An adaptive nonlinear finite element analysis of minimal surface problem based on element energy projection technique. Issue 8 (30th April 2020)

• Record Type: Journal Article
• Authors: Jiang, Kaifeng; Yuan, Si; Xing, Qinyan
• Is Part Of: Engineering computations. Volume 37:Issue 8(2020)
• Page Start: 2847
• Page End: 2869
• Publication Date: 2020-04-30
• Subjects: Nonlinear finite element analysis -- Adaptive strategy -- Element energy projection -- Maximum norm -- Minimal surface problem -- Super-convergence; Computer-aided engineering -- Periodicals; Computer graphics -- Periodicals; 620.00285
• DOI: 10.1108/EC-08-2019-0369
• Languages: English
• ISSNs: 0264-4401

Abstract

Abstract : Purpose: This paper aims to propose a new adaptive strategy for two-dimensional (2D) nonlinear finite element (FE) analysis of the minimal surface problem (MSP) based on the element energy projection (EEP) technique. Design/methodology/approach: By linearizing nonlinear problems into a series of linear problems via the Newton method, the EEP technique, which is an effective and reliable point-wise super-convergent displacement recovery strategy for linear FE analysis, can be directly incorporated into the solution procedure. Accordingly, a posteriori error estimate in maximum norm was established and an adaptive 2D nonlinear FE strategy of h -version mesh refinement was developed. Findings: Three classical known surfaces, including a singularity problem, were analysed. Moreover, an example whose analytic solution is unavailable was considered and a comparison was made between present results and those computed by the MATLAB PDE toolbox. The results show that the adaptively-generated meshes reflect the difficulties inherent in the problems and the proposed adaptive analysis can produce FE solutions satisfying the user-preset error tolerance in maximum norm with a fair adaptive convergence rate. Originality/value: The EEP technique for linear FE analysis was extended to the nonlinear procedure of MSP and can be expected to apply to other 2D nonlinear problems. The employment of the maximum norm makes point-wisely error control on the sought surfaces possible and makes the proposed method distinguished from other adaptive FE analyses.