Augmenting an existing railway bridge monitoring system with additional sensors to create a bridge weigh-in-motion system and digital twin
Accepted version
Peer-reviewed
Repository URI
Change log
Authors
Abstract
Infrastructure asset managers have limited maintenance budgets and require qualitive data on the performance and utilization of their assets in order to prioritize preventative maintenance. A project investigating the potential for using digital twins for infrastructure asset management provided an opportunity to augment an already extensive fiber-optic strain-based bridge structural health monitoring system with additional sensors measuring both deck rotation and axle positions. Data from the new and existing sensors is fed to a database in near real time. In addition to a simple web-based visualization (dashboard), the data from the system can be utilized by a number of different analytical back-ends which together form a Digital Twin of the bridge. The first of these back-ends provides a bridge weigh-in-motion system, but other back ends are possible including statistical finite element analysis models or Bayesian or computer vision-based train categorization systems. This paper details the design and implementation of the augmented system including the additional hardware and software required. Constrains included the requirement to install the new sensors and cabling quickly during a time-limited overnight possession of the bridge. Challenges included the need to correctly timestamp the incoming data from the various separate sensor systems so that the results obtained could be compared and combined correctly. This paper includes some preliminary data demonstrating that the newly augmented system is capable of providing useful data to the asset owner.