Subway track multi-parameter traffic safety monitoring system based on integration of optical fiber sensing and communication

Abstract

This invention discloses a multi-parameter traffic safety monitoring system for subway tracks based on integrated fiber optic sensing and communication. The system comprises, in sequence, a sensing and communication module, a demodulation module, a signal processing and decision-making module, a control module, and an application module. The sensing and communication module collects temperature, vibration, strain, and humidity information in real time through fiber optic sensor arrays within the track and tunnel, and multiplexes the fiber optic cables to achieve co-transmission of sensor data and communication services. The demodulation module converts optical signals into electrical signals and uses time-division multiplexing technology to demodulate the physical parameters. The signal processing and decision-making module performs noise reduction and feature extraction on the signals and uses a pre-set machine learning model to identify abnormal events. Based on the decision results, the control and application modules perform status monitoring, early warning issuance, emergency control, and operation and maintenance functions. This invention provides a highly reliable multi-parameter traffic safety monitoring system for subway tracks based on integrated fiber optic sensing and communication, characterized by comprehensive perception, multi-parameter fusion, and high reliability.

Description

Technical Field

[0001] This invention belongs to the field of rail transit monitoring technology, specifically relating to a multi-parameter traffic safety monitoring system for subway tracks based on integrated fiber optic sensing. Background Technology

[0002] The subway track safety monitoring system is an intelligent system that comprehensively utilizes sensing technology, communication technology, and data analysis technology to monitor and evaluate the status of subway track and tunnel structures in real time, automatically, and continuously. It is a core support for ensuring the efficient operation of urban rail transit networks. By acquiring multi-dimensional data such as deformation, vibration, and temperature of tunnels and tracks in real time, it provides key information for preventing structural damage and optimizing operation and maintenance decisions.

[0003] However, existing monitoring technologies suffer from significant shortcomings in reliability, coordination, and diagnostic capabilities. First, traditional electrical sensors have inherent physical limitations. Electrical devices, such as strain gauges and accelerometers, are susceptible to interference from the strong electromagnetic environment of subways, leading to data drift and even failure. Furthermore, they require dense cabling, increasing maintenance and repair difficulties in high-humidity, high-vibration tunnel environments. Second, insufficient integration of heterogeneous systems creates coordination bottlenecks. Existing solutions often employ an architecture of independently deployed sensing units and communication fibers, resulting in hardware redundancy of 30%-40%. Furthermore, protocol differences between different subsystems create data silos, hindering the spatiotemporal alignment and fusion analysis of multi-source data such as vibration, temperature, and displacement. A deeper contradiction lies in the disconnect between single-parameter sensing and complex operating conditions. Track safety is essentially the result of the coupling of multiple physical fields, including train dynamic loads, track geometric deformation, and tunnel structural stress. Current fiber optic monitoring systems are often limited to single parameters, such as vibration, strain, or temperature, lacking synchronous analysis of the vehicle-track-tunnel coupling mechanism. This leads to incomplete diagnostic conclusions and an inability to provide accurate early warnings.

[0004] The existing monitoring technology system suffers from three major bottlenecks: First, in terms of reliability, it relies on traditional electrical sensors that are susceptible to strong electromagnetic interference and have short lifespans in humid and vibrating environments, resulting in inaccurate data and high maintenance costs. Second, in terms of systemicity, the independent deployment of heterogeneous subsystems such as sensing and communication leads to hardware redundancy and data silos, making it impossible to achieve collaborative analysis and precise linkage of multi-source information. Third, in terms of the most critical diagnostic capability, existing solutions are mostly "blind men and the elephant" type single-parameter sensing, such as measuring only vibration or temperature, which is completely unable to simultaneously capture the complex correlation of multiple physical fields under the coupling effect of "vehicle-track-tunnel", resulting in a serious lack of early warning capability for complex faults.

[0005] Therefore, how to address the shortcomings of existing monitoring technologies in terms of reliability, coordination, and diagnostic dimensions, and to provide a multi-parameter traffic safety monitoring system for subway tracks based on fiber optic sensing that integrates highly reliable perception, multi-system coordination, and multi-parameter intelligent diagnosis, is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to address the problems in the prior art by providing a multi-parameter traffic safety monitoring system for subway tracks based on integrated fiber optic sensing.

[0007] Therefore, the above-mentioned objectives of the present invention are achieved through the following technical solutions:

[0008] A multi-parameter traffic safety monitoring system for subway tracks based on integrated fiber optic sensing is characterized by comprising, in sequence, a sensing and communication module, a demodulation module, a signal processing and decision-making module, a control module, and an application module, wherein...

[0009] The sensing and communication module utilizes an optical fiber sensor array deployed in the subway track and tunnel structure. Based on distributed optical fiber sensing technology, it collects multi-parameter physical information such as temperature, vibration, strain, and humidity in real time. Based on integrated optical fiber sensing technology, it reuses the optical fiber sensor array as a communication transmission medium to achieve the co-transmission of multi-parameter sensing data and subway communication service data in the same optical fiber.

[0010] The demodulation module converts the received optical signal into an electrical signal and uses time-division multiplexing technology to separate signals of different wavelengths in order to demodulate the corresponding physical parameter values.

[0011] The signal processing and decision-making module performs noise reduction and feature extraction on the demodulated signal, and uses a preset machine learning model to identify abnormal events in the track or tunnel structure.

[0012] The control module and the application module are used to receive the decision results of the signal processing and decision-making module and execute at least one of the system's centralized control, status monitoring, early warning release, emergency control and operation and maintenance functions.

[0013] While adopting the above technical solutions, the present invention may also adopt or combine the following technical solutions:

[0014] As a preferred technical solution of the present invention: In the sensing and communication module, an integrated fiber optic grating sensing array cable with DTS, BOTDA, and uwFBG arrays is laid along the subway track and tunnel structure. Among them, DTS based on Raman scattering performs distributed temperature monitoring across the entire domain, BOTDA based on Brillouin scattering performs strain and temperature monitoring, and uwFBG array based on ultra-weak vibration and humidity monitoring of key parts is performed. Multi-parameter sensing includes synchronous acquisition and spatiotemporal correlation analysis of track vibration, track structure strain, and tunnel structure temperature and humidity fields caused by train dynamic loads, so as to achieve a comprehensive assessment of track safety status under vehicle-track-tunnel coupling. The time-division multiplexed fiber optic grating sensing array cable is used to allocate independent wavelengths for the multi-parameter physical information and communication services of the sensing, and finally realizes the synchronous transmission of multi-parameter sensing and communication data of temperature, vibration, strain, and humidity.

[0015] As a preferred technical solution of the present invention: the fiber optic grating sensing array optical cable is prepared by drawing tower, and an ultra-weak fiber optic grating array is engraved inside the optical cable to form a sensing network with a spatial resolution of less than 1 meter and a monitoring distance of not less than 10 kilometers.

[0016] The fiber optic grating sensing array is laid in the following manner: at least three optical cables are laid longitudinally on the center line of the tunnel arch and on both side walls, with the optical cables on both side walls laid at a height of 1.5 to 2 meters above the track surface, so as to realize multi-parameter monitoring of tunnel structure strain, temperature, humidity and track structure vibration and temperature.

[0017] As a preferred technical solution of the present invention: each individual optical fiber adopts time-division multiplexing technology to realize the integration of multi-parameter sensing and communication, and the same optical fiber is divided into independent wavelength channels including a sensing signal processing unit and a data communication unit.

[0018] The sensing signal processing unit includes:

[0019] A distributed temperature sensing module, operating in the 1310 nm band, is used to acquire and process the temperature distribution signal along the single-mode optical fiber.

[0020] A humidity sensing module is optically connected to an ultra-weak fiber grating array inscribed in the single-mode optical fiber. The center wavelength of the ultra-weak fiber grating array is located in the 1500-nanometer band. The module is used to obtain humidity signals by detecting the wavelength drift of the ultra-weak fiber grating.

[0021] A distributed strain sensing module, operating in the 1550 nm band, is used to acquire and process strain distribution signals along the single-mode optical fiber.

[0022] The data communication unit operates in the C-band and transmits communication data through the single-mode optical fiber. It is used to transmit video surveillance, equipment status information and maintenance and testing instructions along the subway line.

[0023] The optical signals in the 1310 nm, 1500 nm, 1550 nm and C-band bands are transmitted in parallel in the same single-mode optical fiber using time-division multiplexing technology, without interfering with each other.

[0024] As a preferred technical solution of the present invention: the demodulation module is connected to the sensing / communication module and is used to receive composite optical signals from the fiber Bragg grating sensing array optical cable and convert the optical signals into electrical signals;

[0025] The demodulation module integrates a time-division multiplexing device to separate the communication wavelength channel in the composite signal from the sensing wavelength channel with different physical parameters;

[0026] The demodulation module is configured to demodulate the separated multi-parameter physical information wavelength channel signal to obtain parameters from vibration frequency, temperature change, strain value, and humidity value.

[0027] As a preferred embodiment of the present invention: the signal processing and decision-making module, connected to the demodulation module, is used to process and make decisions on the demodulated key parameter signals, and includes:

[0028] The signal preprocessing unit performs noise reduction processing on the signal;

[0029] The multi-source data fusion unit is used to perform spatiotemporal alignment and correlation analysis on multi-parameter signals such as temperature, vibration, strain, and humidity to eliminate data silos and achieve collaborative diagnosis of multi-physics fields.

[0030] The feature extraction unit extracts feature information from the denoised signal using at least one of wavelet transform and Fourier analysis methods.

[0031] The intelligent decision-making unit is used to perform pattern recognition and classification on the extracted features by a pre-trained machine learning model, thereby realizing the abnormal diagnosis of track cracks and equipment overheating, and realizing the pattern classification of normal train vibration and track structure abnormalities.

[0032] As a preferred technical solution of the present invention: the pre-trained machine learning model includes a convolutional neural network and a long short-term memory network to implement pattern classification.

[0033] Compared with existing technologies, the multi-parameter traffic safety monitoring system for subway tracks based on integrated fiber optic sensing of the present invention has the following beneficial effects: The multi-parameter traffic safety monitoring system for subway tracks based on integrated fiber optic sensing of the present invention utilizes an integrated fiber optic grating sensor array cable, a wavelength division multiplexing mechanism, and a multi-parameter fusion intelligent machine learning model to systematically solve the technical problems of reliability, coordination, and diagnostic accuracy in existing subway track monitoring technologies. It achieves the technical effects of improving system reliability, enhancing data coordination efficiency, and realizing accurate and intelligent diagnosis, and has great application prospects in the field of subway track and tunnel structure safety monitoring technology. Specifically:

[0034] First, this invention utilizes an integrated architecture of all-fiber sensing and communication to solve the reliability problems of traditional electrical sensors being susceptible to electromagnetic interference and having high system hardware redundancy, achieving the effect of long-term stable system operation and a significant reduction in the total life cycle cost.

[0035] Second, this invention utilizes a time-division multiplexing-based collaborative transmission mechanism to solve the signal conflict and resource competition problems of sensing and communication services when transmitted on the same fiber, achieving the effect of parallel transmission of multi-parameter communication service data and endogenous data collaboration.

[0036] Third, by utilizing multi-mechanism fusion sensing (DTS / BOTDA / uwFBG) and feature-oriented machine learning model mapping strategies, the problem of superficial diagnosis of the complex coupling state of "vehicle-track-tunnel" that cannot be analyzed by single-parameter sensing has been solved, achieving early, accurate, and low-false-alarm identification and early warning of various faults such as track cracks, structural overheating, and track bed settlement. Attached Figure Description

[0037] Figure 1 Schematic diagram of a subway rail transit safety monitoring system based on fiber optic sensing integration;

[0038] Figure 2 Fiber optic cable layout diagram. Detailed Implementation

[0039] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0040] This invention provides a multi-parameter traffic safety monitoring system for subway tracks based on integrated fiber optic sensing, aiming to overcome the problems of poor reliability and difficulty in large-scale reuse of traditional electrical sensors in harsh environments such as strong electromagnetic interference, humidity, and high temperature, and to realize integrated transmission of multi-parameter full-domain perception and communication for subway tracks and tunnel structures.

[0041] like Figure 1As shown, the system includes a sensing and communication module 1, a demodulation module 2, a signal processing and decision-making module 3, a control module 4, and an application module 5 connected in sequence. By using wavelength division multiplexing (WDM) technology, multi-parameter sensing and signal communication are simultaneously achieved in a single optical cable, forming a comprehensive network system with both high-capacity transmission and multi-dimensional sensing capabilities. This significantly improves the utilization rate of optical fiber resources and reduces deployment and maintenance costs.

[0042] The sensing and communication module 1 utilizes an optical fiber sensor array deployed in the subway track and tunnel structure. Based on distributed optical fiber sensing technology, it collects multi-parameter physical information such as temperature, vibration, strain, and humidity in real time. At the same time, based on optical fiber sensing integration technology, the optical fiber sensor array is reused as a communication transmission medium to achieve the co-transmission of multi-parameter sensing data and subway communication service data in the same optical fiber.

[0043] The demodulation module 2 is connected to the sensing and communication module 1 and is used to convert the received optical signal into an electrical signal and use wavelength division multiplexing technology to separate signals of different wavelengths in order to demodulate the corresponding physical parameter values, including vibration frequency, temperature change, strain value and humidity value.

[0044] The signal processing and decision module 3 is connected to the demodulation module 2 and is used to reduce noise and extract features from the demodulated signal. It also uses a preset machine learning model to identify abnormal events in the track or tunnel structure, thereby achieving a progressive evaluation from single-parameter anomaly detection to multi-parameter coupled fault diagnosis.

[0045] The control module 4 and application module 5 are connected to the signal processing and decision-making module 3 and are used to receive decision results and execute at least one of the system's centralized control, status monitoring, early warning release, emergency control and operation and maintenance functions.

[0046] This invention relates to a multi-parameter traffic safety monitoring system for subway tracks based on integrated fiber optic sensing. Fiber optic sensing, leveraging the characteristics of optical signal transmission, possesses resistance to electromagnetic interference and environmental adaptability, enabling stable operation under harsh conditions such as humidity and high temperatures. Through distributed monitoring technologies (such as Brillouin scattering, Raman scattering, and fiber grating arrays), it achieves continuous, blind-spot-free multi-parameter monitoring across the entire line, while reusing existing communication fiber optic resources, significantly reducing deployment costs. Combined with intelligent analysis algorithms (such as machine learning), it can provide real-time early warnings of risks such as track deformation, cable overheating, and foreign object intrusion, and supports fire prevention (such as distributed fiber optic linear temperature detectors) and structural health assessment, providing all-weather, multi-dimensional protection for subway safety operation. The use of fiber optic sensing technology in subway rail transit monitoring has significant advantages.

[0047] The fiber optic sensing-integrated multi-parameter traffic safety monitoring system for subway tracks of the present invention, with its distinctive features of full-domain perception, multi-parameter fusion, and high reliability, effectively makes up for the shortcomings of traditional monitoring systems. It provides a solid data foundation for realizing condition-based repair, preventive maintenance, and intelligent operation of subway infrastructure, and is a key enabling technology to ensure the safe, efficient, and green operation of modern subway traffic.

[0048] Example 1

[0049] like Figures 1-2 As shown, the multi-parameter traffic safety monitoring system for subway tracks based on integrated fiber optic sensing of the present invention is a subway track traffic safety monitoring system based on integrated fiber optic sensing, and its implementation is as follows: Figure 1 As shown, the system includes a sensing / communication module 1, a demodulation module 2, a signal processing / decision module 3, a control module 4, and an application module 5. In the sensing / communication module 1, distributed fiber optic sensing technology (such as Rayleigh scattering, Brillouin scattering, and fiber optic sensor arrays) is used to collect physical parameters such as vibration, temperature, and strain along the subway track and tunnel in real time. Based on integrated fiber optic sensing technology, fiber optic resources are reused, and the collected sensing data and subway communication signals (such as video surveillance and dispatch instructions) are transmitted through the same fiber optic cable, achieving co-transmission of communication and sensing signals. In the demodulation module 2, the optical signals from the sensing module are converted into electrical signals, and key parameters such as vibration frequency, temperature change, strain value, and humidity value are demodulated. Time-division multiplexing (WDM) technology is used to separate the communication and sensing wavelengths to avoid signal crosstalk. In the signal processing / decision module 3, the demodulated signal is denoised and feature extracted (such as wavelet transform and Fourier analysis) to identify abnormal events (such as track cracks and equipment overheating). Pattern classification can be achieved through machine learning algorithms (such as convolutional neural networks and long short-term memory networks), for example, distinguishing between normal train vibrations and track structure anomalies.

[0050] The aforementioned sensing / communication module 1 is constructed using fiber optic grating sensing array optical cables, and is used to simultaneously realize multi-parameter sensing and communication of temperature, vibration, strain, and humidity.

[0051] The fiber optic grating sensing array cable 11 is mainly used for multi-parameter monitoring of tunnel structures (strain, temperature, humidity) and health monitoring of track structures (vibration, temperature).

[0052] The fiber optic grating array sensing cables 11 are all manufactured online using fiber drawing equipment. During the fiber drawing process, ultra-weak fiber drawing gratings are automatically inscribed, and each grating is a sensor. The grating pitch can be adjusted according to actual conditions, the spatial resolution can be less than 1m, and a single fiber can achieve monitoring over ultra-long distances (10km).

[0053] The aforementioned sensing optical cables for multi-parameter monitoring of tunnel structures require at least three cables, with one cable laid longitudinally along the centerline of the arch and on each of the two side walls (1.5-2m above the rail surface) to cover the area under the greatest stress.

[0054] The aforementioned tunnel structure monitoring optical cable is fixed to the concrete surface with epoxy resin adhesive, and U-shaped armored protective sleeves are added to key nodes (such as joints) to prevent mechanical damage.

[0055] The tunnel temperature monitoring uses distributed fiber optic thermometry (DTS) technology based on Raman scattering. By using the intensity ratio of Stokes and anti-Stokes light to demodulate the temperature, temperature monitoring can be achieved in the range of -40℃ to 120℃.

[0056] The tunnel strain monitoring adopts Brillouin optical time-domain analysis (BOTDA) technology, which captures the strain distribution along the optical fiber by Brillouin frequency shift changes, and combines it with a three-dimensional displacement inversion algorithm to realize tunnel deformation monitoring.

[0057] The aforementioned tunnel humidity monitoring is achieved by monitoring the center wavelength shift of the ultra-weak fiber Bragg grating (uwFBG) in the optical cable. The surface of the optical fiber is coated with a moisture-sensitive material coating (polyimide) for humidity response. The moisture-sensitive material will absorb water and expand or lose water and shrink, causing axial strain in the grating, which ultimately affects the change of the center wavelength of the grating.

[0058] The BOTDA technology described above is subject to temperature crosstalk while measuring the strain of optical cables. It compensates for this crosstalk by measuring the temperature of DTS in the same optical cable, thereby achieving high-precision strain measurement.

[0059] The ultra-weak grating is also affected by temperature crosstalk while monitoring tunnel humidity. It is compensated by the temperature measured by DTS in the same optical cable, so as to achieve high-precision humidity measurement.

[0060] The single optical fiber employs time-division multiplexing (WDM) technology to integrate multi-parameter sensing and communication, dividing the same fiber into independent wavelength channels. The 1310nm band transmits DTS temperature signals, a uwFBG array with a center wavelength of 1500nm is used to transmit humidity signals, and the 1550nm band transmits BOTDA strain signals. Light near 1565nm in the C-band (1530-1565nm) transmits communication data (such as Ethernet and 5G signals), used for transmitting video surveillance, equipment status information, and maintenance and testing instructions along the subway line. Since the FBG grating only reflects light at its center wavelength and does not interfere with other wavelength signals, a single optical fiber can achieve integrated multi-parameter sensing and transmission in the tunnel.

[0061] The time-division multiplexing technology uses a four-port WDM coupler to separate signals of different wavelengths, thereby separating sensor data and communication data.

[0062] The aforementioned track health monitoring requires the installation of an optical cable on each of the two tracks. The optical cables are laid bidirectionally along the tracks to form a closed loop. In the event of a single fiber breakage, data integrity can be ensured through the reverse link, thereby improving the resistance to fiber breakage.

[0063] The uwFBG array has a grating spacing of 1-2 meters, which can capture the vibration characteristics of common faults such as loose fasteners and cracked rails. The gratings are densely arranged at turnout switch points, frog points, and rail welded joints, with the spacing reduced to 0.5 meters.

[0064] The optical fiber cable is bonded with epoxy resin and fixed with micro clamps, and integrated with the rail surface using laser welding technology, exhibiting excellent vibration resistance.

[0065] The track temperature monitoring uses distributed fiber optic temperature measurement (DTS) technology based on Raman scattering. By monitoring the track temperature in real time, it can prevent structural damage caused by thermal expansion and contraction of the track and ensure the safety of train operation.

[0066] The uwFBG array senses strain or temperature changes by reflecting wavelength offset. Dynamic strain caused by vibration leads to wavelength offset, and the vibration amplitude and frequency can be calculated by demodulating the offset.

[0067] The uwFBG array is subject to temperature crosstalk while measuring vibration. It is compensated for by measuring the temperature of DTS in the same optical cable, thereby achieving high-precision vibration measurement.

[0068] The single optical fiber employs time-division multiplexing (WDM) technology to integrate multi-parameter sensing and communication, dividing the same fiber into independent wavelength channels. Communication data is transmitted in the 1565nm band, and the Bragg wavelength of the uwFBG is designed to avoid the main detection window of the DTS (DTS uses 1310nm pump light, while uwFBG selects the 1550nm band) to prevent spectral overlap.

[0069] Example 2

[0070] This embodiment describes in detail the specific implementation of the sensing and communication module 1.

[0071] The sensing and communication module 1 is constructed using a fiber optic grating sensing array cable. This cable is fabricated online using fiber drawing equipment, and an ultra-weak fiber optic grating array is automatically inscribed during the fiber drawing process. Each grating serves as a sensor. The grating spacing can be adjusted according to actual monitoring needs, achieving a spatial resolution of less than 1 meter, and a single fiber can monitor distances exceeding 10 kilometers.

[0072] For multi-parameter monitoring of the tunnel structure, at least three fiber optic grating sensor array cables 11 need to be laid, longitudinally along the centerline of the tunnel arch and on both side walls. The cables on the side walls are laid at a height of 1.5 to 2 meters above the rail surface to cover the maximum stress area of ​​the tunnel cross-section. The cables are fixed to the concrete surface with epoxy resin adhesive, and U-shaped armored protective sleeves are added at key nodes such as joints to prevent mechanical damage.

[0073] For health monitoring of the track structure, a fiber optic grating sensor array cable is laid on each of the two tracks, forming a closed loop through bidirectional laying. In the event of a single fiber break, data integrity can be ensured through the reverse link, improving the system's resistance to fiber breakage. The optical cables are bonded with epoxy resin and fixed with miniature clamps, and integrated with the rail surface using laser welding, exhibiting excellent vibration resistance.

[0074] Example 3

[0075] This embodiment describes in detail the specific implementation method of multi-parameter monitoring of tunnel structures.

[0076] (a) Tunnel temperature monitoring

[0077] Tunnel temperature monitoring employs distributed fiber optic temperature measurement technology based on Raman scattering. When a laser pulse propagates through the optical fiber, it generates Stokes light and anti-Stokes light, with the anti-Stokes light being temperature-sensitive. By measuring the intensity ratio of the Stokes light to the anti-Stokes light, the temperature distribution along the fiber optic cable is demodulated. This technology enables continuous temperature monitoring within the range of -40℃ to 120℃, with a measurement accuracy of ±1℃.

[0078] (II) Tunnel Strain Monitoring

[0079] Tunnel strain monitoring employs Brillouin optical time-domain analysis (BDI). When an optical fiber is subjected to axial strain, the Brillouin frequency shift within the fiber changes. By measuring the Brillouin frequency shift, the strain distribution along the fiber can be calculated. Combined with a three-dimensional displacement inversion algorithm, it is possible to monitor tunnel structural deformation, including convergence deformation and settlement deformation.

[0080] (III) Tunnel Humidity Monitoring

[0081] Tunnel humidity monitoring utilizes a moisture-sensitive material coating on the surface of optical fibers to detect humidity changes by monitoring the center wavelength shift of an ultra-weak fiber grating. The moisture-sensitive material is preferably polyimide, whose expansion upon water absorption or contraction upon water loss causes axial strain in the grating, resulting in a center wavelength shift. Quantitative humidity monitoring is achieved by calibrating the relationship between the center wavelength shift and relative humidity.

[0082] Example 4

[0083] This embodiment provides a detailed description of the specific implementation method for multi-parameter monitoring of the track structure.

[0084] (a) Track vibration monitoring

[0085] Track vibration monitoring utilizes ultra-weak fiber optic grating arrays to sense dynamic strain caused by vibration by detecting the reflected wavelength shift. When a train passes, the track vibrates, causing a change in the grating spacing, which in turn leads to a shift in the reflected center wavelength. By acquiring the wavelength shift in real time using high-speed demodulation technology, the vibration amplitude and frequency can be calculated.

[0086] In critical areas such as turnout switch rails, frog points, and rail welded joints, the grating spacing is increased to 0.5 meters to capture vibration characteristics of faults such as loose fasteners and rail cracks. The vibration signals generated when a normal train passes have different time-frequency characteristics than those generated when the track structure is abnormal, which can be used as a basis for fault diagnosis.

[0087] (ii) Track temperature monitoring

[0088] Track temperature monitoring also employs distributed fiber optic temperature measurement technology based on Raman scattering to monitor track temperature in real time. Due to the thermal expansion and contraction characteristics of track materials, temperature changes can cause changes in track structural stress, potentially leading to track deformation or even breakage. Real-time monitoring of track temperature provides data support for preventing structural damage caused by thermal expansion and contraction, ensuring train operation safety.

[0089] Example 5

[0090] This embodiment provides a detailed description of temperature crosstalk compensation technology in multi-parameter monitoring.

[0091] When measuring optical cable strain using Brillouin optical time-domain analysis (OTD), the Brillouin frequency shift is affected by both strain and temperature, and these two factors are coupled together and difficult to distinguish. This embodiment compensates for the strain measurement results using temperature data measured by distributed fiber optic thermometry within the same optical cable, eliminating temperature crosstalk and achieving high-precision strain measurement. The compensation formula is: ε = (Δν_B - C_T·ΔT) / C_ε, where Δν_B is the change in Brillouin frequency shift, ΔT is the temperature change, C_T is the temperature coefficient, and C_ε is the strain coefficient.

[0092] Similarly, ultra-weak fiber optic grating arrays are also affected by temperature crosstalk when monitoring humidity and vibration. The wavelength drift of the grating is affected by both strain and temperature. Compensation can be achieved by using temperature data measured by distributed fiber optic temperature measurement technology in the same optical cable, which can separately realize high-precision measurement of humidity and vibration.

[0093] Example 6

[0094] This embodiment provides a detailed description of the specific implementation of the signal processing and decision-making module 3.

[0095] The signal processing and decision-making module 3 includes a signal preprocessing unit, a feature extraction unit, a multi-source data fusion unit, and an intelligent decision-making unit.

[0096] (a) Signal preprocessing unit

[0097] The signal preprocessing unit performs noise reduction on the demodulated multi-parameter signal to remove environmental and system noise. The methods employed include at least one of wavelet thresholding, Kalman filtering, and moving average filtering to improve the signal-to-noise ratio.

[0098] (ii) Feature extraction unit

[0099] The feature extraction unit extracts time-domain, frequency-domain, and time-frequency-domain feature information from the denoised signal using at least one of the following methods: wavelet transform, Fourier analysis, and empirical mode decomposition. For vibration signals, the extracted features include peak value, root mean square, spectral energy distribution, and wavelet packet energy; for strain, temperature, and humidity signals, the extracted features include rate of change, spatial gradient, and time series statistics.

[0100] (III) Multi-source data fusion unit

[0101] The multi-source data fusion unit is used to perform spatiotemporal alignment and correlation analysis of multi-parameter signals such as temperature, vibration, strain, and humidity. By aligning monitoring data of different parameters in the same spatiotemporal coordinate system, data silos are eliminated, enabling collaborative diagnosis of multiple physics fields. For example, correlating temperature changes with strain changes at the same location can determine the degree of influence of temperature stress on structural deformation.

[0102] (iv) Intelligent Decision-Making Unit

[0103] The intelligent decision-making unit utilizes a pre-trained machine learning model to perform pattern recognition and classification on the extracted multi-source fusion features. The machine learning model includes at least one of convolutional neural networks, long short-term memory networks, and support vector machines.

[0104] The training process of the machine learning model includes: collecting historical monitoring data, including normal operating condition data and various fault operating condition data; labeling the data and establishing a training sample set; extracting multi-parameter features and constructing feature vectors; and using the training sample set to train the machine learning model and optimize the model parameters.

[0105] The trained machine learning model can achieve the following functions:

[0106] First, track crack diagnosis: By analyzing the time-frequency characteristics of vibration signals and combining them with the abrupt changes in strain signals, the location and severity of track cracks can be identified.

[0107] Second, equipment overheating early warning: By monitoring the temperature change trend of key equipment, combined with ambient temperature and load parameters, the risk of equipment overheating is predicted.

[0108] Third, train vibration mode classification: distinguish between normal train vibration and abnormal track structure vibration, and identify fault modes such as fastener loosening and track bed settlement.

[0109] Fourth, multi-parameter coupled fault diagnosis: by integrating multi-parameter information such as temperature, strain, and vibration, it identifies composite faults caused by the coupled effects of multiple factors, such as track structure damage caused by the combined action of temperature stress and train dynamic load.

[0110] In summary, the present invention has the following beneficial effects:

[0111] (i) High reliability: It adopts fiber optic sensing technology and utilizes the characteristics of optical signal transmission to have natural anti-electromagnetic interference capabilities. It can operate stably in harsh environments such as strong magnetic fields, humidity, and high temperature, overcoming the shortcomings of traditional electrical sensors in complex subway environments.

[0112] (ii) Synchronous sensing of multiple parameters: By integrating distributed optical fiber temperature measurement technology, Brillouin optical time domain analysis technology and ultra-weak fiber grating array technology, synchronous monitoring of multiple parameters such as temperature, vibration, strain and humidity is realized in the same optical fiber network, providing multi-dimensional data support for the safety status assessment of rail transit.

[0113] (III) Integrated Sensing and Communication Transmission: Using wavelength division multiplexing technology, multi-parameter sensing data and communication service data are transmitted in the same optical fiber, which greatly improves the utilization rate of optical fiber resources and reduces the system deployment and maintenance costs.

[0114] (iv) High-precision measurement: By using temperature crosstalk compensation technology, the influence of temperature changes on strain, humidity and vibration measurements is eliminated, which significantly improves the measurement accuracy of each parameter.

[0115] (v) Intelligent diagnostic capability: By combining machine learning algorithms to perform pattern recognition on multi-source fusion features, a progressive evaluation from single-parameter anomaly detection to multi-parameter coupled fault diagnosis is realized, which effectively reduces the false alarm rate and the false alarm rate.

[0116] (vi) Full life cycle monitoring: The system can realize continuous and automated monitoring of the entire life cycle of subway tracks and tunnel structures, providing a data foundation for condition-based repair and preventive maintenance. It is a key enabling technology to ensure the safe, efficient and intelligent operation of subway traffic.

[0117] The above specific embodiments are used to explain and illustrate the present invention, and are only preferred embodiments of the present invention, not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made to the present invention within the spirit and scope of the claims shall fall within the protection scope of the present invention.

Claims

1. A multi-parameter traffic safety monitoring system for subway tracks based on integrated fiber optic sensing, characterized in that: It includes a sensing and communication module (1), a demodulation module (2), a signal processing and decision-making module (3), a control module (4), and an application module (5) connected in sequence, wherein, The sensing and communication module (1) utilizes an optical fiber sensing array deployed in the subway track and tunnel structure to collect multi-parameter physical information such as temperature, vibration, strain and humidity in real time based on distributed optical fiber sensing technology; and based on optical fiber sensing integration technology, reuses the optical fiber sensing array as a communication transmission medium to realize the co-transmission of multi-parameter sensing data and subway communication service data in the same optical fiber. The demodulation module (2) converts the received optical signal into an electrical signal and uses time-division multiplexing technology to separate signals of different wavelengths in order to demodulate the corresponding physical parameter values. The signal processing and decision module (3) performs noise reduction and feature extraction on the demodulated signal and uses a preset machine learning model to identify abnormal events in the track or tunnel structure. The control module (4) and the application module (5) are used to receive the decision results of the signal processing and decision module (3) and execute at least one of the system's centralized control, status monitoring, early warning release, emergency control and operation and maintenance functions.

2. The multi-parameter traffic safety monitoring system for subway tracks based on integrated fiber optic sensing as described in claim 1, characterized in that: In the sensing and communication module (1), a fiber optic grating sensing array cable integrating DTS, BOTDA and uwFBG arrays is laid along the subway track and tunnel structure. Among them, DTS based on Raman scattering is used for distributed temperature monitoring across the entire area, BOTDA based on Brillouin scattering is used for strain and temperature monitoring, and uwFBG array based on ultra-weak vibration and humidity monitoring of key parts is used. Multi-parameter sensing includes synchronous acquisition and spatiotemporal correlation analysis of track vibration caused by train dynamic load, track structure strain, tunnel structure temperature field and humidity field, so as to realize a comprehensive assessment of track safety status under vehicle-track-tunnel coupling. The time-division multiplexed fiber optic grating sensing array cable is used to allocate independent wavelengths for the multi-parameter physical information and communication services of the sensing, and finally realizes the synchronous transmission of multi-parameter sensing and communication data of temperature, vibration, strain and humidity.

3. The multi-parameter traffic safety monitoring system for subway tracks based on integrated fiber optic sensing as described in claim 2, characterized in that: The fiber optic grating sensing array optical cable is prepared by drawing a wire tower, and an ultra-weak fiber optic grating array is engraved inside the optical cable to form a sensing network with a spatial resolution of less than 1 meter and a monitoring distance of not less than 10 kilometers. The fiber optic grating sensing array is laid in the following manner: at least three optical cables are laid longitudinally on the center line of the tunnel arch and on both side walls, with the optical cables on both side walls laid at a height of 1.5 to 2 meters above the track surface, so as to realize multi-parameter monitoring of tunnel structure strain, temperature, humidity and track structure vibration and temperature.

4. The multi-parameter traffic safety monitoring system for subway tracks based on fiber optic sensing integration as described in claim 3, characterized in that: Each individual optical fiber employs time-division multiplexing technology to achieve integrated multi-parameter sensing and communication. Independent wavelength channels are divided within the same optical fiber, including sensing signal processing units and data communication units. The sensing signal processing unit includes: A distributed temperature sensing module, operating in the 1310 nm band, is used to acquire and process the temperature distribution signal along the single-mode optical fiber. A humidity sensing module is optically connected to an ultra-weak fiber grating array inscribed in the single-mode optical fiber. The center wavelength of the ultra-weak fiber grating array is located in the 1500-nanometer band. The module is used to obtain humidity signals by detecting the wavelength drift of the ultra-weak fiber grating. A distributed strain sensing module, operating in the 1550 nm band, is used to acquire and process strain distribution signals along the single-mode optical fiber. The data communication unit operates in the C-band and transmits communication data through the single-mode optical fiber. It is used to transmit video surveillance, equipment status information and maintenance and testing instructions along the subway line. The optical signals in the 1310 nm, 1500 nm, 1550 nm and C-band bands are transmitted in parallel in the same single-mode optical fiber using time-division multiplexing technology, without interfering with each other.

5. The multi-parameter traffic safety monitoring system for subway tracks based on fiber optic sensing integration as described in claim 1, characterized in that: The demodulation module (2) is connected to the sensing / communication module (1) and is used to receive composite optical signals from the fiber optic grating sensing array optical cable and convert the optical signals into electrical signals. The demodulation module (2) integrates a time-division multiplexing device to separate the communication wavelength channel in the composite signal from the sensing wavelength channel with different physical parameters; The demodulation module (2) is configured to demodulate the separated multi-parameter physical information wavelength channel signal to obtain parameters from vibration frequency, temperature change, strain value and humidity value.

6. The multi-parameter traffic safety monitoring system for subway tracks based on fiber optic sensing integration as described in claim 1, characterized in that: The signal processing and decision-making module (3), connected to the demodulation module (2), is used to process and make decisions on the demodulated key parameter signals, and includes: The signal preprocessing unit performs noise reduction processing on the signal; The multi-source data fusion unit is used to perform spatiotemporal alignment and correlation analysis on multi-parameter signals such as temperature, vibration, strain, and humidity to eliminate data silos and achieve collaborative diagnosis of multi-physics fields. The feature extraction unit extracts feature information from the denoised signal using at least one of wavelet transform and Fourier analysis methods. The intelligent decision-making unit is used to perform pattern recognition and classification on the extracted features by a pre-trained machine learning model, thereby realizing the abnormal diagnosis of track cracks and equipment overheating, and realizing the pattern classification of normal train vibration and track structure abnormalities.

7. The multi-parameter traffic safety monitoring system for subway tracks based on integrated fiber optic sensing as described in claim 6, characterized in that: Pre-trained machine learning models, including convolutional neural networks and long short-term memory networks, are used to implement pattern classification.

Images