Please use this identifier to cite or link to this item:
https://scholarbank.nus.edu.sg/handle/10635/54725
Title: | A posteriori output bound for partial differential equations based on elemental error bound computing | Authors: | Xuan, Z.C. Lee, K.H. Peraire, J. |
Issue Date: | 2003 | Citation: | Xuan, Z.C.,Lee, K.H.,Peraire, J. (2003). A posteriori output bound for partial differential equations based on elemental error bound computing. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 2667 : 1035-1044. ScholarBank@NUS Repository. | Abstract: | An efficient parallel a posteriori output bound procedure for linear functional of finite element solution of partial differential equations is presented. This procedure is based on independently solving the error bound for finite element solution in local elemental Neumann sub-problems. In each subproblem a modified error residual equation which satisfies consistency without needing any complemental conditions is solved for the error bound for the finite element solution. The error bounds for both primal and dual problems are directly used in the output bound which is obtained from optimizing an augmented Lagrangian with a quadratic energy reformulation of the desired output as the objective and finite element equilibrium conditions and interelement continuity requirements as constraints. The algorithm is verified by an example of 2D Poisson problem in the last of the paper. © Springer-Verlag Berlin Heidelberg 2003. | Source Title: | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | URI: | http://scholarbank.nus.edu.sg/handle/10635/54725 | ISSN: | 03029743 |
Appears in Collections: | Staff Publications |
Show full item record
Files in This Item:
There are no files associated with this item.
Google ScholarTM
Check
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.