含高温度梯度及接触热阻非线性热力耦合问题的谱元法

建立了含高温度梯度及接触热阻的非线性热力耦合问题的谱元法格式, 考虑了温度相关的热导率、弹性模量、泊松比和热膨胀系数, 以及界面应力相关的接触热阻的影响. 谱元法的插值函数基于非等距分布的Lobatto结点集或第二类Chebyshev结点集, 兼具谱方法的高精度和有限元法的灵活性. 数值算例表明, 建立的谱元法计算格式可以高效高精度地求解域内高温度梯度以及含接触热阻的非线性热力耦合问题, 不仅收敛速度快于传统有限元法, 而且用较少的自由度和较短的计算时间即可得到比传统有限元法更高精度的计算结果, 在工程实际热力耦合问题中具有广阔的应用前景.      

TEC结构的三维非线性瞬态温度场分析

热电制冷器(TEC)以其体积小、作用速度快及无噪音等机械制冷无法替代的优点在航空航天和电子工业等领域得到了越来越广泛的应用。本文根据TEC的导热特点,推导了TEC结构稳态温度场的解析解,建立了其瞬态非线性温度场分析的微分方程。利用伽辽金法导出TEC结构热分析的有限元方程,对非线性热分析的有限元方程进行了求解,得到了TEC的稳态温度场和瞬态响应温度场。算例结果表明,本文提出的TEC结构热分析有限元模型具有较高的精度,能够有效地分析TEC的非线性瞬态温度场。     

Reduced‐order modeling for nonlocal diffusion problems

In several settings, diffusive behavior is observed to not follow the rate of spread predicted by parabolic partial differential equations (PDEs) such as the heat equation. Such behaviors, often referred to as anomalous diffusion, can be modeled using nonlocal equations for which points at a finite distance apart can interact. An example of such models is provided by fractional derivative equations. Because of the nonlocal interactions, discretized nonlocal systems have less sparsity, often significantly less, compared with corresponding discretized PDE systems. As such, the need for reduced‐order surrogates that can be used to cheaply determine approximate solutions is much more acute for nonlocal models compared with that for PDEs. In this paper, we consider the construction, application, and testing of proper orthogonal decomposition (POD) reduced models for an integral equation model for nonlocal diffusion. For certain modeling parameters, the model we consider allows for discontinuous solutions and includes fractional Laplacian kernels as a special case. Preliminary computational results illustrate the potential of using POD to obtain accurate approximations of solutions of nonlocal diffusion equations at much lower costs compared with, for example, standard finite element methods. Copyright © 2016 John Wiley & Sons, Ltd.     

Fully Coupled THMC Modeling of Wellbore Stability with Thermal and Solute Convection Considered

Wellbore stability analysis is an important topic in petroleum geomechanics. Analytical and numerical analysis of wellbore stability involves the study of interactions among pressure, temperature and chemical changes, and the mechanical response of the rock, a coupled thermal–hydraulic–mechanical–chemical (THMC) process. Thermal and solute convection have usually been overlooked in numerical models. This is appropriate for shales with extremely low permeability, but for shales with intermediate and high permeability (e.g., shale with a disseminated microfissure network), thermal and solute convection should be considered. The challenge of considering advection lies in the numerical oscillation encountered when implementing the traditional Galerkin finite element approach for transient advection–diffusion problems. In this article, we present a fully coupled THMC model to analyze the stress, pressure, temperature, and solute concentration changes around a wellbore. In order to overcome spurious spatial temperature oscillations in the convection-dominated thermal advection–diffusion problem, we place the transient problem into an advection– diffusion-reaction problem framework, which is then efficiently addressed by a stabilized finite element approach, the subgrid scale/gradient subgrid scale method (SGS/GSGS).     

Simulation of multiphase flows in porous media

The ability to numerically simulate single phase and multiphase flow of fluids in porous media is extremely important in developing an understanding of the complex phenomena governing the flow. The flow is complicated by the presence of heterogeneities in the reservoir and by phenomena such as diffusion, dispersion, and viscous fingering. These effects must be modeled by terms in coupled systems of nonlinear partial differential equations which form the basis of the simulator. The simulator must be able to handle both single and multiphase flows and the transition regimes between the two. A discussion of some of the aspects of modeling dispersion and viscous fingering is presented along with directions for future work.The partial differential equation models are convection-dominated and contain important local effects. An operator-splitting technique is used to address these different effects accurately. Convection is treated by time stepping along the characteristics of the associated pure convection problem, and diffusion is modeled via a Galerkin method for single phase flow and a Petrov-Galerkin technique for multiphase regimes. ELLAM (Eulerian-Lagrangian Localized Adjoint Methods) are discussed to effectively treat the advection-dominated processes. Accurate approximations of the fluid velocities needed in the Eulerian-Lagrangian time-stepping procedure are obtained by mixed finite element methods. Adaptive local grid refinement techniques are then indicated to resolve important local phenomena around wells and large heterogeneities or to resolve the moving internal boundary layers which often govern the mass transfer between phases.     

基于高斯过程回归的有限元应力解的改善研究

本文应用高斯过程回归方法对有限元应力解进行了改善研究．考题是一简化为平面应力问题的各向同性且受均布载荷的等截面悬臂深梁，应力考察量取Mises 应力，高斯积分点为样本点，单元角结点为改善点．4结点单元有限元模型和8 结点单元有限元模型的计算结果表明：(1)改善点的总体误差比样本点的总体误差都小，且4 结点明显、8 结点不明显；(2)边界结点的改善效果均较传统整体应力修匀的效果显著；(3)改善点应力具有置信区间；(4)较传统分片应力修匀方法，高斯过程回归方法可将所选取区域内的所有角结点的应力同时给予改善，且边界角结点改善效果好．     

Stress relaxation of PBI based membrane electrode assemblies

Stress relaxation in the membrane electrode assemblies (MEA) in PEM fuel cells subjected to compressive loads is analyzed. This behavior is important because nonzero contact stress is required to maintain low electric resistivity in the fuel cell stack. Experimental results are used to guide the choice of the viscoelastic properties of the constituents of the MEA, the membrane and the gas diffusion layer (GDL), needed for the model. These properties are incorporated into the model that treats the membrane as a porous-viscoelastic solid, and the gas diffusion layer as a nonlinear elastic solid. Using numerical simulations (finite element method), the stress relaxation curves for the MEA are obtained for different fluid flow boundary conditions, variations in the material properties of the membrane and the GDL. The results are compared to experimental stress relaxation curves. Most of the experimental data were obtained at a temperature of 180 °C, corresponding to operating conditions, so in the model the temperature was considered fixed and equal to this value.     

Patterned bilayer plate microstructures subjected to thermal loading: Deformation and stresses

We studied the deformation of a series of gold/polysilicon patterned plate microstructures fabricated by surface micromachining. The patterned plate microstructures were subjected to a uniform temperature change from 100 °C to room temperature that was intended to induce linear and geometrically nonlinear deformation. We used interferometry to measure full-field deformed shapes of the microstructures. From these measurements we determined the spatially-averaged curvature of the deformed microstructures within individual lines and across the entire plate. The deformation response of the patterned plates can be broadly characterized in terms of the average curvature as a function of temperature change and exhibits linear and geometrically nonlinear behavior. We modeled the deformation response of the patterned plates using geometrically nonlinear plate theory with the finite element method. Good agreement was obtained between predictions and measurements for both local curvature variations across lines and for the evolution of curvature of the entire plate with temperature change. Using a generalized plane strain approach with the finite element method we also modeled the spatial dependence of the stress distribution in the lines and substrate. For thick plates, our results agree with those of previous studies, showing a decrease in the von Mises stress in the metal lines with decreasing linewidth. For thinner substrates, though, we find the behavior with linewidth is opposite and there is a critical substrate thickness (about 10 μm for the system in our study) where the behavior with linewidth changes. These results have important implications in the design of patterned structures for micro-electro-mechanical systems (MEMS) applications where films are of comparable thickness to the underlying substrate.     

On using mesh-based and mesh-free methods in problems defined by Eringen’s non-local integral model: issues and remedies

With the recent success of nonlocal theories in modeling of engineering problems involving small intrinsic length scales, such as modeling of crack propagation, this paper addresses issues pertaining to cost-ineffectiveness of Eringen’s integral model. The cost effectiveness of the computation may be considered as a twofold issue; one pertaining to the non-local model and another pertaining to the numerical tool. First of all, we shall show that during the solution of problems with Eringen’s non-local integral model, there is no need to consider the integral model for the whole computational domain. In fact, the problems may be solved by just using the integral model close to the boundaries, i.e. a boundary layer effect, or around the points with singularities. In this paper we propose a partitioning strategy to remarkably reduce the computational cost. This may be considered as a gateway for solving some types of two-scale problems, e.g. those with macro/micro and nano scales, in which the small scale effects are localized just at parts of the domain. To demonstrate the efficiency of the numerical tools, we examine the performance of the finite element method (FEM), the element free Galerkin method (EFG) and the finite point method (FPM). This paves the way for using mesh-free methods in the solution of problems with non-local integral models. Examples with smooth and non-smooth solutions are considered for examining the efficiency of the methods. It will be shown that, by considering the boundary layer effect, the FEM and FPM will be efficient enough for being used in problems defined by Eringen’s non-local integral model.

     

Simultaneous heat and mass transfer in unsteady free convection from two-dimensional and axisymmetric bodies

The unsteady laminar free convection boundary layer flows around two-dimensional and axisymmetric bodies placed in an ambient fluid of infinite extent have been studied when the flow is driven by thermal buoyancy forces and buoyancy forces from species diffusion. The unsteadiness in the flow field is caused by both temperature and concentration at the wall which vary arbitrarily with time. The coupled nonlinear partial differential equations with three independent variables governing the flow have been solved numerically using an implicit finite-difference scheme in combination with the quasilinearization technique. Computations have been performed for a circular cylinder and a sphere. The skin friction, heat transfer and mass transfer are strongly dependent on the variation of the wall temperature and concentration with time. Also the skin friction and heat transfer increase or decrease as the buoyancy forces from species diffusion assist and oppose, respectively, the thermal buoyancy force, whereas the mass transfer rate is higher for small values of the ratio of the buoyancy parameters than for large values. The local heat and mass transfer rates are maximum at the stagnation point and they decrease progressively with increase of the angular position from the stagnation point.     

On the finite element analysis of woven fabric impact using multiscale modeling techniques

Finite element modeling of the impact of flexible woven fabrics using a yarn level architecture allows the capturing of complex projectile-fabric and yarn–yarn level interactions, however it requires very large computational resources. This paper presents a multiscale modeling technique to simulate the impact of flexible woven fabrics. This technique involves modeling the fabric using a yarn level architecture around the impact region and a homogenized or membrane type architecture at far field regions. The level of modeling resolution decreases with distance away from the impact zone. This results in a finite element model with much lower computational requirements. The yarns are modeled using both solid and shell finite elements. Impedances are matched across all interfaces created between the various regions of the model to prevent artificial reflections of the longitudinal strain waves. A systematic approach is presented to determine geometric and material parameters of the homogenized zone. The multiscale model is extensively validated against baseline models. The limitations of using shell elements to model the yarn level architecture underneath the projectile are addressed.     

Dynamic plasticity and fracture in high density polycrystals: constitutive modeling and numerical simulation

Presented is a constitutive framework for modeling the dynamic response of polycrystalline microstructures, posed in a thermodynamically consistent manner and accounting for finite deformation, strain rate dependence of flow stress, thermal softening, thermal expansion, heat conduction, and thermoelastic coupling. Assumptions of linear and square-root dependencies, respectively, of the stored energy and flow stresses upon the total dislocation density enable calculation of the time-dependent fraction of plastic work converted to heat energy. Fracture at grain boundary interfaces is represented explicitly by cohesive zone models. Dynamic finite element simulations demonstrate the influences of interfacial separation, random crystallographic orientation, and grain morphology on the high-rate tensile response of a realistic two-phase material system consisting of comparatively brittle pure tungsten (W) grains embedded in a more ductile matrix of tungsten-nickel iron (W-Ni-Fe) alloy. Aspects associated with constitutive modeling of damage and failure in the homogenized material system are discussed in light of the computational results.     

Multiscale modeling and simulation of single-crystal MgO through an atomistic field theory

The present work is concerned with the application of an atomistic-continuum field theory (AFT) in modeling and simulation of crystalline materials. Atomistic formulation of the field theory and its finite element implementation are introduced. Single-crystal MgO under mechanical loading is modeled and simulated. With a coarse mesh, the field theory is shown to be able to simulate dynamic and nonlinear behavior of multi-atom crystalline materials without the need of additional numerical treatments. Reducing the finite element mesh to the atomic scale, i.e., the finite element size is equal to the size of the primitive unit cell, atomic-scale critical phenomena, including dislocations nucleation and motion, have been successfully reproduced.     

ANALYSES ON NONLINEAR COUPLING OF MAGNETO-THERMO-ELASTICITY OF FERROMAGNETIC THIN SHELL-Ⅱ: FINITE ELEMENT MODELING AND APPLICATION

Based on the generalized variational principle of magneto-thermo-elasticity of a ferromagnetic thin shell established (see, Analyses on nonlinear coupling of magneto-thermo-elasticity of ferromagnetic thin shell-Ⅰ), the present paper developed a finite element modeling for the mechanical-magneto-thermal multi-field coupling of a ferromagnetic thin shell. The numerical modeling composes of finite element equations for three sub-systems of magnetic, thermal and deformation fields, as well as iterative methods for nonlinearities of the geometrical large-deflection and the multi-field coupling of the ferromagnetic shell. As examples, the numerical simulations on magneto-elastic behaviors of a ferromagnetic cylindrical shell in an applied magnetic field, and magneto-thermo-elastic behaviors of the shell in applied magnetic and thermal fields are carried out. The results are in good agreement with the experimental ones.     

Modeling dynamic plasticity and spall fracture in high density polycrystalline alloys

The dynamic thermomechanical response of a tungsten heavy alloy is investigated via modeling and numerical simulation. The material of study consists of relatively stiff pure tungsten grains embedded within a more ductile matrix alloy comprised of tungsten, nickel, and iron. Constitutive models implemented for each phase account for finite deformation, heat conduction, plastic anisotropy, strain-rate dependence of flow stress, thermal softening, and thermoelastic coupling. The potentially nonlinear volumetric response in tungsten at large pressures is addressed by a pressure-dependent effective bulk modulus. Our framework also provides a quantitative prediction of the total dislocation density, associated with cumulative strain hardening in each phase, and enables calculation of the fraction of plastic dissipation converted into heat energy. Cohesive failure models are employed to represent intergranular fracture at grain and phase boundaries. Dynamic finite element simulations illustrate the response of realistic volume elements of the polycrystalline microstructure subjected to compressive impact loadings, ultimately resulting in spallation of the material. The relative effects of mixed-mode interfacial failure criteria, thermally-dependent fracture strengths, and grain shapes and orientations upon spall behavior are weighed, with interfacial properties exerting a somewhat larger influence on the average pressure supported by the volume element than grain shapes and initial lattice orientations within the bulk material. Spatially resolved profiles of particle velocities at the free surfaces of the volume elements indicate the degree to which the incident and reflected stress waves are altered by the heterogeneous microstructure.     

Small scale model of linepipe

This paper discusses the technique of modeling pressurized pipe using small scale models. Pipes ofD/t ratio of around 64 were modeled using aluminum tubes of 44-mm diameter. Results from tests of small scale models show good agreement with those from large scale tests.     

Contact stresses: a short survey of models and methods of computations

Contact stresses are identified as normal and tangential forces between contacting solids. The normal stresses are modeled using unilateral and complementary conditions, elastic response and normal compliance. Friction laws describe the tangential traction. Friction of materials depends on pressure, sliding velocity, surface temperature, time of contact, surface roughness and presence of wear debris. Phenomenological, micro-mechanical and atomic-scale models as well as non-classical models of anisotropic and heterogeneous friction are important steps in the development of friction modeling. Sophisticated friction models are desirable in vibrating systems, materials processing, rolling contacts, rubber and polymers, geomechanics, bioengineering and living systems. Main numerical methods in contact mechanics are: finite element method, boundary element method and discrete element method. To include specific contact constraints, the following computing techniques are applied: Lagrange multipliers, penalty function, perturbated and augmented Lagrangian methods, mathematical programming methods. The advances of adhesion and impact modeling are outlined in this paper.     

基于Reynolds边界的滑动轴承动力学系数的计算及应用

运用有限元方法,在不需要额外求解Reynolds方程的情况下,求解了具有Reynolds边值条件的流体润滑问题,使得同时完成动力积分过程中非线性油膜力及影响Floquet乘子求解的油膜力Jacobian矩阵的计算成为可能;运用打靶法及预估-校正和打靶法相结合的延续算法考察了轴承-转子系统的非线性不平衡响应及其随轴承设计参数改变而出现的分岔现象,实现了计算量的有效减少。     

A problem for drug release from 2D polymeric systems

A nonlinear diffusion problem for drug release from 2D polymeric systems with finite dissolution rate is considered. The numerical solutions of such problems were obtained under some constraints in respect to the system geometry, the type of adsorption isotherm and concentration dependent diffusivity [Frenning, G., Stromme, M., 2003. Drug release modeled by dissolution, diffusion and immobilization. Int. J. Pharm. 250, 137–145; Frenning, G., Brohede, U., Stromme, M., 2005. Finite element analysis of the release of slowly dissolving drugs from cylindrical matrix systems. J. Control. Release 107, 320–329]. It is derived a numerical approach to solving a generalized problem, which avoids the above limitations. The proposed numerical scheme based on finite element domain approximation and time difference method is used for simulation of 2D drug release under various model parameters. The effects of drug adsorption and concentration dependent drug diffusivity are demonstrated.     

Biothermomechanics of skin tissues

Biothermomechanics of skin is highly interdisciplinary involving bioheat transfer, burn damage, biomechanics and neurophysiology. During heating, thermally induced mechanical stress arises due to the thermal denaturation of collagen, resulting in macroscale shrinkage. Thus, the strain, stress, temperature and thermal pain/damage are highly correlated; in other words, the problem is fully coupled. The aim of this study is to develop a computational approach to examine the heat transfer process and the heat-induced mechanical response, so that the differences among the clinically applied heating modalities can be quantified. Exact solutions for temperature, thermal damage and thermal stress for a single-layer skin model were first derived for different boundary conditions. For multilayer models, numerical simulations using the finite difference method (FDM) and finite element method (FEM) were used to analyze the temperature, burn damage and thermal stress distributions in the skin tissue. The results showed that the thermomechanical behavior of skin tissue is very complex: blood perfusion has little effect on thermal damage but large influence on skin temperature distribution, which, in turn, influences significantly the resulting thermal stress field; the stratum corneum layer, although very thin, has a large effect on the thermomechanical behavior of skin, suggesting that it should be properly accounted for in the modeling of skin thermal stresses; the stress caused by non-uniform temperature distribution in the skin may also contribute to the thermal pain sensation.