<?xml version="1.0"?>
<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:dc="http://purl.org/dc/elements/1.1/"><rdf:Description rdf:about="https://repozitorij.uni-lj.si/IzpisGradiva.php?id=32142"><dc:title>Convergence properties of materially and geometrically non-linear finite-element spatial beam analysis</dc:title><dc:creator>Zupan,	Dejan	(Avtor)
	</dc:creator><dc:creator>Saje,	Miran	(Avtor)
	</dc:creator><dc:subject>non-linear beam theory</dc:subject><dc:subject>non-linear material</dc:subject><dc:subject>reiforced concrete</dc:subject><dc:subject>stress field integration</dc:subject><dc:subject>Newton's iteration</dc:subject><dc:subject>rate of convergence</dc:subject><dc:description>The way the non-linear constitutive equations in the spatial beam formulations are solved, influences the rate of convergence and the computational cost. Three different approaches are studied: (i) the direct global approach, where the constitutive equations are taken to be the iterative part of the global governing equations, (ii) the local (or indirect global) approach, where the constitutive equations are solved separately in each step of the global iteration, and (iii) the partly reduced approach, which is the combination of (i) and (ii). The approaches are compared with regard to the number of global iterations and the total number of floating point operations. The direct global approach is found to be the best choice. (C) 2008 Elsevier B. V. All rights reserved.</dc:description><dc:publisher>Elsevier</dc:publisher><dc:date>2008</dc:date><dc:date>2015-07-10 10:12:26</dc:date><dc:type>Neznano</dc:type><dc:identifier>32142</dc:identifier><dc:language>sl</dc:language></rdf:Description></rdf:RDF>
