Incorporating Noise Effect in Damage Detection With a Curvature Based Method

Damage detection is essential for ensuing safety of bridges, and many damage detection methods have been developed in the past few years, which utilizes static or dynamic responses. In this paper, a curvature-based method is proposed for damage detection. In order to verify its reliability and efficacy, laboratory experiment is conducted on a beam specimen, and the measured displacement is used to identify damage with the proposed method. Damage location and severity were both identified. Taking into account measurement noise, two techniques are proposed to reduce effect of measurement noise. The first one uses averaged multiple measurement, and the other one chooses reasonable interval for curvature calculation. They are both verified with the measurements from the laboratory test.