Home Physical Sciences Dynamic stability of nanocomposite viscoelastic cylindrical shells coating with a piezomagnetic layer conveying pulsating fluid flow
Article Open Access

Dynamic stability of nanocomposite viscoelastic cylindrical shells coating with a piezomagnetic layer conveying pulsating fluid flow

  • Ali Ghorbanpour Arani EMAIL logo , Seyed Abolfazl Mortazavi and Zahra Khoddami Maraghi
Published/Copyright: December 29, 2015

Abstract

In this study, the dynamic stability of an embedded viscoelastic composite cylindrical shell reinforced by boron nitride nanotubes (BNNTs) is investigated. The composite cylindrical shell is coated by a viscoelastic piezomagnetic layer and subjected to combined magneto-electro-mechanical loads. The composite polymer matrix and the coating layer are made of polyvinylidene fluoride (PVDF) and iron oxide (CoFe2O4), respectively. The composite cylindrical shell conveys pulsating fluid flows, which results in harmonic oscillations. The equivalent characteristics of composite are determined using micro-electro-mechanical models. Considering the magneto-electro-mechanical coupling, motion equations are obtained using Hamilton’s principle. Results show the influences of fluid velocity, geometrical parameters of shell, viscoelastic foundation, orientation angle and percentage of BNNTs on the resonance frequency and stability of a PVDF-coated nanocomposite shell. The result of this study may be used for the design of rotating machines, hydraulic systems and motors.

1 Introduction

Composites offer advantageous characteristics of different materials with qualities that none of the constituents possess. Smart composite addressed in this paper has attracted more attention amongst researchers due to the provision of new properties and exploiting unique synergism between materials. Piezoelectric materials produce an electric field when deformed, and undergo deformation when subjected to an electric field. The coupling nature of piezoelectric materials has attracted wide applications in electro-mechanical and electrical devices, such as actuators, sensors and transducers [1].

Since this paper studies the dynamic response of a composite cylinder conveying pulsating fluid, the introduction is divided into three parts including magneto-electro-elastic shells, pulsating fluid and composite structures.

Three-dimensional (3D) solutions for the static analysis of doubly curved functionally graded (FG) magneto-electro-elastic shells were presented by Tsai et al. [2]. They used an asymptotic approach to solve 29 basic equations that reduced to 10 differential equations under elastic, electric and magnetic fields. The governing equations were derived by considering the first-order approximation to the 3D magneto-electro-elasticity. Their results could be used as a reference to develop an advanced approximate shell theory and as the benchmark solutions. Wu and Tsai [3] analyzed a 3D free vibration of doubly curved FG magneto-electro-elastic shells. They used closed-circuit surface conditions and a perturbation method. They introduced a method of multiple scales to eliminate the secular terms in various order problems of their formulation. The coupled classical shell theory was derived in this work as a first-order approximation to the 3D magneto-electro-elasticity. Also some benchmark solutions for the free vibration analysis of FG elastic and piezoelectric plates were used to validate the present asymptotic formulation.

Torsional buckling of a piezoelectric polymeric cylindrical shell made from polyvinylidene fluoride (PVDF) subjected to combined electro-thermo-mechanical loads was analyzed by Mosallaie Barzoki et al. [4]. The core was modeled as an elastic medium by Winkler and Pasternak modules and the shell was reinforced by armchair double-walled boron nitride nanotubes. Mechanical, electrical and thermal characteristics of the composite shell were determined based on micromechanical modeling. Their results indicated that buckling strength increases substantially as harder foam cores are employed. Nonlinear dynamics of pipes conveying pulsating fluid was investigated considering the effect of motion constraints modeled as cubic springs by Wang [5]. He derived the partial differential equation with a higher mean flow velocity and using the Galerkin method and fourth-order Runge-Kutta scheme. He found an analytical model to exhibit rich and variegated dynamical behaviors that include quasi-periodic and chaotic motions. Ariaratnam and Namachchivaya [6] assumed that the flow velocity is harmonically perturbed about a constant mean value and analyzed the dynamic stability of supported cylindrical pipes conveying fluid. They used a method of averaging to explicit stability conditions for perturbations and the Floquet theory to extend the stability boundaries. Their results have shown the effects of the mean flow velocity, dissipative forces, boundary conditions and virtual mass on the extent of parametric instability regions.

Also, nonlinear dynamics of pipes conveying pulsatile fluid was investigated by Panda and Kar [7] where the pipe was subjected to combination and principal parametric resonance. In this research, the fluid velocity harmonically varied to a constant mean velocity. They utilized the method of multiple scales to solve first-order ordinary differential governing equations considering the associated boundary conditions. Their results have shown the response in directly excited and indirectly excited modes due to modal interaction. Liang and Su [8] studied the stability of a single-walled carbon nanotube conveying fluid where the internal flow was pulsating and viscous. Based on nonlocal elasticity theory, partial differential equations were obtained using the Galerkin method as well as averaging equations for the first two modes. They discussed the stability regions in the frequency-amplitude plane and the influences of the nonlocal effect, viscosity and some system parameters. Tan and Tong [9] presented the closed-form formulas for the effective constants of composite materials under various loading assumptions using the linear piezoelectric theory and iso-field assumptions. They found that the predicted electro-elastic constants under various loading assumptions are generally bounded by those obtained under single- and multiple-loading assumptions. In fact, they showed that the effective constants along the fiber direction obtained under the single-loading assumption are the same as those evaluated using the rule of mixtures. They also developed in another work [10] two micromechanics models, referred to as “XY PEMFRC model” and “YX PEMFRC model”, to investigate the electro-magneto-thermo-elastic properties for piezoelectric-magnetic fiber reinforced composite (PEMFRC) materials operating in the linear regime. The theoretical and experimental findings in piezoelectric nanostructures were presented by Fang et al. [11], including piezoelectric nanowires, nanoplates, nanobeams, nanofilms, nanoparticles and piezoelectric heterogeneous materials containing piezoelectric nano-inhomogeneities. First, they delineated the types of piezoelectric nanostructured materials and their wide applications. Then, they illustrated the theoretical foundations including the definition of surface stress, electric displacement, surface constitutive relations, surface equilibrium equations and nonlocal piezoelectricity.

This paper aims to study the dynamic stability of a viscoelastic nanocomposite cylindrical shell conveying pulsating fluid flow made of PVDF and reinforced by boron nitride nanotubes (BNNTs) with a viscoelastic piezomagnetic coating layer of iron oxide (CoFe2O4). In order to control the stability of the system, the cylindrical shell is subjected to applied electric and magnetic fields in the longitudinal direction. The nanocomposite cylindrical shell is embedded in a viscoelastic medium which is simulated by a visco-pasternak model. The governing equations are derived using Hamilton’s principle, which are then solved by the differential quadrature method (DQM) for clamped boundary conditions. The influences of fluid velocity, geometrical parameters of the shell, viscoelastic foundation, orientation angle and percentage of BNNTs in polymer on the resonance frequency of the shell are investigated.

2 Structural definition

A schematic diagram of a double-layered shell embedded in a viscoelastic foundation is illustrated in Figure 1, in which geometrical parameters such as length L, average radius R and thickness h1+h2 are also indicated.

Figure 1: A schematic diagram of a double-layered shell modeled embedded in a viscoelastic foundation.
Figure 1:

A schematic diagram of a double-layered shell modeled embedded in a viscoelastic foundation.

As can be seen from Figure 1, the shell consists of two layers:

  • Viscoelastic nanocomposite cylindrical shell

  • Viscoelastic piezomagnetic coating layer.

From the strain-displacement relations, the stress and moment resultants are obtained separately for each layer. Then the total energy which includes the energies of the viscoelastic nanocomposite cylindrical shell and viscoelastic piezomagnetic coating layer is obtained using Hamilton’s principle in three orthogonal directions of the cylindrical coordinate system.

2.1 Nanocomposite cylindrical shell reinforced by BNNTs

2.1.1 Constitutive equations for piezoelectric materials

In this research, the nanocomposite cylindrical shell is made of PVDF reinforced by BNNTs. In piezoelectric materials, the constitutive equation includes stress (σ) and strain (ε) tensors on the mechanical side, as well as flux density (D) and field strength (E) tensors on the electrostatic side, which may be arbitrarily combined as follows [12]:

(1)σij=Cijklεkl-emijEm(i,j=1,,6)Dm=emijεij+ϵmkEk(m,k=1,2,3)

where Cijkl , emij and εmk are elastic constants, piezoelectric constants and dielectric constants, respectively, and Ek represents the electric field that is defined as a function of the electric potential ϕ, as given below:

(2)Ek=-ϕxk

2.1.2 Micromechanical models

A micromechanical model known as “XY PEFRC” or “YX PEFRC” is employed for modeling of coupled composite microstructures [9]. A representative volume element (RVE) is considered for predicting the elastic, piezoelectric and dielectric properties of the smart system. In this research, both matrix and reinforcements are assumed to be smart and the BNNTs are considered as longitudinal straight fibers in the composite shell. According to the XYPEFRC micromechanical method, the constitutive equations for the electro-mechanical behavior of the selected RVE are expressed as follows [10]:

(3)[σ1σ2σ3σ4σ5σ6]=[C11C12C13000C12C22C23000C13C23C33000000C44000000C55000000C66][ε1ε2ε3ε4ε5ε6]-[00e3100e3200e330e240e1500000][E1E2E3][D1D2D3]=[0000e150000e2400e31e32e33000][ε1ε2ε3ε4ε5ε6]+[ϵ11000ϵ22000ϵ33][E1E2E3]

The coefficients in the above equation in terms of the matrix and piezoelectric fibers properties and volume fraction of the reinforcement are given in Appendix A.

Assuming the plane stress condition and unidirectional electric field along the shell axis in the cylindrical coordinate system, Equation (3) can be reduced to the following relations:

(4){σx(1)σθ(1)τxθ(1)Dx}=[Q]{εxεθγxθEx}[Q]=[C11(1)C12(1)0-e11C12(1)C22(1)0-e1200C66(1)0e11e120ϵ11]

Super index (1) refers to the composite shell.

To consider the effects of the orientation angle of the BNNTs, the following transformation matrix can be employed:

(5)[Q˜]=[T][Q][T]T

The transformation matrix [T] is given as [13] follows:

(6)[T]=[cos2θsin2θ000-sin2θsin2θcos2θ000sin2θ001000000cosθsinθ0000-sinθcosθ0sinθcosθ-sinθcosθ000cos2θ-sin2θ]

where θ is the angle of BNNTs with respect to the shell axis.

2.1.3 Piezomagnetic coating

As previously mentioned, the outer surface of the nanocomposite shell is coated by a thin layer of piezomagnetic material that is sensitive to the magnetic field. In piezomagnetic materials, the constitutive equation includes stress (σ) and strain (ε) tensors on the mechanical side, as well as magnetic flux (B) and magnetic field strength (H) tensors on the magnetostatic side, which may be arbitrarily combined as follows [2]:

(7)σij=Cijklεkl-qmijHm(i,j=1,,6)Bm=qmijεij+μmkHk(m,k=1,2,3)

where Cijkl , qmij and μmk are elastic constants, piezomagnetic constants and magnetic permeability, respectively, and Hk represents the magnetic field that is defined as a function of the magnetic potential ψ, as given below:

(8)Hk=-ψxk

Since the unidirectional magnetic field is along the cylindrical shell axis, the stress-strain relations for the piezomagnetic coating layer are considered as follows:

(9){σx(2)σθ(2)τxθ(2)Bx}=[C11(2)C12(2)0-q11C12(2)C22(2)0-q1200C66(2)0q11q120μ11]{εxεθγxθHx}

Super index (2) refers to the coating layer.

3 Motion equations

3.1 Strain-displacement relationships

In order to calculate the middle-surface strain and curvatures, using Kirchhoff-Love assumptions, the displacement components of an arbitrary point are written as follows [4]:

(10)U(x,θ,z,t)=u(x,θ,t)-zw(x,θ,t)xV(x,θ,z,t)=v(x,θ,t)-zw(x,θ,t)RθW(x,θ,z,t)=w(x,θ,t)

where U, V and W are the displacements of an arbitrary point of the shell in the axial, circumferential and radial directions, respectively; u, v and w are the displacements of points on the middle surface of the shell; and z is the distance from an arbitrary point to the middle surface.

The strain components εx , εθ and γ at an arbitrary point of the shell are related to the middle-surface strains εxm , εθm and γxθm and changes in the curvature and torsion of the middle surface kx , kθ and k as follows:

(11){εxεθγxθ}shell={εxmεθmγxθm}+z{kxkθkxθ}

where z is the distance from the arbitrary point to the middle surface, and assuming Donnell’s hypothesis, the expressions for the middle-surface strains and the changes in the curvature and torsion of the middle surface by using Equation (10) become

(12){εxεθγxθ}shell=(uxvRθ+wRuRθ+vx)-z(2wx22wR2θ222wRxθ).

3.2 Energy method

The total potential energy (Π) of the embedded shell is the sum of the strain energy (U), kinetic energy (K) and the work done by the applied viscoelastic medium (WE) and fluid flow (WF):

(13)Π={KNCS+KPC}-({UNCS+UPC}-{WVM+WF}){Nanocomposite shell: NCSPiezomagnetic coat: PCViscoelastic medium: VMFluid: F

The strain energy U occupying the region V is given by [13]

(14)UNCS=12V(σx(1)εx+σθ(1)εθ+σxθ(1)γxθ-DxEx)dVUPC=12V(σx(2)εx+σθ(2)εθ+σxθ(2)γxθ-BxHx)dV

The kinetic energy of the structure may be expressed as follows:

(15)KNCS=ρNCS2v([Ut]2+[Vt]2+[Wt]2)dVKPC=ρPC2v([Ut]2+[Vt]2+[Wt]2)dV

3.3 External work

3.3.1 Viscoelastic medium

The surrounded viscoelastic medium including both normal and shear modulus, is modeled by considering the damping effect as follows [14]:

(16)WVM=-12A(FVMw)dA=A(-kWw+kg2w-cν[wt])wdA

where kW , kg and cν are spring, shear and damping modulus, respectively.

3.3.2 Fluid structure interaction

The velocity field vector (V=(Vx,Vz)) for the conveying fluid flow in the 3D cylindrical shell model is the relative velocity of the fluid and shell. This vector can be expressed as follows [15]:

(17)vx=ut+ufcos(θ)vθ=utvz=wt+ufsin(θ)

where θ=-wx and uf is the constant velocity of the fluid. In order to calculate the work done by the fluid, the well-known Navies-Stokes equation is used as follows [16]:

(18)ρfDVDt=-P+μ2V+Fbody force

where ρf , P and μ are the fluid density, static pressure and fluid viscosity, respectively.

In the Navies-Stokes equation DDt is defined as follows:

(19)DDt=t+vxx+vθRθ+vzz

Using Equations (17) and (19), Equation (18) can be expanded in the Z direction as follows:

(20)pzz=-ρf(2wt2+2uf2wxt+uf22wx2)+μ(3wx2t+3wR2θ2t+uf(3wx3+3wR2θ2x))

The work done by the fluid can be calculated as follows:

(21)Wfluid=(Ffluid=h1pzz)wdA=(-ρfh1(2wt2+2uf2wxt+vx22wx2))+μh1(3wx2t+3wR2θ2t+uf[3wx3+3wR2θ2x]))wdA

As noted earlier in this paper, both nanocomposite cylindrical shell and piezomagnetic coating layer are the viscoelastic material. Based on Kelvin’s model on elastic materials with the viscoelastic structural damping coefficient g, the elastic coefficients Cij are replaced with the operators Cij (1+g∂/∂t) [17]. Therefore, the equations are rewritten based on the new elastic coefficients.

3.4 Hamilton’s principle

Using Hamilton’s principle, the variational form of the equations of motion can be expressed as follows:

(22)δ0tΠdt=δ0t[{KNCS+KPC}-({UNCS+UPC}-{WVM+WF})]dt=0.

Applying Hamilton’s principle, the coupled motion equations are obtained as follows:

(23)δu:(-ρ¯1-ρ¯2β2β1)2u¯t¯2+(ρ¯2γ3γ3γ1)3w¯ξt¯2+(γ21+C¯11(2)γγ1)22u¯ξ2+g¯(γ+1C¯11(2)γ)23u¯ξ2t¯2+(C¯66(1)β12+C¯66(2)β1β2)2u¯θ2+g¯η(C¯66(1)β1+C¯66(2)β2)3u¯θ2t¯+(C¯12(1)γβ11+C¯66(1)γβ11+C¯12(2)γβ21+C¯66(2)γβ21)2v¯ξθ+γβ11(C¯12(1)+C¯12(2))w¯ξ+g¯(C¯12(1)β1+C¯66(1)β1+C¯12(2)β1+C¯66(2)β1)3v¯ξθt¯+γβ11(C¯12(1)+C¯12(2))2w¯ξt¯-(C¯11(2)γ1γ3)3w¯ξ3-g¯(C¯11(2)γ3)4w¯ξ3t¯-γβ13(C¯12(1)+C¯66(2))3w¯ξθ2-g¯(C¯12(1)β32+2C¯66(2)β3)4w¯θ2ξt¯+(e¯11γ12)2Φ¯ξ2+(q¯11γ1γ2)2Ψ¯ξ2=0
(24)δv:(-ρ¯1-ρ¯2β2β1)2v¯t¯2+(ρ¯2β3β3β1)3w¯θt¯2+(C¯12(1)γβ11+C¯66(1)γβ11)2u¯ξθ2+(+C¯12(2)γβ21+C¯66(2)γβ22)2u¯ξθ2+g¯(C¯12(1)β1+C¯66(1)β1+C¯12(2)β2+C¯66(2)β2)3u¯ξθt¯+(C¯22(1)β12+C¯22(2)β1β2)2v¯θ2+g¯η(C¯22(1)β1+C¯22(2)β2)3v¯θ2t¯+(C¯66(1)γ12+C¯66(2)γ1γ2)2v¯ξ2+g¯(C¯66(1)γ1+C¯66(2)γ1)3v¯ξ2t¯+(C¯22(1)β12+C¯22(2)β1β2)w¯θ+g¯η(C¯22(1)β1+C¯22(2)β2)2w¯θt¯-(C¯12(2)β1γ32+2C¯66(2)β1γ32)3w¯ξ2θ-g¯η(C¯12(2)γ32+2C¯66(2)γ32)4w¯ξ2θt¯-(C¯22(2)β1β32)3w¯θ3-g¯η(C¯22(2)β32)4w¯θ3t¯+(e¯12γ1β1)2Φ¯ξθ+(q¯12γ1β2)2Ψ¯ξθ=0,
(25)δw:(-ρ¯1-ρ¯2β2β1)2w¯t¯2+(ρ¯1γ12121+ρ¯2γ43γ1)4w¯ξ2t¯2+(ρ¯1β12121+ρ¯2β43β1)4w¯θ2t¯2+(2C¯12(2)β1γ32)2w¯ξ2+(-γ14121-C¯11(2)γ1γ43)4w¯ξ4+g¯(-γ13121-C¯11(2)γ43)5w¯ξ4t¯+(-C¯12(1)γ12β126-C¯66(1)γ1β13-2C¯12(2)ηβ1γ43-4C¯66(2)ηβ1γ43)4w¯ξ2θ2+g¯(-C¯12(1)γ12β126-C¯66(1)γ1β123-2C¯12(2)β43η-4C¯66(2)β43η)5w¯ξ2θ2t¯+(-C¯12(1)γ12β12-C¯12(2)γ12β22)(w¯ξ)2+(-C¯22(1)β12-C¯22(2)β1β2)w¯+g¯η(-C¯22(1)β1-C¯22(2)β2)w¯t¯+(2C¯22(2)β1β32)2w¯θ2+(-C¯22(1)β1412+C¯22(2)β1β43)4w¯θ4-(e¯12γ1β1)φ¯θ+(-q¯12β2γ1)ψ¯ξ+(q¯12β32γ1)3ψ¯ξθ2+(-ρ¯21γ1γ32)3u¯ξt¯2+(-C¯12(1)β1γ1-C¯12(2)β2γ1)u¯ξ+(C¯11(2)γ1γ32)3u¯ξ3+g¯(C¯11(2)γ32)4u¯ξ3t¯+g¯(-C¯12(1)β1-C¯12(2)β2)2u¯ξt¯+g¯β32γ1(C¯12(2)+2C¯66(2))3u¯ξθ2+g¯β32(C¯12(2)+2C¯66(2))4u¯ξθ2t¯+(-ρ¯21β1β32)3v¯θt¯2+(-C¯22(1)β1-C¯22(2)β2β1)v¯θ+g¯η(-C¯22(1)β1-C¯22(2)β21)2v¯θt¯+β1γ32(C¯12(2)+2C¯66(2))3v¯ξ2θ+g¯γ32η(C¯12(2)+2C¯66(2))4v¯ξ2θt¯+(C¯22(1)β1β32)3v¯θ3+g¯η(C¯22(1)β32)4v¯θ3t¯-(k¯w)w¯+(k¯gγ1)2w¯ξ2+(k¯gηβ1)2w¯θ2-(c¯v)w¯t¯-2w¯t¯2+(-2γ1u¯f)2w¯ξt¯+(-γ12u¯f2)2w¯ξ2+(γ1μ¯)3w¯ξ2t¯+(γ1ημ¯)3w¯θ2t¯+(γ12μ¯u¯f)3w¯ξ3+(γ1μ¯β1u¯f)3w¯θ2ξ=0
(26)δϕ:e¯11(2u¯ξ2)+e¯12η(2v¯θξ+w¯ξ)-2Φ¯ξ2=0
(27)δψ:q¯11(2u¯ξ2)+q¯12η(2v¯θξ+w¯ξ)-2Ψ¯ξ2-q¯11γ5(3w¯ξ3)-q¯12ηβ5(3w¯ξθ2)=0

where the dimensionless parameters are defined as follows:

(28)γ1=h1l,γ2=h2l,γ3=γ22+γ1γ22,γ4=4γ23+3γ12γ2+6γ1γ2212,3γ5=γ43γ32,β1=h1R1,β2=h2R1,β3=β22+β1β22,β4=4β23+3β12β2+6β1β2212,3β5=β43β32,ξ=xl,η=lR1,u¯=uh1,v¯=vh1,w¯=wh1ρ¯k=ρkρf,Ψ¯=ψΨ0,t¯=th1ρfC11(1),g¯=glC11(1)ρfk¯w=kwh1C11(1),k¯g=kgC11(1)l,c¯v=cvC11(1)ρfΦ¯=φΦ0,u¯f=ufρfC11(1)μ¯=μlC11(1)ρf,Φ0=h1C11(1)ϵ11,e¯11=e11C11(1)ϵ11e¯12=e12C11(1)ϵ11.

4 Solution procedures

The DQM is used in this study to solve the equations of motion. In this method, differential equations change into first algebraic equations with first order and weighting coefficients. In other words, the partial derivatives of a function (F) are approximated by a specific variable, at discontinuous points in the domain as a set of weighting series and its amount represented by the function itself at that point and other points throughout the domain. Let F be a function representing u̅, v̅, w̅ and Φ with respect to variables ξ and θ in the following domain of (0<ξ<L, 0<θ<2π) having Nξ ×Nθ grid points along these variables. The nth-order partial derivative of F(ξ, θ) with respect to ξ, the mth-order partial derivative of F(ξ, θ) with respect to θ and the (n+m)th-order partial derivative of F(ξ, θ) with respect to both ξ and θ may be expressed discretely [18] at the point (ξi , θi) as follows:

(29)dnF(ξi,θj)dξn=k=1NξAik(n)F(ξk,θj)n=1,,Nξ-1
(30)dmF(ξi,θj)dθm=l=1NθBjl(m)F(ξi,θl)m=1,,Nθ-1
(31)dn+mF(ξi,θj)dξndθm=k=1Nξl=1NθAik(n)Bjl(m)F(ξk,θl)

where Aik(n) and Bjl(m) are the weighting coefficients associated with the nth-order partial derivative of F(ξ, θ) with respect to ξ at the discrete point ξi and the mth-order derivative with respect to θ at θi , respectively, whose recursive formulae can be found in Ref. [19]. A more superior choice for the positions of the grid points is Chebyshev polynomials as expressed in [19].

According to the DQM, mechanical clamped and free electrical boundary conditions at both ends of the shell may be written as follows:

(32)  {wi1=vi1=ui1=0j=1NθA2jwji=0wNxi=vNxi=uNxi=0j=1NθA(Nx-1)jwji=0for i=1Nθ

Applying weighting coefficients and using Equations (23)–(27), a set of algebraic governing equations are obtained:

(33)[K]{d}+[C]{d˙}+[M]{d¨}=0

where [K], [M] and [C] are the stiffness, mass and damping matrix, respectively. It is assumed that the flow velocity is harmonically fluctuating and has the following form [19]:

(34)uf=v0[1+αcos(ωt)]

where v0 is the average flow velocity, α is the amplitude harmonic and ω is the frequency.

Substituting the pulsatile fluid velocity in Equation (33), the stiffness matrix coefficients and damping coefficient matrix are separated as follows:

(35)([K]+uf[K]f+uf2[K]ff){d}+([C]+uf[C]f){d˙}+[M]{d¨}=0

Finally, by substituting Equation (34) in Equation (35):

(36)([K]+(uf[1+αcos(ωt)])[K]f+(uf[1+αcos(ωt)])2[K]ff){d}+([C]+(uf[1+αcos(ωt)])[C]f){d˙}+[M]{d¨}=0

where [K]f, [K]ff and [C]f are stiffness and damping coefficient matrix of pulsating fluid. The bulletin helper method is used to solve Equation (36). In this way, the displacement vector {d} can be considered as follows [20]:

(37){d}=k=1,3,[{a}ksinkωt2+{b}kcoskωt2].

Studies show [20] the first dynamic range is the most important and largest range, so by substituting Equation (37) in Equation (36) and separating sin(ωt) and cos(ωt) coefficients, the following form is obtained to determine the scope of system stability:

(38)([K]+(1±α2)[K]f+(1±α+α22)[K]ff)+(±[C]ω22+(αω4±ω2)[C]f)-[M]ω24|=0

Equation (38) is an eigenvalue problem that can be solved by Matlab software.

5 Numerical results and discussion

Despite various research studies on the dynamic response of a cylindrical shell, the smart composite materials with special privileges have received less attention. In this research, the dynamic response of a smart composite cylindrical shell with a piezomagnetic coating layer, conveying pulsation fluid flow, was studied. This is a novel topic that has been studied for the first time and cannot be found in the literature. The material properties of the composite shell and coating layer are presented in Table 1 [2, 21, 22].

Table 1

Electro-magneto-mechanical properties of materials.

PropertyPVDFBNNTPropertyCoFe2O4
C11 (GPa)238.242035C11 (GPa)286.0
C2223.62035C22286.0
C3310.642035C33269.5
C123.98692C12173.0
C132.19692C13170.5
C231.92692C23170.5
C442.15672C4445.3
C554.4692C5545.3
C666.43692C6656.5
e31 (C/m2)-0.130q31 (N/Am)580.3
e32-0.1450q32580.3
e33-0.2760.95q33699.7
e24-0.0090q24550.0
e15-0.1350q15550.0
ε11 (C2/Nm2)110.67e-12110.67e-10μ11 (Ns2/C2)-590.0e-06
ε22106.07e-12110.67e-10μ22-590.0e-06
ε33106.07e-12110.67e-10μ33-157.0e-06

5.1 Effect of the nanotube volume percent on pulsation frequency

Figure 2 shows the effect of the BNNT volume percent in the PVDF matrix when the fiber angle is equal to 60°. As is observed, with the increase in the volume percent, the pulsation frequency increases. Since the elastic stiffness and density of the BNNT are higher than PVDF, with the increase in the volume percent of nanotubes in polymer composites, the equivalent stiffness coefficients and the density of the system increase and the system becomes stronger. It is worth mentioning that the instability zone reduces with the increase in the BNNT volume percent where this zone is the distance between two graphs.

Figure 2: Effect of the volume percent of BNNTs on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.
Figure 2:

Effect of the volume percent of BNNTs on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.

5.2 Effect of the fiber angle in the matrix on pulsation frequency

Figure 3 shows the effect of the BNNT alignment in the PVDF matrix on the dimensionless pulsation amplitude. As can be seen from the figure, with the increase in the fiber angle from 30° to 60° the pulsation frequency reduces significantly and the instability zone increases. On the other hand, the smart composite cylinder becomes susceptible to instability in the higher fiber angle. In fact, the alignment of the fiber with high angle reduces the integrity of the composite, so here 30° is the best option for the optimum design.

Figure 3: Effect of the BNNT orientation angle on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.
Figure 3:

Effect of the BNNT orientation angle on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.

5.3 Effect of the composite thickness ratio on pulsation frequency

Figure 4 illustrates the effect of the thickness to radius ratio of a nanotube reinforced composite, called the thickness ratio, on the dimensionless pulsation amplitude. As shown in the figure, the increase in the thickness ratio leads to the increase in the pulsation frequency. As we know, the composite cylinder with a higher aspect ratio is harder and stronger, and this is the scientific reason for the increase in the pulsation frequency.

Figure 4: Effect of β1 on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.
Figure 4:

Effect of β1 on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.

5.4 Effect of the coating layer thickness ratio on pulsation frequency

Figure 5 displays the effect of the thickness to radius ratio of a piezomagnetic layer as a coat for the composite cylinder on the dimensionless pulsation amplitude. As can be seen, the increase in the thickness ratio of the coating layer leads to the increase in the pulsation frequency like composite thickness ratio because this change increases the thickness of the cylinder and provides a stronger structure. The comparison between Figures 4 and 5 proves that the effect of the composite thickness ratio is more than the piezomagnetic coating layer.

Figure 5: Effect of β2 on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.
Figure 5:

Effect of β2 on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.

5.5 Effect of the aspect ratio on pulsation frequency

Figure 6 shows the effect of the length to the radius ratio of a composite shell called the aspect ratio on the pulsation frequency. As is clear from the figure, the increase in the aspect ratio decreases the pulsation frequency. Instability occurs earlier in high aspect ratio, because the length of the composite shell increases more in comparison with the radius. It is worth to mention that with the increase in the aspect ratio the stability zone increases, while the pulsation frequency reduces.

Figure 6: Effect of η on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.
Figure 6:

Effect of η on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.

5.6 Effect of the fluid flow velocity on pulsation frequency

All figures in this paper are plotted for a dimensionless fluid flow velocity v̅x =1, but Figure 7 shows the effect of different flow velocities on the pulsation frequency. As can be observed in Figure 7, with the increase in fluid velocity, pulsation frequency decreases. So, the greater fluid velocity leads to a large stability zone and low pulsation frequency.

Figure 7: Effect of the fluid velocity on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.
Figure 7:

Effect of the fluid velocity on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.

5.7 Effect of viscoelastic material on pulsation frequency

Figure 8 is examined in this section to present the velocity amplitude versus pulsation frequency for different viscoelastic coefficients using a dimensionless damping parameter g̅. As was expected, the increase in this ratio decreases the pulsation frequency. It is visible from this figure that increasing the viscoelastic coefficient increases the system damping and the pulsation frequency is pulled to the left (the system becomes unstable at lower frequencies).

Figure 8: Effect of the viscoelastic material damping coefficient on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.
Figure 8:

Effect of the viscoelastic material damping coefficient on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.

5.8 Effect of elastic medium on pulsation frequency

5.8.1 Winkler spring modulus

The effect of the dimensionless Winkler modulus (k̅w) on the pulsation frequency is shown in Figure 9. As can be seen, with the increase in the Winkler modulus, pulsation frequency increases. The Winkler elastic modulus represents the elastic medium around the cylindrical shell and causes the hardening of the system where the frequency increases. The instability zone reduces with the increase in the Winkler modulus.

Figure 9: Effect of the spring constants of viscoelastic foundation on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.
Figure 9:

Effect of the spring constants of viscoelastic foundation on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.

5.8.2 Pasternak shear modulus

The effect of the dimensionless Pasternak modulus (k̅g) on the pulsation frequency is shown in Figure 10. This figure reveals that with the increase in the Pasternak shear modulus pulsation frequency increases. Since in Pasternak foundation both shear and normal modulus are considered, so it is expected that the Pasternak model is more efficient than the Winkler model. This result can be observed in comparison with Figures 9 and 10.

Figure 10: Effect of the shear constants of viscoelastic foundation on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.
Figure 10:

Effect of the shear constants of viscoelastic foundation on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.

5.8.3 Damping coefficient

The effect of the dimensionless damping coefficient (c̅v) on the pulsation frequency is shown in Figure 11. It is clear from the figure that by increasing the damping coefficient, the pulsation frequency decreases. c̅v indicates the damping area around the elastic cylindrical shell which reduces the stability of the system.

Figure 11: Effect of the damping constants of viscoelastic foundation on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.
Figure 11:

Effect of the damping constants of viscoelastic foundation on the dimensionless pulsation amplitude versus dimensionless pulsation frequency.

6 Conclusion

In this research, dynamic buckling of a BNNT-reinforced composite conveying pulsating fluid flow was studied. The external surface of the composite cylinder was coated by a piezomagnetic layer and the structure was subjected to the electro-magneto-mechanical loadings. PVDF and CoFe2O4 were selected for the composite matrix and coating layer, respectively. Harmonic oscillations were generated due to pulsating fluid flow. The results obtained by solving the motion equations of the smart coated composite cylindrical shell are listed as follows:

  • Increasing the volume percent of BNNTs in the polymer matrix for the fiber angle of θ=60° leads to an increase in the pulsation frequency, so the system becomes more stable.

  • Increasing the BNNT angle in the polymer matrix leads to a decrease in the pulsation frequency and the smart composite cylinder becomes susceptible to instability.

  • Increasing the thickness ratio of the BNNT-reinforced composite and piezomagnetic coating layer leads to an increase in the pulsation frequency and the stability of the system.

  • The higher values of the aspect ratio, fluid flow velocity, viscoelastic and damping coefficients are the negative factors since they reduce the stability of the system.

  • Winkler spring coefficient and Pasternak shear modulus increase the stability of the system.

The result of this study can be useful for the design and manufacturing of rotating machines, hydraulic systems and motors.


Corresponding author: Ali Ghorbanpour Arani, Department of Solid Mechanics, Faculty of Mechanical Engineering, University of Kashan, Kashan, Iran; and Institute of Nanoscience and Nanotechnology, University of Kashan, Kashan, Iran, e-mail:

Acknowledgments

The authors would like to thank the reviewers for their comments and suggestions to improve the clarity of this paper. This work was supported by University of Kashan (grant number 363443/54). Iranian Nanotechnology Development Committee provided financial support.

Appendix A

The components of the constitutive equation of the PEFRC in terms of matrix and reinforcement properties are presented here according to [9, 10]. The superscripts “m” and “p” indicate the matrix and piezoelectric fiber, respectively.

The components of the elastic stiffness matrix:

(A.1)Cc11=C11pC11mρC11m+(1-ρ)C11p,Cc12=C11[ρC12pC11p+(1-ρ)C12mC11m],Cc13=C11[ρC13pC11p+(1-ρ)C13mC11m],Cc22=ρC22p+(1-ρ)C22m+C122C11-ρ(C12p)2C11p-(1-ρ)(C12m)2C11m,Cc23=ρC23p+(1-ρ)C23m+Cc12Cc13Cc11-ρC12pC13pC11p-(1-ρ)C12mC13mC11m,Cc33=ρC33p+(1-ρ)C33m+C132Cc11-ρ(C13p)2C11p-(1-ρ)(C13m)2C11m,Cc44=ρC44p+(1-ρ)C44m,Cc55=AB2+AC,Cc66=C66pC66mρC66m+(1-ρ)C66p

The components of the piezoelectric matrix:

(A.2)e31=C11[ρe31pC11p+(1-ρ)e31mC11m],e32=ρe32p+(1-ρ)e32m+C12e13C11-ρC12pe31pC11p-(1-ρ)C12me31mC11m,e33=ρe33p+(1-ρ)e33m+C13e31Cc11-ρC13pe31pC11p-(1-ρ)C13me31mC11m,e24=ρe24p+(1-ρ)e24m,e15=BB2+AC

Dielectric constants:

(A.3)ϵ11=CB2+AC,ϵ22=ρϵ22p+(1-ρ)ϵ22m,ϵ33=ρϵ33p+(1-ρ)ϵ33m-e312C11+ρ(e31p)2C11p+(1-ρ)(e31m)2C11m,

where the parameters A, B and C are defined as follows:

(A.4)A=ρC55p(e15p)2+C55pϵ11p+(1-ρ)C55m(e15m)2+C55mϵ11m,B=ρe15p(e15p)2+C55pϵ11p+(1-ρ)e15m(e15m)2+C55mϵ11m,C=ρε11p(e15p)2+C55pϵ11p+(1-ρ)ε11m(e15m)2+C55mϵ11m.

Appendix B

(B.1)δu:{(γ12+C¯11(2)γ1γ2)k=1NxAik(2)+(C¯66(1)β1(2)+C¯66(2)β1β2)k=1NθBjp(2)}u¯(xk,θj)+{(C¯12(1)γ1β1+C¯66(1)γ1β1+C¯12(2)β1γ2+C¯66(2)β1γ2)k=1Nxp=1NθAik(1)Bjp(1)}v¯(xk,θj)+{γ1β1(C¯12(1)+C¯12(2))k=1NxAik(1)-(C¯11(2)γ1γ32)k=1NxAik(3)-γ1β32(C¯12(2)+2C¯66(2))k=1Nxp=1NθAik(1)B(jp)2}w¯(xk,θj)+{(e¯11γ12)k=1NxAik(2)}Φ¯(xk,θj)+{(q¯11γ1γ2)k=1NxAik(2)}Ψ¯(xk,θj)+{(γ1g¯+C¯11(2)γ2g¯)k=1NxAik(2)+(C¯66(1)β1ηg¯+C¯66(2)β2ηg¯)k=1NθBjp(2)}u¯˙(xk,θj)+{(C¯12(1)β1g¯+C¯66(1)β1g¯+C¯12(2)β2g¯+C¯66(2)β2g¯)k=1Nxp=1NθAik(1)Bjp(1)}v¯˙(xk,θj)+{g¯(C¯12(1)β1+C¯12(2)β2)k=1NxAik(1)-(C¯11(2)g¯γ32)k=1NxAik(3)-g¯β32(C¯12(2)+2C¯66(2))k=1Nxp=1NθAik(1)Bjp(2)}w¯˙(xk,θj)+{(-ρ¯1-ρ¯2β2β1)}u¯¨(xk,θj)+{(ρ¯2γ3γ3γ1)k=1NxAik(1)}w¯¨(xk,θj)=0,
(B.2)δv:{(C¯12(1)γ1β1+C¯66(1)γ1β1+C¯12(2)β2γ2+C¯66(2)β2γ1)k=1Nxp=1NθAik(1)Bjp(1)}u¯(xk,θj)+{(C¯22(1)β12+C¯22(2)β1β2)k=1NθBjp(2)+(C¯66(1)γ1(2)+C¯66(2)γ1γ2)k=1NxAik(2)}v¯(xk,θj)+{(C¯22(1)β12+C¯22(2)β1β2)k=1NθBjp(1)-β1γ32(C¯12(2)+2C¯66(2))k=1Nxp=1NθAik(2)Bjp(1)-(C¯22(2)β1β23)k=1NθBjp(3)}w¯(xk,θj)+{(e¯12γ1β1)k=1Nxp=1NθAik(1)Bjp(1)}Φ¯(xk,θj)+{(q¯12γ1β2)k=1Nxp=1NθAik(1)Bjp(1)}Ψ¯(xk,θj)+{g¯(C¯12(1)β1+C¯66(1)β1+C¯12(2)β2+C¯66(2)β2)k=1Nxp=1NθAik(1)Bjp(1)}u¯˙(xk,θj)+{g¯η(C¯22(1)β1+C¯22(2)β2)k=1NθBjp(2)+g¯(C¯66(1)γ1+C¯66(2)γ2)k=1NxAik(2)}v¯˙(xk,θj)+{g¯η(C¯22(1)β1+C¯22(2)β2)k=1NθBjp(1)-g¯ηγ32(C¯12(2)+2C¯66(2))k=1Nxp=1NθAik(2)Bjp(1)(C¯22(2)g¯ηβ23)k=1NθBjp(3)}w¯˙(xk,θj)+{(-ρ¯1-ρ¯2β2β1)}v¯¨(xk,θj)+{(ρ¯2β3β3β1)k=1NθBjp(1)}w¯¨(xk,θj)=0,
(B.3)δw:{γ1(-C¯12(1)β1-C¯12(2)β2)k=1NxAik(1)+(C¯11(2)γ1γ32)k=1NxAik(3)+γ1β32(C¯11(2)+2C¯66(2))k=1Nxp=1NθAik(1)Bjp(2)}u¯(xk,θj)+{(-C¯22(1)β12-C¯22(2)β1β2)k=1NθBjp(1)+β1γ32(C¯12(2)+2C¯66(2))k=1Nxp=1NθAik(2)Bjp(1)+(C¯22(2)β1β32)k=1NθBjp(3)}v¯(xk,θj)+{(2C¯12(2)β1γ32)k=1NxAik(2)+(-γ1412-C¯11(2)γ1γ43)k=1NxAik(4)+(-C¯12(1)γ12β126-C¯66(1)γ1β13-2C¯12(2)β1ηγ43-4C¯66(2)β1ηγ43)k=1Nxp=1NθAik(2)Bjp(2)+(-C¯22(1)β12-C¯22(2)β1β2)+(2C¯22(2)β1β32)k=1NθBjp(2)+(-C¯22(1)β1412-C¯22(2)β1β43)k=1NθBjp(4)-(K¯w)+(K¯gγ1)k=1NxAik(2)-(K¯gηβ1)k=1NθBjp(2)+(γ1γ1v¯x2)k=1NxAik(2)+(γ1γ1μ¯v¯x)k=1NxAik(3)+(β1γ1μ¯v¯x)k=1Nxp=1NθAik(1)Bjp(2)}w¯(xk,θj)+{-(e¯12β1γ1)k=1NxAik(1)}Φ¯(xk,θj)+{(q¯11γ1γ32)k=1NxAik(3)+(-q¯12γ1β2)k=1NxAik(1)+(q¯12γ1β32)k=1Nxp=1NθAik(1)Bjp(2)}Ψ¯(xk,θj)+{g¯[(-C¯12(1)β1-C¯12(2)β2)k=1NxAik(1)+(C¯11(2)γ32)k=1NxAik(3)+β32(C¯11(2)+2C¯66(2))k=1Nxp=1NθAik(1)Bjp(2)]}u¯˙(xk,θj)+{g¯[(-C¯22(1)β1-C¯22(2)β2)k=1NθBjp(1)+γ32(C¯12(2)+2C¯66(2))k=1Nxp=1NθAik(2)Bjp(1)+(C¯22(2)β32)k=1NθBjp(3)]}v¯˙(xk,θj)+{g¯(2C¯12(2)γ32)k=1NxAik(2)+g¯(-γ1312-C¯11(2)γ43)k=1NxAik(4)+g¯(-C¯12(1)γ1β126-C¯66(1)γ1β123-2C¯12(2)β43η-4C¯66(2)β43η)k=1Nxp=1NθAik(2)Bjp(2)+g¯η(-C¯22(1)β1-C¯22(2)β2)+g¯(2C¯22(2)β32)k=1NθBjp(2)+g¯η(-C¯22(1)β1312-C¯22(2)β43)k=1NθBjp(4)-(C¯v)+(-2γ1v¯x)k=1NxAik(1)+(γ1μ¯)k=1NxAik(2)+(β1ημ¯)k=1NθBjp(2)}w¯˙(xk,θj)+{(-ρ¯2γ32γ1)k=1NxAik(1)}u¯¨(xk,θj)+{(-ρ¯2β32β1)k=1NθBjp(1)}v¯¨(xk,θj)+{(-ρ¯1-ρ¯2β2β1)+(ρ¯1γ1212+ρ¯2γ43γ1)k=1NxAik(2)+(ρ¯1β1212+ρ¯2β43β1)k=1NθBjp(2)-(1)}w¯¨(xk,θj)=0,
(B.4)δϕ:{(e¯11)k=1NxAik(2)}u¯(xk,θj)+{(e¯12η)k=1Nxp=1NθAik(1)Bjp(1)}v¯(xk,θj)+{(e¯12η)k=1NxA(1)ik}w¯(xk,θj)-k=1NxAik(2)Φ¯(xk,θj)=0,
(B.5)δψ:{(q¯11)k=1NxAik(2)}u¯(xk,θj)+{(q¯12η)k=1Nxp=1NθAik(1)Bjp(1)}v¯(xk,θj)+{(q¯12η)k=1NxAik(1)+(-q¯11γ5)k=1NxAik(3)+(-q¯12ηβ5)k=1Nxp=1NθAik(1)Bjp(2)}w¯(xk,θj)-k=1NxAik(2)Ψ¯(xk,θj)=0,

References

[1] Jalili N. Piezoelectric-Based Vibration Control from Macro to Micro/Nano Scale Systems. Snell Engineering Center, Northeastern University: Huntington Avenue, Boston, MA, 2010.10.1007/978-1-4419-0070-8Search in Google Scholar

[2] Tsai YH, Wu CP, Syu YS. Eur. J. Mech. A-Solid. 2008, 27, 79–105.10.1016/j.euromechsol.2007.05.002Search in Google Scholar

[3] Wu CP, Tsai YH. Eur. J. Mech. A-Solid 2010, 29, 166–181.10.1016/j.euromechsol.2009.09.004Search in Google Scholar

[4] Mosallaie Barzoki AA, Ghorbanpour Arani A, Kolahchi R, Mozdianfard MR. Appl. Math. Model. 2012, 36, 2983–2995.10.1016/j.apm.2011.09.093Search in Google Scholar

[5] Wang L. Int. J. Nonlinear. Mech. 2009, 44, 115–121.10.1016/j.ijnonlinmec.2008.08.010Search in Google Scholar

[6] Ariaratnam ST, Namachchivaya NS. J. Sound Vib. 1986, 2, 215–230.10.1016/0022-460X(86)90233-6Search in Google Scholar

[7] Panda LN, Kar RC. J. Sound Vib. 2008, 309, 375–406.10.1016/j.jsv.2007.05.023Search in Google Scholar

[8] Liang F, Su Y. Appl. Math. Model. 2013, 37, 6821–6828.10.1016/j.apm.2013.01.053Search in Google Scholar

[9] Tan P, Tong L. Compos. Part A-Appl. S 2002, 33, 631–645.10.1016/S1359-835X(02)00015-5Search in Google Scholar

[10] Tan P, Tong L. Compos. Struct. 2002, 57, 101–108.10.1016/S0263-8223(02)00073-9Search in Google Scholar

[11] Fang XQ, Liu JX, Gupta V. Nanoscale 2013, 5, 1716–1726.10.1039/c2nr33531jSearch in Google Scholar PubMed

[12] Ghorbanpour Arani A, Kolahchi R, Vossough H. Physica B 2012, 407, 4458–4465.10.1016/j.physb.2012.07.046Search in Google Scholar

[13] Ghorbanpour Arani A, Shajari AR, Amir S, Loghman A. Physica E 2012, 45, 109–121.10.1016/j.physe.2012.07.017Search in Google Scholar

[14] Ghorbanpour Arani A, Haghparast E, Heidari Rarani M, Khoddami Maraghi Z. Comp. Mater. Sci. 2015, 96, 448–458.10.1016/j.commatsci.2014.06.013Search in Google Scholar

[15] Ke LL, Wang YS, Wang ZD. Compos. Struct. 2012, 94, 2038–2047.10.1016/j.compstruct.2012.01.023Search in Google Scholar

[16] Wang L, Ni Q. Mech. Res. Commun. 2009, 36, 833–837.10.1016/j.mechrescom.2009.05.003Search in Google Scholar

[17] Pouresmaeeli S, Ghavanloo E, Fazelzadeh SA. Compos. Struct. 2013, 96, 405–410.10.1016/j.compstruct.2012.08.051Search in Google Scholar

[18] Karami G, Malekzadeh P. Comput. Meth. Appl. Mech. Eng. 2002, 191, 3509–3526.10.1016/S0045-7825(02)00289-XSearch in Google Scholar

[19] Civalek Ö. Eng. Struct. 2004, 26, 171–186.10.1016/j.engstruct.2003.09.005Search in Google Scholar

[20] Patel SN, Datta PK, Sheikh AH. Thin Wall Struct. 2006, 44, 321–333.10.1016/j.tws.2006.03.004Search in Google Scholar

[21] Salehi-Khojin A, Jalili N. Compos. Sci. Technol. 2008, 68, 1489–1501.10.1016/j.compscitech.2007.10.024Search in Google Scholar

[22] Cheng ZQ, Limb CW, Kitipornchai S. Int. J. Solids Struct. 2000, 37, 3153–3175.10.1016/S0020-7683(99)00036-0Search in Google Scholar

Received: 2015-4-20
Accepted: 2015-9-26
Published Online: 2015-12-29
Published in Print: 2017-5-1

©2017 Walter de Gruyter GmbH, Berlin/Boston

This article is distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Articles in the same Issue

  1. Frontmatter
  2. Original articles
  3. Wave propagation in functionally graded piezoelectric-piezomagnetic rectangular bars
  4. Graphene/poly(vinylidene fluoride) dielectric composites with polydopamine as interface layers
  5. A novel biaxial double-negative metamaterial for electromagnetic rectangular cloaking operation
  6. Formation of homogenous copper film on MWCNTs by an efficient electroless deposition process
  7. Nano-SiCp/Al2014 composites with high strength and good ductility
  8. Microstrip line-fed monopole antenna on an epoxy-resin-reinforced woven-glass material for super wideband applications
  9. Influence of casting speed on fabricating Al-1%Mn and Al-10%Si alloy clad slab
  10. Thermal insulating epoxy composite coatings containing sepiolite/hollow glass microspheres as binary fillers: morphology, simulation and application
  11. Analysis of influence of fibre type and orientation on dynamic properties of polymer laminates for evaluation of their damping and self-heating
  12. Dynamic stability of nanocomposite viscoelastic cylindrical shells coating with a piezomagnetic layer conveying pulsating fluid flow
  13. Buckling and layer failure of composite laminated cylinders subjected to hydrostatic pressure
  14. One-step preparation and characterization of core-shell SiO2/Ag composite spheres by pulse plating
  15. The failure mechanism of carbon fiber-reinforced composites under longitudinal compression considering the interface
  16. A thermal-plastic model of friction stir welding in aluminum alloy
  17. A model for longitudinal tensile strength prediction of low braiding angle three-dimensional and four-directional composites
  18. Nonlinear stability of shear deformable eccentrically stiffened functionally graded plates on elastic foundations with temperature-dependent properties
  19. Design and multibody dynamics analyses of the novel force-bearing structures for variable configuration spacecraft
Downloaded on 5.3.2026 from https://www.degruyterbrill.com/document/doi/10.1515/secm-2015-0369/html
Scroll to top button