Bridge Safety Evaluation Based on Monitored Live Load Effects

TLDR

The paper proposes a novel method for assessing the safety of existing bridges using monitored structural responses and component conditions. It develops a limit‑state equation for strain data, introduces a condition function α(s,t) to estimate strains at unsensed locations and a prediction function ζ(s,t) to forecast extreme future SHM values, and demonstrates the approach on a Pennsylvania highway bridge monitored from 2001 to 2005. The study shows that SHM data can be integrated into structural safety evaluation, establishing a valid platform for life‑cycle, cost‑oriented, and reliability‑based infrastructure management.

Abstract

A novel approach to evaluating safety of existing bridges based on monitored structural responses and component conditions is presented in this paper. A limit state equation is developed for the measured strain data from structural health monitoring (SHM). The new concepts of the condition function, α(s,t), and prediction function, ζ(s,t), are introduced. The condition function is utilized to estimate the strains at locations other than the strain gauge locations. This function is related to the structural condition assessment results, strain gauge locations, and failure modes under consideration. The prediction function is used to predict the extreme values of the SHM data in the future. An illustration of the proposed approach is provided on an existing highway bridge in Pennsylvania, which had been monitored from 2001 to 2005 by the Advanced Technology for Large Structural Systems Center, a National Engineering Research Center at Lehigh University, Bethlehem, Pa. This study provides the basis for integrating achievable SHM data into structural safety evaluation, and establishes a valid platform for life-cycle, cost-oriented, and reliability-based infrastructure management systems using structural health monitoring.