PSI - Issue 64

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

532 2

load is transferred to the pylon by adjusting the tensions of multiple cables. Structural Health Monitoring (SHM) systems are crucial for monitoring the condition of bridges and infrastructure over time. By continuously assessing a structure's reliability, integrity, and durability, SHM systems help to ensure safe operation throughout a structure's lifetime (Camara et al., 2014). SHM has been widely applied to both new and existing bridges for purposes such as early damage detection, real-time safety evaluations after significant events, prioritizing maintenance schedules, monitoring the effects of repairs, informing future design guidelines, and supporting bridge engineering research, especially in the areas of earthquake and wind resilience using advanced materials (Ko & Ni, 2005). The stay cables are crucial in load-bearing and preserving the structural integrity of stay cable bridges. Continuous monitoring of cable tensions is key for the integrity and safety of the structure during both construction and use. The changes in cable tensions can result from various factors, including vibrations and fatigue due to traffic and wind loads, environmental corrosion, and the change in material properties, which causes the tension to be redistributed to the other cables. This, in turn, can negatively affect the bridge's overall performance. To ensure the integrity and safety of stay cable bridges, it is essential to measure the cable tensions continuously and accurately and monitor the bridge operation (Pipinato et al., 2012). In cable-stayed bridges, the tension of the cables is a critical metric for assessing the structural health and locating any damage. There are two main methods for determining cable tension: one is to measure the tension directly, and the other is to estimate the tension by utilizing the dynamic behavior and characteristics of cables. In the direct measurement method, an adapted device is used to measure the load. The direct methods are lift-off testing (Cho et al., 2013), strain gauge (Guo & Chen, 2011), fiber Bragg grid (FBG) sensors (Li et al., 2009), and elasto-magnetic (EM) vector (Cappello et al., 2018). Since direct methods are mechanical methods, cable tensions can be measured more accurately. However, they are generally used on bridges under construction, considering sensor placement, replacement, and continuity issues. The vibration method is an indirect, cost-effective, and convenient approach commonly used to estimate cable tensions in existing bridges. While not as precise as direct methods, the vibration approach allows for measurements during both construction and bridge use and benefits from more economical and easier-to-install measurement equipment. The most widely used form of this method involves contact accelerometers that record the acceleration response of the structure (Yang et al., 2016). However, installing these sensors, along with their wiring and connections to data acquisition systems on bridges with numerous cables, can be a laborious and expensive process. Wireless sensors present a solution to these challenges due to their flexibility and low installation costs, enabling efficient acceleration response measurements. Nevertheless, these sensors still require physical access for installation, which can be problematic in high-traffic conditions or on hard-to-reach parts of a bridge. Non-contact sensors offer a solution to the drawbacks of contact sensors in monitoring the cable tensions. These methods are the laser Doppler technique (Nassif et al., 2005) and vision-based monitoring systems (Feng et al., 2017). Laser Doppler vibrometers are highly accurate in capturing dynamic responses of the cables but are less cost-effective due to the need for one sensor per cable. Conversely, vision-based systems are cost-effective and straightforward to operate, using industrial or consumer-grade cameras, lenses, and tripods to measure dynamic responses at multiple points simultaneously with the same recording. The convenience, affordability, and ability to cover multiple points simultaneously have made vision-based systems a popular alternative to traditional monitoring methods. Therefore, in this study, cable tensions are indirectly estimated by measuring dynamic responses using a vision based modal identification without using any targets. Cable tensions are determined using the lift-off testing and estimated with the vibration response under four seismic events and three hours before them recorded by the existing monitoring system on the bridge. The study first determines cable tensions on the bridge through the lift-off testing and acceleration response. Subsequently, a vision-based method is utilized to estimate all cable tensions. While the initial methods monitor only 8 out of 42 cables continuously, the vision-based system estimates tensions for all cables, accurately aligning with the existing monitoring system. 2. Methodology In this study, cable tensions are directly determined in the lift-off testing using force sensors and estimated using the vibration method from acceleration response of cables with the existing monitoring system on the bridge. Then,