Drive-By Detection of Scour in a Railway Bridge
Sinem Tola, Eugene J. OBrien, Daniel Cantero, Joaquim Tinoco, José C. Matos, Tulika Bose, Jos Berkers
Abstract
This paper proposes an innovative indirect monitoring system for bridge foundation scour damage detection. The Rail Infrastructure Alignment Acquisition system, a mobile mapping system that can be attached to the front or rear of an operating train, is used for data acquisition. The difference between healthy and damaged pier stiffness responses is utilized as the scour indicator. A cross-entropy optimization algorithm finds the pier stiffness values that minimize the difference between the displacements computed with a numerical model and the measurements. The numerical model adopts the unit load theorem to determine the Moving Reference Influence Lines for a six-axle vehicle crossing the bridge, simulating Rail Infrastructure Alignment Acquisition system displacement measurements on a railway bridge in the UK. The measurements provided belong to the after-repair, i.e., the healthy stage, of the bridge. Due to the absence of scoured-stage measurements, the displacement response of the damaged bridge is synthetically generated using a vehicle–bridge interaction model. Foundation scour is modeled by reducing the vertical pier stiffness. With healthy-bridge field data and synthetically generated damaged-bridge responses superposed with rail irregularities, it is possible to detect the presence of the scour. This approach not only facilitates the early detection of foundation scour but also enhances infrastructure maintenance strategies by enabling the implementation of timely and targeted repairs, thereby improving the safety and resilience of railway operations.