An exact element method

This paper presents a new element based method through a treatment of a one dimensional model problem. The case solved in this work is of a rod with variable cross-sectional area, supported along its length by variable stiffness elastic foundation, and loaded with variable axial load. The variations of all the quantities are taken as general polynomials. Using this method the exact terms in the stiffness matrix are found using the solution for the differential equation (up to any desired accuracy). The exact solution is obtained using one element for each segment with continuously varying properties, and the displacements and stresses are exact all along the rod.     

Exact solutions of axial vibration problems of elastic bars

While exact solutions for linear static analysis of most frame structures can be obtained by the finite element method, it is very difficult to obtain exact solutions for free vibration and harmonic analyses for non‐trivial cases. This paper presents a study on new finite element formulation and algorithms for exact solutions of undamped axial vibration problems of elastic bars. Appropriate shape functions are constructed by using the homogeneous governing equations, and based on the new shape functions, a novel element is formulated. An iterative procedure is proposed for determining both the exact natural frequency values and the corresponding vibration mode shapes. Exact solutions can also be obtained for undamped harmonic response analyses by using the new element, as its stiffness and mass matrices are exact for a specified frequency. Illustrative examples are presented to demonstrate the effectiveness of the proposed element and algorithm. Copyright © 2007 John Wiley & Sons, Ltd.     

Asymptotically approximate model equations for nonlinear dispersive waves in incompressible elastic rods

Summary In literature, nonlinear waves in elastic rods have been studied by many authors. Usually, the Navier-Bernoulli hypothesis (the assumption that plane cross-sections remain planar and normal to the rod axis) is used. Intuitively, one would expect that this would be a good approximation when one is mainly interested in longitudinal waves. However, there are no rigorous theoretical justifications available. Also, a defect of this assumption is that comparing with the exact three-dimensional theory the boundary conditions on the lateral surface can never be satisfied. Recently, three papers have been published to overcome this defect, but they contain some algebraic errors (which implies that the approach adopted there cannot be used to overcome this defect). So, this problem remains open. In this paper, we present our recent research results for this problem, and we have managed to establish asymptotically valid one-dimensional rod equations which are consistent with the lateral boundary conditions. Further, their dispersion relation can match with that of the exact three-dimensional field equations to any asymptotic order in the long-wave limit. For solitary waves in the far field, we derive to the leading order the KdV equation. Comparing its solitary-wave solution with that of the KdV equation obtained through the Navier-Bernoulli hypothesis, we find that the difference is very small. This provides some evidence to the validity of the assumption that plane cross sections remain planar and normal to the rod axis.     

A p‐hierarchical adaptive procedure for the scaled boundary finite element method

This paper describes a p‐hierarchical adaptive procedure based on minimizing the classical energy norm for the scaled boundary finite element method. The reference solution, which is the solution of the fine mesh formed by uniformly refining the current mesh element‐wise one order higher, is used to represent the unknown exact solution. The optimum mesh is assumed to be obtained when each element contributes equally to the global error. The refinement criteria and the energy norm‐based error estimator are described and formulated for the scaled boundary finite element method. The effectivity index is derived and used to examine quality of the proposed error estimator. An algorithm for implementing the proposed p‐hierarchical adaptive procedure is developed. Numerical studies are performed on various bounded domain and unbounded domain problems. The results reflect a number of key points. Higher‐order elements are shown to be highly efficient. The effectivity index indicates that the proposed error estimator based on the classical energy norm works effectively and that the reference solution employed is a high‐quality approximation of the exact solution. The proposed p‐hierarchical adaptive strategy works efficiently. Copyright © 2007 John Wiley & Sons, Ltd.     

A discrete mechanics approach to the Cosserat rod theory—Part 1: static equilibria

A theory of discrete Cosserat rods is formulated in the language of discrete Lagrangian mechanics. By exploiting Kirchhoff's kinetic analogy, the potential energy density of a rod is a function on the tangent bundle of the configuration manifold and thus formally corresponds to the Lagrangian function of a dynamical system. The equilibrium equations are derived from a variational principle using a formulation that involves null‐space matrices. In this formulation, no Lagrange multipliers are necessary to enforce orthonormality of the directors. Noether's theorem relates first integrals of the equilibrium equations to Lie group actions on the configuration bundle, so‐called symmetries. The symmetries relevant for rod mechanics are frame‐indifference, isotropy, and uniformity. We show that a completely analogous and self‐contained theory of discrete rods can be formulated in which the arc‐length is a discrete variable ab initio. In this formulation, the potential energy density is defined directly on pairs of points along the arc‐length of the rod, in analogy to Veselov's discrete reformulation of Lagrangian mechanics. A discrete version of Noether's theorem then identifies exact first integrals of the discrete equilibrium equations. These exact conservation properties confer the discrete solutions accuracy and robustness, as demonstrated by selected examples of application. Copyright © 2010 John Wiley & Sons, Ltd.     

Tangent modulus matrix for finite element analysis of hyperelastic materials

Summary In this study, using the VEC operator [1], compact expressions are formulated for the tangent modulus matrix of hyperelastic materials, in particular elastomers, using Lagrangian coordinates. Compressible, incompressible, and near-compressible materials are considered. Expressions are obtained for the corresponding finite element tangent stiffness matrices. It is observed that the incremental stress-strain relations should be considered anisotropic. Numerical procedures based on Newton iteration are sketched. The limiting case of small strain is developed. Finally, the tangent modulus matrix is presented for the Mooney-Rivlin material, with application to the rubber rod element.     

液化场地中Timoshenko桩自由振动与屈曲的精确数值解

基于单桩的Timoshenko梁模型和桩-土相互作用的Winkler模型,建立考虑轴力效应的具有分布参数的Timoshenko梁模型微分控制方程,确定对应的齐次方程的通解,并以此作为有限单元的基函数。推导得精确形函数矩阵,建立分布参数Timoshenko梁的精确有限单元,根据拉格朗日方程得到有限元离散方程和单元刚度矩阵、几何刚度矩阵和一致质量矩阵。应用建立的精确Timoshenko梁单元于分层液化土中单桩-土-结构系统的自由振动与屈曲模态分析,通过与对应解析解以及常规有限元解的对比,表明精确Timoshenko桩基础单元的可靠性与较常规有限元法的优势。     

Behaviour of Lagrangian triangular mixed fluid finite elements

The behaviour of mixed fluid finite elements, formulated based on the Lagrangian frame of reference, is investigated to understand the effects of locking due to incompressibility and irrotational constraints. For this purpose, both linear and quadratic mixed triangular fluid elements are formulated. It is found that there exists a close relationship between the penalty finite element approach that uses reduced/selective numerical integration to alleviate locking, and the mixed finite element approach. That is, performing reduced/ selective integration in the penalty approach amounts to reducing the order of pressure interpolation in the mixed finite element approach for obtaining similar results. A number of numerical experiments are performed to determine the optimum degree of interpolation of both the mean pressure and the rotational pressure in order that the twin constraints are satisfied exactly. For this purpose, the benchmark solution of the rigid rectangular tank is used. It is found that, irrespective of the degree of mean and the rotational pressure interpolation, the linear triangle mesh, with or without central bubble function (incompatible mode), locks when both the constraints are enforced simultaneously. However, for quadratic triangle, linear interpolation of the mean pressure and constant rotational pressure ensures exact satisfaction of the constraints and the mesh does not lock. Based on the results obtained from the numerical experiments, a number of important conclusions are arrived at.     

Finite element analysis of torsional and torsional–flexural stability problems

Stiffness matrices are formulated for the torsional and lateral stability analysis of structures composed of flexural members by the matrix displacement method. The formulations are based upon approximate displacement fields which represent the action of the element in simple flexure. Example problems, for which exact solutions are known, illustrate the accuracy and convergence characteristics of the derived formulations.     

A shear-deformable two-layer plate element with interlayer slip

A new finite element for two-layer plates with built-in interlayer slip is developed. The new plate element is based on a new plate theory formulated using a new variational principle due to Reissner. The accuracy of the new element is investigated by applying it to the problem of a two-layer plate with a linear slip law in cylindrical bending for which an exact elasticity solution exists. The comparison of the inplane response for several different values of interlayer shear stiffness and layer thickness ratios shows that the new element gives accurate results. The cylindrical bending of a two-layer plate with bilinear interlayer slip law is investigated with the new element and the results obtained are compared with a plane strain analysis. Good agreement is obtained for in-plane normal stresses and displacements. To further assess the simulation capabilities of the proposed element, a rectangular plate under concentrated load is analysed and the growth of slip failure regions is shown.     

Higher order tip enrichment of eXtended Finite Element Method in thermoelasticity

An eXtended Finite Element Method (XFEM) is presented that can accurately predict the stress intensity factors (SIFs) for thermoelastic cracks. The method uses higher order terms of the thermoelastic asymptotic crack tip fields to enrich the approximation space of the temperature and displacement fields in the vicinity of crack tips—away from the crack tip the step function is used. It is shown that improved accuracy is obtained by using the higher order crack tip enrichments and that the benefit of including such terms is greater for thermoelastic problems than for either purely elastic or steady state heat transfer problems. The computation of SIFs directly from the XFEM degrees of freedom and using the interaction integral is studied. Directly computed SIFs are shown to be significantly less accurate than those computed using the interaction integral. Furthermore, the numerical examples suggest that the directly computed SIFs do not converge to the exact SIFs values, but converge roughly to values near the exact result. Numerical simulations of straight cracks show that with the higher order enrichment scheme, the energy norm converges monotonically with increasing number of asymptotic enrichment terms and with decreasing element size. For curved crack there is no further increase in accuracy when more than four asymptotic enrichment terms are used and the numerical simulations indicate that the SIFs obtained directly from the XFEM degrees of freedom are inaccurate, while those obtained using the interaction integral remain accurate for small integration domains. It is recommended in general that at least four higher order terms of the asymptotic solution be used to enrich the temperature and displacement fields near the crack tips and that the J- or interaction integral should always be used to compute the SIFs.     

Microscale heat conduction in homogeneous anisotropic media: a dual-reciprocity boundary element method and polynomial time interpolation approach

In this paper, a dual-reciprocity boundary element method based on some polynomial interpolations to the time-dependent variables is presented for the numerical solution of a two-dimensional heat conduction problem governed by a third order partial differential equation (PDE) over a homogeneous anisotropic medium. The PDE is derived using a non-Fourier heat flux model which may account for thermal waves and/or microscopic effects. In the analysis, discontinuous linear elements are used to model the boundary and the variables along the boundary. The systems of algebraic equations are set up to solve all the unknowns. For the purpose of evaluating the proposed method, some numerical examples with known exact solutions are solved. The numerical results obtained agree well with the exact solutions.     

Use of higher‐order shape functions in the scaled boundary finite element method

The scaled boundary finite element method is a novel semi‐analytical technique, whose versatility, accuracy and efficiency are not only equal to, but potentially better than the finite element method and the boundary element method for certain problems. This paper investigates the possibility of using higher‐order polynomial functions for the shape functions. Two techniques for generating the higher‐order shape functions are investigated. In the first, the spectral element approach is used with Lagrange interpolation functions. In the second, hierarchical polynomial shape functions are employed to add new degrees of freedom into the domain without changing the existing ones, as in the p‐version of the finite element method. To check the accuracy of the proposed procedures, a plane strain problem for which an exact solution is available is employed. A more complex example involving three scaled boundary subdomains is also addressed. The rates of convergence of these examples under p‐refinement are compared with the corresponding rates of convergence achieved when uniform h‐refinement is used, allowing direct comparison of the computational cost of the two approaches. The results show that it is advantageous to use higher‐order elements, and that higher rates of convergence can be obtained using p‐refinement instead of h‐refinement. Copyright © 2005 John Wiley & Sons, Ltd.     

Glued-in-rod timber joints: analytical model and finite element simulation

In this paper, a closed form analytical solution for glued-in-rod (GiR) joints is derived by solving the governing differential equations and accurately applying the boundary conditions in a cylindrical coordinate system for a GiR joint comprising of a rod, adhesive (glue) and timber. The results of the analytical model are compared with 3D continuum finite element simulations and it is shown that the closed-form solution developed can estimate the stress distribution in the adhesive and adherents (rod and timber) with good accuracy. Furthermore, the stiffness of GiR timber joints can be obtained from this analytical model. Closed-form solutions for pull–pull and pull–push test setup configurations are compared and it is shown that the maximum shear stress in the adhesive-adherent interface in a pull–push configuration is around 20% higher than that of the pull–pull counterpart. The typically (around 20%) lower strength of GiR joints in pull–push experiments compared to that of pull–pull tests can be attributed to this higher maximum shear stress which is predicted by the analytical model. A parametric study is carried out using the FE and analytical models and the effects of different variables on the distribution of stresses in the adhesive and adherents are studied.     

A more general investigation for the longitudinal stress waves in microrods with initial stress

In the present work, the propagation of longitudinal stress waves along a nanoscale bar with initial stress is investigated by using a unified nonlocal model with two length scale parameters. In principle, the analysis of wave motion is based on Love rod theory including the effects of lateral deformation. However, here are not ignored the contribution of shear stress components due to lateral deformations in the calculation of total elastic strain energy. By applying Hamilton’s principle, the explicit general solution is obtained, and comparative results containing the different effects are presented and discussed.     

Strain energy release rate determination of prescribed cracks in adhesively-bonded single-lap composite joints with thick bondlines

An analytical model for determining the strain energy release rate due to a prescribed crack in an adhesively-bonded, single-lap composite joint with thick bondlines and subjected to axial tension is presented. An existing analytical model for determining the adhesive stresses within the joint is used as the foundation for the strain energy release rate calculation. In the stress model, the governing equations of displacements within the adherends are formulated using the first-order laminated plate theory. In order to simulate the thick bondlines, the field equations of the adhesive are formulated using the linear elastic theory to allow non-uniform stress distributions through the thickness. Based on the adhesive stress distributions, the equivalent crack tip forces are obtained and the strain energy release rate due to the crack extension is determined by using the virtual crack closure technique (VCCT). The specimen geometry of ASTM D3165 standard test is followed in the derivation. The system of second-order differential equations is solved to provide the adherend and adhesive stresses using the symbolic computational tool, Maple 7. Finite element analyses using J-integral as well as VCCT are performed to verify the developed analytical model. Finite element analyses are conducted using the commercial finite element analysis software ABAQUS™. The strain energy release rates determined using the analytical method correlate well with the results from the finite element analyses. It can be seen that the same prescribed crack has a higher strain energy release rate for the joints with thicker bondlines. This explains the reason that joints with thick bondlines tend to have a lower load carrying capacity.     

An analysis of alternatives for computing axisymmetric element stiffness matrices

This note compares the use of numerical and closed form integration in computation of element stiffness matrices for axisymmetric finite element analysis. Only constant strain elements are considered. Results obtained with Gaussian quadrature and closed form integrators are compared mutually and with an exact solution obtained from classical methods. The FEM global equations estimate the force-displacement behaviour of an elastic continuum with an accuracy that depends on the integration method used. Selection of an integration order minimizing error is particularly critical in the presence of high stress gradients. Best results in the vicinity of the axis of revolution may be obtained with single-point integration rather than higher order approximations or exact integration of the element stiffness matrix. This phenomenon and its consequences are subsequently discussed.     

Hyperbolic and pseudo-hyperbolic equations in the theory of vibration

Longitudinal vibration of bars is usually considered in mathematical physics in terms of a classicalmodel described by the wave equation under the assumption that the bar is thin and relatively long. Moregeneral theories have been formulated taking into consideration the effect of the lateral motion of a relativelythick bar (beam). The mathematical formulation of these models includes higher-order derivatives in theequation of motion. Rayleigh derived the simplest generalization of the classical model in 1894, by includingthe effects of lateral motion and neglecting the shear stress. Bishop obtained the next generalization of thetheory in 1952. The Rayleigh–Bishop model is described by a fourth-order partial differential equation notcontaining the fourth-order time derivative. He took into account the effects of shear stress. Both Rayleigh’sand Bishop’s theories consider lateral displacement being proportional to the longitudinal strain. The Bishopmodel was generalized by Mindlin and Herrmann. They considered the lateral displacement proportional toan independent function of time and longitudinal coordinate. This result is formulated as a system of twodifferential equations of second order, which could be replaced by a single equation of fourth order resolvedwith respect to the highest order time derivative. To obtain a more general class of equations, the longitudinaland lateral displacements are expressed in the form of a power series expansion in the lateral coordinate. In thispaper, all of the above-mentioned equations are considered in the framework of a general theory of hyperbolicequations, with the aim of classifying the equations into general groups. The solvability of the correspondingproblems is also discussed.     

Formulation and performance of variational integrators for rotating bodies

Variational integrators are obtained for two mechanical systems whose configuration spaces are, respectively, the rotation group and the unit sphere. In the first case, an integration algorithm is presented for Euler’s equations of the free rigid body, following the ideas of Marsden et al. (Nonlinearity 12:1647–1662, 1999). In the second example, a variational time integrator is formulated for the rigid dumbbell. Both methods are formulated directly on their nonlinear configuration spaces, without using Lagrange multipliers. They are one-step, second order methods which show exact conservation of a discrete angular momentum which is identified in each case. Numerical examples illustrate their properties and compare them with existing integrators of the literature. Financial support for this work has been provided by grant DPI2006-14104 from the Spanish Ministry of Education and Science.     

A spectral finite element for axial-flexural-shear coupled wave propagation analysis in lengthwise graded beam

A new spectral element (SE) is formulated to analyse wave propagation in anisotropic inhomogeneous beam. The inhomogeneity is considered in the longitudinal direction. Due to this particular pattern of inhomogeneity, the governing partial differential equations (PDEs) have variable coefficients and an exact solution for arbitrary variation of material properties, even in frequency domain, is not possible to obtain. However, it is shown in this work that for exponential variation of material properties, the equations can be solved exactly in frequency domain, when the same parameter governs the variation of elastic moduli and density. The SE is formed using this exact solution as interpolating polynomial. As a result a single element can replace hundreds of finite elements (FEs), which are essential for all wave propagation analysis and also for accurate representation of the inhomogeneity. The developed element is used for eliciting several advantages of the gradation, including mode selection, mode blockage and smoothening of stress waves.