A Digital Twin Framework for Structural Health Monitoring of Existing Large-Span Bridges

Area of Science:

• Civil Engineering
• Structural Health Monitoring
• Cyber-Physical Systems

Background:

• Large-span bridges are vital infrastructure susceptible to degradation from environmental factors and fatigue.
• Existing Digital Twin (DT) implementations often require dense sensor networks and are limited in practical application for bridges.
• Sparse sensing presents a significant challenge for effective bridge monitoring and management.

Purpose of the Study:

• To propose a novel Digital Twin (DT) framework for monitoring and managing large-span bridges under sparse sensing conditions.
• To enhance the scalability and practical applicability of DT technology in civil infrastructure.
• To develop a decision-oriented system for optimizing sensing, inspection, and maintenance strategies.

Main Methods:

• An information-centric DT framework integrating physics-based modeling, data-driven learning, and uncertainty-aware inference.
• Full-field state reconstruction to complement limited physical measurements from sparse sensors.
• Introduction of a synchronized reference configuration (State 0) and a dynamic re-baselining approach (Dynamic State 0).

Main Results:

• Demonstrated a scalable DT framework capable of operating with sparse sensing for large-span bridges.
• Successfully reconstructed the full structural state by combining limited data with advanced modeling and inference.
• Enabled adaptive and decision-oriented recommendations for structural management.

Conclusions:

• The proposed DT framework offers a practical and scalable solution for monitoring aging large-span bridges with limited sensor data.
• Integrating physics-based models, data-driven learning, and uncertainty quantification is key to effective sparse sensing in structural health monitoring.
• The framework supports proactive and optimized maintenance planning, enhancing bridge longevity and safety.