Damage detection of warren truss bridge using frequency change correlation

Abstract

The detection of structural damage in any structure becomes the primary concern in failure analysis. Early detection of structural failure is crucial for preventing hazards and losses caused by structural failure. There are various approaches for detecting structural damage, and these approaches must be developed in large numbers for structural safety.The purpose of this study is to provide a simple and effective method for estimating the damage to a warren truss bridge utilizing frequency change correlation and the results of the modal analysis performed using SAP 2000 FEA software. The numerical results of the first five-mode shapes and frequencies of the Louisville bridge model were validated for modal analysis accuracy in SAP 2000. The primary variable in the modal analysis was the presence of six different sections of hollow mild steel pipe with an outer diameter of 10, 12, 14, 16, 18, and 20 mm and a 2 mm plate thickness. Three different locations for damage have been considered in the warren truss bridge. The first six modes' natural frequencies and modal participation factors were compared for undamaged and damaged warren truss bridges. The results indicate that as the diameter of warren truss bridge sections increases, the natural frequency and modal participation factors of the first six modes remain constant in the undamaged case but increase in the three-damage scenario. Additionally, there was a slight change in natural frequency and modal participation factor for three damage scenarios. As a result of increasing the natural frequency and modal participation factor for damage scenarios, the health of the warren truss bridge may be monitored, and appropriate measures adopted before it fails.