发布时间：2018-03-01

文章标题：【学术成果】一篇通讯作者文章在国际学术期刊Exp Fluids上发表

内容：

 

基于VD技术的三维粒子追踪测速算法

引用格式：

Cui YT, Zhang Y, Jia P, et al (2018) Three-dimensional particle tracking velocimetry algorithm based on tetrahedron vote. Exp Fluids 59:31

 

Fig. The Three-dimensional jet flow reconstructed by PTV algorithm proposed in this paper

 

背景阅读：

粒子追踪测速（Particle Tracking Velocimetry）技术基于拉格朗日视角，通过匹配跨帧粒子来定量重构瞬时流场及粒子轨迹，是从运动学分析通向动力学分析的重要桥梁，近年来在越来越多的力学问题中得到了应用，如气固两相流动[Zhang 2016]、固体弹性形变[Boltyanskiy 2017]和生物流动[Brevis 2011]等。而随着流动的日趋复杂化，PTV的鲁棒性亟待加强，实验中对于测量结果（如维度）丰富性的要求也在提高。相应地，通过多相机系统已经能够重构并输出中度数量（小于10000）的三维粒子瞬时坐标，如V3V[Hill 2008]系统。V3V系统采用基于松弛迭代的NRX算法[Ohmi 2000]进行粒子匹配。NRX的雏形是计算机视觉领域中的运动识别算法[Barnard 1980]，随后被引入粒子流动分析领域[Baek 1996]，并在进一步优化之后形成了目前的NRX。其准确率较高，但是由于存在迭代结构，其计算速率较低，且预设参数较多，用户友好性较低；更为重要的是，这两点不足都不利于处理非定常流动图像序列的海量数据。结合非结构化Delaunay网格的NRX算法[Jia 2013, 2015]在一定程度上提高了算法速率。

有一类基于粒子丛概念所发展出的PTV算法，即以局部粒子团作为最小匹配单元。早期有弹簧模型PTV[Okamoto 1995]和速度梯度张量PTV[Ishikawa 2000]，这两种算法的问题在于，随着流场的变化，粒子团大小难以界定。为了解决这一问题，[Song 1999]提出了基于Delaunay网格划分的DT-PTV，将离散点连接成互不重叠的三角形，从而产生了从三角匹配到粒子匹配的双层匹配结构；[Zhang 2014]将此算法推广到了三维情况。为了克服在粒子浓度较高时三角形的趋同性问题，[Zhang 2015]又发展出了VD-PTV，即采用Delaunay网格的对偶结构Voronoi网格实现匹配。这样一来，算法再次回到了单层匹配结构，而且相比三角形，多边形匹配单元的特异性更强；这两点使得VD-PTV的鲁棒性明显优于DT-PTV。然而，VD-PTV中需要步进旋转多边形以获得最为有效的相关性系数。这不但影响了算法的速率，也使得算法难以拓展到多一个自由度的三维情况中去。本文通过引入Delaunay四面体群选举机制再度实现了DT类PTV的三维化，得到了准确率与NRX相当、计算速率更高的PTV，并通过尾迹流实验证明了此算法的有效性。

 

Baek SJ, Lee SJ (1996) A new two-frame particle tracking algorithm using match probability. Exp Fluids 22-1:23-32.

Barnard ST, Thompson WB (1980) Disparity analysis of images. IEEE Trans Pattern Anal Mach Intell (4):333-340.

Boltyanskiy R, Merrill J W, Dufresne E R (2017) Tracking particles with large displacements using energy minimization. Soft Matter 13(11):2201-2206.

Brevis W, Nino Y, Jirka GH (2011) Integrating cross-correlation and relaxation algorithms for particle tracking velocimetry. Exp Fluids 50(1):135-147.

Hill D, Troolin D, Walters G, Lai W, Sharp K (2008) Volumetric 3-component velocimetry (v3v) measurements of the turbulent flow in stirred tank reactors. In: Proceedings of the 14th international symposium on applications of laser techniques to fluid mechanics. Lisbon, Portugal.

Ishikawa M, Murai Y, Wada A, Iguchi M, Okamoto K, Yamamoto F (2000) A novel algorithm for particle tracking velocimetry using the velocity gradient tensor. Exp Fluids 29:519-531.

Jia P, Wang Y, Zhang Y (2013) Improvement in the independence of relaxation method-based particle tracking velocimetry. Meas Sci Technol 24:055301.

Jia P, Wang Y, Zhang Y, Yang B (2015) Relaxation algorithm-based PTV with dual calculation method and its application in addressing particle saltation. J Vis 18(1):71-81.

Ohmi K, Li H (2000) Particle tracking velocimetry with new algorithms. Meas Sci Technol 11-6:603-616.

Okamoto K, Hassan YA, Schmidl WD (1995) New tracking algorithm for particle image velocimetry. Exp Fluids 19(5):342-347.

Song X, Yamamoto F, Iguchi M, Murai Y (1999) New tracking algorithm of PIV and removal of spurious vectors using Delaunay tessellation. Exp Fluids 26:371-380.

Zhang Y, Wang Y, Jia P (2014) Improving the Delaunay tessellation particle tracking algorithm in the three-dimensional field. Meas 49:1-14.

Zhang Y, Wang Y, Yang B, He WB (2015) A particle tracking velocimetry algorithm based on the Voronoi diagram. Meas Sci Technol 26(7):075302.

Zhang Y, Wang Y, Yang B, Jia P (2016) Measurement of sand creep on a flat sand bed using a high-speed digital camera: Mesoscopic features of creeping grains. Sedimentology 63(3):629-644.