Jtam-A4.dvi JOURNAL OF THEORETICAL AND APPLIED MECHANICS 54, 4, pp. 1205-1217, Warsaw 2016 DOI: 10.15632/jtam-pl.54.4.1205 APPLICATION OF THE DIFFERENTIAL TRANSFORM METHOD TO THE FREE VIBRATION ANALYSIS OF FUNCTIONALLY GRADED TIMOSHENKO BEAMS Özge Özdemir Istanbul Technical University, Faculty of Aeronautics and Astronautics, Maslak, Istanbul, Turkey e-mail: ozdemirozg@itu.edu.tr In this study, free vibration characteristics of a functionally graded Timoshenko beam that undergoes flapwise bending vibration is analysed. The energy expressions are derived by introducing several explanotary figures and tables. Applying Hamilton’s principle to the energy expressions, governing differential equations of motion and boundary conditions are obtained. In the solution part, the equations of motion, including the parameters for rotary inertia, shear deformation, power law index parameter and slenderness ratio are solved using an efficientmathematical technique, called the differential transformmethod (DTM). Natural frequencies are calculated and effects of several parameters are investigated. Keywords: differential transformmethod, functionally graded beam, Timoshenko beam 1. Introduction The concept of Functionally GradedMaterials (FGMs) was originated from a group ofmaterial scientists in Japan asmeans of preparing thermal barriermaterials (Loy et al., 1999). FGMs are special composites that have continuous variation ofmaterial properties in oneormoredirections to provide designers with the ability to distribute strength and stiffness in a desired manner to get suitable structures for specific purposes in engineering and scientific fields such as design of aircraft and space vehicle structures, electronic and biomedical installations, automobile sector, defence industries, nuclear reactors, electronics, transportation sector, etc. As a consequence, it is important to understand static and dynamic behavior of FGMs, so it has been an area of intense research in the recent years. Especially, functionally graded beam (FGB) structures have become a fertile area of research since beam structures have been widely used in aeronautical, astronautical, civil, mechanical and other kinds of installations. Several research papers provide a good introduction and further references on the subject (Alshorbagy et al., 2011; Chakraborty et al., 2003; Giunta et al., 2011; Huang and Li, 2010; Kapuria et al., 2008; Lai et al., 2012; Li, 2008; Loja et al., 2012; Lu and Chen, 2005; Thai and Vo, 2012; Wattanasakulpong et al., 2012; Zhong and Yu, 2007). Due to the increasing application trend of FGMs, several beam theories have been developed to examine the response of FGBs. The Classical Beam Theory (CBT), i.e. Euler Bernoulli BeamTheory, is the simplest theory that can be applied to slender FGBs. The first order shear deformation theory (FSDT), i.e. Timoshenko Beam Theory, is used for the case of either short beams or high frequency applications to overcome the limitations of the CBT by accounting for the tranverse shear deformation effect. Bhimaraddi and Chandrashekhara (1991) derived laminated composite beam equations of motion using the first-order shear deformation plate theory (FSDPT).Dadfarnia (1997) developedanewbeamtheory for laminated composite beams using the assumption that the lateral stresses and all derivatives with respect to the lateral coordinate in the plate equations of motion are ignored. 1206 Ö. Özdemir In this study, which is an extension of the author’s previous works (Kaya and Ozdemir Ozgumus, 2007; Kaya and Ozdemir Ozgumus, 2010; Ozdemir Ozgumus and Kaya, 2013), free vibration analysis of a functionally graded Timoshenko beam that undergoes flapwise bending vibrations is performed.At the beginning of the study, expressions for both kinetic andpotential energies are derived in a detailed way by using explanatory tables and figures. In the next step, governing differential equations of motion are obtained applying Hamilton’s principle. In the solution part, the equations of motion, including the parameters for rotary inertia, shear deformation, power law index parameter and slenderness ratio are solved using an efficient mathematical technique, called the differential transform method (DTM). Natural frequencies are calculated and effects of the parameters, mentioned above, are investigated. The calculated results are compared with the ones in open literature. Consequently, it is observed that there is a good agreement between the results which proves the correctness and accuracy of the DTM. 2. Beam model The governing differential equations of motion are derived for the free vibration analysis of a functionally graded Timoshenko beammodel with a right-handed Cartesian coordinate system which is represented by Fig. 1. Fig. 1. Functionally graded beammodel and the coordinate system Here a uniform, functionally graded Timoshenko beam of length L, height h and width b which has the cantilever boundary condition at point O is shown. The xyz-axes constitute a global orthogonal coordinate systemwith the origin at the root of thebeam.Thex-axis coincides with theneutral axis of thebeamin theundeflectedposition, they-axis lies in thewidthdirection and the z-axis lies in the depth direction. 3. Formulation 3.1. Functionally graded beam formulation Material properties of the beam, i.e. modulus of elasticity E, shear modulus G, Poisson’s ratio ν and material density ρ are assumed to vary continuously in the thickness direction z as a function of the volume fraction, and the properties of the constituent materials according to a simple power law. According to the rule of mixture, the effective material property P(z) can be expressed as follows P(z) =PtVt+PbVb (3.1) where Pt and Pb are the material properties at the top and bottom surfaces of the beam while Vt and Vb are the corresponding volume fractions. The relation between the volume fractions is given by Vt+Vb =1 (3.2) Application of the differential transform method... 1207 The volume fraction of the top constituent of the beam Vt is assumed to be given by Vt = (z h + 1 2 )k k­ 0 (3.3) where k is a non-negative power law index parameter that dictates thematerial variation profile through the beam thickness. Considering Eqs. (3.1)-(3.3), the effective material property can be rewritten as follows P(z) = (Pt−Pb) (z h + 1 2 )k +Pb (3.4) It is evident from Eq.(4) that when z = h/2, E = Et, ν = νt, G = Gt, ρ = ρt and when z=−h/2, E =Eb, ν = νb,G=Gb and ρ= ρb. 3.2. Displacement field and strain field The cross-sectional and the longitudinal views of a Timoshenko beam that undergoes exten- sionandflapwisebendingdeflectionsaregiven inFigs. 2aand2b, respectively.Here, the reference point is chosen, and is represented by P0 before deformation and by P after deformation. Fig. 2. (a) Cross-sectional view, (b) longitudinal view of the Timoshenko beam Here, η is the offset of the reference point from the z-axis, ξ is the offset of the reference point from the middle plane, x is the offset of the reference point from the z-axis, u0 is the elongation, w is the flapwise bending displacement, ϕ is the rotation due to bending and γ is the shear angle. Considering Figs. 2a and 2b, the coordinates of the reference point are obtained as follows: — before deflection (coordinates of P0) x0 =x y0 = η z0 = ξ (3.5) — after deflection (coordinates of P) x1 =x+u0+ ξϕ y1 = η z1 =w+ ξ (3.6) The position vectors of the reference point are represented by r0 and r1 before and after deflec- tion, respectively. Therefore, dr0 and dr1 can be written as follows dr0 = dxi+dηj+dξk dr1 = [(1+u ′ 0+ ξϕ ′)]dxi+dηj+(w′dx+dξ)k (3.7) where (·)′ denotes differentiation with respect to the spanwise coordinate x. 1208 Ö. Özdemir The classical strain tensor εij may be obtained by using the following equilibrium equation given by Eringen (1980) dr1 ·dr1−dr0 ·dr0 =2[dx dη dξ][εij]   dx dη dξ   (3.8) where [εij] =   εxx εxη εxξ εηx εηη εηξ εξx εξη εξξ   (3.9) Substituting Eqs. (3.7) into Eq. (3.8), the components of the strain tensor εij are obtained as follows εxx =u ′ 0+ (u′0) 2 2 + (w′)2 2 +u′0ϕ ′ξ+ϕ′ξ+ (ϕ′)2 2 ξ2 γxη =0 γxξ =(w ′+ϕ)+ϕϕ′ξ−u′0ϕ (3.10) where εxx, γxη and γxξ are the axial strain and the shear strains, respectively. In this work, only εxx, γxη and γxξ are used in the calculations because, as noted byHodges and Dowell (1974) for long slender beams, the axial strain εxx is dominant over the transverse normal strains εηη and εξξ. Moreover, the shear strain γηξ is by two orders smaller than the other shear strains γxξ and γxη. Therefore, εηη, εξξ and γηξ are neglected. In order to obtain simpler expressions for the strain components given byEqs. (3.10), higher order terms can be neglected, so an order of magnitude analysis is performed by using the ordering scheme taken fromHodges and Dowell (1974) and introduced in Table 1. Table 1.Ordering scheme for the Timoshenko beammodel Term Order w′ O(ε) ϕ O(ε) w′+ϕ O(ε2) u′0 O(ε 2) ϕ′ (ε2) Hodges andDowell (1974) used the formulation for anEuler-Bernoulli beam, so in this study their formulation is modified for the Timoshenko beam, and a new expression w′ +ϕ=O(ε2) is added to their ordering scheme as a contribution to literature. Considering Table 1, Eqs. (3.10) are simplified as follows εxx =u ′ 0+ (u′0) 2 2 + (w′)2 2 +ϕ′ξ γxη =0 γxξ =w ′+ϕ (3.11) 3.3. Potential energy The expression for potential energy is given by U = 1 2 l∫ 0 ∫ A (σxxεxx+ τxξγxξ) dAdx= b 2 l∫ 0 h/2∫ −h/2 (σxxεxx+ τxξγxξ) dξdx (3.12) Application of the differential transform method... 1209 The axial forceN, the bendingmomentM and the shear forceQ that act on a laminate at the midplane are expressed as follows (Kollar and Springer, 2003) N = b h/2∫ −h/2 σ dz M = b h/2∫ −h/2 zσ dz Q= b h/2∫ −h/2 τ dz (3.13) Substituting Eqs. (3.11) into Eq. (3.12) and considering Eqs. (3.13), the following expression is obtained U = 1 2 l∫ 0 { Nx [ u′0+ (w′)2 2 ] +Mxϕ ′+Qz(w′+ϕ) } dx (3.14) where Nx =A11u ′ 0+B11ϕ ′ Mx =B11u ′ 0+D11ϕ ′ Q=A55γxξ (3.15) Here, the stiffness coefficients are obtained as follows [A11 B11 D11] = ∫ A E(z)[1 z z2] dA A55 =K ∫ A G(z) dA (3.16) whereK is defined as the shear correction factor that takes the value ofK =5/6 for rectangular cross sections. Substituting Eqs. (3.15) into Eq. (3.14) gives U = 1 2 l∫ 0 [A11(u ′ 0) 2+2B11u ′ 0ϕ ′+D11(ϕ ′)2+A55(w ′+ϕ)] dx (3.17) Referring Eq. (3.17), variation of the potential energy is obtained as follows δU = l∫ 0 [(A11u ′ 0+B11ϕ ′)δu′0+(B11u ′ 0+D11ϕ ′)δϕ′+A55(w ′+ϕ)(δw′ + δϕ)] dx (3.18) 3.4. Kinetic energy The position vector of the point P shown in Fig. 2 is given by r=(x+u0+ ξϕ)i+wk (3.19) Considering Eq. (3.19), the velocity vector of this point is obtained as follows V= ∂r ∂t =(u̇0+ξϕ̇)i+ ẇk (3.20) Hence, the velocity components are Vx = u̇0+ ξϕ̇ Vy =0 Vz = ẇ (3.21) The kinetic energy expression is given by T = 1 2 l∫ 0 ∫ A ρ(z)(V 2x +V 2 y +V 2 z ) dAdx= b 2 l∫ 0 h/2∫ −h/2 ρ(z)(V 2x +V 2 y +V 2 z ) dξdx (3.22) where ρ(z) is the effective material density. 1210 Ö. Özdemir Substituting the velocity components into Eq. (3.22) and taking the variaiton of kinetic energy gives δT = l∫ 0 [I1(u̇0δu̇0+ ẇδẇ)+I2(u̇0δϕ̇+ ϕ̇δu̇0)+ I3ϕ̇δϕ̇] dx (3.23) where I1, I2 and I3 are the inertial characteristics of the beam given by [I1 I2 I3] = ∫ A ρ(z)[1 z z2] dA (3.24) 3.5. Equations of motion and the boundary conditions Hamilton’s principle is expressed as follows t2∫ t1 δ(U −T) dt=0 (3.25) Substituting Eqs. (3.18) (3.23) into Eq. (3.25) gives the equations of motion and the boundary conditions as follows: — equations of motion A11u ′′ 0 +B11ϕ ′′ = I1ü0+ I2ϕ̈ A55(w ′′+ ϕ̇′)= I1ẅ D11ϕ ′′+B11u ′′ 0 −A55(w ′+ϕ)= I2ü0+ I3ϕ̈ (3.26) — boundary conditions x=0 u0(0, t)=w(0, t) =ϕ(0, t) = 0 x=L A11u ′ 0(L,t)+B11ϕ ′(L,t)= 0 A55[w ′(L,t)+ϕ(L,t)] = 0 D11ϕ(L,t)+B11u ′ 0(L,t)−A55(w ′+ϕ)= 0 (3.27) In order to investigate free vibration of the beam model considered in this study, a sinusoidal variation of u0, w and ϕ with a circular natural frequency ω is assumed, and the functions are approximated as u0(x,t)=u(x)e iωt w(x,t)=w(x)eiωt ϕ(x,t) =ϕ(x)eiωt (3.28) SubstitutingEqs. (3.28) into the equations ofmotion, i.e. Eqs. (3.26), and into the boundary conditions, i.e. Eqs.(3.27), the following dimensionless equations are obtained as follows: — equations of motion γ2ũ∗∗+α2ϕ̃∗∗+λ2(ũ+µ2ϕ̃)= 0 w̃∗∗+ ϕ̃ τ2 +λ2w̃=0 τ2(α2ũ∗∗+ ϕ̃∗∗+µ2λ2ũ)+(r2τ2λ2−1)ϕ̃− w̃∗ =0 (3.29) — boundary conditions x=0 ũ(0, t) = w̃(0, t)= ϕ̃(0, t)= 0 x=L γ2ũ∗+α2ϕ̃∗(L,t)= 0 1 τ2 [w̃∗(L,t)+ ϕ̃(L,t)] = 0 α2ũ∗(L,t)+ ϕ̃∗(L,t)= 0 (3.30) Application of the differential transform method... 1211 Here, the dimensionless parameters are defined as w̃= w L ũ= u L ϕ̃=ϕ γ2 = A11L 2 D11 τ2 = D11 A55L 2 λ2 = I1L 4ω2 D11 µ2 = I2 I1L r2 = I3 I1L2 α2 = B11L D11 (3.31) 4. Differential Transform Method The Differential Transform Method (DTM) is a transformation technique based on the Taylor series expansion and is a useful tool to obtain analytical solutions of differential equations. In this method, certain transformation rules are applied, and the governing differential equations and the boundary conditions of the system are transformed into a set of algebraic equations in terms of the differential transforms of the original functions, and the solution of these algebraic equations gives the desired solution of the problem. Consider a function f(x) which is analytical in a domain D and let x = x0 represent any point inD. The function f(x) is then represented by apower serieswhose center is located atx0. The differential transform of the function f(x) is given by F[k] = 1 k! (dkf(x) dxk ) x=x0 (4.1) where f(x) is the original function and F[k] is the transformed function. The inverse transfor- mation is defined as f(x)= ∞∑ k=0 (x−x0) kF[k] (4.2) Combining Eq. (4.1) and Eq. (4.2), we get f(x)= ∞∑ k=0 (x−x0) k k! (dkf(x) dxk ) x=x0 (4.3) Considering Eq. (4.3), it is noticed that the concept of differential transform is derived from the Taylor series expansion. However, the method does not evaluate the derivatives symbolically. In actual applications, the function f(x) is expressed by a finite series and Eq. (4.3) can be written as follows f(x)= m∑ k=0 (x−x0) k k! (dkf(x) dxk ) x=x0 (4.4) which means that the rest of the series f(x)= ∞∑ k=m+1 (x−x0) k k! (dkf(x) dxk ) x=x0 (4.5) is negligibly small. Here, the value of m depends on the convergence of natural frequencies. Theorems that are frequentlyused in the transformationprocedureare introduced inTable 2, and theorems that are used for boundary conditions are introduced in Table 3. After applying DTM to Eqs. (3.29) and (3.30), the transformed equations of motion and boundary conditions are obtained as follows: 1212 Ö. Özdemir Table 2.DTM theorems used for equations of motion Original function Transformed function f(x)= g(x)±h(x) F[k] =G[k]±H[k] f(x)=λg(x) F[k] =λG[k] f(x)= g(x)h(x) F[k] = ∑k l=0G[k− l]H[l] f(x)= dng(x) dxn F[k] = (k+n)! k! G[k+n] f(x)=xn F[k] = δ(k−n)= { 0 if k 6=n 1 if k=n Table 3.DTM theorems used for boundary conditions x=0 x=1 Original B.C. Transformed B.C. Original B.C. Transformed B.C. df(0) dx =0 F(0)= 0 f(1)= 0 ∑ ∞ k=0F(k) = 0 df dx (0)= 0 F(1)= 0 df dx (1)= 0 ∑ ∞ k=0kF(k) = 0 d2f dx2 (0)= 0 F(2)= 0 d2f dx2 (1)= 0 ∑ ∞ k=0k(k−1)F(k) = 0 d3f dx3 (0)= 0 F(3)= 0 d3f dx3 (1)= 0 ∑ ∞ k=0(k−1)(k−2)kF(k) = 0 — equations of motion γ2(k+1)(k+2)U[k+2]+α2(k+1)(k+2)ϕ[k+2]+λ2(U[k]+µ2ϕ[k]) = 0 1 τ2 (k+1)(k+2)W [k+2]+λ2W [k]+ 1 τ2 (k+1)ϕ[k+1]=0 α2(k+1)(k+2)U[k+2]+(k+1)(k+2)ϕ[k+2]+λ2µ2U[k]+ ( r2λ2− 1 τ2 ) ϕ[k] − 1 τ2 (k+1)W [k+1]=0 (4.6) — boundary conditions x=0 U[k] =W [k] =ϕ[k] = 0 x=L γ2 ∞∑ k=0 kU[k]+α2 ∞∑ k=0 kϕ[k] = 0 1 τ2 ( ∞∑ k=0 (kW [k]+ϕ[k]) ) =0 α2 ∞∑ k=0 kU[k]+ ∞∑ k=0 kϕ[k] = 0 (4.7) 5. Results and discussions In the numerical analysis, two cases are studied. In the first case, natural frequencies of a pure aluminum Timoshenko beam with simply-simply supported (SS) end conditions and, in the second case, a fuctionally gradedTimoshenkobeamwith clamped free (CF)boundaryconditions are calculated. Effects of the slenderness ratioL/h and the power law index parameter k on the Application of the differential transform method... 1213 natural frequencies are investigated. The results are presented in related tables. In order to validate the calculated results, comparisons with the studies in open literature are made and a very good agreement between the results is observed, which proves the correctness and accuracy of the Differential Transform Method. It is believed that the tabulated results can be used as references by other researchers to validate their results. Case 1. Pure aluminum simply supported beam Table 4.Material properties of the aluminumTimoshenko beam Property Aluminum (Al) Elasticity modulusE 70GPa Material density ρ 2700kg/m3 Poisson’s ratio ν 0.23 Variation of the first five natural frequencies of the S-S pure aluminum Timoshenko beam with respect to the slenderness ratio L/h is given in Table 5. When the calculated results are compared with the ones given by Sina et al. (2009), a very good agreement between the results is observed. Table 5.Dimensionless natural frequencies of the pure aluminumTimoshenko beam Frequency Slenderness ratioL/h λ= ωL 2 h √ ρm Em 10 20 30 40 50 100 Fundamental 2.87896 2.91515 2.92204 2.92447 2.92559 2.92709 Sina et al. (2009) 2.879 – 2.922 – – 2.927 2nd NF 10.9963 11.5159 11.6224 11.6606 11.6784 11.7024 3rd NF 23.1528 25.3995 25.9107 26.0988 26.1876 26.3078 4th NF 32.2814 43.9854 45.4901 46.0634 46.3380 46.7137 5th NF 38.0919 64.5627 69.9845 71.3222 71.9741 72.8788 Case 2. FG Timoshenko beam The FG beam is made of aluminum (Al) at the top and alumina (Al2O3) at the bottom. The effective beam properties change through the beam thickness according to the power law. Thematerial properties of the FG beam are displayed in Table 6. Table 6.Material properties of the FG beam Property Aluminum (Al) Alumina (Al2O3) Elasticity modulusE 70GPa 380GPa Material density ρ 2702kg/m3 3960kg/m3 Poisson’s ratio ν 0.3 0.3 Variation of the fundamental natural frequency of the C-F functionally graded Timoshenko beamaccording to thepower law exponent forL/h=20 is given inTable 7.When the calculated results are compared with the ones given by Şimsek (2010), a very good agreement between the results is observed. 1214 Ö. Özdemir Table 7.Dimensionless fundamental frequencies of the C-F FGTimoshenko beam Frequency Power law exponent k λ= ωL 2 h √ ρm Em 0 0.2 0.5 1 2 5 10 Full metal Fundamental 1.94955 1.81407 1.66026 1.50103 1.36966 1.30373 1.26493 1.01297 Şimsek (2010) 1.94957 1.81456 1.66044 1.50104 1.36968 1.30375 1.26495 1.01297 InTable 8, variation of the dimensionless natural frequencies of the C-F functionally graded Timoshenko beam with respect to the power law exponent k and the slenderness ratio L/h is presented. Table 8. Variation of the dimensionless natural frequencies of the C-F functionally graded Timoshenko beamwith respect to the power law exponent k and the slenderness ratioL/h k L/h 0 0.2 0.5 1 2 5 10 Full metal 3 1.80329 1.6829 1.54468 1.39885 1.27348 1.19956 1.15684 0.936972 8.21514 7.72196 7.12189 6.44425 5.77913 5.22620 4.96530 4.26852 9.06941 8.675 8.23136 7.7081 7.05971 6.19459 5.64279 4.71239 17.9802 16.9703 15.7499 14.3438 12.9094 11.6647 11.0314 9.34237 26.4033 25.1248 23.5417 21.6204 19.4397 17.1022 15.8590 – 4 1.86385 1.73735 1.59278 1.44141 1.31344 1.24242 1.20111 0.96844 9.42868 8.8443 8.15412 7.39468 6.68493 6.15895 5.8858 4.89906 12.0925 11.5548 10.931 10.1889 9.27327 8.07187 7.35535 6.28319 21.5877 20.3297 18.8307 17.1413 15.4804 14.1093 13.3796 11.2168 34.5928 32.7263 30.451 27.8043 25.0665 22.6209 21.4285 18.0754 5 1.89441 1.76476 1.61692 1.46276 1.33353 1.26419 1.2237 0.98432 10.2025 9.55154 8.79239 7.97167 7.23018 6.72424 6.44766 5.30114 15.1157 14.4404 13.6505 12.7061 11.5406 10.0258 9.14605 7.85398 24.2839 22.8225 21.0981 19.1875 17.3683 15.9509 15.1753 12.6177 40.3144 38.0031 35.2454 32.1226 29.0171 26.3775 24.9471 20.9484 10 1.93806 1.80382 1.65126 1.49308 1.36215 1.29547 1.25629 1.007 11.6155 10.8294 9.92996 8.98688 8.18692 7.73783 7.4778 6.03531 30.2314 28.5306 26.2115 23.7423 21.5708 19.8231 18.204 15.708 30.5505 28.8901 27.3023 25.3998 23.0609 20.4233 19.5414 15.8738 55.4176 51.8978 47.8058 43.39 39.4041 36.6649 35.1095 28.7945 15 1.94655 1.81139 1.65791 1.49895 1.3677 1.30157 1.26267 1.01141 11.9506 11.1299 10.1949 9.22147 8.40836 7.97825 7.72685 6.20943 32.4399 30.247 27.7372 25.1042 22.8681 21.6068 20.8769 16.8555 45.347 43.3142 40.921 38.0455 34.4964 29.9319 27.3397 23.5619 60.9894 56.9502 52.3063 47.394 43.1404 40.553 39.057 31.6896 20 1.94955 1.81407 1.66026 1.50103 1.36966 1.30373 1.26493 1.01297 12.0753 11.2415 10.293 9.30821 8.49032 8.06791 7.8202 6.27423 33.2016 30.9307 28.3406 25.6387 23.3723 22.1529 21.4418 17.2513 60.4627 57.7447 54.1446 49.0496 44.62 39.8473 36.4316 31.4159 63.4443 59.1681 54.6767 50.7988 46.1299 42.3338 40.8514 32.9651 In Fig. 3, convergence of the first five natural frequencies with respect to the number of terms N used in DTM application is shown, where L/h = 5 and k = 0.5. To evaluate up to the fifth natural frequency to five-digit precision, it has been necessary to take 45 terms. Application of the differential transform method... 1215 Additionally, it is seen that higher modes appear when more terms are taken into account in DTM application. Thus, depending on the order of the required modes, one must try a few values for the term number at the beginning of the calculations in order to find the adequate number of terms. For instance, onlyN =50 is enough for the results given in Tables 5, 7 and 8. Fig. 3. Convergence of the first five natural frequencies with respect to the number of termsN 6. Conclusion In this study, formulation of a functionally graded Timoshenko beam that undergoes flapwise bendingvibration is derived by introducing several explanotary figures and tables.ApplyingHa- milton’s principle to the obtained energy expressions, governing differential equations of motion and the boundaryconditions are derived. In the solution part, the equations ofmotion, including the parameters for rotary inertia, shear deformation, power law indexparameter and slenderness ratio are solved using an efficient mathematical technique, called the differential transformme- thod (DTM). Natural frequencies are calculated and effects of the abovementioned parameters are investigated. Considering the calculated results,the following conclusions are reached: • As the slenderness ratioL/h increases, the natural frequencies increase; • The effect of the slenderness ratio on the frequencies is negligible forlong FG beams (i.e., L/h­ 20); • The natural frequencies decrease as the value of the power-law exponent k increases. 1216 Ö. Özdemir 7. Future work According to the author’s knowledge, the Differential Transform Method has not been applied to functionally graded Timoshenko beams in literature before. Therefore, this gap is aimed to be fulfilled in this paper. However, in this study, a functionally graded Timoshenko beam with a power-law gradient is considered and the efficiency of DTM has not been examined for other gradients such as exponent gradient (Tang et al., 2014; Hao andWei, 2016; Li et al., 2013;Wang et al., 2016). The examination of the DTMefficiency for other gradient types can be considered as a challenging future work. References 1. Alshorbagy A.E., Eltaher M.A., Mahmoud F.F., 2011, Free vibration characteristics of a functionally graded beam by finite element method,Applied Mathematical Modelling, 35, 412-425 2. Bhimaraddi A., Chandrashekhara K., 1991, Some observation on themodeling of laminated composite beams with general lay-ups,Composite Structures, 19, 371-380 3. Chakraborty A., Gopalakrishnan S., Reddy J.N., 2003, A new beam finite element for the analysis of functionally graded materials, International Journal of Mechanical Sciences, 45, 519-539 4. Dadfarnia M., 1997, Nonlinear forced vibration of laminated beam with arbitrary lamination, M.Sc. Thesis, Sharif University of Technology 5. Deng H.D., Wei C., 2016, Dynamic characteristics analysis of bi-directional functionally graded Timoshenko beams,Composite Structures, in press 6. EringenA.C., 1980,Mechanics of Continua,RobertE.KriegerPublishingCompany,Huntington, NewYork 7. Giunta G., Crisafulli D., Belouettar S., Carrera E., 2011, Hierarchical theories for the free vibration analysis of functionally graded beams,Composite Structures, 94, 68-74 8. Hodges D.H., Dowell E.H., 1974, Nonlinear equations of motion for the elastic bending and torsion of twisted nonuniform rotor blades,NASA Technical Report, NASA TND-7818 9. HuangY., Li X.F., 2010,A new approach for free vibration of axially functionally graded beams with non-uniform cross-section, Journal of Sound and Vibration, 329, 2291-2303 10. Kapuria S., Bhattacharyya M., Kumar A.N., 2008, Bending and free vibration response of layered functionally graded beams: a theoretical model and its experimental validation,Composite Structures, 82, 390-402 11. KayaM.O.,OzdemirOzgumusO., 2007,Flexural-torsional-coupledvibrationanalysis of axially loaded closed-sectioncompositeTimoshenkobeambyusingDTM, Journal of Sound andVibration, 306, 495-506 12. Kaya M.O., Ozdemir Ozgumus O., 2010, Energy expressions and free vibration analysis of a rotating uniform timoshenko beam featuring bending-torsion coupling, Journal of Vibration and Control, 16, 6, 915-934 13. Kollar L.R., Springer G.S., 2003,Mechanics of Composite Structures, Cambridge University Press, United Kingdom 14. Lai S.K., Harrington J., Xiang Y., Chow K.W., 2012, Accurate analytical perturbation approach for large amplitude vibration of functionally graded beams, International Journal of Non-Linear Mechanics, 47, 473-480 15. Li X.F., 2008, A unified approach for analyzing static and dynamic behaviors of functionally graded Timoshenko and Euler-Bernoulli beams, Journal of Sound and Vibration, 318, 1210-1229 Application of the differential transform method... 1217 16. LiX.F., KangY.A.,Wu J.X., 2013,Exact frequency equation of free vibration of exponentially funtionally graded beams,Applied Acoustics, 74, 3, 413-420 17. Loja M.A.R., Barbosa J.I.,Mota Soares C.M., 2012,A study on themodelling of sandwich functionally graded particulate composite,Composite Structures, 94, 2209-2217 18. Loy C.T., Lam K.Y., Reddy J.N., 1999, Vibration of functionally graded cylinderical shells, International Journal of Mechanical Science, 41, 309-324 19. LuC.F., ChenW.Q., 2005, Free vibration of orthotropic functionally graded beamswith various end conditions, Structural Engineering and Mechanics, 20, 465-476 20. Ozdemir Ozgumus O., Kaya M.O., 2013, Energy expressions and free vibration analysis of a rotatingTimoshenko beam featuring bending-bending-torsion coupling,Archive of AppliedMecha- nics, 83, 97-108 21. SinaS.A.,NavaziH.M.,HaddadpourH., 2009,Ananalyticalmethod for freevibrationanalysis of functionally graded beams,Materials and Design, 30, 741-747 22. Şimsek M., 2010, Fundamental frequency analysis of functionally graded beams by using different higher-order beam theories,Nuclear Engineering and Design, 240, 697-705 23. Tang A.Y., Wu J.X., Li X.F., Lee K.Y., 2014, Exact frequency equations of free vibration of exponentially non-uniform functionally graded Timoshenko beams, International Journal of Mechanical Sciences, 89, 1-11 24. Thai H.T.,VoT.P., 2012,Bending and free vibration of functionally gradedbeams using various higher-order shear deformation beam theories, International Journal of Mechanical Sciences, 62, 57-66 25. Wang Z., Wang X., Xu G., Cheng S., Zeng T., 2016, Free vibration of two directional func- tionally graded beams,Composite Structures, 135, 191-198 26. Wattanasakulpong N., Prusty B.G., Kelly D.W., Hoffman M., 2012, Free vibration analysis of layered functionally graded beams with experimental validation,Materials and Design, 36, 182-190 27. Zhong Z., YuT., 2007, Analytical solution of a cantilever functionally graded beam,Composites Science and Technology, 67, 481-488 Manuscript received November 29, 2015; accepted for print February 18, 2016