Parameter optimization for Pivot hysteresis model for reinforced concrete columns with different failure modes

Reliable hysteretic modeling is essential for the evaluation of existing reinforced concrete (RC) buildings. Also, tall buildings and other structures of high importance require reasonably accurate seismic nonlinear time-domain analysis for design, which includes load-displacement behavior and hysteretic modeling. However, the hysteretic behavior of RC members is very complex and challenging to predict. Currently, the Pivot hysteresis model is used to predict the hysteretic response of members with parameters alpha and beta representing the unloading stiffness and pinching, respectively, determined using the geometric and reinforcement details in column members. However, the hysteretic response is fairly different for flexure and shear failure modes and must be accounted for to derive accurate hysteretic predictions. In this paper, an effort has been made to improve the existing approach for columns by deriving alpha and beta separately for flexure and shear modes of failure. The chosen variables to control alpha and beta are axial load ratio (ALR), longitudinal and transverse reinforcement indices. Equations for alpha and beta were calibrated through optimization of energy dissipation from 113 column specimens subjected to cyclic loading, using an advanced optimization algorithm known as Simulated Annealing. The proposed formulations can improve the accuracy of the Pivot hysteresis model for column members by capturing pinching and stiffness degradation more reasonably under different failure modes. The proposed approach can aid design engineers in performing a more reliable and accurate analysis of complex structures.