A Real Time Stochastic Scour Detection For Railway BridgesRIDGES

Abstract

This study aims to propose a real-time stochastic scour detection (SSA) of railway bridges by means of ambient vibration monitoring. A remote ambient vibration monitoring system is deployed in a railway bridge to measure ambient vibration. To detect the scour from the ambient vibrations, a non-Gaussian approach is proposed since the identified frequencies are scattering and the probability distribution of the identified frequencies sometimes show non-Gaussian distribution with fat tails. Observations from the ambient vibration data before and after a flood event showed that target frequency was slightly decreased under the increase of water surface during swollen river water period. However, the decreased frequency recovered once the water surface level decreased to normal river water level. The proposed real-time SSA showed an extremely low probability of scour of the target bridge due to the swollen river water period.