SEISMIC AND WIND BEHAVIOR OF PLANAR COMPOSITE PLATE SHEAR WALLS CONCRETE FILLED (C-PSWCF)

Abstract

Composite plate shear walls / concrete filled (C-PSW/CF) are being considered to resist lateral loads (seismic and wind loads) in the design of mid- to high-rise buildings. In recognition of its modularity, construction speed, and influence on the overall project schedule and economy, C-PSW/CF is referred as SpeedCore by the American Institute of Steel Construction (AISC). Planar C-PSW/CF consist of a long rectangular steel box (made up of flange and web plates) filled with plain concrete. The steel web plates govern the length, and the flange plates govern the thickness of the planar wall. The steel web plates are connected to each other using tie bars or steel shapes, and composite interaction between the steel plates and concrete infill is achieved using tie bars or a combination of tie bars and shear studs. The steel modules consisting of plates, tie bars and shear studs are prefabricated and transported to the site. After assembly and erection, the steel modules serve as falsework for construction activities and formwork for concrete casting, which significantly reduces construction schedule. There are no conventional steel reinforcing bars in the concrete core, and the web and flange plates serve as primary steel reinforcement of the system. Experimental and numerical investigations were conducted on planar C-PSW/CF specimens to evaluate their seismic and wind behavior. Planar C-PSW/CF specimens were subjected to constant axial force and cyclic lateral loading (seismic or wind loading protocol) up to failure. This thesis presents the experimental results, test observations, lateral load-displacement responses, moment-rotation responses, moment-curvature relationships, 3D finite element results, and proposed phenomenological (effective) stress-strain relationships of the tested planar C-PSW/CF specimens. Seven large-scale C-PSW/CF specimens were tested to evaluate the effects of the following parameters on the seismic response: (i) the axial compression force levels, (ii) steel reinforcement ratios, (iii) steel plate slenderness ratios, (iv) tie bar reinforcement ratios, (v) tie bar spacing-to-wall thickness ratios, (vi) plain or threaded tie bars connections (welded or double nuts connections), (vii) flange (closure) plates or rolled shapes as boundary elements, and (viii) welded or embedded wall-to-basemat connections. Two large-scale C-PSW/CF specimens with welded wall-to-basemat connections were tested to evaluate the effects of the wind loading protocol, low cycle fatigue, and powder actuated fasteners (PAFs) in the plastic hinge region. All planar C-PSW/CF specimens exceeded their nominal flexural capacities estimated using the fiber section analysis method or the plastic stress distribution method, while accounting for the effects of axial force. The seismic lateral load-displacement behavior of planar C-PSW/CF can be characterized by: elastic uncracked (concrete) behavior, cracked (concrete) behavior, yielding of the flange plates, inelastic local buckling of the flanges, extensive local buckling of the flange and web plates, fracture initiation of flanges, propagation of fracture into the web plates, and complete fracture of flanges. In accordance with the experimental results, increasing the axial force level resulted in the higher lateral stiffness and capacity of C-PSW/CF specimens, but it decreased marginally the ductility. Local buckling of the flanges occurred after yielding of flanges; however, the inelastic local buckling of flanges did not decrease the lateral load capacity of C-PSW/CF. Fracture initiation of flanges resulted in the decrease of lateral load capacity of the specimens. The C-PSW/CF specimens could resist the wind lateral loading protocol consisting of 2,162 load cycles with no fracture failure. Hairline cracks were detected using magnetic particle inspection (MPI) at wall-to-foundation connection weld after the completion of the wind loading protocol. The hairline cracks were due to low-cycle fatigue damage caused by inelastic strain ranges. By contrast, powder actuated fasteners (PAFs) did not initiate fracture, which still initiated from the wall-to-foundation connection weld. Detailed 3D nonlinear finite element models of planar C-PSW/CF specimens were developed and benchmarked using the experimental results. The benchmarked 3D finite element results were used to develop phenomenological (effective) stress-strain relationships for the compression concrete and steel flange plates of the C-PSW/CF specimens. These effective stress-strain relationships implicitly account for the various behavioral phenomena such as steel yielding, local buckling, biaxial stress state, steel fracture, composite action, tie bar reinforcement ratio, concrete plasticity, concrete cracking, and concrete confinement. These effective stress-strain relationships were used in 2D finite element and fiber-based models to simulate the behavior of C-PSW/CF specimens with reasonable accuracy. These simpler finite element models (2D or fiber-based) can be used efficiently in analysis and design of C-PSW/CF in high-rise buildings.