Sensor Technology for Damage Monitoring of Wheelset Axles in Rolling Stock (SAFE On-Line)
The project “SAFE On-Line” aims to develop a hardware and software system for real-time monitoring of wheelset axle damages in rolling stock.
Using Ultrasonic Guided Waves combined with Principal Component Analysis, the system enables continuous condition-based monitoring, replacing conventional stationary ultrasonic inspections.
This innovation increases operational safety, reduces maintenance costs, and improves train availability.
The project addresses experimental validation of sensors under harsh operating conditions, energy-efficient signal generation, data acquisition, transmission, and evaluation.
Research and development are carried out in NRW, strengthening regional know-how and creating jobs.
Project description
Wheelset axle damage in rolling stock is unpredictable and can lead to critical safety issues.
Current maintenance relies on periodic ultrasonic inspections in workshops, which are time-consuming and expensive.
The “SAFE On-Line” project seeks to develop an innovative system that continuously monitors axle conditions using Ultrasonic Guided Waves and Principal Component Analysis.
The project focuses on designing a robust sensor and measurement system capable of operating under harsh conditions (shocks, moisture, dirt, varying temperatures).
The system will enable real-time damage detection, replacing conventional inspections, and allowing for condition-based maintenance (CBM).
For railway operators, this technology promises significant safety improvements, cost reduction, and higher train availability.
The project also emphasizes NRW regional development by conducting all research, development, manufacturing, and application processes locally, contributing to job creation and SME know-how growth.
The University of Siegen’s Embedded Systems department contributes its expertise in FPGA and embedded controllers, focusing on software and hardware integration, functional safety, and validation of sensor systems in operational environments.
Bullet points for focus points of the project
- Continuous real-time monitoring of wheelset axles using ultrasonic sensors
- Application of Principal Component Analysis for damage detection on Embedded Plattforms
- Replacement of stationary, workshop-based inspections with automated condition-based maintenance
- Validation of sensor functionality under harsh operational environments
- Signal generation and sensor management on FPGA
- Data acquisition, transmission, and real-time analysis
- Regional technology development and job creation in NRW
- Ensuring functional safety in embedded hardware and software systems
All-in-one overview
-
Project duration
01/05/2017 to 30/04/2020 Duration Time: 36 Month -
Keyword for areas Wheelset Axle Monitoring, Ultrasonic Guided Waves, Condition-Based Maintenance (CBM), Functional Safety, Real-Time Damage Detection
-
-
Financing
European Fund and the State of NRW for Regional Development (EFRE) Total Funding: 2.103.476,63 € Funding USI: 795.106,13 €
Methodology
Method 1:
Sensor and Hardware Development Design and prototype robust ultrasonic sensors suitable for harsh railway environments.
Ensure reliable operation under shocks, vibrations, dirt, moisture, and temperature fluctuations.
Develop energy-efficient signal generation and sensor management strategies.
Method 2:
Data Acquisition and Analysis
Implement a measurement and analysis system using Ultrasonic Guided Waves combined with Principal Component Analysis.
Develop software for data acquisition, transmission, and evaluation.
Validate system functionality and accuracy under real-world operating conditions.
Method 3:
Functional Safety Analysis and System Integration Integrate sensors, FPGA/embedded controllers, and software into a complete monitoring system.
Conduct tests and validation to ensure compliance with functional safety requirements.
Provide evidence for reliable condition-based maintenance applications.
Method 4:
Field Testing and Validation
Deploy the system in operational railway vehicles to evaluate performance, robustness, and usability.
Adjust hardware and software based on real-world data to achieve reliable damage detection and operational efficiency.
