Particle-based Fluid Simulation Research Articles

An evaluation of noise reduction algorithms for particle-based fluid simulations in multi-scale applications

Filtering of particle-based simulation data can lead to reduced computational costs and enable more efficient information transfer in multi-scale modelling. This paper compares the effectiveness of various signal processing methods to reduce numerical noise and capture the structures of nano-flow systems. In addition, a novel combination of these algorithms is introduced, showing the potential of hybrid strategies to improve further the de-noising performance for time-dependent measurements. The methods were tested on velocity and density fields, obtained from simulations performed with molecular dynamics and dissipative particle dynamics. Comparisons between the algorithms are given in terms of performance, quality of the results and sensitivity to the choice of input parameters. The results provide useful insights on strategies for the analysis of particle-based data and the reduction of computational costs in obtaining ensemble solutions.

A novel coupling of noise reduction algorithms for particle flow simulations

Proper orthogonal decomposition (POD) and its extension based on time-windows have been shown to greatly improve the effectiveness of recovering smooth ensemble solutions from noisy particle data. However, to successfully de-noise any molecular system, a large number of measurements still need to be provided. In order to achieve a better efficiency in processing time-dependent fields, we have combined POD with a well-established signal processing technique, wavelet-based thresholding. In this novel hybrid procedure, the wavelet filtering is applied within the POD domain and referred to as WAVinPOD. The algorithm exhibits promising results when applied to both synthetically generated signals and particle data. In this work, the simulations compare the performance of our new approach with standard POD or wavelet analysis in extracting smooth profiles from noisy velocity and density fields. Numerical examples include molecular dynamics and dissipative particle dynamics simulations of unsteady force- and shear-driven liquid flows, as well as phase separation phenomenon. Simulation results confirm that WAVinPOD preserves the dimensionality reduction obtained using POD, while improving its filtering properties through the sparse representation of data in wavelet basis. This paper shows that WAVinPOD outperforms the other estimators for both synthetically generated signals and particle-based measurements, achieving a higher signal-to-noise ratio from a smaller number of samples. The new filtering methodology offers significant computational savings, particularly for multi-scale applications seeking to couple continuum informations with atomistic models. It is the first time that a rigorous analysis has compared de-noising techniques for particle-based fluid simulations.

Adaptively constructing liquid surface from particle-based simulation

Natural phenomena simulation attracts a lot of research attention and interest in virtual reality. The simulation for liquid has become a research focus in both computer graphics and computational physics, because of the difficulties in dynamic modelling and high computational complexity. We introduce the main research achievements in recent years with regard to liquids, which is a common natural phenomenon. A hybrid modelling approach for dynamic liquid simulation is proposed, followed by a surface reconstruction method using simulated results. In particle-based fluid simulation, surface construction is one procedure to contour the implicit function which determines an isosurface in a scalar field. We describe an adaptive polygonisation approach, by adaptively constructing an unconstrained octree structure, to reduce excessive subdivision which is required by traditional methods like Marching Cubes. Dual Marching Cubes is then applied to perform surface extraction on a new grid which is topologically dual to the octree structure. Experiment shows the advantages of our approach in generating satisfactory effects without excessively fine-grained grid structure. Satisfactory visual effect is achieved and this application can be used in computer games, movie making and virtual simulation in medical areas.

Efficient extracting surfaces approach employing anisotropic kernels for SPH fluids

Particles are disordered throughout the entire process of fluid simulation using particle-based methods, extracting surfaces through following up particles is unlikely to achieve. Therefore, it is reasonably necessary to extract fluid surfaces called surface reconstruction which has been research focus in particle-based fluid simulation for decades. To construct more smooth surfaces and enhance reconstruction efficiency in fluid simulation, this paper addresses an efficient anisotropic surface reconstruction method for particle-based fluid simulation. First, we simplify and modify the construction of traditional anisotropic kernel function. Second, we divide particles into near-surface particles and internal particles according to the analysis of particles’ eigenvectors. Finally, near-surface particles are involved in the calculation of surface reconstruction while internal particles are directly assigned color field values through the number of neighbor particles. Experimental results show that this algorithm ensures smoothness and geometric characteristics of fluid surfaces reconstructed. Compared to existing algorithms, this approach is simple and easy to implement and greatly improves the operation efficiency.

Particle-based parallel fluid simulation in three-dimensional scene with implicit surfaces

We propose an algorithm for fluid simulation in three-dimensional scenes with obstacles represented as implicit surfaces. The fluid simulation is performed based on the smoothed particle hydrodynamics method and the behavior of fluid particles in the vicinity of obstacles is defined by introducing a new model of particle motion specific to implicitly representation. For effective parallelization on the graphics processing units, we apply the grid of polynomial method to express the implicitly defined obstacles. The proposed method takes advantage of the properties of implicit representation such as smoothness, shape modelling and the expression of deforming objects.

Moving particle semi-implicit method for fluid simulation with implicitly defined obstacles

We developed a particle-based fluid simulation method with obstacles defined in implicit form. Our fluid simulation is based on the Moving Particle Semi-implicit (MPS) method, a typical particle-based algorithm achieving incompressible flow. In general, the particle-based fluid simulation is performed with obstacles that are defined as a set of particles located along the boundaries. We employ the implicit representation for the geometric information of obstacles which requires a new formulation of particle motion at the vicinity of boundaries. The main difficulty of MPS-based simulation with implicit obstacles is the lack of boundary particles that are required for the computation of two quantities: particle number density and particle force determined by pressure field. In our formulation, new definitions of the two quantities giving good estimation of the original ones is developed and incorporated in the MPS algorithm. In addition, we provide a modified algorithm for the construction of discrete linear system specific to the implicit representation for the computation of particle pressure. Our numerical tests show that the proposed approximation techniques provide adequate particle motion at the vicinity of the implicitly defined obstacles.

이상적인 점탄성체 항복 조건을 이용한 폭발 시뮬레이션

In particle-based fluid simulation, the yield stress is required for the deformation of the viscoelastic material like gel. von Mises's yield condition has been proposed to implement deformation of viscoelastic objects, but did not express the explosion. Furthermore, von Mises's yield condition is hard to approximate. We propose an ideal yield condition for viscoelastic object that reference from Tresca's yield condition. Unlike conventional particle-based simulation approximate the external power by the deformed length of the object, this paper is approximate the external power by area of the object. We check up that explosion was realistic when a viscoelastic object is compressed under the ideal yield condition.

유체 시뮬레이션에서 부자연스러운 쇄파의 발생을 완화하기 위한 파티클 움직임 보간 방법

파티클 기반의 유체 시뮬레이션에서 파티클 들이 경계면에 부딪쳐 쇄파를 일으키는 경우 과도한 움직임으로 인해 자연스러운 흐름을 표현하기 어렵다. 파티클이 이동하는 시간 간격을 세분화하여 선형보간 함으로써 이 문제를 해결할 수 있다. 이렇게 하면 경계면에 모여 압력이 높아진 파티클 들의 가속도 벡터가 부드럽게 변한다. 하지만 보간을 하기위한 최소 샘플수가 높기 때문에 파티클 들이 점성을 가진 액체처럼 뭉치는 문제가 발생한다. 본 연구에서는 이 문제를 해결하기 가중치 보간 방식을 사용한다. 가중치 보간은 파티클의 현재 가속벡터와 이전 가속벡터에 서로 다른 가중치를 주며 차례로 더해 이동벡터를 구한다. 가중치 보간 방식을 쓰면 비슷한 흐름을 표현하는데 필요한 샘플수가 선형 보간 방식보다 적다. 그 결과로 점성을 가진 액체처럼 뭉치는 선형보간 방식의 문제점을 해결할 수 있다.

GPUスパコンにおける動的負荷分散を用いた粒子法による大規模流体シミュレーション

GPUスパコンにおける動的負荷分散を用いた粒子法による大規模流体シミュレーション

Real-time Subsurface Scattering for Particle-based fluids using Finite Volume Method

We present a real-time subsurface scattering simulation to perform real-time rendering of translucent particle-based fluids. After particle-based fluid simulation, we immediately build voxelized fluids, ca lled Voronoi fluids, with particle locations and neighbour lists using GPUs. And then, we perform a multiple subsurface scattering simulation over the Voronoi fluids with the diffusion equation (DE). We employ Finite Volume Methods to solve DE efficiently and rapidly with th e voxelized fluids. In our implementation, DE is solved on GPUs with the diffusive source boundary condition and query function for computing the outgoing radiance. We demonstrate a milk demo that show our method can be performed in real-time with CUDA-based fluid simulation.

No-slip boundary conditions and forced flow in multiparticle collision dynamics

Multiparticle collision dynamics (MPCD) is a particle-based fluid simulation technique that is becoming increasingly popular for mesoscale fluid modeling. However, some confusion and conflicting results persist in literature regarding several important methodological details, in particular the enforcement of the no-slip condition and thermostatting in forced flow. These issues persist in simple flows past stationary boundaries, which we exclusively focus on here. We discuss the parametrization of MPCD fluids and its consequences for fluid-solid boundaries in great detail, and show that the method of virtual particles proposed by Lamura et al. and adopted by many others is required only for parameter choices that lead to viscosities dominated by collisional contributions. We test several implementations of the virtual particle method and discuss how to completely eliminate slip at stationary boundaries. We also show that stochastic boundary reflection rules are inherently problematic for forced flow and suggest a possible remedy. Finally, we discuss the most robust way to achieve forced flow and evaluate several thermostatting methods in the process. All discussion is limited to solid objects that do not move as a result of collisions with MPCD particles (i.e., walls). However, the results can be extended to solutes that experience forces and torques due to interactions with MPCD particles (e.g., colloids). The detailed analysis presented for this simple case provides the level of rigor and accuracy to the MPCD method required for the study of more complex systems.

Local Poisson SPH For Viscous Incompressible Fluids

AbstractEnforcing 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).

Incorporating stochastic turbulence in particle-based fluid simulation

We propose a new fluid modeling technique aimed at incorporating stochastic turbulence into a widely used Lagrangian fluid solver, the Smoothed Particle Hydrodynamics (SPH) method. We add to each S...

Dissipative-particle dynamics simulations of flow over a stationary sphere in compliant channels

Dissipative-particle dynamics (DPD), a particle-based fluid-simulation approach, is employed to simulate isothermal pressure-driven flow across a sphere in compliant cylindrical channels. The sphere is represented by frozen DPD particles, while the surrounding fluid is modeled using simple fluid particles. The channel walls are made up of interconnected finite extensible nonlinear elastic bead-spring chains. The wall particles at the inlet and outlet ends of the channel are frozen so as to hinge the channel. The model is assessed for accuracy by computing the drag coefficient CD in shear flow past a uniform sphere in unbounded flow, and comparing the results with those from correlations in literature. The effect of the aspect ratio λ of the channel, i.e., the ratio of the sphere diameter d to the channel diameter D, on the drag force FD on the sphere is investigated, and it is found that FD decreases as λ decreases toward the values predicted by the correlations as λ approaches zero. The effect of the elasticity of the wall is also studied. It is observed that as the wall becomes more elastic, there is a decrease in FD on the sphere.

A Hierarchical Hashing Scheme for Nearest Neighbor Search and Broad-Phase Collision Detection

Increasing computational power allows computer graphics researchers to model spectacular phenomena such as fluids and their interactions with deformable objects and structures. Particle-based (or Lagrangian) fluid and solid simulations are commonly managed separately and mixed together for the collision detection phase. We present a unified dynamic acceleration model to be used for particle neighborhood queries and broad-phase collision detection, based on a hierarchical hash table data structure. Our method is able to significantly reduce computations in large, empty areas, and thus gives better results than existing acceleration techniques, such as multilevel hashing schemes or KD-trees, in most situations.

Porous flow in particle-based fluid simulations

This paper presents the simulation of a fluid flowing through a porous deformable material. We introduce the physical principles governing porous flow, expressed by the Law of Darcy, into the Smoothed Particle Hydrodynamics (SPH) framework for simulating fluids and deformable objects. Contrary to previous SPH approaches, we simulate porous flow at a macroscopic scale, making abstraction of individual pores or cavities inside the material. Thus, the number of computational elements is kept low, while at the same time realistic simulations can be achieved. Our algorithm models the changing behavior of the wet material as well as the full two-way coupling between the fluid and the porous material. This enables various new effects, such as the simulation of sponge-like elastic bodies and water-absorbing sticky cloth.

Adaptively sampled particle fluids

We present novel adaptive sampling algorithms for particle-based fluid simulation. We introduce a sampling condition based on geometric local feature size that allows focusing computational resources in geometrically complex regions, while reducing the number of particles deep inside the fluid or near thick flat surfaces. Further performance gains are achieved by varying the sampling density according to visual importance. In addition, we propose a novel fluid surface definition based on approximate particle-to-surface distances that are carried along with the particles and updated appropriately. The resulting surface reconstruction method has several advantages over existing methods, including stability under particle resampling and suitability for representing smooth flat surfaces. We demonstrate how our adaptive sampling and distance-based surface reconstruction algorithms lead to significant improvements in time and memory as compared to single resolution particle simulations, without significantly affecting the fluid flow behavior.

Adaptively sampled particle fluids

We present novel adaptive sampling algorithms for particle-based fluid simulation. We introduce a sampling condition based on geometric local feature size that allows focusing computational resources in geometrically complex regions, while reducing the number of particles deep inside the fluid or near thick flat surfaces. Further performance gains are achieved by varying the sampling density according to visual importance. In addition, we propose a novel fluid surface definition based on approximate particle-to-surface distances that are carried along with the particles and updated appropriately. The resulting surface reconstruction method has several advantages over existing methods, including stability under particle resampling and suitability for representing smooth flat surfaces. We demonstrate how our adaptive sampling and distance-based surface reconstruction algorithms lead to significant improvements in time and memory as compared to single resolution particle simulations, without significantly affecting the fluid flow behavior.

1601 商用映像制作への粒子法シミュレーションの適用(キーノート講演,OS-16A,OS-16 CGと計算力学)

1601 商用映像制作への粒子法シミュレーションの適用(キーノート講演,OS-16A,OS-16 CGと計算力学)

剛体との相互作用を伴う水の実時間アニメーション

Real-time rendering of realistic motion of fluids is a method that immerses a player into an interactive application such as computer games. Interaction of fluids with rigid bodies is important because fluids and rigid bodies move influencing each other. Fluid simulation based on Computational Fluid Dynamics (CFD) is useful for rendering a visually plausible behaviour of fluids. However, due to the high computational cost of CFD, real-time rendering of fluids needs a fast simulation. This paper describes the particle-based fluid simulation based on Smoothed Particle Hydrodynamics which includes interactions between fluids and rigid bodies, its fast implementation, and the manner of rendering a realistic water surface with optical phenomena such as reflection, refraction, and the Fresnel effect. The proposed method enables real-time animation of water with rigid body interaction.