Kinematic Modelling of the Shear Behaviour of Coupling Beams in Wall Structures

Abstract

Kinematic Modelling of the Shear Behaviour of Coupling Beams in Wall Structures University of Liège - Faculty of Applied Sciences Franssen Renaud - Master student in Civil Engineering Academic Year 2015-2016 Promoter - Boyan Mihaylov

The aim of this thesis is to develop a kinematic model capable of predicting the shear strength and deformation patterns of reinforced concrete coupling beams in coupled wall systems. To assist in the development of the model, data from tests of coupling beams has been collected from literature. The obtained database contains 30 tests of conventionally reinforced coupling beams featuring different failure modes.

Three ways of modelling coupling beams are summarized from the simplest one, strut-and-tie models, to the most complex one, non-linear finite element models (FEM). The intermediate approach discussed in the thesis is a three-parameter kinematic theory (3PKT)\up{\citep{3PKT}} for the shear behaviour of continuous deep beams. Because of the similarities in shear behavior between deep beams and short coupling beams, this approach is used as a foundation for the development of a kinematic model for coupling beams.

To improve the understanding of the behavior of coupling beams and to assist in the development of the kinematic model, non-linear finite element analyses were performed of selected series of coupling beams from the database. These analyses were performed using the program VecTor2 based on the modified compression field theory\up{\cite{mcft1,mcft2}}. It is found that this approach provides reliable predictions of failure load, failure mode, and load-displacement response of coupling beams.

In the main part of the thesis, the kinematic model for continuous deep beams is modified to capture the shear strength of coupling beams. Appropriate physical assumptions are made to reflect the specific features of the shear behaviour of coupling beams. The modified model applies to diagonal shear failures occurring prior to yielding of the longitudinal reinforcement. Six tests from the database exhibited such failures and for these tests the model produced adequate predictions of shear strength.

To extend the validation of the proposed kinematic model, new shear strength data was generated by performing parametric studies with FE models. By comparing the results from the two approaches, it is shown that the relatively simple kinematic model approximates very well the FEM shear strength predictions for coupling beams with variable the aspect ratios, shear reinforcement ratios, and concrete strength.