A stable space-time FE method for the shallow water equations

We consider the finite element (FE) approximation of the two dimensional shallow water equations (SWE) by considering discretizations in which both space and time are established using a stable FE method. Particularly, we consider the automatic variationally stable FE (AVS-FE) method, a type of discontinuous Petrov-Galerkin (DPG) method. The philosophy of the DPG method allows us to establish stable FE approximations as well as accurate a posteriori error estimators upon solution of a saddle point system of equations. The resulting error indicators allow us to employ mesh adaptive strategies and perform space-time mesh refinements, i.e., local time stepping. We establish a priori error estimates for the AVS-FE method and linearized SWE and perform numerical verifications to confirm corresponding asymptotic convergence behavior. In an effort to keep the computational cost low, we consider an alternative space-time approach in which the space-time domain is partitioned into finite sized space-time slices. Hence, we can perform adaptive mesh refinements on each individual slice to preset error tolerances as needed for a particular application. Numerical verifications comparing the two alternatives indicate the space-time slices are superior for simulations over long times, whereas the solutions are indistinguishable for short times. Multiple numerical verifications show the adaptive mesh refinement capabilities of the AVS-FE method, as well the application of the method to some commonly applied benchmarks for the SWE.

     

Three-dimensional finite element analysis for soil slopes stabilisation using piles

This paper presents the analytical methods of slope-stabilising piles using the three-dimensional (3-D) finite element (FE) analysis with the strength reduction method (SRM). This 3-D FE model is employed to overcome the limitations observed in two-dimensional (2-D) FE analysis. The solutions obtained from 3-D FE analyses are verified to be less conservative in this paper. The 3-D analysis is considered to be of particular importance to pile-slope problems. The soil that flows between piles cannot be taken account properly in the 2-D FE analysis. The method adopted in this paper can avoid the assumption of soil movement and the pressure distribution along the piles subjected to soil movement. The numerical analysis employs the Mohr–Coulomb failure criterion with the strength reduction technique for soil and an elastic member for piles. The spacing effect of the pile is considered in the 3-D model, the S/D (S: centre to centre, D: diameter of pile) ratio, equal to 4.0, is found to be equivalent to the single pile stabilisation. The middle portion of the slope is identified as the optimal location to place the piles. The proper length of the pile, which can be used to stabilise the slope, is also examined using 3-D FE analyses. It is concluded that L/H greater or equal 0.70 is recommended (L: pile length, H: slope height). The numerical analyses are conducted based on a coupled analysis, which simultaneously considers both the slope stability and the pile response. The failure mechanisms of the pile-slope system subjected to the pile locations, pile head conditions and pile length are each discussed. The contact pressure, shear force and moment along the piles are presented to illustrate the pile stabilising mechanism herein.     

6节点无界元及其应用

推导了6节点无界元的相关公式,并采用VC++和Matlab混合编程的手段,设计了基于Windows操作平台的面向对象的有限元与无界元耦合分析软件。应用该软件,成功的对某大型水电站地下硐室群的稳定性进行了有限元分析和有限元与无界元耦合分析,验证了公式的正确性。分析表明,在无限域和半无限域的岩土工程问题中引入无界元可以达到消除“边界效应”和提高计算效率的目的。     

横观各向同性软土地基上大圆筒防波堤稳定性研究

对于横观各向同性软土地基上沉入式大圆筒防波堤结构提出了一种准三维上限极限分析方法,所假设的破坏机制为大圆筒结构绕筒体内中轴线上某点发生转动失稳,泥面处形成楔体破坏,而筒底部形成圆弧滑裂面。本方法可以考虑土的三轴压缩、拉伸强度与直剪强度的差异。利用ABAQUS分别进行平面应变以及三维有限元分析,软土采用Hill本构模型,所得到的破坏模式以及大圆筒结构水平承载力与上限极限分析方法吻合较好,同时可以得出考虑地基土各向异性的大圆筒结构承载力比不考虑时有较大降低。     

A 2.5-D coupled FE–BE model for the dynamic interaction between saturated soil and longitudinally invariant structures

A two-and-a-half-dimensional (2.5-D) coupled finite element–boundary element (FE–BE) model is presented to simulate the three-dimensional dynamic interaction between saturated soils and structures with longitudinally invariant geometries. A regularized 2.5-D boundary integral equation for saturated porous media is derived that avoids the evaluation of singular traction integrals. The 2.5-D coupled FE–BE model is established by using the continuity conditions on the soil–structure interface. The developed model is verified through comparison with an existing semi-analytical method. Two case studies of a tunnel embedded in a poroelastic half-space and the efficiency of a vibration isolating screen are presented.     

A 3D composite finite element used to model a vertical drain and its surrounding remoulded soil

This work addresses an enrichment technique for the three-dimensional (3D) finite element (FE) analysis of a vertical drain foundation because (1) 1D and 2D simulations are insufficient to integrally describe the consolidation behaviour and (2) drains are small both in spacing and size, resulting in enormous computational costs for a traditional 3D FE analysis. Based on the idea of the semi-analytical finite element method (FEM), which combines general FEM with the high accuracy of a closed-form solution, a new spatial element that contains a drain well and its neighbouring smear zone is presented. This new combined element is depicted by eight global independent nodes and two local dependent nodes, and a classical analytical theory is introduced to set up the relationship between the two kinds of nodes. Because permeability diversity between the drain and the smear zone is considered, both the effects of smearing and well resistance are taken into account with the composite element method (CEM). A detailed derivation of the CEM is performed using the weighted residual method. The accuracy of the proposed method is validated with a totally penetrating, single-drain ground analysis for seven calculation conditions. Additionally, the proposed CEM saves 1/4–1/2 mesh elements and helps to avoid slender elements for the FEM analysis of the drained foundation.     

Some aspects of coupled FEBEM analysis of underground openings

The finite element method (FEM) and the boundary element method (BEM) are two well established numerical methods used for the analysis of underground openings. The advantages of both the methods are utilized by adopting FEBEM in which finite elements are coupled with boundary elements. A coupling procedure is presented in this paper. In using FEBEM, the effect of the location of interface boundary between finite element and boundary element regions, effect of Poisson's ratio and effect of stress ratio are discussed. It is shown that Poisson's ratio and stress ratio have significant effect on the accuracy of the results. Different discretization schemes are discussed to study their effect on accuracy and computation time. The use of different material properties in the FE region is presented. A comparative study is made with FEM for all the cases. It is shown that use of FEBEM is more advantageous than FEM.     

CDFSIM: Efficient Stochastic Simulation Through Decomposition of Cumulative Distribution Functions of Transformed Spatial Patterns

Simulation of categorical and continuous variables is performed using a new pattern-based simulation method founded upon coding spatial patterns in one dimension. The method consists of, first, using a spatial template to extract information in the form of patterns from a training image. Patterns are grouped into a pattern database and, then, mapped to one dimension. Cumulative distribution functions of the one-dimensional patterns are built. Patterns are then classified by decomposing the cumulative distribution functions, and calculating class or cluster prototypes. During the simulation process, a conditioning data event is compared to the class prototype, and a pattern is randomly drawn from the best matched class. Several examples are presented so as to assess the performance of the proposed method, including conditional and unconditional simulations of categorical and continuous data sets. Results show that the proposed method is efficient and very well performing in both two and three dimensions. Comparison of the proposed method to the filtersim algorithm suggests that it is better at reproducing the multi-point configurations and main characteristics of the reference images, while less sensitive to the number of classes and spatial templates used in the simulations.     

Parametric study on seismic ground response by finite element modelling

In this paper the results of 2D FE analyses of the seismic ground response of a clayey deposit, performed adopting linear visco-elastic and visco-elasto-plastic constitutive models, are presented. The viscous and linear elastic parameters are selected according to a novel calibration strategy, leading to FE results comparable to those obtained by 1D equivalent-linear visco-elastic frequency-domain analyses. The influence of plasticity on the numerical results is also investigated, with particular reference to the relation between the hysteretic and viscous damping effects. Finally, different boundary conditions, spatial discretisation and time integration parameters are considered and their role on the FE results discussed.     

方形平板锚抗拉承载力的大变形有限元分析

基于网格重分和改进的REP应力恢复技术,建立了三维大变形有限元方法研究拉力作用下方形平板锚与黏性土地基的相互作用。与常规的小变形有限元不同,大变形分析能够完整模拟平板锚的上拔过程,如果平板锚底面与土体始终保持接触,三维大变形计算得到的方板与圆板抗拉力相差很小;在无重土中的平板在加载初始即与土体脱离时,方板的承载力略低于圆板。大变形分析给出的立即脱离承载力系数与模型试验数据基本吻合,而小变形有限元与下限分析忽略了方形平板锚的长距离上拔过程对其抗拉力的影响,可能高估深锚的承载力。改进估计方形平板锚抗拉承载力的简化方法,方便于工程应用。     

Finite element analyses of two-tier geosynthetic-reinforced soil walls: Comparison involving centrifuge tests and limit equilibrium results

This study presents the procedure and results of the finite element (FE) analyses of a series of centrifuge tests on geosynthetic-reinforced soil (GRS) two-tier wall models with various offset distances. The objectives of this study were to evaluate the applicability of FE for analyzing GRS two-tier walls with various offset distances and to investigate the performance and behavior of GRS two-tier walls in various stress states. The FE simulations were first verified according to the centrifuge test results by comparing the locations of failure surfaces. The FE results were then used to investigate the effective overburden pressure, mobilization and distribution of reinforcement tensile loads, and horizontal deformation at the wall faces. The interaction between two tiers was investigated based on the FE results, which were also used to examine the modeling assumption of reinforcement tensile loads in limit equilibrium (LE) analysis and to evaluate the design methods in current design guidelines. This study demonstrated favorable agreement between FE and the centrifuge model in locating the failure surface. The FE results indicated that as the offset distance increased, the reinforcement tensile load and wall deformation decreased in both the upper and lower tiers, suggesting the attenuation of interaction between the two tiers. The maximum tensile loads of all reinforcement layers at the wall failure predicted using FE analysis and LE method assuming uniform distribution of reinforced tensile loads were comparable. Compared with the FE results, the Federal Highway Administration (FHWA) design guidelines are conservative in determining the effect of overburden pressure, required tensile strength, location of maximum tension line (for designing the reinforcement length), and the critical offset distance. Furthermore, the FHWA design guidelines do not account for the influence of the lower tier on the upper tier that was observed in this study.     

土工织物加筋垫层的应用研究

在研究土工织物加筋机理和室内外载荷试验的基础上,提出用改进的Binquet法进行加筋垫层的设计, 并在8幢4～ 7 层建筑物软基的处理中得到成功的应用。     

双列块法——大型、稀疏、对称线代数方程组的一种有效解法

双列块法用于有限元分析计算中求解大型线代数方程组。目前在有限元计算中最通常使用的求解大型线代数方程组的方法是分块的三角分解法。由于这种方法对方程组系数矩阵的分块受带宽的限制‘以至目前一般计算机的内存贮量限制了大带宽问题的求解,使一些大型的有限元课题,特别是三维问题的解算难以实现。双列块法解决了这个西难。双列块法对方程组系数矩阵实行按计算机许可的容量划分列块,按列分解,可以完全不受带宽的限制。方法是成功的、有效的。用双列块法在PE-3220小型机上仅用686K字节的内存解算了最大半带宽为1746,系数矩阵存贮量为5000K以上字节的三维(124个20节点等参元,8个16节点等参节理元、831个节点)有限元课题,这种方法使小型和高档微机用于大型有限元的解算成为可能。     

The Double Column Block Method,an Effective Solution for Large sparse symmetric System of Linear Equations

The double Column block method applies to solute of large sparse linear simultaneous equations in finite element analysis. At present the portioned triangular decomposition method is generally used for solving large systems of linear equations in FE. With the partition of the coefficient matrix dependent on its bandwidth, the solution of problems with a large bandwidth is restricted owing to the computer core storage and hence it is difficult for this method to solve some large-scal FE problems, especially the three-dimensional problems. This difficulty has been overcome by the double column block method, completely independent of the bandwidth, by means of our method the coefficient matrix is partitioned according to computer core storage capacity. It is proved that this approach is successful and effective. Based on this method, the solution has been completed of the three-dimensional FE problem with 124 twenty-node and 8 sixteen-node isoparametric elements, and 831 nodes, having a maximum half-bandwidth of 1746 and global stiffness matrix storage of more than 5000 K bytes. The computation was performed on PE-3220 minicomputer and only a core storage of 685 K bytes was used. The double column block method makes it possible for minicomputer and high-level microcomputer to be applied to calculations in large-scale FE problems.     

A new approximation in determination of vertical displacement behavior of a concrete-faced rockfill dam

A new simplified method based on one-dimensional displacement theory and 2-D finite element (FE) analysis was developed to predict the vertical displacement behavior of a concrete-faced rockfill dam. The FE analyses were carried out at the end of construction (EOC) and the end of first filling of reservoir. The proposed method was calibrated by using continuously monitored vertical displacement of the dam’s body to determine the mobilized modulus of elasticity of the rockfills at the EOC. The prediction capability of the method was demonstrated using field measurements against the findings from the 2-D FE analysis simulating characteristics of construction stages of the dam. The validity of the method was also examined on another membrane-faced rockfill dam by comparing the geodetic measurements of vertical displacement measurements of the dam’s body with the calculated vertical displacements from 2-D FE analysis at the EOC.     

Viscoelastic functionally graded finite element method with recursive time integration and applications to flexible pavements

The finite‐element (FE) method is used for modeling geotechnical and pavement structures exhibiting significant non‐homogeneity. Property gradients generated due to non‐homogeneous distributions of moisture is one such example for geotechnical materials. Aging and temperature‐induced property gradients are common sources of non‐homogeneity for asphalt pavements. Investigation of time‐dependent behavior combined with functionally graded property gradation can be accomplished by means of the non‐homogeneous viscoelastic analysis procedure. This paper describes the development of a generalized isoparametric FE formulation to capture property gradients within elements, and a recursive formulation for solution of hereditary integral equations. The formulation is verified by comparison with analytical and numerical solutions. Two application examples are presented: the first describes stationary crack‐tip fields for viscoelastic functionally graded materials, and the second example demonstrates the application of the proposed procedures for efficient and accurate simulations of interfaces between layers of flexible pavement. Copyright © 2011 John Wiley & Sons, Ltd.     

Finite element study of patterns of shear zones in granular bodies during plane strain compression

Numerical investigations of patterns of shear zones in granular bodies obtained during quasi-static plane strain compression tests were performed. The effect of a spatially correlated stochastic distribution of the initial void ratio and roughness of horizontal plates was analyzed. To describe a mechanical behavior of a cohesionless granular material during a monotonic deformation path in a plane strain compression test, a micro-polar hypoplastic constitutive model was used. FE calculations were carried out with both initially dense and initially loose cohesionless sand. A Latin hypercube method was applied to generate Gaussian truncated random fields of initial void ratio in a granular specimen. A weak correlation of the initial void ratio in both directions and its large standard deviation were assumed for all specimens. The horizontal boundaries were either ideally smooth or very rough. The FE results show similar patterns of shear zones as compared to experiments.     

Nonlinear dynamic finite element simulation of Alibey earth dam

In this study, dynamic behavior and earthquake resistance of Alibey earth dam was investigated. The dam was modeled with four node plane-strain finite elements (FE) and displacement-pore pressure coupled FE analyses were performed. Nonlinear material models such as pressure dependent and independent multi yield materials were implemented during the analyses. Transient dynamic FE analyses were performed with Newmark method. The Newton-Raphson solution scheme was adopted to solve the equations. Liquefaction and/or cyclic mobility effects were considered during the analysis. For the FE analyses, OpenSees (Open System for Earthquake Engineering Simulation) framework was adopted.     

Seismic analysis of laterally loaded pile under influence of vertical loading using finite element method

An efficient analytical approach using the finite element (FE) method, is proposed to calculate the bending moment and deflection response of a single pile under the combined influence of lateral and axial compressive loading during an earthquake, in both saturated and dry homogenous soil, and in a typical layered soil. Applying a pseudo-static method, seismic loads are calculated using the maximum horizontal acceleration (MHA) obtained from a seismic ground response analysis and a lateral load coefficient (a) for both liquefying and non-liquefying soils. It is observed that for a pile having l/d ratio 40 and embedded in dry dense sand, the normalized moment and displacement increase when the input motion becomes more severe, as expected. Further increasing of a from 0.1 to 0.3 leads to increase in the normalized moment and displacement from 0.033 to 0.042, and 0.009 to 0.035, respectively. The validity of the proposed FE based solution for estimating seismic response of pile is also assessed through dynamic centrifuge test results.     

Frictional slip plane growth by localization detection and the extended finite element method (XFEM)

This paper is concerned with developing a numerical tool for detecting instabilities in elasto‐plastic solids (with an emphasis on soils) and inserting a discontinuity at these instabilities allowing the boundary value problem to proceed beyond these instabilities. This consists of implementing an algorithm for detection of strong discontinuities within a finite element (FE) framework. These discontinuities are then inserted into the FE problem through the use of a displacement field enrichment technique called the extended finite element method (XFEM). The newly formed discontinuities are governed by a Mohr–Coulomb frictional law that is enforced by a penalty method. This implementation within an FE framework is then tested on a compressive soil block and a soil slope where the discontinuity is inserted and grown according to the localization detection. Copyright © 2010 John Wiley & Sons, Ltd.