DAMAGE LOCALIZATION SYSTEM FOR SINGLE SPAN SINGLE SPAN BRIDGE STRUCTURE

• Main Author: Dritama, Ryan
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/56882
• Institution: Institut Teknologi Bandung

Bridge inspections are currently still carried out manually and visually observed, making it difficult to locate bridge damage and inefficient in terms of time. This damage if left unchecked can get worse so that it has the potential to cause the bridge to collapse so that it can cause casualties. To solve this problem we need a system that is able to detect the location of damage to the bridge automatically and inform the user. The system can predict the location of the damage by observing the dynamic response of the bridge structure in the form of the natural frequency of the bridge generated by passing trucks. The natural frequency response of the bridge is then further processed using machine learning techniques so that the system is able to predict the location of the bridge damage. The dataset used to train the machine learning model was collected through a laboratory-scale tesbed bridge experiment. Prediction of the location of the damage is divided into 8 segments of the bridge surface area. The results of testing the machine learning model have been able to estimate the location of the bridge damage.