Abstract

The use of C-shaped reinforced concrete (RC) core walls as the main lateral force resisting system for building structures is a popular choice for medium- to high-rise buildings. These cores are typically closed on three sides (C-shaped) and are either open or partially open on the fourth side. Despite the frequent use of C-shaped RC core walls as the primary seismic force resisting system (SFRS) for multi-story buildings, there are still challenges in estimating their inelastic seismic response, whether they are separate or coupled C-shaped walls. The eccentricity between the centre of mass and the centre of rigidity tends to induce a significant torsional seismic demand on such walls. To address this issue, several design codes have proposed recommendations for evaluating the effect of this eccentricity. Although National Building Code of Canada (NBCC, 2015) specifies provisions to consider the accidental eccentricity and the torsional sensitivity of the structure in the design, there is a gap in the knowledge of the inelastic structural response of C-shaped RC core walls for new buildings. On the other hand, many existing RC C-shaped walls are in need for retrofitting in order to meet current seismic design codes, or to meet increased demands due to change in the use and occupancy of the building, or to retrofit post-earthquake damages. In recent years, application of fibre-reinforced polymer (FRP) composites in retrofitting of RC walls has considerably increased due to its advantages such as high strength to weight ratio and its fast and easy installation.
The objectives of this thesis are to: (i) investigate, numerically, the effectiveness of FRP retrofitting on the seismic performance of RC shear wall systems, (ii) evaluate the deficiency of seismic design provisions of C-shaped shear wall structures with high torsional sensitivity, (iii) examine, experimentally, the seismic response of FRP-retrofitted C-shaped RC walls to quantify the efficiency of the FRP retrofitting in repairing the RC core walls.
To achieve the first objective, a simplified modelling approach was proposed for analyzing the FRP retrofitted RC walls in order to be used as a simple and efficient method in practice. Nonlinear Incremental Dynamic Analysis (IDA) of a typical twelve story RC building structure, following the FEMA P695 methodology, showed that although the increase of torsional sensitivity has no significant effect on the inter-story drift ratios of the building, it could significantly decrease the Collapse Margin Ratio (CMR).
To achieve the second objective, building structures with different levels of height and torsional sensitivity were studied. Results showed that although response spectrum analysis (RSA) provides consistent predictions for story shear demand in regular buildings, significant underestimation of design forces might be obtained for buildings with a torsional sensitivity of B ≥ 2.0. Dual Plastic Hinge (DPH) method was found to be an efficient alternative in reducing the story shear demand in structures with high torsional sensitivity, compared to structures designed based on the common Single Plastic Hinge (SPH) method.
To achieve the third objective, a previously tested large-scale C-shaped RC wall was retrofitted using carbon fibre-reinforced polymer (CFRP) composite sheets and was tested under the same condition and loading protocol of the original wall to quantify the efficacy of FRP retrofitting on the wall’s response. Furthermore, multi-directional cyclic tests were conducted in order to evaluate the complete nonlinear response of the FRP retrofitted C-shaped wall up to failure. The assessment was based on experimental measurements and observations in terms of 3D displacements, strains (both from strain gauges and Digital Image Correlation, DIC, system), crack pattern, ductility, curvature profiles and mode of failures. The test showed that the FRP retrofitting scheme used in the current work performed very well by enhancing both the strength and ductility of the retrofitted wall from its damaged state, while holding its stiffness very close, compared to those of the control wall.