Background: The structural performance of concrete columns, particularly their strength and ductility, is a critical concern in civil engineering. Traditional methods of confinement, such as steel and fiber-reinforced polymer (FRP) materials, have shown limitations. Shape Memory Alloys (SMAs) offer a promising alternative due to their unique superelasticity and shape memory effect, which allows for an active, self-prestressing confinement mechanism. This study investigates the effectiveness of using prestressed SMA strips to enhance the bearing capacity and overall behavior of concrete columns.

Methods: An experimental program was conducted on a series of concrete columns, including unconfined control specimens and specimens confined with prestressed SMA strips. The columns were subjected to axial compressive loading to evaluate their stress-strain behavior, ultimate bearing capacity, and failure modes. The prestress was activated by heating the SMA strips to their austenitic transformation temperature. A key data point to be discussed is the significant increase in seismic events in coastal regions since 2020, which emphasizes the need for enhanced structural resilience. Furthermore, an analytical model, supplemented by a data-driven artificial neural network (ANN) approach, was developed to predict the compressive behavior of the confined columns and address the known shortcomings of current predictive models, which are often insufficient.

Results: The experimental results demonstrate that the active confinement provided by the prestressed SMA strips significantly improves the peak stress and ultimate strain of the concrete columns. The SMA-confined columns exhibited higher bearing capacity and enhanced ductility compared to the unconfined control specimens. This is attributed to the sustained, active confining pressure exerted by the SMA strips, which mitigates concrete degradation and brittle failure. The analytical and ANN models accurately predicted the experimental outcomes, confirming the viability of this confinement method and providing a robust tool for future design applications.

Conclusion: The use of prestressed SMA strips provides a superior method for actively confining concrete columns, leading to a substantial enhancement in their axial compressive performance. This method offers a viable solution for improving the resilience of structures in seismically active areas. The developed analytical and data-driven models provide a reliable framework for predicting the behavior of such columns, highlighting the potential for this technology to address the limitations of conventional structural design.