Volume 26, Issue 5
Detecting Particle Clusters in Particle-Fluid Systems by a Density Based Method

Tian Tian, Han Wang, Wei Ge & Pingwen Zhang

Commun. Comput. Phys., 26 (2019), pp. 1617-1630.

Published online: 2019-08

Preview Full PDF 16 952
Export citation
  • Abstract

In this paper, the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) method is proposed to detect particle clusters in particle-fluid systems. The particles are grouped in one cluster when they are connected by a dense environment. The parameters that define the dense environment are determined by analyzing the structure of the system, therefore, our approach needs little human intervention. The method is illustrated by identifying the clusters in two kinds of simulation trajectories of different particle-fluid systems. The robustness of cluster identification in terms of statistical properties of clusters in the steady state is demonstrated

  • Keywords

Particle-fluid system, cluster, DBSCAN method.

  • AMS Subject Headings

6207, 70C20, 76T25

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address

1501110025@pku.edu.cn (Tian Tian)

wang han@iapcm.ac.cn (Han Wang)

wge@ipe.ac.cn (Wei Ge)

pzhang@pku.edu.cn (Pingwen Zhang)

  • References
  • Hide All
    View All

@Article{CiCP-26-1617, author = {Tian , Tian and Wang , Han and Ge , Wei and Zhang , Pingwen }, title = {Detecting Particle Clusters in Particle-Fluid Systems by a Density Based Method}, journal = {Communications in Computational Physics}, year = {2019}, volume = {26}, number = {5}, pages = {1617--1630}, abstract = {

In this paper, the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) method is proposed to detect particle clusters in particle-fluid systems. The particles are grouped in one cluster when they are connected by a dense environment. The parameters that define the dense environment are determined by analyzing the structure of the system, therefore, our approach needs little human intervention. The method is illustrated by identifying the clusters in two kinds of simulation trajectories of different particle-fluid systems. The robustness of cluster identification in terms of statistical properties of clusters in the steady state is demonstrated

}, issn = {1991-7120}, doi = {https://doi.org/10.4208/cicp.2019.js60.09}, url = {http://global-sci.org/intro/article_detail/cicp/13278.html} }
Copy to clipboard
The citation has been copied to your clipboard