Deep Learning Model for Railroad Structural Health Monitoring via Distributed Acoustic Sensing

Abstract

Railway infrastructure plays a vital role in modern transportation systems, facilitating the efficient movement of people and goods. However, the integrity and performance of railroad structures are subject to various external forces and aging processes, which necessitate continuous monitoring to ensure safety and operational efficiency. This research focused on the structural health monitoring of the railroad using Distributed Acoustic Sensing (DAS) data collected from a High Tonnage Loop (HTL). An investigation on applying a deep learning model, long-shot-term memory (LSTM), and gated recurrent Unit(GRU) is presented to identify and classify railroad conditions.