Seismic Performance of FRP Retrofitted Exterior RC Beam-Column Joints under Varying Axial and Bidirectional Loading
Thesis DisciplineCivil Engineering
Degree GrantorUniversity of Canterbury
Degree NameDoctor of Philosophy
Most of the experimental studies available in literature on the seismic assessment and retrofit of existing, poorly detailed, reinforced concrete (RC) beam-column joints typical of pre-1970s construction practice have concentrated on the two-dimensional (2D) response, using unidirectional cyclic loading testing protocols and a constant column axial load. Even more limited information is available on the performance of corner three-dimensional (3D) RC beam-column joints with substandard detailing subjected to a bidirectional-loading regime. In addition, little effort has been dedicated to the development of simple but reliable analysis and design procedure for FRP-strengthened joints. This thesis aims to (1) investigate the effects of varying axial and bidirectional loading on the seismic performance of deficient exterior RC beam-column joints before and after retrofit and (2) develop performance-based seismic assessment and FRP-based retrofit procedures for exterior and corner beam-column joints. For this purpose, following a critical review on the seismic vulnerability of both existing and retrofitted exterior and corner beam-column joints under varying and bidirectional loading demands, a comprehensive experimental programme along with analytical and numerical studies are carried out. A performance-based retrofit approach was adopted in order to achieve the desired ductile failure mode by modifying the hierarchy of strength within the beam-column joint system. In order to achieve this, existing retrofit design methodology available in literature was refined and a step-by-step procedure was proposed for the assessment of the as-built and proceeding retrofit design of FRP-retrofitted exterior beam-column joints. In addition, the role and importance of accounting for the correct demand conditions (e.g., the variation of axial loads) in the assessment of the existing joint and the design of the FRP retrofit scheme were discussed. In order to assist the retrofit design and assessment procedure a semi-empirical analytical model was developed to evaluate the joint shear resistance after FRP retrofit. Both the proposed assessment methodology and design procedure along with the analytical procedure were verified by experimental studies performed in this thesis and experimental tests available in the literature. Parametric analyses were also carried out to indicate the Umut Akguzel Seismic Performance of FRP Retrofitted Exterior RC Beam-Column Joints under Varying Axial and Bidirectional Loading effectiveness of strengthening with different materials, configurations and failure limit states. The feasibility and efficiency of a retrofitting intervention using GFRP composites were investigated based on the quasi-static cyclic tests conducted on four 2D exterior (plane frame) and two 3D corner (space frame) RC joints. All specimens were of 2/3 scale, designed according to pre-1970s construction practice. The properties of the specimens are summarized briefly as follows: (1) Specimen 2D1 was tested under varying axial load as an exterior 2D benchmark unit; (2) Specimen 2D2 was tested under constant axial load with minimum retrofit solution designed according to the proposed methodology; (3) Specimen 2D3 was tested under varying axial load with the same retrofit scheme adopted in 2D2; (4) Specimen 2D4 was tested under varying axial load with improved retrofit scheme; (5) Specimen 3D1 was tested under bidirectional loading with varying axial load as a benchmark; (6) Specimen 3D2 was tested under bidirectional loading with varying axial load with improved retrofit scheme adopted in Specimen 2D4. The test outcomes highlighted the potentially unconservative effects of neglecting the actual multiaxial load demand, when assessing the behaviour of existing beam-column joints and designing a proper retrofit intervention. Corner beam-column joints within a frame building were confirmed to be particularly vulnerable. However, with an adequate retrofit design accounting for the multiaxial load effects, the implemented retrofit solutions provided the necessary improvements of the behaviour of the as-built specimens. This resulted in the development of a more appropriate hierarchy of strength with the formation of plastic hinges in the beam and protection of the weaker mechanisms. The experimental findings were also used to identify the critical damage limit states and engineering parameters to be adopted within the framework of performance-based seismic retrofit design. In addition to the experimental and analytical studies, FEM numerical studies based on microplane concrete model approach were carried out. Three-dimensional finite element (FE) models for retrofitted 2D exterior joint (Specimen 2D2) and as-built 3D corner joint (Specimen 3D1) were developed and analysed. The analysis results showed good agreement with the experimental results.