Seismic retrofitting strategies for buildings using innovative

In the research, a comprehensive study on innovative approaches for retrofitting buildings equipped with various lateral force-resisting systems will be performed, including steel moment-resisting frames (SMRFs), reinforced concrete moment-resisting frames (RC MRFs), buckling-restrained braced frames (BRBFs), infilled masonry walls (IMWs), and RC shear walls (RCSWs). The main goal is to facilitate the seismic modeling and retrofitting of buildings with novel ideas to reduce the time of retrofitting and improve the seismic performance of structures, as well as to use the predictive machine learning (ML) models to evaluate seismic behavior, seismic performance levels, seismic failure probabilities, and risk assessment. The study considers both external and internal earthquake-induced forces. External forces from structural pounding between adjacent buildings will be analyzed, and probability factors will be proposed to quantify their effects. Additionally, fluid viscous dampers (FVDs) will be suggested to absorb the impact energies for the retrofitting scheme. The research evaluates the seismic performance of SMRFs with rigid and semi-rigid joints, incorporating shape memory alloy (SMA) bolts to improve their performance. Additionally, the influence of IMWs on the seismic performance of SMRFs will be examined considering soil structure interaction (SSI) effects, which confirms their influence on seismic behavior. A new concept involving SMA materials in buckling-restrained braces (BRBs) will be introduced to reduce residual drifts after earthquakes. Recognizing the limitations of existing case-based studies, the research will advance the use of conventional and ensemble ML methods to improve predictive models. These models will be developed to estimate seismic limit-state capacities, seismic performance levels, and evaluate seismic risk in various structural systems. Based on the promising results of the proposed ML models, a user-friendly graphical interface (GUI) will be developed, which facilitates future enhancements and broadens the applicability of the ML models for retrofitting purposes.