SMA-braced steel frames influenced by temperature: Practical modelling strategy and probabilistic performance assessment

Abstract

This study presents a practical and efficient modelling approach for shape memory alloy (SMA)-braced steel frames under various temperatures, and examines their probabilistic performance affected by temperature. A one-dimensional constitutive model is developed in OpenSees to capture the superelasticity (SE), partial SE, shape memory effect (SME), and permanent yielding behaviors of SMA elements subjected to varying temperatures. The proposed modelling approach is partly validated against an experimental study newly conducted in the present work. The material-level investigation is followed by a comprehensive probabilistic performance assessment of a typical steel frame employing SMA-based self-centering braces (SMA-SCBs), where a practical and realistic modelling strategy for the SMA-SCB is developed to capture the degradation behavior, especially at low temperatures. Seismic risk assessment is conducted to evaluate the collapse resistance and resilience performance of the considered structures, with the influence of temperature highlighted. Among other important findings, it is shown that the decrease in environment temperature significantly increases the peak inter-story drift (PID) and residual inter-story drift (RID) responses. Under the maximum considered earthquake (MCE), the probability of collapse increases from 2.7% to 19.7% when the temperature decreases from 30 °C to −15 °C. From a self-centering capability point of view, the probabilistic of exceedance of 0.5% RID threshold soars from negligible to nearly 100%, considering the same temperature variation. The study highlights the fact that temperature plays a significant role in the annual rate of exceedance of all damage states, and warns that the idealized flag-shaped modelling approach for SMA components is extremely unsafe at low temperatures.