Structural Health Dignosis/Monitoring Using Neural Networks for Cable-Stayed Bridge Management System

The cable-stayed bridge is generally a highly statically indeterminate structure. The structural performance of the bridge is highly sensitive to the load distribution among major components of the bridge. Therefore, the stayed cables of the cable–stayed bridge should be monitored to prevent bridge damage due to earthquake, strong wind, differential settlement, fatigue/defect of the material or loose of tension within the cables. That makes the structure health monitoring and diagnosis of the cable forces for the optimum structural performance very important in a cable-stayed bridge maintenance procedure. This study proposes a structural health diagnosis/monitoring management system for cable-stayed bridges using Neural Networks and field measurement data. The neural networks were used to 1. Analyze reversely the corresponding axial forces of all the stayed cables using sets of rotation measured from the pylon and 2. Determine the type and degree (scope) of the damaged bridge with ease and efficiency. Based on the cable force evaluated, the structural behavior including the deformation and stress state of the bridge can be traced successfully. Also, the damage state of the cable-stayed bridge can be identified using neural networks through the measured cable forces within stayed-cables. A few cases were studied and the results obtained could be applied for Cable-Stayed Bridge Management System