An Online Classification Method for Fault Diagnosis of Railway Turnouts

Abstract

Railway turnout system is a key infrastructure to railway safety and efficiency. However, it is prone to failure in the field. Therefore, many railway departments have adopted a monitoring system to monitor the operation status of turnouts. With monitoring data collected, many researchers have proposed different fault-diagnosis methods. However, many of the existing methods cannot realize real-time updating or deal with new fault types. This paper—based on imbalanced data—proposes a Bayes-based online turnout fault-diagnosis method, which realizes incremental learning and scalable fault recognition. First, the basic conceptions of the turnout system are introduced. Next, the feature extraction and processing of the imbalanced monitoring data are introduced. Then, an online diagnosis method based on Bayesian incremental learning and scalable fault recognition is proposed, followed by the experiment with filed data from Guangzhou Railway. The results show that the scalable fault-recognition method can reach an accuracy of 99.11%, and the training time of the Bayesian incremental learning model reduces 29.97% without decreasing the accuracy, which demonstrates the high accuracy, adaptability and efficiency of the proposed model, of great significance for labor-saving, timely maintenance and further, safety and efficiency of railway transportation.

The Fundamental Approach of the Digital Twin Application in Railway Turnouts with Innovative Monitoring of Weather Conditions

“…Many of the existing fault-diagnosis methods cannot realize real-time updating or deal with new fault types [ 20 ]. Ou, D. et al [ 20 ] note that due to its complicated mechanical and electrical structure, exposure to the outdoor environment and the need to be frequently pulled, the turnout is more prone to failure.…”

“…Many of the existing fault-diagnosis methods cannot realize real-time updating or deal with new fault types [ 20 ]. Ou, D. et al [ 20 ] note that due to its complicated mechanical and electrical structure, exposure to the outdoor environment and the need to be frequently pulled, the turnout is more prone to failure. Conducting research on the automation of turnout condition monitoring methods is very important not only to improve the safety and reliability of railroad traffic but also to reduce maintenance costs [ 21 , 22 ].…”

“…Some researchers have also proposed different incremental methods for classification problems [ 27 ]. However, many of these methods cannot be combined effectively with the field experience of workers (especially for the turnout system), and, also, some of these methods cannot update themselves, which may reduce the efficiency of the model when new samples or fault types appear since the training samples are quite limited from the very beginning [ 20 ]. In the case of applying digital twins in the monitoring of railway turnouts based on taking into account the data coming from the measurement station T szyn WS1 WIFI together with the fundamental data, the created model will be dynamic and will be a reflection of the turnout system and the process existing in the real world.…”

“…The temperature distribution of CWR rails is not uniform, which significantly complicates the precise definition of measurement principles. The selection of a representative measuring point should take into account the average deformation associated with an uneven temperature distribution [ 19 , 20 ]. To perform the tests, an external DS18B20 temperature sensor was placed inside the crown of the rail type S49 (49E1).…”

Railroad Turnout Wear Diagnostics

“…The results of tests of turnout geometry, profile from the first stage are compared with the results of dynamic tests from the second stage of the implemented research programme. The comparative analysis of the test results is aimed at the evaluation of the applied research methods for the early diagnostics of the turnout failures [ 50 , 51 , 52 , 53 , 54 , 55 ].…”