RM?PTV中选取候选粒子的形心DT网格方法  

Determining Candidate Particles with Centroid DT Method in RM-PTV

作　　者：贾攀[1] 王元[1] 张洋[1] 

机构地区：[1]西安交通大学流体机械及工程系,西安710049

出　　处：《西安交通大学学报》2013年第1期80-84,共5页Journal of Xi'an Jiaotong University

基　　金：国家自然科学基金资助项目(10972176;11272252)

摘　　要：为提高松弛迭代粒子追踪测速法(RM-PTV)提取瞬态流场的便捷性,提出了一种形心Delaunay三角剖分(DT)方法,并用其代替固定阈值法来选取与目标粒子预匹配的候选粒子,以降低RM-PTV对待测流场特性的依赖。通过对典型流型的自定义及标准人工粒子流场图像的匹配计算进行全面、有效地验证表明:形心DT网格可有效、便捷地选取候选粒子,保证RM-PTV得到准确的匹配结果,兼顾匹配准确率和效率,省去流场速度的预先估计,减小算法对流场及具体粒子分布情况的依赖,使RM-PTV更独立,应用更方便。To improve the convenience of the relaxation based particle tracking velocimetry（RM-PTV） in obtaining the instantaneous flow structure,a centroid Delaunay triangulation（DT） based technique was proposed to determine candidate particles for the target particles in the matching process,and typical self-defined particle images and standard particle images were analyzed to test the validity of the new method.The results show that the centroid DT method can select the candidate particles effectively and conveniently.Moreover,consideration is given to balance accuracy and efficiency,and the previous estimation of the flow field can be avoided,which makes the application of RM-PTV more independent,especially for the complicated flow situation and inexperienced users.

关 键 词：松弛迭代粒子追踪测速法 候选粒子 固定阈值 DELAUNAY三角剖分 

分 类 号：O359[理学—流体力学] X169[理学—力学]