Founded in 1541 and designated a UNESCO World Heritage Site in 1991, the historic center of Morelia, Mexico, includes more than 1000 architectural monuments that are vulnerable to severe seismic damage because of their age, structural types, and material degradation. This study assesses the seismic vulnerability and fragility of the façades of 577 residential and commercial masonry buildings at this heritage site. Unlike standard large-scale investigations, this study uses a refined, parameter-based methodology to incorporate microstructural material properties, digital building-height assessment, and detailed conservation-state data from visual inspections. Field data are supplemented by 3D micro-computed tomography (micro-CT) of architectural stone samples to evaluate stone porosity, which correlates with degradation of mechanical strength. Vulnerability indices are subsequently categorized using the Jenks natural breaks optimization method to minimize intra-class variance and enhance spatial mapping. Structural fragility curves are derived for four distinct damage states by coupling an adapted macroscopic damage model with localized seismic demand scenarios. The results indicate that, while the evaluated building inventory exhibits homogeneous vulnerability and acceptable performance under a standard 475-year-return-period design earthquake, an extraordinary seismic event would trigger extensive nonlinear structural damage across the building façades. These findings offer a high-resolution, georeferenced technical framework to help decision-makers implement targeted structural retrofitting, urban prevention policies, and risk-mitigation strategies in heritage cities.
Earthquakes pose significant challenges to the operational continuity and rapid restoration of substation systems, yet existing recovery strategies often lack the ability to accurately represent functional degradation and to support fast decision-making across diverse damage scenarios. This study develops a system-level functional degradation model that jointly captures equipment capacity loss and the evolution of power-transfer pathways, enabling a more realistic representation of post-earthquake supply capacity. Building on this model, a hierarchical intelligent decision-making framework is proposed. At the scenario-specific level, a reinforcement-learning-guided genetic algorithm (RL-GA) adaptively adjusts genetic operators according to the search state, achieving efficient repair-path optimization under predefined seismic scenarios. At the scenario-general level, a policy-making reinforcement learning model (RL-PM) is trained on multi-scenario samples to learn state–decision mappings, allowing rapid cross-scenario inference and supporting both post-disaster emergency repair and pre-disaster reinforcement planning. Case studies on a typical 220 kV substation demonstrate that the proposed framework reveals systematically different functional degradation characteristics, accelerates system recovery under multiple PGA levels, and reveals the differentiated roles of redundancy configuration, equipment reinforcement, and repair-resource allocation. The results highlight its potential as a unified and engineering-oriented decision-support approach for enhancing seismic resilience of critical power infrastructure.
To mitigate the construction complexity and high costs associated with conventional self-centering systems, this study proposes a novel post-tensioned (PT) precast L-shaped concrete column system, in which only unbonded PT bars traverse the column-base interface with low initial prestressing force, while the column-base joint is intentionally designed to uplift freely, thus activating a rocking mechanism. Four full-scale quasi-static cyclic tests were conducted to evaluate and compare the seismic performance of the proposed system against conventional cast-in-place columns. The results demonstrate that the novel self-centering system exhibits enhanced ductility and higher ultimate drift ratios, with minimal damage to the column itself. At failure, the system maintained a residual drift ratio of approximately 1/200(0.5%)—about 40% of that observed in cast-in-place specimen, demonstrating robust self-centering. This study further develops an analytical capacity model for the column-base interface from fundamental kinematics and section equilibrium. Closed-form computational procedures were systematically derived to identify key feature points such as the decompression, interface-opening, yielding and peak points, and to predict the force–deformation envelope. The analytical predictions matched the experimental backbone with high fidelity, indicating that the proposed analytical approach provides accurate predictions and can serve as a reliable design basis. Moreover, a set of performance-based design indices—lateral force, inter-story drift ratio, damage extent, and residual drift ratio—is established specifically for the system, providing a direct and implementable basis for performance-based seismic design. An experimentally validated finite element model was then developed to simulate the seismic response, accurately capturing the key characteristics of the self-centering behavior, including rocking-induced re-centering and damage localization, thereby demonstrating the feasibility of the proposed simplified PT-only connection for repairable L-shaped precast columns with enhanced seismic resilience. These findings indicate a constructible, cost-effective, and repairable solution and provide an analytical and numerical basis for performance-based seismic design of resilient precast concrete structures.
To overcome the brittle shear failure and repair difficulties of conventional precast concrete beam–column joints, this study proposes a novel precast beam–column connection featuring a single-sided slotted beam end and a bottom-embedded replaceable sliding friction damper, rooted in the design philosophies of damage control and equivalent monolithic performance. Quasi-static tests show that the joint achieves the ”strong-joint–weak-member” failure mode, and the beam-end damage of the joint is significantly alleviated compared with the monolithic joint. However, its initial stiffness and cumulative energy dissipation decrease by approximately 62% and 30%, respectively. A numerical model was developed in OpenSees and validated against the experimental data, based on which parametric analyses were conducted to investigate the effects of beam depth and damper slip force on seismic performance. The parametric analysis indicated that, With the other geometric and mechanical parameters kept unchanged, beam depth primarily affected the initial stiffness, load-carrying capacity and energy dissipation of the joint, whereas an excessive beam depth reduced the ultimate displacement. The damper slip force significantly influenced the load-carrying capacity, energy-dissipation performance and displacement ductility. Therefore, the beam depth and damper slip force should be coordinated according to the required load-carrying capacity, deformation capacity and energy-dissipation performance of the joint.
