dc.contributor.author |
Papadrakakis, M |
en |
dc.contributor.author |
Dracopoulos, MC |
en |
dc.date.accessioned |
2014-03-01T01:08:12Z |
|
dc.date.available |
2014-03-01T01:08:12Z |
|
dc.date.issued |
1991 |
en |
dc.identifier.issn |
0045-7825 |
en |
dc.identifier.uri |
https://dspace.lib.ntua.gr/xmlui/handle/123456789/10339 |
|
dc.subject |
Element By Element |
en |
dc.subject.classification |
Engineering, Multidisciplinary |
en |
dc.subject.classification |
Mathematics, Interdisciplinary Applications |
en |
dc.subject.classification |
Mechanics |
en |
dc.subject.other |
Computer Programming |
en |
dc.subject.other |
Mathematical Techniques - Finite Element Method |
en |
dc.subject.other |
Structural Analysis - Computer Applications |
en |
dc.subject.other |
Coefficient Matrix-Vector Products |
en |
dc.subject.other |
Element-Element Solution Method |
en |
dc.subject.other |
Iterative Method Convergence Properties |
en |
dc.subject.other |
Iterative Method Global Preconditioner |
en |
dc.subject.other |
Preconditioning Techniques |
en |
dc.subject.other |
Mathematical Techniques |
en |
dc.title |
A global preconditioner for the element-by-element solution methods |
en |
heal.type |
journalArticle |
en |
heal.identifier.primary |
10.1016/0045-7825(91)90090-S |
en |
heal.identifier.secondary |
http://dx.doi.org/10.1016/0045-7825(91)90090-S |
en |
heal.language |
English |
en |
heal.publicationDate |
1991 |
en |
heal.abstract |
Two computational strategies are proposed in this paper to be implemented for preconditioning of basic iterative methods in the finite element environment. The preconditioning technique employed is called partial preconditioning, which uses a drop-off parameter to control the elements to be retained in the preconditioning matrix without making any alterations to the initial coefficient matrix. We develop a framework for constructing the preconditioning matrices based on the partial preconditioning concept without the need to store the complete coefficient matrix. In those cases where, due to storage limitations, the coefficient matrix-vector products are performed on the element level, as well as in a number of nonlinear applications where the formation of the coefficient matrix is not necessary, the proposed techniques produce a global preconditioner for accelerating an iterative element-by-element solution procedure. The first technique is general and can handle any type of finite element, while the second is particularly adjusted to problems comprising only one-dimensional elements. © 1991. |
en |
heal.publisher |
ELSEVIER SCIENCE SA LAUSANNE |
en |
heal.journalName |
Computer Methods in Applied Mechanics and Engineering |
en |
dc.identifier.doi |
10.1016/0045-7825(91)90090-S |
en |
dc.identifier.isi |
ISI:A1991FY93200001 |
en |
dc.identifier.volume |
88 |
en |
dc.identifier.issue |
3 |
en |
dc.identifier.spage |
275 |
en |
dc.identifier.epage |
286 |
en |