An efficient maximum bound principle preserving p-adaptive operator-splitting method for three-dimensional phase field shape transformation model. (15th August 2022)

• Record Type: Journal Article
• Authors: Wang, Yan; Xiao, Xufeng; Feng, Xinlong
• Is Part Of: Computers & mathematics with applications. Volume 120(2022)
• Page Start: 78
• Page End: 91
• Publication Date: 2022-08-15
• Subjects: Shape transformation -- Operator splitting method -- p-adaptive algorithm -- Maximum bound principle -- Parallel algorithm; Electronic data processing -- Periodicals; Mathematics -- Data processing -- Periodicals; 510.28541
• DOI: 10.1016/j.camwa.2022.06.015
• Languages: English
• ISSNs: 0898-1221

Abstract

Abstract: In this paper, a novel numerical algorithm for efficient modeling of three-dimensional shape transformation governed by the modified Allen-Cahn (A-C) equation is developed, which has important significance for computer science and graphics technology. The new idea of the proposed method is as follows. Firstly, the operator splitting method is used to decompose the three-dimensional problem into a series of one-dimensional subproblems that can be solved in parallel in the same direction. Secondly, a temporal p-adaptive strategy, which is based on the extrapolation technique, is proposed to improve the convergence order in time and preserve the computational efficiency simultaneously. Finally, a parallel least distance modification technique is developed to force the discrete maximum bound principle. The proposed method achieves high precision and high efficiency at the same time. Numerical examples include the effectiveness of the p-adaptive method and the bound preserving least distance modification, and a series of complex three-dimensional shape transformation modelings. Highlights: The operator splitting method is used to solve 3D shape transformation PDE. A temporal p-adaptive strategy is developed to improve computational efficiency. A least-distance modification is developed to force the discrete maximum bound.