Determination of hanger tension force with vibration measurements

Abstract

Suspension bridges are one of the preferred bridge types for crossing long spans. The internal forces of the cable elements used in these structures have a significant impact on the structural geometry and the fatigue life of the elements. Therefore, periodic monitoring of cable forces is important for the structural health of suspension bridges. In many suspension bridges built today, hanger forces can be determined instantly with structural health and monitoring systems. Hanger forces are easily monitored with the Load Cells placed between the gusset plate on deck and the hanger lower sockets. However, due to the lack of modern structural health monitoring systems in many suspension bridges, especially those built in the 20th hanger forces cannot be monitored instantly. In these bridges, hanger forces can be calculated by vibration measurements, which is one of the alternative methods. In this study, vibration measurements were carried out on two different suspension bridges. The first bridge in the study is the Osman Gazi Bridge (Kocaeli / Türkiye), where the structural health monitoring system is located and the hanger forces are instantly monitored by load cells. The second bridge is Fatih Sultan Mehmet Bridge (Istanbul / Türkiye), where hanger replacement works were carried out and new hanger forces were measured with calibrated hydraulic jacks. Three-dimensional acceleration data recorded on hangers of different lengths and cross-sections in two different suspension bridges were used to obtain the natural vibration frequencies of the hangers with Fast Fourier Transform (FFT). Using the obtained frequency values, hanger forces were calculated with Taut Spring Theory, Axially Loaded Beam Theory, Effective Vibration Length Method, and Unified Practical Formula. Hanger forces obtained with these theories were compared with known hanger forces and the differences were determined. However, for the FSM Bridge hangers that are less than 5.6 meters the hanger tension forces calculated with inconsistent results. In order to identify the main cause of the problem which depends on the FFT results or calculation theories, the finite element model of the hangers are modelled with SAP2000. The modal frequencies of the hangers were calculated with SAP2000 and these results are used for tension force calculation theories. When the finite element results used in the theories, the hanger tension force is calculated with high accuracy. As a result, the main source of the error is identified as FFT. To enhance the precision of the FFT analysis, Low Pass and High Pass filters were applied along with windowing functions such as Hann, Hamming, and Blackman. However, these operations did not yield the anticipated level of improvement in the results. Subsequently, Ultra Precise FFT was used to determine the system's mode frequencies. Despite these steps, similar to the outcomes observed with the windowing functions, there was no notable change in the FFT results. In this study, a single accelerometer was utilized for determining the mode frequencies of the hangers. However, employing multiple accelerometers for mode shape-aided calculation with multi-acceleration measurements can significantly enhance the accuracy of the mode frequencies.