.png){kind=icon}
.png){kind=icon}
Return
Train Identification Using Distributed Acoustic Sensing
.png){kind=icon}
OA
AI
DOI：10.1007/978-3-031-88974-5_64.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Over the past three decades, Distributed Acoustic Sensing (DAS) has garnered considerable attention, primarily due to its potential across a wide range of applications, especially in the railway industry. In particular, DAS has proven to be an effective tool for train tracking and the detection of anomalies in railway platforms and track structures. In the present work, we explain how to use the DAS system in order to perform train classification. Our approach is based on the estimation of the bogie displacement frequency of the operating train which allows to estimate its car count which is an efficient parameter to identify the train's class. To obtain such estimation, we employ a frequency analysis of the temporal signals collected by the DAS interrogator at various positions during a train's passage using Continuous Wavelet Transform (CWT). The performance of the proposed algorithm was tested using real data from measurements taken on the french rail network. These tests resulted in a great classification accuracy of around 96%, demonstrating the effectiveness of our approach.
Keywords:
Distributed Acoustic Sensing (DAS)
Train classification
Optical fiber sensors
Track monitoring
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
T
IF:
0
Papers:
121
Citations:
0
Organization
S