Deep Q Network-Based Optimization Algorithm for Planar Delaunay Mesh

被引：0
|
作者：
Zhang H. [1 ]
Liu X. [1 ]
Li H. [1 ]
机构：
[1] Engineering Mechanics Institute, Nanjing Tech University, Nanjing
来源：
Jisuanji Fuzhu Sheji Yu Tuxingxue Xuebao/Journal of Computer-Aided Design and Computer Graphics | 2022年 / 34卷 / 12期
关键词：
deep Q network; deep reinforcement learning; Delaunay mesh; mesh optimization;
D O I：
10.3724/SP.J.1089.2022.19247
中图分类号：
学科分类号：
摘要：
It is necessary to conduct mesh optimization after generating Delaunay mesh, which is essential to ensure the reliability of numerical simulation. To improve the quality of Delaunay mesh on the plane domain, a mesh optimization algorithm based on deep Q network (DQN) is proposed. Firstly, the quality of the initial mesh is evaluated, and the element nodes that do not meet the quality requirements are selected and arranged in ascending order of their quality. Secondly, the node movement is described as Markov decision process, and the DQN model is established and trained. Thirdly, the empirical parameters of the model training are used to accelerate the optimal mesh quality. Finally, several test examples from practical tunnel, cylinder block, mechanical parts, etc., are employed to verify the applicability and reliability of the proposed algorithm. Compared with the existing typical algorithms, the test results show that the proposed algorithm can significantly improve the quality of poor elements, the quality distribution of optimized mesh will be more concentrated, and no invalid elements are produced during the optimization process. © 2022 Institute of Computing Technology. All rights reserved.
引用
收藏
页码：1943 / 1950
页数：7
相关论文
共 25 条
  • [1] Guan Zhenqun, Song Chao, Gu Yuanxian, Et al., Recent advances of research on finite element mesh generation methods, Journal of Computer-Aided Design & Computer Graphics, 15, 1, pp. 1-14, (2003)
  • [2] Zheng Yao, Chen Jianjun, Unstructured mesh generation: theory, algorithms and applications, pp. 22-55, (2016)
  • [3] Owen S J., A survey of unstructured mesh generation technology
  • [4] Field D A., Laplacian smoothing and Delaunay triangulations, Communications in Applied Numerical Methods, 4, 6, pp. 709-712, (1988)
  • [5] Freitag L A., On combining Laplacian and optimization-based mesh smoothing techniques
  • [6] Gong Xiangfei, Zhang Shudao, One mesh smoothing algorithm combining Laplacian and local optimization-based mesh smoothing techniques, Transactions of Beijing Institute of Technology, 30, 5, pp. 616-621, (2010)
  • [7] Lee E K, Shin M, Kim J., Improved simultaneous mesh untangling and quality improvement methods of 2D triangular meshes, International Journal of Computational Methods, 16, 8, (2019)
  • [8] Kim J, Panitanarak T, Shontz S M., A multiobjective mesh optimization framework for mesh quality improvement and mesh untangling, International Journal for Numerical Methods in Engineering, 94, 1, pp. 20-42, (2013)
  • [9] Mittal K, Fischer P., Mesh smoothing for the spectral element method, Journal of Scientific Computing, 78, 2, pp. 1152-1173, (2019)
  • [10] Kim J., An efficient approach for solving mesh optimization problems using Newton’s method, Mathematical Problems in Engineering, 2014, (2014)
123