Intelligent monitoring system and method for deformation of long tunnel

Abstract

The invention relates to an intelligent monitoring system and method for deformation of a long tunnel, and the system comprises a multi-dimensional optical fiber sensing network, a data collection and demodulation unit, a data processing and intelligent analysis center, and a safety early warning module, and the multi-dimensional optical fiber sensing network comprises a distributed linear optical fiber sensing system, an optical fiber grating sensing element, and an optical fiber inertia measurement unit. The distributed linear optical fiber sensing system is used for tunnel longitudinal strain distribution measurement and vibration and acoustic signal sensing; the fiber bragg grating sensing element is used for monitoring circumferential strain and longitudinal strain of a key point and longitudinal differential settlement of a tunnel; the optical fiber inertial measurement unit measures overall axis deflection and torsional deformation; the data acquisition and demodulation unit is used for acquiring and demodulating the sensing signal; the data processing and intelligent analysis center is used for deformation category identification, trend prediction and graded early warning; and the safety early warning module sends early warning information to operation management personnel. According to the invention, full-section, high-precision and real-time online monitoring and intelligent early warning of the long tunnel can be realized.

Description

Technical Field

[0001] This invention relates to the field of tunnel engineering health monitoring and safety early warning technology, and in particular to an intelligent monitoring system and method for deformation of long tunnels. Background Technology

[0002] Long tunnels (such as railway tunnels, highway tunnels, and water conveyance tunnels) are susceptible to various factors during operation, including changes in geological conditions, groundwater erosion, disturbance from surrounding construction, and material aging. These factors can lead to structural defects such as convergence deformation, settlement, and misalignment, seriously threatening the safety of tunnel operation. Therefore, effective deformation monitoring of long tunnels is crucial to ensuring their safe operation.

[0003] Traditional tunnel deformation monitoring methods mainly rely on equipment such as total stations and convergence meters, and are conducted through regular manual inspections. However, these traditional methods have many drawbacks: First, they are discrete point measurements, which can only acquire data from a limited number of measuring points and cannot reflect the continuous deformation state of the entire tunnel, easily missing key defects. Second, they have poor timeliness; the long cycle of manual inspections makes it difficult to achieve real-time, continuous, and automated monitoring, and cannot provide timely warnings of sudden deformations. Third, the measurement accuracy is easily affected by the environment; environmental factors such as humidity, dust, and vibration inside the tunnel can interfere with the measurement accuracy of traditional optical or electronic instruments, leading to inaccurate results. Fourth, single monitoring methods cannot accurately distinguish between different types of deformation in the tunnel, such as longitudinal settlement, lateral convergence, and longitudinal tension / compression, which is not conducive to the analysis of the causes of defects and targeted treatment.

[0004] Therefore, how to provide a comprehensive, accurate, and intelligent monitoring system and method for the deformation of long tunnels is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] This invention provides an intelligent monitoring system and method for deformation of long tunnels to solve the above-mentioned technical problems.

[0006] To address the aforementioned technical problems, this invention provides an intelligent monitoring system for deformation in long tunnels, comprising a multi-dimensional fiber optic sensor network, a data acquisition and demodulation unit, a data processing and intelligent analysis center, and a safety early warning module. The multidimensional fiber optic sensing network includes a distributed linear fiber optic sensing system, fiber optic grating sensing elements, and fiber optic inertial measurement units. The distributed linear fiber optic sensing system is densely laid on the surface of the tunnel lining or embedded in the tunnel lining in a zigzag or spiral pattern, and is used for measuring the longitudinal strain distribution and sensing vibration and acoustic signals along the entire length of the tunnel. The fiber optic grating sensing elements are deployed at key sections of the tunnel, including fiber optic grating strain sensors and fiber optic grating hydrostatic levels, which are used to monitor the circumferential strain, longitudinal strain, and longitudinal uneven settlement of the tunnel at key points, respectively. The fiber optic inertial measurement units are deployed at preset intervals within the tunnel, using a fiber optic gyroscope / inclinometer system to provide an absolute attitude reference for the monitoring system and measure the overall axial deflection and torsional deformation of the tunnel. The data acquisition and demodulation unit integrates a BOTDA demodulator, a DAS demodulator, and an FBG demodulator, and is used to acquire and demodulate the sensing signals output by the multidimensional fiber optic sensing network. The data processing and intelligent analysis center includes a data fusion module, a deformation field reconstruction module, an intelligent diagnosis module, and an early warning module. The data fusion module is used to perform spatiotemporal registration and fusion of BOTDA distributed strain data, FBG point strain and settlement data, and fiber optic inertial measurement unit absolute displacement data. The deformation field reconstruction module establishes a tunnel structure deformation field model based on the fused data and visualizes a three-dimensional deformation cloud map of the entire tunnel. The intelligent diagnosis module and the early warning module realize deformation category identification, trend prediction, and graded early warning through feature extraction units and machine learning models. The safety early warning module is used to send early warning information to operation and management personnel in real time.

[0007] Preferably, the distributed linear fiber optic sensing system employs BOTDA and DAS technologies, wherein: the BOTDA technology is used to measure the longitudinal strain distribution along the entire length of the tunnel, and the DAS technology is used to realize real-time sensing of vibration and acoustic signals along the tunnel and locate abnormal events.

[0008] Preferably, the distributed linear fiber optic sensing system includes a BOTDA subsystem and a DAS subsystem; The BOTDA subsystem includes a first laser, an electro-optic modulator, a frequency scanning and synchronization control unit, a circulator / coupler, a photodetector, and a signal processing and demodulation unit. The first laser generates a frequency-stable narrow-linewidth laser, which is modulated by the electro-optic modulator into a pulse probe beam and a continuous pump beam. The frequency scanning and synchronization control unit controls the frequency difference between the two beams and scans them. The circulator / coupler separates the injected and returned signals. The photodetector detects the power change of the returned pump beam. The signal processing and demodulation unit reconstructs the Brillouin gain spectrum and extracts the Brillouin frequency shift. Combined with the temperature compensation data of the reference optical cable, the strain distribution is calculated. The DAS subsystem includes a second laser, an acousto-optic modulator, an optical fiber amplifier, an interferometer, a high-speed data acquisition card, and a digital signal processor. The second laser generates coherent laser light, which is modulated into high-frequency narrow pulse light by the acousto-optic modulator. The optical fiber amplifier increases the power of the input optical light. The interferometer detects the phase change of the scattered light. The high-speed data acquisition card acquires the light intensity signal. The digital signal processor reconstructs the three-dimensional information of "position-time-vibration amplitude" through phase unwrapping, differential calculation, and digital filtering.

[0009] Preferably, the feature extraction unit of the intelligent diagnosis module and the early warning module extracts key features from strain, settlement, and vibration data, and the key features include at least deformation rate, deformation curvature, and frequency of abnormal events; the machine learning model uses LSTM temporal network and CNN spatial network algorithms to analyze historical data and real-time data.

[0010] Preferably, the deformation category identification includes automatically identifying typical deformation patterns, which at least include convergence deformation, settlement deformation, and axis offset; the trend prediction is based on time series analysis to predict the deformation development trend over a future period; the graded early warning includes four levels: blue observation, yellow warning, orange alert, and red emergency, and the corresponding level of early warning is triggered according to the deformation amount, deformation rate, and pattern recognition results.

[0011] Preferably, the distributed linear fiber optic sensing system is laid out at a density of one ring every 5 meters, and the fiber optic grating sensing elements are arranged at intervals of 200 meters.

[0012] This invention also provides an intelligent monitoring method for deformation of long tunnels based on multi-source optical fiber sensing, which applies the monitoring system described above and includes the following steps: Step 1: Deploy the multi-dimensional fiber optic sensing network by laying three sensing optical cables tightly in a zigzag or spiral pattern along both sides of the tunnel and the arch. Deploy the fiber optic strain sensor and fiber optic static level at key sections, and install one fiber optic inertial measurement unit at each preset interval. Step 2: The data acquisition and demodulation unit synchronously acquires BOTDA sensor signals, DAS sensor signals, FBG sensor signals and fiber optic inertial measurement unit signals, and demodulates the signals using the integrated demodulator. Step 3: The data processing and intelligent analysis center receives the demodulated signal and performs spatiotemporal registration and multi-source data fusion through the data fusion module; Step 4: Based on the fused data, establish a deformation field model of the tunnel structure through the deformation field reconstruction module to generate a three-dimensional deformation cloud map; Step 5: The intelligent diagnosis module and the early warning module extract data features, use the machine learning model to identify deformation patterns, predict deformation development trends, and trigger corresponding level early warnings based on deformation amount, deformation rate, and pattern recognition results. Step 6: The safety warning module sends warning information in real time through audible and visual alarms, SMS, or App push notifications.

[0013] Preferably, step 2, the signal acquisition and demodulation of the BOTDA sensor signal includes the following steps: Step 11: Inject continuous pump light and frequency-tunable probe light from both ends of the optical fiber, respectively; Step 12: Scan the frequency difference between the two beams within the set frequency range and record the spatiotemporal data of the amplification of pump light power during the propagation of the probe light pulse at each frequency difference; Step 13: Generate a "power-frequency difference" Brillouin gain spectrum for each location of the optical fiber, and extract the Brillouin frequency shift ν_B corresponding to the peak value; Step 14: Compare the measured Brillouin frequency shift ν_B with the initial reference ν_B0, and calculate the strain change Δε using the formula ν_B(ε,T)=ν_B0+C_ε×Δε+C_T×ΔT, where C_ε is the strain coefficient, C_T is the temperature coefficient, and ΔT is the temperature change. Temperature compensation is performed by arranging an unstressed reference optical cable.

[0014] Preferably, step 2, the signal acquisition and demodulation of the DAS sensor signal includes the following steps: Step 21: Inject a laser pulse of a set frequency into the sensing fiber; Step 22: Receive the Rayleigh backscattered signal generated during pulse propagation and interfere with the reference light in the interferometer; Step 23: The high-speed data acquisition card records the backscattering curve of the entire optical fiber corresponding to each laser pulse; Step 24: Compare the backscattering curves of multiple consecutive pulses, analyze the phase difference or amplitude change at a specific position, determine the vibration occurrence, and reconstruct the vibration waveform.

[0015] Preferably, the deformation modes in step 5 include convergent deformation, settlement deformation, and axis offset, and the trend prediction is based on time series analysis to predict the deformation development trend over a future period of time.

[0016] Compared with existing technologies, the intelligent monitoring system and method for deformation of long tunnels provided by this invention have the following advantages: 1. Monitoring accuracy has been greatly improved. (1) Multi-source data fusion: Combining distributed strain, point-based high-precision strain / settlement, and absolute reference displacement data, mutual verification and supplementation are achieved, avoiding the limitations of a single technology and improving the reliability of monitoring data; (2) Temperature compensation mechanism: Temperature compensation is achieved by arranging a reference optical cable, which effectively eliminates the interference of temperature changes on the measurement results and ensures the accuracy of strain measurement; (3) Dynamic reference correction: The cumulative error in displacement calculation by strain integral is corrected by the absolute attitude reference provided by the fiber optic inertial measurement unit, which significantly improves the absolute accuracy of displacement measurement.

[0017] 2. Achieve full coverage and real-time monitoring (1) The distributed sensor network is laid with a density of 5 meters per circle, realizing continuous monitoring of "every meter" of the tunnel, with no blind spots, and completely solving the problem that key defects are easily missed by traditional discrete point measurement. (2) The entire system operates automatically, enabling real-time monitoring and response. It is highly timely and can provide early warning of sudden deformations.

[0018] 3. High level of intelligence (1) It can not only "perceive" tunnel deformation, but also "diagnose" the type, cause and development stage of deformation through AI algorithm, realizing the leap from "monitoring" to "early warning" and then to "prediction"; (2) The graded early warning mechanism makes management measures more targeted and effective. Managers can take corresponding measures according to the early warning level to improve the efficiency of tunnel safety management.

[0019] 4. Excellent durability and economy (1) Fiber optic sensors have a long lifespan (up to 30 years), and are characterized by electromagnetic interference resistance, corrosion resistance, intrinsic safety, etc. They have low maintenance costs, and the total life cycle cost is far lower than that of frequent manual inspections and traditional equipment replacement. (2) A system integrates multiple monitoring functions and can simultaneously monitor multiple parameters such as strain, settlement, vibration, and attitude. It is cost-effective and reduces the overall investment in tunnel monitoring. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the layout of the intelligent monitoring system for deformation of long tunnels in a specific embodiment of the present invention; Figure 2 This is a schematic diagram of the working logic of a distributed linear fiber optic sensing system in a specific embodiment of the present invention. Figure 3 This is a side view of a tunnel (sensor element layout) in a specific embodiment of the present invention. Figure 4 This is a flowchart of the data processing and intelligent analysis center in a specific embodiment of the present invention.

[0021] In the diagram: 001-Tunnel cross-section, 111-First laser, 112-Electro-optic modulator, 113-Frequency scanning and synchronization control unit, 114-Circulator / coupler, 115-Photodetector, 116-Signal processing and demodulation unit, 121-Second laser, 122-Acousto-optic modulator, 123-Fiber optic amplifier, 124-Interferometer, 125-High-speed data acquisition card, 126-Digital signal processor, 131-Fiber Bragg grating strain sensor, 132-Fiber Bragg grating hydrostatic level, 140-Fiber optic inertial measurement unit, 200-Data acquisition and demodulation unit, 300-Data processing and intelligent analysis center, 310-Data fusion module, 320-Deformation field reconstruction module, 330-Intelligent diagnostic module, 340-Early warning module. Detailed Implementation

[0022] To illustrate the technical solutions of the invention in more detail, specific embodiments are listed below to demonstrate the technical effects; it should be emphasized that these embodiments are used to illustrate the invention and not to limit the scope of the invention.

[0023] The intelligent monitoring system for deformation of long tunnels provided by this invention, such as Figures 1 to 4 As shown, it includes a multi-dimensional fiber optic sensor network, a data acquisition and demodulation unit 200, a data processing and intelligent analysis center 300, and a security early warning module, wherein: The multidimensional fiber optic sensing network includes a distributed linear fiber optic sensing system, fiber optic grating sensing elements, and a fiber optic inertial measurement unit 140. The distributed linear fiber optic sensing system is laid tightly on the lining surface of the tunnel section 001 in a zigzag or spiral pattern or embedded within the tunnel lining, and is used for measuring the longitudinal strain distribution and sensing vibration and acoustic signals along the entire length of the tunnel. The fiber optic grating sensing elements are deployed at key tunnel sections (such as sections with poor geological conditions, tunnel entrances, and near connecting passages), including a fiber optic grating strain sensor 131 and a fiber optic grating hydrostatic level 132, which are used to monitor the circumferential strain, longitudinal strain, and longitudinal uneven settlement of the tunnel at key points, respectively. Specifically, the fiber optic grating strain sensor 131 monitors the circumferential and longitudinal strain of key points at high frequency and with high precision, serving as a supplement and verification of the accuracy of the distributed system; the fiber optic grating hydrostatic level 132 is deployed on the tunnel floor and is used to accurately measure the longitudinal uneven settlement of the tunnel. The fiber optic inertial measurement unit 140 is deployed in the tunnel at preset intervals (e.g., 500 meters). It uses a fiber optic gyroscope / inclinometer system to provide an absolute attitude reference for the monitoring system, measure the overall axial deflection and torsional deformation of the tunnel, and correct the errors generated when calculating displacement from accumulated strain.

[0024] The data acquisition and demodulation unit 200 integrates a BOTDA demodulator, a DAS demodulator, and an FBG demodulator, and is used to synchronously, at high speed, and continuously acquire and demodulate all sensing signals output by the multidimensional fiber optic sensing network.

[0025] The data processing and intelligent analysis center 300 includes a data fusion module 310, a deformation field reconstruction module 320, an intelligent diagnosis module 330, and an early warning module 340. The data fusion module 310 is used to perform spatiotemporal registration and fusion of BOTDA distributed strain data, FBG point strain and settlement data, and absolute displacement data of fiber optic inertial measurement unit 140. The deformation field reconstruction module 320 establishes a tunnel structure deformation field model based on the fused data and visualizes the three-dimensional deformation cloud map of the entire tunnel (including convergence, settlement, and axis deflection). The intelligent diagnosis module 330 realizes deformation category identification, trend prediction, and graded early warning through feature extraction unit and machine learning model.

[0026] The safety early warning module is used to send early warning information to operation and management personnel in real time through sound and light alarms, SMS, App push, and other means.

[0027] This invention enables full-dimensional monitoring of "point-line-surface", solving the problem of easy omission of diseases in traditional discrete point measurement; multi-module collaboration realizes full-process automation from data acquisition to early warning, replacing manual inspection and improving monitoring timeliness.

[0028] In some embodiments, please refer to the following: Figure 2The distributed linear fiber optic sensing system employs BOTDA (Brillouin Optical Time Domain Analysis) and DAS (Rayleigh Scattering Distributed Acoustic Sensing) technologies. BOTDA is used to measure the longitudinal strain distribution along the entire tunnel length, while DAS is used to achieve real-time sensing of vibration and acoustic signals along the tunnel, locating abnormal events. Specifically, by monitoring the continuous change of the Brillouin frequency shift along the fiber optic cable, the longitudinal strain distribution along the entire tunnel length is accurately measured, acquiring tensile or compressive deformation signals. Real-time sensing of vibration and acoustic signals generated along the tunnel by structural cracking, spalling, and rock loosening is used to locate abnormal events (such as rockfalls or collisions). The two technologies complement each other, covering the monitoring needs of "static strain + dynamic vibration," improving the comprehensiveness of tunnel defect identification.

[0029] Specifically, the distributed linear fiber optic sensing system includes a BOTDA subsystem and a DAS subsystem; The BOTDA subsystem includes a first laser 111, an electro-optic modulator 112, a frequency scanning and synchronization control unit 113, a circulator / coupler 114, a photodetector 115, and a signal processing and demodulation unit 116. The first laser 111 serves as a light source, generating a frequency-stable narrow-linewidth laser. The electro-optic modulator 112 modulates the laser into a high-power pulsed light (probe light) and a continuous pump light. The frequency scanning and synchronization control unit 113 controls the frequency difference between the two beams and performs scanning. The circulator / coupler 114 separates the injected and returned signals. The photodetector 115 detects the power change of the returned pump light. The signal processing and demodulation unit 116 reconstructs the Brillouin gain spectrum and extracts the Brillouin frequency shift at each position. Combined with the temperature compensation data of the reference optical cable, the strain distribution is calculated.

[0030] Specifically, a continuous pump light is injected into one end of an optical fiber, and a pulsed probe light is injected into the other end. When the two beams meet in the optical fiber, and the difference between the frequency of the probe light and the frequency of the pump light is exactly equal to the Brillouin frequency shift of the local optical fiber, the strongest energy transfer can occur due to the stimulated Brillouin amplification effect.

[0031] By scanning the frequency difference between the probe light and the pump light, and recording the time from the emission of the probe pulse to the receipt of the amplified signal at each frequency point, the location and magnitude of the strain / temperature change can be determined simultaneously.

[0032] The position is calculated as follows: L = (c × t) / (2 × n). Where L is the distance, c is the speed of light, t is the time, and n is the refractive index of the optical fiber.

[0033] The strain / temperature change is calculated as follows: ν_B(ε,T)=ν_B0+C_ε×Δε+C_T×ΔT. Where: ν_B is the measured Brillouin frequency shift; ν B0 is the initial (no strain, at reference temperature) Brillouin frequency shift; C ε is the strain coefficient (approximately 0.05 MHz / με); Δε is the strain change (microstrain); C_T is the temperature coefficient (approximately 1.0 MHz / °C); ΔT is the temperature change (degrees Celsius).

[0034] The DAS subsystem includes a second laser 121, an acousto-optic modulator 122, an optical fiber amplifier 123, a Michelson or Mach-Zehnder interferometer 124, a high-speed data acquisition card 125, and a digital signal processor 126. The second laser 121 generates coherent laser light with extremely low phase noise. The acousto-optic modulator 122 modulates the continuous laser light into a narrow pulse light with a high frequency and high repetition rate. The optical fiber amplifier 123 amplifies the pulse light, increases the input optical power, and enhances the backscattered signal. The interferometer 124 is the core demodulation unit used to detect the phase change of the scattered light. The high-speed data acquisition card 125 acquires the light intensity signal output by the interferometer at an extremely high sampling rate. The digital signal processor 126 is used to execute complex algorithms (such as phase unwrapping, differential calculation, and digital filtering) to restore the large amount of spatiotemporal data acquired into three-dimensional information of "position-time-vibration amplitude".

[0035] In some embodiments, please refer to the following: Figure 4 The feature extraction unit of the intelligent diagnostic module 330 extracts key features from strain, settlement, and vibration data. The key features include at least deformation rate, deformation curvature, and frequency of abnormal events. The machine learning model uses LSTM temporal network and CNN spatial network algorithms to analyze historical and real-time data. This embodiment achieves an upgrade from "data acquisition" to "intelligent diagnosis" through quantified features and intelligent algorithms, replacing manual experience judgment and improving the accuracy of deformation recognition and the foresight of trend prediction.

[0036] In some embodiments, please refer to Figure 4 The deformation category identification includes automatically identifying typical deformation patterns, which at least include convergence deformation, settlement deformation, and axis offset; the trend prediction is based on time series analysis to predict the deformation development trend over a future period; the graded early warning includes four levels: blue observation, yellow warning, orange alert, and red emergency, which are triggered according to the deformation amount, deformation rate, and pattern recognition results. The graded early warning mechanism is used to match different risk levels with appropriate handling strategies to avoid over-warning or under-warning; deformation pattern identification clarifies the type of damage, providing a basis for targeted reinforcement.

[0037] In some embodiments, please refer to the following: Figure 3The distributed linear fiber optic sensing system is laid out at a density of 5 meters per ring, and the fiber optic grating sensing elements are deployed at a spacing of 200 meters. Standardized deployment parameters ensure the consistency of monitoring data, and encrypted risk sections improve the accuracy of disease identification. Reasonable deployment intervals balance monitoring accuracy and cost investment.

[0038] This invention also provides an intelligent monitoring method for deformation of long tunnels based on multi-source optical fiber sensing, which applies the monitoring system described above and includes the following steps: Step 1: Deploy the multi-dimensional fiber optic sensing network by laying three sensing optical cables tightly in a zigzag or spiral pattern along the side walls and arch of the tunnel; deploy the fiber optic strain sensor 131 and fiber optic static level 132 at key cross sections, and install one fiber optic inertial measurement unit 140 at each preset interval. Step 2: The data acquisition and demodulation unit 200 synchronously acquires the BOTDA sensor signal, DAS sensor signal, FBG sensor signal and the signal from the fiber optic inertial measurement unit 140, and demodulates the signal using the integrated demodulator. Step 3: The data processing and intelligent analysis center 300 receives the demodulated signal and performs spatiotemporal registration and multi-source data fusion through the data fusion module 310; Step 4: Based on the fused data, a deformation field model of the tunnel structure is established through the deformation field reconstruction module 320 to generate a three-dimensional deformation cloud map; Step 5: The intelligent diagnostic module 330 extracts data features, uses the machine learning model to identify deformation patterns, predicts deformation development trends, and triggers corresponding level warnings based on deformation amount, deformation rate, and pattern recognition results. Step 6: The safety warning module sends warning information in real time through audible and visual alarms, SMS, or App push notifications.

[0039] By adopting the above method, 24 / 7 uninterrupted monitoring was achieved, improving the efficiency of tunnel safety management.

[0040] In some embodiments, step 2, the signal acquisition and demodulation of the BOTDA sensor signal includes the following steps: Step 11: Inject continuous pump light and frequency-tunable probe light from both ends of the optical fiber.

[0041] Step 12: Scan the frequency difference between the two beams within the set frequency range and record the spatiotemporal data of the pump light power being amplified when the probe light pulse propagates at each frequency difference.

[0042] Step 13: Generate a "power-frequency difference" curve, i.e., a Brillouin gain spectrum, for each location of the optical fiber. The frequency corresponding to the peak of this spectrum is the Brillouin frequency shift ν_B at that point.

[0043] Step 14: Compare the measured Brillouin frequency shift ν_B with the initial reference ν_B0, and calculate the strain change Δε using the formula ν_B(ε,T)=ν_B0+C_ε×Δε+C_T×ΔT. Temperature compensation is performed by arranging an unstressed reference optical cable to eliminate environmental interference and improve the reliability of strain measurement.

[0044] Preferably, step 2, the signal acquisition and demodulation of the DAS sensor signal includes the following steps: Step 21: Inject a laser pulse of a set frequency output by the second laser 121 into the sensing fiber.

[0045] Step 22: Receive the Rayleigh backscattered signal generated during pulse propagation and interfere with the reference light in interferometer 124.

[0046] Step 23: The high-speed data acquisition card 125 records the backscattering curve of the entire optical fiber corresponding to each laser pulse.

[0047] Step 24: By comparing the backscatter curves of multiple consecutive pulses and analyzing the phase difference or amplitude change at specific positions between adjacent curves, the vibration occurrence can be determined and the vibration waveform can be restored by the digital signal processor 126 to ensure high-resolution sensing of the vibration signal.

[0048] In some embodiments, the deformation modes in step 5 include convergent deformation, settlement deformation, and axis offset. The trend prediction is based on time series analysis to predict the deformation development trend over a future period of time. The trend prediction provides a time window for taking reinforcement measures in advance to avoid the deterioration of the disease.

[0049] The following is a specific implementation method for a certain construction scenario: 1. Deployment: Three sensing optical cables are laid in a spiral pattern (one loop every 5 meters) along both sides of the tunnel and the arch. Strain sensor arrays (5 arranged in a circumferential direction) and one fiber optic static level 132 are deployed at critical sections every 200 meters. One fiber optic inertial measurement unit 140 is installed every 500 meters.

[0050] 2. Data Acquisition: All demodulation equipment is installed in equipment boxes on the side of the tunnel and connected to the data processing and intelligent analysis center 300 in the central control room via optical cables.

[0051] 3. Operation: The system continuously collects data. When the data processing and intelligent analysis center 300 detects a continuous increase in strain in a certain section, simultaneously monitors minute cracking sound signals, and the sensor at that section also confirms that the strain exceeds the limit, the system determines it as "accelerated convergence deformation," immediately issues a yellow warning, and prompts the system to pay close attention to that section, thus gaining time for timely reinforcement measures.

[0052] In summary, the intelligent deformation monitoring system and method for long tunnels provided by this invention, through the collaboration of a multi-dimensional fiber optic sensor network, a data acquisition and demodulation unit 200, a data processing and intelligent analysis center 300, and a safety early warning module, achieves full-section, high-precision, real-time online deformation monitoring and intelligent early warning for long tunnels.

[0053] Obviously, those skilled in the art can make various modifications and variations to the invention without departing from the spirit and scope of the invention. Therefore, if these modifications and variations fall within the scope of the claims of the invention and their equivalents, the invention is also intended to include these modifications and variations.

Claims

1. A smart monitoring system for deformation of long tunnels, characterized in that, It includes a multi-dimensional fiber optic sensor network, a data acquisition and demodulation unit, a data processing and intelligent analysis center, and a security early warning module. The multidimensional fiber optic sensing network includes a distributed linear fiber optic sensing system, fiber optic grating sensing elements, and fiber optic inertial measurement units. The distributed linear fiber optic sensing system is densely laid on the surface of the tunnel lining or embedded in the tunnel lining in a zigzag or spiral pattern, and is used for measuring the longitudinal strain distribution and sensing vibration and acoustic signals along the entire length of the tunnel. The fiber optic grating sensing elements are deployed at key sections of the tunnel, including fiber optic grating strain sensors and fiber optic grating hydrostatic levels, which are used to monitor the circumferential strain, longitudinal strain, and longitudinal uneven settlement of the tunnel at key points, respectively. The fiber optic inertial measurement units are deployed at preset intervals within the tunnel, using a fiber optic gyroscope / inclinometer system to provide an absolute attitude reference for the monitoring system and measure the overall axial deflection and torsional deformation of the tunnel. The data acquisition and demodulation unit integrates a BOTDA demodulator, a DAS demodulator, and an FBG demodulator, and is used to acquire and demodulate the sensing signals output by the multidimensional fiber optic sensing network. The data processing and intelligent analysis center includes a data fusion module, a deformation field reconstruction module, an intelligent diagnosis module, and an early warning module. The data fusion module is used to perform spatiotemporal registration and fusion of BOTDA distributed strain data, FBG point strain and settlement data, and fiber optic inertial measurement unit absolute displacement data. The deformation field reconstruction module establishes a tunnel structure deformation field model based on the fused data and visualizes the three-dimensional deformation cloud map of the entire tunnel. The intelligent diagnosis module and early warning module achieve deformation category identification, trend prediction and hierarchical early warning through feature extraction unit and machine learning model; The safety early warning module is used to send early warning information to operation and management personnel in real time.

2. The intelligent monitoring system for deformation of long tunnels as described in claim 1, characterized in that, The distributed linear fiber optic sensing system employs BOTDA and DAS technologies. The BOTDA technology is used to measure the longitudinal strain distribution along the entire length of the tunnel, while the DAS technology is used to realize real-time sensing of vibration and acoustic signals along the tunnel and locate abnormal events.

3. The intelligent monitoring system for deformation of long tunnels as described in claim 1, characterized in that, The distributed linear fiber optic sensing system includes a BOTDA subsystem and a DAS subsystem; The BOTDA subsystem includes a first laser, an electro-optic modulator, a frequency scanning and synchronization control unit, a circulator / coupler, a photodetector, and a signal processing and demodulation unit. The first laser generates a frequency-stable narrow-linewidth laser, which is modulated by the electro-optic modulator into a pulse probe beam and a continuous pump beam. The frequency scanning and synchronization control unit controls the frequency difference between the two beams and scans them. The circulator / coupler separates the injected and returned signals. The photodetector detects the power change of the returned pump beam. The signal processing and demodulation unit reconstructs the Brillouin gain spectrum and extracts the Brillouin frequency shift. Combined with the temperature compensation data of the reference optical cable, the strain distribution is calculated. The DAS subsystem includes a second laser, an acousto-optic modulator, an optical fiber amplifier, an interferometer, a high-speed data acquisition card, and a digital signal processor. The second laser generates coherent laser light, which is modulated into high-frequency narrow pulse light by the acousto-optic modulator. The optical fiber amplifier increases the power of the input optical light. The interferometer detects the phase change of the scattered light. The high-speed data acquisition card acquires the light intensity signal. The digital signal processor reconstructs the three-dimensional information of "position-time-vibration amplitude" through phase unwrapping, differential calculation, and digital filtering.

4. The intelligent monitoring system for deformation of long tunnels as described in claim 1, characterized in that, The feature extraction unit of the intelligent diagnosis module and the early warning module extracts key features from strain, settlement and vibration data. The key features include at least deformation rate, deformation curvature and abnormal event frequency. The machine learning model employs LSTM temporal network and CNN spatial network algorithms to analyze historical and real-time data.

5. The intelligent monitoring system for deformation of long tunnels as described in claim 1, characterized in that, The deformation category identification includes automatically identifying typical deformation patterns, which at least include convergence deformation, settlement deformation, and axis offset. The trend prediction is based on time series analysis to predict the deformation development trend over a future period of time. The tiered early warning system includes four levels: blue observation, yellow warning, orange alert, and red emergency. The corresponding level of early warning is triggered based on the amount of deformation, the deformation rate, and the pattern recognition results.

6. The intelligent monitoring system for deformation of long tunnels as described in claim 1, characterized in that, The distributed linear fiber optic sensing system is laid out at a density of one ring every 5 meters, and the fiber optic grating sensing elements are deployed at intervals of 200 meters.

7. A method for intelligent monitoring of deformation in long tunnels based on multi-source optical fiber sensing, using the monitoring system as described in any one of claims 1 to 6, characterized in that, Includes the following steps: Step 1: Deploy the multi-dimensional fiber optic sensing network by laying three sensing optical cables tightly in a zigzag or spiral pattern along both sides of the tunnel and the arch. Deploy the fiber optic strain sensor and fiber optic static level at key sections, and install one fiber optic inertial measurement unit at each preset interval. Step 2: The data acquisition and demodulation unit synchronously acquires BOTDA sensor signals, DAS sensor signals, FBG sensor signals and fiber optic inertial measurement unit signals, and demodulates the signals using the integrated demodulator. Step 3: The data processing and intelligent analysis center receives the demodulated signal and performs spatiotemporal registration and multi-source data fusion through the data fusion module; Step 4: Based on the fused data, establish a deformation field model of the tunnel structure through the deformation field reconstruction module to generate a three-dimensional deformation cloud map; Step 5: The intelligent diagnosis module and the early warning module extract data features, use the machine learning model to identify deformation patterns, predict deformation development trends, and trigger corresponding level early warnings based on deformation amount, deformation rate, and pattern recognition results. Step 6: The safety warning module sends warning information in real time through audible and visual alarms, SMS, or App push notifications.

8. The intelligent monitoring method for deformation of long tunnels as described in claim 7, characterized in that, Step 2, the signal acquisition and demodulation of the BOTDA sensor signal includes the following steps: Step 11: Inject continuous pump light and frequency-tunable probe light from both ends of the optical fiber, respectively; Step 12: Scan the frequency difference between the two beams within the set frequency range and record the spatiotemporal data of the amplification of pump light power during the propagation of the probe light pulse at each frequency difference; Step 13: Generate a "power-frequency difference" Brillouin gain spectrum for each location of the optical fiber, and extract the Brillouin frequency shift ν_B corresponding to the peak value; Step 14: Compare the measured Brillouin frequency shift ν_B with the initial reference ν_B0, and calculate the strain change Δε using the formula ν_B(ε,T)=ν_B0+C_ε×Δε+C_T×ΔT, where C_ε is the strain coefficient, C_T is the temperature coefficient, and ΔT is the temperature change. Temperature compensation is performed by arranging an unstressed reference optical cable.

9. The intelligent monitoring method for deformation of long tunnels as described in claim 7, characterized in that, Step 2, the signal acquisition and demodulation of the DAS sensor signal includes the following steps: Step 21: Inject a laser pulse of a set frequency into the sensing fiber; Step 22: Receive the Rayleigh backscattered signal generated during pulse propagation and interfere with the reference light in the interferometer; Step 23: The high-speed data acquisition card records the backscattering curve of the entire optical fiber corresponding to each laser pulse; Step 24: Compare the backscattering curves of multiple consecutive pulses, analyze the phase difference or amplitude change at a specific position, determine the vibration occurrence, and reconstruct the vibration waveform.

10. The intelligent monitoring method for deformation of long tunnels as described in claim 7, characterized in that, The deformation modes mentioned in step 5 include convergent deformation, settlement deformation, and axis offset. The trend prediction is based on time series analysis to predict the deformation development trend over a future period of time.

Images