Abstract
Efficacy in railway track maintenance is not only critical for safe operation, but by reducing delays and increasing capacity, can result in economic benefits for rail operators.
Currently, rail infrastructure managers have dedicated vehicles to carry out track measurement. These vehicles are, however, expensive to procure and run as they need specially trained crews to operate them. Furthermore, they require uninterrupted access to the track, and so are difficult to schedule around other services. Consequently, the UK only has 6 such measurement vehicles and the mainline network is only inspected on a 4-6 weekly cycle (less regularly for other services).
This work demonstrates the application of an inexpensive inertial measurement unit (IMU), installed on an in-service passenger train, and using novel algorithms, to provide continuous monitoring of the track and its degradation. Two IMUs were fitted, one on the bogie and one in the cab.
Inertial data from the bogie mounted IMU was collected and processed to provide vertical and lateral displacement profiles of sections of track from multiple passes of a rail vehicle. A modified Kalman filter was used to produce the associated dataset profiles to within 0.2 mm location accuracy. Profiles calculated from multiple passes of the same area of track were compared to monitor track degradation.
Inertial data from the IMU mounted in a driver’s cab was also collected and used to calculate the ride comfort using BS EN 12299. The ride comfort indices were compared to recommended limits on acceleration and change of acceleration (jerk).
Analysis of the results were used to determine whether trains on the measured routes meet current standards for ride comfort.
Reported issues on the measured routes from the Network Rail Fault Management System (FMS) and data from New Measurement Train (NMT) were used to evaluate the performance of the IMU based system against current UK industry track monitoring practices and results were comparable. The findings from this project show that an IMU mounted on the bogie of an in-service railway vehicle allows track geometry data to be measured at more frequent intervals than with a dedicated track measurement vehicle and at a fraction of the cost.
Currently, rail infrastructure managers have dedicated vehicles to carry out track measurement. These vehicles are, however, expensive to procure and run as they need specially trained crews to operate them. Furthermore, they require uninterrupted access to the track, and so are difficult to schedule around other services. Consequently, the UK only has 6 such measurement vehicles and the mainline network is only inspected on a 4-6 weekly cycle (less regularly for other services).
This work demonstrates the application of an inexpensive inertial measurement unit (IMU), installed on an in-service passenger train, and using novel algorithms, to provide continuous monitoring of the track and its degradation. Two IMUs were fitted, one on the bogie and one in the cab.
Inertial data from the bogie mounted IMU was collected and processed to provide vertical and lateral displacement profiles of sections of track from multiple passes of a rail vehicle. A modified Kalman filter was used to produce the associated dataset profiles to within 0.2 mm location accuracy. Profiles calculated from multiple passes of the same area of track were compared to monitor track degradation.
Inertial data from the IMU mounted in a driver’s cab was also collected and used to calculate the ride comfort using BS EN 12299. The ride comfort indices were compared to recommended limits on acceleration and change of acceleration (jerk).
Analysis of the results were used to determine whether trains on the measured routes meet current standards for ride comfort.
Reported issues on the measured routes from the Network Rail Fault Management System (FMS) and data from New Measurement Train (NMT) were used to evaluate the performance of the IMU based system against current UK industry track monitoring practices and results were comparable. The findings from this project show that an IMU mounted on the bogie of an in-service railway vehicle allows track geometry data to be measured at more frequent intervals than with a dedicated track measurement vehicle and at a fraction of the cost.
| Original language | English |
|---|---|
| Title of host publication | World Congress on Railway Research |
| Publication status | Published - 29 Oct 2019 |
| Event | World Congress on Railway Research: Railway Research to Enhance the Customer Experience - Tokyo International Forum, Tokyo, Japan Duration: 28 Oct 2019 → 1 Nov 2019 Conference number: 12 https://wcrr2019.org |
Conference
| Conference | World Congress on Railway Research |
|---|---|
| Abbreviated title | WCRR2019 |
| Country/Territory | Japan |
| City | Tokyo |
| Period | 28/10/19 → 1/11/19 |
| Internet address |