01
+9.281e−01
+1.193e+00 −3.445e+01
−2.703e+01
+7.955e−01
+1.061e+00 −3.817e+01
−3.074e+01
+6.629e−01
+9.281e−01 −4.188e+01
−3.445e+01
+5.303e−01
+7.955e−01 −4.559e+01
−3.817e+01
+3.978e−01
+6.629e−01 −4.930e+01
−4.188e+01
+2.652e−01
+5.303e−01 −5.301e+01
−4.559e+01
+1.326e−01
+3.978e−01 −5.672e+01
−4.930e+01
+0.000e+00
+2.652e−01 −6.043e+01
−5.301e+01
+1.326e−01 −5.672e+01
+0.000e+00 −6.043e+01

Y Y
Y Step: Step−1 Y Step: Step−1

0.0 ms 0.3 ms
Increment 0: Step Time = 0.0 Increment 263: Step Time = 3.0002E−04
Z Step: Step−1
XPrimary Var: U, U3 Z Step: Step−1
XPrimary Var: U, U3

0.0 ms 0.3 ms
IncrementVar: U0: Deformation
Deformed Step Time = Scale
0.0 Factor: +1.000e+00 IncrementVar: U
Deformed 263:Deformation
Step Time =Scale
3.0002E−04
Factor: +1.000e+00
Z Primary
XStatus Var:
Var: U, U3
STATUS Z Primary
XStatus Var:
Var: U, U3
STATUS
Deformed Var: U Deformation Scale Factor: +1.000e+00 Deformed Var: U Deformation Scale Factor: +1.000e+00
Status Var: STATUS Status Var: STATUS
U, U3 U, U3
−1.949e+01 +2.779e+00
U, U3
−2.792e+01 U, U3
−1.226e+01
−3.636e+01
−1.949e+01 −2.730e+01
+2.779e+00
−4.479e+01
−2.792e+01 −4.234e+01
−1.226e+01
−5.322e+01
−3.636e+01 −5.738e+01
−2.730e+01
−6.166e+01
−4.479e+01 −7.242e+01
−4.234e+01
−7.009e+01
−5.322e+01 −8.746e+01
−5.738e+01
−7.852e+01
−6.166e+01 −1.025e+02
−7.242e+01
−8.696e+01
−7.009e+01 −1.175e+02
−8.746e+01
−9.539e+01
−7.852e+01 −1.326e+02
−1.025e+02
−1.038e+02
−8.696e+01 −1.476e+02
−1.175e+02
−1.123e+02
−9.539e+01 −1.627e+02
−1.326e+02
−1.207e+02
−1.038e+02 −1.777e+02
−1.476e+02
−1.123e+02 −1.627e+02
−1.207e+02 −1.777e+02

Y Y
Y Step: Step−1 Y Step: Step−1

0.6 ms 0.9 ms
Increment 524: Step Time = 6.0083E−04 Increment 781: Step Time = 9.0091E−04
Z Step: Step−1
XPrimary Var: U, U3 Z Step: Step−1
XPrimary Var: U, U3

0.6 ms 0.9 ms
IncrementVar: U
Deformed 524:Deformation
Step Time =Scale
6.0083E−04
Factor: +1.000e+00 IncrementVar: U
Deformed 781:Deformation
Step Time =Scale
9.0091E−04
Factor: +1.000e+00
Z XPrimary
Status Var:
Var: U, U3
STATUS Z XPrimary
Status Var:
Var: U, U3
STATUS
Deformed Var: U Deformation Scale Factor: +1.000e+00 Deformed Var: U Deformation Scale Factor: +1.000e+00
Status Var: STATUS Status Var: STATUS

Figure 2.
2. Deformation series
series of the
the bird projectile
projectile impacting the
the plate with
with material 1.
1.
Figure 2. Deformation
Figure Deformation series of
of the bird
bird projectileimpacting
impacting theplate
plate withmaterial
material 1.

Figure
Figure 3. Deflection histories
3. Deflection histories at
at the
the plate
plate centers
centers with
with the
the three
three different
different materials.
materials.
Figure 3. Deflection histories at the plate centers with the three different materials.
 Materials 2020, 13, 129 6 of 22
Materials 2020, 13, x FOR PEER REVIEW 6 of 22

U, Magnitude
+6.047e+01
+5.543e+01
+5.039e+01
+4.536e+01
+4.032e+01
+3.528e+01
+3.024e+01
+2.520e+01
+2.016e+01
+1.512e+01
+1.008e+01
+5.039e+00
+0.000e+00

(a)

Y Y

Step: Step−1 Z X
Z X
Increment 283: Step Time = 3.0100E−04
Primary Var: U, Magnitude
Deformed Var: U Deformation Scale Factor: +1.000e+00
Status Var: STATUS

U, Magnitude
+1.906e+02
+1.748e+02
+1.589e+02
+1.430e+02
+1.271e+02
+1.112e+02
+9.532e+01
+7.943e+01
+6.355e+01
+4.766e+01
+3.177e+01
+1.589e+01
+0.000e+00

(b)

Y Y

Step: Step−1 Z X
Z X
Increment 917: Step Time = 9.9093E−04
Primary Var: U, Magnitude
Deformed Var: U Deformation Scale Factor: +1.000e+00
Status Var: STATUS

U, Magnitude
+2.165e+02
+1.984e+02
+1.804e+02
+1.624e+02
+1.443e+02
+1.263e+02
+1.082e+02
+9.020e+01
+7.216e+01
+5.412e+01
+3.608e+01
+1.804e+01
+0.000e+00

(c)

Y Y

Step: Step−1
Z X
Increment 1582: Step Time = 1.7104E−03 Z X
Primary Var: U, Magnitude
Deformed Var: U Deformation Scale Factor: +1.000e+00
Status Var: STATUS

Figure 4. Projectile
Figure deformation
4. Projectile deformationseries
seriesin
inthe
the model oflaminate
model of laminatewith
with material
material 3: (a)
3: (a) 0.3 ms;
0.3 ms; (b)ms;
(b) 1.0 1.0 ms;
(c) ms.
(c) 1.7 1.7 ms.
Materials 2020, 13, 129 7 of 22
Materials 2020, 13, x FOR PEER REVIEW 7 of 22

Rotation YZ Rotation ZX
UR, UR1 UR, UR2
0.304 0.222
0.254 0.186
0.205 0.149
0.155 0.112
0.105 0.075
0.056 0.038
0.006 0.001
−0.044 −0.035
−0.094 −0.072
−0.143 −0.109
−0.193 −0.146
−0.243 −0.183
−0.292 −0.220

Y Y

Step: Step−1 Step: Step−1

0.18 ms 0.18 ms
Z X
Increment 159: Step Time = 1.8057E−04 Z X
Increment 159: Step Time = 1.8057E−04
Primary Var: UR, UR1 Primary Var: UR, UR2
Deformed Var: U Deformation Scale Factor: +1.000e+00 Deformed Var: U Deformation Scale Factor: +1.000e+00
Status Var: STATUS Status Var: STATUS

UR, UR1 UR, UR2

0.257 0.196
0.214 0.164
0.171 0.131
0.128 0.099
0.086 0.066
0.043 0.034
−0.000 0.001
−0.043 −0.031
−0.086 −0.064
−0.128 −0.096
−0.171 −0.129
−0.214 −0.161
−0.257 −0.194

Y Y

Step: Step−1 Step: Step−1

0.30 ms 0.30 ms
Z X
Increment 263: Step Time = 3.0002E−04 Z X
Increment 263: Step Time = 3.0002E−04
Primary Var: UR, UR1 Primary Var: UR, UR2
Deformed Var: U Deformation Scale Factor: +1.000e+00 Deformed Var: U Deformation Scale Factor: +1.000e+00
Status Var: STATUS Status Var: STATUS

UR, UR1 UR, UR2

0.233 0.249
0.193 0.207
0.153 0.166
0.113 0.124
0.073 0.082
0.033 0.040
−0.007 −0.001
−0.047 −0.043
−0.087 −0.085
−0.127 −0.126
−0.167 −0.168
−0.206 −0.210
−0.246 −0.251

Y Y

Step: Step−1 Step: Step−1

Z X Z X
0.42 ms 0.42 ms
Increment 368: Step Time = 4.2072E−04 Increment 368: Step Time = 4.2072E−04
Primary Var: UR, UR1 Primary Var: UR, UR2
Deformed Var: U Deformation Scale Factor: +1.000e+00 Deformed Var: U Deformation Scale Factor: +1.000e+00
Status Var: STATUS Status Var: STATUS

UR, UR1 UR, UR2

0.339 0.279
0.282 0.234
0.225 0.188
0.168 0.142
0.111 0.097
0.054 0.051
−0.003 0.006
−0.060 −0.040
−0.117 −0.085
−0.174 −0.131
−0.231 −0.177
−0.288 −0.222
−0.346 −0.268

Y Y

Step: Step−1 Step: Step−1

0.60 ms 0.60 ms
Z X
Increment 524: Step Time = 6.0083E−04 Z X
Increment 524: Step Time = 6.0083E−04
Primary Var: UR, UR1 Primary Var: UR, UR2
Deformed Var: U Deformation Scale Factor: +1.000e+00 Deformed Var: U Deformation Scale Factor: +1.000e+00
Status Var: STATUS Status Var: STATUS

Figure 5. Rotational deformations of the laminates with material 1.

According to the numerical results, the projectile deformations of the three models are similar
to each other, as shown in Figure 6, which depicts the projectile deformations (magnitudes of the
 Materials 2020, 13, x FOR PEER REVIEW 8 of 22

Materials 2020, 13, 129 Figure 5. Rotational deformations of the laminates with material 1. 8 of 22

According to the numerical results, the projectile deformations of the three models are similar
to each other, as shown
particle displacement) at 1.2inms Figure
in the6,models
which depicts the projectile
with different plate’sdeformations (magnitudes
materials. Therefore, the of the
different
particle materials
composite displacement)
haveatlittle
1.2 mseffect
in theon
models with different
projectile plate’s materials.
deformations. This fact Therefore, the different
can be further indicated
composite materials have little effect on projectile deformations. This fact can be further
in the time histories of the deflections, as shown in Figure 3. Before the plates go back to the neutral indicated in
the time histories of the deflections, as shown in Figure 3. Before the plates go back to the neutral
position, the time histories of the deflections at the plate’s centers are quite similar to each other.
position, the time histories of the deflections at the plate’s centers are quite similar to each other.
During this stage, the projectile contacted on the plate surface. Thus, the projectile deformations are
During this stage, the projectile contacted on the plate surface. Thus, the projectile deformations are
also similar. Furthermore, in order to highlight the deformation differences among the analyzed plate
also similar. Furthermore, in order to highlight the deformation differences among the analyzed plate
models, Figure
models, 7 compares
Figure 7 compares thetheout-of-plane
out-of-planedisplacements for the
displacements for theplates
platesatatdifferent
different time
time steps,
steps, which
which
can can
provide global
provide information
global information of the
of structural impact
the structural response.
impact Results
response. cancan
Results alsoalso
verify thatthat
verify the the
global
displacements were quite similar in the square laminated plates made of different
global displacements were quite similar in the square laminated plates made of different material material systems.
systems.

U, Magnitude
+2.047e+02
+1.876e+02
+1.706e+02
+1.535e+02
+1.365e+02
+1.194e+02
+1.023e+02
+8.529e+01
+6.823e+01
+5.117e+01
+3.412e+01
+1.706e+01
+0.000e+00

(a)

Step: Step−1
Z X
Increment 1035: Step Time = 1.2002E−03
Primary Var: U, Magnitude
Deformed Var: U Deformation Scale Factor: +1.000e+00
Status Var: STATUS

U, Magnitude
+2.037e+02
+1.867e+02
+1.697e+02
+1.528e+02
+1.358e+02
+1.188e+02
+1.018e+02
+8.487e+01
+6.789e+01
+5.092e+01
+3.395e+01
+1.697e+01
+0.000e+00

(b)

Step: Step−1
Z X
Increment 1195: Step Time = 1.2005E−03
Primary Var: U, Magnitude
Deformed Var: U Deformation Scale Factor: +1.000e+00
Status Var: STATUS

U, Magnitude
+2.063e+02
+1.891e+02
+1.719e+02
+1.547e+02
+1.375e+02
+1.203e+02
+1.031e+02
+8.594e+01
+6.875e+01
+5.156e+01
+3.438e+01
+1.719e+01
+0.000e+00

(c)

Step: Step−1
Z X
Increment 1110: Step Time = 1.2004E−03
Primary Var: U, Magnitude
Deformed Var: U Deformation Scale Factor: +1.000e+00
Status Var: STATUS

Figure 6. Projectile deformations at 1.2 ms in the models with different plate’s materials: (a) material 1;
(b) material 2; (c) material 3.
 Materials 2020, 13, x FOR PEER REVIEW 9 of 22

Materials 2020, 13, 129

Figure 6. Projectile deformations at 1.2 ms in the models with different plate’s materials: (a) material 9 of 22
1; (b) material 2; (c) material 3.

Figure
Figure 7. The
7. The out-of-plane
out-of-plane (U3)displacements
(U3) displacements of
ofthe
theplates
plateswith
withdifferent material
different systems.
material systems.

The envelope
The envelope quantity
quantity was was
usedused to compare
to compare the impact
the impact damages
damages of different
of the the different material
material systems,
systems, i.e., the envelope damage of the 24 plies in total. Figure 8 depicts the
i.e., the envelope damage of the 24 plies in total. Figure 8 depicts the fiber tension damages (the term fiber tension damages
(the term DAMAGEFT reported in the legend) of the three models after bird impact. The colors from
DAMAGEFT reported in the legend) of the three models after bird impact. The colors from blue to red
blue to red indicate the damage states ranging from the undamaged to the maximum damaged. If
indicate the damage states ranging from the undamaged to the maximum damaged. If the damage
the damage index of an element equals 1, this element has completely lost its load-carrying capacity.
indexThe of an element
damaged equals
area of the1,laminate
this element
with has completely
material 3 is the lost its (≈100%),
largest load-carrying capacity.
while that The damaged
of material 2 is
area of the laminate with material 3 is the largest (≈100%), while that of material
the smallest (≈80%). The reason can be ascribed to the different properties of the fiber tension strength 2 is the smallest
(≈80%). The
XT, for reason
which can be
material 2 isascribed to the
the best and different
material 3 is the properties of the fiber
worst. Therefore, tension
the fiber damageX
strength
tension isT , for
T
whichmainly
material 2 is the
governed bybest and tension
the fiber material 3 is thewhile
strength, worst. theTherefore,
longitudinal thetensile
fiber fracture
tensionenergy
damage G1cishas
mainly
T
little effect
governed by theonfiber
it because
tension G1c strength,
of materialwhile
3 is approximately
the longitudinal 23 times higher
tensile than that
fracture GT1c has1 little
of materials
energy
and 2. Figure 9 depicts
T the fiber tension damages at different plies (front,
effect on it because G1c of material 3 is approximately 23 times higher than that of materials middle, and back plies)1ofand 2.
Figure 9 depicts the fiber tension damages at different plies (front, middle, and back plies) ofofthe
the plate with material 2. Results show that both the center of the back side and the boundary theplate
with material 2. Results show that both the center of the back side and the boundary of the front side
experience relatively high fiber-tension damage, while the middle ply experiences relatively small
damage. These distribution laws agree well with the experimental observations from the previous
literature dealing with the other composite material systems under high-velocity impact [16,22,40].
Materials 2020, 13, x FOR PEER REVIEW 10 of 22

front side experience relatively high fiber-tension damage, while the middle ply experiences
relatively small damage. These distribution laws agree well with the experimental observations from
Materials
the 2020, 13,
previous 129
literature 10 of 22
dealing with the other composite material systems under high-velocity impact
[16,22,40].

DAMAGEFT
Envelope (max abs)
(Avg: 75%)
+2.581e−01
+2.366e−01
+2.151e−01
+1.936e−01
+1.721e−01
+1.506e−01
+1.290e−01
+1.075e−01
+8.603e−02
+6.452e−02
+4.301e−02
+2.151e−02
+0.000e+00

(a)

Step: Step−1
Z X
Increment 2569: Step Time = 3.0000E−03
Primary Var: DAMAGEFT
Deformed Var: not set Deformation Scale Factor: not set
Status Var: STATUS

DAMAGEFT
Envelope (max abs)
(Avg: 75%)
+2.579e−01
+2.364e−01
+2.149e−01
+1.934e−01
+1.720e−01
+1.505e−01
+1.290e−01
+1.075e−01
+8.598e−02
+6.448e−02
+4.299e−02
+2.149e−02
+0.000e+00

(b)

Step: Step−1
Z X
Increment 2960: Step Time = 3.0000E−03
Primary Var: DAMAGEFT
Deformed Var: not set Deformation Scale Factor: not set
Status Var: STATUS

DAMAGEFT
Envelope (max abs)
(Avg: 75%)
+2.586e−01
+2.370e−01
+2.155e−01
+1.939e−01
+1.724e−01
+1.508e−01
+1.293e−01
+1.077e−01
+8.618e−02
+6.464e−02
+4.309e−02
+2.155e−02
+0.000e+00

(c)

Step: Step−1
Z X
Increment 2758: Step Time = 3.0000E−03
Primary Var: DAMAGEFT
Deformed Var: not set Deformation Scale Factor: not set
Status Var: STATUS

Figure 8.
Figure Thefiber
8. The fiber tension
tension damages of the three
three models
modelsafter
afterbird
birdimpact:
impact:(a)
(a)material
material1;1;(b)
(b)material
material2;
(c)(c)
2; material 3. 3.
material
Materials 2020, 13, 129 11 of 22
Materials 2020, 13, x FOR PEER REVIEW 11 of 22

Front
DAMAGEFT
Envelope (max abs)
(Avg: 75%)
+2.579e−01
+2.364e−01
+2.149e−01
+1.934e−01
+1.720e−01
+1.505e−01
+1.290e−01
+1.075e−01
+8.598e−02
+6.448e−02
+4.299e−02
+2.149e−02
+0.000e+00

Step: Step−1
Z X
Increment 2960: Step Time = 3.0000E−03
Primary Var: DAMAGEFT
Deformed Var: not set Deformation Scale Factor: not set
Status Var: STATUS

Middle Back

Figure 9. The fiber tension damages at different plies of the plate with material 2.
Figure 9. The fiber tension damages at different plies of the plate with material 2.

In terms of failure analysis of composite materials, it requires the mesh size to be small enough in
In terms of failure analysis of composite materials, it requires the mesh size to be small enough
order to well capture the fracture properties, but in the explicit dynamic analysis of structures under
in order to well capture the fracture properties, but in the explicit dynamic analysis of structures
impact,
underthe meshthe
impact, size cannot
mesh size be too small,
cannot be tooso as tosoavoid
small, as to the numerical
avoid instability.
the numerical The reason
instability. is that
The reason
theisstability limit ∆t limit Δtstable is approximately proportional to the smallest element dimensionmesh,
that the stabilitystable is approximately proportional to the smallest element dimension in the in
which can bewhich
the mesh, expressed
can be ∆tstable ≈ Las
as expressed minΔ/C d , where Lmin
t s ta b le ≈ L m in C d , is the
where smallest
L min is theelement
smallestdimension
element in the mesh,
dimension
andinCthe
d is mesh,
the wave
and speed of the
Cd is the wavematerial.
speed In of brief, in the non-linear
the material. In brief, insimulation of structural
the non-linear impact,
simulation of
westructural
should strike a balance between computational burden and numerical
impact, we should strike a balance between computational burden and numerical accuracy. To demonstrate
present meshes
accuracy. applicable for
To demonstrate the damage
present analysis, Figure
meshes applicable for the10 compares
damage the fiber-tension
analysis, damages
Figure 10 compares the of
thefiber-tension
laminate with material 1 by using different meshes. It can be validated
damages of the laminate with material 1 by using different meshes. It can be validatedthat the relatively coarse
meshes
that thearerelatively
applicable to predict
coarse meshesthe aredamage
applicable distributions.
to predict the damage distributions.
Materials 2020, 13, 129 12 of 22
Materials 2020, 13, x FOR PEER REVIEW 12 of 22

DAMAGEFT DAMAGEFT
Envelope (max abs) Envelope (max abs)
(Avg: 75%) (Avg: 75%)
+2.581e−01 +2.587e−01
+2.366e−01 +2.371e−01
+2.151e−01 +2.156e−01
+1.936e−01 +1.940e−01
+1.721e−01 +1.725e−01
+1.506e−01 +1.509e−01
+1.290e−01 +1.294e−01
+1.075e−01 +1.078e−01
+8.603e−02 +8.623e−02
+6.452e−02 +6.468e−02
+4.301e−02 +4.312e−02
+2.151e−02 +2.156e−02
+0.000e+00 +0.000e+00

(a) (b)

Figure
Figure 10. The
10. The comparison
comparison of fiber-tension
of fiber-tension damages
damages in laminate
in the the laminate (material
(material 1) with
1) with different
different meshes:
meshes: (a) relatively coarse mesh,
(a) relatively coarse mesh, (b) dense mesh.(b) dense mesh.

Figure
Figure 11 depicts
11 depicts thethe fiber
fiber compressiondamages
compression damages of of the
the three
threemodels
modelsafterafterbird
birdimpact.
impact. It can be be
It can
seen that the damage distributions in materials 1 and 2 are similar to each other
seen that the damage distributions in materials 1 and 2 are similar to each other (located at the center (located at the center
andandthe the boundaries
boundaries of of
thethe plate),but
plate), butmaterial
material 11 hashas a
a larger
larger damaged
damagedarea, area,mainly
mainly due
dueto the lower
to the lower
value of fiber compression strength XC. Results also show significant difference between the fiber
value of fiber compression strength XC . Results also show significant difference between the fiber
compression damage distributions of materials 1, 2, and that of material 3. The reason can be ascribed
compression damage distributions of materials 1, 2, and that of material C 3. The reason can be ascribed
to the relatively lower longitudinal compressive fracture energy (G1cC) of material 3. Therefore, the
to the relatively lower longitudinal compressive fracture energy (G1c ) of material 3. Therefore, the
compressive strength parameters can be considered to determine the severity of fiber compression
compressive strength
damage, while parametersfracture
the compressive can be energy
considered to determine
determines the damage the severity
distribution of fiber compression
patterns.
damage,Figure
while 12thedepicts
compressive fracture energy determines the damage distribution
the matrix compression damages of the three models after bird impact. It can patterns.
beFigure 12 depicts
seen that the matrix
the damage compression
distributions damages
in the three of the
materials arethree
similarmodels
to eachafter
otherbird impact.
(located It can
in the
be seen
area that the the
around damage
initialdistributions
impact contact). in the three materials
However, results show are similar to each
significant otherin(located
increase the matrix in the
areacompression
around thedamage
initial of
impact
materialcontact).
3. From However,
the initiation results
criteriashow significant
of Hashin’s damage, increase
it can beinseen
the that
matrix
the matrixdamage
compression compression damage3.mainly
of material From depends on thecriteria
the initiation longitudinal shear strength
of Hashin’s damage,(Sit12)can andbe theseen
thattransverse
the matrixshear strengthdamage
compression (S13) when transverse
mainly dependscompression strength shear
on the longitudinal (YC) is higher (Sthan
strength 12 ) the
and the
transverse shear strength (S ). Herein, both S and S of materials 1 and 2 are
transverse shear strength (S13 ) when transverse compression strength (YC ) is higher than the transverse
13 12 13 higher than those of
shearmaterial 3. Consequently,
strength (S13 ). Herein,the both most
S12severe
and Smatrix compression damage appeared in material 3.
13 of materials 1 and 2 are higher than those of material 3.
Figure 13 depicts the matrix tension
Consequently, the most severe matrix compression damage damages of the three models
appeared inafter bird 3.
material impact. It can be
observed that the damage distributions are similar on the whole, and relatively wide damaged area
Figure 13 depicts the matrix tension damages of the three models after bird impact. It can be
appeared in the model with material 3. In the damage initiation criteria, the transverse tensile
observed that the damage distributions are similar on the whole, and relatively wide damaged area
strength (YT) and the longitudinal shear strength (S12) are the beneficial factors for the matrix
appeared in the model with material 3. In the damage initiation criteria, the transverse tensile strength
compression damage. Comparing the material properties, it can be concluded that YT and S12 are the
(YT )most
and influential
the longitudinal
factors shear strength
for matrix (S12damage,
tension ) are thewhile
beneficial factors
the tensile for theenergies
fracture matrix compression
have little
damage. Comparing
influence. the material properties, it can be concluded that Y T and S 12 are the most influential
factors for matrix tension damage, while the tensile fracture energies have little influence.
Materials 2020, 13, 129 13 of 22
Materials 2020, 13, x FOR PEER REVIEW 13 of 22

DAMAGEFC
Envelope (max abs)
(Avg: 75%)
+2.585e−01
+2.370e−01
+2.154e−01
+1.939e−01
+1.723e−01
+1.508e−01
+1.293e−01
+1.077e−01
+8.617e−02
+6.463e−02
+4.308e−02
+2.154e−02
+0.000e+00

(a)

Step: Step−1
Z X
Increment 2569: Step Time = 3.0000E−03
Primary Var: DAMAGEFC
Deformed Var: not set Deformation Scale Factor: not set
Status Var: STATUS

DAMAGEFC
Envelope (max abs)
(Avg: 75%)
+2.586e−01
+2.370e−01
+2.155e−01
+1.939e−01
+1.724e−01
+1.508e−01
+1.293e−01
+1.077e−01
+8.620e−02
+6.465e−02
+4.310e−02
+2.155e−02
+0.000e+00

(b)

Step: Step−1
Z X
Increment 2960: Step Time = 3.0000E−03
Primary Var: DAMAGEFC
Deformed Var: not set Deformation Scale Factor: not set
Status Var: STATUS

DAMAGEFC
Envelope (max abs)
(Avg: 75%)
+2.586e−01
+2.371e−01
+2.155e−01
+1.940e−01
+1.724e−01
+1.509e−01
+1.293e−01
+1.078e−01
+8.621e−02
+6.465e−02
+4.310e−02
+2.155e−02
+0.000e+00

(c)

Step: Step−1
Z X
Increment 2758: Step Time = 3.0000E−03
Primary Var: DAMAGEFC
Deformed Var: not set Deformation Scale Factor: not set
Status Var: STATUS

Figure 11. The

Thefiber
fibercompression
compressiondamages
damagesofof
the three
the models
three after
models bird
after impact:
bird (a) material
impact: 1; (b)
(a) material 1;
material
(b) 2; (c)
material 2; material 3. 3.
(c) material
Materials 2020, 13, 129 14 of 22
Materials 2020, 13, x FOR PEER REVIEW 14 of 22

DAMAGEMC
Envelope (max abs)
(Avg: 75%)
+2.584e−01
+2.369e−01
+2.154e−01
+1.938e−01
+1.723e−01
+1.507e−01
+1.292e−01
+1.077e−01
+8.614e−02
+6.461e−02
+4.307e−02
+2.154e−02
+0.000e+00

(a)

Step: Step−1
Z X
Increment 2569: Step Time = 3.0000E−03
Primary Var: DAMAGEMC
Deformed Var: not set Deformation Scale Factor: not set
Status Var: STATUS

DAMAGEMC
Envelope (max abs)
(Avg: 75%)
+2.585e−01
+2.369e−01
+2.154e−01
+1.939e−01
+1.723e−01
+1.508e−01
+1.292e−01
+1.077e−01
+8.616e−02
+6.462e−02
+4.308e−02
+2.154e−02
+0.000e+00

(b)

Step: Step−1
Z X
Increment 2960: Step Time = 3.0000E−03
Primary Var: DAMAGEMC
Deformed Var: not set Deformation Scale Factor: not set
Status Var: STATUS

DAMAGEMC
Envelope (max abs)
(Avg: 75%)
+2.586e−01
+2.371e−01
+2.155e−01
+1.940e−01
+1.724e−01
+1.509e−01
+1.293e−01
+1.078e−01
+8.620e−02
+6.465e−02
+4.310e−02
+2.155e−02
+0.000e+00

(c)

Step: Step−1
Z X
Increment 2758: Step Time = 3.0000E−03
Primary Var: DAMAGEMC
Deformed Var: not set Deformation Scale Factor: not set
Status Var: STATUS

Figure 12.
12. The
Thematrix
matrixcompression
compressiondamages
damagesofof
thethe
three models
three after
models bird
after impact:
bird (a) material
impact: 1; (b)
(a) material 1;
material 2; (c)
(b) material 2; material 3. 3.
(c) material
Materials 2020, 13, 129 15 of 22
Materials 2020, 13, x FOR PEER REVIEW 15 of 22

DAMAGEMT
Envelope (max abs)
(Avg: 75%)
+2.586e−01
+2.371e−01
+2.155e−01
+1.940e−01
+1.724e−01
+1.509e−01
+1.293e−01
+1.078e−01
+8.620e−02
+6.465e−02
+4.310e−02
+2.155e−02
+0.000e+00

(a)

Step: Step−1
Z X
Increment 2569: Step Time = 3.0000E−03
Primary Var: DAMAGEMT
Deformed Var: not set Deformation Scale Factor: not set
Status Var: STATUS

DAMAGEMT
Envelope (max abs)
(Avg: 75%)
+2.585e−01
+2.370e−01
+2.155e−01
+1.939e−01
+1.724e−01
+1.508e−01
+1.293e−01
+1.077e−01
+8.618e−02
+6.464e−02
+4.309e−02
+2.155e−02
+0.000e+00

(b)

Step: Step−1
Z X
Increment 2960: Step Time = 3.0000E−03
Primary Var: DAMAGEMT
Deformed Var: not set Deformation Scale Factor: not set
Status Var: STATUS

DAMAGEMT
Envelope (max abs)
(Avg: 75%)
+2.588e−01
+2.373e−01
+2.157e−01
+1.941e−01
+1.725e−01
+1.510e−01
+1.294e−01
+1.078e−01
+8.627e−02
+6.471e−02
+4.314e−02
+2.157e−02
+0.000e+00

(c)

Step: Step−1
Z X
Increment 2759: Step Time = 3.0000E−03
Primary Var: DAMAGEMT
Deformed Var: not set Deformation Scale Factor: not set
Status Var: STATUS

Figure 13.The
Figure13. Thematrix
matrixtension damages
tension of theofthree
damages the models after bird
three models impact:
after bird (a) material
impact: (a)1;material
(b) material 2;
1; (b)
(c) material
material 3. material 3.
2; (c)

Figures
Figures14–16
14–16show
show the the
timetime
histories of kinetic
histories energies,energies,
of kinetic strain energies,
strain and damage
energies, dissipation
and damage
energies in the three models, respectively. The time histories of kinetic energies and strain energies
dissipation energies in the three models, respectively. The time histories of kinetic energies and strain in
the three models are close to each other, which also indicates that the global deformations are
energies in the three models are close to each other, which also indicates that the global deformations similar in
different materials.
are similar However,
in different the However,
materials. damage dissipation
the damage energies are quite
dissipation different
energies in thedifferent
are quite three models,
in the
with material 2 the lowest and material 3 the highest. A preliminary explanation
three models, with material 2 the lowest and material 3 the highest. A preliminary explanationis that material 2 is
is that
characterized by high stiffness and high strength. On the contrary, it can be noted that
material 2 is characterized by high stiffness and high strength. On the contrary, it can be noted thatmaterial 3 is
characterized by the lowestby
material 3 is characterized strength parameters.
the lowest strength parameters.
Materials 2020, 13, 129 16 of 22
Materials 2020,
Materials 2020, 13,
13, xx FOR
FOR PEER
PEER REVIEW
REVIEW 16 of
16 of 22
22

3.0x1077
3.0x10
Material 1
Material 1
2.8x1077
2.8x10 Material 2
Material 2
2.6x1077 Material 3
Material 3
2.6x10
(mJ)
energy(mJ) 2.4x1077
2.4x10

2.2x1077
2.2x10
Kineticenergy

2.0x1077
2.0x10

1.8x1077
1.8x10
Kinetic

1.6x1077
1.6x10

1.4x1077
1.4x10

1.2x1077
1.2x10

1.0x1077
1.0x10
0.000
0.000 0.001
0.001 0.002
0.002 0.003
0.003
Time (s)
Time (s)

Figure 14.
Figure 14. Time
Time histories of
Time histories of the
the kinetic
kinetic energy
energy in
in the
the three
three models.
models.

1x1077
1x10
Material 1
Material 1
9x1066
9x10 Material 2
Material 2
8x1066 Material 3
Material 3
8x10

7x1066
7x10
(mJ)
energy(mJ)

6x1066
6x10
Strainenergy

5x1066
5x10

4x1066
4x10
Strain

3x1066
3x10

2x1066
2x10

1x1066
1x10
0
0
0.000
0.000 0.001
0.001 0.002
0.002 0.003
0.003
Time (s)
Time (s)

Figure 15.
Figure 15. Time
Time histories
histories of the strain
of the strain energy
energy in
in the
the three
three models.
models.
Materials 2020, 13, 129 17 of 22
Materials 2020, 13, x FOR PEER REVIEW 17 of 22

5x1055

dissipation energy (mJ)

energy (mJ) 4x1055

3x1055
Damage dissipation

2x1055
Damage

1x1055 Material 1
Material 2
Material 3
0
0.000 0.001 0.002 0.003
Time (s)

Figure
Figure 16. Time histories
16. Time histories of
of the
the damage
damage dissipation
dissipation energy
energy in
in the
the three
three models.
models.

Energy output is particularly important in checking the accuracy of the solution in an explicit
Energy output is particularly important in checking the accuracy of the solution in an explicit
dynamic analysis. In order to further validate the global quality of the numerical simulations, the
dynamic analysis. In order to further validate the global quality of the numerical simulations, the
artificial strain energies of the three models are plotted in Figure 17, which belongs to a kind of
artificial strain energies of the three models are plotted in Figure 17, which belongs to a kind of
“artificial” energy and should be negligible compared to “real” energies such as the strain energy and
“artificial” energy and should be negligible compared to “real” energies such as the strain energy
the kinetic energy in explicit dynamic analysis. Theoretically speaking, the total energy in explicit
and the kinetic energy in explicit dynamic analysis. Theoretically speaking, the total energy in explicit
dynamic analysis should be a constant, i.e., the total energy balance. In numerical computation, it can
dynamic analysis should be a constant, i.e., the total energy balance. In numerical computation, it can
usually be approximate to a constant, as long as the error is small enough. The results in Figure 18
usually be approximate to a constant, as long as the error is small enough. The results in Figure 18
indicate that the total energies of present simulations are close to a constant. From Figures 14–18, the
indicate that the total energies of present simulations are close to a constant. From Figures 14–18, the
effectiveness of present numerical simulations can also be verified.
effectiveness of present numerical simulations can also be verified.

Figure
Figure 17. Time histories
17. Time histories of
of the
the artificial
artificial strain
strain energy
energy in
in the
the three
three models.
models.
Materials2020,
Materials 2020,13,
13,129
x FOR PEER REVIEW 18of
18 of22
22

Figure18.
Figure 18. Time
Timehistories
historiesof
ofthe
thetotal
totalenergy
energyin
inthe
thethree
threemodels.
models.

Figure
Figure19 19shows
showsthe themaximum
maximum damage damage indices
indices for for different
different damage damage modes modes in in the
the three
three models.
models.
ItItcan
canbe beseenseenthat thatlocal
localmaximum
maximumdamages damagesare arealmost
almostidentical
identicalfor forthethefour
fourdamage
damagemodes modesof ofthe
the
considered
consideredthree threematerials.
materials. This This phenomenon
phenomenon can can bebeascribed
ascribedto tothe
thesame
samestructural
structuralconfigurations
configurations
and
and the theidentical
identicalinput inputimpactimpactenergy
energy(E(E impact)) considered
impact considered in inpresent
presentsimulations.
simulations. The The bird-strike
bird-strike
impact
impact energy energy (E (Ebird-strike ) isdefined
bird-strike) is defined bybythethe impact
impact velocity
velocity andand projectile
projectile massmassas Eas
bird-strike = 0.5 ×=m0.5
Ebird-strike bird

×× m 2
× V 2 , where m is the mass of bird and V is the relative velocity between the
Vbird
relative , where
relativembird is the bird mass of bird and Vrelative is the relative velocity between the impacted
relative
impacted
structurestructure and the projectile.
and the projectile. In presentIn present
simulations,simulations, Ebird-strike
Ebird-strike approximatelyapproximately equalskJ.
equals 29.65 29.65 kJ.
In the
In the considered
considered impact impact duration,
duration, the the total
total energy
energy absorption(E
absorption (Eabsorption)
absorption )
of of
thethe impacted
impacted plate
plate can be
be
determined from the equation Eabsorption =  21mVrelative  , where m is the mass of SPH
P 1 2 −1 2 mV2 residual
1 2
determined from the equation Eabsorption = 
2
N  mVrelative −P mVresidual  , where m is the mass of SPH
particle; V residual is the residual velocity; and N the 2
 symbol depicts the2  sum of all SPH particles (N).
particle; V residual is the residual velocity; and the symbol ∑ depicts the sum of all SPH particles (N). The
The initial impact energy of the projectile was partly absorbed by the plate and partly carried by the
initial impact
particles (residual energy of theThe
velocity). projectile
former is was partly
mainly in absorbed
the form ofbydeformation
the plate and andpartly
damage; carried by the
the latter is
particles (residual velocity). The former is mainly in the form
mainly in the form of the residual velocities of each projectile particle and the vibrating plate. of deformation and damage; the latter
is mainly
From in thethe form of
damage the residualand
distributions velocities
the local of maximum
each projectile damage, particlesome and the vibrating
similar laws canplate. be found
with the From the damageresults
experimental distributions and the
of previous local maximum
literature. Wagner damage,et al. [20]some havesimilar
concludedlawsthat can different
be found
with the experimental results of previous literature. Wagner et
carbon fiber composites have very similar ballistic limits but differ in terms of delamination behavior.al. [20] have concluded that different
carbon fiber composites have very similar ballistic limits but
In present simulations, the local maximum damage induces are highly similar, but the damage areas differ in terms of delamination behavior.
In present
differ greatly simulations,
for different the local maximum
composite materials. damage induces are highly similar, but the damage areas
differ greatly for different composite materials.
Materials 2020, 13, 129 19 of 22
Materials 2020, 13, x FOR PEER REVIEW 19 of 22

Material 1
0.30 Material 2
Material 3
0.25
Maximum damage index

0.20

0.15

0.10

0.05

0.00
Fiber tension Fiber compression Matrix tension Matrix compression
Damage mode

Figure 19. Maximum local damage indices for different damage modes in the three models.
Figure 19. Maximum local damage indices for different damage modes in the three models.

Lastly, considering the fact that modal parameters are greatly useful to characterize the failure
Lastly, considering the fact that modal parameters are greatly useful to characterize the failure
resistance under dynamic loadings [41,42], modal analysis was conducted to possibly explain the
resistance under dynamic loadings [41,42], modal analysis was conducted to possibly explain the
damage results. Structural modal behaviors are determined by the mass matrix (M) and stiffness
damage results. Structural modal behaviors are determined by the mass matrix (M) and stiffness
matrix (K) together in FEM format. As indicated in the multiple degree-of-freedom (n) system, the
matrix (K) together in FEM format. As indicated in the multiple degree-of-freedom (n) system, the
frequency equation det K − ω 2M = 0 can determine the characteristic values (ω1 , ω2 , ···, ωn ) and
2
frequency equation det ( K − ω M ) = 0 can determine the characteristic values (ω1, ω2, ∙∙∙, ωn) and can
can hence reveal the inherent dynamic properties of a system. The matrix K − ω2 M is called dynamic
hence reveal
stiffness when the inherent
structural dynamic
damping canproperties of a which
be neglected, system. Thecase
is the matrix K − ω 2 M is called
for high-velocity impact.dynamic
Modal
stiffness when structural damping can be neglected, which is the case
frequencies can be considered as an indicator to evaluate the mass distribution and the for high-velocity impact. Modal
equivalent
frequencies
stiffness. can2 be
Table considered
gives the totalas an indicator
masses to evaluatefrequencies
and fundamental the mass distribution
of the threeand the equivalent
plates. Figure 20
stiffness. Table 2 gives the total masses and fundamental frequencies of the
compares the first five modal frequencies of the three plates with different materials. Therefore,three plates. Figure
from20
compares
the viewpointthe first five modal
of dynamics, it frequencies
can be seen of thematerial
that three plates
2 haswith
the different materials.
better stiffness Therefore,
properties thanfrom
the
other two. In addition, in spite of the largest mass and lowest frequencies, the laminate of materialthe
the viewpoint of dynamics, it can be seen that material 2 has the better stiffness properties than 1
other
has two.impact
better In addition, in spite
resistance thanof the
thatlargest mass and
of material lowestdue
3, mainly frequencies, the laminate
to the higher strengthofproperties.
material 1
has
In better
brief, theimpact
dynamic resistance than
properties andthat of material
strength 3, mainly
properties jointlydue to the higher
determine strength
the impact properties.
damage In
resistance
brief, the dynamic properties and strength properties jointly determine the impact
of composite laminates under bird strike. Future optimization study can be considered from these damage resistance
of composite
two aspects. laminates under bird strike. Future optimization study can be considered from these
two aspects.
Table 2. The total masses and modal frequencies of the three plates.
Table 2. The total masses and modal frequencies of the three plates.
Material System Material 1 Material 2 Material 3
Material System Material 1 Material 2 Material 3
Total Mass (kg) 1.44 1.41 1.26
Total
Fundamental Mass (kg)
Frequency (Hz) 139.28 1.44 159.09 1.41 152.251.26
Fundamental Frequency (Hz) 139.28 159.09 152.25
Materials 2020, 13, 129 20 of 22
Materials 2020, 13, x FOR PEER REVIEW 20 of 22

Material 1
500
Material 2
Material 3
400

Frequency (Hz)
300

200

100

0
1 2 3 4 5
Mode

Figure20.
Figure Modalfrequencies
20.Modal frequenciesof
ofthe
theplates
plateswith
withdifferent
differentmaterials.
materials.

4. Conclusions
4. Conclusions
By means of SPH-FEM modelling, this study numerically investigated the bird-impact behaviors
By means of SPH-FEM modelling, this study numerically investigated the bird-impact behaviors
(deformations and in-plane damages) of composite laminates with three different materials for
(deformations and in-plane damages) of composite laminates with three different materials for
aeronautical applications. The numerical models were validated from various perspectives, and the
aeronautical applications. The numerical models were validated from various perspectives, and the
numerical results were comparatively investigated in detail. The following conclusions can be drawn:
numerical results were comparatively investigated in detail. The following conclusions can be drawn:
1. Results show that the different composite materials (with the same reinforced fibers) have little
1. Results show that the different composite materials (with the same reinforced fibers) have little
effect on projectile deformations during the bird impact;
effect on projectile deformations during the bird impact;
2. The fiber tension damage is mainly governed by the fiber tension strength, while the longitudinal
2. The fiber tension damage is mainly governed by the fiber tension strength, while the
tensile fracture energy has little effect on it;
longitudinal tensile fracture energy has little effect on it;
3. The
3. The compressive
compressive strength
strengthparameters
parametersdetermine
determine thethe
severity of fiber
severity compression
of fiber damage,
compression while
damage,
the compressive fracture energy determines the damage distribution
while the compressive fracture energy determines the damage distribution patterns; patterns;
4. The
4. Thethree
threecomposite
compositematerials
materialshavehave similar
similar distribution
distribution patterns
patterns inin the
the matrix
matrix compression
compression
damage, while the most severe matrix compression damage appeared in material 3,3,due
damage, while the most severe matrix compression damage appeared in material duetotoits
its
weakest shear strengths;
weakest shear strengths;
5.
5. The parametersYYTTand
Theparameters andSS1212are
arethe
themost
mostinfluential
influentialfactors
factorsfor
forthe
thematrix
matrixtension
tensiondamage,
damage,while
while
the tensile fracture energies have little influence;
the tensile fracture energies have little influence;
6. Modal frequency was
6. Modal wastentatively
tentatively used to explain
used the damage
to explain behavior
the damage of the composite
behavior laminates,
of the composite
for it can manifest both the mass and stiffness characteristics of a dynamic structure.
laminates, for it can manifest both the mass and stiffness characteristics of a dynamic structure. The dynamic
properties
The dynamic and strength and
properties properties
strength jointly determine
properties jointlythedetermine
impact damage resistance
the impact damage of resistance
composite
oflaminates
composite under bird strike.
laminates under bird strike.
Author Contributions: Conceptualization, Y.Z.; Data curation, Y.Z.; Formal analysis, Y.Z.; Funding acquisition,
Author
Y.S.; Contributions:
Investigation, Y.Z.;Conceptualization,
Methodology, Y.Z.;Y.Z.;Project
Data curation, Y.Z.; Formal
administration, T.H.;analysis,
Resources,Y.Z.;Y.S.;
Funding acquisition,
Software, Y.Z.;
Y.S.; Investigation,
Supervision, Y.Z.; Methodology,
Y.S.; Validation, Y.Z.; Project Y.Z.;
T.H.; Visualization, administration, T.H.; Resources,
Writing—original Y.S.;Writing—review
draft, Y.Z.; Software, Y.Z.; Supervision,
& editing,
Y.S.; Validation, T.H.; Visualization, Y.Z.; Writing—original draft, Y.Z.; Writing—review & editing, Y.Z. All authors
Y.Z.
haveAll authors
read have read
and agreed and
to the agreed to
published the published
version version of the manuscript.
of the manuscript.
Funding:
Funding: The authors acknowledge
The authors acknowledgethe thefinancial
financial supports
supports from from the jointly
the jointly funded
funded projectproject of the National
of the National Natural
Natural
Science Science Foundation
Foundation and the and
Civilthe Civil Aviation
Aviation Administration
Administration of Chinaof China (U1333119);
(U1333119); ShanghaiShanghai Engineering
Engineering Research
Center of Center
Research Civil Aircraft
of CivilHealth Monitoring
Aircraft (GCZX-2015-05);
Health Monitoring the Defense the
(GCZX-2015-05); Industrial
DefenseTechnology
IndustrialDevelopment
Technology
Program (JCKY2013605B002); Fundamental Research Funds for the Central Universities (56XBC18206,
Development Program (JCKY2013605B002); Fundamental Research Funds for the Central Universities 56XBA18201,
IEA180017A19); National Science and Technology Major Project (2017-VIII-0003-0114).
(56XBC18206, 56XBA18201, IEA180017A19); National Science and Technology Major Project (2017-VIII-0003-
Acknowledgments: The authors greatly appreciate the Editor and anonymous Reviewers for their insightful
0114).
comments and suggestions to improve the manuscript.
Acknowledgments: The authors greatly appreciate the Editor and anonymous Reviewers for their insightful
comments and suggestions to improve the manuscript.
Materials 2020, 13, 129 21 of 22

Data Availability Statement: The data used to support the findings of this study are available from the
corresponding author upon request.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Abrate, S. Soft impacts on aerospace structures. Prog. Aerosp. Sci. 2016, 81, 1–17. [CrossRef]
2. Zhang, D.; Fei, Q. Effect of bird geometry and impact orientation in bird striking on a rotary jet-engine fan
analysis using SPH method. Aerosp. Sci. Technol. 2016, 54, 320–329. [CrossRef]
3. Mao, R.; Meguid, S.A.; Ng, T.Y. Finite Element Modeling of a Bird Striking an Engine Fan Blade. J. Aircr.
2007, 44, 583–596. [CrossRef]
4. Liu, J.; Zhong, D.; Li, Y.; Tang, Z.; Gao, X.; Zhang, Z.; Huang, F. Numerical simulation and test on damage of
rotary engine blades impacted by bird. Int. J. Crashworthines 2019, 24, 106–120. [CrossRef]
5. Vignjevic, R.; Orłowski, M.; De Vuyst, T.; Campbell, J.C. A parametric study of bird strike on engine blades.
Int. J. Impact Eng. 2013, 60, 44–57. [CrossRef]
6. Guida, M.; Sellitto, A.; Marulo, F.; Riccio, A. Analysis of the Impact Dynamics of Shape Memory Alloy
Hybrid Composites for Advanced Applications. Materials 2019, 12, 153. [CrossRef]
7. Di Caprio, F.; Cristillo, D.; Saputo, S.; Guida, M.; Riccio, A. Crashworthiness of wing leading edges under
bird impact event. Compos. Struct. 2019, 216, 39–52. [CrossRef]
8. Guida, M.; Marulo, F.; Polito, T.; Meo, M.; Riccio, M. Design and Testing of a Fiber-Metal-Laminate
Bird-Strike-Resistant Leading Edge. J. Aircr. 2009, 46, 2121–2129. [CrossRef]
9. Diamantakos, I.; Fotopoulos, K.; Jamin, M.; Eberhardt, A.; Lampeas, G. Investigation of bird strike events on
composite wing panels. Fatigue Fract. Eng. M. 2017, 40, 1538–1550. [CrossRef]
10. McCarthy, M.A.; Xiao, J.R.; McCarthy, C.T.; Kamoulakos, A.; Ramos, J.; Gallard, J.P.; Melito, V. Modelling
of Bird Strike on an Aircraft Wing Leading Edge Made from Fibre Metal Laminates—Part 2: Modelling of
Impact with SPH Bird Model. Appl. Compos. Mater. 2004, 11, 317–340. [CrossRef]
11. Kermanidis, T.; Labeas, G.; Sunaric, M.; Johnson, A.F.; Holzapfel, M. Bird strike simulation on a novel
composite leading edge design. Int. J. Crashworthines 2006, 11, 189–202. [CrossRef]
12. Cai, J.; Bao, H.; Zuo, H.; Huang, Y. Safety evaluation of airworthiness requirement of bird-strike on aeroplane.
Eng. Fail. Anal. 2019, 102, 407–416. [CrossRef]
13. Grimaldi, A.; Sollo, A.; Guida, M.; Marulo, F. Parametric study of a SPH high velocity impact analysis – A
birdstrike windshield application. Compos. Struct. 2013, 96, 616–630. [CrossRef]
14. Hedayati, R.; Ziaei-Rad, S.; Eyvazian, A.; Hamouda, A.M. Bird strike analysis on a typical helicopter
windshield with different lay-ups. J. Mech. Sci. Technol. 2014, 28, 1381–1392. [CrossRef]
15. Hedayati, R.; Sadighi, M. Effect of Using an Inner Plate between Two Faces of a Sandwich Structure in
Resistance to Bird-Strike Impact. J. Aerosp. Eng. 2016, 29, 04015020. [CrossRef]
16. Mohagheghian, I.; Wang, Y.; Zhou, J.; Yu, L.; Guo, X.; Yan, Y.; Charalambides, M.N.; Dear, J.P. Deformation and
damage mechanisms of laminated glass windows subjected to high velocity soft impact. Int. J. Solids Struct.
2017, 109, 46–62. [CrossRef]
17. Mohagheghian, I.; Charalambides, M.N.; Wang, Y.; Jiang, L.; Zhang, X.; Yan, Y.; Kinloch, A.J.; Dear, J.P.
Effect of the polymer interlayer on the high-velocity soft impact response of laminated glass plates. Int. J.
Impact Eng. 2018, 120, 150–170. [CrossRef]
18. Kaboglu, C.; Mohagheghian, I.; Zhou, J.; Guan, Z.; Cantwell, W.; John, S.; Blackman, B.R.K.; Kinloch, A.J.;
Dear, J.P. High-velocity impact deformation and perforation of fibre metal laminates. J. Mater. Sci. 2018,
53, 4209–4228. [CrossRef]
19. Liu, X.; Gu, W.; Liu, Q.; Lai, X.; Liu, L. Damage of Hygrothermally Conditioned Carbon Epoxy Composites
under High-Velocity Impact. Materials 2018, 11, 2525. [CrossRef]
20. Wagner, T.; Heimbs, S.; Franke, F.; Burger, U.; Middendorf, P. Experimental and numerical assessment of
aerospace grade composites based on high-velocity impact experiments. Compos. Struct. 2018, 204, 142–152.
[CrossRef]
21. Sun, M.; Chang, M.; Wang, Z.; Li, H.; Liu, Y. Simulation of Eccentric Impact of Square and Rectangular
Composite Laminates Embedded with SMA. Materials 2018, 11, 2371. [CrossRef] [PubMed]
Materials 2020, 13, 129 22 of 22

22. Liu, J.; Liu, H.; Kaboglu, C.; Kong, X.; Ding, Y.; Chai, H.; Blackman, B.R.K.; Kinloch, A.J.; Dear, J.P. The Impact
Performance of Woven-Fabric Thermoplastic and Thermoset Composites Subjected to High-Velocity Soft-
and Hard-Impact Loading. Appl. Compos. Mater. 2019, 26, 1389–1410. [CrossRef]
23. Zhou, Y.; Sun, Y.; Huang, T.; Cai, W. SPH-FEM simulation of impacted composite laminates with different
layups. Aerosp. Sci. Technol. 2019, 95, 105469. [CrossRef]
24. Zhou, Y.; Sun, Y.; Huang, T. SPH-FEM Design of Laminated Plies under Bird-Strike Impact. Aerospace 2019,
6, 112. [CrossRef]
25. Zhou, Y.; Sun, Y.; Cai, W. Bird-striking damage of rotating laminates using SPH-CDM method. Aerosp. Sci.
Technol. 2019, 84, 265–272. [CrossRef]
26. Zhou, Y.; Sun, Y.; Huang, T. Impact responses of slender composite plates for bird-strike testing of fan blades.
Lat. Am. J. Solids Struct. 2019, 16. [CrossRef]
27. Zhou, Y.; Sun, Y.; Huang, T. Impact-Damage Equivalency for Twisted Composite Blades with Symmetrical
Configurations. Symmetry 2019, 11, 1292. [CrossRef]
28. Ali, M.; Joshi, S.C.; Sultan, M.T.H. Palliatives for Low Velocity Impact Damage in Composite Laminates.
Adv. Mater. Sci. Eng. 2017, 2017, 1–16. [CrossRef]
29. Heimbs, S. Computational methods for bird strike simulations: A review. Comput. Struct. 2011, 89, 2093–2112.
[CrossRef]
30. Heimbs, S.; Bergmann, T. High-Velocity Impact Behaviour of Prestressed Composite Plates under Bird Strike
Loading. Int. J. Aerosp. Eng. 2012, 2012, 1–11. [CrossRef]
31. Zhang, Z.; Li, L.; Zhang, D. Effect of arbitrary yaw/pitch angle in bird strike numerical simulation using SPH
method. Aerosp. Sci. Technol. 2018, 81, 284–293. [CrossRef]
32. Dar, U.A.; Awais, M.; Mian, H.H.; Sheikh, M.Z. The effect of representative bird model and its impact
direction on crashworthiness of aircraft windshield and canopy structure. Proc. Inst. Mech. Eng. Part G J.
Aerosp. Eng. 2019, 233, 5150–5163. [CrossRef]
33. Riccio, A.; Cristiano, R.; Saputo, S.; Sellitto, A. Numerical methodologies for simulating bird-strike on
composite wings. Compos. Struct. 2018, 202, 590–602. [CrossRef]
34. Sun, F.; Sun, Q.; Ni, L.; Liang, K. Numerical analysis of anti-bird strike performance in structural connection
design for a vertical tail leading edge. Thin-Walled Struct. 2019, 144, 106319. [CrossRef]
35. Zhou, J.; Liu, J.; Zhang, X.; Yan, Y.; Jiang, L.; Mohagheghian, I.; Dear, J.P.; Charalambides, M.N. Experimental
and numerical investigation of high velocity soft impact loading on aircraft materials. Aerosp. Sci. Technol.
2019, 90, 44–58. [CrossRef]
36. Bogenfeld, R.; Kreikemeier, J.; Wille, T. Review and benchmark study on the analysis of low-velocity impact
on composite laminates. Eng. Fail. Anal. 2018, 86, 72–99. [CrossRef]
37. Riccio, A.; Ricchiuto, R.; Saputo, S.; Raimondo, A.; Caputo, F.; Antonucci, V.; Lopresto, V. Impact behaviour
of omega stiffened composite panels. Prog. Aerosp. Sci. 2016, 81, 41–48. [CrossRef]
38. Smojver, I.; Ivančević, D. Bird strike damage analysis in aircraft structures using Abaqus/Explicit and coupled
Eulerian Lagrangian approach. Compos. Sci. Technol. 2011, 71, 489–498. [CrossRef]
39. Liu, J.; Li, Y.; Gao, X. Bird strike on a flat plate: Experiments and numerical simulations. Int. J. Impact Eng.
2014, 70, 21–37. [CrossRef]
40. Roberts, G.D.; Pereira, J.M.; Revilock, D.M., Jr.; Binienda, W.K.; Xie, M.; Braley, M. Ballistic Impact of Braided
Composites with a Soft Projectile. J. Aerosp. Eng. 2005, 18, 3–7. [CrossRef]
41. Filippatos, A.; Gude, M. Influence of Gradual Damage on the Structural Dynamic Behaviour of Composite
Rotors: Experimental Investigations. Materials 2018, 11, 2421. [CrossRef] [PubMed]
42. Zhou, Y.; Tao, J. Theoretical and numerical investigation of stress mode shapes in multi-axial random fatigue.
Mech. Syst. Signal Pr. 2019, 127, 499–512. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).