Railway experts at the University of Huddersfield (England) Insititute of Railway Research (IRR) are working with Siemens to develop an inexpensive and easily-fitted sensor that could turn virtually every rail vehicle into a track monitor, detecting and transmitting vital information about the condition of rails and rail bed throughout the network.
According to the experts, the result would not only be improvements in safety and reliability, but also major efficiency gains and cost savings for network operators, as well as improved ride quality for passengers.
Every train in the UK – and many overseas – is fitted with a GSM-R cab radio system, some of which are produced by Siemens. It is now possible to retrofit an inexpensive Tracksure sensor card to the Siemens cab radios and by picking up vibrations they transmit information – received by a control centre – that can detect under-track voids.
These voids are gaps that have developed between sleepers and ballast. In serious cases, they can lead to an increased risk of rail breaks, along with poor vehicle ride performance. The monitoring system would therefore provide an early warning of problems – especially at switches and crossings and at the transition to other high value assets such as bridges.
University of Huddersfield’s IRR developed an all-new algorithm that could be programmed into the sensors, so they detect under-track voids.
“It was very challenging,” said team member Dr Farouk Balouchi.
“Initially we used simulation to identify what type of sensors and what accuracy and sensitivity would be needed for the Tracksure prototype. This led on to us developing a highly efficient algorithm which can process large quantities of acceleration data in a short space of time to detect the location and severity of potential track voids.”
At a conference showcasing the solution, Roy Formston of Siemens UK detailed how techniques such as machine learning, additional functionality could be included to improve the detection accuracy by making use of multiple train runs over the same section of track.
The IRR intends to further collaborate with Siemens to develop a concept that has the potential to provide levels of “big data” that could provide a huge boost to rail safety and cost-efficiency.
“Through this further work we are investigating additional functionalities of the system, whereby we can detect other anomalies from the track and the vehicle,” said Dr Adam Bevan, another IRR team member.
For example, in addition to detecting voids, sensors could give early warning of problems such as corrugation of the track or wheel flats – distortions in wheel shape caused by factors such as the lack of adhesion. Actual vehicle suspension faults could also be picked up by sensors.