基于四叉树直角坐标网格的运动界面追踪方法

运动界面的追踪是两相流模拟中需要解决的主要问题之一。基于四叉树直角坐标网格实现了Level set界面追踪方法和网格的局部自适应控制。通过模拟3个给定速度场(平移流场、旋转流场、剪切流场)的运动界面追踪问题,证明了自适应四叉树直角坐标网格与Level set方法结合的优势。     

基于八叉树自适应网格技术的LevelSet运动界面追踪方法

Level Set方法因能有效地处理界面处复杂的拓扑结构变化以及大变形问题,广泛应用于界面追踪领域。在Level Set方法追踪运动界面时引入八叉树网格技术,通过八叉树网格的细化和粗化技术减少计算网格数量和计算内存并提高计算效率和计算精度。因为八叉树网格为非均匀网格,其相邻网格的层数值可能不相同,所以不能直接采用WENO格式离散Level Set函数得到网格处的函数值,进而提出八叉树网格离散模型解决这一问题,并提出基于八叉树网格距离场重新初始化方法减少Level Set方法的质量损失,最后将基于八叉树网格技术的Level Set方法应用于两个给定速度场的运动界面模拟算例以及基准件方腔的铸造充型过程的模拟。模拟结果表明该方法可以提高界面的精度,同时改善质量守恒性。     

基于八叉树自适应网格技术的Level Set运动界面追踪方法

Level Set方法因能有效地处理界面处复杂的拓扑结构变化以及大变形问题,广泛应用于界面追踪领域。在Level Set方法追踪运动界面时引入八叉树网格技术,通过八叉树网格的细化和粗化技术减少计算网格数量和计算内存并提高计算效率和计算精度。因为八叉树网格为非均匀网格,其相邻网格的层数值可能不相同,所以不能直接采用WENO格式离散Level Set函数得到网格处的函数值,进而提出八叉树网格离散模型解决这一问题,并提出基于八叉树网格距离场重新初始化方法减少Level Set方法的质量损失,最后将基于八叉树网格技术的Level Set方法应用于两个给定速度场的运动界面模拟算例以及基准件方腔的铸造充型过程的模拟。模拟结果表明该方法可以提高界面的精度,同时改善质量守恒性。     

三维熔体前沿界面的Level Set追踪

给出三维Level Set方程,采用五阶加权本质无振荡格式进行空间离散,通过算例验证了该算法的正确性及追踪三维运动界面的准确性。进而将Level Set算法和同位网格有限体积法进行耦合,模拟了注塑成型充填阶段的三维流动过程,准确追踪到了不同时刻熔体前沿界面,预测并分析了流动过程中不同时刻的压力、速度等重要流动特征。数值结果表明,该方法可追踪三维熔体前沿界面,预测充填过程中的重要流动特征。     

Pseudo-Anosov映射作用下拓扑混沌数值模拟

对方腔内3个方形搅拌轴两两交替运动引发的Stokes流动的混合问题进行了数值模拟研究。提出了两种搅拌轴方形运动路径来引发相空间内Pseudo-Anosov（pA）映射。采用有限体积方法求解速度场,搅拌轴的周期性速度边界借助叠加网格技术来实现。流动区域采用交错网格划分,控制方程组采用具有二阶精度的中心差分格式离散。粒子运动前锋追踪计算采用具有四阶精度的Runge-Kutta方法实现。得到了Poincaré截面,表明pA起作用的空间尺度几乎覆盖整个方腔,只有4个角区域除外。计算了不同初始位置的示踪剂演化图像来表征混合过程。示踪剂增长将经历指数增长,其增长指数大于pA映射矩阵的预测值,这是由于搅拌轴形状、运动路径等流场细节导致的示踪剂界面局部二次折叠拉伸造成的。     

气辅注射成型运动界面的Level Set方法数值模拟

给出气液两相流数学模型,选取Cross-WLF模型作为熔体的黏度模型,采用Level Set/SIMPLEC方法模拟了气体辅助注射成型中气体穿透过程,追踪到了不同时刻的运动界面(气熔界面和熔体前沿界面),描述了运动过程中不同时刻速度和温度等重要物理量的分布情况,分析了熔体温度、气体延迟时间和注射压力对气体穿透时间和穿透长度的影响。数值结果表明,Level Set/SIMPLEC方法可以准确追踪气体穿透过程中的两个运动界面;熔体温度、延迟时间和气体注射压力对气体穿透长度有显著影响。     

熔体充模过程动态模拟及流场分析

将Level Set／Ghost方法应用于聚合物成型研究，实现了非等温情况下注射成型聚合物熔体充模阶段的动态模拟；得到了正确的流线分布和不同时刻的温度、压力等值线分布。对Level Set／Ghost方程的求解，空间方向采用高分辨率、稳定且无振荡的5WENO（the fifth—order weighted essentially non—oscillatory）格式进行离散，时间方向采用稳定的TVD-RK（total variation diminishing Runge Kutta）方法进行离散。物理量控制方程采用一般的有限差分格式进行数值求解。结果表明，Level Set／Ghost方法可以准确追踪非等温聚合物熔体前沿界面的位置，并能精确描述前沿界面的形状，同时可以实现动态流场物理量的准确模拟。     

水平集方法数值模拟单液滴的生成过程(英文)

液滴生成的动力学过程和机理,对溶剂萃取、喷墨打印机设计、飞行器保护等有重要意义。虽然已有很多相关的实验、理论和计算研究,但对涉及拓扑变形界面的该问题研究仍富有挑战性,常用的有限元等模拟方法还有待改进。1988年提出的水平集(Level set)方法,可以比较简方便地模拟多相变形界面问题,已成功用于多相流、结晶、浇铸、模式识别等过程的计算。作者在对水平集方法进行改进并用于相间传质模拟的基础上,本文数值模拟了单个液滴在毛细管口的生长和脱离过程。假设流体均为不可压缩液体或气体作层流流动,采用二维轴对称欧拉坐标和交错网格,利用控制容积法和SIMPLE算法离散求解藕合水平集函数的运动方程组。水平集函数的发展方程和重新初始化方程的空间与时间离散格式,分别采用5阶WENO(Weighted essentially nonoscillatory)和3阶TVD(Total variation diminishing)Runge-Kutta格式。初步计算了不同参数的单液滴在气体或另一不互溶液体中的生成过程,液滴的拓扑形状、两相流场和液滴大小与实验或文献报道的模拟结果基本吻合,为今后拓展模拟三维空间的不稳定运动液滴或气泡生成过程的流动和传递奠定了基础。     

气泡在不互溶两相流体界面运动行为的数值模拟

基于水平集法(Level set method),提出了一种处理三相流问题的计算方法.计算了钢-渣-气三相流条件下气泡在钢渣界面的行为,考察了不同大小的气泡穿过钢渣界面时的拓扑变化以及导致钢渣界面流场的变化,初步分析了大气泡上浮对卷渣的影响.计算结果表明,气泡直径从0.1~3mm,上浮过程中变形程度随直径的增大而愈发剧烈,钢渣界面处的流场差异也随之增大,可能成为卷渣的原因之一.     

三维非等温非牛顿流体充模过程的建模与模拟

为了避免使用速度延拓法处理自由表面边界条件,更真实地反映非牛顿熔体的充模过程,耦合Navier-Stokes方程组,建立了三维实体气液两相流模型。采用基于SIMPLE算法的同位网格有限体积法求解流场物理量,并结合Level Set界面追踪技术实现了非等温熔体充模阶段的动态模拟。为防止出现三维情形下棋盘型压力场,利用同位网格动量插值技术,解决压力与速度失耦的问题。分析了充模过程中,流场物理量的变化情况,研究了注射速率、入口温度等对充模过程及熔体凝固层厚度的影响,得出了与实验相吻合的模拟结果。数值结果表明,三维实体气液两相流模型能更准确地描述聚合物熔体的充模过程;当注射速率较大、入口温度较高时,充模时间较短,熔体凝固层的厚度较小。     

An Efficient Calculation of Multidimensional Freeze-Drying Problems Using Fixed Grid Method

Abstract

A calculation procedure based on a fixed grid method streamlined for multidimensional freeze-drying problems was developed, considering coupled heat and mass transfer, sublimation of ice, and motion of ice front. Efficiency of the fixed grid calculation in solving one- and two-dimensional freeze-drying problems was evaluated by comparing its accuracy and calculation cost with those of a conventional moving grid calculation. The numerical investigation showed that the fixed grid method can produce accurate results comparable to those obtained by the moving grid method, with almost equal calculation cost in case of one-dimensional freeze-drying problems. But the fixed grid method required about 70% more computation time than the moving grid method in case of two-dimensional freeze-drying problems. It is thought that the fixed grid method may be advantageous over conventional moving grid methods when freeze-drying problems involving cumbersome moving interface geometries have to be considered.     

An Efficient Calculation of Multidimensional Freeze-Drying Problems Using Fixed Grid Method

A calculation procedure based on a fixed grid method streamlined for multidimensional freeze-drying problems was developed, considering coupled heat and mass transfer, sublimation of ice, and motion of ice front. Efficiency of the fixed grid calculation in solving one- and two-dimensional freeze-drying problems was evaluated by comparing its accuracy and calculation cost with those of a conventional moving grid calculation. The numerical investigation showed that the fixed grid method can produce accurate results comparable to those obtained by the moving grid method, with almost equal calculation cost in case of one-dimensional freeze-drying problems. But the fixed grid method required about 70% more computation time than the moving grid method in case of two-dimensional freeze-drying problems. It is thought that the fixed grid method may be advantageous over conventional moving grid methods when freeze-drying problems involving cumbersome moving interface geometries have to be considered.     

模拟变形液滴和气泡运动的改进水平集算法

An improved level set approach for computing the incompressible two-phase flow with significantly deformed free interface in presented.The control volume formulation with the semi-implicit method for pressure-linked equations consistent(SIMPLEC)algorithm incroporated is used to solve the governing equations on a staggered grid.Several improvements concerning the convergence and numerical stability.The motion of a bubble or drop in a liquid with large density ratio,viscosity ratio and surface tension in numerically simulated.The computational results are in good agreement with the reported experimental data.     

Moving mesh generation for tracking a shock or steep moving front

For tracking a shock or steep moving front in the numerical solution of Partial Differential Algebraic Equations (PDAEs), an accurate spatial discretization method, Weighted Essentially Non-Oscillatory (WENO) scheme, is combined with moving grid techniques so that spacing of moving meshes is smoothed locally and globally. Several monitor functions, as metric criteria of node concentration, are examined. While the fixed grid method (uniform grid size) needs many mesh points to obtain enough solution accuracy, the moving grid method (non-uniform grid size) enhances accuracy even at small mesh numbers but it may be prohibitive owing to the addition of complex and non-linear mesh equations into physical PDAEs. The combination of the WENO scheme (based on an adaptive stencil idea) with the moving grid techniques improves stability and accuracy in the numerical solution over the commonly used moving grid method of central discretization. To locate adequate grid position in the moving mesh method, suitable monitor function according to problems must be selected.     

Numerical simulation of the sequential coinjection molding process based on level set method

The sequential coinjection molding (SCIM) process has always been regarded as a challenging multiphase flow problem, which includes skin and core polymer melts together with gas. Thus, this article presents a 3D mathematical model for it, in which the governing equations of gas, skin, and core melts are united into a system namely the generalized Navier‐Stokes equations. By doing this, the model can be solved simply by applying the finite volume SIMPLE method on the collocated grid. The core penetration process is simulated by level set method, which can capture two different types of moving interface simultaneously at different time. By simulating the SCIM process of a centrally gated rectangular plate and then comparing the numerical results with available experiment results, the proposed mathematical model is validated and the influences of skin/core volume ratio, injection temperature, and core injection delay on the depth of core penetration are analyzed in detail. Then, the sequential coinjection of a line‐gated plate is investigated numerically, obtaining some important information in the gap‐wise direction, which cannot be caught by 2.5D model. All the numerical results show that the multiphase flow model proposed in this article is effective and can be used to describe the flow behaviors of polymer melt in the SCIM process. POLYM. ENG. SCI. 55:1707–1719, 2015. © 2014 Society of Plastics Engineers     

Study of the effects of heat removal on the copying accuracy of the electrochemical machining process

The ECM process with a moving cathode tool is simulated with the aim to study the influence of the temperature on the uniformity or copying quality of the removal rate. Two temperature dependent electrode reaction models are presented and used for solving the charge conservation and internal energy balance equations, in combination with the Navier–Stokes equations for viscous incompressible flow. The level set method is used for tracking the anode interface. The heat conduction through electrodes and the heat production by the electrode reactions are found to play an important role.     

SCR反应器入口流场均匀性模拟研究

采用Fluent软件对SCR反应器的入口部分流场进行了模拟,分析了不同导流板形式对烟气速度分布的影响。对整流格栅的画法进行了研究,分别对比了采用多孔介质模型和直接画出整流格栅两种方式,发现采用多孔介质模型所得模拟结果与实际物理场有偏差,最后选定直接画出整流格栅的方式进行模拟。对比了在流道转弯处设置不同导流板形式对流场的影响,发现当转弯处设置10块弧形等长导流板时流场分布最为均匀。