1 INTRODUCTION
Revolution shell structures involving functionally graded material (FGM) conical shells potentially have wide application in many modern industry fields such as aerospace, airplane, missile, booster and other aerospace vehicles (^{Koizumi M. 1993}, ^{Shen HS. 2009}). Therefore, the free vibration of rotating truncated conical shells is one of interesting and important problems and has received considerable attention of researchers.
For unstiffened conical shells, much significant results are obtained. ^{Chandrasekharan and Ramamurti (1981} and 1982) studied axisymmetric and asymmetric free vibrations of laminated conical shells using the Rayleigh-Ritz procedure. ^{Shu (1996}) presented an efficient approach for analyzing the free vibration of conical shells. The same author (1996) investigated the free vibration of composite laminated conical shells by generalized differential quadrature. By using the Galerkin and harmonic balance methods, ^{Xu et al. (1996}) studied the nonlinear free vibration of a symmetrically laminated, cross-ply, geometrically perfect thick conical shell with its two ends both clamped and both simply supported. ^{Lam and Hua (1997} and ^{1999}) analyzed the vibration and influences of boundary conditions on the frequency characteristics of rotating truncated circular conical shells. Following this direction, the frequency analysis of rotating truncated circular orthotropic conical shells with different boundary conditions was studied by ^{Hua (2000}). That same author (2000), based on the Love first approximation theory and Galerkin procedure, and taking into account the influences of centrifugal and Coriolis accelerations, investigated frequency characteristics of a rotating truncated circular layered conical shell. With the improved generalized differential quadrature method, ^{Lam et al. (2002}) analyzed free vibration characteristics of truncated conical panels. Using the classical thin shell theory and the element-Free kp-Ritz method, ^{Liew et al. (2005}) investigated the free vibration of thin conical shells. ^{Civalek (2006}) proposed the discrete singular convolution method for analyzing the free vibration of rotating conical shells in which a regularized Shannon's delta kernel is selected as the singular convolution to illustrate his algorithm. ^{Sofiyev et al. (2010}), by using the Donnell shell theory and Galerkin method, presented an analytical procedure to study the free vibration and stability of homogeneous and non-homogeneous orthotropic truncated and complete conical shells with clamped edges under external pressure.
For unstiffened FGM conical shells, there are many available results. ^{Tornabene (2009}) and ^{Tornabene et al. (2009}), based on the first order shear deformation theory and 2-D differential quadrature solution, studied the free vibration analysis of functionally graded conical, cylindrical and annular plates structures using 2-D differential quadrature solution. ^{Sofiyev (2009}) analyzed the vibration and stability behavior of freely supported FGM conical shells subjected to external pressure by Galerkin method. The same author (2012) analyzed the nonlinear vibration of FGM truncated conical shells by analytical approach. For unstiffened FGM conical panels, ^{Bich et al. (2012}) investigated by analytical method the linear mechanical buckling of that structure using the classical shell theory and Galerkin method. The investigation on the linear buckling of truncated hybrid FGM conical shells with piezoelectric layers subjected to combined action of thermal and electrical loads was reported by ^{Torabi et al. (2013}). Based on the First order shear deformation theory (FSDT), ^{Malekzadeh and Heydarpour (2013}) studied effects of centrifugal and Coriolis, of geometrical and material parameters on the free vibration behavior of rotating FGM un-stiffened truncated conical shells subjected to different boundary conditions.
As can be seen the above introduced results only relate to unstiffened structures. However, in practice, plates and shells including conical shells usually are reinforced by stiffeners system to provide the benefit of added load carrying capability with a relatively small additional weight. Thus, the study on dynamic behavior of those structures is significant practical problem.
For stiffened isotropic or orthotropic conical shells, ^{Weingarten (1965}) studied the free vibration of a ring-stiffened simply supported conical shell by considering an equivalent orthotropic shell and using Galerkin method. Applying the energy approach, ^{Crenwelge and Muster (1969}) investigated the resonant frequencies of simply supported ring-stiffened, and ring and stringer-stiffened conical shells. ^{Rao and Reddy (1981}) studied the minimum weight design of axially loaded simply supported stiffened conical shells with natural frequency constraints. The influence of placing the stiffeners inside as well as outside the conical shell on the optimum design is considered. The expressions for the critical axial buckling load and natural frequency of vibration of conical shell also are obtained. By applying structural symmetry techniques, ^{Mustaffa and Ali (1987}) investigated free vibration characteristics of stiffened cylindrical and conical shells. ^{Srinivasan and Krishnan (1989}), using the integral equation for the space domain and mode superposition for the time domain, obtained the results on the dynamic response analysis of stiffened conical shell panels in which the effect of eccentricity is taken into account. ^{Mecitoglu (1996}) studied the vibration characteristics of stiffened truncated conical shells based on the Donnell-Mushtari thin shell theory, the stiffeners smeared technique and the collocation method. The problem on the free vibration of rotating composite conical shells with stringer and ring stiffeners was solved by ^{Talebitooti et al. (2010}).
For stiffened FGM shells, ^{Najafizadeh et al. (2009}) with the linearized stability equations in terms of displacements studied buckling of FGM cylindrical shell reinforced by rings and stringers under axial compression. The stiffeners and skin, in their work, are assumed to be made of functionally graded materials and its properties vary continuously through the thickness direction. Following this direction, ^{Dung and Hoa (2015}) obtained the results on the static nonlinear buckling and post-buckling analysis of eccentrically stiffened FGM circular cylindrical shells under torsional loads in thermal environment. The material properties of shell and stiffeners are assumed to be continuously graded in the thickness direction. Galerkin method was used to obtain closed-form expressions to determine critical buckling loads. ^{Dung et al. (2014}) investigated the static buckling of FGM conical shells reinforced by FGM stiffeners under axial compressive load and external pressure by analytical method. The change of distance between stringers is considered in that work. By considering homogenous stiffeners, ^{Bich et al. (2013}) obtained the results on the nonlinear static and dynamic analysis of eccentrically stiffened FGM cylindrical shells and doubly curved thin shallow shells based on the classical shell theory. The governing equations of motion were derived by using the smeared stiffeners technique and the classical shell theory with von Karman geometrical nonlinearity. The nonlinear critical dynamic buckling load is found according to the Budiansky-Roth criterion. ^{Dung et al. (2013}) studied a mechanical buckling of eccentrically stiffened functionally graded (ES-FGM) thin truncated conical shells subjected to axial compressive load and uniform external pressure load based on the smeared stiffeners technique and the classical shell theory and considering homogenous stiffeners. ^{Dung and Nam (2014}) solved the problem on the nonlinear dynamic analysis of eccentrically stiffened functionally graded circular cylindrical thin shells under external pressure and surrounded by an elastic medium. The nonlinear critical dynamic buckling load is found according to the Budiansky-Roth criterion.
From the review of the literature, as can be seen the free vibration of rotating FGM truncated conical shell reinforced by eccentric stiffener system and taking into account the influences of centrifugal force and Coriolis acceleration still is not investigated. The aim of this paper is to investigate just mentioned problem. We focus on three new contributions as: FGM truncated conical shells are reinforced by FGM stringers and rings; A change of spacing between stringer stiffeners is considered; A centrifugal force and Coriolis acceleration are taken into account. The governing equations are derived using Donnell shells theory and smeared stiffeners technique. For stiffened FGM conical shells, it is difficult that these equations are a couple set of three variable coefficient partial differential equations, while for stiffened cylindrical shells those governing equations only are a couple set of constant coefficient partial differential equations. This difficulty is overcome in the paper. The sixth order polynomial equation for frequency w is obtained by applying Galerkin method. Numerical simulations are been done to show effects of geometrics parameters, vibration modes and rotating speed, reinforcement stiffener on frequency characteristics of the shell. The present results are validated by comparing with those in the literature.
2 MATERIAL PROPERTIES OF SHELL AND STIFFENERS
Consider a thin circular truncated conical shell with the semi-vertex angle a thickness h, length L and small base radius r as shown in Fig. 1. The curvilinear coordinate system is defined as (x, θ, z), where the origin is located in the middle surface of the shell, x is in the generatrix direction measured from the vertex of shell, θ is in the circumferential direction and the axes z is perpendicular to the plane (x, θ) and its direction is the outwards normal direction of the conical shell. Denotes x _{0}- the distance from the vertex to small base, and u, v and w - the displacement components of a point in the middle surface in the direction x, q and z, respectively.
Assume that the FGM conical shell rotates about its vertical symmetrical axis with a constant angular velocity denoted by Ω. Also assume that the shell is stiffened by closely spaced circular rings and longitudinal stringers and the stiffeners and skin are made of functionally graded materials varying continuously through the thickness direction of the shell with the power law. Two cases are investigated in this work.
Case 1: Conical shell with ceramic outside surface and metal inside surface and inside stiffener.
Case 2: Conical shell with ceramic outside surface and metal inside surface and outside stiffener.
For case 1, Young Modulus and densities of FGM shell and FGM stiffeners are given by (^{Najafizadeh et al., 2009}; ^{Dung et al., 2014})
where ν _{sh} = ν _{s} = ν _{r} = ν = const, k, k _{2} and k _{3} are volume fractions indexes of shell, stringer and ring, respectively and subscripts c, m, sh, s and r denote ceramic, metal, shell, longitudinal stringers and circular ring, respectively. It is evident that, from Eqs. (1)-(6), a continuity between the shell and stiffeners is satisfied. Note that the thickness of the stringer and the ring are respectively denoted by h _{1}, and h _{2}; and E_{c}, E_{m} are Young Modulus of the ceramic and metal; and E_{sh}, E_{s}, E_{r}, ρ_{sh}, ρ_{s}, ρ_{r} are Young Modulus and densities of shell, of stiffener in the x-direction and θ-direction, respectively. The coefficient ν is Poison's ratio.
Note k _{2} = k _{3} = k, when k _{2} →∞, k _{3} →∞ leads to homogeneous stiffener.
The Poisson's ratio ν is assumed to be constant.
Young Modulus and densities for the case 2 are given in Appendix I.
3 FUNDAMENTAL EQUATIONS
According to the Donnell shell theory, the strain components at distance z from the reference surface of shell are of the form (^{Brush and Almroth, 1975}; ^{Reddy, 2004}; ^{Volmir, 1972})
where ε _{xm} and ε_{θ} _{m} are the middle surface strains in meridional and circumferential direction, and γ _{x} _{θ} _{m} is the shear strain of the middle surface of the shell, and k_{x}, k _{θ} and k_{x} _{θ} are the change of curvatures and twist, respectively. They are related to the displacement components u, v, w as (^{Brush and Almroth, 1975}; ^{Volmir, 1972}; ^{Hua, 2000})
Using Hooke's Law, the stress-strain relations are expressed by:
For the conical shell
For the stringer and ring stiffeners
where the subscripts sh and st denote shell and stiffeners, respectively, and E_{s}, E_{r} are Young Modulus of stringer and ring stiffener respectively.
The contribution of stiffeners is taken into account by using the smeared stiffener technique. In addition, the change of spacing between stringers in the meridional direction also is considered. Integrating the above stress-strain equations and their moments through the thickness of the shell, the expressions for force and moment resultants of ES-FGM truncated conical shells are defined as (^{Brush and Almroth, 1975}; ^{Najafizadeh et al., 2009}; ^{Dung et al., 2014})
where C _{1} ^{0}, C _{2}, d _{1}, d _{2}, E _{1} _{s} , E _{1} _{r} , E _{3} _{s} , E _{3} _{r} , A_{ij}, B_{ij}, D_{ij} can be found in Appendix I.
The fundamental equations for the vibration of rotating truncated conical shells, basing on the Donnell shell theory, are as (^{Hua, 2000}; ^{Chen et al., 1993})
where ρ_{1} can be found in Appendix I, and the quality N_{θ} ^{0} is defined as the initial hoop tension due to the centrifugal force effect and given by (^{Hua, 2000})
Note that the system of equations (13) not only consists of three terms of relative accelerations ∂^{2} u/∂t ^{2}, ∂^{2} v/∂t ^{2} and ∂^{2} w/∂t ^{2} but also four terms of Coriolis accelerations 2Ω sin α∂v/∂t, 2Ω sin α∂u/∂t, 2Ω cos α∂w/∂t and 2Ω cos α∂v/∂t.
Introducing Eqs. (7-10) into Eqs. (11-12), then substituting the resulting equations and Eq. (14) into Eq. (13), the vibration equations in terms of displacements for rotating ES-FGM truncated conical shell are obtained as
here R_{ij} are partial differential operators and defined as following
The system of Eqs. (15-17) is used to analyze the frequency characteristics of rotating ES-FGM truncated conical shells. It is difficult that these equations are a couple set of three variable coefficient partial differential equations. This is a main difference between the free vibration analyses of rotating conical shell and cylindrical shell. This is also a reason why the investigations on the vibration of a rotating ES-FGM conical shell are very limited. This difficulty will be got over below.
4 BOUNDARY CONDITIONS AND SOLUTION OF THE PROBLEM
Assuming that the stiffened FGM truncated conical shell is simply supported at both ends. Thus the boundary conditions are expressed in the following form
The displacement components satisfying accurately the above mentioned geometric boundary conditions and the force boundary conditions in the average sense, may be chosen as
where m is the number of half-waves along a generatrix and n is the number of circumferential full-waves, ω (rad/s) is the natural circular frequency of the rotating ES-FGM conical shell, and Ψ_{1}, Ψ_{2} and Ψ_{3} are un-known constants.
As above noted, it is difficult to use the trial function (28) and Eqs. (15-17) to obtain directly the polynomial equation of frequency. Therefore, a different procedure is proposed as follows. Because x _{0} ≤ x ≤ x _{0} + L; 0 ≤ θ ≤ 2π, so we can carry out following equivalent transformations. Firstly multiplying Eq. (15) by x ^{2} and Eqs. (16, 17) by x ^{3}, then applying Galerkin method for the resulting equations, leads to
Substituting expressions (28) into Eqs. (15-17) then into Eq. (29), after integrating longer and some rearrangements, we obtain
where the coefficients H ^{0} _{ij} , H ^{1} _{ij} are given in Appendix II.
For Eq. (30) to have nontrivial solution, the determinant of the characteristic matrix should be set equal to zero. Developing that determinant and solving resulting equation for frequency ω, yields
Eq. (31) is the sixth order polynomial equation for ω and it is used to analyze the frequency characteristics of rotating ES-FGM conical. Numerical results below will show that this equation will have two roots whose absolute values are smallest and they are real numbers, one positive and the other negative. The positive value corresponds to the forward wave and the negative value corresponds to the backward wave.
5 NUMERICAL RESULTS AND DISCUSSION
5.1 Verification of the Present Method
Before stating the analysis of shell frequency characteristics, the validity of the present study should be ensured. Table 1 compares the frequency parameter results of this paper for unstiffened isotropic truncated conical shell with the frequency parameter given by (^{Hua, 2000})
f | α=30^{0} | α=45^{0} | α=60^{0} | |||
N | Present | Hua [14] | Present | Hua [14] | Present | Hua [14] |
2 | 0.8360 | 0.8420 | 0.7589 | 0.7655 | 0.6322 | 0.6348 |
3 | 0.7365 | 0.7376 | 0.7175 | 0.7212 | 0.6223 | 0.6238 |
4 | 0.6378 | 0.6362 | 0.6725 | 0.6739 | 0.6138 | 0.6145 |
5 | 0.5550 | 0.5528 | 0.6322 | 0.6323 | 0.6106 | 0.6111 |
6 | 0.4962 | 0.4950 | 0.6034 | 0.6035 | 0.6161 | 0.6171 |
7 | 0.4652 | 0.4661 | 0.5908 | 0.5921 | 0.6327 | 0.6350 |
8 | 0.4624 | 0.4660 | 0.5967 | 0.6001 | 0.6618 | 0.6660 |
9 | 0.4854 | 0.4916 | 0.6216 | 0.6273 | 0.7036 | 0.7101 |
Computations have been carried out for a statonary isotropic conical shell with the following data base as:
k = 0, Ω = 0, m = 1, h/R = 0.01, L = 0.25R/sin α, E = 4.8265 × 10^{9} (Pa), ν = 0.3, ρ = 1314 (kg/m^{3}).
Frequency ω is found from Eq. (31).
It can be observed a good agreement is obtained in this comparison.
5.2 ES-FGM Truncated Conical Shells
In the following subsections, the materials used are Alumina with E_{c} = 380 Gpa, ρ _{c} = 3800 kg/m^{3} and Aluminum with E_{m} = 70 Gpa, ρ _{m} = 2702 kg/m^{3} and ν=0.3. The stiffeners are FGM.
5.2.1 Effect of Circumferential Wave Number n
Table 2 and Fig. 2a and 2b show effects of circumferential wave number n on frequency ω in case forward wave (Line(a)) and backward wave (Line(b)) with volume fraction index k = 1; 5 and semi-vertex angle α = 30º; 45º; 60º. Consider stiffeners attached to inside of the shell. The input parameters are taken as h=0.004 m, r/h=30, L/r=2.5, b _{1} =0.002 m, h _{1}=0.004 m, n_{s}=40, b _{2} =0.002 m, h _{2}=0.004 m, n_{r}=40, m=1, Ω=150 rad/s, k=k_{2}=k_{3}=1. As can be seen that at the same rotating velocity Ω when n ≤ 3 the frequency ω decreases with the increase of circumferential wave number n. But the frequency ω increases with the increase of circumferential wave number n when n ≥ 5. We also can be seen that frequency ω has the minimum value at mode (m, n)=(1,4).
ω | α=30^{0} | α=45^{0} | α=60^{0} | |||
---|---|---|---|---|---|---|
n=1 | ^{(a)}0.2582e+5 | -^{(b)}0.2564e+5 | ^{(a)}0.1983e+5 | -^{(b)}0.1974e+5 | ^{(a)}0.1317e+5 | -^{(b)}0.1314e+5 |
2 | 0.1673e+5 | -0.1656e+5 | 0.1400e+5 | -0.1389e+5 | 0.0987e+5 | -0.0982e+5 |
3 | 0.1156e+5 | -0.1143e+5 | 0.1003e+5 | -0.0994e+5 | 0.0742e+5 | -0.0738e+5 |
4 | 0.1044e+5 | -0.1034e+5 | 0.0868e+5 | -0.0861e+5 | 0.0674e+5 | -0.0671e+5 |
5 | 0.1276e+5 | -0.1268e+5 | 0.0989e+5 | -0.0983e+5 | 0.0792e+5 | -0.0789e+5 |
6 | 0.1720e+5 | -0.1714e+5 | 0.1287e+5 | -0.1282e+5 | 0.1039e+5 | -0.1037e+5 |
7 | 0.2299e+5 | -0.2294e+5 | 0.1696e+5 | -0.1693e+5 | 0.1371e+5 | -0.1370e+5 |
8 | 0.2986e+5 | -0.2982e+5 | 0.2191e+5 | -0.2188e+5 | 0.1770e+5 | -0.1769e+5 |
9 | 0.3773e+5 | -0.3769e+5 | 0.2760e+5 | -0.2757e+5 | 0.2229e+5 | -0.2228e+5 |
10 | 0.4656e+5 | -0.4652e+5 | 0.3400e+5 | -0.3398e+5 | 0.2745e+5 | -0.2744e+5 |
Forward wave-Line(a), Backward wave-Line(b).
5.2.2 Effect of Rotating Speed Ω
Table 3 describes effects of rotating speed Ω on the critical frequency. It is clear that the critical frequency ω increases with the increase of Ω. For example, in Table 3, for the forward wave (with α = 60º) when the rotating speed Ω varies the values from 0 to 2000 (rad/s), the value ω_{cr} increases from 0.0670e+5 to 0.0970e+5. This increase is considerable about 30.9 %.
ωcr | α=15^{0} | α=30^{0} | α=45^{0} | α=60^{0} | ||||
---|---|---|---|---|---|---|---|---|
Ω=0 rad/s | 0.1191e+5 (3) | 0.1037 e+5 (4) | 0.0863e+5 (4) | 0.0670e+5 (4) | ||||
Ω=50 rad/s | 0.1194e+5 (3)* | -0.1188e+5 (3) | 0.1039e+5 (4) | -0.1036e+5 (4) | 0.0864e+5 (4) | -0.0862e+5 (4) | 0.0671e+5 (4) | -0.0670e+5 (4) |
Ω=100 rad/s | 0.1197e+5 (3) | -0.1186e+5 (3) | 0.1042e+5 (4) | -0.1035e+5 (4) | 0.0866e+5 (4) | -0.0861e+5 (4) | 0.0672 e+5 (4) | -0.0670e+5 (4) |
Ω=200 rad/s | 0.1203e+5 (3) | -0.1181e+5 (3) | 0.1047e+5 (4) | -0.1033e+5 (4) | 0.0871e+5 (4) | -0.0862e+5 (4) | 0.0677e+5 (4) | -0.0672e+5 (4) |
Ω=500 rad/s | 0.1226e+5 (3) | -0.1171e+5 (3) | 0.1072e+5 (4) | -0.1038e+5 (4) | 0.0896e+5 (4) | -0.0873e+5 (4) | 0.0704e+5 (4) | -0.0693e+5 (4) |
Ω=1000 rad/s | 0.1277e+5 (3) | -0.1167e+5 (3) | 0.1140e+5 (4) | -0.1071e+5 (4) | 0.0969e+5 (4) | -0.0923e+5 (4) | 0.0789e+5 (4) | -0.0766e+5 (4) |
Ω=1500 rad/s | 0.1343e+5 (3) | -0.1177e+5 (3) | 0.1238e+5 (4) | -0.1134e+5 (4) | 0.1075e+5 (4) | -0.1006e+5 (4) | 0.0880e+5 (3) | -0.0839e+5 (3) |
Ω=2000 rad/s | 0.1421e+5 (3) | -0.1201e+5 (3) | 0.1358e+5 (4) | -0.1220e+5 (4) | 0.1206e+5 (4) | -0.1115e+5 (4) | 0.0970e+5 (3) | -0.0916e+5 (3) |
* Circumferential wave number n
h=0.004 m, r/h=30, L/r=2.5, b _{1} =0.002 m, h _{1}=0.004 m, n_{s}=40, b _{2} =0.002 m, h _{2}=0.004 m, n_{r}=40, m=1, k = k _{2} = k _{3}.
Graphically, Figs. 3 and 4 also show that the critical frequency ω increases with the increase of Ω. In addition when the conical shell is a stationary state (for Ω = 0) a standing wave occurs. However, forward and backward waves will appear when the conical shell starts to rotate.
5.2.3 Effect of Semi-Vertex Angle α
Table 4 presents effects of semi-vertex angle α on critical frequency ω_{cr} . The geometrical parameters of shell and stiffener are given by h=0.004 m, r/h=30, L/r=2.5, b _{1} =0.002 m, h _{1}=0.004 m, n_{s}=40, b _{2} =0.002 m, h _{2}=0.004 m, n_{r}=40, m=1, Ω=150 rad/s, k_{2}=k_{3}=k.
ω_{cr} | k=0.5 | k=1 | k=5 | |||
---|---|---|---|---|---|---|
α=10^{0} | 0.1295e+5 (3) | -0.1278e+5 (3) | 0.1215e+5 (3) | -0.1198e+5 (3) | 0.0974e+5 (3) | -0.0957e+5 (3) |
α=15^{0} | 0.1283e+5 (3) | -0.1267e+5 (3) | 0.1200e+5 (3) | -0.1183e+5 (3) | 0.0956e+5 (3) | -0.0940e+5 (3) |
α=20^{0} | 0.1258e+5 (4) | -0.1245e+5 (4) | 0.1192e+5 (3) | -0.1176e+5 (3) | 0.0946e+5 (3) | -0.0930e+5 (3) |
α=30^{0} | 0.1105e+5 (4) | -0.1095e+5 (4) | 0.1044e+5 (4) | -0.1034e+5 (4) | 0.0848e+5 (4) | -0.0837e+5 (4) |
α=45^{0} | 0.0922e+5 (4) | -0.0915e+5 (4) | 0.0868e+5 (4) | -0.0861e+5 (4) | 0.0701e+5 (4) | -0.0694e+5 (4) |
α=60^{0} | 0.0715e+5 (4) | -0.0712e+5 (4) | 0.0674e+5 (4) | -0.0671e+5 (4) | 0.0547e+5 (4) | -0.0544e+5 (4) |
α=75^{0} | 0.0464e+5 (3) | -0.0463e+5 (3) | 0.0434e+5 (3) | -0.0433e+5 (3) | 0.03505e+5 (3) | -0.03494e+5 (3) |
α=80^{0} | 0.0364e+5 (3) | -0.0364e+5 (3) | 0.0342e+5 (3) | -0.0342e+5 (3) | 0.02811e+5 (3) | -0.02807e+5 (3) |
As can be observed that the critical frequency of shell decreases when the semi-vertex angle increases. This decrease is significant. For example, for the forward wave with Ω=150 rad/s, k=1 when the semi-vertex angle varies the values from 10^{0} to 80^{0}, the critical frequency ω_{cr} decreases from 0.1215e +5 to 0.0342e+5.
Graphically, Fig. 5 also shows that critical frequency decreases when α increases.
5.2.4 Effect of Stiffeners
Table 5 describes effects of stiffener on frequency with r/h=100, r=0.08 m, L/r=2, b _{1} =0.002 m, h _{1}=0.004 m, b _{2} =0.002 m, h _{2}=0.004 m, m=1, Ω=150 rad/s, k=k_{2}=k_{3}=1, α=45^{0}. From obtained results as can be seen with the same stiffener numbers, the critical frequency ω _{cr} of orthogonally stiffened shell is the biggest and the critical frequency ω _{cr} of the shell with the stringer is the smallest.
ω×10^{-5} | Stringer | Ring | Orthogonal | |||
---|---|---|---|---|---|---|
ns=80, nr=0 | ns=0, nr=80 | ns=40, nr=40 | ||||
n=1 | 0.2779 | -0.2772 | 0.2253 | -0.2236 | 0.2579 | -0.2564 |
n=2 | 0.2091 | -0.2082 | 0.1278 | -0.1266 | 0.1550 | -0.1539 |
n=3 | 0.1527 | -0.1518 | 0.0915 | -0.0908 | 0.1098 | -0.1091 |
n=4 | 0.1135 | -0.1127 | 0.1074 | -0.1070 | 0.1190 | -0.1185 |
n=5 | 0.0888 | -0.0882 | 0.1565 | -0.1562 | 0.1666 | -0.1663 |
n=6 | 0.0759 | -0.0753 | 0.2239 | -0.2238 | 0.2352 | -0.2350 |
n=7 | 0.0719 | -0.0715 | 0.3055 | -0.3055 | 0.3190 | -0.3189 |
n=8 | 0.0745 | -0.0741 | 0.4002 | -0.4004 | 0.4164 | -0.4165 |
n=9 | 0.0814 | -0.0810 | 0.5076 | -0.5079 | 0.5271 | -0.5273 |
n=10 | 0.0913 | -0.0909 | 0.6274 | -0.6278 | 0.6507 | -0.6510 |
Figs 6 describes effects of stiffener on frequency with two cases r/h=100 and r/h=200. As can be observed the ω - n curves of stiffened FGM rotating conical shell are lower than those of un-stiffened FGM rotating conical shell when the value of n is smaller any value n _{*} (In this example n _{*} = 4) and inverse trend for n > n _{*}.
5.2.5 Effect of Volume Fraction Index k
Table 6 considers effects of index volume k on the critical frequency for a stiffened FGM truncated conical shell when the rotating speed Ω=0 rad/s, 150 rad/s and 400 rad/s.
ω_{cr} | Ω=0 rad/s | Ω=150 rad/s | Ω=400 rad/s | ||
---|---|---|---|---|---|
Ceramic | 0.1185e+5 (4) | 0.1192e+5(4) | -0.1182e+5 (4) | 0.1209e+5 (4) | -0.1182e+5 (4) |
k=0.5 | 0.1098e+5 (4) | 0.1105e+5 (4) | -0.1095e+5 (4) | 0.1123e+5 (4) | -0.1095e+5 (4) |
k=1 | 0.1037e+5 (4) | 0.1044e+5 (4) | -0.1034e+5 (4) | 0.1063e+5 (4) | -0.1035e+5 (4) |
k=5 | 0.0840e+5 (4) | 0.0848e+5 (4) | -0.0837e+5 (4) | 0.0868e+5 (4) | -0.0840e+5 (4) |
k=10 | 0.0760e+5 (4) | 0.0768e+5 (4) | -0.0757e+5 (4) | 0.0790e+5 (4) | -0.0762e+5 (4) |
Metal | 0.0603e+5 (4) | 0.0611e+5 (4) | -0.0601e+5 (4) | 0.0636e+5 (4) | -0.0609e+5 (4) |
h=0.004 m, r/h=30, L/r=2.5, b_{1}=0.002 m, h_{1}=0.004 m, n_{s}=40, b_{2}=0.002 m, h_{2}=0.004 m, n_{r}=40, m=1, α=30^{0}, k_{2}=k_{3}=k.
Fig. 7 plots 4 lines ω _{cr} - k corresponding to L/r=2; 2.5; 3; 4.
Fig. 8 plots 4 lines ω _{cr} - k corresponding to α = 15º; 30º; 45º; 60º.
It is found that the critical frequency ω_{cr} decreases when k increase. This is reasonable because the value increase of k implies the increase of metal constituent in shell. So the stiffness of shell decreases and that leads to the critical frequency ω_{cr} decreases.
5.2.6 Effect of r/h and L/r
Tables 7, 8 and Fig. 9 illustrate effects of the radius-to-thickness ratio r/h and length-to-radius ratio L/r on critical frequency ω_{cr} of stiffened FGM truncated conical shells with the parameters given by r=0.08 m, b_{1}=0.002 m, h_{1}=0.004 m, n_{s}=40, b_{2}=0.002 m, h_{2}=0.004 m, n_{r}=40, m=1, Ω=150 rad/s, k=k_{2}=k_{3}=1, α=30^{0}.
ω_{cr} | L/r=1 | 2 | 2.5 | 3 | 4 | 5 |
---|---|---|---|---|---|---|
r/h=20 | 0.3775e+5 (3) | 0.2166e+5 (3) | 0.1781 (3) | 0.1504e+5 (3) | 0.1135e+5 (3) | 0.0903e+5 (3) |
50 | 0.2632e+5 (3) | 0.1639e+5 (3) | 0.1392 (3) | 0.1207e+5 (3) | 0.0948e+5 (3) | 0.0765e+5 (4) |
100 | 0.2052e+5 (3) | 0.1307e+5 (3) | 0.1130 (3) | 0.0997e+5 (3) | 0.0805e+5 (3) | 0.0673e+5 (3) |
200 | 0.1614e+5 (3) | 0.1029e+5 (3) | 0.0899 (3) | 0.0802e+5 (3) | 0.0661e+5 (3) | 0.0563e+5 (3) |
500 | 0.1224e+5 (3) | 0.0765e+5 (3) | 0.0668 (3) | 0.0597e+5 (3) | 0.0497e+5 (3) | 0.0428e+5 (3) |
ω_{cr} | L/r=1 | 2 | 2.5 | 3 | 4 | 5 |
---|---|---|---|---|---|---|
r/h=20 | -0.3765e+5 (3) | -0.2153e+5 (3) | -0.1768 (3) | -0.1491e+5 (3) | -0.1121e+5 (3) | -0.0889e+5 (3) |
50 | -0.2622e+5 (3) | -0.1627e+5 (3) | -0.1380 (3) | -0.1195e+5 (3) | -0.0935e+5 (3) | -0.0754e+5 (4) |
100 | -0.2041e+5 (3) | -0.1295e+5 (3) | -0.1118 (3) | -0.0984e+5 (3) | -0.0792e+5 (3) | -0.0660e+5 (3) |
200 | -0.1603e+5 (3) | -0.1017e+5 (3) | -0.0887 (3) | -0.0789e+5 (3) | -0.0649e+5 (3) | -0.0550e+5 (3) |
500 | -0.1213e+5 (3) | -0.0754e+5 (3) | -0.0656 (3) | -0.0585e+5 (3) | -0.0484e+5 (3) | -0.0415e+5 (3) |
Figs. 10 and 11 show the effects of r/h ratio and L/r ratio on frequency ω of the shell. It is observed that the critical frequency ω_{cr} as well ω decrease markedly with the increase of those ratios. This decrease is considerable. For example in Table 7, the value ω _{cr} = 0.3775e+5 (with r/h=20, L/r=1) decreases about 1.8 times in comparison with ω _{cr} = 0.2052e+5 (with r/h=100, L/r=1).
5.2.7 Comparison Between Outside FGM Stiffener and Inside FGM Stiffener
Table 9 compares effects of outside FGM stiffener and inside FGM stiffener on the critical frequency with k=0.5; 1; 5 and α=10^{0}; 30^{0}; 45^{0}; 60^{0}; 75^{0} and h=0.004 m, r/h=30, L/r=2.5, b_{1}=0.002 m, h_{1}=0.004 m, n_{s}=40, b_{2}=0.002 m, h_{2}=0.004 m, n_{r}=40, m=1, Ω=150 rad/s, k_{2}=k_{3}=1/k. We can see with the same input parameter the critical frequency ω_{cr} for inside FGM stiffener is bigger than outside FGM stiffener when 10º ≤ α ≤ 45º while 60º ≤ α ≤ 75º the critical frequency ω_{cr} for inside FGM stiffener is smaller than outside FGM stiffener.
ω_{cr} | k=0.5 | k=1 | k=5 | |||
---|---|---|---|---|---|---|
α=10^{0} | ^{a}0.1246e+5 (3) | -0.1228e+5 (3) | 0.1140e+5 (4) | -0.1125e+5 (4) | 0.0903e+5 (3) | -0.0885e+5 (3) |
^{b}0.1295e+5 (3) | -0.1278e+5 (3) | 0.1215e+5 (3) | -0.1198e+5 (3) | 0.0974e+5 (3) | -0.0957e+5 (3) | |
α=30^{0} | ^{a}0.1045e+5 (4) | -0.1033e+5 (4) | 0.0956e+5 (4) | -0.0944e+5 (4) | 0.0761e+5 (4) | -0.0749e+5 (4) |
^{b}0.1105e+5 (4) | -0.1095e+5 (4) | 0.1044e+5 (4) | -0.1034e+5 (4) | 0.0848e+5 (4) | -0.0837e+5 (4) | |
α=45^{0} | ^{a}0.0919e+5 (4) | -0.0911e+5 (4) | 0.0834e+5 (5) | -0.0827e+5 (5) | 0.0671e+5 (4) | -0.0663e+5 (4) |
^{b}0.0922e+5 (4) | -0.0915e+5 (4) | 0.0868e+5 (4) | -0.0861e+5 (4) | 0.0701e+5 (4) | -0.0694e+5 (4) | |
α=60^{0} | ^{a}0.0728e+5 (4) | -0.0724e+5 (4) | 0.0669e+5 (4) | -0.0665e+5 (4) | 0.0534e+5 (4) | -0.0530e+5 (4) |
^{b}0.0715e+5 (4) | -0.0712e+5 (4) | 0.0674e+5 (4) | -0.0671e+5 (4) | 0.0547e+5 (4) | -0.0544e+5 (4) | |
α=75^{0} | ^{a}0.0505e+5 (4) | -0.0504e+5 (3) | 0.0460e+5 (4) | -0.0458e+5 (4) | 0.0371e+5 (3) | -0.0370e+5 (3) |
^{b}0.0464e+5 (3) | -0.0463e+5 (3) | 0.0434e+5 (3) | -0.0433e+5 (3) | 0.03505e+5 (3) | -0.03494e+5 (3) |
a Outside FGM stiffener, b Inside FGM stiffener
Table 10 compares effects of inside FGM stiffener and outside FGM stiffener on the critical frequency with k=0.5; 1; 5; 10 and Ω=0; 150; 400 rad/s. As can be seen that the critical frequency ω_{cr} of an inside FGM stiffener attached shell is bigger than one of outside FGM stiffener attached shell.
ω_{cr} | Ω=0 rad/s | Ω=150 rad/s | Ω=400 rad/s | ||
---|---|---|---|---|---|
k=0.5 | ^{a}0.1038e+5 (4) | 0.1045e+5 (4) | -0.1033e+5 (4) | 0.1064e+5 (4) | -0.1033e+5 (4) |
^{b}0.1098e+5 (4) | 0.1105e+5 (4) | -0.1095e+5 (4) | 0.1123e+5 (4) | -0.1095e+5 (4) | |
k=1 | ^{a}0.0948e+5 (4) | 0.0956e+5 (4) | -0.0944e+5 (4) | 0.0976e+5 (4) | -0.0944e+5 (4) |
^{b}0.1037e+5 (4) | 0.1044e+5 (4) | -0.1034e+5 (4) | 0.1063e+5 (4) | -0.1035e+5 (4) | |
k=5 | ^{a}0.0753e+5 (4) | 0.0761e+5 (4) | -0.0749e+5 (4) | 0.0784e+5 (4) | -0.0752e+5 (4) |
^{b}0.0840e+5 (4) | 0.0848e+5 (4) | -0.0837e+5 (4) | 0.0868e+5 (4) | -0.0840e+5 (4) | |
k=10 | ^{a}0.0701e+5 (4) | 0.0709e+5 (4) | -0.0697e+5 (4) | 0.0733e+5 (4) | -0.0702e+5 (4) |
^{b}0.0760e+5 (4) | 0.0768e+5 (4) | -0.0757e+5 (4) | 0.0790e+5 (4) | -0.0762e+5 (4) |
a Outside FGM stiffener, b Inside FGM stiffener
h=0.004 m, r/h=30, L/r=2.5, b_{1}=0.002 m, h_{1}=0.004 m, n_{s}=40, b_{2}=0.002 m, h_{2}=0.004 m, n_{r}=40, m=1, α=30^{0}, k_{2}=k_{3}=1/k.
5.2.8 Comparison Between FGM Stiffener and Homogeneous Stiffener (Inside Stiffener)
Table 11 compares the critical frequencies of homogeneous stiffener attached shell with those of FGM stiffener attached shell when the volume fraction index k=0.5; 1; 5 and α varies the values from 10^{o} to 60^{o}.
ω_{cr} | k=0.5 | k=1 | k=5 | |||
---|---|---|---|---|---|---|
α=10^{0} | ^{a}0.1173e+5 (4) | -0.1159e+5 (4) | 0.1115e+5 (4) | -0.1101e+5 (4) | 0.0913e+5 (3) | -0.0896e+5 (3) |
^{b}0.1295e+5 (3) | -0.1278e+5 (3) | 0.1215e+5 (3) | -0.1198e+5 (3) | 0.0974e+5 (3) | -0.0957e+5 (3) | |
α=15^{0} | ^{a}0.1100e+5 (4) | -0.1087e+5 (4) | 0.1041e+5 (4) | -0.1027e+5 (4) | 0.0915e+5 (4) | -0.0902e+5 (4) |
^{b}0.1283e+5 (3) | -0.1267e+5 (3) | 0.1200e+5 (3) | -0.1183e+5 (3) | 0.0956e+5 (3) | -0.0940e+5 (3) | |
α=20^{0} | ^{a}0.1055e+5 (4) | -0.1042e+5 (4) | 0.0993e+5 (4) | -0.0980e+5 (4) | 0.0854e+5 (4) | -0.0841e+5 (4) |
^{b}0.1258e+5 (4) | -0.1245e+5 (4) | 0.1192e+5 (3) | -0.1176e+5 (3) | 0.0946e+5 (3) | -0.0930e+5 (3) | |
α=30^{0} | ^{a}0.0992e+5 (4) | -0.0981e+5 (4) | 0.0928e+5 (4) | -0.0917e+5 (4) | 0.0772e+5 (4) | -0.0761e+5 (4) |
^{b}0.1105e+5 (4) | -0.1095e+5 (4) | 0.1044e+5 (4) | -0.1034e+5 (4) | 0.0848e+5 (4) | -0.0837e+5 (4) | |
α=45^{0} | ^{a}0.0839e+5 (5) | -0.0833e+5 (5) | 0.0796e+5 (5) | -0.0790e+5 (5) | 0.0660e+5 (4) | -0.0652e+5 (4) |
^{b}0.0922e+5 (4) | -0.0915e+5 (4) | 0.0868e+5 (4) | -0.0861e+5 (4) | 0.0701e+5 (4) | -0.0694e+5 (4) | |
α=60^{0} | ^{a}0.0673e+5 (4) | -0.0670e+5 (4) | 0.0628e+5 (4) | -0.0625e+5 (4) | 0.0514e+5 (4) | -0.0511e+5 (4) |
^{b}0.0715e+5 (4) | -0.0712e+5 (4) | 0.0674e+5 (4) | -0.0671e+5 (4) | 0.0547e+5 (4) | -0.0544e+5 (4) |
a Homogeneous stiffener, b FGM stiffener
h=0.004 m, r/h=30, L/r=2.5, b_{1}=0.002 m, h_{1}=0.004 m, n_{s}=40, b_{2}=0.002 m, h_{2}=0.004 m, n_{r}=40, m=1, Ω=150 rad/s.
Table 12 also compares the critical frequencies of homogeneous stiffener attached shell with those of FGM stiffener attached shell when the volume fraction index k=0.5; 1; 5 and the rotating speed Ω=0; 150; 400 (rad/s).
ω_{cr} | Ω=0 rad/s | Ω=150 rad/s | Ω=400 rad/s | ||
---|---|---|---|---|---|
k=0.5 | a0.0985e+5 (4) | 0.0992e+5 (4) | -0.0981e+5 (4) | 0.1011e+5 (4) | -0.0982e+5 (4) |
b0.1098e+5 (4) | 0.1105e+5 (4) | -0.1095e+5 (4) | 0.1123e+5 (4) | -0.1095e+5 (4) | |
k=1 | a0.0921e+5 (4) | 0.0928e+5 (4) | -0.0917e+5 (4) | 0.0948e+5 (4) | -0.0919e+5 (4) |
b0.1037e+5 (4) | 0.1044e+5 (4) | -0.1034e+5 (4) | 0.1063e+5 (4) | -0.1035e+5 (4) | |
k=5 | a0.0765 e+5 (4) | 0.0772e+5 (4) | -0.0761e+5 (4) | 0.0794e+5 (4) | -0.0765e+5 (4) |
b0.0840e+5 (4) | 0.0848e+5 (4) | -0.0837e+5 (4) | 0.0868e+5 (4) | -0.0840e+5 (4) |
a Homogeneous stiffener, b FGM stiffener
h=0.004 m, r/h=30, L/r=2.5, b_{1}=0.002 m, h_{1}=0.004 m, n_{s}=40, b_{2}=0.002 m, h_{2}=0.004 m, n_{r}=40, m=1, α=30^{0}.
It is found that the critical frequency corresponding to FGM stiffener attached shell is bigger than one of homogeneous stiffener attached shell in these two cases.
6 CONCLUSIONS
An analytical solution is presented, in this paper, to investigate the free vibration of rotating eccentrically stiffened functionally graded truncated conical. Some new contributions are obtained as follows:
FGM truncated conical shells are reinforced by FGM stringers and rings in which a change of spacing between stringer stiffeners is considered;
A centrifugal force and Coriolis acceleration are taken into account.
The sixth order polynomial equation for ω is obtained analytically and it is used to analyze the frequency characteristics of rotating ES-FGM conical shells.
Effects of stiffener, geometrics parameters, cone angle, vibration modes and rotating speed on frequency characteristics of the shell forward and backward wave are discussed in detail.