Application of new error estimators based on gradient recovery and external domain approaches to 2D elastostatics problems

Two new error estimators for the BEM in 2D potential problems were recently presented by the authors. This work extends these two error estimators for 2D elastostatics problems. The first approach involves a local error estimator based on a gradient recovery procedure in which the error function is based on differences between smoothed and non-smoothed rates of change of boundary variables in the local tangential direction. The second approach is associated with the external problem formulation and gives both local and global measures of the error, depending on a choice of the external evaluation point. These approaches are post-processing procedures. Both estimators show consistency with mesh refinement and give similar qualitative results. The error estimator using the gradient recovery approach presents a more general characteristic as its formulation does not rely on an ‘optimal’ choice of an external parameter, such as in the case of the external domain error estimator. Also, the external domain error estimator can be used only for domains in which an exterior region exists. For example, the external domain error estimator cannot be used for an infinite domain with a crack, because a point in the exterior region (inside the crack) will not be at a finite distance to the crack surface.     

Simulation of two‐ and three‐dimensional viscoplastic flows using adaptive mesh refinement

This paper presents a finite element solver for the simulation of steady non‐Newtonian flow problems, using a regularized Bingham model, with adaptive mesh refinement capabilities. The solver is based on a stabilized formulation derived from the variational multiscale framework. This choice allows the introduction of an a posteriori error indicator based on the small scale part of the solution, which is used to drive a mesh refinement procedure based on element subdivision. This approach applied to the solution of a series of benchmark examples, which allow us to validate the formulation and assess its capabilities to model 2D and 3D non‐Newtonian flows.     

A two-level mesh repartitioning scheme for the displacement-based lower-order finite element methods in volumetric locking-free analyses

We present an approach for repartitioning existing lower-order finite element mesh based on quadrilateral or triangular elements for the linear and nonlinear volumetric locking-free analysis. This approach contains two levels of mesh repartitioning. The first-level mesh re-partitioning is an h-adaptive mesh refinement for the generation of a refined mesh needed in the second-level mesh coarsening. The second-level mesh coarsening involves a gradient smoothing scheme performed on each pair of adjacent elements selected based on the first-level refined mesh. With the repartitioned mesh and smoothed gradient, the equivalence between the mixed finite element formulation and the displacement-based finite element formulation is established. The extension to nonlinear finite element formulation is also considered. Several linear and non-linear numerical benchmarks are solved and numerical inf-sup tests are conducted to demonstrate the accuracy and stability of the proposed formulation in the nearly incompressible applications.     

Automatic adaptive refinement for shell analysis using nine-node assumed strain element

An automatic adaptive refinement procedure for the analysis of shell structures using the nine-node degenerated solid shell element is suggested. The basic adaptive refinement principle and the effects of singularities and boundary layers on the convergence rate of the nine-node element used are discussed. A new stress recovery procedure based on the patch convective co-ordinate system concept is developed for the construction of a continuous smoothed stress field over the shell domains. The stress recovery procedure is easy to implement, requires a modest computational effort and needs only local patch information. It can be applied to shells with non-uniform thickness as well as to multi-layered shell structures. The smoothed recovered stress obtained is then used with the Zienkiewicz and Zhu error estimator for a posteriori error estimation during the adaptive refinement analysis. Numerical results which are in good agreement with theoretical predictions are obtained and they indicate that the current adaptive refinement procedure can eliminate the effect of singularities inside the problem domains so that a near-optimal convergence rate is achieved in all the numerical examples. This also indicates that the stress recovery procedure can produce an accurate stress field and as a result the error estimator can reflect the error distribution of the finite element solution. Even though in the current study only one type of element is used in the analysis, the whole adaptive refinement scheme can be readily applied to any other types of degenerated solid element. © 1997 John Wiley & Sons, Ltd.     

An automatic adaptive refinement procedure for the reproducing kernel particle method. Part I: Stress recovery and a posteriori error estimation

In this study, an adaptive refinement procedure using the reproducing kernel particle method (RKPM) for the solution of 2D elastostatic problems is suggested. This adaptive refinement procedure is based on the Zienkiewicz and Zhu (ZZ) error estimator for the a posteriori error estimation and an adaptive finite point mesh generator for new point mesh generation. The presentation of the work is divided into two parts. In Part I, concentration will be paid on the stress recovery and the a posteriori error estimation processes for the RKPM. The proposed error estimator is different from most recovery type error estimators suggested previously in such a way that, rather than using the least-squares fitting approach, the recovery stress field is constructed by an extraction function approach. Numerical studies using 2D benchmark boundary value problems indicated that the recovered stress field obtained is more accurate and converges at a higher rate than the RKPM stress field. In Part II of the study, concentration will be shifted to the development of an adaptive refinement algorithm for the RKPM.     

Effective stress-based finite element error estimation for composite bodies

This paper presents a discretization error estimator for displacement-based finite element analysis applicable to multi-material bodies such as composites. The proposed method applies a specific stress continuity requirement across the intermaterial boundary consistent with physical principles. This approach estimates the discretization error by comparing the discontinuous finite element effective stress function with a smoothed (C⁰ continuous) effective stress function for non-intermaterial boundary elements with a smoothed pseudo-effective stress function for elements which lie on the intermaterial boundary. Examples are presented which illustrate the effectiveness of the multi-material error estimator. The pointwise pseudo-effective stress and the L₂ norm of the estimated stress error are seen to converge with mesh refinement, while Zienkiewicz and Zhu's error estimator failed to converge for elements on the intermaterial boundary due to the physically admissible stress discontinuities that exist on the intermaterial boundary.     

Adaptive superposition of finite element meshes in non‐linear transient solid mechanics problems

An s‐adaptive finite element procedure is developed for the transient analysis of 2‐D solid mechanics problems with material non‐linearity due to progressive damage. The resulting adaptive method simultaneously estimates and controls both the spatial error and temporal error within user‐specified tolerances. The spatial error is quantified by the Zienkiewicz–Zhu error estimator and computed via superconvergent patch recovery, while the estimation of temporal error is based on the assumption of a linearly varying third‐order time derivatives of the displacement field in conjunction with direct numerical time integration. The distinguishing characteristic of the s‐adaptive procedure is the use of finite element mesh superposition (s‐refinement) to provide spatial adaptivity. Mesh superposition proves to be particularly advantageous in computationally demanding non‐linear transient problems since it is faster, simpler and more efficient than traditional h‐refinement schemes. Numerical examples are provided to demonstrate the performance characteristics of the s‐adaptive method for quasi‐static and transient problems with material non‐linearity. Copyright © 2007 John Wiley & Sons, Ltd.     

Automatic adaptive FE analysis of thin‐walled structures using 3D solid elements

In this study, a new automatic adaptive refinement procedure for thin‐walled structures using 3D solid elements is suggested. This procedure employs a specially designed superconvergent patch recovery (SPR) procedure for stress recovery, the Zienkiewicz and Zhu (Z–Z) error estimator for the a posteriori error estimation, a new refinement strategy for new element size prediction and a special mesh generator for adaptive mesh generation. The proposed procedure is different from other schemes in such a way that the problem domain is separated into two distinct parts: the shell part and the junction part. For stress recovery and error estimation in the shell part, special nodal coordinate systems are used and the stress field is separated into two components. For the refinement strategy, different procedures are employed for the estimation of new element sizes in the shell and the junction parts. Numerical examples are given to validate the effectiveness of the suggested procedure. It is found that by using the suggested refinement procedure, when comparing with uniform refinement, higher convergence rates were achieved and more accurate final solutions were obtained by using fewer degrees of freedoms and less amount of computational time. Copyright © 2008 John Wiley & Sons, Ltd.     

Remeshing for metal forming simulations—Part II: Three‐dimensional hexahedral mesh generation

In the present study, a hexahedral mesh generator was developed for remeshing in three‐dimensional metal forming simulations. It is based on the master grid approach and octree‐based refinement scheme to generate uniformly sized or locally refined hexahedral mesh system. In particular, for refined hexahedral mesh generation, the modified Laplacian mesh smoothing scheme mentioned in the two‐dimensional study (Part I) was used to improve the mesh quality while also minimizing the loss of element size conditions. In order to investigate the applicability and effectiveness of the developed hexahedral mesh generator, several three‐dimensional metal forming simulations were carried out using uniformly sized hexahedral mesh systems. Also, a comparative study of indentation analyses was conducted to check the computational efficiency of locally refined hexahedral mesh systems. In particular, for specification of refinement conditions, distributions of effective strain‐rate gradient and posteriori error values based on a Z² error estimator were used. From this study, it is construed that the developed hexahedral mesh generator can be effectively used for three‐dimensional metal forming simulations. Copyright © 2002 John Wiley & Sons, Ltd.     

A posteriori error estimation in non‐linear FE analyses of shell structures

During the last decade, significant scientific efforts were made in the area of quality assurance of numerical results obtained by means of the finite element method (FEM). These efforts were based on adaptive remeshing controlled by an estimated error. This paper reports on the extension of error estimation to non‐linear shell analysis involving strain‐hardening and softening plasticity. In the context of incremental‐iterative analyses, an incremental error estimator is introduced. It is based on the rate of work. The stress recovery technique proposed by Zienkiewicz and Zhu (Int. J. Numer. Meth. Engng 1992; 33 :1331) is modified to allow for discontinuities of certain stress components in case of localization arising from, e.g. cracking of concrete. The developed error estimator is part of a calculation scheme for adaptive non‐linear FE analysis. If the estimated error exceeds a prespecified threshold value in the course of an adaptive analysis, a new mesh is generated. After mesh refinement the state variables are transferred from the old to the new mesh and the calculation is restarted at the load level which was attained with the old mesh. The performance of the proposed error estimator is demonstrated by means of adaptive calculations of a reinforced concrete (RC) cooling tower. The influence of the user‐prescribed error threshold on the numerical results is investigated. Copyright © 2002 John Wiley & Sons, Ltd.     

An error‐estimate‐free and remapping‐free variational mesh refinement and coarsening method for dissipative solids at finite strains

A variational h‐adaptive finite element formulation is proposed. The distinguishing feature of this method is that mesh refinement and coarsening are governed by the same minimization principle characterizing the underlying physical problem. Hence, no error estimates are invoked at any stage of the adaption procedure. As a consequence, linearity of the problem and a corresponding Hilbert‐space functional framework are not required and the proposed formulation can be applied to highly non‐linear phenomena. The basic strategy is to refine (respectively, unrefine) the spatial discretization locally if such refinement (respectively, unrefinement) results in a sufficiently large reduction (respectively, sufficiently small increase) in the energy. This strategy leads to an adaption algorithm having O(N) complexity. Local refinement is effected by edge‐bisection and local unrefinement by the deletion of terminal vertices. Dissipation is accounted for within a time‐discretized variational framework resulting in an incremental potential energy. In addition, the entire hierarchy of successive refinements is stored and the internal state of parent elements is updated so that no mesh‐transfer operator is required upon unrefinement. The versatility and robustness of the resulting variational adaptive finite element formulation is illustrated by means of selected numerical examples. Copyright © 2008 John Wiley & Sons, Ltd.     

Different approaches for mixed LSFEMs in hyperelasticity: Application of logarithmic deformation measures

We present geometrically nonlinear formulations based on a mixed least‐squares finite element method. The L²‐norm minimization of the residuals of the given first‐order system of differential equations leads to a functional, which is a two‐field formulation dependent on displacements and stresses. Based thereon, we discuss and investigate two mixed formulations. Both approaches make use of the fact that the stress symmetry condition is not fulfilled a priori due to the row‐wise stress approximation with vector‐valued functions belonging to a Raviart‐Thomas space, which guarantees a conforming discretization of H(div). In general, the advantages of using the least‐squares finite element method lie, for example, in an a posteriori error estimator without additional costs or in the fact that the choice of the polynomial interpolation order is not restricted by the Ladyzhenskaya‐Babu?ka‐Brezzi condition (inf‐sup condition). We apply a hyperelastic material model with logarithmic deformation measures and investigate various benchmark problems, adaptive mesh refinement, computational costs, and accuracy.     

On using degenerated solid shell elements in adaptive refinement analysis

Three different degenerated shell elements are studied in an adaptive refinement procedure for the solution of shell problems. The stress recovery procedure expressed in a convective patch co‐ordinate system is used for the construction of continuous smoothed stress fields for the a posteriori error estimation. The performance of the stress recovery procedure, the error estimator and the adaptive refinement strategy are tested by solving three benchmark shell problems. It is found that when adaptive refinement is used, the adverse effects of boundary layers and stress singularities are eliminated and all the elements tested are able to achieve their optimal convergence rates. It is also found that the accuracy of the shell elements increases with the number of polynomial terms included in the stress and strain approximations. In addition, if complete Lagrangian polynomial terms are used, the element will be less sensitive to shape distortion than the one in which only complete polynomial terms are employed. Copyright © 1999 John Wiley & Sons, Ltd.     

Extended isogeometric analysis based on PHT‐splines for crack propagation near inclusions

Adaptive local refinement is one of the main issues for isogeometric analysis (IGA). In this paper, an adaptive extended IGA (XIGA) approach based on polynomial splines over hierarchical T‐meshes (PHT‐splines) for modeling crack propagation is presented. The PHT‐splines overcome certain limitations of nonuniform rational B‐splines–based formulations; in particular, they make local refinements feasible. To drive the adaptive mesh refinement, we present a recovery‐based error estimator for the proposed method. The method is based on the XIGA method, in which discontinuous enrichment functions are added to the IGA approximation and this method does not require remeshing as the cracks grow. In addition, crack propagation is modeled by successive linear extensions that are determined by the stress intensity factors under linear elastic fracture mechanics. The proposed method has been used to analyze numerical examples, and the stress intensity factors results were compared with reference results. The findings demonstrate the accuracy and efficiency of the proposed method.     

Multigrid solver with automatic mesh refinement for transient elastoplastic dynamic problems

This paper presents an adaptive refinement strategy based on a hierarchical element subdivision dedicated to modelling elastoplastic materials in transient dynamics. At each time step, the refinement is automatic and starts with the calculation of the solution on a coarse mesh. Then, an error indicator is used to control the accuracy of the solution and a finer localized mesh is created where the user‐prescribed accuracy is not reached. A new calculation is performed on this new mesh using the non‐linear ‘Full Approximation Scheme’ multigrid strategy. Applying the error indicator and the refinement strategy recursively, the optimal mesh is obtained. This mesh verifies the error indicator on the whole structure. The multigrid strategy is used for two purposes. First, it optimizes the computational cost of the solution on the finest localized mesh. Second, it ensures information transfer among the different hierarchical meshes. A standard time integration scheme is used and the mesh is reassessed at each time step. Copyright © 2010 John Wiley & Sons, Ltd.     

Adaptive superposition of finite element meshes in elastodynamic problems

An adaptive finite element procedure is developed for modelling transient phenomena in elastic solids, including both wave propagation and structural dynamics. Although both temporal and spatial adaptivity are addressed, the novel feature of the formulation is the use of mesh superposition to produce spatial refinement (referred to as s‐adaptivity) in transient problems. Spatial error estimation is based on superconvergent patch recovery of higher‐order accurate stresses and is used to guide mesh adaptivity, while the temporal error estimation is based on the assumption of linearly varying third‐order time derivatives of the displacement field and is used to adjust the time step size for the HHT‐α variant of the Newmark direct numerical integration method. Spatial adaptivity of the mesh is performed using a hierarchical h‐refinement scheme that is efficiently implemented using a structured version of finite element mesh superposition. This particular spatial adaptivity scheme is extremely fast and consequently makes it feasible to repeatedly update both the mesh and the time increment as required in an adaptive transient analysis. This work represents the initial effort in applying this type of spatial adaptivity to transient problems. Three example problems are given to demonstrate the performance characteristics of the s‐adaptive procedure. Copyright © 2005 John Wiley & Sons, Ltd.     

Adaptive error analysis in seepage problems

Numerical experiments in adapting variations of a computationally simple error estimator (the Zienkiewicz-Zhu estimator) to an existing finite element code are shown. The error estimator used allows both overall and local errors to be estimated. From the local estimates of error, refinements of the mesh are calculated to reach a prescribed error tolerance. These calculated refinements are used by a mesh refiner to produce a modified mesh which lowers the overall error to the prescribed value while keeping the mesh as crude as possible. The physical example on which these numerical experiments are performed is that of free surface flow through an earth dam with a toe drain. It is also shown how the problem formulation affects the error analysis and how the choice of computational scheme affects the mesh adaptation.     

Multiscale and hysteresis effects in vortex pattern simulations for Ginzburg–Landau problems

The behavior and properties of vortex pattern solutions to a benchmark Ginzburg–Landau model are investigated using hybrid continuation algorithms in conjunction with parallel adaptive mesh refinement (AMR) schemes to resolve the local vortices. The model is related to phase transition models arising in superconductivity and superfluids. The approach is based on a coupled variational formulation and finite element approximation scheme for the complex‐valued solution. The associated algorithms implement continuation treatments based on the vortex scale coherence parameter and the winding number parameter in this model. Simulation results demonstrate the behavior of non‐unique solutions, as characterized by different vortex configurations, and energy plots are used to display hysteresis effects. The complex‐valued nature of the solution also serves to illustrate some interesting open questions related to AMR strategies and error indicators for complex‐valued solution fields as well as other implications for such coupled systems. Copyright © 2009 John Wiley & Sons, Ltd.     

Reissner–Mindlin shell implementation and energy conserving isogeometric multi‐patch coupling

A shear‐flexible isogeometric Reissner–Mindlin shell formulation using non‐uniform rational B‐splines basis functions is introduced, which is used for the demonstration of a coupling approach for multiple non‐conforming patches. The six degrees of freedom formulation uses the exact surface normal vectors and curvature. The shell formulation is implemented in an isogeometric analysis framework for computation of structures composed of multiple geometric entities. To enable local model refinement as well as non‐matching domains coupling, a conservative multi‐patch approach using Lagrange multipliers for structured non‐uniform rational B‐splines patches is presented. Here, an additional border frame mesh is used to couple geometries during structural analyses. This frame interface approach avoids the problem of excessive constraints when multiple patches are coupled at one point. First, the shell formulation is verified with several reference cases. Then the influence of the frame interface discretization and frame penalty stiffness on the smoothness of the results is investigated. The effects of the perturbed Lagrangian method in combination with the frame interface approach is shown. In addition, results of models with T‐joint interface connections and perpendicular stiffener patches are presented. Copyright © 2016 John Wiley & Sons, Ltd.     

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.