3D Interface Element Research Articles

3D finite element-based modeling of discontinuities

A rock mass is composed of blocks, and the interfaces of various scale blocks represent different kind discontinuities. Such structure is also associated with nonuniformity of stresses. The stress–strain behavior of rock mass in the Khibiny apatite–nepheline massif in the course of mining is governed by natural geological and induced faulting. This study considers modification of the finite element method in the stress–strain analysis of rocks with regard to deformation at interfaces of different-modulus media. After 2D tests of interface elements, an optimal type of the interface element was selected for the 3D modification implementation. The latter can improve reliability of geomechanical forecasts in mineral mining in complicated geological and geodynamic conditions. From the test data on modification of interface elements, the optimal interface element is assumed to be the six-node interface element proposed by V. Kalyakin and Jianchao Li. The six-node interface element is introduced in the model of a tunnel with simulation of an unloading line at the boundary. The adequate results on adjacent rock deformation are obtained. The 3D interface element modification reveals its peculiarities and limitations as regards introduction in finite element models of mineral deposits and enclosing rock mass. The ways of solving these problems are proposed.

The effect of residual stress due to interference fit on the fatigue behavior of a fastener hole with edge cracks

Abstract The fatigue property of riveted lap joint is greatly related to the riveting-induced residual stress. However, an accurate study of the fatigue property considering the influence of residual stress quantitatively can be very difficult. A 3D interface element based on the virtual crack closure technology (VCCT) was developed to calculate the stress intensity factor (SIF) for through cracks at the hole edge. The riveting process was analyzed prior to the tensile test in the finite element code, so that the residual stress can be taken into account to get the eventual value of SIF. The result shows that, with the presence of fatigue cracks, the initial stress-strain state in the structure would change, especially near the crack tip, where great compressive stress can be found. The eventual residual stress cannot be derived by simply superimposing the riveting-induced residual stress with the crack-induced residual stress. The 3D-VCCT interface element shows strong ability to solve the SIFs. The FE analysis results agree well with the reported models both in the fatigue crack growth rate (FCGR) and in the shape of the crack front. However, when the crack is extremely short, not only the reported models, but also the present numerical model would fail. Besides, unlike Elber's model and Schijve's model, this study shows that the crack opening stress should not be a function of the stress ratio solely, but also with the crack length included.

A numerical investigation of the interplay between cohesive cracking and plasticity in polycrystalline materials

The interplay between cohesive cracking and plasticity in polycrystals is herein investigated. A unified finite element formulation with elasto-plastic elements for the grains and interface elements for the grain boundaries is proposed. This approach is suitable for the analysis of polycrystalline materials with a response ranging from that of brittle ceramics to that of ductile metals. Crystal plasticity theory is used for 3D computations, whereas isotropic von Mises plasticity is adopted for the 2D tests on plane strain cross-sections. Regarding the grain boundaries, a cohesive zone model (CZM) accounting for Mode Mixity is used for the constitutive relation of 2D and 3D interface elements. First, the analysis of the difference between 3D and 2D simulations is proposed. Then, considering all the nonlinearities in the model, their interplay is numerically investigated. It is found that the CZM nonlinearity prevails over plasticity for low deformation levels. Afterwards, plasticity prevails over CZM. Finally, for very large deformation, failure is ruled by the CZM formulation which induces softening. The meso-scale numerical results show that the simultaneous use of cohesive interface elements for the grain boundaries and plasticity theory for the grains is a suitable strategy for capturing the experimental response of uniaxial tensile tests.

A new 3D interface element for three dimensional finite element analysis of FRP strengthened RC beams

The analysis of interfacial stresses in structural component has been the subject of several investigations but it still requires more effort and studies. In this study a general three-dimensional interface element has been formulated for stress and displacement analyses in the interfacial area between two adjacent plate bending element and brick element. Interface element has 16 nodes with 5 degrees of freedom (DOF) in each node adjacent to plate bending element and 3 DOF in each node adjacent to brick element. The interface element has ability to transfer three translations from each side of interface element and two rotations in the side adjacent to the plate element. Stiffness matrix of this element was formulated and implemented in three-dimensional finite element code. Application of this element to the reinforced concrete (RC) beam strengthened with fiber reinforced polymer (FRP) including variation of deflection, slip between plate and concrete, normal and shear stresses distributions in FRP plates have been verified using experimental and numerical work of strengthened RC beams carried out by some researchers. The results show that this interface element is effective and can be used for structural component with these types of interface elements.

Assessment of computational performance for a vector parallel implementation: 3D probabilistic model discrete cracking in concrete

This work presents an assessment of the computational performance of a vector-parallel implementation of probabilistic model for concrete cracking in 3D. This paper shows the continuing efforts towards code optimization as reported in earlier works Paz, et al. (2002a,b and 2003). The probabilistic crack approach is based on the direct Monte Carlo method. Cracking is accounted by means of 3D interface elements. This approach considers that all nonlinearities are restricted to interface elements modeling cracks. The heterogeneity governs the overall cracking behavior and related size effects on concrete fracture. Computational kernels in the implementation are the inexact Newton iterative driver to solve the non-linear problem and a preconditioned conjugate gradient (PCG) driver to solve linearized equations, using an element by element (EBE) strategy to compute matrix-vector products. In particular the paper analyzes code behavior using OpenMP directives in parallel vector processors (PVP), such as the CRAY SV1 and CRAY T94. The impact of the memory architecture on code performance, and also some strategies devised to circumvent this issue are addressed by numerical experiment.

Concrete—steel bond under repeated loading

This article reports an experimental study to characterise the concrete–steel bond under repeated loading. Three loading patterns, namely partial unloading, full unloading and multiple unloading were performed. In addition, the non-linear bond behaviour of the concrete–steel interface was studied whereby the initial elastic bonding, non-linearity and softening behaviour of the interface were simulated in a numerical analysis. The complete tangential traction–slip curves, as well as the unloading/reloading routes in the repeated loading cases, were calculated and compared with the experimental curves. 3D interface elements and a damage model are used in the numerical analysis. The small number of 3D interface elements used and the high accuracy achieved by the numerical simulation ascertained the viability of the proposed approach for analysing composite structures.

Concrete?steel bond under repeated loading

This article reports an experimental study to characterise the concrete–steel bond under repeated loading. Three loading patterns, namely partial unloading, full unloading and multiple unloading were performed. In addition, the non-linear bond behaviour of the concrete–steel interface was studied whereby the initial elastic bonding, non-linearity and softening behaviour of the interface were simulated in a numerical analysis. The complete tangential traction–slip curves, as well as the unloading/reloading routes in the repeated loading cases, were calculated and compared with the experimental curves. 3D interface elements and a damage model are used in the numerical analysis. The small number of 3D interface elements used and the high accuracy achieved by the numerical simulation ascertained the viability of the proposed approach for analysing composite structures.

Behavior of eight-bolt large capacity endplate connections

Only a limited number of experimental and analytical reports exist concerning large capacity, eight-bolt, extended endplate connections. This paper describes the structural properties, both stiffness and strength, of an extended endplate connected to a column flange, with no stiffeners in either tension or compression region, using eight high strength fully and partially prestressed bolts. The case studies considered in this paper offer greater insight into the significance of endplate and column flange interaction and bolt positioning; besides the displacement and stress distributions, the prying phenomenon is also addressed. The methodology is based on a finite element modeling developed previously for the standard four-bolt array. ANSYS, version 4.4, a large-scale general purpose code is selected for equivalent three-dimensional (3D) analysis. The end-plate, beam and column flanges and web are represented as plate elements. Each bolt shank is modeled using six spar elements. 3D interface elements are used to model the boundary between column flange and back of the end-plate that may make or break contact.

3D interface elements for modelling complex potential drops. Comparison with a boundary elements method

Voltage drops due to complex surface impedance and/or surface current sources are modelled by a 3D finite elements method. Two specific formulations are given with these special 3D curvilinear second order elements: conduction in a low frequency marine electrometer and thermoelectric Seebeck effect on a solidification front. Simultaneous normal and tangential interfacial discontinuities can be computed. The method is validated by comparison with a boundary elements method and experimental values. >

A new three-dimensional interface (contact) element for fiber pull-out behavior in composites

A new three-dimensional (3D) interface element has been developed to study the nature and properties of the resistance to fiber-matrix interfacial debonding in composites composed of ductile fibers in a brittle or elastic matrix. In order to model various types of interface characteristics, an experimental/assumed nonlinear bond stress-slip relationship is used to calculate the stiffness of the discrete pull out springs connecting the steel fiber and the concrete element nodes. The cementitious matrix and the steel fiber are discretized by using 3D eight-noded isoparametric hexahedron elements and two-noded bar elements, respectively, with three translational degrees of freedom at each node. Other 3D elements such as 20-noded quadratic hexahedron, four-noded tetrahedron, six-noded linear, and 15-noded quadratic prism elements, and three-noded quadratic bar elements are incorporated in the program. The von Mises and Mohr-Coulomb yield criteria are used in the material nonlinearity. An incremental total Lagrangian formulation for concrete and an updated Lagrangian formulation for steel fiber and the interface element, are employed in the geometric nonlinearity with large deformation. The effects of fiber size and shape such as straight, hooked-end, and crimped fibers, and the influence of number of fibers being pulled out simultaneously from the same area are examined. Apparently, this study constitutes the first in which the fiber pull-out behavior in fiber-reinforced composites has been studied by using a novel approach in the nonlinear 3D interface element.