Damage detection using curvatures obtained from vehicle measurements

Abstract

This paper describes a new procedure for bridge damage identification through drive-by monitoring. Instantaneous Curvature (IC) is presented as a means to determine a local loss of stiffness in a bridge through measurements collected from a passing instrumented vehicle. Moving Reference Curvature (MRC) is compared with IC as a damage detection tool. It is assumed that absolute displacements on the bridge can be measured by the vehicle. The bridge is represented by a finite element (FE) model. A Half-car model is used to represent the passing vehicle. Damage is represented as a local loss of stiffness in different parts of the bridge. 1% random noise and no noise environments are considered to evaluate the effectiveness of the method. A generic road surface profile is also assumed. Numerical simulations show that the local damage can be detected using IC if the deflection responses can be measured with sufficient accuracy. Damage quantification can be obtained from MRC.

The Feasibility of Using Laser Doppler Vibrometer Measurements from a Passing Vehicle for Bridge Damage Detection

“…Curvature. Curvature at each point in time can be estimated from deflection measurements at three different positions on the bridge [29]. Figure 6 shows a contour plot of deflections for the healthy bridge without considering a road profile.…”

“…The approach is able to detect localized damage, but vehicle calibration is needed to use the method in field applications. OBrien et al [29] propose using curvatures of the bridge deflections at a moving coordinate for damage detection. They assume that the bridge deflection can be measured from a passing vehicle.…”

“…The method is based on the concept of Instantaneous Curvature (IC). It is shown that curvature is sensitive to damage and can detect the presence and location of damage [14,29,30]. Since velocities are measured on a TSD, Rate of Instantaneous Curvature (RIC), the first time derivative of IC, is introduced for damage detection.…”

Proposal for loading tests of bridges - simulation and laboratory experiments

“…Identification of the dynamic properties of bridges and damage detection is a current topic of research in many institutions (e.g. O´Brien, 2017). The damage diagnosis as a successive step can apply also many ways how to reveal the possible damage reaching from directly measured time series, identified natural frequencies (Kashangaki, 1991) and mode shapes (Padney, 1995;overview by Salgado, 2008) over the response functions (Sampaio, 2003) or power spectral densities (Li, 2015), of course also under consideration of various statistical methods (Faber, 2000;Mainini, 2017).…”

“…However, the available time for a precise analysis of experiments during vehicle passage is a limiting factor at higher vehicle speeds and the intensity of vibrations induced by the driving vehicle is a limiting factor at lower speeds due to the environmental noise (or other sources of vibrations). Another limiting factor is said to be also the surface roughness (O´Brien, 2017). Therefore, a solution of dynamic excitation was sought that provides us with a higher intensity of dynamic load also at lower speeds of the passing loading-testing vehicle, and possibly not sensitive to the surface roughness.…”