Reliability-Based Condition Assessment of Bridge Deck Using Long-Term Monitoring Data

Abstract

We propose a reliability-based approach for progressive assessment of structural condition of bridge deck sections, and apply the proposed method to the suspension Tsing Ma Bridge with the use of long-term strain monitoring data. After performing wavelet multi-resolution analysis to extract the live-load-caused strain components from raw measurement data, the strain sequences acquired from different structural members on the same deck section are synthesized to derive the dynamic internal forces (axial force, bending moment, and shear force) of the monitored deck section. The Weibull mixture modeling is then carried out to characterize the probability density distribution of the peaks of each internal force component, and the corresponding reliability index is obtained. The reliability index of the deck section is evaluated by considering the different internal force components in series connection. The evaluation result of reliability index at deck section level is obtained year by year with cumulated strain monitoring data, thus tracking the variation in structural condition of the bridge deck evolutionarily.