Damage-Sensitive Feature for Long-Span Steel Box-Girder Suspension Bridges Under Service Loads Based on Long-Term Monitoring

Abstract

To extract valuable data from long-term monitoring and identify structural damage to ensure the serviceability and safety of bridges, a commonly used method in large bridge maintenance. This article developed a time-domain energy-based method for identifying bridge damage based on long-term structural health monitoring noisy output measurements. This method first calculated an energy threshold within the structural frequency bandwidth of natural vibration at measurement points. Then, the structural damage was assessed by computing the time-domain energy accumulation. The improved procedure of this proposed method involved extracting modal frequencies from the signal for each modal order, determining the structural frequency bandwidth of natural vibration, and applying the square root envelope method to smooth vibration data and mitigate the impact of abnormal signals. Later, the applicability of the proposed method for damage identification was verified by comparing it with conventional modal flexibility and modal curvature methods. Finally, a case study on the Runyang bridge, a supported steel-box-girder suspension bridge with a main span of 1490 m in China, was made for demonstration. The results indicated that the area between 1/4-span and 1/2-span of the bridge was more sensitive to damage under service loads. The method performed well in identifying potential damage locations and assessing vulnerability. The calculation results using the method proposed in this article were consistent with those of traditional methods. The research offers methodological backing for utilizing extensive data monitoring in evaluating structural damage.