Application of SPH-GPU Parallel Computing in Wind and Sand Flow

doi:10.3778/j.issn.1002-8331.2007-0196

Abstract

Abstract: In order to realize the realistic simulation of the wind and sand movement in a small scale, the smooth particle hydrodynamics smooth particle hydrodynamics（SPH） method based on the Lagrangian mechanics meshless form is used to solve the grid based on the Euler grid method various problems which caused by large deformation or deformation boundary and overcome the difficulty of tracking the trajectory of any single particle with a fixed Euler grid. Therefore, this method has a unique advantage in the study of sand movement. However, with the increase of the number of SPH particles in the flow of wind and sand, the calculation efficiency of this method is low while the defects of large calculation scale are especially obvious in the process of wind and sand simulation. In order to improve its computational efficiency, software and hardware platform, a two-dimensional gas-sand two-phase coupled model of SPH-GPU parallel acceleration is established on the CUDA. Firstly, the serial hotspot program is analyzed to find the most time-consuming and suitable parallel hotspot program. Secondly, the GPU parallel computing model is verified. The temporal and spatial variation laws of the movement of the sand particles are obtained macroscopically while the jumping trajectories and the sharp trajectories of typical sand particles are obtained microscopically. Finally, the computational efficiency of CPU and GPU under three different particle numbers is compared. The simulation results prove that the SPH-GPU parallel computing method can be further applied to the numerical simulation study of wind and sand flow.

Key words: smooth particle hydrodynamics（SPH） method, sandstorm, parallel acceleration, computational efficiency

摘要： 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。

关键词: 光滑粒子流体动力学（SPH）方法, 风沙运动, 并行加速, 计算效率

LIANG Lanbo, JIN Afang, WEN Tengteng. Application of SPH-GPU Parallel Computing in Wind and Sand Flow[J]. Computer Engineering and Applications, 2022, 58(1): 248-254.

梁岚博, 金阿芳, 闻腾腾. SPH-GPU并行计算在风沙流中的应用[J]. 计算机工程与应用, 2022, 58(1): 248-254.

References

[1] 倪晋仁，李振山.风沙两相流理论及其应用[M].北京：科学出版社，2006：1-204.
NI J R，LI Z S.The theory of wind-sand two-phase flow and its application[M].Beijing：Science Press，2006：1-204.
[2] RALAIARISOA J L，BESNARD J B，FURIERI B.transition from saltation to collisional regime in windblown sand[J].Physical Review Letters，2020，124：198501.
[3] WU J S，ZHANG H，YANG R.Numerical modeling of dam-break flood through intricate city layouts including underground spaces using GPU-based SPH method[J].Journal of Hydrodynamics，2013，25（6）：818-828.
[4] 买雪洁，石杰元，童倩倩.基于周期性校正神经网络的血流血管壁耦合[J].计算机工程与应用，2019，55（24）：178-183.
MAI X J，SHI J Y，TONG Q Q.Coupling of blood flow and vessel wall based on periodic-corrected network[J].Computer Engineering and Applications，2019，55（24）：178-183.
[5] 金阿芳，买买提明·艾尼，胡国玉，等.基于SPH方法的风沙流中沙粒速度分布数值计算[J].干旱区研究，2018，35（1）：107-113.
JIN A F，MAIMAI TIMING A N，HU G Y，et al.SPH-based numerical calculation of sand grain velocity distribution in wind-blown sand flow[J].Arid Zone Research，2018，35（1）：107-113.
[6] 逯博，买买提明·艾尼，金阿芳，等.基于SPH的风沙运动的数值模拟[J].力学学报，2013，45（2）：177-182.
LU B，MAIMAI TIMING A N，JIN A F，et al.Numerical simulation of sand movement based on SPH[J].Acta Mechanica Sinica，2013，45（2）：177-182.
[7] 牟阳阳，金阿芳，别艺楠.坡面风沙的SPH数值模拟[J].干旱区研究，2015，32（6）：1140-1145.
MOU Y Y，JIN A F，BIE Y N.SPH numerical simulation of windblown sand movement over the slope surface[J].Arid Zone Research，2015，32（6）：1140-1145.
[8] 陈福振，强洪夫，高巍然.风沙运动问题的SPH-FVM耦合方法数值模拟研究[J].物理学报，2014，63（13）：14-26.
CHEN F Z，QIANG H F，GAO W R.Simulation of aerolian sand transp ort with SPH-FVM coupled method[J].Acta Physica Sinica，2014，63（13）：14-26.
[9] CHEN J Y，LIEN F S，PENG C，et al.GPU-accelerated smoothed particle hydrodynamics modeling of granular flow[J].Powder Technology，2020，359：94-106.
[10] 徐锋.基于众核架构的并行SPH算法的研究与实现[D].上海：上海交通大学，2013.
XU F.Research and implementation of parallel SPH algorithm based on many-core architecture[D].Shanghai：Shanghai Jiaotong University，2013.
[11] SASANI BABAK A.Numerical study of wave run-up and overtopping considering bed roughness using SPH-GPU[J].Coastal Engineering Journal，2019，61（4）：502-519.
[12] CHEN J Y，LIEN F S，PENG C，et al.GPU-accelerated smoothed particle hydrodynamics modeling of granular flow[J].Powder Technology，2020，359：94-106.
[13] GINGLOD R A，MONAGHAN J J.Smoothed particle hydrodynamics：Theory and application to non-spherical star[J].Monthly Notices of the Royal Astronomical Society，1977，181：375-389.
[14] MONAGHAN J J，GINGOLD R A.Shock simulation by the particle method SPH[J].Journal of Computational Physics，1983，52：374-389.
[15] 韩旭.光滑粒子流体动力学：一种无网格粒子法[M].长沙：湖南大学出版社，2005.
HAN X.Smooth particle hydrodynamics：A meshless particle method[M].Changsha：Hunan University Press，2005.
[16] 李志强，王元，张洋.跃移沙粒运动参数的数字图像分析[J].西安交通大学学报，2010，44（7）：109-113.
LI Z Q，WANG Y，ZHANG Y.Digital image analysis on motion parameters of saltating sand grains[J].Journal of Xi’an Jiaotong University，2010，44（7）：109-113.