Condition Monitoring of Railway Tracks and Vehicles with Distributed Fiber Optic Sensing

Increased use of mass transportation with high capacity and high headway causes growing burden to the whole railway infrastructure. Therefore, it is essential to offer permanent surveillance of these highly frequented railways. The early detection of possible damages of the railway facilities due to wear and tear, natural causes like mudflow, avalanches, floods, landslides, fallen trees, or due to trespassers or unauthorized persons on the railroad, can prevent secondary damage by fast and correct actions. The Austrian Federal Railways (ÖBB) currently monitor the state of the rail tracks via biannual test drives and by the train drivers during regular operation. In the railway stations and in the industrial premises the site staff pays attention to the condition of the rail facilities. Irregularities are reported to the operational management. Because of the use of local monitoring systems not all railways can be monitored in real time. The aim of this research project is to study the detection properties of complementary sensor technologies, which enable a complete and permanent monitoring over the entire rail track. The sensor technologies used for this purpose are optical fiber systems applied to the rail tracks as well as already existing systems. The fiber optic measurement systems include optical time domain reflectometry (OTDR) and the distributed Brillouin measuring systems. This will enable the detection of local changes, noise events and vibration as well as local temperature changes and strains over a distance of up to 40 km with each single instrument. The optical strain gauges on the rail tracks are used for the detection of tensions on the tracks and serve as isolated references. The already existing systems, such as avalanche and landslide warning systems will be integrated and complemented with additional sensors, if necessary. The automatic interpretation of the large amount of sensor data is carried out by means of suitable algorithms for signal processing and pattern analysis. The research project will investigate and demonstrate the most appropriate sensor technologies and types for detecting maintenance-relevant conditions and changes on an ÖBB rail track.