Fleet of sensors for autonomous railway condition monitoring
Working towards an autonomous maintenance strategy for railway fasteners.
Facts
Researchers: Matti Rantatalo (PL), Praneeth Chandran, Florian Thiery, Johan Odelius
Project Sponsor: Vinnova
Goal
The purpose of the project was to work towards an autonomous maintenance strategy for railway fasteners using condition monitoring via trains in operation. By using passenger and freight trains as carriers of robust and autonomous sensors, a greater amount of information about the status of the facility will be created. The goal of the project was to test and develop a cloud-based solution and business model for a scalable implementation of the magnetic field-based sensor for the inspection of railway fasteners.
Project status and results
The project developed a method for analysing and managing information from railway fastener inspection sensors in a context where a fleet of sensors is installed across various trains and operators. This approach enables the synchronization of different measurements, allowing for effective monitoring of changes in the condition of fasteners. Additionally, the project provided a framework for implementing such a sensor system in Sweden or internationally, considering diverse business interests and incentives.
The project was divided into two main parts. The first part focused on developing a cloud-based algorithm to synchronize measurement data from multiple sensors installed on operational trains. This work relied on real-time measurements from trains in service and involved addressing several technical and practical challenges, including adapting the measurements to locomotive schedules. This required not only software development but also modifications to hardware solutions. The second part, which addressed the business landscape, was more theoretical and involved fewer practical challenges.
Figure1. Project overview
Figure1. Project overview
Figure 2. Aligned track signatures of multiple train measurements (represented as Peak-to-Peak feature) with respect to sleeper number
Contact
Matti Rantatalo
- Professor
- 0920-492124
- matti.rantatalo@ltu.se
- Matti Rantatalo
Updated: