NON-SINGULAR SOMIGLIANA STRESS IDENTITIES IN ELASTICITY

The paper presents two fully equivalent and regular forms of the hypersingular Somigliana stress identity in elasticity that are appropriate for problems in which the displacement field (and resulting stresses) is C^1,α continuous. Each form is found as the result of a single decomposition process on the kernels of the Somigliana stress identity in three dimensions. The results show that the use of a simple stress state for regularization arises in a direct manner from the Somigliana stress identity, just as the use of a constant displacement state regularization arose naturally for the Somigliana displacement identity. The results also show that the same construction leads naturally to a finite part form of the same identity. While various indirect constructions of the equivalents to these findings are published, none of the earlier forms address the fundamental issue of the usual discontinuities of boundary data in the hypersingular Somigliana stress identity that arise at corners and edges. These new findings specifically focus on the corner problem and establish that the previous requirements for continuity on the densities in the hypersingular Somigliana stress identity are replaced by a sole requirement on displacement field continuity. The resulting regularized and finite part forms of the Somigliana stress identity leads to a regularized form of the stress boundary integral equation (stress-BIE). The regularized stress-BIE is shown to properly allow piecewise discontinuity of the boundary data subject only to C^1,α continuity of the underlying displacement field. The importance of the findings is in their application to boundary element modeling of the hypersingular problem. The piecewise discontinuity derivation for corners is found to provide a rigorous and non-singular basis for collocation of the discontinuous boundary data for both the regularized and finite part forms of the stress-BIE. The boundary stress solution for both forms is found to be an average of the computed stresses at collocation points at the vertices of boundary element meshes. Collocation at these points is shown to be without any unbounded terms in the formulation thereby eliminating the use of non-conforming elements for the hypersingular equations. The analytical findings in this paper confirm the correct use of both regularized and finite part forms of the stress-BIE that have been the basis of boundary element analysis previously published by the first author of the current paper.     

Weakly singular stress‐BEM for 2D elastostatics

A weakly singular stress‐BEM is presented in which the linear state regularizing field is extended over the entire surface. The algorithm employs standard conforming C⁰ elements with Lagrangian interpolations and exclusively uses Gaussian integration without any transformation of the integrands other than the usual mapping into the intrinsic space. The linear state stress‐BIE on which the algorithm is based has no free term so that the BEM treatment of external corners requires no special consideration other than to admit traction discontinuities. The self‐regularizing nature of the Somigliana stress identity is demonstrated to produce a very simple and effective method for computing stresses which gives excellent numerical results for all points in the body including boundary points and interior points which may be arbitrarily close to a boundary. A key observation is the relation between BIE density functions and successful interpolation orders. Numerical results for two dimensions show that the use of quartic interpolations is required for algorithms employing regularization over an entire surface to show comparable accuracy to algorithms using local regularization and quadratic interpolations. Additionally, the numerical results show that there is no general correlation between discontinuities in elemental displacement gradients and solution accuracy either in terms of unknown boundary data or interior solutions near element junctions. Copyright © 1999 John Wiley & Sons, Ltd.     

A Hermite interpolation algorithm for hypersingular boundary integrals

This paper presents a conforming C¹ boundary integral algorithm based on Hermite interpolation. This work is motivated by the requirement that the surface function multiplying a hypersingular kernel be differentiable at the collocation nodes. The unknown surface derivatives utilized by the Hermite approximation are determined, consistent with other boundary values, by writing a tangential hypersingular equation. Hypersingular equations are primarily invoked for solving crack problems, and the focus herein is on developing a suitable approximation for this geometry. Test calculations for the Laplace equation in two dimensions indicate that the algorithm is a promising technique for three-dimensional problems.     

A 2D time-domain BEM for transient wave scattering analysis by a crack in anisotropic solids

A two-dimensional (2D) time-domain boundary element method (BEM) is presented in this paper for transient analysis of elastic wave scattering by a crack in homogeneous, anisotropic and linearly elastic solids. A traction boundary integral equation formulation is applied to solve the arising initial-boundary value problem. A numerical solution procedure is developed to solve the time-domain boundary integral equations. A collocation method is used for the temporal discretization, while a Galerkin-method is adopted for the spatial discretization of the boundary integral equations. Since the hypersingular boundary integral equations are first regularized to weakly singular ones, no special integration technique is needed in the present method. Special attention of the analysis is devoted to the computation of the scattered wave fields. Numerical examples are given to show the accuracy and the reliability of the present time-domain BEM. The effects of the material anisotropy on the transient wave scattering characteristics are investigated.     

Numerical integration schemes for evaluation of (hyper)singular integrals in 2D BEM

We consider hypersingular integral equations associated with 2D boundary value problems and defined on domains given by piecewise smooth parametric representations. In particular, given any (polynomial) local basis, we consider all integrals whose evaluation is required when the equations are solved by Galerkin BEM. In order to compute these integrals we use very efficient numerical schemes, recently proposed, which require the user to define a mesh, not necessarily uniform, on the boundary and to specify the local degrees of the approximant. Therefore these rules are quite suitable for the construction of p- and h-p versions of Galerkin BEM.     

Nearly singular approximations of CPV integrals with end- and corner-singularities for the numerical solution of hypersingular boundary integral equations

A local numerical approach to cope with the singular and hypersingular boundary integral equations (BIEs) in non-regularized forms is presented in the paper for 2D elastostatics. The approach is based on the fact that the singular boundary integrals can be represented approximately by the mean values of two nearly singular boundary integrals and on the techniques of distance transformations developed primarily in previous work of the authors. The nearly singular approximations in the present work, including the normal and the tangential distance transformations, are designed for the numerical evaluation of boundary integrals with end-singularities at junctures between two elements, especially at corner points where sufficient continuity requirements are met. The approach is very general, which makes it possible to solve the hypersingular BIE numerically in a non-regularized form by using conforming C⁰ quadratic boundary elements and standard Gaussian quadratures without paying special attention to the corner treatment.With the proposed approach, an infinite tension plate with an elliptical hole and a pressurized thick cylinder were analyzed by using both the formulations of conventional displacement and traction boundary element methods, showing encouragingly the efficiency and the reliability of the proposed approach. The behaviors of boundary integrals with end- and corner-singular kernels were observed and compared by the additional numerical tests. It is considered that the weakly singularities remain but should have been cancelled with each other if used in pairs by the corresponding terms in the neighboring elements, where the corner information is included naturally in the approximations.     

Three-dimensional crack problem analysis using boundary element method with finite-part integrals

A set of hypersingular integral equations of a three-dimensional finite elastic solid with an embedded planar crack subjected to arbitrary loads is derived. Then a new numerical method for these equations is proposed by using the boundary element method combined with the finite-part integral method. According to the analytical theory of the hypersingular integral equations of planar crack problems, the square root models of the displacement discontinuities in elements near the crack front are applied, and thus the stress intensity factors can be directly calculated from these. Finally, the stress intensity factor solutions to several typical planar crack problems in a finite body are evaluated. This revised version was published online in August 2006 with corrections to the Cover Date.     

Einführung in moderne Galerkin-Randeiementmethoden mit einer Anwendung aus dem Maschinenbau

The Galerkin-type boundary element method (BEM) is an discretization procedure for integral equations, represents itself however compared with classical integral equation methods as an universal tool for the solution of practical engineering problems and can be coupled very easily with finite element substructures. The BEM, whose main advantage lies in the fact that only a surface mesh must be generated, is superior to FEM in special applications, i.e. in elastostatics (notch problems) and fracture mechanics. In this paper the individual steps to solving an elliptical boundary value problem of 3-D linear elasticity theory by way of an equivalent system of boundary integral equations will be explained. For the mathematical investigation of elliptical differential equations and integral equations, the theory of Sobolev spaces has proved to be especially suitable. Basic terms to Sobolev spaces will be introduced so that the reader does not have to refer to textbooks for new terms. The transformation of elliptical boundary value problems to systems of singular and hypersingular integral equations will be explained with help of a Calderón projector, which is defined by using fundamental solutions. The discretization of the obtained integral equations with the Galerkin-type BEM will be presented. Finally the approximation of non-linear problems by using the Galerkin-type BEM will be shown. A numerical test for a strength problem will be discussed shortly.     

Analytical study and numerical experiments for degenerate scale problems in the boundary element method for two‐dimensional elasticity

For a plane elasticity problem, the boundary integral equation approach has been shown to yield a non‐unique solution when geometry size is equal to a degenerate scale. In this paper, the degenerate scale problem in the boundary element method (BEM) is analytically studied using the method of stress function. For the elliptic domain problem, the numerical difficulty of the degenerate scale can be solved by using the hypersingular formulation instead of using the singular formulation in the dual BEM. A simple example is shown to demonstrate the failure using the singular integral equations of dual BEM. It is found that the degenerate scale also depends on the Poisson's ratio. By employing the hypersingular formulation in the dual BEM, no degenerate scale occurs since a zero eigenvalue is not embedded in the influence matrix for any case. Copyright © 2002 John Wiley & Sons, Ltd.     

Dual boundary element analysis using complex variables for potential problems with or without a degenerate boundary

The dual boundary element method in the real domain proposed by Hong and Chen in 1988 is extended to the complex variable dual boundary element method. This novel method can simplify the calculation for a hypersingular integral, and an exact integration for the influence coefficients is obtained. In addition, the Hadamard integral formula is obtained by taking the derivative of the Cauchy integral formula. The two equations (the Cauchy and Hadamard integral formula) constitute the basis for the complex variable dual boundary integral equations. After discretizing the two equations, the complex variable dual boundary element method is implemented. In determining the influence coefficients, the residue for a single-order pole in the Cauchy formula is extended to one of higher order in the Hadamard formula. In addition, the use of a simple solution and equilibrium condition is employed to check the influence matrices. To extract the finite part in the Hadamard formula, the extended residue theorem is employed. The role of the Hadamard integral formula is examined for the boundary value problems with a degenerate boundary. Finally, some numerical examples, including the potential flow with a sheet pile and the torsion problem for a cracked bar, are considered to verify the validity of the proposed formulation. The results are compared with those of real dual BEM and analytical solutions where available. A good agreement is obtained.     

Numerical p‐version refinement studies for the regularized stress‐BEM

In the development of the boundary element method (BEM) and the finite element method (FEM) researchers have typically selected similar basis functions. That is, both methods typically employ low‐order interpolations such as piece‐wise linear or piece‐wise quadratic and rely on h‐version refinement to increase accuracy as required. In the case of the FEM, the decision to use low‐order elements is made for computational efficiency as an attractive compromise between local modeling accuracy and sparseness of the resulting linear system. However, in many BEM formulations, low‐order elements may be the only practical choice given the complexity of using analytic integration formulae in conjunction with special integral interpretations. Unlike their efficient use in the FEM, fine meshes of low‐order elements in the BEM are highly inefficient from a computational standpoint given the dense nature of BEM systems. Moreover, owing to singularities in the kernel functions, the BEM should be expected to benefit more so than the FEM from very high levels of local accuracy. Through the use of regularized algorithms which only require numerical integration, p‐version refinement in the BEM is easily extended to include any set of basis functions with no significant increase in programming complexity. Numerical results show that by using interpolations as high as 12th and 16th order, one can expect reductions in error by as many as five orders of magnitude over comparable algorithms based on similar system size. For two‐dimensional problems, it is also shown that, for a given level of error, one can expect reductions in system size by an order of magnitude, thus leading to a reduction in computational expense for conventional algorithms by three orders of magnitude. Copyright © 2003 John Wiley & Sons, Ltd.     

Application of bicubic splines in the boundary element method

In this paper, we proposed the method for improving the accuracy of BEM, which is based on application of bicubic splines for interpolation functions. Application of bicubic splines ensures continuity of class C¹ at the boundaries of element. Such an interpolation results in smooth approximation of the surface sources leading to high accuracy of computation. Set of integral equations is solved by implementation of Galerkin method for determination of unknown coefficients.The accuracy of the proposed approach is illustrated by comparison of the solution of electric field in thin-plate capacitor by BEM using bicubic splines, second-order polynomial, linear and piecewise-constant interpolation.     

A mixed boundary value problem for the unsteady Stokes system in a bounded domain in Rn

A mixed boundary value problem for the unsteady Stokes system is studied from the point of view of the theory of hydrodynamic potentials. Existence and uniqueness results as well as boundary integral representations of the classical solution are given for bounded domains having compact but not connected boundaries of class C^1,α (0<α≤1).     

NEW NUMERICAL INTEGRATION SCHEMES FOR APPLICATIONS OF GALERKIN BEM TO 2-D PROBLEMS

We consider hypersingular integral formulation of some elasticity and potential boundary value problems on 2-D domains. In particular, we consider all integrals whose evaluation is required when the equations are solved by a Galerkin BEM based on piecewise polynomial approximants of arbitrary local degrees. In order to compute these integrals, we use very efficient formulas recently proposed, which require the user to define a mesh, not necessarily uniform, on the boundary and specify the local degrees of the approximant. These rules are quite suitable for the construction of h–p version of the BEM. Implementation of h−, p− and h–p methods are applied to some classical problems and several numerical results are presented. © 1997 by John Wiley & Sons, Ltd.     

A hypersingular time‐domain BEM for 2D dynamic crack analysis in anisotropic solids

A hypersingular time‐domain boundary element method (BEM) for transient elastodynamic crack analysis in two‐dimensional (2D), homogeneous, anisotropic, and linear elastic solids is presented in this paper. Stationary cracks in both infinite and finite anisotropic solids under impact loading are investigated. On the external boundary of the cracked solid the classical displacement boundary integral equations (BIEs) are used, while the hypersingular traction BIEs are applied to the crack‐faces. The temporal discretization is performed by a collocation method, while a Galerkin method is implemented for the spatial discretization. Both temporal and spatial integrations are carried out analytically. Special analytical techniques are developed to directly compute strongly singular and hypersingular integrals. Only the line integrals over an unit circle arising in the elastodynamic fundamental solutions need to be computed numerically by standard Gaussian quadrature. An explicit time‐stepping scheme is obtained to compute the unknown boundary data including the crack‐opening‐displacements (CODs). Special crack‐tip elements are adopted to ensure a direct and an accurate computation of the elastodynamic stress intensity factors from the CODs. Several numerical examples are given to show the accuracy and the efficiency of the present hypersingular time‐domain BEM. Copyright © 2008 John Wiley & Sons, Ltd.     

Determination of the natural frequencies and natural modes of a rod using the dual BEM in conjunction with the domain partition technique

In this paper, the dual BEM in conjunction with the domain partition technique is employed to solve both natural frequencies and natural modes of a rod. In this new approach, there exists no spurious eigenvalue using the complex-valued singular or hypersingular equation alone. In the derivation of the singular and hypersingular integral equations, if only the real parts of the kernel functions are chosen, the resulting eigenequations have spurious eigenvalues. Such spurious eigenvalues stem from adding the dummy links into the interior structures considered. Although the spurious eigenvalues exist in this approach which uses the real-valued kernel functions, the possible indeterminacy of eigenmodes using the conventional real-valued singular or real-valued hypersingular equations disappears when the domain partition technique is adopted. The conventional real-valued multiple reciprocity BEM results in spurious eigenvalues for the mixed boundary conditions and indeterminacy of eigenmodes owing to insufficient rank of the leading coefficient matrix for the Dirichlet and Neumann boundary conditions. Such problems can be solved by combining the singular and hypersingular equations together; however, they also can be treated by using the real-valued singular or hypersingular equation alone if the domain partition technique is adopted. Three examples including the Dirichlet, Neumann and mixed type boundary conditions are investigated to show the validity of current approach.     

A 2D time-domain collocation-Galerkin BEM for dynamic crack analysis in piezoelectric solids

A time-domain boundary element method (TDBEM) for transient dynamic analysis of two-dimensional (2D), homogeneous, anisotropic and linear piezoelectric cracked solids is presented in this paper. The present analysis uses a combination of the strongly singular displacement boundary integral equations (BIEs) and the hypersingular traction boundary integral equations. The spatial discretization is performed by a Galerkin-method, while a collocation method is implemented for the temporal discretization. Both temporal and spatial integrations are carried out analytically. In this way, only the line integrals over a unit circle arising in the time-domain fundamental solutions are computed numerically by standard Gaussian quadrature. An explicit time-stepping scheme is developed to compute the unknown boundary data including the generalized crack-opening-displacements (CODs) numerically. Special crack-tip elements are adopted to ensure a direct and an accurate computation of the dynamic field intensity factors (IFs) from the CODs. Several numerical examples involving stationary cracks in both infinite and finite solids under impact loading are presented to show the accuracy and the efficiency of the developed hypersingular time-domain BEM.     

Evaluation of polymer fracture parameters by the boundary element method

The boundary element method (BEM) for two-dimensional linear viscoelasticity is applied to polymer fracture. The time-dependence of stress intensity factors is assessed for various viscoelastic models as well as loading and support conditions. Various representations of the energy release rate under isothermal conditions are adopted. Additional boundary integral equations for the displacement gradient in the domain are linked to algorithms for the evaluation of path-independent J-integrals. The consistency of BEM predictions and their good agreement with other analytical results confirms BEM as a valid modelling tool for viscoelastic fracture characterisation and failure assessment under complex geometric and loading conditions.     

Accuracy of desingularized boundary integral equations for plane exterior potential problems

In this article, computational results from boundary integral equations and their normal derivatives for the same test cases are compared. Both kinds of formulations are desingularized on their real boundary. The test cases are chosen as a uniform flow past a circular cylinder for both the Dirichlet and Neumann problems. The results indicate that the desingularized method for the standard boundary integral equation has a much larger convergence speed than the desingularized method for the hypersingular boundary integral equation. When uniform nodes are distributed on a circle, for the standard boundary integral formulation the accuracies in the test cases reach the computer limit of 10⁻¹⁵ in the Neumann problems; and O(N⁻³) in the Dirichlet problems. However, for the desingularized hypersingular boundary integral formulation, the convergence speeds drop to only O(N⁻¹) in both the Neumann and Dirichlet problems.     

A simple a-posteriori error estimation for adaptive BEM in elasticity

In this paper, the properties of various boundary integral operators are investigated for error estimation in adaptive BEM. It is found that the residual of the hyper-singular boundary integral equation (BIE) can be used for a-posteriori error estimation for different kinds of problems. Based on this result, a new a-posteriori error indicator is proposed which is a measure of the difference of two solutions for boundary stresses in elastic BEM. The first solution is obtained by the conventional boundary stress calculation method, and the second one by use of the regularized hyper-singular BIE for displacement derivative. The latter solution has recently been found to be of high accuracy and can be easily obtained under the most commonly used C ⁰ continuous elements. This new error indicator is defined by a L ₁ norm of the difference between the two solutions under Mises stress sense. Two typical numerical examples have been performed for two-dimensional (2D) elasticity problems and the results show that the proposed error indicator successfully tracks the real numerical errors and effectively leads a h-type mesh refinement procedure.