Real -Time Damage Forecasting in Smart Bridge Structures Using Digital Twins and Deep Learning: A Case Study of an Urban Bridge in Tehran

Urban bridge infrastructures face increasing demands for safety, reliability, and real-time responsiveness, especially in high-density metropolitan areas such as Tehran. This study presents a novel framework for real-time damage forecasting in smart bridge structures by integrating digital twin technology with deep learning algorithms. A digital twin model of a reinforced concrete urban bridge in Tehran was developed using structural monitoring data including strain, acceleration, and displacement sensors. The system continuously synchronizes virtual and physical environments to capture evolving structural behaviors under both operational and extreme loading conditions. Convolutional neural networks (CNNs) and long short-term memory (LSTM) networks were trained on historical SHM data to detect early signs of deterioration and predict damage progression in real time. The results demonstrate that the proposed hybrid framework achieves high prediction accuracy and enables proactive maintenance scheduling. This integration of data-driven modeling and virtual-physical synchronization holds promise for enhancing resilience and service life in critical bridge assets within smart city infrastructure systems.

کلیدواژه ها:

Digital Twin ، Deep Learning ، Structural Health Monitoring (SHM) ، Real-Time Damage Forecasting ، Smart Bridge ، Urban Infrastructure ، Tehran Case Study