Molecular-Dynamics Method for the Simulation of Grain-Boundary Migration

A molecular-dynamics method for the simulation of the intrinsicmigration behavior of individual, flat grain boundaries is introducedand validated. A constant driving force for grain-boundary migrationis generated by imposing an anisotropic elastic strain on a bicrystalsuch that the elastic-energy densities in its two halves aredifferent. For the model case of a large-planar-unit-cell, high-angle(001) twist boundary in Cu we show that an elastic strain of1%–4% is sufficient to drive thecontinuous, viscous movement of the boundary at temperatures wellbelow the melting point. The driving forces thus generated (at thehigh end of the experimentally accessible range) enable aquantitative evaluation of the migration process during the timeframe of 10^-9 s typically accessible bymolecular-dynamics simulation. For this model high-angle grainboundary we demonstrate that (a) the drift velocity is, indeed,proportional to the applied driving force thus enabling us todetermine the boundary mobility, (b) the activation energy forgrain-boundary migration is distinctly lower than that forgrain-boundary self-diffusion or even self-diffusion in the melt and(c) in agreement with earlier simulations the migration mechanisminvolves the collective reshuffling during local disordering(melting) of small groups of atoms and subsequentresolidification onto the other crystal.     

界面捕捉Level Set方法的(AMR)数值模拟

在流体力学方程的计算中采用高精度WENO格式,用AMR(adaptive mesh refinement)方法提高流场局部分辨率,在采用Level Set函数标定物质界面的计算中用GFM(ghost fluid method)方法进行界面处理,尝试将AMR技术与界面追踪技术相互融合并应用于数值模拟,对不同的模拟结果进行了比较.     

沟槽结构飞片的波剖面测量与数值模拟研究

应用线、面结合激光速度干涉仪(VISAR)技术,对点爆散心爆轰产物驱动带沟槽飞片的自由面速度进行了实验测量,获得了飞片表面长12mm的一条线上点的速度-时间历程,以及飞片起跳后1、3、5μs 3个特定时刻飞片表面10 mm范围内速度的二维分布。同时用LS-DYNA对实验模型进行了数值模拟,所得结果与实验测量结果符合较好,表现出线、面结合VISAR技术在研究非均匀材料动态响应上的应用潜力。     

Molecular-Dynamics Simulation of Grain-Boundary Diffusion Creep

Molecular-dynamics (MD) simulations are used, for the first time, to study grain-boundary diffusion creep of a model polycrystalline silicon microstructure. Our fully dense model microstructures, with a grain size of up to 7.5 nm, were grown by MD simulations of a melt into which small, randomly oriented crystalline seeds were inserted. In order to prevent grain growth and thus to enable steady-state diffusion creep to be observed on a time scale accessible to MD simulations (of typically 10^-9s), our input microstructures were tailored to (i) have a uniform grain shape and a uniform grain size of nm dimensions and (ii) contain only high-energy grain boundaries which are known to exhibit rather fast, liquid-like self-diffusion. Our simulations reveal that under relatively high tensile stresses these microstructures, indeed, exhibit steady-state diffusion creep that is homogenous (i.e., involving no grain sliding), with a strain rate that agrees quantitatively with that given by the Coble-creep formula.     

Numerical Simulation of Physical Fields by the Collocation Method

Russian Physics Journal - The article describes the collocation method for the numerical solution of mathematical physics boundary value problems. By arranging the collocation nodes in a special...     

改进粒子水平集法及其应用

针对传统粒子水平集方法的缺陷提出一种虚拟粒子对距离函数的修正方法.改进算法针对虚拟粒子与待修正网格点的各种相对位置进行误差修正,可以避免虚拟粒子在非界面法线方向移动而产生的误差.数值求解涡旋流场问题显示:改进算法的界面捕捉性能得到显著提高.在此基础上,结合投影法求解Navier-Stokes方程组,数值模拟存在高密度比(10³量级)及粘性比(10²量级)的不可压缩两相流问题(气泡上升及自由面不稳定性问题).计算结果与已有结果吻合良好,详细分析主要控制参数对瑞利-泰勒不稳定性现象的影响.     

非结构网格上Level Set方程的间断有限元解法

针对间断有限元弱形式难于求解可压缩流场中Level Set方程的问题提出间断有限元强形式,从而在统-框架下解决Level Set方程在可压缩与不可压缩流场中的求解问题.通过非结构网格上采用Legendre-Gauss-Lobatto节点构造基函数,在复杂区域上可以达到任意高阶的精度.将若干-、二、三维算例与已有文献或解析解比较,验证方法追踪自由界面的有效性.结果表明,该方法适合各种情形下Level Set方程求解,易于在复杂区域的非结构网格上实施,精度高、分辨率高且具有高质量守恒性,既能避免重新初始化过程又方便向高维扩展.     

基于水平集的两步MCMC方法对河道形态的识别

首先运用符号距离函数刻画河道的复杂几何形态，然后通过求解水平集演化方程和两步马尔科夫链蒙特卡罗（MCMC）算法拟合生产历史数据，逐步更新河道的边界.在两步MCMC方法的第一步，应用流线模拟计算的敏感性矩阵获取近似的似然函数，修改MCMC的推荐概率分布；第二步，为确保MCMC算法的严密性，对通过第一步的油藏模型进行完整的数值模拟以获取精准的似然函数，并用更改的接受概率作为模型接受的判断准则.最后通过二维计算实例验证该方法的有效性.     

基于改进水平集方法的翼型拓扑优化

将改进的水平集方法与灵敏度分析方法相结合,发展了一种研究Navier-Stokes方程拓扑优化的新方法。在拓扑优化过程中,引入无需样条参数化的网格重新划分方法,实现水平集函数演化和流场求解的结合。最后,通过对低雷诺数条件下Navier-Stokes翼型优化问题的研究,表明该方法具有良好的全局性,并能有效节约优化问题的计算工作量。     

Numerical Simulation of Rogue Waves by the Local Discontinuous Galerkin Method

We study rogue waves described by nonlinear Schr6dinger equations. Such wave solutions are so different from conventional soliton solutions that classic methods such as the Crank-Nicolson scheme cannot work for these cases. Fortunately, we find that the local discontinuous Galerkin method equipped with Dirichlet boundary conditions can simulate rogue waves very well. Several numerical examples are presented to show such interesting wave solutions.     

Numerical Simulation of Two-Dimensional Flow over Three Cylinders by LatticeBoltzmann Method

The numerical simulation using the multiple relaxation time lattice Boltzmann method (MRT-LBM) is carried out for the purpose of investigating the two-dimensional flow around three circular cylinders. Among these three circular cylinders, one of the three cylinders on which a forced in-line vibrating is used to do this research and attempt to find out the effects of the movingcylinder and the other two rigid cylinders on the wake characteristics and vortex formation. As a benchmark problem to discuss the problem of lift coefficient r.m.s for these cylinders with spacing ratios T/D between other rigid side-by-side cylinders, and the calculation is carried out with two compared cases at Reynolds number of 100, two of the cylinders are rigid and the other one is an in-line vibrated cylinder lying downstream, in addition, forced vibrating amplitude and frequency are A/D= 0.5 and f_ν=0.4 (where A is the forced amplitude, D is the cylinder diameter, and f_ν stands for the vibrating frequency, respectively). The calculated results not onlyindicate that the spacing ratios T/D (T is the center-to-center spacing between the two upstream cylinders) have influence on the wake patterns and the formation of vortex shedding, but also analyze the lift coefficient r.m.s for the three cylinders with the spacing ratios S/D (where S is the center-to-center spacing between the center of upstream two side-by-side cylinders and downstream cylinder).     

颗粒碰撞的直接模拟算法

为研究细微颗粒间平均碰撞率,构建了直接模拟三维空间内颗粒碰撞的数值方法.对剪切流内夹带的无惯性颗粒和自由运动颗粒的碰撞过程分别进行了数值模拟,得到的颗粒平均碰撞率的模拟值与理论值的相对误差小于1%.在碰撞颗粒对的搜索过程中,引入网络排除法等优化措施,使计算量降低两个量级以上,而不增大相对误差.采用不同的Δt模拟同一颗粒群的运动,发现较小的时间步长可以避免因为截断误差增大的相对误差.在不遗失颗粒碰撞的条件下,较密的计算网格可明显减少计算量.     

一种保持界面位置不动的水平集函数隐式重新初始化方法

以保持界面不动为主要目标,推导水平集函数重新初始化方程中光滑参数的取值公式,得到一种新型的隐式重新初始化方法.证明应用该方法进行重新初始化时,保证界面附近水平集函数节点值符号不变时间步长只需满足原始的CFL条件即可.最后,将该方法与原有初始化方法结合,得到一种准确且快速的隐式重新初始化方法.用数值算例验证该方法的有效性.     

重新初始化的LS方法跟踪二维可压缩多介质流界面运动

 用非耦合求解方法计算Level Set函数方程与流体力学方程组，应用重新初始化的Level Set函数确保距离函数性质，流体力学方程组采用二阶精度多介质流波传播差分格式计算，重新初始化方程采用五阶WENO格式计算。并给出了二维可压缩多介质流界面运动的计算结果。     

Numerical Simulation of Plasma Antenna with FDTD Method

We adopt cylindrical-coordinate FDTD algorithm to simulate and analyse a 0.4-m-long column configuration plasma antenna. FDTD method is useful for solving electromagnetic problems, especially when wave characteristics and plasma properties are self-consistently related to each other. Focus on the frequency from 75 MHz to 400 MHz, the input impedance and radiation efficiency of plasma antennas are computed. Numerical results show that, different from copper antenna, the chaxacteristics of plasma antenna vary simultaneously with plasma frequency and collision frequency. The property can be used to construct dynamically reconfigurable antenna. The investigation is meaningful and instructional for the optimization of plasma antenna design.     

SPH方法在液固撞击数值模拟中的应用

对液滴分别采用光滑粒子流体动力学(SPH)和任意拉格朗日-欧拉(ALE)两种方法模拟液滴高速冲击有机玻璃(PMMA)过程中液滴的变形、固体结构的损伤.结果表明:两种算法中结构表面的损伤情况与Brunton实验数据吻合,证明了数值分析方法的可行性和精确性;此外,对比发现SPH方法模拟液固撞击过程更具有优势.利用SPH方法分析实际工程问题,验证了两种工程处理技术可以有效地提高螺旋桨抗雨水的冲击能力.     

二维光子晶体中的无序数值模拟研究

应用on-shell方法研究弱无序情况下的二维光子晶体，无序的引入使得原来不透电磁波的禁带中，出现了共振模，即某一波长的电磁波可以透过，计算结果表明共振模的出现位置，条数多少及透过率随着无序程度的不同而变化。     

高斯光束的数值模拟新方法

基于高斯光束的单叶双曲面特征,给出了一种新的模拟高斯光束的数值方法.该方法将高斯光束视为由无数个共轴单叶双曲面构成的双曲线体,由束腰截面发射的光子传播方向是在经过该光子发射点的双曲面内两条直母线中等概率选择.这不仅符合光子沿直线传播,而且意味着大量光子的“平均”传播方向与相应高斯光束的波印廷矢量相同.数值模拟结果表明：该方法可高准确度地模拟高斯光束的几何构形、光强分布、光子的光程分布和波前曲率半径,而且模拟效率与现有的方法相近.