Trackside Optical Fibre Acoustic Sensing (TOFAS)

Distributed fibre optic acoustic and dynamic strain sensing has important applications in the security, energy, environment and transport industries. Examples of such applications include intruder detection, leak detection in oil/gas pipelines and nuclear power reactor systems, monitoring shock waves caused by fracking, and tracking and listening to moving trains.
The key advantage of all distributed fibre optic sensing is that a measurand can be detected at every point along the fibre. In this way a large number of discrete sensors can be simply replaced by a single optical fibre. The basic operation is based on sending pulses of light down the optical sensing fibre and detecting the changes in the backscattered light, caused by the parameters to be measured. The Rayleigh backscattered light is sensitive to the sound pressure induced strain on the fibre. Since each point on the Rayleigh backscattered trace corresponds to one section of the sensing fibre, the acoustic wave field along the sensing fibre can be mapped by launching optical pulses into the sensing fibre at a regular intervals and monitoring the changes in the backscattered traces. With the appropriate optical setup and the digital signal processing that we have developed, the acoustically induced strain in terms of its frequency, phase and amplitude can be spatially resolved along the entire length of the sensing fibre. The repetition rate of the pulses determines the frequency at which the measurement is repeated and hence the detection bandwidth of the acoustic signal. In essence, the single optical fibre can perform the same function as multiple (~10000) microphones but with much reduced cost and complexity of installation.
The proposed research is to develop a distributed fibre optic acoustic and dynamic strain sensor technology with capabilities far in excess of what has currently been achieved in order to improve its applicability to a number of key applications, but in particular to the rail transport industry for monitoring the health of track and trains. The improvement will stem from modification of the optical configuration and introduction of new hardware and software for data handling and processing.
Whilst future predicted growth in rail travel will inevitably require additional growth in rail infrastructure, it is imperative that the industry continues to strive to improve the efficiency of existing train services, whilst maintaining the highest of safety standards. This proposal is concerned with developing the state of the art distributed fibre optic acoustic sensing and with the goal of enabling
i) Accurate determination of the location and speed of trains which will allow train density to be optimised;
ii) Abnormal sounds to be detected, providing early indication of potential problems such as intruders, cable theft, loose and rattling components, etc, facilitating timely maintenance or preventative action to minimize disruptions; and
iii) The condition of track-side machines such as level crossing motors and remote generators to be monitored, ensuring safe and efficient operation.
Achieving these goals will help to provide safe, efficient and reliable rail transport that maximises the capacity of the existing infrastructure.

Economic Impact:
The economic impact of this work falls into two categories: reduction in the railway maintenance cost and cable-theft related expenses. Railway infrastructure maintenance is an expensive and labour intensive procedure which relies on basic visual inspection of the railway line. Despite the recent proposal of introducing video monitoring from a moving train, complete deployment of this technology has yet to be implemented and inspection either on foot or on motorized trolleys is still required. Using a distributed acoustic sensor can significantly reduce costs by avoiding unnecessary visual inspection. In addition, the costs arising from rail-service disruption can be reduced or eliminated by regular remote inspection of the health and condition of the railways.
Distributed acoustic sensor can also be used as an intruder detection sensor. In a report published by the transport committee of Parliament, direct and indirect cost to the Network Rail from cable theft between 2008 to 2010 was estimated to be in excess of £72M. Using the distributed acoustic sensor, any acoustic excitation in the vicinity of the railway track can be quantified and analysed. Following the analysis, any abnormal acoustic excitation can be flagged for further investigation. The advantage of our proposed system over conventional optical fibre intruder detection systems is that our system will be able to fully identify the acoustic characteristics of the intruder and not just its location.

Safety Impact:
In addition to the economic impacts, distributed acoustic sensing has potential to introduce a wide range of safety benefits. Monitoring the location and speed of trains every metre across the rail network using distributed acoustic sensors can provide useful input to significantly improve the current signalling system. At the same time, the same sensing mechanism can also be used to detect and warn of rock falls along the railway track.
As mentioned earlier, the distributed acoustic system can be used as a remote health monitoring system. Therefore, the assessment of the railway track condition can be carried out continuously during the day which will reduce the likelihood of accidents such as the train derailment that occurred at Argyll in 2010.
In addition, this sensing system can also be used to measure a number of essential variables about the train status such as bearing wear. Bearing faults can be detected by acoustic signals specific to the bearings. There are lineside detectors in use for this but a continuous i.e. distributed system would be beneficial. With the need to reduce expenditure of Network Rail on maintenance, the introduction of additional safety features in the form of a distributed acoustic sensor can significantly improve the safety of the rail transport system.
Finally, it should be mentioned that the sensing mechanism relies on optical fibre which is robust, durable, and has little sensitivity to electromagnetic perturbations. Therefore, an optical fibre based sensor can be considered as a vital backup system for signalling and mapping the location and speed of the trains across the railway network in situations such as solar storm and flood, when conventional systems may fail.

Academic Impact:
We will be publishing our results in scientific journals and presenting our work at national and international conferences with the aim of publicising our work to help attract interest from both the academic community and industry, which is anticipated to lead to future collaboration. .