Unsupervised Machine Learning for Improved Delaunay Triangulation
Physical oceanography models rely heavily on grid discretization. It is known that unstructured grids perform well in dealing with boundary fitting problems in complex nearshore regions. However, it is time-consuming to find a set of unstructured grids in specific ocean areas, particularly in the ca...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-12-01
|
Series: | Journal of Marine Science and Engineering |
Subjects: | |
Online Access: | https://www.mdpi.com/2077-1312/9/12/1398 |
_version_ | 1797503220156727296 |
---|---|
author | Tao Song Jiarong Wang Danya Xu Wei Wei Runsheng Han Fan Meng Ying Li Pengfei Xie |
author_facet | Tao Song Jiarong Wang Danya Xu Wei Wei Runsheng Han Fan Meng Ying Li Pengfei Xie |
author_sort | Tao Song |
collection | DOAJ |
description | Physical oceanography models rely heavily on grid discretization. It is known that unstructured grids perform well in dealing with boundary fitting problems in complex nearshore regions. However, it is time-consuming to find a set of unstructured grids in specific ocean areas, particularly in the case of land areas that are frequently changed by human construction. In this work, an attempt was made to use machine learning for the optimization of the unstructured triangular meshes formed with Delaunay triangulation in the global ocean field, so that the triangles in the triangular mesh were closer to equilateral triangles, the long, narrow triangles in the triangular mesh were reduced, and the mesh quality was improved. Specifically, we used Delaunay triangulation to generate the unstructured grid, and then developed a K-means clustering-based algorithm to optimize the unstructured grid. With the proposed method, unstructured meshes were generated and optimized for global oceans, small sea areas, and the South China Sea estuary to carry out data experiments. The results suggested that the proportion of triangles with a triangle shape factor greater than 0.7 amounted to 77.80%, 79.78%, and 79.78%, respectively, in the unstructured mesh. Meanwhile, the proportion of long, narrow triangles in the unstructured mesh was decreased to 8.99%, 3.46%, and 4.12%, respectively. |
first_indexed | 2024-03-10T03:47:26Z |
format | Article |
id | doaj.art-00194684fbcf4d2ebc9268c6764169a0 |
institution | Directory Open Access Journal |
issn | 2077-1312 |
language | English |
last_indexed | 2024-03-10T03:47:26Z |
publishDate | 2021-12-01 |
publisher | MDPI AG |
record_format | Article |
series | Journal of Marine Science and Engineering |
spelling | doaj.art-00194684fbcf4d2ebc9268c6764169a02023-11-23T09:03:12ZengMDPI AGJournal of Marine Science and Engineering2077-13122021-12-01912139810.3390/jmse9121398Unsupervised Machine Learning for Improved Delaunay TriangulationTao Song0Jiarong Wang1Danya Xu2Wei Wei3Runsheng Han4Fan Meng5Ying Li6Pengfei Xie7College of Computer Science and Technology, China University of Petroleum, Qingdao 266580, ChinaCollege of Computer Science and Technology, China University of Petroleum, Qingdao 266580, ChinaGuangdong Laboratory of Marine Science and Engineering, Zhuhai 519080, ChinaCollege of Computer Science and Technology, China University of Petroleum, Qingdao 266580, ChinaCollege of Computer Science and Technology, China University of Petroleum, Qingdao 266580, ChinaSchool of Geosciences, China University of Petroleum, Qingdao 266580, ChinaCollege of Computer Science and Technology, China University of Petroleum, Qingdao 266580, ChinaCollege of Computer Science and Technology, China University of Petroleum, Qingdao 266580, ChinaPhysical oceanography models rely heavily on grid discretization. It is known that unstructured grids perform well in dealing with boundary fitting problems in complex nearshore regions. However, it is time-consuming to find a set of unstructured grids in specific ocean areas, particularly in the case of land areas that are frequently changed by human construction. In this work, an attempt was made to use machine learning for the optimization of the unstructured triangular meshes formed with Delaunay triangulation in the global ocean field, so that the triangles in the triangular mesh were closer to equilateral triangles, the long, narrow triangles in the triangular mesh were reduced, and the mesh quality was improved. Specifically, we used Delaunay triangulation to generate the unstructured grid, and then developed a K-means clustering-based algorithm to optimize the unstructured grid. With the proposed method, unstructured meshes were generated and optimized for global oceans, small sea areas, and the South China Sea estuary to carry out data experiments. The results suggested that the proportion of triangles with a triangle shape factor greater than 0.7 amounted to 77.80%, 79.78%, and 79.78%, respectively, in the unstructured mesh. Meanwhile, the proportion of long, narrow triangles in the unstructured mesh was decreased to 8.99%, 3.46%, and 4.12%, respectively.https://www.mdpi.com/2077-1312/9/12/1398unstructured grid generation and optimizationK-means clusteringglobal ocean modelDelaunay triangulationgrid quality |
spellingShingle | Tao Song Jiarong Wang Danya Xu Wei Wei Runsheng Han Fan Meng Ying Li Pengfei Xie Unsupervised Machine Learning for Improved Delaunay Triangulation Journal of Marine Science and Engineering unstructured grid generation and optimization K-means clustering global ocean model Delaunay triangulation grid quality |
title | Unsupervised Machine Learning for Improved Delaunay Triangulation |
title_full | Unsupervised Machine Learning for Improved Delaunay Triangulation |
title_fullStr | Unsupervised Machine Learning for Improved Delaunay Triangulation |
title_full_unstemmed | Unsupervised Machine Learning for Improved Delaunay Triangulation |
title_short | Unsupervised Machine Learning for Improved Delaunay Triangulation |
title_sort | unsupervised machine learning for improved delaunay triangulation |
topic | unstructured grid generation and optimization K-means clustering global ocean model Delaunay triangulation grid quality |
url | https://www.mdpi.com/2077-1312/9/12/1398 |
work_keys_str_mv | AT taosong unsupervisedmachinelearningforimproveddelaunaytriangulation AT jiarongwang unsupervisedmachinelearningforimproveddelaunaytriangulation AT danyaxu unsupervisedmachinelearningforimproveddelaunaytriangulation AT weiwei unsupervisedmachinelearningforimproveddelaunaytriangulation AT runshenghan unsupervisedmachinelearningforimproveddelaunaytriangulation AT fanmeng unsupervisedmachinelearningforimproveddelaunaytriangulation AT yingli unsupervisedmachinelearningforimproveddelaunaytriangulation AT pengfeixie unsupervisedmachinelearningforimproveddelaunaytriangulation |