GPU中的流体场景实时模拟算法

为了实时模拟真实的大规模流体场景,提出一种基于平滑粒子流体力学(SPH)进行流体场景模拟的算法.首先提出了新的精细程度函数作为非均匀采样的依据,以减少实际模拟时所需的粒子数,提高模拟的速度;然后引入一种三维空间网格划分算法和改进的并行基数排序算法,以加快模拟过程中对邻域粒子和边界的查找及其相互作用的计算;最后使用最新的NVIDIA(CUDA(架构,将SPH的全部模拟计算分配到GPU流处理器中,充分利用GPU的高并行性和可编程性,使得对SPH方法的流体计算和模拟达到实时.实验结果表明,采用文中算法能对流体场景的计算模拟达到实时,并实现比较真实的模拟效果.与已有的SPH流体CPU模拟方法相比,其加速比达到2个数量级以上,同时相比已有GPUSPH方法,能模拟出更为丰富的细节效果.     

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.     

非均匀粒子流体模拟

基于物理模型的流体模拟能精确描述流体的运动,但计算量大、实时性较差。根据流体的物理模型,提出了一种基于光滑粒子动力学(SPH)的非均匀粒子模拟流体的方法。提出了一种多因素决定粒子精细度的方法,显著减少了实际模拟需要的流体粒子数目,提高了模拟的速度而又不降低模拟真实感。流体的表面重构采用了一种新的基于屏幕空间的高真实感、高速的重构算法。实验结果表明,与传统的SPH方法相比,该方法在模拟同样场景情况下能大幅降低需要的粒子数目,提升了算法的效率。     

基于CPU-GPU混合加速的SPH流体仿真方法

基于光滑粒子流体力学SPH的流体仿真是虚拟现实技术的重要研究内容,但SPH流体仿真需要大量的计算资源,采用一般计算方法难以实现流体仿真的实时性。流体仿真通常由物理计算、碰撞检测和渲染等部分组成,借助GPU并行加速粒子的物理属性计算和碰撞过程使SPH方法的实时流体仿真成为可能。为了满足流体仿真应用中的真实性和实时性需求,提出一种基于CPU GPU混合加速的SPH流体仿真方法,流体计算部分采用GPU并行加速,流体渲染部分采用基于CPU的OpenMP加速。实验结果表明,基于CPU GPU混合加速的SPH流体仿真方法与CPU实现相比,能显著地减少流体仿真单帧计算时间且能更快速地完成渲染任务。     

三维修正SPH方法实现的海水运动效果模拟

围绕SPH的标准控制方程,对SPH的标准核函数使用了所谓的线性再生核近似方法进行修正,这种方式所取得的梯度效果令人满意。修正的SPH方法模型用来模拟流体自由表面运动。仿真效果展现了水平面上波的形成和传播过程。最后在Open GL环境下实现了的逼真模拟效果。     

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。     

一种面向流体仿真的场景处理方法

在流体仿真中,场景处理是将场景模型转化为可计算流体边界的重要途径.针对传统算法效率低的问题,提出一种面向流体仿真的场景处理方法.通过挖掘流体仿真计算的规律,定义面向流体计算特征的场景语义;利用该语义建立最优场景图来组织仿真场景,并对场景进行高效的处理.将文中方法应用到火灾仿真和流体动画中的实验结果表明,该方法能够提高场景处理效率,满足流体仿真计算的需求.     

一个SPH流体实时模拟的全GPU实现框架

怎样实时地进行高度逼真的大规模流体模拟是图形学要研究的一个重要内容。流体的模拟由物理计算、碰撞检测、表面重构和渲染几个部分组成,因此有大量工作针对流体模拟中的各个部分算法进行GPU加速。提出一整套基于GPU的SPH流体模拟加速框架。在利用平滑粒子动力学(SPH)求解Navier-Stokes方程的基础上,借助基于GPU的空间划分PSS(Parallel Spatial Subdivision)来大幅度提升粒子碰撞的检测速度。同时,设计一种基于几何着色器(Geometry Shader)的流体表面信息的重建算法,并进一步地实现基于索引的优化,使得在流体表面重建过程无须遍历不包含表面的区域。实验结果表明,该方法能实时模拟出具有较好真实感的流体场景。     

一种新的有效模拟不可压缩流体的SPH方法

提出了一种基于光滑粒子流体动力学(SPH)来模拟不可压缩流体的有效方法.传统的SPH方法是针对可压缩流体设计的,而该方法是传统SPH方法的一个扩展.提出了一种新的可以满足不可压缩性的压强计算方法,讨论了压力和粘性力的新型计算方法.实验结果表明,提出的方法与以前的方法相比,能够更真实地模拟不可压缩流体.     

GPU通用计算平台上的SPH流体模拟

针对流体模拟需要大量计算资源从而很难达到实时模拟的问题,提出一种完全在GPU上实现的基于平滑粒子流体动力学的流体模拟方法.首先通过在GPU上构造基于哈希函数的空间均匀网格来实现任意大小场景的快速邻近粒子查找,并在GPU上并行求解SPH流体方程来实现流体模拟;渲染流体时,通过在顶点着色器中进行纹理采样,利用粒子坐标缓存数据直接更新流体粒子系统的顶点缓存,从而避免了CPU—GPU之间的数据传输,充分利用了GPU的并行性.实验对比表明,与纯CPU实现以及CPU和GPU混合实现的模拟结果相比,采用该方法能显著地减少单个时间片的计算时间,大幅度提高流体模拟和渲染的整体性能.     

Monolithic Friction and Contact Handling for Rigid Bodies and Fluids Using SPH

We propose a novel monolithic pure SPH formulation to simulate fluids strongly coupled with rigid bodies. This includes fluid incompressibility, fluid–rigid interface handling and rigid–rigid contact handling with a viable implicit particle-based dry friction formulation. The resulting global system is solved using a new accelerated solver implementation that outperforms existing fluid and coupled rigid–fluid simulation approaches. We compare results of our simulation method to analytical solutions, show performance evaluations of our solver and present a variety of new and challenging simulation scenarios.     

IISPH‐FLIP for incompressible fluids

We propose to use Implicit Incompressible Smoothed Particle Hydrodynamics (IISPH) for pressure projection and boundary handling in Fluid‐Implicit‐Particle (FLIP) solvers for the simulation of incompressible fluids. This novel combination addresses two issues of existing SPH and FLIP solvers, namely mass preservation in FLIP and efficiency and memory consumption in SPH. First, the SPH component enables the simulation of incompressible fluids with perfect mass preservation. Second, the FLIP component efficiently enriches the SPH component with detail that is comparable to a standard SPH simulation with the same number of particles, while improving the performance by a factor of 7 and significantly reducing the memory consumption. We demonstrate that the proposed IISPH‐FLIP solver can simulate incompressible fluids with a quantifiable, imperceptible density deviation below 0.1%. We show large‐scale scenarios with up to 160 million particles that have been processed on a single desktop PC using only 15GB of memory. One‐ and two‐way coupled solids are illustrated.     

基于位置的流体实时交互仿真

针对基于光滑粒子动力学方法的流体交互仿真过程中效率低,交互细节不够真实等问题,提出了采用基于位置的流体来模拟刚体工具与流体的交互方法. 该方法在传统的光滑粒子动力学算法的基础上进行改进,以基于CUDA并行计算平台实时模拟交互过程,并结合力觉交互设备实时输出交互力. 实验结果表明仿真过程中的交互力符合预期,在保证流体模拟的精度的前提下验证了交互力的连续性以及稳定性.     

基于SPH与FEM的结构入水分析方法

建立基于光滑粒子动力学（smoothed particle hydrodynamics, SPH）、有限元法（finite element method, FEM）和无反射边界耦合的结构入水分析方法,将无限水域利用无反射边界条件截断成有限水域,将有限水域分为流体变形大的SPH区域、流体变形小的FEM区域和声学流体FEM区域,结构用FEM离散。采用通用接触算法模拟SPH与FEM的耦合,采用声固耦合方法处理FEM区域之间的耦合,建立流固耦合的SPH FEM分析方法。该方法结合SPH模拟大变形的优点和FEM的高效性,可实现含自由液面变形、液体飞溅和无限水域等特点的流固耦合问题的模拟,为结构入水分析缩小离散区域、降低自由度和SPH粒子数等提供一种有效的分析方法。     

SPH haptic interaction with multiple-fluid simulation

Physics-based fluid interaction plays an important role in computer animation, with wide applications in virtual reality, computer games, digital entertainment, etc. For example, in virtual reality education and games, we often need fluid interactions like acting as an alchemist to create a potion by stirring fluid in a crucible. The traditional input devices such as a mouse and keyboard can basically input 2D information without feedback. In recent years, the continuous development of haptic device not only can achieve six degrees-of-freedom input, but also can calculate the force in virtual scenes and feedback to the user to make a better virtual experience. How to use haptic device in different kinds of virtual fluid scenarios to provide better experience is an important issue in the field of virtual reality. On the other hand, the researches on multiple-fluid interaction especially based on smoothed particle hydrodynamics (SPH) method are very lacking. Therefore, we study the key techniques of haptic interaction with SPH multiple-fluid to compensate this defect in computer graphics community. Different from the single-phase flow, interaction with multiple-fluid flow has difficulties in the realization of properties of different phases. After adding the multiple-fluid simulation, it is also important to keep haptic interaction real time. Our research is based on the mixture model. We guarantee the authenticity of multiple-fluid mixing effect while changing the drift velocity solver to improve efficiency. We employ a unified particle model to achieve rigid body–liquid coupling, and use FIR filter to smooth feedback force to the haptic device. Our novel multiple-fluid haptic simulation can provide an interactive experience for mixing liquid in virtual reality.     

基于CUDA的弱可压SPH流体建模与仿真

为了实现小尺度范围流体场景的实时、真实感模拟,采用弱可压SPH方法对水体进行建模,提出了流体计算的CPU GPU混合架构计算方法。针对邻域粒子查找算法影响流体计算效率的问题,采用三维空间网格对整个模拟区域进行均匀网格划分,利用并行前缀求和和并行计数排序实现邻域粒子的查找。最后,采用基于CUDA并行加速的Marching Cubes算法实现流体表面提取,利用环境贴图表现流体的反射和折射效果,实现流体表面着色。实验结果表明,所提出的流体建模和模拟算法能实现小尺度范围流体的实时计算和渲染,绘制出水的波动、翻卷和木块在水中晃动的动态效果,当粒子数达到1 048 576个时,GPU并行计算方法相较CPU方法的加速比为60.7。     

Direct raytracing of a closed-form fluid meniscus

We present a direct raytracing method for implicitly described fluid surfaces that takes into account the effects of capillary solid coupling at the boundaries. The method is independent of the underlying fluid simulation method and solely based on distance fields. We make use of the closed-form solution of the meniscus shape at the fluid interface to achieve the effect of surface tension exerted by the solid object. The shape of the liquid at these boundaries is influenced by various physical properties such as the force of gravity and the affinity between the liquid and the solid material. We generate contact angles at the boundaries without the need for computationally intensive small-scale simulation. At render time, we combine the closed-form solution for a small-scale effect with the numerical solution of a large-scale simulation. Our method is applicable for any implicit representation of the fluid surface and does not require an explicit extraction of the surface geometry. Therefore, it is especially useful for particle-based simulations. Furthermore, the solution is guaranteed to yield the correct contact angle and, for certain scenarios, it delivers the entirely correct solution throughout the interface; even in general scenarios, it yields plausible results. As for an example, we implemented and tested the proposed method in the setting of a smoothed particle hydrodynamics (SPH) fluid simulation.     

Surface Tension Model Based on Implicit Incompressible Smoothed Particle Hydrodynamics for Fluid Simulation

In order to capture stable and realistic microscopic features of fluid surface, a surface tension and adhesion method based on implicit incompressible SPH (smoothed particle hydrodynamics) is presented in this paper. It gives a steady and fast tension model and can solve the problem of not considering adhesion. Molecular cohesion and surface minimization are considered for surface tension, and adhesion is added to show the microscopic characteristics of the surface. To simulate surface tension and adhesion stably and efficiently, the surface tension and adhesion model is integrated to an implicit incompressible SPH method. The experimental results show that the method can better simulate surface features in a variety of scenarios compared with previous methods and meanwhile ensure stability and efficiency.     

Real-time adaptive fluid simulation with complex boundaries

In this paper, we present a new adaptive model for real-time fluid simulation with complex boundaries based on Smoothed Particle Hydrodynamics (SPH) framework. Firstly, we introduce an adaptive SPH framework that is based on our character field function composed of four factors: geometrical complexity, boundary condition, physical complexity, and complementary condition in terms of the neighboring particle number. Meanwhile, the rule for particle adaptation is presented. We also present a two-step method to fast detect collision with complex boundary. The first step is voxelization on the complex scene. In the second step, based on the result of voxelization, we propose a three-phase method to fast detect collisions between complex boundaries and particles. By using this method, we avoid most of the useless intersection detection computation and greatly enhance the computation efficiency. In addition, a subdivision of boundary is precomputed before the collision interaction method so that fluid in a scene with complex boundary can still be simulated at relatively high speed and system stability risk is reduced greatly. To further accelerate the simulation, a highly parallel fluid algorithm is presented and implemented using GPU so that we can simulate dynamic fluid with mutual interaction between fluid and complex boundary at a considerably fast speed without compromising realism.     

Simulating colliding flows in smoothed particle hydrodynamics with fractional derivatives

We propose a new method based on the use of fractional differentiation for improving the efficiency and realism of simulations based on smoothed particle hydrodynamics (SPH). SPH represents a popular particle‐based approach for fluid simulation and a high number of particles is typically needed for achieving high quality results. However, as the number of simulated particles increase, the speed of computation degrades accordingly. The proposed method employs fractional differentiation to improve the results obtained with SPH in a given resolution. The approach is based on the observation that effects requiring a high number of particles are most often produced from colliding flows, and therefore, when the modeling of this behavior is improved, higher quality results can be achieved without changing the number of particles being simulated. Our method can be employed to reduce the resolution without significant loss of quality, or to improve the quality of the simulation in the current chosen resolution. The advantages of our method are demonstrated with several quantitative evaluations. Copyright © 2013 John Wiley & Sons, Ltd.