DAMAGE DETECTION IN BRIDGE STRUCTURE USING ENERGY METHOD AND EXPANSION OF MODE SHAPE FROM LIMITED MODAL DATA

• Main Author: Wiwiek Chandra, Willy
• Format: Theses
• Language: Indonesia
• Subjects: Teknik sipil
• Online Access: https://digilib.itb.ac.id/gdl/view/47931
• Institution: Institut Teknologi Bandung

The development of damage detection system for bridge structure is an important study to complete the structural health and monitoring system. The limited modal data due to the small number of sensors in the structural health and monitoring system poses challenges in the development of damage detection system. In this article, a new damage detection system is developed which consists of two stages: the expansion of mode shapes using Bayesian method and the localization of the damages using modal strain energy method. The damage detection system by combining the two methods is applied to a simply supported multi-girder concrete bridge structure with numerical simulation data and field-test data. Numerical simulation data consists of single and multiple damage simulations with variations in damage location and damage level. Numerical simulation data is also varied in the amount of known modal data. These variations aim to understand the accuracy of the combination of the two proposed method. In addition, field-test data is used to understand the ability of the combination of the two proposed method to be implemented in a real bridge structure. Field-test data is obtained from a bridge structure that has been damaged due to a fire. In addition, real damage record data from field inspection is used for validation. The results show that the combination of the two proposed method is able to provide a prediction of the damage location with a good correlation to the damage simulation and real damage on a simply supported multi-girder concrete bridge structure.