高质量点集的快速局部网格生成算法

高效及高质量的局部网格生成算法是基于节点有限元并行方法设计的关键。泡泡布点算法能够在复杂区域上不经过人工干预生成高质量的节点集,本文提出了基于该方法所生成的节点集的快速局部网格生成算法。该算法充分利用泡泡布点方法提供的节点集及节点邻接链表信息,避免了桶数据结构的建立以及节点的局部搜索过程,只需应用Delaunay三角剖分的外接圆准则从中心节点的邻接链表中去除极少数的非卫星点,可快速地生成局部网格,比现有的局部网格生成算法更为快捷。算例结果表明,该算法高效可靠,生成网格与Delaunay三角剖分网格一致。     

一种全四边形网格生成方法——改进模板法

首先对全四边形单元网格自动剖分算法中的模板法进行了探讨，并提出了相应的改进方法。在此基础上提出了一种新的全四边形单元网格自动生成方法。该方法允许在两个方向上存在网格疏密过渡，并可以提高单元的密度要求自动计算亲单元每条边上的结点数，有效地对局部实施加密处理。     

面向复杂几何模型的并行四面体网格生成方法

面向大规模工程计算等数值模拟领域,提出了一种支持复杂几何模型的大规模四面体网格并行生成方法。该方法以复杂几何模型作为输入,首先采用串行网格生成方法生成初始四面体网格,然后通过两级区域分解方法将初始网格分解为多个子网格并分配到相应的进程中,进程间并行地提取出子网格的表面网格,并基于几何模型对面网格进行贴体加密,最后对加密后的面网格采用Delaunay方法重新生成四面体网格,该方法可以更好地适应高性能计算机体系结构,较好地克服了并行方法中并行性能和网格质量不能兼顾的问题。对三峡大坝模型进行测试和验证,证明该方法具有良好的并行效率和可扩展性,可以在数万处理器核上并行生成数十亿高质量四面体网格。     

Automatic,unstructured mesh generation for tidal calculations in a large domain

An automated procedure is described for the production of unstructured, finite element meshes to perform depth-integrated, hydrodynamic calculations in an ocean-scale, two-dimensional domain. Three relatively coarse meshes with nearly identical boundaries are automatically produced by basing internal size guidelines on a localized truncation error analysis that was performed using results from a highly resolved mesh.

Qualitative and quantitative comparisons of model performance are made at 150 historical tidal stations. The coarsest mesh is shown to meet or exceed the overall accuracy of the other meshes, including a highly resolved mesh that has over six times as many computational points. The automated procedure quickly and easily produces a computationally efficient and accurate finite element mesh that is reproducible. In addition, the methodology is shown to have potential for assessing the importance and accuracy of and bathymetric details and evaluating historical hydrodynamic data.     

并行有限元网格生成方法及其应用

大型工程数值仿真中,在前处理阶段需要生成千万甚至亿量级的网格,传统的串行网格生成方法由于内存和时间的限制,难以处理如此规模的网格。针对此问题,本文提出了一种大规模网格并行生成方法。首先基于推进波前法对几何模型进行初始体网格划分,接着利用图论理论进行区域分解,并通过表面单元恢复保持其几何精度,然后通过分裂法进行网格的并行生成。将所述方法应用到实际大型工程数值仿真前处理阶段,结果表明所述方法可以获得较好的并行效率,同时所产生的网格质量可以满足后续计算需要。     

A fast mesh deformation method for marine propeller flow

ABSTRACT

A fast mesh deformation method for propeller flow is developed based on the elastic solid method. The flow field of a propeller is assumed to be fulfilled with a kind of pseudo elastic solid which does not influence the flow. The vibration equation for the propeller blade-pseudo elastic solid system is derived. During fluid-structure coupling, the nodal displacement for the blade and the flow mesh is computed by modal superposition of the first several modes. Fluid-structure coupling is performed for a highly skewed propeller. The computing time for the dynamic mesh by the present method is about 0.017% of the computing time by the existing elastic solid method. The computing time for the fluid-structure coupling using the present method is 52% less than the computing time by the existing elastic solid method.     

基于射线穿透法的GPU并行阶梯型有限差分网格生成算法

三维大规模有限差分网格生成技术是三维有限差分计算的基础,网格生成效率是三维有限差分网格生成的研究热点。传统的阶梯型有限差分网格生成方法主要有射线穿透法和切片法。本文在传统串行射线穿透法的基础上,提出了基于GPU （graphic processing unit）并行计算技术的并行阶梯型有限差分网格生成算法。并行算法应用基于分批次的数据传输策略,使得算法能够处理的数据规模不依赖于GPU内存大小,平衡了数据传输效率和网格生成规模之间的关系。为了减少数据传输量,本文提出的并行算法可以在GPU线程内部相互独立的生成射线起点坐标,进一步提高了并行算法的执行效率和并行化程度。通过数值试验的对比可以看出,并行算法的执行效率远远高于传统射线穿透法。最后,通过有限差分计算实例可以证实并行算法能够满足复杂模型大规模数值模拟的需求。

     

三维六面体有限元网格自动划分中的一种单元转换优化算法

本文采用十节点曲边四面体转换为六面体网格,并采用非线性约束优化算法取Ｌａｐｌａｃｉａｎ光滑处理算法有效地提高六面体单元的质量,实现了对任意实体的六面体网格自动划分。     

脆性材料破坏模拟的网格生成算法

脆性材料的破坏过程具有随机性,当前的网格生成算法没有充分考虑脆性材料破坏时裂纹扩展和碎块生成的随机性。在Persson网格生成算法与Delaunay随机网格剖分理论基础上,提出了一种可根据模拟需要动态控制网格品质的网格生成算法。通过对随机分布点的Delauna三角化,生成初始网格,然后将网格体系比拟为桁架结构,网格节点即为桁架节点。桁架节点在虚拟力作用下可动态调整位置,并最终达到整个体系受力平衡。对Persson 算法中的尺寸分布函数和收敛条件进行了修正,从而提高了收敛速度,并适用于任意形状对象的网格剖分。 基于VC++平台开发了算法程序。通过实例对算法进行了验证,表明算法能够满足脆性材料破碎模拟的需要。     

基于AFT-Delaunay的二维解耦并行网格生成算法

面向平面任意几何区域网格生成,提出了一种将波前法AFT(Advancing Front Technique)与Delaunay法相结合的解耦并行网格生成算法。算法主要思想是沿着求解几何区域惯性轴,采用扩展的AFT-Delaunay算法生成高质量三角形网格墙,递归地将几何区域动态划分成多个彼此解耦的子区域;采用OpenMP多线程并行技术,将子区域分配给多个CPU并行生成子区域网格;子区域内部的网格生成复用AFT-Delaunay算法,保证了生成网格的质量、效率和一致性要求。本算法优先生成几何边界与交界面网格,有利于提高有限元计算精度;各个子区域的网格生成彼此完全解耦,因此并行网格生成过程无需通信。该方法克服了并行交界面网格质量恶化难题,且具有良好的并行加速比,能够全自动、高效率地并行生成高质量的三角网格。     

一种新的求解对流占优问题的自适应网格细化方法

在均匀网格上求解对流占优问题时,往往会产生数值震荡现象,因此需要局部加密网格来提高解的精度。针对对流占优问题,设计了一种新的自适应网格细化算法。该方法采用流线迎风SUPG(Petrov-Galerkin)格式求解对流占优问题,定义了网格尺寸并通过后验误差估计子修正来指导自适应网格细化,以泡泡型局部网格生成算法BLMG为网格生成器,通过模拟泡泡在区域中的运动得到了高质量的点集。与其他自适应网格细化方法相比,该方法可在同一框架内实现网格的细化和粗化,同时在所有细化层得到了高质量的网格。数值算例结果表明,该方法在求解对流占优问题时具有更高的数值精度和更好的收敛性。     

一种新的钢筋薄膜组合单元及其网格自动生成

提出了一种新的钢筋薄膜组合单元模型和薄膜单元结点信息生成技术。该模型能考虑任意方向的钢筋薄膜,使得单元网格的划分不必考虑钢筋的方位,克服了现有钢筋薄膜组合单元所存在的缺陷。应用本文提出的薄膜结点生成技术,只要给出钢筋薄膜四个角点的坐标,即可自动形成薄膜单元在组合单元中的信息。算例表明该模型方便可行,单元划分效率高,且能达到相当的精度。     

A finite volume method for numerical grid generation

A novel method to generate body‐fitted grids based on the direct solution for three scalar functions is derived. The solution for scalar variables ξ, η and ζ is obtained with a conventional finite volume method based on a physical space formulation. The grid is adapted or re‐zoned to eliminate the residual error between the current solution and the desired solution, by means of an implicit grid‐correction procedure. The scalar variables are re‐mapped and the process is reiterated until convergence is obtained. Calculations are performed for a variety of problems by assuming combined Dirichlet–Neumann and pure Dirichlet boundary conditions involving the use of transcendental control functions, as well as functions designed to effect grid control automatically on the basis of boundary values. The use of dimensional analysis to build stable exponential functions and other control functions is demonstrated. Automatic procedures are implemented: one based on a finite difference approximation to the Cristoffel terms assuming local‐boundary orthogonality, and another designed to procure boundary orthogonality. The performance of the new scheme is shown to be comparable with that of conventional inverse methods when calculations are performed on benchmark problems through the application of point‐by‐point and whole‐field solution schemes. Advantages and disadvantages of the present method are critically appraised. Copyright © 1999 National Research Council of Canada.     

A large-scale parallel hybrid grid generation technique for realistic complex geometry

High-Performance Computing (HPC) systems and Computational Fluid Dynamics (CFD) have made significant progress in recent years; however, as the basis of the large-scale parallel computing, the massive grid generation of billions of cells has become a bottleneck problem. In this study, a parallel grid generation technique is proposed to generate large-scale mixed grids with arbitrary cell types and scales. The basic idea of our method is analogous to the global mesh refinement technique. An initial coarse grid with arbitrary cell types is regarded as a background mesh which is partitioned into subzones, and subzones are assigned onto different CPU cores. After the cells and faces in each subzone are split, the inserted new points of the solid wall are projected onto the original CAD entities to preserve the geometry accurately. Finally, the tangled cells caused by the projection in the boundary layer are untangled by a local Radial Basis Function mesh deformation technique. Furthermore, a parallel partition approach and an efficient wall distance computing technique for massive grids are developed also to shorten the preprocessing time. The tests show that the preprocessing efficiency has been increased by two or three orders compared with traditional methods. Billions of grids are generated for the AIAA JSM high-lift model and the Chinese CHN-T1 transport model to test the ability of the parallel grid generation technique. The maximum scale up to 19 billion mixed elements is generated using 16 384 CPU cores in parallel, and the mesh quality is acceptable for CFD simulations.     

A structured tri-tree search method for generation of optimal unstructured finite element grids in two and three dimensions

A new method for generating finite element grids in two and three dimensions is developed. The method is based on a new search tree structure. The search tree is built upon triangles in two dimensions and tetrahedra in three dimensions. The density of elements can be varied throughout the computational domain. Efficient search algorithms for finding points in space and for finding the boundary of the domain have been developed. The speed of the grid algorithm will permit adaptive gridding during computation. The grid algorithm is generally applicable to both hydrodynamic as well as aerodynamic finite element computations. The technique has been used with success for gridding the North Sea-Skagerrak area.     

一种快速稳健的并行多块结构动网格方法

为解决传统网格处理方法不能满足复杂外形在大设计空间内进行优化时对网格质量的要求的问题,提出了一种并行多块动网格方法,该方法基于初始外形的多块结构网格,根据优化过程中个体外形与初始外形拓扑结构相近的特性,利用体样条插值方法来拟合多块结构网格各块顶点的位移,得到几何外形变化后的拓扑结构,再利用无限插值方法并行地移动初始外形多块结构网格的边、面和块内的网格点,进行光顺处理后得到变形后几何外形的空间网格;该方法在保证网格质量的同时,可以极大地提高网格生成效率,本文以某翼身组合体为例结果表明,该方法在大设计空间的复杂外形设计问题中具有很强的实用性。     

应用于非凸设计区域桁架拓扑优化的一种基结构生成方法

为了提高桁架拓扑优化设计中非凸区域生成基结构的效率和准确性,介绍了一种基于几何和网格信息生成初始基结构的方法。通过直接将已有几何边界作为碰撞检测的标准,避免对原有边界扩展或构建限制区域,简化了基结构生成过程并可以避免杆件识别不准确的问题;通过减少需要与凹边界碰撞检测的杆件个数来提升整体基结构的生成效率;最后,针对已生成的基结构进行基于塑性准则的桁架拓扑优化。     

A stable and efficient hybrid method for aeroacoustic sound generation and propagation

We discuss how to combine the node based unstructured finite volume method widely used to handle complex geometries and nonlinear phenomena with very efficient high order finite difference methods suitable for wave propagation dominated problems. This fully coupled numerical procedure reflects the coupled character of the sound generation and propagation problem. The coupling procedure is based on energy estimates and stability can be guaranteed. Numerical experiments using finite difference methods that shed light on the theoretical results are performed. To cite this article: J. Nordström, J. Gong, C. R. Mecanique 333 (2005).     

A novel method for automated grid generation of ice shapes for local‐flow analysis

Modelling a complex geometry, such as ice roughness, plays a key role for the computational flow analysis over rough surfaces. This paper presents two enhancement ideas in modelling roughness geometry for local flow analysis over an aerodynamic surface. The first enhancement is use of the leading‐edge region of an airfoil as a perturbation to the parabola surface. The reasons for using a parabola as the base geometry are: it resembles the airfoil leading edge in the vicinity of its apex and it allows the use of a lower apparent Reynolds number. The second enhancement makes use of the Fourier analysis for modelling complex ice roughness on the leading edge of airfoils. This method of modelling provides an analytical expression, which describes the roughness geometry and the corresponding derivatives. The factors affecting the performance of the Fourier analysis were also investigated. It was shown that the number of sine–cosine terms and the number of control points are of importance. Finally, these enhancements are incorporated into an automated grid generation method over the airfoil ice accretion surface. The validations for both enhancements demonstrate that they can improve the current capability of grid generation and computational flow field analysis around airfoils with ice roughness. Copyright © 2004 John Wiley & Sons, Ltd.     

基于知识图谱和GPT模型的可靠性代码自动生成方法

工程结构服役中广泛使用可靠性分析进行结构安全评估,但可靠性分析方法种类多、分析程序代码自动化程度低且复用难,需要研究可靠性代码自动生成方法。生成式预训练转换器GPT(Generative Pre-trained Transformer)模型已经在大量替代编程手工作业,进行代码自动生成。但是,其在工程领域中的应用受限于可学习数据量小和问题匹配度不高。本文提出了一种结合多种类可靠性知识图谱、基于GPT的代码自动完成模型进行Matlab可靠性代码预测的方法,使用精心设计的源代码预处理降噪策略,以及知识图谱传播模拟密集型任务解释意图;采用条件代码生成训练,有效提升了小数据样本量的学习性能,实现高准确率、问题匹配的可靠性代码自动生成。最后通过三个可靠性知识图谱案例验证了所提方法的有效性。