In this dissertation, the capability of different distributed plasticity numerical models to simulate the seismic behavior of reinforced concrete (RC) axial-flexural elements, namely planar walls and square columns, is evaluated based on previous experimental results of isolated wall and column test specimens as a well as a 7-story wall building subassembly. Greater emphasis is placed on slender (flexure-dominant) structures with softening post-peak behavior (due to concrete crushing with or without rebar buckling) since this type of behavior has been commonly observed during experimental tests and after earthquakes and its prediction can be highly sensitive to the mesh size used in numerical modeling.
Results obtained in this dissertation include: 1) modeling recommendations to simulate the cyclic global lateral force-displacement (F-D) behavior of slender and squat RC walls, including a consistent mesh-sensitivity investigation; 2) a new metric to quantitatively evaluate simulated hysteretic F-D curves as compared with measured curves; 3) plastic hinge integration models that use material regularization (i.e., regularized plastic hinge models) to accurately simulate and obtain objective (i.e., mesh-independent) global and local (i.e., material strains and section curvatures) behaviors of slender planar RC walls and square columns; 4) a new confined concrete crushing energy equation for regularized constitutive models to simulate RC columns through failure; and 5) quantification of variability in simulated dynamic seismic performance of RC wall structures from different distributed plasticity models.
The results of this investigation are intended to improve current numerical modeling guidelines for practitioners conducting nonlinear analysis of axial-flexural RC walls and columns as part of performance-based seismic design. The results are also aimed for researchers conducting detailed analysis to predict global and local wall and column behaviors under seismic loading.