Dynamic seismic performance assessment of buildings based on digital twin technology

Abstract

The increasing complexity and vulnerability of urban structures under seismic excitation demand real-time, accurate, and adaptive assessment methodologies. Traditional earthquake response evaluations, primarily based on static finite element analysis or delayed post-event data processing, often fail to capture the evolving structural states during an earthquake. This study proposes a dynamic seismic performance assessment framework grounded in digital twin technology, which integrates real-time sensor feedback with physics-based simulation. The system continuously updates internal model parameters through a synchronization mechanism, enabling accurate tracking of displacement, strain, and inter-story drift under seismic events. Experimental validation on three typical structural types—steel frames, shear walls, and hybrid systems—under both historical and synthetic ground motions demonstrates the proposed method’s superior accuracy, robustness, and responsiveness compared to static models, sensor-only estimations, and data-driven alternatives. The results indicate that the digital twin framework is well-suited for practical deployment in earthquake-prone regions, providing timely and reliable decision support for infrastructure resilience.