Improvising Communication System by High Density Fibre Optic Cable using IoT Notifications

Abstract

We present a nondedicated bridge health monitoring (BHM) system that uses distributed
acoustic sensors embedded in existing telecommunication fiber-optic cables to gather bridge
dynamic strain responses. Our telecommunication cable-based system enables effective and
affordable BHM without the need for on-site sensor installation and maintenance. However, due
to the high measurement noise and error propagation in this nondedicated strain data, it is
difficult to extract bridge damage-sensitive information (such as natural frequencies and mode
shapes). We create a physics-guided system identification technique that simulates strain mode
to get around the problem. Forms based on bridge dynamics-derived parametric mode shape
functions that are governed by physics. We next analytically double-integrate the modelled
strain mode shape to determine the displacement mode shape function. By limiting strain and
displacement mode forms with respect to bridge dynamics, our technique increases the
precision of calculating bridge damage-sensitive features and decreases error propagation. On a
continuous three-span concrete bridge in San Jose, California, we tested our technique. With a
precision of one metre, our algorithm was able to rebuild the strain and displacement mode
forms and effectively identify the first three natural frequencies.