Adaptive finite element analysis of axisymmetric freezing

The paper describes an adaptive finite element analysis of the transient axisymmetric freezing process. Adaptivity schemes are applied to both space and time tessellations. Error equidistribution adjusts the nodal positions and Taylor series analysis of time truncation error selects the time step. In implicit time integration, the location of the freezing front and the nodal temperatures at the next time is solved by full Newton all at once. The method appears to be accurate; in cases for which closed-form solutions are available, it agrees well with them. It also avoids the problem found in the Modified Isotherm Migration Method where the freezing front tends to retrogress when solid just forms on the outer surface cooled by convection.     

Energy distributions estimation using stochastic finite element

In this paper two problems are treated: (1) estimation of the mean value of a random function Z(x), defined in a stochastic finite element (SFE) v, z_v=1/v∫_vZ(x) dx, where the distributions of Z(x) at each node are known; and (2) Kriking solution with SFE under the non-stationary hypothesis: E(Z(x))=m(x), C(x, h)=E{Z(x+h)Z(x)}−m(x+h)m(x). Several temperature distribution results are presented using a plane SFE. Finally, the conclusions are given underlining SFE applications in energy, hydrology, geology etc., generally in whatever disciplines the distributions are used.     

机车复轨器的有限元分析

复轨器是救援脱轨机车车辆的专用必备装置.在救援过程中,它能引导脱轨机车车辆的轮对滚动到钢轨顶部,复位到钢轨上,从而达到救援排障的目的,此间复轨器须承受较大的载荷.为了达到既满足强度和刚度的要求,又能减轻重量、节约材料等目的,对复轨器作较为精确的应力、位移等分析具有一定的实用价值.本文利用有限元分析软件ANSYS对某公司生产制造的人字型复轨器进行有限元计算分析,并且根据计算分析结果为以后此类复轨器的改进或重新设计提供重要的理论参考和工程实例.人字型复轨器由左"人"右"入"两个单件组成,其结构和组成如图1所示.     

An elasto-plastic finite element for steel pipelines

An efficient finite element for the modeling of inelastic behaviour of three-dimensional pipe systems is presented. The formulation is based on a two-node pipe element with 12 degrees of freedom. The element consists of an elastic portion and two potentially plastic 3D hinges of zero-length lumped at both nodes. The behaviour of the plastic hinges is characterized using recently derived and experimentally validated plastic interaction relations for pipe sections. The normality condition of plasticity is applied to the analytically derived yield hyper-surface at the stress resultant level in order to approximately simulate material elasto-plastic behaviour. The element models shear deformation effects both in the elastic and plastic ranges. Thus, it is suitable for predicting pipe behaviour subjected to high shearing forces. The model captures the essential features of pipe behaviour using a remarkably small number of degrees of freedom and is particularly suited for the analysis of long pipeline systems. Solutions for simple problems are provided and compared to several other well-established elements in the ABAQUS library in order to assess the validity of the results and demonstrate their scope of applicability.     

Adaptive finite element analysis of microwave driven convection

This study describes an adaptive finite element methodology for heat transfer by convection applied to microwave heating of liquids. This is the first attempt to model such type of problems employing the concepts of error estimation and mesh adaptivity. The proposed methodology is generic and can be applied to steady-state, transient, linear and nonlinear problems involving heat transfer by conduction and convection. There was very good agreement between simulation and experimental results.     

A finite element solution of double diffusive convection

The finite element method is used to analyze double diffusive convection in a rectangular box. The dimensions of the domain are chosen such that it matches exactly with a unit cell. Three cases are studied which correspond to three sets of boundary conditions on the walls. Numerical results are presented for Prandtl numbers from 0.1 to 10.0 and thermal Rayleigh number ranging from 2,000 to 10,000. Solute Rayleigh number is held constant at 1,000 throughout the study. The type of element used is the eight noded element with all dependent variables except the pressure interpolated quadratically and the pressure interpolated linearly. Numerical results are presented in terms of isotherms, streamlines, and isohals. Oscillatory solutions are obtained for a certain range of thermal Rayleigh numbers from the steady state algorithm. The oscillations can be seen in terms of the numerical solution oscillating from iteration to iteration about a mean solution.     

A finite element analysis for inverse heat conduction problems

This paper presents an efficient inverse analysis technique based on a sensitivity coefficient algorithm to estimate the unknown boundary conditions of multidimensional steady and transient heat conduction problems. Sensitivity coefficients were used to represent the temperature response of a system under unit loading conditions. The proposed method, coupled with the sensitivity analysis in the finite element formulation, is capable of estimating both the unknown temperature and heat flux on the surface provided that temperature data are given at discrete points in the interior of a solid body. Inverse heat conduction problems are referred to as ill-posed because minor inaccuracy or error in temperature measurements cause a drastic effect on the predicted surface temperature and heat flux. To verify the accuracy and validity of the new method, two-dimensional steady and transient problems are considered. Their surface temperature and heat flux are evaluated. From a comparison with the exact solution, the effects of measurement accuracy, number and location of measuring points, a time step, and regularization terms are discussed. © 1998 Scripta Technica. Heat Trans Jpn Res, 26(6): 345–359, 1997     

Non-linear finite element analysis and high temperature defect assessment

In recent years a methodology has been developed for predicting the complex phenomenon of creep-fatigue crack growth. This is of considerable importance in a wide variety of modern components and structures, for example gas turbines, high temperature pressure vessels and pipework. Essentially the technique involves calculating rates of crack growth for the limiting cases of rapid cycling and steady loading, and then adding the rates to give combined effects. As well as creep data and the well known Paris Law for fatigue crack growth, the method requires linear and non-linear fracture mechanics quantities K and C* to be calculated. K is the well known elastic crack tip stress intensity factor. C* is the creep analogy of the J-integral, defined by:

The paper describes the methodology for determining creep-fatigue crack growth and the application of ABAQUS to determine K and C* for a test specimen and a turbine blade. The use of approximate techniques is also evaluated.     

Experiences using three-dimensional finite element analysis for leak-before-break assessment

Because of practical limitations, analytical problems in fracture mechanics have often been solved using simplified geometries (e.g. plane stress/strain, shell models). We have applied the Leak-Before-Break approach extensively to the large diameter piping of a new nuclear power plant. Various piping components such as elbows, tee and branch connections with postulated cracks were analyzed. Since no credible geometric simplification was possible, fully three-dimensional (3D) analytical models were found to be essential.

The paper describes our experiences in performing 3D Finite Element (FE) analysis of these components. Included are comparisons of numerical and test results of compact specimens, material modeling considerations, handling of 3D effects, such as the variation of the J-integral along a crack front, and especially, the effects of plasticity.

The overall intent of the paper is not simply to present specific numerical results, but rather to give some perspective on the effort required and results attainable.     

Direct finite element route for design-by-analysis of pressure components

In the new European standard for unfired pressure vessels, EN 13445-3, there are two approaches for carrying out a Design-by-Analysis that cover both the stress categorization method (Annex C) and the direct route method (Annex B) for a check against global plastic deformation and against progressive plastic deformation. This paper presents the direct route in the language of limit and shakedown analysis. This approach leads to an optimization problem. Its solution with Finite Element Analysis is demonstrated for mechanical and thermal actions. One observation from the examples is that the so-called 3f (3Sm) criterion fails to be a reliable check against progressive plastic deformation. Precise conditions are given, which greatly restrict the applicability of the 3f criterion.     

基于参数化有限元分析方法的装配型架优化

装配型架的设计对飞机产品的制造与装配具有重要意义.基于参数化的型架有限元分析法能很好地克服现有装配型架设计方法工作效率低下的弱点.基于飞机装配型架设计的基本原理,在模型简化原则的基础上,将参数化设计与有限元相结合,通过改变型架参数自动生成分析模型,并完成型架有限元分析,从而实现型架骨架的界面优化.该方法在一定程度上提高了型架分析的效率,可缩短实际生产工作周期,避免了大量人力物力的浪费.     

Modeling hydrogen dragging by mobile dislocations in finite element simulations

Finite element simulation modeling permits to predict hydrogen concentration for various initial boundary-values problems, but the results depend on the underlying transport mechanisms accounted for. Trapping process is a key factor in the apparent hydrogen diffusion, and the case of mobile traps as dislocations needs modification of the hydrogen transport equation usually considered in the literature. An extension of this model is proposed where hydrogen dragging by mobile traps is modeled by reaction-diffusion equations, involving trapping and detrapping kinetic, and is applied for evolving trap density with plastic strain. The consequences of trapped hydrogen mobility on diffusive hydrogen repartition in a reference Small Scale Yielding configuration are focused on, especially in term of acceleration of hydrogen transport. The potentiality of the model is illustrated by the modeling of the trapped hydrogen breakaway from fast moving dislocations.     

Compressible heat transfer computations by an adaptive finite element method

This paper presents adaptive finite element computations of laminar jet impingement heat transfer. Variable fluid properties and compressibility effects are considered. A unified formulation of the equations is used to treat the simultaneous presence of three flow regimes: incompressible (ρ=constant), compressible (ρ=ρ(p,T)), and anelastic (ρ=ρ(T)). The error estimator uses a local least squares projection method and accounts for errors in velocity, pressure and temperature. The performance of the methodology is verified by solving a problem possessing a closed form solution. Several applications are then considered. We study two different gases (air and CO₂), different conditions (heated, cooled or constant properties), compressibility and inlet velocity profile effects. Heat transfer is a key element of the study. Results indicate that the methodology can produce grid independent solutions even for derived quantities and in thin boundary layers.     

Direct finite element design optimisation of the cagelessreluctance synchronous machine

The finite element analysis method is used directly in optimisation algorithms to optimise in multidimensions the design of the cageless reluctance synchronous machine. Two optimisation methods are evaluated to minimise or maximise the function value. These are the direction set method of Powell and the quasi-Newton algorithm. Both methods proved to be successful, with some advantages and disadvantages. Using these methods at a power level below 10 kW, results are given of structures of the reluctance synchronous machine which have been optimised according to specific criteria. Calculated and measured results show that the maximum torque optimum designed reluctance synchronous machine has the advantages of high power density and high efficiency     

A vortex model for Darrieus turbine using finite element techniques

Since 1970 several aerodynamic prediction models have been formulated for the Darrieus turbine. We can identify two families of models: stream-tube and vortex. The former needs much less computation time but the latter is more accurate. The purpose of this paper is to show a new option for modelling the aerodynamic behaviour of Darrieus turbines. The idea is to combine a classic free vortex model with a finite element analysis of the flow in the surroundings of the blades. This avoids some of the remaining deficiencies in classic vortex models. The agreement between analysis and experiment when predicting instantaneous blade forces and near wake flow behind the rotor is better than the one obtained in previous models.     

基于显式有限元的机车碰撞模型研究

结合DF8B型机车,利用ANSYS/LS-DYNA软件分别建立了考虑转向架结构和考虑转向架质量的机车以10m/s的速度正面碰撞刚性墙的仿真模型,并对两种模型的仿真分析结果进行了比较。结果表明,考虑转向架实际结构的机车碰撞刚性墙有限元模型,能更准确地模拟机车的正面碰撞过程。     

Hygrothermal coupling analysis of magneto-electroelastic beams using finite element methods

In this article, the finite element (FE) method has been used to assess the coupled static behavior of hygro-thermo-magneto-electroelastic (HTMEE) beam. Influence of externally applied hygrothermal loads on the direct (displacements, electric and magnetic potentials) and derived quantities (stresses, electric displacement and magnetic flux densities) of HTMEE beam have been studied in detail. The principle of total potential energy and the coupled constitutive equations of HTMEE material are used for the FE formulation. A generalized condensation technique is adopted to solve the global FE equations of motion. Numerical examples are discussed to examine the effect of hygrothermal loads and distinct effect of moisture concentration on the behavior of the beam. Particular emphasis has been placed to analyze the influence of temperature and moisture dependent elastic stiffness coe?cients associated with empirical constants. Considering the independent effect of temperature and moisture on the coupled static responses, the most significant combination of the empirical constants corresponding to temperature dependency and moisture dependency are explored. Extensive computational examples are considered to examine the significant effect of boundary conditions, temperature gradient, moisture concentration gradient and empirical constants on the static behavior of HTMEE beam. It is observed that the static behavior of HTMEE beam is significantly influenced by the hygrothermal loads and empirical constants. The results presented in this article would serve as a benchmark results in design and analysis of HTMEE structures for sensors and actuators applications.     

Stochastic finite element modeling of transient heat transfer in layered composites

The stochastic finite element analysis of the transient heat transfer in composite materials is the main subject of the paper presented. The approach proposed is based on the second order perturbation second central probabilistic moment method. The methodology proposed is illustrated with the example of transient heat transfer in the three-component layered composite with randomly defined thermal properties.     

Plastic limit pressures for cracked pipes using finite element limit analyses

Based on detailed finite element (FE) limit analyses, the present paper provides approximations for plastic limit pressure solutions for plane strain pipes with extended inner axial cracks; axi-symmetric (inner) circumferential cracks; axial through-wall cracks; axial (inner) surface cracks; circumferential through-wall cracks; and circumferential (inner) surface cracks. In particular, for surface crack problems, the effect of the crack shape, semi-elliptical or rectangular, on the limit pressure is quantified. Comparisons with existing analytical and empirical solutions show a large discrepancy for short circumferential through-wall cracks and for surface cracks (both axial and circumferential). Being based on detailed 3D FE limit analysis, the present solutions are believed to be accurate, and thus to be valuable information not only for plastic collapse analysis of pressurised piping but also for estimating non-linear fracture mechanics parameters based on the reference stress approach.     

Thermo-mechanical modelling of a single-bead-on-plate weld using the finite element method

This work describes a three-dimensional thermo-mechanical finite element analysis of a single weld bead-on-plate of austenitic stainless steel performed as part of the NeT programme. The overall aim is to validate the use of finite element (FE) weld simulations to accurately predict residual stress states for use in the assessment of welded components. Here, the final residual stresses in the plate are predicted, which can later be verified through comparison with measured distributions. A one-way coupled analysis is carried out with the thermal and mechanical problems solved in sequence. For the thermal analysis, two approaches are adopted to simulate the welding process. In one case sections of the weld bead's elements are sequentially deposited, while in the other the whole bead is deposited simultaneously. A moving heat source is used to simulate the torch traversal over the weld bead in both cases. Predicted thermal profiles, residual stresses and equivalent plastic strains, due to the welding process are presented at certain locations in the plate.