Real-Time Dynamic Warning on Deflection Abnormity of Cable-Stayed Bridges Considering Operational Environment Variations

Abstract

Long-span bridges face various threats every day due to their sophisticated operational environments. To investigate the operational state of bridges, a real-time dynamic warning method on the abnormity of cable-stayed bridges is proposed based on deflection measurements. The generalized Pareto distribution (GPD) model and finite-element (FE) calculation are used to determine the basic warning threshold. Considering the complicated components of the monitoring deflection signals (e.g., measurement noise, thermal effect, and vehicle load-induced effect), each signal component should be further modeled and investigated to improve the accuracy and effectiveness of the warning system. Because thermal effects are usually not the signal component of interest that may cover the signal fluctuation induced by abnormity, they should be separated from the raw signals. Then, in view of the variations of operational condition over time, the warning threshold is updated termly according to the operational period by using statistical approaches. Finally, the effectiveness of the proposed warning methodology is validated by analyzing the recorded deflection data from the third Nanjing Yangtze River Bridge. As a result, the proposed dynamic warning method is more effective than the conventional static one and will help bridge management departments identify emergencies in time and take inspection or maintenance decisions to ensure structural safety.