高可扩展可容错的无网格/粒子程序petaPar及其测试

petaPar 粒子模拟程序面向千万亿次级计算,在统一框架下实现两种广受关注的粒子模拟算法：光滑粒子流体动力学（Smoothed Particle Hydrodynamics,SPH）和物质点法（Material Point Method,MPM)。代码支持多种材料模型、强度模型和失效模型,适合模拟大变形、高应变率和流固耦合问题。支持纯 MPI 和 MPI+X 混合两种并行模型。系统具有可容错性,支持无人值守变进程重启。在Titan 上测试表明,petaPar 可线性扩展到 26 万 CPU 核,SPH 和 MPM 算法并行效率相对 8 192 核分别为 87% 和 90%。     

民用飞机水上迫降数值模拟分析

针对民用飞机水上迫降撞击压力问题缺乏合适的理论分析方法的问题,基于Dytran考虑水一空气两相流体与飞机结构之间的相互耦合作用,通过一般耦合算法实现流固耦合计算;对民用飞机水上迫降过程中机身底部着水压力载荷和着水过程中的飞机俯仰角进行有效数值模拟与水上迫降载荷处理．飞机结构入水时压力在初期达到峰值,然后衰减,在峰值过后会出现小幅波动．     

大变形柔性管道两相流流致振动研究

针对柔性管道内段塞流引起的结构大变形流致振动问题,本文采用分区强流固耦合方法建立了面向大变形两相流输运管道的双向流固耦合数值计算模型.基于流体体积法对气液两相流动界面进行追踪并结合任意拉格朗日-欧拉(ALE)动网格方法考虑流体域网格变形,同时采用有限元方法建立了柔性管道动力学模型,根据流体和管道壁面的相互作用构建强流固耦合计算模型.研究表明,在两相流作用下柔性管道的振动主要以类似一阶和二阶振动模态响应为主且会发生模态切换;模态切换与管内的液塞长度、液塞流动频率以及气液塞在管内的轴向分布有关;管道的大变形振动促进了短气塞的融合并显著改变了液塞的长度和频率,进而影响管道的振动和流型转变界限.     

基于ALE方法的楔形体入水分析

基于ALE方法分析楔形体入水问题,楔形体采用拉格朗日网格离散,空气和水采用ALE网格离散。将楔形体视为刚体,空气和水的力学行为分别采用Gamma定律和GRUNEISEN状态方程模拟。讨论ALE流固耦合关键字中罚函数罚因子的取值方法,提出相应的建议原则;分析水域截断边界对楔形体响应的影响,给出模拟无限水域的截断边界位置的建议值;分析楔形体表面压力振荡的原因,提出楔形体表面压力获取方法。通过试验结果对比分析,验证方法的合理性。     

复杂场景流固耦合的高效模拟

大规模复杂流体场景的高效交互模拟技术在灾难仿真、虚拟现实和影视特效制作等领域都有重要的应用价值.针对现有基于物理的流体算法大多只适于尺度较小的场景,难以有效地模拟复杂流体大场景中流固耦合时的固体破碎效果的问题,提出一种复杂场景的固液耦合高效模拟方法.首先提出一种基于流体隐式粒子法和散度为零的SPH法相混合的计算框架,在确保流体物理属性的同时充分利用细粒度的隐式粒子来丰富流体运动细节,提高场景模拟的真实感;然后采用一种多维度的流固耦合分合计算策略来进一步提高流固耦合效率;为实现流体冲击下的固体破碎效果,采用一种物理与几何相混合的破碎方法:以基于断裂力学中的应变能密度模型来确定固体破碎时碎片的分布,采用基于几何的质心Voronoi方法快速生成碎片,最终实现百万量级粒子参与的复杂流体场景的交互模拟,以及高速流体冲击下的固体破碎效果的高效模拟.     

流-固交互及可变形体破裂的真实感模拟

为了模拟流体与动态环境的相互作用,提出一种流-固交互及可变形体破裂的真实感建模与绘制的算法.该算法使用光滑粒子流体动力学(SPH)与有限单元法(FEM)分别对流体与变形固体进行建模;再根据流-固交互作用的特点,给出一种快速分离液体表面粒子与固体表面网格的交互方法,并采用虚节点的流-固耦合模型模拟了液-固相互作用力.文中算法可用于多个流-固交互破裂的现象,如水管崩裂、水冲堤坝等.     

基于SPH方法的滴水涟漪动画模拟

光滑粒子流体动力学（Smoothed Particle Hydrodynamics,SPH）方法是一种新近发展的可用于流体模拟的无网格数值方法。文中基于SPH方法的基本原理,利用SPH方法求解描述水流现象的二维浅水波方程,根据具体模型使用Mon-aghan人工粘性的变形形式,有效地防止了相互靠近粒子的穿透,消除了SPH方法在模拟流体动力学问题时产生的数值振荡。通过使用可变光滑长度,使邻近粒子的数量保持相对稳定,提高了求解的计算效率和精度。同时,对光滑长度进行了修正以获取对称光滑长度,保持了粒子间相互作用对称性。全面考虑了各种定解条件的设置,对水滴的运动进行了模拟,SPH模拟结果与有限差分法、有限体积法结果非常吻合,验证了方法的准确性,为SPH方法的进一步发展和广泛运用奠定了基础。     

SPH-GPU并行计算在风沙流中的应用

为了实现小尺度范围风沙运动的真实感模拟,采用基于拉格朗日力学无网格形式的光滑粒子流体动力学（smooth particle hydrodynamics,SPH）方法解决了基于欧拉网格法因网格大变形或者变形边界等引起的各种问题,并克服了不能用固定欧拉网格追踪任意单颗粒子运动轨迹的困难,因此该方法在研究风沙运动方面有着独特的优势。然而,随着风沙流动中SPH粒子数目的增加,该方法计算效率低,计算规模大的缺陷在风沙模拟过程中尤为明显。为了提高其计算效率,在CUDA软硬件平台上,建立SPH-GPU并行加速的二维气沙两相耦合模型,对串行的热点程序进行分析,找出最耗时且适合并行的热点程序;其次对GPU并行计算模型进行验证,宏观上得到了沙粒群运动的时空变化规律,微观上得到了典型沙粒的跃移轨迹和变异的尖角轨迹;最后对比了三种不同粒子数下CPU与GPU的计算效率。模拟结果证明SPH-GPU并行计算方法能够进一步应用在风沙流的数值模拟研究中。     

基于INTESIM-GISCI的流固耦合 仿真软件技术及应用

为突破传统商业软件流固耦合分析仅局限于内部预先集成的流体和结构求解器的约束,使流固耦合分析更具开放性和可拓展性,以耦合驱动程序INTESIM GISCI为框架体系,在开源流体求解器SU2的源代码上加入时间同步点和相关函数功能形成INTESIM-SU2,使INTESIM-SU2可以与原有的结构求解器INTESIM-Structure通过耦合界面传递数据的方式进行耦合分析,从而实现基于耦合驱动程序INTESIM-GISCI的流固耦合仿真软件开发。将多个流固耦合分析案例与其他商业软件进行对比,证明基于INTESIM-GISCI的流固耦合仿真软件可广泛应用于实际工程问题的仿真分析。     

离散元-有限差分跨尺度耦合下桩与土体稳定性细观数值分析

为准确分析桩体宏观变形与周围土颗粒细观力学行为,采用离散元-有限差分的跨尺度耦合进行桩和土体接触过程中的稳定性分析,研究桩下沉过程中周围土体的细观变形、应力分布和桩体自身变化情况,通过FLAC3D建立桩和外部土体有限差分网格单元,对桩周围侧土体应用PFC3D离散元建立土颗粒微观结构模型.研究结果表明:离散元与有限差分耦...     

Realistic and stable simulation of turbulent details behind objects in smoothed‐particle hydrodynamics fluids

This paper presents a novel realistic and stable turbulence synthesis method to simulate the turbulent details generated behind objects in smoothed particle hydrodynamics (SPH) fluids. Firstly, by approximating the boundary layer theory on the fly in SPH fluids, we propose a vorticity production model to identify which fluid particles shed from object surfaces and which are seeded as vortex particles. Then, we employ an SPH‐like summation interpolant formulation of the Biot–Savart law to calculate the fluctuating velocities stemming from the generated vorticity field. Finally, the stable evolution of the vorticity field is achieved by combining an implicit vorticity diffusion technique and an artificial dissipation term. Moreover, in order to efficiently catch turbulent details for rendering, we propose an octree‐based adaptive surface reconstruction method for particle‐based fluids. The experiment results demonstrate that our turbulence synthesis method provides an effect way to model the obstacle‐induced turbulent details in SPH fluids and can be easily added to existing particle‐based fluid–solid coupling pipelines. Copyright © 2014 John Wiley & Sons, Ltd.     

SPH particle boundary forces for arbitrary boundaries

This paper is concerned with approximating arbitrarily shaped boundaries in SPH simulations. We model the boundaries by means of boundary particles which exert forces on a fluid. We show that, when these forces are chosen correctly, and the boundary particle spacing is a factor of 2 (or more) less than the fluid particle spacing, the total boundary force on a fluid SPH particle is perpendicular to boundaries with negligible error. Furthermore, the variation in the force as a fluid particle moves, while keeping a fixed distance from the boundary, is also negligible. The method works equally well for convex or concave boundaries. The new boundary forces simplify SPH algorithms and are superior to other methods for simulating complicated boundaries. We apply the new method to (a) the rise of a cylinder contained in a curved basin, (b) the spin down of a fluid in a cylinder, and (c) the oscillation of a cylinder inside a larger fixed cylinder. The results of the simulations are in good agreement with those obtained using other methods, but with the advantage that they are very simple to implement.     

一种基于SPH流体模拟的固液边界改进算法

光滑粒子流体动力学(SPH)法是一种无网格的流体模拟方法,固液边界处理是SPH法模拟流体行为的重点和难点。本文提出一种单层加密粒子法进行固液边界处理。与虚拟粒子法将边界假设为静止的流体粒子不同,本文将边界假设为具有一定密度的固体粒子,依靠物理约束进行流体计算。这种方法能够有效降低模拟中穿越边界的粒子数量,使得流体边界处的模拟更加符合真实情况。本文采用仿真流体数据对提出的算法进行验证,并对其有效性进行分析讨论。     

3D IEM formulation with an IEM/FEM coupling scheme for solving elastostatic problems

In this paper, a detailed three-dimensional infinite element methodology (IEM) formulation with an infinite element (IE)–finite element (FE) coupling scheme for investigating elastostatic problems is presented. This method is equally well suited for a regular perfect domain and a domain with geometric singularity; for example, domains with cracks. In this method, the primary problem domain is subdivided into two sub-domains modeled separately using IEM and finite element method (FEM), respectively. All degrees of freedom related to the IE sub-domain, except for those associated with the coupling interface, are condensed and transformed to form a finite master IE with the master nodes on the sub-domain boundary. Finally, a symmetrical IE stiffness matrix containing only master node degrees of freedom is assembled into the system stiffness matrix for the FE sub-domain. A very fine mesh pattern can be established using these efficient numerical techniques without increasing the d.o.f.'s of the global FEM solution. Numerical examples are presented and compared with the corresponding analytical or numerical solutions to show the performance of the proposed methodology.     

Modeling accidental-type fluid–structure interaction problems with the SPH method

This paper presents the validation aspects of a unified numerical framework based on SPH formulation and devoted to the modeling of fluid–structure interaction problems involving large motion of the fluid and large deformation with a possible failure of the structure. The fluid domain is modeled according to an updated Lagrangian formulation. The solid domain (3D and shell models) uses the total Lagrangian formulation. The fluid–structure interaction is treated via a unilateral contact algorithm adapted to SPH context. The SPH framework is verified on academic test cases and validated by simulating an experiment involving the reservoir leakage.     

Local Poisson SPH For Viscous Incompressible Fluids

Enforcing fluid incompressibility is one of the time‐consuming aspects in SPH. In this paper, we present a local Poisson SPH (LPSPH) method to solve incompressibility for particle based fluid simulation. Considering the pressure Poisson equation, we first convert it into an integral form, and then apply a discretization to convert the continuous integral equation to a discretized summation over all the particles in the local pressure integration domain determined by the local geometry. To control the approximation error, we further integrate our local pressure solver into the predictive‐corrective framework to avoid the computational cost of solving a pressure Poisson equation globally. Our method can effectively eliminate the large density deviations mainly caused by the solid boundary treatment and free surface topological change, and show advantage of a higher convergence rate over the predictive‐corrective incompressible SPH (PCISPH).     

固体入水的边界压力处理及表面粒子位置的修正

基于传统光滑粒子流体动力学(SPH)方法的边界力法、虚粒子法或耦合力法处理固体入水时,流 体与固体交互界面的粒子密度不连续、压力不稳定、固体边界处会发生部分流体粒子穿透或分离等现象,而流 体表面因为受到力的作用,表面破碎后,液面较粗糙。针对上述问题,结合边界力和虚粒子的优点,对耦合力 法进行改进,处理运动固体边界,阻止流体粒子穿透固体边界;改进交互界面的压力计算方法,提高计算精度, 稳定交互界面压力场;对流体表面的粒子位置进行校正,提升流体表面自由流动液面边界的模拟效果。通过经 典的二维固体入水实验,对该方法进行了验证,实验结果表明,本文方法在流体粒子与固体粒子交互后,交互 界面压力稳定,界面分离清晰无穿透,表面流体粒子分布均匀,流场的运动真实自然。     

关于液滴撞击弹性固体变形预测仿真

在液滴撞击弹性固体问题中,由于流固两相的动力学特性复杂且传统网格法求解困难,研究中通常将固体结构视为刚性壁面,不考虑固体在冲击下的变形情况及变形对液滴的影响,导致数值仿真精度较低。根据光滑粒子动力学(Smoothed Particle Hydrodynamics,SPH)方法的基本原理和理论,采用连续介质力学的控制方程,引入非牛顿流体和弹性固体所遵循的本构关系,分析流固两相的相互作用,提出一种流固边界的耦合处理算法,建立流固耦合动力学模型,对非牛顿液滴撞击弹性固体的动态过程进行数值仿真。仿真结果表明,上述数值方法能够精细地预测出撞击过程中非牛顿液滴的形态变化和弹性固体结构的微变形情况,并探讨了具有不同弹性模量的固体在液滴撞击下的可变形性及对液滴的影响。     

Interaction of flexible multibody systems with fluids analyzed by means of smoothed particle hydrodynamics

The present work deals with a computational approach to fluid-structure interaction (FSI) problems by coupling of flexible multibody system dynamics and fluid dynamics. Since the methods for the numerical modeling are well known, both for the structural and the fluid part, the focus of this work lies on the coupling formalism. Moreover, the applicability of the presented approach to arbitrary geometries and high structural stiffness is studied, as well as an easy model setup. No restriction should be made on the topology of the structure or the complexity of motion.For the fluid part a meshless method, known as smoothed particle hydrodynamics (SPH) is applied, which fulfills the above requirements. While an explicit time integration scheme in SPH provides a fast simulation of the fluid dynamics, advanced methods from flexible multibody dynamics provide a variety of benefits for the simulation of the solid part. Amongst these are specialized structural finite elements for both small and large deformation bodies, joints, stable implicit time-integration schemes, and model reduction techniques.A rule for the interaction between fluids and structures is derived from imposing a distributed potential over boundary segments of the structures, which the fluid particles respond to. The work is concluded by illustrative examples, demonstrating the successful coupling of flexible multibody systems with fluids.     

GPU‐assisted real‐time coupling of blood flow and vessel wall

The vessel wall and the blood flow interact and influence each other, and real‐time coupling between them is of great importance to the virtual surgery as well as the research and diagnosis of vascular disease. On the basis of smoothed particle hydrodynamics (SPH), we present a new approach to solve non‐Newtonian viscous force of blood and a parallel mixed particles‐based coupling method for blood flow and vessel wall. Meanwhile, we also design a proxy particle‐based vessel wall force visualization method. Our method is as follows. Firstly, we solve the non‐Newtonian viscous forces of blood through the SPH method to discretize the Casson equation. Secondly, in each time step, we combine blood particles and sampling proxy particles on the blood vessel wall to form mixed particles and calculate the interaction forces through the SPH method between every pair of the neighboring mixed particles inside the graphics processing unit. Thirdly, the forces of the proxy particles will be mapped to the color display of the proxy particle. Experimental results demonstrate that our method is able to implement real‐time sizeable coupling of blood flow and vessel wall while mainly ensuring physical authenticity and it can also provide real‐time and obvious information about vessel wall force distribution. Copyright © 2015 John Wiley & Sons, Ltd.