Application and Research of Distributed Acoustic Sensing in Vibration Monitoring of Subway Tunnels

With the booming development of the rail transit industry and the increase in the construction and operation of mileage, the health status of infrastructure such as vehicles, tracks, and trains, as well as operational safety, have received widespread attention. Distributed optical fiber sensing technology has the characteristics of long monitoring distance, strong real-time performance, and highly economical, and has broad prospects in rail transit applications. This paper adopts Distributed Acoustic Sensing technology to obtain vibration information along the subway tunnel, analyzes the time-domain signal and frequency-domain energy characteristics of train vibration, as well as the passing time, direction, and location of the train, providing technical support for the safe operation of the subway.


Introduction
Distributed Acoustic Sensing (DAS) technology uses optical fiber as a sensing and transmission medium to monitor signals along the fiber.It offers high sensitivity and does not have blind spots, as every point along the fiber is a sensitive unit.DAS has been widely applied in various fields such as perimeter security [1], pipeline monitoring [2][3], and railway safety monitoring [4][5].Therefore, the application of distributed optical fiber sensing technology in the monitoring of operational safety in subway tunnels has become a new research hotspot.
Different requirements in subway monitoring lead to the application of different distributed optical fiber sensing technologies.For example, dynamic monitoring of train operation (speed and positioning) utilizes DAS based on Phase-Sensitive Optical Time-Domain Reflectometry (φ-OTDR) .Structural health monitoring of track and bridges, on the other hand, utilizes Distributed Strain Sensing (DSS) based on Brillouin Optical Time-Domain Analysis/Reflectometry (BOTDA/R) .
This study focuses on the application of DAS technology to monitor vibration information in subway tunnels, providing a basis for its engineering application during subway operation and maintenance.

Monitoring Principle
DAS employs φ -OTDR technology to detect vibrations by utilizing the sensitivity of Rayleigh scattered light interference wave phase to sound waves and vibration characteristics caused by the transmission of coherent light in optical fibers.
When using OTDR systems for vibration detection, the signal-to-noise ratio can be low, and in some cases, vibration signals may not be detected at all.This is because OTDR systems use broadband lasers, and the interference between different linewidths of light in the backscattered Rayleigh signal can drown out the useful vibration information.Additionally, as the linewidth of the laser increases, the sensitivity of the system decreases [6].
To improve the the response ability of OTDR technology to external vibration interference, φ -OTDR technology was introduced.In contrast to traditional OTDR, φ -OTDR technology uses coherent light pulses as the probing light source.As a result, the output of this sensing technology is the coherent superposition of the backscattered Rayleigh light from different scattering centers within the pulse width range.
Similar to traditional OTDR, φ -OTDR technology measures the time difference between the injected light pulse and the corresponding Rayleigh signal to determine the position of a disturbance point.When external vibrations act on the sensing fiber, the length and refractive index of the fiber at the corresponding position will change, leading to a change in the phase of the backscattered Rayleigh light at the position.Since the output of φ-OTDR technology is the result of Rayleigh scattered light interference in the pulse width range, the final interference result will change and correspond to the location of the intrusion.By analyzing the changes in the Rayleigh waveform, it is possible to detect phase variations in the fiber caused by external vibrations [7].
Using optical heterodyne detection technology, the detection light containing the measured information and the referenced local oscillator light are performed optical mixing on the photodetector under the precondition of interference.Since the frequency response range of current photodetectors is much lower than that of light, the output differential frequency signal contains amplitude, frequency, and phase information of the measured signal [7].
In optical heterodyne detection, by optically mixing the signal bandwidth from optical frequency to the frequency band that the photodetector can respond to, the output current is proportional to the amplitude of the signal light and local oscillator light.With a fixed reference signal, the frequency and phase variation information of the measured signal can be obtained, providing more reference information for signal detection [8]. Figure 1 shows the structure of a Φ-OTDR system based on coherent detection.Unlike systems based on direct detection, the returned Rayleigh scatter light from the sensing fiber interferes with the local oscillator light emitted by a narrow linewidth laser in a 2x2 fiber coupler.The interference signal is then detected by a balanced photodetector.The balanced photodetector consists of two photodiodes with similar performance.By designing the circuitry to subtract the currents converted from the two optical signals, the DC component is removed, improving the system's common-mode rejection ratio and detection sensitivity.The alternating component, which contains the phase and intensity information of the disturbance signal, is preserved [9].
As the optical pulse propagates forward in the sensing fiber, it continuously generates backward Rayleigh scatter signals.Therefore, by measuring the temporal variation of the backward Rayleigh scatter light, the vibration signal at different positions can be obtained.Demodulating the vibration signal yields the vibration information at that specific position.

Field Testing
The author and their team have installed vibration optical cables in a tunnel section of a subway line in Nanjing.DAS is adopted to monitor real-time vibration information along the subway tunnel through fiber optic sensing and signal transmission.Data processing and analysis are conducted to study the characteristics and patterns of vibration signals, exploring the feasibility of using distributed acoustic sensing technology for subway safety monitoring.
The effective monitoring section of the vibration sensing optical cable is from K17+42 to K17+460. Figure 2 shows a schematic diagram of the deployment of the vibration sensing fiber cable in the subway tunnel

Statistical Analysis.
Data from all alarm incidents within a day, from 00:00 to the next day's 00:00, are selected.Figure 3 shows the analysis results of several hours.The horizontal axis represents the time distribution, and the vertical axis represents the position points of the fiber cables (vertical coordinate multiplied by 10 represents the length position points of the cables).
In Figure 3, each blue circle represents an alarm incident at a specific position point.A vertical column of blue circles represents the alarm of the whole monitoring area, and the alarm interval is about 6 minutes, with an alarm time interval of approximately 6 minutes.As shown in the red box in Figure 3, two different alarm times for the same position point are observed: 2022-12-30 06:02:18 and 2022-12-30 06:08:39, with a time gap of 6 minutes and 21 seconds, matching the actual train schedule.Therefore, it can be concluded that the vertical column of alarm signals represents the vibration signals generated by the passing trains in the subway.From Figure 4-17, the FBE results show the first two frequency bands (0-250Hz) account for the majority of energy, some of which are the first frequency band (0-125Hz) and some of which are the second frequency band (126-250Hz).The proportions of the remaining frequency bands are mostly below 0.3, with the maximum not exceeding 0.5.

Conclusion
1.The use of DAS can capture the vibration information of subway trains passing by, and analysis can reveal the signal characteristic.
2. Through data analysis, the time-frequency domain information and energy distribution characteristics of train signals have been obtained, and the results show that the energy is mainly concentrated in the 0-250Hz frequency band.
3. In addition to the train signal, other vibration signals in the subway tunnel need to be further analyzed and studied.

Figure 1 .
Figure 1.The schematic diagram of heterodyne coherent detection optical path structure.

Figure 2 .
Figure 2. The schematic diagram of vibration sensing fiber cable.
Figure 4~17 shows the time-domain waveform and frequency band energy (FBE) calculation results of train vibration signals at different times and dates, which is used to analyze the characteristics of the train signals.Due to the large amount of data, this article only lists seven sets of data.For the sake of unified comparison, the FBE has been normalized here.