Condition Monitoring, Fault Diagnosis and Predictive Maintenance in Railway Systems

Working Group

Condition Monitoring, Fault Diagnosis and Predictive Maintenance in Railway Systems

Introduction

The modern railway has evolved into a sophisticated and interdisciplinary field in which there is an increasing tendency to automation. Today, railway involves disciplines with them it had virtually no connection in the past, for instance, mechatronics and computer science.

One major point in terms of the safety and availability of railways involves maintenance, which represents up to 50% of the total life cycle cost of a railway project. Currently, maintenance of the railway track and vehicles occurs on a planned schedule basis; the frequency of maintenance actions is given mainly by historical records, statistics of failure time, and the experience of the railway operator.

Today, the objective of all railway companies around the world is to ensure the safety and improve the performance and profitability of railway systems while considering at the same time the life cycle cost with the purpose to reduce it. A window of opportunity to achieve this goal lies in the maintenance phase via condition monitoring.

Condition monitoring is based on the continuous and automatic tracking (via measurements and analysis of physical data) of the state (condition) of a system or process to detect early signs of deterioration and developing faults. Moreover, CM implies only a marginal additional cost with respect to the overall project cost, while it can help to significantly reduce the maintenance costs via condition-based and predictive maintenance policy.

Aware of the modernization tendency of the modern railway, the Institute of Railway and Transportation Engineering established the working group Condition Monitoring, Fault Diagnosis and Predictive Maintenance in Railway Systems for advanced research on railway systems monitoring. The focus of the working group is to apply modern algorithms (including model-based, data-driven and hybrid) to perform data analysis and system modelling for vehicle-based fault detection and isolation of the railway track.

The four stages of an Intelligent Maintenance System: (1) Monitoring, (2) Diagnosis, (3) Prognosis, (4) Assessment and Decision Making.
The four stages of an Intelligent Maintenance System: (1) Monitoring, (2) Diagnosis, (3) Prognosis, (4) Assessment and Decision Making.

Source: Modern Tendencies in Vehicle-Based Condition Monitoring of the Railway Track https://ieeexplore.ieee.org/document/10041173

Thematic Focus

  • Railway instrumentation and vehicle-based infrastructure monitoring
  • Hybrid dynamic modelling of railway vehicles
  • Model-based and Machine Learning algorithms
  • Fault detection and isolation on the railway track
  • Predictive maintenance and optimal maintenance scheduling

Models and Systems

  • Vehicle-track scale model [Insert link]
  • Vehicle-based track quality monitoring system – Reallabor “Tälesbahn” [Insert link]

Related research projects

  • Determination of the characteristics of punctual instabilities considering the occurring cause and load (EPIB 1.1) [Insert link]
    • Funded by the Deutsche Forschungsgemeinschaft (DFG)
    • Project period: 11.2017 – 04.2020
  • Track condition monitoring via inertial sensors on regular in-service vehicles
    • Reallabor short-term recommissioning
    • Project duration: 01.2022 – 07.2022
  • Efficient Sensor-based Condition Monitoring Methodology for the Detection and Localization of Faults on the Railway Track (ConMoRAIL)

Publications

Working Group Members

Dieses Bild zeigt Xiaoyue Chen

Xiaoyue Chen

M.Sc.

Akademischer Mitarbeiterin, Labor

Dieses Bild zeigt Héctor Alberto Fernández Bobadilla

Héctor Alberto Fernández Bobadilla

M. Eng.

Akademischer Mitarbeiter, Doktorand

Dieses Bild zeigt Euiyoul Kim

Euiyoul Kim

M.Sc.

Doktorand

Zum Seitenanfang