Free Vibration and Buckling Analysis of

Stiffened Sandwich Plates with Repeated

Fold

Subhankar Pramanik, Sreyashi Das &

Arup Guha Niyogi

Journal of The Institution of

Engineers (India): Series C
Mechanical, Production, Aerospace and
Marine Engineering

ISSN 2250-0545

J. Inst. Eng. India Ser. C

DOI 10.1007/s40032-020-00627-x

1 23
Your article is protected by copyright and all
rights are held exclusively by The Institution
of Engineers (India). This e-offprint is for
personal use only and shall not be self-
archived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.

1 23
 Author's personal copy
J. Inst. Eng. India Ser. C
https://doi.org/10.1007/s40032-020-00627-x

ORIGINAL CONTRIBUTION

Free Vibration and Buckling Analysis of Stiffened Sandwich

Plates with Repeated Fold
Subhankar Pramanik1 • Sreyashi Das1 • Arup Guha Niyogi1

Received: 15 May 2020 / Accepted: 3 October 2020

Ó The Institution of Engineers (India) 2020

Abstract In this paper, complex-shaped sandwich folded relatively lower than complex curved shell structures.
plate structures with and without stiffeners have been These structures provide highest stiffness compared to
analyzed using first-order shear deformation theory. Finite plate due to the presence of the ridges formed by intro-
element analysis using eight-noded isoparametric ducing folds. On the other hand, sandwich construction
serendipity elements has been used. Regular plate elements with soft embedded core provides excellent bending stiff-
have been applied to model the plates and stiffeners. ness. It improves dynamic response and stability while
Transformation matrices have been used to form the global maintaining high stiffness-to-weight ratio.
stiffness matrix. Effect of different boundary conditions, Lot of works have been reported on isotropic folded
fiber angles and core thicknesses has been discussed. Dif- plates. Goldberg et al. [1] solved folded plate structure
ferent stiffener arrangements have been analyzed to study using method of elasticity. Cheung et al. [2] used finite
their effects. Numerical results reveal that the fixed edge of strip method to analyze folded box girder and prismatic
folded plate increases the stiffness prominently. Also, plate structures. Dynamic analysis of isotropic folded plate
proper positioning of stiffeners can significantly increase structures has been performed by Irie et al. [3]. Liu et al.
the stiffness and hence the natural frequency and the [4] analyzed one- and two-folded plates using finite ele-
buckling load. ment transfer matrix method. Niyogi et al. [5] carried out
finite element-based free vibration analysis of laminated
Keywords Stiffened folded plate  Corrugated roof  composite folded plate structures using first-order shear
Sandwich material  Finite element deformation theory. In the same year, Meunier et al. [6]
studied dynamic behavior of fiber-reinforced plastic (FRP)
composite materials using Reddy’s first- and higher-order
Introduction shear deformation theories. Duan and Miyamoto [7]
developed a mixed finite element method to analyze folded
The folded plate structures are abundantly used in struc- plate structures in 2002. Pal et al. [8] worked on free and
tural engineering applications such as box bridges, cul- forced vibration response of laminated composite and
verts, aerospace-, naval- and building-structures. It is also sandwich folded plates with and without stiffeners using
used as cladding system and often gives more economy nine-noded Lagrangian element. Lee et al. [9] investigated
over conventional material. The spans covered by folded the dynamic behavior of multiple-folded composite lami-
plates are more compared to flat plates of similar thickness. nates using higher-order plate theory. Later, they analyzed
Cost and effort required for constructing such structures are [10] dynamic response of a pre-stressed concrete box girder
bridge subjected to moving loads using folded plate ele-
ments in 2005. Peng et al. [11] analyzed stiffened and un-
& Sreyashi Das stiffened folded plates using mesh-free Galerkin method.
[email protected]
Later Peng [12] analyzed symmetric laminated composite
1
Department of Civil Engineering, Jadavpur University, folded plate structures for free vibration using an element-
Kolkata 700032, India free Galerkin method. Haldar et al. [13] investigated

123
 Author's personal copy
J. Inst. Eng. India Ser. C

laminated composite plates of different shapes using high z

precision shear deformable element. Anh et al. [14] pro- y
x
posed a coupled numerical method for the optimization of
fundamental frequency of folded laminated composite
plates in 2015. In the same year, static and dynamic anal-
yses of stiffened folded plates were done by Minh et al. y
[15] using discrete shear gap method (CS-FEM-DSG3). x
Very recently in 2018, they developed a homogenization Fig. 1 Positive senses of displacements
method [16] using cell-based smoothed Mindlin plate ele-
ment (CS-MIN3) for free vibration analysis of corrugated
panels. uðx; y; zÞ ¼ u0 ðx; yÞ þ zhy
Very few papers are available for the buckling analysis vðx; y; zÞ ¼ v0 ðx; yÞ  zhx
of sandwich folded plate structures with stiffeners. Liew w ¼ w0 ð1Þ
et al. [17] investigated the elastic buckling behavior of
;x ¼ hy þ w;x
stiffened and un-stiffened folded plates, with fixed modulus
of elasticity and Poisson’s ratio, under partial in-plane edge ;y ¼  hx þ w;y :
loads using meshfree Galerkin method. Barbero et al. [18] Here, u0 ðx; yÞ; v0 ðx; yÞ and w0 are corresponding mid-
developed a mixed isostatic quadrilateral three-dimen- plane displacements. ;x and ;y are shear strain cxz and cyz ,
sional plate element for static and buckling analyses of respectively. Generalized linear strains in terms of mid-
laminated composite folded plates. Very recently in 2019, plane strains are expressed as
Guo et al. [19] analyzed nonlinear vibration characteristics
of a Z-shaped folded plate analytically and experimentally. x ¼ u;x ¼ u0;x þ zhy;x ¼ 0x þ zKx
They used Hamilton principle, von Kármán equations and y ¼ v;y ¼ v0;y  zhx;y ¼ 0y þ zKy
the classical laminate plate theory to analyze the nonlinear  
cxy ¼ u;y þ v;x ¼ u0;y þ v0;x þ z hy;y  hx;x ¼ 0xy þ zKxy
dynamic model of folded plates. The review reveals lack of
research on open literature on buckling and free vibration ;x ¼ cxz ¼ u0;z þ w0;x
on complex multi-folded stiffened sandwich plate ;y ¼ cyz ¼ v0;z þ w0;y
structures. ez ¼ 0;
Hence, this paper is directed at dynamic-stability anal-
ð2Þ
ysis of complex folded plate structures. Finite element
method using eight-noded serendipity element has been where Kx and Ky are the curvatures in x–z and y–z
used in the analysis. Parametric studies have been per- planes, respectively, while Kxy is the cross-curvature in x–
formed for variation in core thickness, lamination angle of y plane.
face sheets and boundary conditions. It has been shown that The constitutive matrix [D] is given by
the performance of folded plates can be enhanced by 2 3
A11 A12 A16 B11 B12 B16 0 0
judicious use of stiffeners. 6 A12 A22 A26 B12 B22 B26
6 0 0 7 7
6 7
6 A16 A26 A66 B16 B26 B66 0 0 7
Theoretical Formulation 6 7
6B 0 7
6 11 B12 B16 D11 D12 D16 0 7
½D  ¼ 6 7:
6 B12 B22 B26 D21 D22 D26 0 0 7
In the present research, first-order transverse shear defor- 6 7
6B 0 7
mation theory has been used that follows the Yang–Norris– 6 16 B26 B66 D16 D26 D66 0 7
6 7
Stavsky theory [20] and includes the effect of rotary inertia 4 0 0 0 0 0 0 A44 A45 5
of plate elements while computing the element mass 0 0 0 0 0 0 A45 A55
matrix. The displacements of a flat Mindlin plate are fully Here;
described by five components, u, v; w; hx and hy , where
n Zk
z
displacement and rotations follow right-hand cork screw X  
Aij ; Bij ; Dij ¼ c0k 2
ij 1; z; z k dz ði; j ¼ 1; 2 and 6Þ
rule with the z-axis directed upward from the mid-plane of
k¼1
the plate element. Positive displacements u, v and w are zk1

n Zk
z
directed along positive x, y and z directions, while positive X
hx and hy rotations are shown in Fig. 1. and Aij ¼ a c0k
ij dz ði; j ¼ 4; 5Þ;
k¼1
The displacements and rotations at various points within zk1
the plate element are given by, ð3Þ

123
 Author's personal copy
J. Inst. Eng. India Ser. C

2 3
where c0k Ni;x 0 0 0 0
ij = off-axis stiffness coefficients of kth lamina [5]
6 Ni;y 0 0 0 0 7
and a is the shear correction factor taken as 5/6. 6 7
6 0 Ni;x 0 0 0 7
Using eight-noded isoparametric element, the linear 6 7
6 0 Ni;y 0 0 0 7
stiffness matrix is given by 6 7
6 0 0 Ni;x 0 0 7
6 7
Z1 Z1 6 0 0 Ni;y 0 0 7
½H  ¼ 6 7 ði ¼ 1 to 8Þ ð8Þ
½Ke  ¼ ½BT ½D½BjJ jdndg; ð4Þ 6 0
6 0 0 Ni;x 0 7 7
6 0 0 0 Ni;y 0 7
1 1 6 7
6 0 0 0 0 Ni;x 7
where ½Bi  is linear strain–displacement matrix given by, 6 7
6 0 0 0 0 Ni;y 7
2 3 6 7
Ni;x 0 0 0 0 4 0 0 0 Ni 0 5
6 0 Ni;y 0 0 0 7 0 0 0 0 Ni
6 7
6 Ni;y Ni;x 0 0 0 7
6 7 and [Sa  can be written as,
6 0 0 0 0 Ni;x 7
6
½ Bi  ¼ 6 7 ði ¼ 1 to 8Þ: 2 3
6 0 0 0 Ni;y 0 7 7
Nxa
6 Nxy
a 7
Nya
6 0 0 0 Ni;x Ni;y 7 6
6 0
7
7
6 7 6 0 Nxa 7
4 0 0 Ni;x 0 Ni 5 6 0
6 0 a
Nxy symmetric 7
7
6 0 0 0 0 Nxa 7
0 0 Ni;y Ni 0 6 7
6 0 0 0 a
0 Nxy Nya 7
6 7
6 Nxa t2 7
ð5Þ ½S  ¼ 6
a
6 0 0 0 0 0 0 7;
7 ð9Þ
6 12 7
6 a 2
Nxy t Nya t2 7
6 7
Geometric Stiffness Matrix 6 0
6
0 0 0 0 0
12 12
7
7
6 Nxa t2 7
6 0 0 0 0 0 0 0 0 7
6 7
6 12 7
When a folded plate is subjected to in-plane edge load, the 4 a 2
Nxy t a 25
N t
y
0 0 0 0 0 0 0 0
resultant strain and stress distribution in the folded plate is 12 12
non-uniform. Assuming that transverse deflection, w, does where Nxa ; Nya and Nxy
a
are applied in-plane loads.
not vary along depth, the nonlinear strain components of
the plate can be expressed as [21]. Mass Matrix
1h    i
exnl ¼ u20;x þ v20;x þ w2;x þ 2z u0;x hy;x  v0;x hx;x þ z2 h2y;x þ h2x;x
2 The element mass matrix with rotary inertia is given by
1h   [24, 25],
eynl ¼ u20;y þ v20;y þ w2;y þ 2z u0;y hy;y  v0;y hx;y ZZ
2  i
þz2 h2x;y þ h2y;y ½Me  ¼ ½N T ½q½N dxdy; ð10Þ
2 3
 I
exynl ¼ u0;x u0;y þ v0;x v0;y 60 I Symm 7
    6 7
þw;x w;y þ z u0;y hy;x þ u0;x hy;y  z v0;y hx;x þ v0;x hx;y where inertia matrix ½q ¼ 6 6 0 0 I 7,
7
  4P 0 0 Q 5
þz2 hy;x hy;y þ hx;x hx;y : where
0 P 0 0 Q
n Zk
z
ð6Þ X  
I; P; Q ¼ qk 1; Z; Z 2 dz; ð11Þ
The potential energy of in-plane stress, produced by k¼1
zk1
applied in- plane load, for the element can be expressed as
[22, 23] qk being the density of the kth lamina.
ZZZ
1  a 
Uae ¼ fde gT KGE fde gdv; ð7Þ Finite Element Transformation
2 ve

where A transformation [T] is applied to transform local element

Z1 Z1 linear stiffness, geometric stiffness and mass matrices into
  global stiffness and mass matrices [26] such that
a
KGe ¼ ½H T ½Sa ½H jJ jdndg
1 1 fug ¼ ½T fu0 g; ð12Þ

is the element geometric stiffness matrix due to in-plane where 

T
stresses produced by applied in-plane load. For eight-noded fug ¼ nu; v; w; hx ; hy ; hz o=T local displacements
isoparametric element, ½H  can be expressed as [23], fu0 g ¼ u0 ; v0 ; w0 ; h0x ; h0y ; h0z = global displacements

123
 Author's personal copy
J. Inst. Eng. India Ser. C

2 3
cosðx0 ; xÞ cosðy0 ; xÞ cosðz0 ; xÞ 0 0 0 global Y-axis is retained as local Y-axis, and the local
6 cosðx0 ; yÞ cosðy0 ; yÞ cosðz0 ; yÞ 0 0 0 7
6 7 X-axis is assumed to be normal to local Y-axis, in the plane
6 cosðx0 ; zÞ cosðy0 ; zÞ cosðz0 ; zÞ 0 0 0 7
6
½T  ¼ 6 7:
6 0 0 0 cosðx ; xÞ cosðy ; xÞ cosðz ; xÞ 7
0 0 0
7
of the element. Free vibration as well as buckling investi-
4 0 0 0 cosðx ; yÞ cosðy ; yÞ cosðz ; yÞ 5
0 0 0
0 0 0
gations have been done. New set of results have been
0 0 0 cosðx ; zÞ cosðy ; zÞ cosðz ; zÞ
obtained to study variation in core thicknesses, stack
sequence and boundary conditions.
The positive directions of the translations and rotations
of the eight-noded plate elements are shown by [26] (refer
Mesh Convergence Study
Fig. 2).
The stiffeners too are modeled as folded plate finite
The corrugated plate of thickness 10 mm as shown in
elements. Thus, the local element stiffness and mass
Fig. 4 has been analyzed. Length of the inclined stretch
matrices (unprimed) are transformed to global (primed)
(= AB = CD = EF = GH) of the folded plate is taken as
coordinates using [27]
0.5 m and that of horizontal part (= BC = DE = FG) is
½K0e ¼ ½TT ½K e ½T ; 1 m. Let the number of elements along local X direction be
½M0e ¼ ½TT ½M e ½T ; ð13Þ denoted by Nx and that along Y by Ny. The edge along
 0a  T a  Y = 1 plane is taken as clamped. All other sides are set free.
and; KG e ¼ ½T KGE e ½T : Sandwich plates with fiber-reinforced plastic face sheets,
  made of glass polyester resins and HEREX C70.130 PVC
Here, TT ¼ T1 since ½T  is orthogonal.
To apply this transformation, the element arrays are foam core [6] with properties shown in Table 1 are taken
inflated to incorporate a very small positive quantity, a for the present analysis.
thousandth of the least quantities in the leading diagonal, in Two layers of e-glass epoxy composites with fiber angle
locations corresponding to the hz terms [27] to render (0°/90°) have been accounted as face sheet. The face sheets
element matrices completely three-dimensional. Finally, are assumed to be 2 mm each with a core thickness of
these arrays are evaluated for each element, assembled and 6 mm. First five natural frequencies (Hz) for different Nx
updated for boundary conditions before solving. and Ny are presented in Table 2. From Table 2, it is clear
The governing equations for free vibration and buckling that Nx = 40 (four and eight elements along each inclined
analysis are given by: and straight stretch, respectively) and Ny = 4 can be used
  for further analysis.
For free vibration : ½K0x2n ½M0 fdg ¼ 0 ð14aÞ
  0a 
For buckling analysis : ½K0  l KG fdg ¼ 0: ð14bÞ Validation Study for Folded Plate Formulation

The natural frequencies and critical buckling loads are To the best of authors’ knowledge, no existing literature
extracted, respectively, from Eqs. (14a) and (14b). For has been found where the formulation for free vibration
clarity, the flowchart of the algorithm is provided in Fig. 3. and buckling analysis can be validated for sandwich folded
plates. Hence, a validation study has been made for both
isotropic and sandwich folded plates and compared with
Numerical Analysis
ANSYS. A onefold cantilever plate with crank angle 90°,
as shown in Fig. 5, and subjected to uniaxial load is taken
A MATLAB program has been developed to analyze fol-
for the study. The plate dimension is given as follows:
ded plates with repetitive geometry as shown in Fig. 4.
AB = EF = 1 m, AC = CE = BD = DF = 0.5 m.
Eight-noded serendipity element has been used in the
Thickness h = 0.01 m. The plate is clamped along ACE.
analysis. While defining local axes for finite elements, the
For isotropic material, the properties taken are

Fig. 2 Transformation of translations and rotation from local xi to global xi’-axes

123
 Author's personal copy
J. Inst. Eng. India Ser. C

Fig. 3 Flowchart of the program

E = 10.92 9 109 Pa, Poisson’s ratio t = 0.3, density

q = 1000 kg/m3. Table 3 shows the first three natural fre-
quencies (Hz) and non-dimensional critical buckling load
kcr = k a2/p2D, where k is buckling loads and D = Eh3/
12(1 - t2). The results are compared with results produced
through ANSYS 19 R3 academic version. From Table 3, it
is seen that the result matches very well with ANSYS.
Fig. 4 Geometry of corrugated plate

Table 1 Material properties for the sandwich material used [6]

E1 (GPa) E2 (GPa) G12 = G13 (GPa) G23 (GPa) m12 = m21 q (kg/m3)

Face 24.51 7.77 3.34 1.34 0.078 1800

Core 0.104 0.104 0.05 0.05 0.33 130

123
 Author's personal copy
J. Inst. Eng. India Ser. C

Table 2 First five undamped natural frequencies (Hz) computed for different mesh sizes
Mode no. Nx = 10, Ny = 1 Nx = 20, Ny = 2 Nx = 20, Ny = 4 Nx = 40, Ny = 4

1 20.68 19.47 18.99 18.95

2 20.77 19.52 19.04 19.00
3 34.10 32.93 32.22 32.12
4 38.44 37.02 36.25 36.08
5 39.74 38.39 37.60 37.39

A Table 4 Natural frequencies (Hz) of a three-layer simply supported

rectangular sandwich plate
Modes Experimental [30] HSDT [29] Present
L B
1 - 23 23.37
C
2 45 45 44.97
D 3 69 71 71.93
90o 4 78 80 81.29
L
5 92 93 93.74
E 6 129 128 131.38
L

F
Skin: E1 = 24.51 GPa, E2 = 7.77 GPa G12 = G13-
Fig. 5 Onefold plate model for validation study = G23 = 3.34 GPa, t = 0.078, s = 1800 kg/m3
Core: E1 = E2 = 0.104 GPa, G12 = G13 = G23-
For validation of sandwich plate structures, a three-layer = 0.05 GPa, t = 0.33, q = 130 kg/m3.
rectangular simply supported sandwich plate has been In Table 5, the first three natural frequencies (Hz) and
taken [28]. The dimension of the plate is 1.83 m 9 1.22 m. non-dimensional critical buckling loads have been shown.
Skin plate thickness = 4.06 9 10–4 m, Es = 68.9 GPa, From the results, it is noticed that MATLAB formulation
m = 0.3, qs = 2.77 9 103 Ns2/m4; For core layer, thick- provides slightly conservative results. It could be due to the
ness = 0.0064 m, E = 0.134 sGPa, G = 0.052 GPa, qc- fact that the present element is based on an equivalent two-
= 0.122 9 103 Ns2/m4. The result from the current dimensional formulation with five degrees of freedom
formulation, HSDT [29] and experimental results [30] are (DOF) per node, whereas in ANSYS, SHELL281 allows
given in Table 4. six DOF per node and used with layered shell definition
From Table 4, it is seen that the result matches quite with definable number of integration points through the
well with experiments and with HSDT for three-layer thickness of the layers.
sandwich plate.
Since no case of sandwich folded plate was available in Case Study 1: Free Vibration Study of the Sandwich
the open literature, the first model (Fig. 5) has been ana- Corrugated Plate
lyzed using sandwich material with same arrangement of
mesh convergence study and compared with ANSYS using Free vibration analysis of the sandwich folded plate
SHELL281 layered element. The material properties used (Fig. 4) made up of material as explained in Table 1 is
are as follows: conducted and reported in Table 6. The total thickness is

Table 3 The first three modes from free vibration and buckling for isotropic folded plate
Mode number MATLAB (Hz) ANSYS (Hz) MATLAB critical buckling load (kc) ANSYS critical buckling load (kc)

1 13.477 13.456 1.796 1.792

2 27.099 27.083
3 49.032 48.970

123
 Author's personal copy
J. Inst. Eng. India Ser. C

Table 5 The first three modes of free vibration (Hz) and critical CCCC FCFC CFCF
buckling loads [kc = ka2/E2h3] for the sandwich folded plate SSSS CFFF
60

Natural Frequency (Hz)

Mode MATLAB ANSYS MATLAB ANSYS
number (Hz) (Hz) buckling load buckling load 50
(kc) (kc) 40
30
1 13.170 12.713 1.628 1.609
20
2 28.300 27.560
10
3 52.821 50.334
0
30 45 60 75 90
Fibre angle θ (degree)

Fig. 6 Fundamental frequencies (Hz) plotted against changing fiber

kept as 10 mm. The layup for the sandwich is taken as 0/ angles for core thickness of 6 mm
h/core/h/0 where h is the inclination of fibers with respect
1. With increase in core thickness, natural frequency
to local X-axis (Fig. 4). h is varied as 30°, 45°, 60°, 75° and
increases. However, the gain in frequency drops as the
90° and the boundary conditions taken are as follows.
weaker and lighter core thickness rises. When the core
CCCC—Clamped at all edges thickness is 6 mm, the face thickness is 2 mm each.
SSSS—Simply supported at all edges Gradual increase in core thickness to 7 mm or 8 mm
FCFC—Clamped along two opposite straight edges, i.e., increases the lever arm distance at the cost of face
along y direction of Fig. 4 at x = 0,4. sheet thickness. The mass drops linearly as core
CFCF—Clamped along two opposite folded edges, i.e., widens. Thus, there is a rise that shows a negative
along x direction of Fig. 4 at y = 0, s1 gradient.
CFFF—Clamped along folded edge along y = 1 plane 2. It is evident that for CCCC and CFCF cases, the
only of Fig. 4. natural frequencies increase with increase in h (Fig. 6).
But for all other boundary conditions, the changes in
Variation of natural frequencies for various boundary
natural frequency are very nominal. For CCCC and
conditions with respect to fiber angle h is plotted for 6 mm
CFCF case, the structure spans in shorter direction, and
thick core in Fig. 6.
as h rises, the fibers too orient themselves along the
From Table 6 and Fig. 6, the following observations can
shorter direction. However, for the remaining cases,
be made:
just the reverse holds true.

Table 6 First three natural frequencies (Hz) for different boundary conditions and fiber angles
h Mode CCCC SSSS FCFC CFCF CFFF
Core thickness Core thickness Core thickness Core thickness Core thickness
6 mm 7 mm 8 mm 6 mm 7 mm 8 mm 6 mm 7 mm 8 mm 6 mm 7 mm 8 mm 6 mm 7 mm 8 mm

30° 1 50.5 52.1 52.5 40.8 42.0 42.3 2.4 2.4 2.4 40.8 42.1 42.6 19.2 19.7 19.8
2 52.4 53.9 54.3 42.5 43.8 44.0 3.0 3.1 3.1 40.8 42.2 42.6 20.8 21.4 21.6
3 54.0 55.6 55.9 44.7 46.0 46.2 5.2 5.3 5.4 51.3 52.9 53.3 34.6 35.4 35.4
45° 1 51.4 53.1 53.5 40.6 41.8 42.1 2.3 2.3 2.3 42.5 44.0 44.6 19.3 19.8 19.8
2 53.0 54.7 55.1 42.2 43.4 43.6 3.1 3.2 3.2 42.5 44.0 44.6 20.6 21.2 21.4
3 54.5 56.1 56.5 44.2 45.4 45.6 5.0 5.1 5.1 52.1 53.8 54.2 33.6 34.3 34.2
60° 1 52.3 54.1 54.6 39.8 41.0 41.2 2.2 2.2 2.2 44.2 45.9 46.6 19.2 19.7 19.8
2 53.8 55.6 56.1 41.4 42.5 42.7 3.0 3.0 3.1 44.2 45.9 46.6 20.0 20.6 20.8
3 55.2 56.9 57.4 43.3 44.4 44.5 4.8 4.9 4.9 53.0 54.7 55.3 32.8 33.4 33.2
75° 1 53.0 54.8 55.4 38.9 40.0 40.1 2.2 2.2 2.2 45.5 47.4 48.2 19.0 19.5 19.5
2 54.5 56.3 56.9 40.5 41.5 41.6 2.8 2.9 2.9 45.5 47.4 48.2 19.4 19.9 19.9
3 55.8 57.6 58.2 42.5 43.5 43.6 4.8 4.9 4.8 53.6 55.5 56.1 32.3 32.8 32.6
90° 1 53.2 55.1 55.8 38.5 39.5 39.6 2.2 2.2 2.2 46.0 48.0 48.8 18.9 19.4 19.4
2 54.7 56.6 57.2 40.1 41.1 41.2 2.7 2.8 2.8 46.0 48.0 48.9 19.0 19.5 19.5
3 56.1 57.9 58.5 42.1 43.1 43.1 4.8 4.8 4.8 53.9 55.8 56.4 32.1 32.6 32.4

123
 Author's personal copy
J. Inst. Eng. India Ser. C

MODE 1 MODE 2 MODE 3

CCCC

CFFF

FCFC

CFCF

SSSS

Fig. 7 First three mode shapes of the corrugated roof model (0°/90°/core/90°/0°), with core thickness 6 mm, for free vibration analysis (Case
study 1)

3. From Table 6, it is observed that CCCC boundary

condition makes the structure stiffest followed by
CFCF, SSSS, CFFF, FCFC conditions. In case of
CFCF condition, the unsupported length is 1 m, thus
providing higher stiffness than FCFC condition where
unsupported length is 4 m.
First three mode shapes of corrugated roof model with
6 mm core thickness and stack sequence 0°/90°/core/90°/
Fig. 8 Boundary conditions and uniaxial load on model 1 for CFFF 0° for different boundary conditions are plotted in Fig. 7.
condition

123
 Author's personal copy
J. Inst. Eng. India Ser. C

Table 7 Non-dimensional buckling load (= ka2/E2h3) for 0°/h/core/h/0° sandwich corrugated plates with different core thicknesses
Fiber angle CFFF FCFC CFCF
Core thickness Core thickness Core thickness
h 6 mm 7 mm 8 mm 6 mm 7 mm 8 mm 6 mm 7 mm 8 mm

30° 36.73 30.84 23.02 1.83 1.55 1.16 85.98 72.61 54.49
45° 37.50 31.51 23.54 1.92 1.62 1.22 93.17 79.11 59.64
60° 36.93 30.98 23.10 1.80 1.51 1.13 100.56 85.89 65.06
75° 36.99 31.02 23.12 1.59 1.33 0.98 106.66 91.58 69.68
90° 39.12 32.94 24.63 1.50 1.24 0.92 109.18 93.96 71.64

CFFF FCFC CFCF

Fig. 9 Buckling mode shape for the structure with 6 mm core and layup 0°/90°/core/90°/0° (Case study 2)

Fig.10 Geometry of stiffened corrugated plate

Figure 7 shows that in all mode shapes the horizontal Case Study 2: Buckling Analysis of the Sandwich
plates have deformed predominantly leaving the stiffer Corrugated Plate
ridge and the inclined plates undeformed. The first mode in
CFFF is antisymmetric, unlike CCCC case. The tip of the In this section, buckling analyses of the corrugated roof
central flat element in CFFF shows out of the plane model as shown in Fig. 4 have been performed. Total
deformation only in the third mode. FCFC condition is the thickness of the plate is maintained at 10 mm. The material
weakest with two further ends clamped and the longer sides properties used are as given in Table 1. Stacking sequence
free, showing one-way bending in first mode, torsion in the 0/h/core/h/0 has been used in the analysis. A uniaxial in-
second and a wavy form in the third mode. In SSSS plane edge load has been applied as shown in Fig. 8. Three
boundary conditions, the horizontal plate elements are most boundary conditions CFFF, FCFC and CFCF have been
disturbed, with symmetry in first and second modes, specified. The buckling load is tabulated in Table 7 for
antisymmetry in third. In CFCF case, the mode shape different h values and core thicknesses.
comes quite similar to CFFF case.

123
 Author's personal copy
J. Inst. Eng. India Ser. C

Table 8 Natural frequencies (Hz) and the non-dimensional buckling load (= ka2/E2h3) of stiffened cantilever sandwich corrugated plate
Natural frequency (Hz) Critical buckling load (non-dimensional)
1 2 3 4 5 6 7 8 9 10
h Mode No Stiffener Stiffener Stiffener position No Stiffener Stiffener Stiffener position
stiffener position 1 position 2 1 and 2 Stiffener position 1 position 2 1 and 2

30° 1 19.25 27.68 32.59 52.03 36.73 36.96 52.78 64.94

2 20.78 29.81 32.93 52.34
3 34.59 41.24 45.54 73.34
45° 1 19.29 27.89 33.05 51.86 37.5 37.59 54.49 68.99
2 20.61 30.07 33.81 52.62
3 33.56 40.80 45.68 74.82
60° 1 19.24 27.64 33.33 51.85 36.93 36.93 55.18 72.86
2 20.05 29.45 34.34 52.80
3 32.77 40.09 45.49 75.60
75° 1 19.02 27.27 33.67 52.00 36.99 36.96 56.85 77.55
2 19.37 28.35 34.42 52.66
3 32.28 39.40 45.16 75.78
90° 1 18.95 27.43 34.10 52.26 39.12 39.13 59.72 80.88
2 19.00 27.48 34.13 52.35
3 32.12 39.10 45.01 75.75

From Table 7, it is observed that for an increase in core plate. Three cases have been studied with cantilever
thickness and decrease in the corresponding skin thickness, boundary condition (CFFF).
the buckling load reduces noticeably for all boundary
a. Stiffener is attached at y = 0.5 m (position 1) with
conditions. This is justified since the effective stiffness
increase in mass by 10%
decreases as the share of weak core increases.
b. Stiffener is attached at y = 0.0 m (position 2) with
Here, for CFFF and CFCF boundary conditions, 0°/90°
increase in mass by 10%
layup gives maximum buckling load compared to other
c. Both stiffeners are attached with increase in mass by
values of h. The reason can be attributed to fixed folded
20%
edge. For FCFC boundary condition, 0°/90° skin gives the
lowest critical buckling load. Further in the FCFC case, the The thickness of the plate is taken as 10 mm with core
lengthy unsupported zone between the two fixed supports thickness 6 mm. 0/h/core/h/0 layup sequence is used once
renders it weaker than the CFFF case, and hence have very again. The natural frequencies in Hz and the non-dimen-
poor buckling resistance compared to CFCF and CFFF sional buckling load for different layup arrangements are
cases. FCFC boundary condition allows to-and-fro move- tabulated in Table 8. First three mode shapes for free
ment in the Y direction (Fig. 4). The buckling mode shape vibration and buckling mode shapes are plotted in Fig. 11.
at critical buckling load for 0°/90°/core/90°/0° and 6 mm Comparing columns 3 and 6 of Table 8, one can infer
core thickness is plotted in Fig. 9 for different boundary that a 20% rise in mass in the form of stiffeners increased
conditions. From the figure, it is noticed that CFFF and fundamental frequencies to the tune of 260–275%, which is
CFCF conditions manifest symmetry, unlike FCFC case, quite commendable. When the stiffener is attached at the
where the central flat panel moved to the right and upward. free end, the change in stiffness is maximum (70–80%).
Applying both stiffeners simultaneously increases stiffness
Case Study 3: Analysis of the Sandwich Corrugated Plate above 100%.
with Stiffener In case of buckling, applying intermediate stiffener
nominally changes the buckling load since there has been a
Here, the corrugated folded plate is stiffened with two 0.1- rise of mass-specific stiffness. But critical buckling load
m-wide stiffeners in the X–Z plane as shown in Fig. 10 so increases to a greater extent (40–50%) for stiffeners posi-
as to control the out of the plane deformation. The stiff- tioned at free end. It is observed that natural frequency and
eners are assumed to have the same property as of the main non-dimensional buckling load are maximum for stiffener
position at 2 with h = 90°.

123
 Author's personal copy
J. Inst. Eng. India Ser. C

STIFFENER POSITION 1(Y=0.5) STIFFENER POSITION 2(Y=0.0) STIFFENER POSITION 1 AND 2

MODE 1

MODE 2

MODE 3

BUCKLING
MODE
SHAPE

Fig. 11 First three mode shapes (free vibration) and critical buckling mode of stiffened sandwich corrugated folded plate with 6 mm core and
layup 0°/90°/core/90°/0° for different stiffener positions (Case study 3)

From Fig. 11, no prominent changes are found in first and rotary inertia has been used. This has been converted
two mode shapes of the stiffened sandwich folded plate. into a six degree of freedom per node by introducing a
Only the central flat plate is subjected to bending defor- small hz seed in the otherwise blank hz locations to apply
mation in the third mode. Comparing with Fig. 7, it may be the full three-dimensional transformation applicable to
inferred that the stiffeners at position 2 have resisted flat folded plate elements.
plate displacements to a great extent. From the studies, it is observed that with increase in the
core thickness, stiffness of the structure decreases but mass
decreases even more rapidly for the cases considered. As a
Conclusion result, fundamental frequency has increased for the case
studies presented. However, the gain in natural frequency
This study deals with the free vibration and buckling
reduces asymptotically since frequencies have joint con-
analysis of sandwich folded plates of self-replicating nature
tribution of stiffness and mass. Natural frequencies depend
with and without stiffeners. An eight-noded regular five
on the boundary condition and the layup sequence of face
degree of freedom per node Mindlin-type plate element
sheets too. As expected, when the folded plates are
considering first-order transverse shear deformation theory
clamped on all edges, the stiffness of the structure

123
 Author's personal copy
J. Inst. Eng. India Ser. C

increased radically. Also, in this case at fiber angle 14. L. Le-Anh, T. Nguyen-Thoi, V. Ho-Huu, H. Dang-Trung, T. Bui-
h = 90°, fundamental frequency attained maxima. Xuan, Static and frequency optimization of folded laminated
composite plates using an adjusted differential evolution algo-
However, reduction in core thickness has been observed rithm and a smoothed triangular plate element. Compos. Struct.
to reduce the critical buckling load. This is obvious since 127, 382–394 (2015)
the buckling load depends on geometry and stiffness of the 15. N. Nguyen-Minh, T. Nguyen-Thoi, T. Bui-Xuan, T. Vo-Duy,
structure and not on the mass. Further, it is seen that Static and free vibration analyses of stiffened folded plates using
a cell-based smoothed discrete shear gap method (CS-FEM-
changes in boundary conditions bring drastic reduction in DSG3). Appl. Math. Comput. 266, 212–234 (2015)
critical buckling load values. For stiffened corrugated 16. N. Nguyen-Minh, N. Tran-Van, T. Bui-Xuan, T. Nguyen-Thoi,
plates, proper selection of stiffener location could enhance Free vibration analysis of corrugated panels using homogeniza-
performance of the structure, and thus, judicious use of tion methods and a cell-based smoothed Mindlin plate element
(CS-MIN3). Thin-Walled Struct. 124, 184–201 (2018)
stiffeners could be quite lucrative in boosting mass-specific 17. K.M. Liew, L.X. Peng, S. Kitipornchai, Buckling of folded plate
stiffness. structures subjected to partial in-plane edge loads by the FSDT
meshfree Galerkin method. Int. J. Numer. Methods Eng. 65,
Acknowledgements The authors gratefully acknowledge the finan- 1495–1526 (2006)
cial support extended by the Council of Scientific and Industrial 18. E.J. Barbero, A. Madeo, G. Zagari, R. Zinno, G. Zucco, A mixed
Research, India (File No: 09/096 (0974)/ (2019)/EMRI) for con- isostatic 24 dof element for static and buckling analysis of lam-
ducting this research. inated folded plates. Compos. Struct. 116, 223–234 (2014)
19. G. Xiangying, Z. Yang, Z. Wei, Theoretical and experimental
investigation on the nonlinear vibration behavior of Z-shaped
References
folded plates with inner resonance. Eng. Struct. 182, 123–140
(2019)
1. J.E. Goldberg, H.L. Leve, Theory of prismatic folded plate 20. P.C. Yang, C.H. Norris, Y. Stavsky, Elastic wave propagation in
structures. J. IABSE 17, 59–86 (1957) heterogeneous plates. Int. J. Solids Struct. 2, 665–684 (1966)
2. Y.K. Cheung, Folded plate structures by finite strip method. 21. R.M. Jones, Mechanics of Composite Materials, 2nd edn. (Taylor
J. Struct. Eng. ASCE 1969(95), 2963–2979 (1969) & Francis, Milton Park, 1999)
3. T. Irie, G. Yamada, Y. Kobayashi, Free vibration of a cantilever 22. R. Cook, D. Markus, M. Plesha, Concepts and Applications of
folded plate. J. Acoust. Soc. Am. 76(6), 1743–1748 (1984) Finite Element Analysis, 4th edn. (Wiley, Hoboken, 2002)
4. W.H. Liu, C.C. Huang, Vibration analysis of folded plates. 23. K.S. Ram, P.K. Sinha, Hygrothermal effects on the buckling of
J. Sound Vib. 157(1), 123–137 (1992) laminated composite plates. Compos. Struct. 21(4), 233–247
5. A. Guha Niyogi, M.K. Laha, P.K. Sinha, Finite element vibration (1992)
analysis of laminated composite folded plate structures. Shock 24. J.N. Reddy, Mechanics of Laminated Composite Plates and
Vib. 6, 273–283 (1999) Shells: Theory and Analysis, 2nd edn. (CRC Press, Boca Raton,
6. M. Meunier, R.A. Shenoi, Free vibration analysis of composite 2003)
sandwich plates. Proc. Inst. Mech. Eng. Part C J. Mech. Eng. Sci. 25. J.T. Oden, Mechanics of Elastic Structures (McGraw-Hill, New
213(7), 715–727 (1999) York, 1967)
7. M. Duan, Y. Miyamoto, Effective hybrid/mixed finite elements 26. S. Das, A. Guha Niyogi, Free-vibration analysis of epoxy-based
for folded-plate structures. J. Eng. Mech. 128(2), 202–208 (2002) cross-ply laminated composite folded plates subjected to hygro-
8. S. Pal, A. Guha Niyogi, Application of folded plate formulation thermal loading. J. Inst. Eng. India Ser. C (2020).
in analysing stiffened laminated composite and sandwich folded https://doi.org/10.1007/s40032-020-00573-8
plate vibration. J. Reinf. Plast. Compos. 27(7), 693–710 (2008) 27. K.J. Bathe, Finite Element Procedures (Prentice Hall of India
9. S.Y. Lee, S.C. Wooh, S.S. Yhim, Dynamic behavior of folded Pvt. Ltd., New Delhi, 1996)
composite plates analyzed by the third order plate theory. Int. 28. A. Nayak, S. Moy, R. Shenoi, Free vibration analysis of com-
J. Solids Struct. 41, 1879–1892 (2004) posite sandwich plates based on Reddy’s higher-order theory.
10. S.Y. Lee, S.S. Yhim, Dynamic behavior of long-span box girder Compos. Part B Eng. 33(7), 505–519 (2002)
bridges subjected to moving loads: numerical analysis and 29. M. Meunier, Dynamic analysis of composite sandwich plates
experimental verification. Int. J. Solids Struct. 42(18–19), using higher-order theory. PhD Thesis, University of
5021–5035 (2005) Southampton, April 2001.
11. L.X. Peng, S. Kitipornchai, K.M. Liew, Free vibration analysis of 30. M.E. Raville, C.E.S. Ueng, Determination of natural frequencies
folded plate structures by the FSDT mesh-free method. Comput. of vibration of a sandwich plate. Exp. Mech. 7, 490–493 (1967)
Mech. 39, 799–814 (2007)
12. L.X. Peng, Free vibration analysis of symmetrically laminated
folded plate structures using an element-free Galerkin method, Publisher’s Note Springer Nature remains neutral with regard to
Mathematical Problems in Engineering 2015; Article ID 124296. jurisdictional claims in published maps and institutional affiliations.
13. S. Haldar, A.H. Sheikh, Free vibration analysis of isotropic and
composite folded plates using a shear flexible element. Finite
Elem. Anal. Des. 42(3), 208–226 (2005)

123