Titre : | A nonlinear finite element study of reinforced concrete beams | Type de document : | texte imprimé | Auteurs : | Akthem Abdulkarim Al Manaseer, Auteur ; D. V. Phillips, Directeur de thèse | Editeur : | Glasgow : University of Glasgow | Année de publication : | 1983 | Importance : | 410 f. | Présentation : | ill. | Format : | 27 cm. | Note générale : | Thèse de Doctorat : Génie Civil : Royaume-Uni, University of Glasgow : 1983
Bibliogr. [28] f. Annexe [16] f | Langues : | Anglais (eng) | Mots-clés : | Nonlinear -- finite element method
Reinforced -- concrete beams
Smeared -- crack approach
Endochronic -- theory
Isoparametric -- elements
Modified -- newton-raphson approach
Tension -- stiffening effects | Index. décimale : | D003383 | Résumé : | This thesis describes the development of plane stress, nonlinear finite element method of analysis for reinforced concrete beams.
These include simple, deep, and T-beams, failing in flexure and in shear.
The nonlinear response is assumed to be caused by concrete cracking, nonlinear biaxial stress-strain relations, and by the yielding of steel reinforcement.
The smeared crack approach was used with two models for post cracking behaviour, one with a tension stiffening effect and the other with no-tension stiffening.
The endochronic theory with some adaptations was used to account for all other uncracked zones.
8-noded isoparametric elements were used for concrete representation and 3-noded isoparametric elements for steel.
A modified Newton-Raphson approach, was used for solving the nonlinear problem with both the constant and variable stiffness methods.
This was based on the evaluation of a tangential elasticity matrix.
The unbalanced nodal forces were obtained by the method of residual forces and convergence was checked using either a force or a displacement criteria.
A nonlinear finite element program was developed where all, the required aspects to model the reinforced concrete structures were included.
It was a main contention of this work that the nonlinear solution parameters had such an important influence on the solution process that an extensive study was required to determine their effects.
This was carried out on simple and complex beams, and suitable guide lines were established.
In particular tension stiffening effects were investigated and rejected in favour of a method with no-tension stiffening used in conjunction with controls on other solution parameters.
Other parameters such as order of Gauss rule, shear retention factor, convergence tolerance, etc. were also studied in detail.
An investigation into the behaviour of a range of deep beams including perforated deep beams and beams which were heavily reinforced, was undertaken.
These beams failed both in shear and flexure.
In most cases crack pattern, stress distribution, and load deflection curves were used to validate, (or otherwise), the performance of the proposed models and program.
Finally a method is proposed for analysing T-beams using plane stress elements where the flange is treated separately and is connected to the web by a fictitious element.
Other approximations are introduced in order to treat the problem as a two dimensional structure. |
A nonlinear finite element study of reinforced concrete beams [texte imprimé] / Akthem Abdulkarim Al Manaseer, Auteur ; D. V. Phillips, Directeur de thèse . - Glasgow : University of Glasgow, 1983 . - 410 f. : ill. ; 27 cm. Thèse de Doctorat : Génie Civil : Royaume-Uni, University of Glasgow : 1983
Bibliogr. [28] f. Annexe [16] f Langues : Anglais ( eng) Mots-clés : | Nonlinear -- finite element method
Reinforced -- concrete beams
Smeared -- crack approach
Endochronic -- theory
Isoparametric -- elements
Modified -- newton-raphson approach
Tension -- stiffening effects | Index. décimale : | D003383 | Résumé : | This thesis describes the development of plane stress, nonlinear finite element method of analysis for reinforced concrete beams.
These include simple, deep, and T-beams, failing in flexure and in shear.
The nonlinear response is assumed to be caused by concrete cracking, nonlinear biaxial stress-strain relations, and by the yielding of steel reinforcement.
The smeared crack approach was used with two models for post cracking behaviour, one with a tension stiffening effect and the other with no-tension stiffening.
The endochronic theory with some adaptations was used to account for all other uncracked zones.
8-noded isoparametric elements were used for concrete representation and 3-noded isoparametric elements for steel.
A modified Newton-Raphson approach, was used for solving the nonlinear problem with both the constant and variable stiffness methods.
This was based on the evaluation of a tangential elasticity matrix.
The unbalanced nodal forces were obtained by the method of residual forces and convergence was checked using either a force or a displacement criteria.
A nonlinear finite element program was developed where all, the required aspects to model the reinforced concrete structures were included.
It was a main contention of this work that the nonlinear solution parameters had such an important influence on the solution process that an extensive study was required to determine their effects.
This was carried out on simple and complex beams, and suitable guide lines were established.
In particular tension stiffening effects were investigated and rejected in favour of a method with no-tension stiffening used in conjunction with controls on other solution parameters.
Other parameters such as order of Gauss rule, shear retention factor, convergence tolerance, etc. were also studied in detail.
An investigation into the behaviour of a range of deep beams including perforated deep beams and beams which were heavily reinforced, was undertaken.
These beams failed both in shear and flexure.
In most cases crack pattern, stress distribution, and load deflection curves were used to validate, (or otherwise), the performance of the proposed models and program.
Finally a method is proposed for analysing T-beams using plane stress elements where the flange is treated separately and is connected to the web by a fictitious element.
Other approximations are introduced in order to treat the problem as a two dimensional structure. |
|