Applied Mathematical Modelling 39 (2015) 2489–2508

Contents lists available at ScienceDirect

Applied Mathematical Modelling

journal homepage: www.elsevier.com/locate/apm

A new hyperbolic shear deformation theory for bending and

free vibration analysis of isotropic, functionally graded,
sandwich and laminated composite plates
Amale Mahi a,⇑, El Abbas Adda Bedia b, Abdelouahed Tounsi b
a
Laboratoire des Sciences Aéronautiques, Université de Blida 1, BP 270, 09000 Blida, Algeria
b
Laboratoire des Matériaux et Hydrologie, Université de Sidi Bel Abbes, BP 89 Ben M’hidi, 22000 Sidi Bel Abbes, Algeria

a r t i c l e i n f o a b s t r a c t

Article history: A new hyperbolic shear deformation theory applicable to bending and free vibration
Received 15 March 2013 analysis of isotropic, functionally graded, sandwich and laminated composite plates is pre-
Received in revised form 29 June 2014 sented. This new theory has five degrees of freedom, provides parabolic transverse shear
Accepted 17 October 2014
strains across the thickness direction and hence, it does not need shear correction factor.
Available online 20 November 2014
Moreover, zero-traction boundary conditions on the top and bottom surfaces of the plate
are satisfied rigorously. The energy functional of the system is obtained using Hamilton’s
Keywords:
principle. Analytical solutions of deflection and stresses are obtained using Navier-type
Functionally graded plate
FG sandwich plate
procedure. Free vibration frequencies are then accurately calculated using a set of bound-
Laminated composite plate ary characteristic orthogonal polynomials associated with Ritz method. Numerical compar-
Bending isons are conducted to verify and to demonstrate the accuracy and efficiency of the present
Free vibration theory. Excellent agreement with the known results in the literature has been obtained.
Shear deformation plate theory Ó 2014 Elsevier Inc. All rights reserved.

1. Introduction

Functionally graded materials (FGMs) are a new kind of inhomogeneous composite materials which have a smooth and
continuous variation of material properties along one or more directions. At each interface, the material is chosen according
to specific applications and environment loadings. Gradually varying the material properties can prevent from interface
cracking, disbonding and residual stresses and thus maintain structural integrity to a desirable level. These highly heteroge-
neous materials are also used as thermal barrier, wear coating and corrosion resistant coating in many engineering
applications.
In 1984–1985, a group of Japanese scientist proposed, for the first time, the concept of FGM. Five years later, the first
international conference [1] was held at Sendai-City in Japan. The interest was such that a rapid progress of FGM research
in Japan was noticed from 1984 to 1996 [2]. Since then, the increased attention given by many industrial engineers and
researchers to these new composite materials was motivated by the real possibilities of improvement that can be achieved
for the mechanical properties of composites to overcome problems arising in composite structures such as delamination due
to high thermal gradients. Nowadays, these innovative materials are increasingly being used in the manufacture of aerospace
and aeronautical components designed to support high temperature environments. They are also used in the manufacture of

⇑ Corresponding author.
E-mail address: amel.mahi@gmail.com (A. Mahi).

http://dx.doi.org/10.1016/j.apm.2014.10.045
0307-904X/Ó 2014 Elsevier Inc. All rights reserved.
2490 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508

components that serve critical functions in aircrafts, space rockets, launchers and space shuttles. Further extensive applica-
tions of FGMs can be found in nuclear reactors, chemical plants, and in other applications such as biomechanical, optical,
automotive, electronic, mechanical, civil, and shipbuilding industries.
FGMs are typically made of a ceramic because of its thermal stability, good thermal shock resistance, mechanical strength,
high-temperature creep resistivity, low density, excellent chemical resistance, good stiffness-to-weight ratio, and a metal
because of its good fracture toughness. The functional association of these two high performance materials provides to avi-
ation and aerospace manufacturers the ability to face successfully a growing demand in achieving optimal performance
while satisfying stringent safety standards. A good overview of the concept of FGMs, their processing, and their applications
can be found in Refs. [3–5].
In order to solve problems of structures composed of structural elements it is necessary to find the right theory that
describes correctly the static and dynamic behaviors of the structure. This has led many researchers to develop more theories
as well as new and effective methods of resolution.
On basis of Euler–Bernoulli beam theory that relies on plane cross-sections hypothesis Kirchhoff [6] built a new theory for
thin plates in 1850, on the assumption that normals to the undeformed midplane remain straight and normal to the
deformed midplane and unstretched in length. It was then extended to thin shells by Love [7] in 1888. These two theories
together are referred as classical theory of plates and shells (also called Kirchhoff–Love theory). Since this theory neglects the
transverse shear flexibility, this leads to an overestimation of the bending stiffness and consequently an overestimation of
vibrational frequencies and underestimation of deflections.
The theory of moderately thick plates, known as first order shear deformation plate theory (FSDPT), was consolidated by
Mindlin from the works of Rayleigh [8], Timoshenko [9,10], Reissner [11] and Uflyand [12]. The works of Mindlin [13] were
based, mainly, on observations made before by Rayleigh and Timoshenko. In fact, Rayleigh stated in 1877, that it is important
to include the rotary inertia in any vibration analysis system. The works of Timoshenko [9,10] have shown that taking into
account the effects of rotary inertia and transverse shear affects the natural frequencies of bending beams. These two effects
tend to reduce the calculated resonant frequencies, due to the increase in inertia and flexibility of the system. Since the first
order theory of plates do not satisfy the boundary conditions at the top and bottom surfaces of the plate (zero traction
boundary conditions), a shear correction factor across the thickness is necessary for FSDPTs [9,10,13]. The value of the shear
correction factor depends on the geometry of the plate, on the variation of Poisson’s ratio through the thickness, on the
applied loading and on boundary conditions. Around 1970, Whitney [14], Whitney and Sun [15], Pagano [16] and Srinivas
and Rao [17] have shown, through their respective works, that the effects of transverse shear are more important for lam-
inated plates than for isotropic plates. In the literature, three approaches of approximation models for moderately thick
plates were reported [13,18–21]. Reissner’s approach [22] involves a stress–displacement field containing in-plane normal
and shear stresses. The approach of Henchy–Mindlin [8] assumes a displacement field, while the approach of Ambartsumyan
[21] assumes a field of normal and shear stresses.
It is well known that the effect of transverse shear deformation and normal stress in the thickness direction becomes
important above a certain value of the thickness-to-side ratio. To avoid the use of a shear correction factor and to rely on
more realistic assumptions than those of Kirchhoff and Mindlin theories, new mathematical models were needed. The
aim of these models is to better capture static and dynamic behaviors of plates. To address this challenging need, several
refined theories have been proposed by various authors [23–34], which are all based on the expansion of displacements
through the thickness using power series (see [35,36]). Most of the existing higher order shear deformation theories are
based on approaches of Reissner, Henchy–Mindlin and Ambartsumyan and are applicable to various plates’ models
(e.g. [20,23–28,30,31,37–41]), while some are applicable to multilayer FGM plates as well as to thick FGM plates
(e.g. [15,28,42–50]). The unified formulation (known as CUF) proposed by Carrera and Petrolo [51,52] was recently extended
by the noteworthy works of Demasi [53–57]. CUF permits a systematic assessment of a large number of plate and shell
models, whose accuracy has been well demonstrated in the literature.
New formulations involving higher order shear deformation theories have emerged in recent years for static and dynamic
analysis of beams and plates. Among all these new theories, Aydogdu [58] proposed a new exponential shear deformation
theory for laminated composite plates. El Meiche et al. [59] developed a new hyperbolic shear deformation theory for buck-
ling and vibration of FG sandwich plates. Most of new emerging higher order theories rely on five unknowns leading to five
governing equations [28,29,33,34,49,58–64]. They all neglect the thickness stretching by considering the transverse dis-
placement independent of the thickness coordinate (transverse inextensibility), contrary to new quasi-3D hyperbolic and
sinusoidal shear deformation theories developed recently by Neves et al. [65,66]. In their works, the out-of-plane displace-
ment is defined as quadratic in the thickness direction while the in-plane displacements are of hyperbolic sine or sinusoidal
type, leading to best predicting static and dynamic behaviors of plates at the cost of solving a system of nine governing equa-
tions. Ghugal and Sayyad [67] studied the free vibration of thick orthotropic plates involving trigonometric shear deforma-
tion theory. In-plane displacements are of sine type while the transverse displacement is of cosine type. According to these
authors, their theory is not only capable to produce frequencies of thickness stretch mode, but also yields the exact value of
dynamic shear correction factor from the thickness shear motion of vibration. It is interesting to note that Kant [47], when
studying bending behavior of isotropic plate, adopted a formulation that incorporates a quadratic variation of transverse
shearing strains and a linear variation of transverse normal strain through the thickness of the plate besides the use of
the three-dimensional Hooke’s law.
 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508 2491

In order to reduce the number of variables used in number of existing formulations, Shimpi [68] developed a refined
model for isotropic plates (called refined plate theory). The most interesting features of this theory are that it requires no
shear correction factor, has strong similarities with classical plate theory in many aspects such as equations of motion,
boundary conditions, and stress resultants. RPT was then successfully extended to orthotropic plates analysis and to free
vibration plates by Shimpi and Patel [69,70], to buckling analysis of plates by Kim et al. [71] and to Free vibration analysis
of FGM plates resting on Winkler–Pasternak elastic foundations by Atmane et al. [72]. The reliability of this refined theory
was also demonstrated when applied to free vibration of laminated composite plates and to buckling analysis of orthotropic
plates by Thai and Kim [73,74], and to Free vibration of functionally graded sandwich plates by Hadji et al. [75]. More
interesting extensions of four variables refined plate theory on basis of Shimpi’s works were also more recently carried
out successfully by different authors [76–82].
This paper presents a new higher order shear deformation plate theory. The in-plane displacements use a combination of
hyperbolic tangent function and polynomial ones. The thickness stretching effect is ignored. This new theory has five degrees
of freedom, provides parabolic transverse shear strains across the thickness direction and hence, does not need shear correc-
tion factor. Moreover, zero-traction boundary conditions on the top and bottom surfaces of the plate are satisfied rigorously.
The assessment of this theory is performed by numerical comparisons with the available data based on the three-
dimensional theory of linear elasticity and with those based on the well-known refined shear deformation plate theories
wherever applicable.

2. Theoretical formulation

2.1. Displacement field

In order to obtain a compact formulation which takes into account different displacement-based shear deformation
functions, the displacement field, which can capture the mechanical behavior of thin, moderately thick and thick plates,
may be written as follows:
 
 ðx; y; z; tÞ ¼ uðx; y; tÞ  z @w
u @x
þ uu ðzÞ wx ðx; y; tÞ þ @w
@x
;
h i
v ðx; y; z; tÞ ¼ v ðx; y; tÞ  z @x þ uv ðzÞ wy ðx; y; tÞ þ @y ;
@w @w ð1Þ
 y; z; tÞ ¼ wðx; y; tÞ;
wðx;
where u, v and w are displacements on the middle surface of the plate, wx and wy are rotations of transverse normal about y
axis and x axis of the plate respectively, and u is the shape function of the unified theory which determines the distribution
of the transverse shear strains and stresses (Figs. 1 and 2) through the thickness of the plate and is defined as:.
 z  3
h 4 z
uu ðzÞ ¼ uv ðzÞ ¼ uðzÞ ¼ tanh 2  2 2
: ð2Þ
2 h 3cosh ð1Þ h
As it can be seen, the shape function of the present theory uses a combination of hyperbolic tangent function and
polynomial ones.

0.5

0.4

0.3

0.2

0.1
z/h

-0.1

-0.2
Present
TSDT (Touratier)
-0.3 PSDT (Reddy)
ESDT (Karama)
-0.4

-0.5
-0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4
φ (z/h)

Fig. 1. Strain shape function of different shear deformation plate theories.

2492 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508

0.5

0.4

0.3

0.2

Present
0.1
TSDT (Touratier)
PSDT (Reddy)

z/h
0
ESDT (Karama)

-0.1

-0.2

-0.3

-0.4

-0.5
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
φ' (z/h)

Fig. 2. Transverse stress profile of different shear deformation plate theories.

It should be noted that the present shear deformation model is based on the same assumptions as those of Reddy’s theory
[28]. By doing simple derivations, it can be easily verified that the present theory accounts for parabolic variation of trans-
verse shear strains and consequently shear stresses through the thickness of the plate (see Fig. 2). Furthermore, the present
model satisfies rigorously the zero traction boundary conditions on the top and bottom surfaces of the plate:

rxz ðx; y; zÞjz¼h2 ¼ ryz ðx; y; zÞz¼h ¼ 0: ð3Þ
2

The above displacement field yields the following u dependent linear strains relations:
Normal strains:
" #
@u @w @wx @ 2 w
ex ¼ z þ uðzÞ þ 2 ;
@x @x @x @x
" #
@v @w @wy @ 2 w ð4Þ
ey ¼ z þ uðzÞ þ 2 ;
@y @y @y @y
ez ¼ 0:
Shear strains:

@u @ v @w @wx @wy @w
cxy ¼ þ  2z þ uðzÞ þ þ2 ;
@y @x @x@y @y @x @x@y
 
@u @w
cxz ¼ wx þ ; ð5Þ
@z @x
 
@u @w
cyz ¼ wy þ :
@z @y

2.2. Constitutive equations

2.2.1. Case of isotropic and sandwich FG plate

Consider two types of FGM plates of length a, constant thickness h and width b, as shown on Figs. 3 and 4:
The material properties of the FGM plate are assumed to be continually graded through the thickness direction according
to the following well-known rule of mixture:
PðzÞ ¼ Pb þ ðPt  P b ÞVðzÞ: ð6Þ
The volume fraction V(z) is defined for different FGM types as follows:
P-FGM plate (Fig. 3):
 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508 2493

z
Pt
x
h/2
-h/2
Pb
a

Fig. 3. P-FGM plate.

z metal
h3=h/2
h2 FGM x
Ceramic
h1
FGM
h0=-h/2
metal
a

Fig. 4. FG sandwich plate.

n
VðzÞ ¼ ðz=h þ 1=2Þ ; h=2  z  h=2 ð7Þ
FG Sandwich plate [83] (Fig. 4):
8 n
< ½ðz  h0 Þ=ðh1  h0 Þ ; h0  z  h1 ;
>
VðzÞ ¼ 1; h1  z  h2 ; ð8Þ
>
: n
½ðz  h3 Þ=ðh1  h3 Þ ; h2  z  h3 ;
where Pt and Pb are corresponding properties at the top and at the bottom surfaces respectively, and n is the volume fraction
index. P(z) represents the effective material property such as Young’s modulus E, Poisson’s ratio m and mass density q.
The state of the linear stress for an FGM plate is determined using the generalized Hook’s law, which is used to obtain
normal and transverse shear stresses in the following matrix form:
8 9 2 38 9
>
>
>
rx >> Q
> 6 11
Q 12 0 0 0 > > ex >>
>
> ry >> 7>
>e > >
>
< >
= 6 Q 21 Q 22 0 0 0 7>>
< y >>
=
6 7
ryz ¼ 6
6 0 0 Q 44 0 0 77 cyz ð9Þ
>
> > 6 7>
>c >
>
>
>
>
rxz >
>
>
>
>
4 0 0 0 Q 55 0 5>>
>
>
>
>
xz >
>
>
: ; :
rxy 0 0 0 0 Q 66 cxy ;
where the well-known engineering constants Q ij , which are functions of coordinate z, are given by:
EðzÞ EðzÞ
Q 11 ¼ Q 22 ¼ ; Q 12 ¼ Q 21 ¼ mðzÞQ 11 ; Q 44 ¼ Q 55 ¼ Q 66 ¼ : ð10Þ
1  m2 ðzÞ 2½1 þ mðzÞ

2.2.2. Case of laminated composite plate

The composite laminated plate is constituted by a finite number of orthotropic layers. Constitutive equations for a kth
layer, in the orthotropic local coordinate system, have the same form as those of above FGM plate. The engineering constants
defined in the material axes of the layer are given in this case by [84]:
E1 m12 E2 E2
Q 11 ¼ ; Q 12 ¼ ; Q 22 ¼ ;
1  m12 m21 1  m12 m21 1  m12 m21
ð11Þ
E2
Q 66 ¼ G12 ; Q 44 ¼ G23 ; Q 55 ¼ G13 ; m21 ¼ m12 ;
E1
in which E1, E2, m12, m21, G12, G23 and G13 are the material properties of the lamina.
The stress–strain relations in the global (x, y, z) coordinate system can be written as:
8 9 ðkÞ 2 ðkÞ ðkÞ 3 8 9
>
> rx >> Q 11 Q 12 Q 16 0 0 >
> ex >>
>
> >
> 6 7 >
> >
>
>
> ry >> 6 Q 12 Q 22 Q 26 0 0 7 > > ey >>
< = 6 7 < =
rxy ¼6
6 Q 16 Q 26 Q 66 0 0 7 7 cxz : ð12Þ
>
> >
> 6 7 > > >
>
>
> >
ryz > 4 0 0 0 Q 44 Q 45 5 > > >
cyz >
>
> >
> >
> >
>
:
rxz ; 0 0 0 Q 45 Q 55
:
cxz ;
2494 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508

The expressions of the transformed material constants Q ij can be found in Ref. [84]

2.3. Energy equations

The strain energy Ud of the plate is expressed by:

Z Z Z h=2
1 1
Ud ¼ rij eij dm ¼ ðrx ex þ ry ey þ ryz cyz þ rxz cxz þ rxy cxy ÞdAdz: ð13Þ
2 m 2 A h=2

The kinetic energy Ec of freely vibrating plate is given by:

Z Z Z (   2  2 )
1 @ui @ui 1 h=2  2
@u @ v 
@w
Ec ¼ qðzÞ dm ¼ qðzÞ þ þ dAdz: ð14Þ
2 m @t @t 2 A h=2 @t @t @t

The work Vq done by the transverse distributed load q can be written as:
Z
Vq ¼  qwdA: ð15Þ
A

2.4. Forces and moments resultants

2.4.1. Case of isotropic and sandwich FG plate

The forces and moments resultants acting on the FGM plate are defined as:
Z h=2 Z h=2
ðNx ; Mx ; M ax Þ ¼ ð1; z; uÞrx dz; ðNy ; M y ; May Þ ¼ ð1; z; uÞry dz;
h=2 h=2
Z h=2 Z h=2
ð16Þ
@u
ðQ xy ; M xy ; M axy Þ ¼ ð1; z; uÞrxy dz; ðQ axz ; Q ayz Þ ¼ ðrxz ; ryz Þ dz:
h=2 h=2 @z
Using expressions (9) and (10) in Eqs. (16), the forces and moments resultants acting on the FGM plate, which are related
to displacements and rotations of middle surface of the plate, can be written in matrix form as follows:
8 9 2 38 9
>
> Nx >
> A11 A12 B11 B12 Ba11
Ba12 > u;x >
>
> >
> >
> >
>
>
> 6 >
> 7>
> >
>
>
> Ny 6 A12
>
> A22 B12 B22 Ba12 Ba22 7 >
> v >
>
>
> 6 >
> 7> ;y >
>
> 6 >
> 7>
>
>
>
>
>
>
<M 6B>
= B12 a
D11 D12 D11 D12 7 a 7> < w;xx > =
x 6 11
¼6
6
7
7 ;
>
> My >> 6 B12 B22 D12 D22 Da12 Da22 7> > w;yy > >
>
> >
> 6 7>
> >
>
>
> > > >
> a>> 6 a 7> >
>
> M >
> 6B Ba12 Da11 Da12 Daa Daa 7>
> wx;x þ w;xx >>
>
> x >
> 4 11 11 12 5>>
>
>
>
>
ð17Þ
>
> >
> >
: >
;
: Ma ;
y Ba12 Ba22 a a
D12 D22 D12 D22 aa aa
wy;y þ w;yy
8 9 2 38 9
> Q xy > A66 B66 Ba66 > u;y þ v ;x > ( a ) " a #( )
>
< >
= 6 >
< >
= Q xz A55 0 wx þ w;x
7
M xy ¼ 6 B D66 a 7
D66 5 2w ; ¼ ;
> 4 66
;xy
>
> > > a
Q yz Aa44 wy þ w;y
: a > ; >
: >
; 0
M xy Ba66 Da66 Daa66
wx;y þ wy;x þ 2w;xy

in which, the extensional, coupling, bending and transverse shear rigidities Aij, Bij , Dij and Aaij respectively, are defined as
follows:
  Z h=2
Aij ; Bij ; Baij ¼ Q ij ð1; z; uÞdz; ði; j ¼ 1; 2; 6Þ;
h=2
  Z h=2
Dij ; Daij ; Daa
ij ¼ Q ij ðz2 ; uz; u2 Þdz; ði; j ¼ 1; 2; 6Þ; ð18Þ
h=2
Zh=2  2
@u
Aaij ¼ Q ij dz; ði; j ¼ 4; 5Þ:
h=2 @z

2.4.2. Case of laminated composite plate

Similarly, substituting expressions (11) and (12) into Eqs. (16), the forces and moments resultants acting on the laminated
composite plate, which are related to displacements and rotations of middle surface of the plate, can be written in matrix
form as follows:
 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508 2495

8 9 2 38 9
>
> Nx > > A11 A12 A16 B11 B12 B16 Ba11 Ba12 Ba16 > u;x >
>
> > > >
>
> Ny > >
> 6 A12 A22 A26 B12 B22 B26 Ba12 Ba22 Ba26 7
>
> v ;y >
>
>
> >
> 66 7>>
>
>
>
>
>
>
>
>
Q xy > >
> 6
> A16 A26 A66 B16 B26 B66 Ba16 Ba26 Ba66 7
7>
>
>
> u;y þ v ;x >
>
>
>
>
> Mx > > 6 7 > >
>
< >
= 6 B B12 B16 D11 D12 D16 Da11 Da12 Da16 7>> w;xx >
>
6 11 7< =
My ¼6
6 12 B22 B26 D12 D22
B D26 Da12 Da22 Da26 7
7> w ;yy ;
>
> >
> 6 > >
>
>
>
>
M xy >
> 6 B16 B26 B66 D16 D26 D66 Da16 Da26 Da66 7
7>> 2w;xy >
> ð19Þ
>
> a >>
> 6 a 7>> >
>
> Mx >
> > 6 B Ba12 Ba16 Da11 Da12 Da16 Daa Daa Daa 7>>
> wx;x þ w;xx >
>
>
>
> a >> 6 11 11 12 16 7>
aa 5>
>
>
> > 4B a
Ba22 Ba26 Da12 Da22 Da26 Daa Daa D66 > >
> My >
>
> > 12 16 26 >
> wy;y þ w;yy >
>
: a > ; a
: ;
M B Ba26 Ba66 Da16 Da26 Da66 Daa Daa Daa wx;y þ wy;x þ 2w;xy
( xy) 12
 
16 26 66

Q axz Aa55 Aa45 wx þ w;x

a ¼ ;
Q yz Aa45 Aa44 wy þ w;y

where
  Z h=2
Aij ; Bij ; Baij ¼ Q ij ð1; z; uÞdz; ði; j ¼ 1; 2; 6Þ;
h=2
  Z h=2
Dij ; Daij ; Daa
ij ¼ Q ij ðz2 ; uz; u2 Þdz; ði; j ¼ 1; 2; 6Þ; ð20Þ
h=2
Z h=2  2
@u
Aaij ¼ Q ij dz; ði; j ¼ 4; 5Þ:
h=2 @z
Here, the additional higher-order coupling and bending rigidities Baij , Daij and Daa
ij are related to the shape function u(z)
which was introduced in the displacement field expression. It can also be noted that the higher-order moments M ax , M ay
and M axy are also related to the same shear distribution function u(z).
The inertia terms obtained upon developing kinetic energy equation (14) are expressed as follows:
Z h=2 Z h=2
ðI1 ; I2 ; I3 Þ ¼ qðzÞð1; z; z2 Þdz; ðIa2 ; Ia3 ; Iaa
3 Þ ¼ qðzÞuðzÞð1; z; uÞdz; ð21Þ
h=2 h=2

where I1 ; I2 ; I3 are axial, coupling and rotary inertia, respectively. Ia2 ; Ia3 ; Iaa
3 are higher-order axial, coupling and rotary
inertia, respectively.

3. Governing equations

The governing equations and the associated boundary conditions are derived using the dynamic version of principle of
virtual work known as Hamilton’s principle:
Z t
0¼ ðdU d þ dV q  dEc Þ: ð22Þ
0

By integrating by parts with respect to t as well as x and y, and by collecting the coefficients of the virtual displacements
du, dv, dw, dwx and dwy, we arrive at the following governing equations of motion in terms of stress resultants:

@Nx @Q xy @2u @ 2 wx a  @2w

du : þ ¼ I1 2 þ Ia2 2
þ I2  I2 ; ð23aÞ
@x @y @t @t @x@t 2

@Ny @Q xy @2v @ 2 wy a  @2w

dv : þ ¼ I1 2 þ Ia2 þ I 2  I 2 ; ð23bÞ
@y @x @t @t 2 @y@t2

@2  @2   @2   @Q a @Q ayz
dw : 2
Mx  Max þ 2 M xy  Maxy þ 2 My  May þ xz
þ þq
@x @x@y @y @x @y
! ! !
a
 @3u @3v a aa
 @ 3 wx @ 3 wy @2w a aa
 @4w @4w
¼ I2  I2 þ þ I3  I 3 þ þ I1 2  I3  2I3 þ I3 þ ; ð23cÞ
@x@t 2 @y@t 2 @x@t2 @y@t2 @t @x2 @t2 @y2 @t 2

a
@M ax @Mxy @2u @ 2 wx aa  @3w
dwx : þ  Q axz ¼ Ia2 2 þ Iaa
3 2
þ I3  Ia3 ; ð23dÞ
@x @y @t @t @x@t 2

@M ay @Maxy @2v @ 2 wy aa  @3w

dwy : þ  Q ayz ¼ Ia2 2 þ Iaa
3 2
þ I3  Ia3 : ð23eÞ
@y @x @t @t @y@t 2
2496 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508

The system of Eq. (23) can be expressed in terms of displacements and rotations (u, v, w, wx, wy) by simply substituting for
the stress resultants from Eq. (17) or Eq. (19).

4. Solution procedure

4.1. Ritz method

Approximation methods are used when exact solutions of a problem of mechanics of deformable solids cannot be easily
obtained. The Ritz method is a powerful approximate method based on a variational approach. It is an efficient method for
obtaining solutions accurately with few computational efforts. According to this method, the admissible displacements and
rotations of the middle surface of the plate are written in the form of an infinite double series of continues functions P Fi ðnÞ
and P Fj ðgÞ defined for each variable field:
X
1 X
1
ðU; V; W; Wx ; Wy Þ ffi ða ~ ; ~c ; d
~ij ; b ~ ; ~e ÞPF ðnÞPF ðgÞ sinðxtÞ; ð24Þ
ij ij ij ij i j
i¼1 j¼1

where a ~ij , ~cij , d

~ij , b ~ij and ~eij are undetermined Ritz parameters, P F ðnÞ and P F ðgÞ are characteristic orthogonal polynomials. x is
i j
the angular natural frequency of vibration.
The free vibration problem of the plate is solved by truncating the series (24) and substituting them into energy equations
(13) and (14) to obtain U max d and Emax
c respectively with the periodic component eliminated and by minimizing the Lagrang-
ian energy functional L with respect to the unknown Ritz parameters:
@L @L @L @L @L
¼ ¼ ¼ ¼ ¼ 0; i ¼ 1; 2; . . . ; M; j ¼ 1; 2; . . . ; N; ð25Þ
@a ~
~ij @ b @ ~
c ij ~
@ dij @ ~eij
ij

in which the total energy L in terms of maximum kinetic energy Emax

c and maximum strain energy U max
d , stored in a vibratory
cycle, is expressed as:

L ¼ Emax
c  U max
d : ð26Þ
The minimizing conditions (25) along with energy functional equation (26) yield the following generalized eigenvalue
problem:

½K  x2 Mfqg ¼ f0g; ð27Þ

where mass matrix K and stiffness matrix K are of order (5(M  N)  5(M  N)).
It should be noted that each element of the vector q is composed of sub-matrix of Ritz parameters of order
((M  N)  (M  N)). Also, the static bending problem can be recovered by setting x = 0 in Eq. (27).
In this paper, various boundary conditions are considered, which can be satisfied by employing appropriate admissible
Ritz functions. To this end, the Gram–Schmidt orthogonalization process along with boundary polynomial functions is
chosen to obtain a complete set of trial functions [85,86]:

P1 ðnÞ ¼ XðnÞ; P1 ðgÞ ¼ YðgÞ;

P2 ðfÞ ¼ ðf  B2 ÞP1 ðfÞ; ð28Þ
Pk ðfÞ ¼ ðf  Bk ÞPk1 ðfÞ  Ck Pk2 ðfÞ; f ¼ n; g; k > 2;

with
R1 R1
WðfÞ½Pk1 ðfÞ2 fdf WðfÞPk1 ðfÞPk2 ðfÞfdf
Bk ¼ R01 ; Ck ¼ 0
R1 ; f ¼ n; g; ð29Þ
0
WðfÞ½Pk1 ðfÞ2 df 0
WðfÞ½Pk2 ðfÞ2 df

where WðfÞ is a weighting function used in the orthogonalization process. X ðnÞ and YðgÞ are the appropriate admissible
boundary polynomial Ritz functions satisfying the boundary conditions and having the following form:

XðnÞ ¼ np ð1  nÞq ; YðgÞ ¼ gr ð1  gÞs ; n ¼ x=a; g ¼ y=b: ð30Þ

Here, the positive integers p and r take the values of 0, 1, or 2 in order to define free, simply supported or clamped conditions,
respectively at f = 0 and g = 0. Similarly, q and s take the values of 0, 1, or 2 in order to define free, simply supported or
clamped conditions, respectively at f = 1 and g = 1.
By increasing sufficiently the number of these trial functions in series (24), one can guaranty a convergence of upper-
bound approximate solutions towards the exact natural frequencies of vibration.

4.2. Boundary conditions

The plate can have the following boundary conditions or a combination of them:
 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508 2497

Simply supported edges:

 

W ¼ Nx ¼ Mx ¼ Max ¼ 0n¼0;1 and W ¼ Ny ¼ My ¼ M ay ¼ 0 : ð31Þ
g¼0;1

Clamped edges:

U ¼ V ¼ W ¼ Wx ¼ 0jn¼0;1 and U ¼ V ¼ W ¼ Wy ¼ 0g¼0;1 : ð32Þ

Free edges:
 

Q axz ¼ Nx ¼ Mx ¼ M ax ¼ 0n¼0;1 and Q ayz ¼ Ny ¼ My ¼ M ay ¼ 0 : ð33Þ
g¼0;1

4.3. Analytical solutions for FG plates

As the plate have a simple geometry (square or rectangle) and is subjected only to simply supported boundary conditions,
the determination of displacements and stresses can be performed very accurately using Navier approach. This method
requires that the displacements and rotation must be expanded in Fourier series to give:
X
1 X
1
uðx; y; tÞ ¼ U mn cos ax sin by sin xt;
m¼1 n¼1
X1 X 1
v ðx; y; tÞ ¼ V mn sin ax cos by sin xt;
m¼1 n¼1
X1 X 1
wðx; y; tÞ ¼ W mn sin ax sin by sin xt; ð34Þ
m¼1 n¼1
X1 X 1
wx ðx; y; tÞ ¼ X mn cos ax sin by sin xt;
m¼1 n¼1
X1 X 1
wy ðx; y; tÞ ¼ Y mn sin ax sin by sin xt:
m¼1 n¼1

The applied transverse load q is also expanded in the double Fourier series as:
X
1 X
1
qðx; yÞ ¼ Q mn sin ax sin by; ð35Þ
m¼1 n¼1

where a = mp/a and b = np/b. Umn, Vmn, Wmn, Xmn, Ymn, and Qmn are the unknown amplitudes of the series expansions. x is the
natural frequency of vibration.
The coefficients Qmn are given by:
Z Z (
a b q0 for sinusoidal load
4
Q mn ¼ qðx; yÞ sin ax sin by dxdy ¼ 16q0 ð36Þ
ab 0 0 p2 mn for uniform load

The expressions (34) satisfy automatically the following boundary conditions:

 

V ¼ W ¼ wy ¼ Nx ¼ Mx ¼ M ax ¼ 0n¼0;1 and U ¼ W ¼ wx ¼ W ¼ Ny ¼ My ¼ M ay ¼ 0 : ð37Þ
g¼0;1

Using Eqs. (17) and substituting series (34) and (35) into Eqs. (23), the Navier solutions can be easily obtained from the
resulting system of algebraic equations:
02 3 2 318 9 8 9
k11 k12 k13 k14 k15 m11 m12 m13 m14 m15 >
> U mn >
> >
> 0 >
>
B6 7 6 7C>> > >
> > >
>
B6 k21 k22 k23 k24 k25 7 6 m21 m22 m23 m24 m25 7C>>
< V mn >
>
=
>
>
< 0 >
>
=
B6 7 6 7C
B6 k31 k32 k33 k34 k35 7  x 2 6 m31 m32 m33 m34 m35 7 C W mn ¼ Q mn ; ð38Þ
B6 7 6 7C> >
B6 7 6 7C> > >
> >
>
@4 k41 k42 k43 k44 k45 5 4 m41 m42 m43 m44 m45 5A>>
> X > > >
> 0 >
>
> mn >
: > >
; >
:
>
>
;
k51 k52 k53 k54 k55 m51 m52 m53 m54 m55 Y mn 0
where the coefficients of each matrix are given in Appendix A.

4.4. Analytical solutions for antisymmetric cross-ply laminates

For simply supported laminated plates, the equations of motion (23) can be solved analytically, when lamination scheme
is of antisymmetric cross-ply [0°/90°/0°/90°. . .] or symmetric cross-ply [0°/90°. . .]s type. In the case of antisymmetric
cross-ply laminates, the Navier solutions exist if the following stiffnesses are identically zero:
2498 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508

a
A16 ¼ A26 ¼ B16 ¼ Ba16 ¼ B26 ¼ Ba26 ¼ D16 ¼ Da16 ¼ Daa a aa
16 ¼ D26 ¼ D26 ¼ D26 ¼ A45 ¼ 0: ð39Þ
Similarly, by using Eq. (19) and by substituting series (34) and (35) into Eq. (23), the analytical solutions can also be easily
obtained from Eq. (38). The corresponding coefficients applicable to this case are also given in Appendix A.

5. Numerical results and discussion

The present hyperbolic shear deformation plate theory is applied to several static bending and free vibration examples
based on 2-D and 3-D elasticity solutions, involving various boundary conditions and aspect ratios. To this end, some numer-
ical examples are presented in order to verify the accuracy and efficiency of the present theory.

5.1. Convergence examination

Table 1 shows the fundamental frequency parameter FGM square thick plate (Al/ZrO2, n = 10, a/h = 5) for different bound-
ary conditions. The solutions presented in Ref. [87] have been performed using 3-D linear elasticity theory and are used as
basis for ascertaining the convergence of present solution procedure. All results are determined by truncating the series (24).
Table 1 compares the frequencies calculated by using different numbers of terms in the series expansion to benchmark
results of Ref. [87]. By using M = N = 6, it is shown that the results converge well. In addition, higher accuracy can be obtained
by adding more terms in the series solutions (24). Here, the percentage error is determined as follows:
 
Published  Present
%Error ¼  100: ð40Þ
Published
The average relative error is about 0.148% in all cases, which is acceptable for most engineering applications. The excel-
lent numerical stability of solutions is also evident. Therefore, and for the convenience of calculation, the displacement and
the rotation expansions will be simply truncated to M = N = 8 in all subsequent comparisons. It is noted that, if the last
sequence number of the modes used as basis for comparison taken from the available literature exceed the default trunca-
tion, the number of terms of series (24) is increased accordingly to maintain the accuracy up to last mode number shown.

5.2. Bending of FGM plate

In this section, the present shear hybrid model is tested for static flexure problem of FG plate. To this end, a comparison
study is conducted and the obtained results are compared with quasi-3D solutions given by Wu et al. [88]. The through
thickness variation of dimensionless displacement and stresses of thick (a/h = 5) FG plate (Al/Al2O3) for n = 4 is depicted
on Fig. 5. The plate material and various nondimensional parameters used can be found in Ref. [88]. Fig. 5 shows a good
agreement between results. However, the difference between the results observed on Fig. 5d is due to the fact that the pres-
ent model neglects the stretching effect which affects the evaluation of the transverse shear stresses. As quasi-3D models
[88] take into account this thickness stretching, they provide a better prediction of the transverse shear stress than the
present theory.

5.3. Bending of FGM Sandwich plate

The aim of this section is to demonstrate the accuracy and efficiency of the present theory in predicting deflection and
stresses of FG (Al/ZrO2) square sandwich plate for various volume fraction power indexes and schemes. Quasi-3D solutions
provided by Zenkour [83] are used as benchmark results and same material is considered in the calculations. This enhanced
2-D plate theory takes into account the stretching effect and gives more accurate results that are very close to 3-D elasticity
ones. Tables 2–4 compare the maximum dimensionless deflection and stresses of a moderate (a/h = 10) thick FG square
sandwich plate under sinusoidal distributed load. The various nondimensional parameters used can be found in Ref. [83].
To have the same comparison basis (displacement field and method of resolution), we present in Tables 2–4 numerical

Table 1 pffiffiffiffiffiffiffiffiffiffiffiffi

 ¼ x a2 =h qt =Et , for square FG plates (Al/ZrO2, n = 10, a/h = 5) for different classical boundary
Convergence study of fundamental frequency parameter, x
conditions.

Source Grid size SSSS CCCC FFFF SCSC CFCF CSCF

3D-Chebyshev [87] 222 1.5937 2.6076 1.0440 2.1377 1.8049 1.8340
333 1.4662 2.1840 1.0440 1.8520 1.4577 1.5126
444 1.3711 2.0976 0.9327 1.7635 1.3957 1.4491
555 1.3711 2.0880 0.9327 1.7578 1.3852 1.4387
666 1.3711 2.0832 0.9219 1.7549 1.3820 1.4352
Present 66 1.3694 2.0853 0.9231 1.7570 1.3790 1.4323
%Error 0.121% 0.102% 0.127% 0.119% 0.219% 0.203%
 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508 2499

Fig. 5. Through thickness variation of dimensionless displacement and stresses of thick FGM plate (Al/Al2O3, n = 4, a/h = 5).

results obtained by using the present theory, the third shear deformation (TSPDT) of Reddy and the sinusoidal shear defor-
mation theory (SSDPT) of Karama. Here, it is clear that the present hybrid model gives slightly more accurate results than
those obtained by using SSDPT and TSDPT. Tables 2–4 shows that when the volume fraction power index n increase, the
maximum vertical deflection, axial stress, and shear stress increase. This is because the rigidity of the plate is reduced when
n increases due to the greater proportion of metal whose Young’s modulus is lower than that of ceramic. The effect of
increasing of core thickness with respect to the total thickness of the plate is to decrease the deflection and stresses because
the proportion of ceramic becomes more predominant and the plate is more rigid. Also, for the non-symmetric (2-2-1) FGM
plat, the 2-D plate theories give the highest percent error for axial and shear stresses. In this case, the need to enhance these
plates’ theories is obvious.

5.4. Bending of laminated composite plate

In this example, symmetric cross-ply (0°/90°/90°/0°) square laminates under sinusoidal distributed load are considered.
The computation of bending and stress results is performed for various values of thickness parameter a/h (ranging from thick
to thin plate). 3D-Elasticity solutions obtained by Pagano and Hatfield [89] are used as benchmark results. The same material
is used in the calculations. The parabolic shear deformation theory results of Reddy [28] are also included in the comparison
study. Table 5 shows dimensionless maximum center deflection and stresses in four-layer (0°/90°/90°/0°) square laminates
subject to sinusoidal distributed load under simply supported boundary conditions. The various nondimensional parameters
used can be found in Ref. [64]. It is seen from Table 5 that the present theory is capable to produce more accurate results than
ones obtained by employing Reddy’s theory [28]. It seems that a combination of hyperbolic tangent function and polynomial
ones produce a better model than Reddy’s ones. However, the shear stress values are not accurate enough because the 2-D
theories make simplifying assumptions on displacement field which are unnecessary in model based on 3-D elasticity
theory. The simplifying assumptions can have a significant effect on the accuracy of results for a given problem. The only
obvious advantage is that they can improve the computational cost because of reducing the number of unknowns and
governing equations and make the shear model simple to use.
2500 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508

Table 2
Comparison of maximum dimensionless deflection of the FG (Al/ZrO2) square sandwich plate (a/h = 10).

n Source rz w(a/2, b/2, 0)

1-0-1 2-1-2 1-1-1 2-2-1 1-2-1
1 Quasi-3D [83] –0 0.32001 0.30275 0.28867 0.27760 0.26815
Present (HSDPT) =0 0.32344 0.30620 0.29191 0.28080 0.27092
Present (SSDPT) =0 0.32349 0.30624 0.29194 0.28082 0.27094
Present (TSDPT) =0 0.32358 0.30632 0.29199 0.28085 0.27093
%Error (HSDPT) 1.072% 1.140% 1.122% 1.153% 1.033%
%Error (SSDPT) 1.087% 1.153% 1.133% 1.160% 1.040%
%Error (TSDPT) 1.116% 1.179% 1.150% 1.171% 1.037%
2 Quasi-3D [83] –0 0.36891 0.34737 0.32816 0.31152 0.29874
Present (HSDPT) =0 0.37309 0.35210 0.33275 0.31607 0.30259
Present (SSDPT) =0 0.37319 0.35218 0.33280 0.31611 0.30260
Present (TSDPT) =0 0.37335 0.35231 0.33289 0.31617 0.30263
%Error (HSDPT) 1.133% 1.362% 1.399% 1.461% 1.289%
%Error (SSDPT) 1.160% 1.385% 1.414% 1.473% 1.292%
%Error (TSDPT) 1.204% 1.422% 1.441% 1.493% 1.302%
5 Quasi-3D [83] –0 0.40532 0.38612 0.36546 0.34361 0.32966
Present (HSDPT) =0 0.40891 0.39147 0.37120 0.34944 0.33471
Present (SSDPT) =0 0.40905 0.39160 0.37128 0.34950 0.33474
Present (TSDPT) =0 0.40927 0.39183 0.37145 0.34960 0.33480
%Error (HSDPT) 0.886% 1.386% 1.571% 1.697% 1.532%
%Error (SSDPT) 0.920% 1.419% 1.593% 1.714% 1.541%
%Error (TSDPT) 0.975% 1.479% 1.639% 1.743% 1.559%
10 Quasi-3D [83] –0 0.41448 0.39856 0.37924 0.35577 0.34259
Present (HSDPT) =0 0.41738 0.40365 0.38520 0.36193 0.34810
Present (SSDPT) =0 0.41751 0.40380 0.38531 0.36200 0.34815
Present (TSDPT) =0 0.41772 0.40407 0.38551 0.36212 0.34824
%Error (HSDPT) 0.700% 1.277% 1.572% 1.731% 1.608%
%Error (SSDPT) 0.731% 1.315% 1.601% 1.751% 1.623%
%Error (TSDPT) 0.782% 1.382% 1.653% 1.785% 1.649%

Table 3
Comparison of dimensionless axial stress of the FG (Al/ZrO2) square sandwich plate (a/h = 10).

n Source rz rx(a/2, b/2, h/2)

1-0-1 2-1-2 1-1-1 2-2-1 1-2-1
1 Quasi-3D [83] –0 1.57004 1.48833 1.41781 1.30907 1.31204
Present (HSDPT) =0 1.54512 1.46367 1.39479 1.28926 1.29248
Present (SSDPT) =0 1.54487 1.46342 1.39453 1.28900 1.29151
Present (TSDPT) =0 1.54416 1.46274 1.39383 1.28829 1.29222
%Error (HSDPT) 1.587% 1.657% 1.624% 1.513% 1.491%
%Error (SSDPT) 1.603% 1.674% 1.642% 1.533% 1.565%
%Error (TSDPT) 1.648% 1.719% 1.691% 1.587% 1.511%
2 Quasi-3D [83] –0 1.81509 1.72030 1.62591 1.46372 1.47421
Present (HSDPT) =0 1.78454 1.68751 1.59466 1.43772 1.44786
Present (SSDPT) =0 1.78429 1.68727 1.59440 1.43745 1.44758
Present (TSDPT) =0 1.78357 1.68660 1.59370 1.43671 1.44684
%Error (HSDPT) 1.683% 1.906% 1.922% 1.776% 1.787%
%Error (SSDPT) 1.697% 1.920% 1.938% 1.795% 1.806%
%Error (TSDPT) 1.737% 1.959% 1.981% 1.845% 1.857%
5 Quasi-3D [83] –0 1.97912 1.91504 1.82018 1.60953 1.63906
Present (HSDPT) =0 1.95101 1.87768 1.78226 1.57700 1.60539
Present (SSDPT) =0 1.95077 1.87748 1.78202 1.57672 1.60510
Present (TSDPT) =0 1.95003 1.87686 1.78138 1.57598 1.60437
%Error (HSDPT) 1.420% 1.951% 2.083% 2.021% 2.054%
%Error (SSDPT) 1.432% 1.961% 2.096% 2.038% 2.072%
%Error (TSDPT) 1.470% 1.994% 2.132% 2.084% 2.116%
10 Quasi-3D [83] –0 2.00692 1.97075 1.89162 2.18558 1.67350
Present (HSDPT) =0 1.98458 1.93485 1.84994 1.62920 1.67097
Present (SSDPT) =0 1.98431 1.93466 1.84972 1.62893 1.67070
Present (TSDPT) =0 1.98347 1.93407 1.84911 1.62818 1.66998
%Error (HSDPT) 1.113% 1.822% 2.203% 25.457% 0.151%
%Error (SSDPT) 1.127% 1.831% 2.215% 25.469% 0.167%
%Error (TSDPT) 1.168% 1.861% 2.247% 25.504% 0.210%
 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508 2501

Table 4
Comparison of dimensionless transverse shear stress of the FG (Al/ZrO2) square sandwich plate (a/h = 10).

n Source rz rxz(0, b/2, 0)

1-0-1 2-1-2 1-1-1 2-2-1 1-2-1
1 Quasi-3D [83] –0 0.36531 0.34366 0.32853 0.31785 0.30845
Present (HSDPT) =0 0.30315 0.28166 0.27215 0.27101 0.26436
Present (SSDPT) =0 0.29907 0.27774 0.26809 0.26680 0.25258
Present (TSDPT) =0 0.29203 0.27104 0.26117 0.25951 0.26004
%Error (HSDPT) 17.016% 18.041% 17.161% 14.737% 14.294%
%Error (SSDPT) 18.133% 19.182% 18.397% 16.061% 18.113%
%Error (TSDPT) 20.060% 21.131% 20.503% 18.355% 15.695%
2 Quasi-3D [83] –0 0.41778 0.38601 0.36417 0.34824 0.33543
Present (HSDPT) =0 0.33672 0.29772 0.28182 0.28031 0.26966
Present (SSDPT) =0 0.33285 0.29422 0.27807 0.27627 0.26543
Present (TSDPT) =0 0.32622 0.28838 0.27188 0.26939 0.25834
%Error (HSDPT) 19.403% 22.872% 22.613% 19.507% 19.608%
%Error (SSDPT) 20.329% 23.779% 23.643% 20.667% 20.869%
%Error (TSDPT) 21.916% 25.292% 25.343% 22.642% 22.982%
5 Quasi-3D [83] –0 0.46890 0.42723 0.39918 0.37791 0.36234
Present (HSDPT) =0 0.39782 0.32215 0.29466 0.29271 0.27548
Present (SSDPT) =0 0.39370 0.31930 0.29150 0.28895 0.27153
Present (TSDPT) =0 0.38634 0.31454 0.28643 0.28265 0.26512
%Error (HSDPT) 15.159% 24.596% 26.184% 22.545% 23.972%
%Error (SSDPT) 16.038% 25.263% 26.975% 23.540% 25.062%
%Error (TSDPT) 17.607% 26.377% 28.245% 25.207% 26.831%
10 Quasi-3D [83] –0 0.49051 0.44435 0.41385 0.39045 0.37390
Present (HSDPT) =0 0.44679 0.33954 0.30301 0.30057 0.27867
Present (SSDPT) =0 0.44158 0.33693 0.30018 0.29693 0.27492
Present (TSDPT) =0 0.43206 0.33242 0.29566 0.29083 0.26894
%Error (HSDPT) 8.913% 23.587% 26.783% 23.020% 25.469%
%Error (SSDPT) 9.975% 24.175% 27.466% 23.952% 26.472%
%Error (TSDPT) 11.916% 25.190% 28.559% 25.514% 28.072%

Table 5
Dimensionless deflections and stresses in four-layer (0°/90°/90°/0°) square laminates under sinusoidal distributed load (b = a).

a/h Source w rx ry rxy rxz ryz

4 3D-Elasticity [89] 1.954 0.720 0.663 0.047 0.219 0.291
TSDPT [28] 1.893 0.665 0.632 0.044 0.206 0.239
HTSDT [64] 1.894 0.664 0.631 0.044 0.206 0.239
Present (HSDPT) 1.916 0.689 0.636 0.045 0.222 0.250
%Error (Present) 1.958% 4.346% 4.139% 2.677% 1.219% 14.024%
%Error (Reddy) 3.122% 7.625% 4.646% 5.782% 5.753% 17.904%
10 3D-Elasticity [89] 0.743 0.559 0.401 0.028 0.301 0.196
TSDPT [28] 0.715 0.546 0.389 0.027 0.264 0.153
HTSDT [64] 0.715 0.545 0.388 0.027 0.264 0.153
Present (HSDPT) 0.723 0.550 0.392 0.027 0.287 0.161
%Error (Present) 2.746% 1.603% 2.362% 1.491% 4.811% 17.806%
%Error (Reddy) 3.809% 2.397% 3.042% 2.545% 12.292% 21.888%
20 3D-Elasticity [89] 0.517 0.543 0.308 0.023 0.328 0.156
TSDPT [28] 0.506 0.539 0.304 0.023 0.283 0.123
HTSDT [64] 0.507 0.539 0.304 0.023 0.282 0.123
Present (HSDPT) 0.509 0.540 0.305 0.023 0.307 0.129
%Error (Present) 1.557% 0.492% 0.916% 0.348% 6.262% 17.019%
%Error (Reddy) 2.128% 0.681% 1.201% 0.870% 13.872% 20.897%
100 3D-Elasticity [89] 0.439 0.539 0.276 0.022 0.337 0.141
TSDPT [28] 0.434 0.538 0.271 0.021 0.290 0.112
HTSDT [64] 0.435 0.539 0.271 0.021 0.289 0.112
Present (HSDPT) 0.435 0.539 0.271 0.021 0.316 0.117
%Error (Present) 0.741% 0.072% 1.884% 0.972% 6.323% 17.135%
%Error (Reddy) 0.958% 0.148% 1.884% 1.389% 14.036% 20.780%
2502 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508

Table 6 pffiffiffiffiffiffiffiffiffiffiffiffi
 ¼ xðb2 =p2 Þ qh=D, of SSSS isotropic square plate.
Comparison of first eight frequency parameters, x

h/b Source Mode number

1 2 3 4 5 6 7 8
0.01 Present 1.9993 4.9955 4.9955 7.9885 9.9839 9.9839 12.9711 12.9712
Classical theory [90] 2 5 5 8 10 10 13 13
3-D Chebyshev–Ritz [96] 1.9972 4.9999 4.9999 7.9996 9.9995 9.9995 13.0000 13.0000
3-D DQ [97] 1.9952 4.9977 5.0081 7.9957 10.0240 10.0290 13.0110 13.0120
0.1 Present 1.9319 4.6098 4.6098 6.5234 6.5234 7.0755 8.6238 8.6238
Mindlin theory [92] 1.9311 4.6048 4.6048 – – 7.0637 8.6049 8.6049
Reddy–Ritz [95] 1.9317 4.6088 4.6088 6.5233 6.5233 7.0731 8.6189 8.6189
3-D Exact solutions [98] 1.9342 4.6222 4.6222 – – 7.1030 8.6617 8.6617
3-D Ritz [99] 1.9342 4.6222 4.6222 6.5234 6.5234 7.1030 8.6617 8.6617
3-D DQ [100] 1.9342 4.6222 4.6222 6.5234 6.5234 7.1030 8.6617 8.6617
3-D Chebyshev–Ritz [96] 1.9342 4.6222 4.6222 6.5234 6.5234 7.1030 8.6617 8.6617
3-D DQ [97] 1.9342 4.6250 4.6250 6.5234 6.5234 7.1064 8.6932 8.6932
0.2 Present 1.7689 3.2617 3.2617 3.8724 3.8724 4.6127 5.6055 6.5234
Mindlin theory [92] 1.7659 – – 3.8576 3.8576 – 5.5729 6.5809
3-D Exact solutions [98] 1.7557 – – 3.8991 3.8991 4.6128 5.6527 –
3-D Ritz [99] 1.7558 3.2617 3.2617 3.8991 3.8991 4.6128 5.6524 6.5234
3-D DQ [100] 1.7558 3.2617 3.2617 3.8991 3.8991 4.6127 5.6524 6.5234
3-D Chebyshev–Ritz [96] 1.7558 3.2617 3.2617 3.8991 3.8991 4.6128 5.6524 6.5234
3-D DQ [97] 1.7758 3.2617 3.2617 3.8999 3.8999 4.6127 5.6533 6.5236
0.5 Present 1.2476 1.3047 1.3047 1.8451 2.3186 2.3186 2.6094 2.6094
Reddy–Ritz [95] 1.2451 1.3047 1.3047 1.8451 2.3079 2.3079 2.6094 2.6094
3-D Exact solutions [98] 1.2590 1.3047 1.3047 1.8451 2.3312 2.3312 2.6094 2.6094
3-D Ritz [99] 1.2590 1.3047 1.3047 1.8451 2.3312 2.3312 2.6094 2.6094
3-D Chebyshev–Ritz [96] 1.2451 1.3047 1.3047 1.8451 2.3079 2.3079 2.6094 2.6094

Table 7 pffiffiffiffiffiffiffiffiffiffiffiffi
 ¼ xðb2 =p2 Þ qh=D, of CCCC isotropic square plate.
Comparison of first eight frequency parameters, x

h/b Source Mode number

1 2 3 4 5 6 7 8
0.01 Present 3.6420 7.4214 7.4214 10.9336 13.2911 13.3553 16.6565 16.6565
Mindlin–Ritz [93] 3.6421 7.4254 7.4254 10.9510 13.2940 13.3560 – –
3-D Ritz solutions [101] 3.6490 7.4352 7.4352 10.9530 13.3150 13.3790 16.6820 16.6820
3-D DQ [97] 3.6671 7.4599 7.4705 10.9950 13.3670 13.4300 16.7480 16.7480
0.1 Present 3.3092 6.3317 6.3317 8.8957 10.5036 10.5959 12.4783 12.4783
Classical theory [91] 3.3099 6.3249 6.3249 8.8977 10.4550 10.5440 – –
Mindlin–Ritz [93] 3.2954 6.2858 6.2858 8.8098 10.3790 10.4780 12.5530 12.5530
Reddy–Ritz [95] 3.3025 6.3107 6.3107 8.8583 10.4464 10.5418 12.4784 12.4784
3-D Ritz solutions [101] 3.3184 6.3402 6.3402 8.8961 10.4900 10.5900 12.5190 12.5190
3-D DQ [97] 3.3282 6.3547 6.3547 8.9135 10.4930 10.5920 12.5240 12.5240
3-D B-spline Ritz [102] 3.3182 6.3404 6.3404 8.8961 10.4969 10.5972 12.5182 12.5182
0.5 Present 2.7251 4.8021 4.8021 6.2392 6.2392 6.4762 7.4167 7.4322
3-D Ritz solutions [101] 2.7247 4.7706 4.7706 6.2727 6.2727 6.4163 7.3219 7.4225
3-D DQ [97] 2.7288 4.7762 4.7762 6.2754 6.2754 6.4224 7.3254 7.4258
3-D B-spline Ritz [102] 2.7244 4.7702 4.7702 6.2724 6.2724 6.4159 7.3229 7.4235

5.5. Free vibration of isotropic plate

In this section, a square isotropic plate having thickness ratio (h/b) ranging from 0.01 to 0.5 (from very thin to very thick
plate) is examined. Here, three general boundary conditions, namely SSSS, CCCC and FFFF, are considered for comparison
purpose. In Tables 6–8, free dimensionless natural frequency parameters are given up to the eighth mode for SSSS and CCCC
boundary conditions and up to the tenth mode for FFFF boundary condition. All results are compared to those computed by
considering classical plate theory [90,91], Mindlin plate theory [92–94], third-order shear deformation plate theory [95] and
three-dimensional linear elastic plate theory [96–103]. For all thickness ratios shown, the agreement is found to be good.
Indeed, for plate with SSSS boundary conditions and having h/b = 0.1, the average relative error is about 0.240% for the pres-
ent theory and 0.260% for Reddy theory [95] when compared with the exact three-dimensional solution. For the case of plate
with h/b = 0.5, the average relative error is about 0.248% for the present theory and 0.388% for Reddy theory [95] when com-
pared with the exact three-dimensional solution. According to the Table 7, the average relative deviation is about 0.173% for
the present theory and 0.419% for Reddy theory [95] when compared with the exact three-dimensional solution for plate
 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508 2503

Table 8 pffiffiffiffiffiffiffiffiffiffiffiffi
 ¼ xðb2 qh=DÞ, of FFFF isotropic square plate.
Comparison of first ten frequency parameters, x

h/b Source Mode number

1 2 3 4 5 6 7 8 9 10
0.01 Present 13.454 19.589 24.260 34.759 34.759 61.042 61.042 63.576 69.148 77.068
3-D B-spline Ritz [102] 13.419 19.589 24.258 34.669 34.669 61.016 61.016 63.355 68.984 76.932
Classical theory [90] 13.489 19.789 24.432 35.024 35.024 61.526 61.526 – – –
0.05 Present 13.199 19.424 24.018 33.896 33.896 59.470 59.470 61.187 66.493 74.231
3-D B-spline Ritz [102] 13.147 19.425 24.018 33.727 33.727 59.477 59.477 60.739 66.299 74.104
0.1 Present 12.739 18.946 23.330 32.008 32.008 55.397 55.397 55.978 60.750 67.799
3-D B-spline Ritz [102] 12.723 18.954 23.345 31.955 31.955 55.490 55.490 55.821 60.760 67.875
3-D Ritz [103] 12.726 18.955 23.347 31.965 31.965 55.493 55.493 55.853 60.767 67.882
Reddy–Ritz [95] 12.722 18.944 23.325 31.931 31.931 55.358 55.358 55.741 60.655 67.694
Mindlin–Ritz [94] 12.719 18.945 23.323 31.922 31.922 55.351 55.351 55.715 60.632 67.674
0.2 Present 11.715 17.407 21.207 27.625 27.625 40.194 42.810 42.810 45.265 45.332
3-D B-spline Ritz [102] 11.710 17.433 21.252 27.647 27.647 40.192 42.775 42.775 45.308 45.526
3-D Ritz [103] 11.710 17.433 21.252 27.647 27.647 40.191 42.776 42.776 45.310 –
Mindlin–Ritz [94] 11.701 17.400 21.194 27.573 27.573 – – – 45.105 –
0.3 Present 10.658 15.623 18.857 23.602 23.602 26.796 28.540 28.540 30.351 34.495
3-D B-spline Ritz [102] 10.648 15.657 18.914 23.613 23.613 26.793 28.488 28.488 30.351 34.376
0.4 Present 9.677 13.948 16.732 20.097 20.319 20.319 21.405 21.405 22.763 25.872
3-D B-spline Ritz [102] 9.658 13.980 16.781 20.093 20.297 20.297 21.333 21.333 22.763 25.697
0.5 Present 8.810 12.491 14.932 16.078 17.124 17.124 17.693 17.693 18.210 20.697
3-D B-spline Ritz [102] 8.780 12.515 14.961 16.072 17.030 17.030 17.631 17.631 18.210 20.451
3-D Ritz [103] 8.780 12.515 14.962 16.073 17.030 17.030 17.631 17.631 18.211 –
Reddy–Ritz [95] 8.797 12.479 14.909 16.077 17.124 17.124 17.650 17.650 18.211 21.502

Table 9 pffiffiffiffiffiffiffiffiffiffiffiffi
 ¼ xðb2 =hÞ qt =Et , of SSSS P-FGM plate (Al/Al2O3, a/b = 1).
Comparison of first five frequency parameters, x

h/b n Source Mode number

1 2 3 4 5
0.02 0 Present 5.965 14.880 14.880 23.756 29.659
3-D Chebyshev–Ritz [104] 5.965 14.880 14.880 23.760 29.660
0.2 Present 5.536 13.812 13.812 22.054 27.536
3-D Chebyshev–Ritz [104] 5.536 13.810 13.810 22.060 27.540
5 Present 3.925 9.785 9.785 15.614 19.487
3-D Chebyshev–Ritz [104] 3.925 9.787 9.787 15.620 19.490
0.2 0 Present 5.283 9.742 9.742 11.566 11.566
3-D Chebyshev–Ritz [104] 5.304 9.742 9.742 11.650 11.650
0.2 Present 4.929 9.280 9.280 10.839 10.840
3-D Chebyshev–Ritz [104] 4.950 9.280 9.280 10.920 10.920
5 Present 3.390 6.296 6.296 7.282 7.284
3-D Chebyshev–Ritz [104] 3.406 6.296 6.296 7.347 7.347

with CCCC boundary conditions and having h/b = 0.1. When the plate is under FFFF boundary conditions and having h/b = 0.1
(Table 8), the average relative deviations are 0.131% and 0.154% for the present model and Reddy model, respectively. For
thick plate (h/b = 0.5) under the same boundary conditions, the average relative errors are 0.377% and 0.732% for the present
model and Reddy model, respectively. As it can see, the present hybrid model gives slightly more accurate results than
Reddy’s ones. It should be pointed out that the fundamental mode provided by 3-D solutions should present the lowest
percentage deviation when compared with each other. This is not the case as shown on Table 6 for h/b = 0.2 where the
fundamental frequency parameter provided by Liew and Teo [97] deviate about 1.126% when compared to that of Malik
and Bert [100].

5.6. Free vibration of FGM plate

A square FGM plate undergoing free vibration is considered. Material properties and geometrical dimensions are those of
literatures [104–106] used for comparison study. A number of comparisons were conducted in order to show the effective-
ness of the present theory when predicting natural frequency parameters of FGM square and rectangular plates for different
boundary conditions.
The first five frequency parameters of a square FG plate (Al/Al2O3) under SSSS, CFFF, and CFCF boundary conditions
respectively are given in Tables 9–11. The obtained solutions, when compared with the 3-D elasticity solutions provided
2504 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508

Table 10 pffiffiffiffiffiffiffiffiffiffiffiffi
 ¼ xðb2 =hÞ qt =Et , of CFFF P-FGM (Al/Al2O3, a/b = 1).
Comparison of first five frequency parameters, x

h/b n Source Mode number

1 2 3 4 5
0.1 0 Present 1.039 2.444 6.092 6.585 7.731
3-D Chebyshev–Ritz [104] 1.042 2.446 6.107 6.601 7.732
0.2 Present 0.965 2.273 5.668 6.274 7.191
3-D Chebyshev–Ritz [104] 0.968 2.275 5.682 6.289 7.193
5 Present 0.682 1.594 3.960 4.277 5.023
3-D Chebyshev–Ritz [104] 0.684 1.595 3.969 4.287 5.026
0.2 0 Present 1.012 2.233 3.293 5.354 6.821
3-D Chebyshev–Ritz [104] 1.016 2.233 3.306 5.361 6.828
0.2 Present 0.941 2.083 3.137 5.004 6.364
3-D Chebyshev–Ritz [104] 0.945 2.084 3.150 5.012 6.372
5 Present 0.661 1.431 2.146 3.391 4.337
3-D Chebyshev–Ritz [104] 0.664 1.432 2.154 3.396 4.347

Table 11 pffiffiffiffiffiffiffiffiffiffiffiffi
 ¼ xðb2 =hÞ qt =Et , of CFCF P-FGM plate (Al/Al2O3, a/b = 1).
Comparison of first five frequency parameters, x

h/b n Source Mode number

1 2 3 4 5
0.2 0 Present 5.364 6.092 8.880 9.556 12.346
3-D Chebyshev–Ritz [104] 5.382 6.089 8.893 9.536 12.280
0.2 Present 5.030 5.716 8.461 8.954 11.633
3-D Chebyshev–Ritz [104] 5.050 5.715 8.473 8.940 11.580
5 Present 3.391 3.823 5.766 5.977 7.642
3-D Chebyshev–Ritz [104] 3.400 3.820 5.774 5.976 7.609

Table 12  pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 ¼ x b2 =p2 Im =Dm , of CCCC P-FGM plate (SUS304/Si3N4, n = 2, a = 0.2, a/b = 1, a/h = 10).
Comparison of first eight frequency parameters, x

Source Mode number

1 2 3 4 5 6 7 8
Presenta 4.1083 7.8556 7.8565 11.0386 13.0737 13.1852 15.3905 15.3918
Presentb 4.1089 7.8690 7.8698 11.0691 13.0884 13.1960 15.3982 15.3995
Presentc 4.1262 7.9151 7.9160 11.1460 13.2164 13.3183 15.3922 15.3933
3-D Chebyshev–Ritz [106] 4.1658 7.9389 7.9389 11.1212 13.0973 13.2234 15.3627 15.3627
HSDPT-Ritz [107] 4.1165 7.9696 7.9696 11.2198 13.1060 13.2089 15.9471 15.9471
DQ-Galerkin, HSDPT [108] 4.1062 7.8902 7.8902 11.1834 12.5881 13.1867 15.4530 16.0017
a
Characteristics orthogonal polynomials (Gram–Schmidt).
b
Chebyshev polynomials.
c
Hierarchical trigonometric functions.

by Huang et al. [104], are found to be good for different thickness ratios and volume fraction power indexes shown. Table 12
compares the first eight frequency parameters of square FGM plate (SUS304/Si3N4, n = 2) for a/h = 10 and having all edges
clamped. According to Table 12, it can be observed that the numerical results obtained when using hierarchical trigonomet-
ric functions as trials Ritz functions are more accurate when compared to 3-D elasticity solutions provided by Li et al. [106].
Indeed, the average relative deviations obtained by employing characteristics orthogonal polynomials, Chebyshev polynomi-
als, and hierarchical trigonometric functions are 0.88%, 0.54%, and 0.47% respectively while the average relative deviation
obtained by Kim [107], when using Reddy’s theory and beam functions in Ritz procedure, is about 1.328% for all modes
shown. It should be point out that our computed values shown were calculated using only 8  8 points while the ones pro-
vided by Kim [107] were obtained upon using a grid size equal to 10  10. It is clear here that the present model, associated
with the present resolution method, gives better results. The last example in this section (Table 13) compares the non-
dimensional fundamental frequency of a rectangular (b = 2a) FG plate for six different classical boundary conditions (SSSS,
SCSC, SFSF, SCSF, SSSC, and SSSF). The plate is made of metal and ceramic (Al/Al2O3). The percentage average deviations
of fundamental frequency parameter under boundary conditions SSSS, SCSC, SFSF, SCSF, SSSC, and SSSF when compared with
the exact solution are 0.201%, 0.322%, 0.062%, 0.035%, 0.246%, and 0.026% respectively. All computed values are very close to
those of 3-D elasticity solutions obtained by Jin et al. [105]. It can be concluded that for all cases presented, the numerical
results are good compared to solutions obtained using three-dimensional linear plate theory [104–106].
 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508 2505

Table 13 pffiffiffiffiffiffiffiffiffiffiffiffi

 ¼ x a2 =h qt =Et , of rectangular FG plate (Al/Al2O3) with different classical boundary conditions.
Comparison of fundamental frequency parameters, x

b/a h/a n SSSS SCSC SFSF SCSF SSSC SSSF

Present 3-Da Present 3-Da Present 3-Da Present 3-Da Present 3-Da Present 3-Da
0.1 0 0.0717 0.0719 0.0790 0.0793 0.0569 0.0568 0.0607 0.0607 0.0749 0.0751 0.0600 0.0600
1 0.0549 0.0550 0.0605 0.0608 0.0435 0.0435 0.0464 0.0465 0.0573 0.0575 0.0459 0.0459
2 0.0498 0.0499 0.0549 0.0552 0.0395 0.0395 0.0422 0.0422 0.0520 0.0522 0.0417 0.0417
5 0.0470 0.0471 0.0516 0.0519 0.0373 0.0372 0.0398 0.0398 0.0490 0.0492 0.0393 0.0393
2 0.2 0 0.2703 0.2713 0.2929 0.2941 0.2165 0.2166 0.1914 0.1914 0.2804 0.2814 0.1914 0.1914
1 0.2080 0.2088 0.2260 0.2271 0.1663 0.1665 0.1592 0.1592 0.2160 0.2169 0.1592 0.1592
2 0.1881 0.1888 0.2041 0.2050 0.1506 0.1507 0.1438 0.1438 0.1952 0.1960 0.1438 0.1438
5 0.1748 0.1754 0.1886 0.1895 0.1404 0.1405 0.1243 0.1243 0.1810 0.1817 0.1243 0.1243
0.5 0 0.9570 0.9570 0.9570 0.9570 0.9570 0.9570 0.4785 0.4785 0.9570 0.9570 0.4785 0.4785
1 0.7939 0.7937 0.7939 0.7937 0.7939 0.7937 0.3978 0.3978 0.7939 0.7937 0.3978 0.3978
2 0.7151 0.7149 0.7151 0.7149 0.7151 0.7149 0.3591 0.3591 0.7151 0.7149 0.3591 0.3591
5 0.6169 0.6168 0.6169 0.6168 0.6169 0.6168 0.3102 0.3101 0.6169 0.6168 0.3102 0.3101
a
Jin et al. [105].

It is seen from all tables and figures that the presented results are accurate enough to prove the efficiency of the present
model in predicting deflection, stresses and natural frequency parameters of isotropic, functionally graded, sandwich and
laminated composite plates. Moreover, the present theory, which combines hyperbolic tangent function and polynomial
ones, produce a better shear model than Reddy’s ones for all cases considered in comparison study.

6. Conclusions

In this article, a new hyperbolic shear deformation theory applicable to bending and free vibration analysis of isotropic,
functionally graded, sandwich and laminated composite plates is presented. It is variationally consistent, based on the same
assumptions as those of the third-order shear deformation plate theory of Reddy [28] and have five degrees of freedom.
Moreover, it accounts for parabolic variation of transverse shear strains and shear stresses through the thickness of the plate
and satisfies rigorously the zero traction boundary conditions on the top and bottom surfaces of the plate. All results
obtained reveal that the present theory can be considered as a good alternative to some 2-D theories for approximating
the tedious and time consuming three-dimensional isotropic, composite and FGM plate problems.

Appendix A

The coefficients of each matrix appearing in Eq. (38) for both FGM plate and antisymmetric cross-ply laminates are
defined as follows:

k11 ¼ A11 a2 þ A66 b2 ; k12 ¼ ðA12 þ A66 Þab;

 
k13 ¼ a3 Ba11  B11 þ 2Ba66  2B66 þ Ba12  B12 ab2 ; ðA1Þ

k14 ¼ Ba11 a2 þ b2 Ba66 ; k15 ¼ Ba66 þ Ba12 ab;

k21 ¼ k12 ; k22 ¼ A22 b2 þ A66 a2 ; k23 ¼ a2 bð2Ba66  2B66 þ Ba12  B12 Þ þ b3 ðBa22  B22 Þ;

k24 ¼ ðBa66 þ Ba12 Þab; k25 ¼ Ba66 a2 þ Ba22 b2 ;

k31 ¼ k13 ; k32 ¼ k23 ;

2 a 2 a
k33 ¼ ðD11  2Da11 þ Daa 4 a aa a aa 2 2 a aa 4
11 Þa þ 2ð2D66  4D66 þ 2D66 þ D12  2D12 þ D12 Þa b þ ðD22  2D22 þ D22 Þb þ a A55 þ b A44 ;
a
k34 ¼ ðDaa a 3 aa a aa a 2
11  D11 Þa þ ðD12  D12 þ 2D66  2D66 Þab þ A55 a;
a
k35 ¼ ðDaa a 3 aa a aa a 2
22  D22 Þb þ ðD12  D12 þ 2D66  2D66 Þa b þ A44 b;

a
k41 ¼ k14 ; k42 ¼ k24 ; k43 ¼ k34 ; k44 ¼ Daa 2 aa 2
11 a þ D66 b þ A55 ; k45 ¼ ðDaa aa
12 þ D66 Þab;
a
k51 ¼ k15 ; k52 ¼ k25 ; k53 ¼ k35 ; k54 ¼ k45 ; k55 ¼ a 2
Daa
66 þ b Daa
2
22 þ A44 ;
2506 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508

m11 ¼ I1 ; m12 ¼ 0; m13 ¼ ðIa2  I2 Þa; m14 ¼ Ia2 ; m15 ¼ 0;

m21 ¼ m12 ; m22 ¼ I1 ; m23 ¼ ðIa2  I2 Þb; m24 ¼ 0; m25 ¼ Ia2 ;
m31 ¼ m13 ; m32 ¼ m23 ; m33 ¼ I1 þ ðIaa
3 þ I3  2Ia3 Þð 2
a þ b2 Þ;
ðA2Þ
m34 ¼ ðIaa
3  Ia3 Þ a; m35 ¼ ðIaa
3  Ia3 Þb;
m41 ¼ m14 ; m42 ¼ m24 ; m43 ¼ m34 ; m44 ¼ Iaa
3 ; m45 ¼ 0;
m51 ¼ m15 ; m52 ¼ m25 ; m53 ¼ m35 ; m54 ¼ m45 ; m55 ¼ Iaa
3 :

References

[1] M. Yamanouchi, M. Koizumi, I. Shiota, in: Proceedings of the First International Symposium on Functionally Gradient Materials, Japan, 1990.
[2] M. Koizumi, FGM activities in Japan, Compos. B (1997) 1–4.
[3] M. Koizumi, The Concept of FGM, Functionally Gradient Materials, The American Ceramic Society, Westerville, 1993, pp. 3–10.
[4] S. Suresh, A. Mortensen, Fundamentals of functionally graded materials, in: Processing and Thermomechanical Behaviour of Graded Metals and
Metal–Ceramic Composites, IOM Communications, Cambridge, UK, 1998.
[5] Y. Miyamoto, W.A. Kaysser, B.H. Rabin, A. Kawasaki, R.G. Ford, Functionally Graded Materials: Design, Processing and Applications, Kluwer Academic
Publishers, Hingham, MA (US), United States, 1999.
[6] G.R. Kirchhoff, Uber das Gleichgewicht und die Bewegung einer Elastischen Scheibe, J. Reine Angew Math. (Crelle) 40 (1850) 51–88.
[7] A.E.H. Love, On the Small Free Vibrations and Deformations of Elastic Shells, Philosophical Transactions of the Royal Society, London, 1888.
[8] L. Rayleigh, Theory of Sound, Dover Publication, Inc, New York, USA, 1945.
[9] S. Timoshenko, On the correction of transverse shear deformation of the differential equations for transverse vibrations of prismatic bars, Phil. Mag. 41
(6) (1921) 744–746.
[10] S. Timoshenko, On the transverse vibrations of bars of uniform cross-section, Phil. Mag. 43 (6) (1922) 125–131.
[11] E. Reissner, On the theory of bending of elastic plates, J. Math. Phys. (1944) 184–191.
[12] Y.S. Uflyand, The propagation of waves in the transverse vibrations of bars and plates, Prikladnaia Mathematika i Mekhanica 12 (1948) 287–300.
[13] R.D. Mindlin, Influence of rotatory inertia and shear on flexural motions of isotropic elastic plates, Trans. ASME, J. Appl. Mech. (1951) 31–38.
[14] J.M. Whitney, The effect of transverse shear deformation on the bending of laminated plates, J. Compos. Mater. 3 (1969) 534–547.
[15] J.M. Whitney, C.T. Sun, A higher-order theory for extensional motion of laminated composites, J. Sound Vib. (1973) 85–97.
[16] J.N. Pagano, Exact solutions for rectangular bidirectional composites and sandwich plates, J. Compos. Mater. 4 (1970) 20–35.
[17] S. Srinivas, A.K. Rao, Bending, vibration and buckling of simply supported thick orthotropic rectangular plates and laminates, Int. J. Solids Struct.
(1970) 1463–1481.
[18] E. Reissner, On the analysis of first and second order shear deformation effects for isotropic elastic plates, Trans. ASME – J. Appl. Mech. (1980) 959–
961.
[19] R. Schmidt, A refined nonlinear theory of plates with transverse shear deformation, Ind. Math. 27 (1) (1977) 23–38.
[20] M. Levinson, An accurate simple theory of the statics and dynamics of elastic plates, Mech. Res. Commun. (1980) 343–350.
[21] S.A. Ambartsumyan, Theory of Anisotropic Plates, Technomic Publishing Company, Stanford, 1970.
[22] E. Reissner, The effect of transverse shear deformation on the bending of elastic plates, J. Appl. Mech. – Trans. ASME (1945) 69–77.
[23] V.Z. Vlasov, The method of initial function in problems of theory of thick plates and shells, in: Ninth Congress of International Mechanical
Applications, 1957, pp. 321–330.
[24] K.T.S.R. Iyengar, K. Chandrashekhara, V.K. Sebastian, On the analysis of thick rectangular plates, Ing. Arch. 43 (5) (1974) 317–330.
[25] A.V. Krishna Murty, Higher order theory for vibrations of thick plates, AIAA J. 15 (12) (1977) 1823–1824.
[26] N.D. Kuzentov, G.G. Kartashov, Natural vibrational modes of laminated anisotropic plates and flat shells, Sov. Appl. Mech. (1981) 334–339.
[27] K.T.S.R. Iyengar, S.K. Pandya, Vibration of orthotropic rectangular thick plates, Int. J. Solids Struct. (1982) 149–156.
[28] J.N. Reddy, A simple higher-order theory for laminated composite plates, ASME J. Appl. Mech. (1984) 745–752.
[29] J.N. Reddy, N.D. Phan, Analysis of laminated composite plates using a higher-order shear deformation theory, Int. J. Numer. Meth. Eng. (1985)
2201–2219.
[30] A.V. Krishna Murty, Toward a consistent plate theory, AIAA J. 24 (6) (1986) 1047–1048.
[31] A.V. Krishna Murty, Flexural of composite plates, Compos. Struct. (1987) 161–177.
[32] A.V. Krishna Murty, S. Vellaichamy, On higher-order shear deformation theory of laminated composite panels, Compos. Struct. (1987) 247–270.
[33] J.N. Reddy, Theory and Analysis of Elastic Plates, Taylor & Francis, Philadelphia, 1999.
[34] J.N. Reddy, Analysis of functionally graded plates, Int. J. Numer. Meth. Eng. 47 (2000) 663–684.
[35] J.N. Reddy, A review of refined theories of laminated composite plates, Shock Vib. Digest 22 (7) (1990) 3–17.
[36] M. Mallikarjuna, T. Kant, A critical review and some results of recently developed refined theories of fiber-reinforced laminated composites and
sandwiches, Compos. Struct. 23 (4) (1993) 293–312.
[37] L. Librescu, On the theory of anisotropic elastic shells and plates, Int. J. Solids Struct. (1967) 53–68.
[38] A. Bhimaraddi, L.K. Stevens, A higher order theory for free vibration of orthotropic, homogeneous and laminated rectangular plates, J. Appl. Mech. –
Trans. ASME (1984) 195–198.
[39] J.G. Ren, A new theory of laminated plate, Compos. Sci. Technol. (1986) 225–239.
[40] T. Kant, B.N. Pandya, A simple finite element formulation of a higher-order theory for unsymmetrically laminated composite plates, Compos. Struct.
(1988) 215–264.
[41] M. Touratier, A refined theory for thick composite plates, Mech. Res. Commun. 15 (4) (1988) 229–236.
[42] V.K. Prokopov, Application of the symbolic method to the derivation of the equations of the theory of plates, J. Appl. Math. Mech. (1965) 1064–1183.
[43] C.T. Sun, J.M. Whitney, Theories for dynamic response of laminated plates, AIAA J. (1973) 178–183.
[44] R. Nelson, D. Lorch, A refined theory for laminated orthotropic plates, ASME J. Appl. Mech. (1974) 177–183.
[45] K.H. Lo, R.M. Christensen, E.M. Wu, A high-order theory of plate deformation. Part I: homogeneous plates, ASME J. Appl. Mech. (1977) 663–668.
[46] K.H. Lo, R.M. Christensen, E.M. Wu, A high-order theory of plate deformation. Part II: laminated plates, ASME J. Appl. Mech. (1977) 669–676.
[47] T. Kant, Numerical analysis of thick plates, Comput. Methods Appl. Mech. Eng. (1982) 1–18.
[48] N.J. Pagano, S.R. Soni, Global–local laminate variational model, Int. J. Solids Struct. (1983) 207–228.
[49] M. Karama, B.A. Harb, S. Mistou, S. Caperaa, Bending, buckling and free vibration of laminated composite with a transverse shear stress continuity
model, Compos. B 29 (3) (1998) 223–234.
[50] M. Karama, K.S. Afaq, S. Mistou, Mechanical behaviour of laminated composite beam by the new multi-layered laminated composite structures model
with transverse shear stress continuity, Int. J. Solids Struct. 40 (2003) 1525–1546.
[51] E. Carrera, Theories and finite elements for multilayered plates and shells: a unified compact formulation with numerical assessment and
benchmarking, Arch. Comput. Meth. Eng. 9 (3) (2003) 215–296.
[52] E. Carrera, M. Petrolo, Guidelines and recommendations to construct theories for metallic and composite plates, AIAA J. 48 (12) (2010) 2852–2866.
 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508 2507

[53] L. Demasi, Mixed plate theories based on the generalized unified formulation. Part II: layerwise theories, Compos. Struct. 87 (1) (2009) 12–22.
[54] L. Demasi, Mixed plate theories based on the generalized unified formulation. Part IV: Zig-zag theories, Compos. Struct. 87 (3) (2009) 195–205.
[55] L. Demasi, Mixed plate theories based on the generalized unified formulation. Part I: governing equations, Compos. Struct. 87 (1) (2009) 1–11.
[56] L. Demasi, Mixed plate theories based on the generalized unified formulation. Part III: advanced mixed high order shear deformation theories,
Compos. Struct. 87 (3) (2009) 183–194.
[57] L. Demasi, Mixed plate theories based on the generalized unified formulation. Part V: results, Compos. Struct. 88 (1) (2009) 1–16.
[58] M. Aydogdu, A new shear deformation theory for laminated composite plates, Compos. Struct. (2009) 94–101.
[59] N. El Meiche, A. Tounsi, N. Ziane, I. Mechab, E.A. Adda Bedia, A new hyperbolic shear deformation theory for buckling and vibration of functionally
graded sandwich plate, Int. J. Mech. Sci. 53 (4) (2011) 237–247.
[60] M. Touratier, An efficient standard plate theory, Int. J. Eng. Sci. (1991) 901–916.
[61] K.P. Soldatos, A transverse shear deformation theory for homogeneous monoclinic plates, Acta Mech. 94 (1992) 195–220.
[62] J.L. Mantari, C. Guedes Soares, Bending analysis of thick exponentially graded plates using a new trigonometric higher order shear deformation
theory, Compos. Struct. 94 (6) (2012) 1991–2000.
[63] J.L. Mantari, A.S. Oktem, C. Guedes Soares, A new higher order shear deformation theory for sandwich and composite laminated plates, Compos. B Eng.
43 (3) (2012) 1489–1499.
[64] J.L. Mantari, A.S. Oktem, C. Guedes Soares, A new trigonometric shear deformation theory for isotropic, laminated composite and sandwich plates, Int.
J. Solids Struct. 49 (1) (2012) 43–53.
[65] A.M.A. Neves, A.J.M. Ferreira, E. Carrera, M. Cinefra, C.M.C. Roque, R.M.N. Jorge, et al, A quasi-3D hyperbolic shear deformation theory for the static and
free vibration analysis of functionally graded plates, Compos. Struct. 94 (5) (2012) 1814–1825.
[66] A.M.A. Neves, A.J.M. Ferreira, E. Carrera, C.M.C. Roque, M. Cinefra, R.M.N. Jorge, et al, A quasi-3D sinusoidal shear deformation theory for the static and
free vibration analysis of functionally graded plates, Compos. B Eng. 43 (2) (2012) 711–725.
[67] Y.M. Ghugal, A.S. Sayyad, Free vibration of thick orthotropic plates using trigonometric shear deformation theory, Latin Am. J. Solids Struct. 8 (2011)
229–243.
[68] R.P. Shimpi, Refined plate theory and its variants, AIAA J. (2002) 137–146.
[69] R.P. Shimpi, H.G. Patel, Free vibrations of plate using two variable refined plate theory, J. Sound Vib. (2006) 979–999.
[70] R.P. Shimpi, H.G. Patel, A two variable refined plate theory for orthotropic plate analysis, Int. J. Solids Struct. (2006) 6783–6799.
[71] S.E. Kim, H.T. Thai, J. Lee, Buckling analysis of plates using the two variable refined plate theory 47 (4) (2009) 455–462.
[72] H.A. Atmane, A. Tounsi, I. Mechab, E.A. Adda Bedia, Free vibration analysis of functionally graded plates resting on Winkler–Pasternak elastic
foundations using a new shear deformation theory, Int. J. Mech. Mater. Des. (2010) 113–121.
[73] H.-T. Thai, S.-E. Kim, Free vibration of laminated composite plates using two variable refined plate theory, Int. J. Mech. Sci. (2010) 626–633.
[74] H.-T. Thai, S.-E. Kim, Levy-type solution for buckling analysis of orthotropic plates based on two variable refined plate theory, Compos. Struct. (2011)
1738–1746.
[75] L. Hadji, H.A. Atmane, A. Tounsi, I. Mechab, E.A. Adda Bedia, Free vibration of functionally graded sandwich plates using four-variable refined plate
theory, Appl. Math. Mech. (2011) 925–942.
[76] A. Benachour, H.D. Tahar, H.A. Atmane, A. Tounsi, M.S. Ahmed, A four variable refined plate theory for free vibrations of functionally graded plates
with arbitrary gradient, Compos. B Eng. 42 (6) (2011) 1386–1394.
[77] M. Bourada, A. Tounsi, M.S.A. Houari, E.A. Adda Bedia, A new four-variable refined plate theory for thermal buckling analysis of functionally graded
sandwich plates, J. Sandwich Struct. Mater. 14 (1) (2011) 5–33.
[78] A. Hamidi, M. Zidi, M.S.A. Houari, A. Tounsi, A new four variable refined plate theory for bending response of functionally graded sandwich plates
under thermomechanical loading, Compos. Part B: Eng. (2012).
[79] T.H. Daouadji, A. Hadj Henni, A. Tounsi, E.A. Adda Bedia, A new hyperbolic shear deformation theory for bending analysis of functionally graded
plates, Modell. Simul. Eng. 2012 (2012) 1–10. Hindawi Publishing Corporation.
[80] H.-T. Thai, S.-E. Kim, A simple higher-order shear deformation theory for bending and free vibration analysis of functionally graded plates, Compos.
Struct. 96 (2013) 165–173.
[81] J.L. Mantari, C. Guedes Soares, A novel higher-order shear deformation theory with stretching effect for functionally graded plates, Compos. B Eng. 45
(1) (2013) 268–281.
[82] H.-T. Thai, S.-E. Kim, A simple quasi-3D sinusoidal shear deformation theory for functionally graded plates, Compos. Struct. 99 (2013) 172–180.
[83] A.M. Zenkour, Bending analysis of functionally graded sandwich plates using a simple four-unknown shear and normal deformations theory, J.
Sandwich Struct. Mater. (2013).
[84] J.N. Reddy, Mechanics of Laminated Composite Plates and Shells: Theory and Analysis, second ed., CRC Press, 2004.
[85] R.B. Bhat, Natural frequencies of rectangular plates using characteristic orthogonal polynomials in Rayleigh–Ritz method, J. Sound Vib. (1985)
493–499.
[86] S. Chakraverty, Vibration of Plates, CRC Press, Taylor & Francis Group, New York, 2009.
[87] B. Uymaz, M. Aydogdu, Three-dimensional vibration analyses of functionally graded plates under various boundary conditions, J. Reinf. Plast. Compos.
26 (18) (2007) 1847–1863.
[88] C.-P. Wu, K.-H. Chiu, Y.-M. Wang, RMVT-based meshless collocation and element-free Galerkin methods for the quasi-3D analysis of multilayered
composite and FGM plates, Compos. Struct. 93 (2) (2011) 923–943.
[89] N.J. Pagano, H.J. Hatfield, Elastic behavior of multilayered bidirectional composites, AIAA J. 10 (7) (1972) 931–933.
[90] A.W. Leissa, The free vibration of rectangular plates, J. Sound Vib. 31 (3) (1973) 257–293.
[91] D.J. Gorman, Accurate free vibration analysis of clamped Mindlin plates using the method of superposition, J. Sound Vib. 98 (1996) 341–353.
[92] R.D. Mindlin, A. Schacknow, H. Deresiewicz, Flexural vibrations of rectangular plates, ASME J. Appl. Mech. 23 (1956) 430–436.
[93] O.L. Roufael, D.J. Dawe, Vibration analysis of rectangular Mindlin plates by the finite strip method, Comput. Struct. 12 (1980) 833–842.
[94] K.M. Liew, Y. Xiang, S. Kitipornchai, Transverse vibration of thick rectangular plates-I. Comprehensive sets of boundary conditions, Comput. Struct. 49
(1993) 1–29.
[95] C.W. Lim, K.M. Liew, S. Kitipornchai, Numerical aspects for free vibration of thick plates. Part I: formulation and verification, Comput. Methods Appl.
Mech. Eng. 156 (1998) 15–29.
[96] D. Zhou, Y.K. Cheung, F.T.K. Au, S.H. Lo, Three-dimensional vibration analysis of thick rectangular plates using Chebyshev polynomial and Ritz method,
Int. J. Solids Struct. 39 (26) (2002) 6339–6353.
[97] K.M. Liew, T.M. Teo, Three-dimensional vibration analysis of rectangular plates based on differential quadrature method, J. Sound Vib. 220 (4) (1999)
577–599.
[98] S. Srinivas, C.V.J. Rao, A.K. Rao, An exact analysis for vibration of simply-supported homogeneous and laminated thick rectangular plates, J. Sound Vib.
12 (1970) 187–199.
[99] A.W. Leissa, Z.D. Zhang, On the three-dimensional vibrations of the cantilevered rectangular parallelepiped, J. Acoust. Soc. Am. 73 (1983) 2013–2021.
[100] M. Malik, C.W. Bert, Three-dimensional elasticity solutions for free vibrations of rectangular plates by the differential quadrature method, Int. J. Solids
Struct. 35 (1998) 299–319.
[101] K.C. Hung, A treatise on three dimensional vibration of a class of elastic solids, Nanyang Technological University, Singapore, China, 1996.
[102] H. Nagino, T. Mikami, T. Mizusawa, Three-dimensional free vibration analysis of isotropic rectangular plates using the B-spline Ritz method, J. Sound
Vib. 317 (2008) 329–353.
2508 A. Mahi et al. / Applied Mathematical Modelling 39 (2015) 2489–2508

[103] K.M. Liew, K.C. Hung, M.K. Lim, Three-dimensional vibration of rectangular plates: effects of thickness and edge constraints, J. Sound Vib. 182 (1995)
709–727.
[104] C.S. Huang, P.J. Yang, M.J. Chang, Three-dimensional vibration analyses of functionally graded material rectangular plates with through internal
cracks, Compos. Struct. 94 (9) (2012) 2764–2776.
[105] G. Jin, Z. Su, S. Shi, T. Ye, S. Gao, Three-dimensional exact solution for the free vibration of arbitrarily thick functionally graded rectangular plates with
general boundary conditions, Compos. Struct. 108 (2014) 565–577.
[106] Q. Li, V.P. Iu, K.P. Kou, Three-dimensional vibration analysis of functionally graded material plates in thermal environment, J. Sound Vib. 324 (3–5)
(2009) 733–750.
[107] Y.-W. Kim, Temperature dependent vibration analysis of functionally graded rectangular plates, J. Sound Vib. 284 (3–5) (2005) 531–549.
[108] J. Yang, H.-S. Shen, Vibration characteristics and transient response of shear-deformable functionally graded plates in thermal environments, J. Sound
Vib. 255 (2002) 579–602.