الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 189 |  |  |
| | | from | 189 | | |
Abstract
This study investigates numerically the behaviour of substandard detailed reinforced concrete (RC)inverted beam-column knee joints under a closing moment. The role of such joints in internal forces redistribution and stability of RC frames is evaluated using both linear and nonlinear finite element analysis. RC frames are commonly used system in RC buildings. The philosophy behind the proper design for this type of frames is to provide them with sufficient ductility. The structural ductility of a frame is mainly determined by the ductility of its components, i.e. the beams, columns and joints forming this frame. Previous research shows that building’s damage may be caused by inadequate reinforcement details of its beam-column joints. Deficiency in joints performance is related to inadequate codes guidelines or to exceptional event configurations, for example, the removal of a ground corner column turns an exterior joint into an inverted knee joint. Code provisions for exterior joints are indeed not sufficient to knees ones. The effects of the resulted deficiencies, i.e. the absence of joint vertical stirrups and improper column bar anchorage, are evaluated in this study. The first part of this research focuses on the global numerical modelling of a RC frame after the loss of one of its ground corner columns, using SAP 2000 software. The role of the inverted knee joint above the removed column in redistributing internal forces and in resisting progressive collapse is discussed. For validation purposes, the numerical models are used to reproduce the response of previously studied RC frames; the first reference frame is selected because of the limited level of deformations as the frame is only subjected to dead loads; the second reference is selected because of two reasons: first, the availability of more detailed nonlinear analysis results using a layered beam approach; second, the limited length for bar anchorage or the absence of joint vertical stirrups will have an obvious impact on frame response as the frame has dropped beams (beams with thicknesses bigger than slab thickness). Parametric studies are conducted to determine the variance in the models response based on various plastic hinges properties established according to joints reinforcement conditions. Nonlinear static and dynamic analyses are performed to predict plastic hinge formation sequences and to evaluate the level of plastic deformations in each hinge. The plastic hinges formation sequence is considered as an indication for the frame resistance against progressive collapse.More detailed investigations of the role of other parameters in joint ultimate capacity are presented in the second part. The second part of this research focuses on the local numerical simulation of a standard beam column knee joint with reference to published results of an experimentally tested joint. Micro and macro models are developed using LS-DYNA and the OpenSees platform. These models are based on a suitable concrete constitutive model, a reinforcing bar plasticity model and validated algorithms for bond-slip formulation. The micro models are able to capture the joint ultimate capacity, the joint shear strength and the cracking failure mechanisms. These numerical models are then used to simulate the substandard beam-column joints appearing in RC frames after ground corner column removal. The macro models with a modified joint panel shear stress-strain relationship are able to assess the joints real capacity and these models are extended to evaluate other parameters: concrete compressive strength, beam reinforcement ratio and beam bars anchorage type. This study ends with the development of strut and tie models (STMs) for a substandard knee joint to evaluate the role of joint transverse reinforcement and improper columns bar anchorage. My work contributes to the study of measures to avoid progressive collapse of RC framed structures. The contribution focuses on the development and exploitation of numerical models for these frames and especially the joint above a lost corner column.My study combines the codes SAP2000, LS-DYNA and OpenSees to provide a multilevel approach in numerical simulations. The global RC frame modelling with SAP2000 covers pushdown analysis for the investigation of the RC frame ultimate load carrying capacity and establishment of collapse modes. To assess the actual damage of a RC frame, three different characteristics for plastic hinges expected to be formed at the joint above the lost column, are used and termed normal, shear deficient and anchorage deficient. LS-DYNA is used as it includes explicit time integration solver techniques which preserveconvergence even at high nonlinearity level.OpenSees is chosen for the analysis of the considered knee joints, since its library includes a more detailed joint model that can explicitly consider various joint failure modes. This joint model is also compatible with nonlinear beam-column element formulations. A modified joint panel shear.