PSI - Issue 64

7

Cevdet Enes Cukaci et al. / Procedia Structural Integrity 64 (2024) 531–538 Cukaci and Soyoz / Structural Integrity Procedia 00 (2024) 000 – 000

537

Video recordings for vision-based modal identification are taken from three locations and five different angles, on supports strategically chosen to minimize structural vibration response and focus on cable responses. This method enables the analysis of 42 cables' displacement responses and natural frequencies without attaching any targets to the structure. A single camera, an optical lens and a tripod setup makes it feasible to measure all cables' responses efficiently. Fig. 9 shows the displacement response from vision-based modal identification and PSDs as examples.

Fig. 9 Selected vision-based modal identification results for cable-210.

Comparing results from the lift-off testing, vibration method with accelerometers, and vision-based modal identification, discrepancies in cable tension determination are minimal: less than 9% between the lift-off and accelerometer-based methods, under 10% between the vision-based method and the lift-off testing, and less than 5% between the two vibration methods. Differences may arise due to single-strand measurements with force sensors in lift-off testing. However, the consistency of the vision-based method confirms its validity in estimating cable tensions, as indicated by the close results compared to other methods as shown in Fig. 10.

Fig. 10 Comparison of cable tension between existing monitoring system and vision-based modal identification results.

6. Conclusion Civil infrastructures like buildings and bridges experience material degradation over time, necessitating maintenance for public safety. SHM systems play a vital role in ensuring the structural health of these structures, especially cable-stayed bridges where cable tension is critical. The decrease in cable tension due to factors like fatigue, corrosion, and environmental conditions can shorten a structure’s service life. Traditional monitoring methods that require individual sensors for each cable are expensive and labor-intensive for continuous use. This study presents an alternative using a vision-based method with a consumer-grade camera to assess the dynamic response of cables without any target. The study applied this vision-based technique to the Komurhan Cable-stayed Bridge, favoring its simplicity and cost-effectiveness. This process involved edge detection and movement tracking from various recording angles, successfully capturing the dynamic responses of 42 cables. The vision-based modal identification gives consistent results with the lift-off testing and vibration method obtained from the existing accelerometers. This method's advantages include measuring multiple cables simultaneously with a single camera setup, greatly reducing the need for extensive sensor networks and associated costs. Despite potential accuracy trade-offs, such as