凝固模拟中有限元网格向有限差分网格的转换

如果能将有限元网格转换成为有限差分网格,就可以利用现有的商业软件进行网格剖分,然后进行网格转化,再应用到开发的温度场模拟软件了.在此方面做了一些探索工作,以解决凝固模拟中有限元网格向有限差分网格转换的问题.     

基于有限差分网格的铸件热应力分析

提出了有限差分网格向有限元网格转换的方法,进而在该有限元模型基础上采用FDM/FEM集成铸件热应力分析系统实现了铸件的热应力分析。采用典型的应力框试件,将该方法计算结果和直接剖分得到的有限元网格基础上的计算结果进行了对比,结果表明该方法是可行的。     

基于STL的射线穿透法网格剖分的研究

介绍基于STL的射线穿透法来剖分析有限差分网格的基本思路和流程;开发了网格自动剖分软件;实际运用结果表明,采用射线穿透法对记录实体信息的STL文件进行剖分是1种切实可行的办法。     

Auto CAD下的铸件三维网格自动剖分

本文以Ａｕｔｏ ＣＡＤ为图形支撑平台，利用截面组合法，结合Ａｕｔｏ ＣＡＤ的图形接口文件ＤＸＦ开发了铸件有限差分三维网格自动剖分软件。     

用ADS开发铸件实体网格剖分应用程序

在用AutoCADR12进行铸件实体几何造型基础上，用AutoCAD语言开发系统ADS开发了网格剖分应用程序，可实现对任意复杂形状铸件的有限差分网格剖分，具有强的实用性。     

用ADS开发铸件实体网格部分应用程序

在用ＡｕｔｏＣＡＤＲ１２进行铸件实体几何造型基础上，用ＡｕｔｏＣＡＤ的Ｃ语言开发系统ＡＤＳ开发了网格剖分应用程序，可实现对任意复杂形状铸件有限差分网格剖分，具有强的实用性。     

基于有限差分网格的发动机缸体铸件热应力模拟

对发动机气缸体铸件从浇注完到开箱落砂后的铸造过程应力场进行计算机数值模拟.为了实现形状复杂铸件的热应力分析,应用了有限差分网格向有限元网格的转换方法,得到缸体的有限元模型,并将该模型与经过大量简化后直接剖分得到的有限元模型进行了比较.在此有限元模型的基础上,采用FDM/FEM集成铸件热应力分析系统实现了对发动机缸体铸件热应力分析,模拟结果和实际生产情况相吻合.     

基于IDEAS的三维有限差分网格自动剖分

在分析ＩＤＥＡＳ数据结构的基础上,采用端点排序和扫描线填充算法开发了适用于任意复杂形体的有限差分网格自动剖分程序。通过实际应用证明了程序的实用性。     

有限差分问题中的变网格技术

本文针对有限差分问题中的均匀网格技术存在的弊端提出了变网格技术,并提出了区分粗细网格的假设与原则,给出了区分粗细网格在算法上实现的流程图。     

两种不同约束条件下发动机缸体铸件热应力场的数值模拟

采用FDM/FEM集成热应力分析系统(其中采用铸造之星FT-STAR进行温度场计算,ANSYS进行热应力场计算,采用FT-STRESS进行有限差分网格向有限元网格转换和有限差分/有限元温度载荷转换),对某厂柴油机发动机缸体铸件进行从浇注到冷却至室温全过程的温度场、热应力场数值模拟,得到冷却变形情况及残余应力分布,并研究比较了将气缸处砂芯考虑为部分刚性,和将砂芯考虑为完全退让性的两种不同约束模拟方案对计算结果的影响。前者应力值和变形值远远大于后者的结果。     

Grain boundary finite length faceting

We study symmetrical and asymmetrical aluminium grain boundary faceting with molecular dynamics simulations employing two embedded atom method potentials. Facet formation, coarsening, and the reversible phase transition of $\sum 3\{110\}$ boundary into $\sum 3\{112\}$ twin, and vice versa, are demonstrated in the simulations and the results are consistent with earlier experimental studies and theoretical models. The $\sum 11\{002{\}}_1/\{667{\}}_2$ boundary shows faceting into $\{225{\}}_1/\{441{\}}_2$ and $\{667{\}}_1/\{001{\}}_2$ boundaries and coarsens with a slower rate when compared to $\sum 3\{112\}$ facets. However, facets formed by $\{111{\}}_1/\{112{\}}_2$ and $\{001{\}}_1/\{110{\}}_2$ boundaries from a $\{116{\}}_1/\{662{\}}_2$ boundary are stable against finite temperature annealing. In the above faceted boundary, elastic strain energy induced by atomic mismatch across the boundary creates barriers to facet coarsening. Grain boundary tension is too small to stabilize the finite length faceting in both $\sum 3\{112\}$ twin and asymmetrical $\{111{\}}_1/\{112{\}}_2$ and $\{001{\}}_1/\{110{\}}_2$ facets. The observed finite facet sizes are dictated by facet coarsening kinetics which can be strongly retarded by deep local energy minima associated with atomic matching across the boundary.     

Computational analysis of power frequency devices by a novel hybrid quasi-static finite difference–finite element technique

A hybrid explicit numerical methodology for the analysis of power frequency devices, characterized by large differences in permeability and conductivity values, is presented. It is based on the separation of the physical problem into two distinct ones according to the properties of the materials comprising the structure under investigation. A finite difference time-advancing scheme, specially modified for eddy-current analysis, is implemented in magnetic/conducting media, whereas in non-magnetic/non-conducting regions, like air, a finite element procedure is utilized. The two techniques are separately implemented and combined via appropriate interface boundary conditions into an efficient, systematic time domain procedure. The derived hybrid scheme is proved to be ideal for 2D power frequency problems.     

Spectral analysis of surface waves in single and multi-layer slabs with finite thickness using finite element modelling

This paper presents the numerical modelling of spectral analysis of surface waves (SASW) method to find the surface wave velocities and to investigate if the numerical modelling of a multi-layer slab system can show the dispersive characteristics. In the modelling of a single-layer slab, the average surface wave velocities obtained from the finite element mode superposition analysis are reasonably close to the experimental value. The refinement of the finite element mesh from element size of 25.4×25.4 mm to 12.7×12.7 mm improves the surface wave velocity results. The numerical modelling of a three-layer slab shows that the dispersive characteristics of surface wave exist in the layer system and can be modelled by the finite element analysis. This study further advances the application of the SASW method to the assessment of existing concrete slab structures.     

Temperatures in deep grinding of finite workpieces

This paper investigates the diverse thermal effects generated in high efficiency deep grinding (HEDG). Using a new thermal model of circular arc contact with transient analysis, the transient behaviour of the maximum contact temperature has been analysed for various grinding conditions. It is found that steady state conditions can be achieved for the conditions of sufficient workpiece length and high workspeeds. The effect is important for the understanding of the deep grinding process and for the prediction of satisfactory grinding conditions. HEDG conditions also have very apparent effects on the depth of heat penetration to the workpiece. The parameters investigated include mean contact angle, Peclet number and the heat source distribution. Experimental results are presented for specific energy, energy partition and mean temperature for high efficiency deep grinding.     

Theory of finite amplitude machine tool instability

A non-linear theory of machine tool chatter is presented. It is shown that even when the machine tool structure is linear or only slightly non-linear (as appears to be the case), large non-linearity is introduced by two causes. Firstly, by the chatter amplitudes exceeding a certain value, dependent on the mean chip thickness and the vibrating tool leaving the workpiece. Secondly, by a non-linearity of the cutting force characteristics.By combining these non-linear effects with the linear operative receptance of the machine tool structure, non-linear stability conditions are derived which describe a surface in the three-dimensional space defined by rotational speed, depth of cut and chatter amplitude. Sections through this surface provide the traditional (linear) stability chart relating depth of cut and rotational speed, as well as non-linear charts relating chatter amplitude with depth of cut. The latter are used for discussing the mechanism of stabilisation of chatter amplitudes and the phenomenon of finite amplitude instability. It is shown that the physical cause of finite amplitude instability lies in the non-linearity of the cutting force characteristic.The theory presented leads to conclusions of practical significance. The most important of these is that when the cutting force characteristic is non-linear the cutting process may be stabilised by a large increase of the mean chip thickness.Theoretical work is supported by experiments and good correlation between these is noted.     

基于有限元逆向法和正向法的板料成形过程中合理毛坯形状的确定

合理的毛坯形状可以提高板料成形质量,降低成本.有限元逆向法确定简单形状零件的毛坯具有很高的可靠性,但不适合于单独用来确定带有拉深筋及切边余量大的覆盖件拉深成形板料毛坯.本文采用有限元逆向法和正向法结合的方法,确定覆盖件零件的成形毛坯,并用物理实验验证了其合理性.     

一种快速有限元网格的生成方法

借助汽车后大梁零件回弹现场调试,利用CAE软件(Dynaform)与逆向工程软件(Imageware)之间的数据接口,介绍了直接将扫描点云引入CAE软件内进行网格处理的方法与技巧,节省了大量的传统建模时间,在缩短模具生产周期的同时也降低了调试成本。     

利用有限元仿真技术优化模具结构

建立起精锻件模具结构优化的数学评估模型,结合有限元仿真技术对不同模具结构参数(凹模入口角、工作带长度、锥形角等)下的金属成形过程进行分析,找出最佳的模具结构参数组合,从而达到实现模具结构优化的目的。最后,本文还以汽车球头销锻件为例,对球头预成形工序锥形角大小的控制进行分析,证实该方法的科学性和有效性。     

渗碳问题有限元模拟收敛性分析

采用Laplace变换法对一维渗碳问题进行了理论求解,在此基础上使用SYSWORLD有限元分析软件进行了渗碳问题有限元模拟收敛性分析。本文用DH表示渗碳影响距离。结果表明,对于一维渗碳问题,以理论解为参考,网格尺寸小于DH/3时,模拟计算值较精确;对于二维渗碳问题,四边形网格的计算精度较三角形网格高些,网格尺寸应小于DH/4;三维渗碳模拟时,网格尺寸应小于DH/5。