Contact point response-based indirect bridge health monitoring using robust empirical mode decomposition

Abstract

Indirect bridge health monitoring (iBHM) has emerged as a promising technology for effective and inexpensive monitoring of bridge infrastructure. Driving speed, vehicle frequency, and bridge frequency are among the main contributors to vehicle response while it traverses over the bridge. The vehicle response is criticized as the presence of vehicle frequency can make the vehicle scanning method ineffective. On the other hand, the contact point (CP) response of the vehicle is free from vehicle frequency and can enable the extraction of more number of bridge frequencies. However, this approach suffers from accurate bridge condition assessment, including damage detection under real-life noisy vehicle measurements. In this study, the CP response of a vehicle is explored in tandem with a novel time-frequency decomposition method known as Robust Empirical Mode Decomposition (REMD) for the modal identification of bridges. A numerical study is undertaken to test the feasibility of the proposed method. The simulated parametric study involves parameters such as vehicle speed, measurement noise, and structural damage and their effect on the performance of the proposed method. Furthermore, a field application of the proposed method is verified using a full-scale study of a medium-span bridge. The CP response of a full-scale bridge is calculated from the vehicle response collected from a test vehicle. It shows that the statistical average of the results of the data collected using multiple passes of a test vehicle over the bridge can provide reasonably accurate results, similar to the direct BHM and thereby proving its capability of bridge condition assessment.