Seismic Behavior and Modeling of T-shaped Reinforced Concrete Walls under High Axial Force Ratios

This paper presents large-scale quasi-static tests conducted to investigate the seismic behavior of slender T-shaped walls under a high axial compression force ratio of approximately 0.19. Four specimens were designed per various codes and design methods for comparison. All specimens failed in a flexural mode, characterized by the crushing of concrete and buckling or fracture of boundary longitudinal rebars at the web toe. The specimens designed per the US code ACI 318-19 provisions and designed using displacement-based method exhibited satisfactory deformation capacity with an ultimate drift exceeding 2.0%, which validates the effectiveness of the provisions and displacement-based method for T-shaped walls under high axial force ratios. However, the specimen designed per the Chinese code GB 50011-2010 experienced a sudden drop in strength after 1.0% drift in flange-in-tension loading, indicating insufficient design of the special boundary element at the web toe and the necessity of improvement of the current Chinese code provisions. The flexural strength of the T-shaped wall specimens can be accurately estimated using cross-sectional analysis. Analysis of the test data indicated that the effective stiffness of the T-shaped RC walls with an axial force ratio of no greater than 0.1 was significantly lower than 0.35EcIg, while that of the T-shaped walls with an axial force ratio of 0.19 reached 0.5-0.7EcIg. Existing equations did not provide accurate estimate of the effective stiffness of T-shaped walls. The lateral drift limit formulation developed by Segura and Wallace could reasonably estimate the lateral drift capacities for the T-shaped RC wall specimens under high axial force ratios. Finally, three conceptually different models, namely the MVLEM-3D, SFI-MVLEM-3D, and multi-layer shear element models were adopted for the simulation of T-shaped RC walls. All three models reasonably predicted the flexural strength capacity (mostly less than 15% error) and the effective stiffness (13%-43% error) of the T-shaped walls.