Identification of Damage in a Truss Bridge Using a Moving Instrumented Vehicle from Limited Measurements

Abstract

In recent years, there has been a significant increase in the number of bridges which are being instrumented and monitored on an ongoing basis. This is in part due to the introduction of bridge management systems designed to provide a high level of protection to the public and early warning if the bridge becomes unsafe. This paper investigates a novel alternative; a low-cost method consisting of the use of a vehicle to monitor the dynamic behaviour of bridges. A truss vehicle-bridge model is used to test the effectiveness of the approach in identifying the damping ratio of the bridge with the time-delay neural networks (TNNs). A simulation study has been carried out for the incomplete measurement data. It has been observed that TNNs have performed better than traditional neural networks(NN) in this application and the arithmetic of the TNNs is simple.