Downhole trapped personnel life state detection and positioning system based on distributed optical fiber acoustic wave sensing

Abstract

The application discloses a downhole trapped personnel life state detection and positioning system based on distributed optical fiber acoustic sensing, which comprises a downhole sensing layer, the downhole sensing layer comprises armored single-mode sensing optical fiber, the sensing optical fiber adopts a double deployment mode of permanent pre-burial combined with temporary cloth supplement, the permanent pre-burial is performed along the whole process of downhole roadways, working faces, refuge chambers and key pipelines, the temporary cloth supplement is performed through drilling and rescue channels to quickly lower and drag, the sensing optical fiber is fixed to the roadway wall or the anchor rod, and is tightly acoustically bound to the downhole surrounding rock through a coupling agent, without downhole power supply and electronic nodes, and the sensing optical fiber has intrinsic safety and explosion-proof characteristics, can stably work in the downhole environment of high dust, high humidity, high temperature and high pressure and disaster collapse, greatly improves the coupling efficiency of weak life signals and the transmission signal-to-noise ratio, solves the pain points of easy damage and limited coverage range of traditional wired / wireless detection equipment after disasters, and greatly expands the life detection coverage range.

Description

Technical Field

[0001] This invention belongs to the field of underground safety technology, specifically relating to a system for detecting and locating the life status of trapped personnel underground based on distributed fiber optic acoustic sensing. Background Technology

[0002] The underground trapped personnel life status detection and positioning system is a specialized emergency rescue equipment used to quickly search for, accurately locate, and assess the vital signs of trapped personnel after accidents such as collapses, water inrushes, and explosions occur in underground work sites such as mines, tunnels, and underground integrated pipe corridors.

[0003] Currently, traditional methods for underground life detection and location mainly rely on personnel locator cards, portable audio detectors, bio-radar, and emergency wired communication. These methods have significant limitations in actual disaster scenarios: First, the coverage is limited and discontinuous, mostly single-point or short-range detection, making it difficult to achieve long-distance, full-tunnel, and full-area monitoring. Second, their post-disaster survivability is poor; traditional equipment relies on base stations, cables, power supplies, and electronic nodes, which are easily damaged or have their signals interrupted after a disaster, making stable operation difficult. Third, their anti-interference capability is weak; underground dust, water accumulation, metal supports, mechanical vibrations, and strong environmental noise can easily lead to life-threatening situations. The system suffers from several drawbacks: first, it misjudges and misses vital signs, making it unable to effectively extract weak signs such as breathing and heartbeat; second, it can only determine the presence of personnel, not assess their vital status, as most devices cannot identify breathing and heartbeat rhythms, distinguish between unconscious, seriously injured, and alive states, and struggle to simultaneously locate and differentiate multiple individuals; and third, its positioning accuracy is insufficient, with traditional radar and audio equipment only providing approximate locations and failing to output precise three-dimensional coordinates, leading to a high degree of blindness and low efficiency in rescue operations. Therefore, a better detection system based on distributed fiber optic acoustic sensing for detecting and locating the vital status of trapped personnel in underground wells is proposed. Summary of the Invention

[0004] The purpose of this invention is to provide a system for detecting and locating the vital signs of trapped personnel in underground mines based on distributed fiber optic acoustic sensing, so as to solve the problems mentioned above.

[0005] To achieve the above objectives, this invention provides a system for detecting and locating the vital signs of trapped personnel in underground mines based on distributed fiber optic acoustic sensing, comprising:

[0006] The underground sensing layer includes armored single-mode sensing optical fibers. The sensing optical fibers are deployed using a dual method combining permanent pre-embedding and temporary patching. During permanent pre-embedding, they are laid along the entire length of underground roadways, working faces, refuge chambers, and key pipelines. During temporary patching, they are quickly lowered and pulled through boreholes and rescue channels. The sensing optical fibers are fixed to the roadway wall or anchor bolts and achieve tight acoustic bonding with the surrounding rock underground through a coupling agent. They do not require underground power supply or electronic nodes, and have intrinsic safety and explosion-proof characteristics. They can work stably in underground environments with high dust, high humidity, high temperature and high pressure, and catastrophic collapse, significantly improving the coupling efficiency and transmission signal-to-noise ratio of weak life signals.

[0007] The ground demodulation and acquisition layer includes a DAS demodulation host and a data preprocessing unit. The DAS demodulation host is based on the phase-sensitive optical time-domain reflectometry (φ-OTDR) principle and incorporates a narrow-linewidth laser, pulse modulator, optical coupler, photodetector, and high-speed data acquisition unit. It can inject narrow-linewidth pulsed laser into the armored single-mode sensing fiber, acquire the Rayleigh scattered light signal after the fiber is in place, and complete the photo-to-electric signal conversion, outputting continuous vibration / acoustic wave time-domain and frequency-domain data along the entire length of the fiber. The data preprocessing unit uses adaptive noise reduction and wavelet threshold denoising algorithms to filter out background environmental noise such as downhole fans, water flow, and surrounding rock vibration, achieving preliminary enhancement of life signals and anomaly screening.

[0008] The host computer analysis and application layer includes a life signal recognition module, a three-dimensional positioning module, a status assessment module, and a visualization command module. Each module works in concert to complete life signal extraction, positioning calculation, status assessment, and rescue guidance, respectively.

[0009] The downhole sensing layer, the surface demodulation and acquisition layer, and the host computer analysis and application layer are sequentially connected by signals.

[0010] In one or more embodiments of the present invention, the key parameters of the DAS demodulation host are: detection distance ≥ 20km, spatial resolution ≤ 1m, sampling rate ≥ 5kHz, phase demodulation accuracy < 1mrad, capturing full-band signals from 1Hz to 20kHz, covering all types of life signals including human breathing (0.1~0.5Hz), heartbeat (0.8~2Hz), micro-motion (1~10Hz), and voice / knocking for help (200Hz~5kHz), which is used to overcome the technical limitations of traditional detection equipment in short distance and narrow frequency band.

[0011] In one or more embodiments of the present invention, the life signal recognition module employs a fusion algorithm combining time-frequency analysis (STFT, wavelet transform), deep learning (CNN, LSTM), and pattern matching to construct a locally trained life signal recognition model. This model can accurately extract four core life signals—breathing, heartbeat, micro-movements, and voice calls for help—from a noisy environment. It can effectively distinguish human life signals from underground mechanical interference and environmental noise. Simultaneously, through multi-feature fusion and clustering analysis techniques, it solves the technical problem of signal superposition and confusion among multiple trapped individuals, achieving an accuracy rate of ≥95%. This overcomes the shortcomings of traditional systems that can only determine "whether there are people" but cannot distinguish signal types.

[0012] In one or more embodiments of the present invention, the three-dimensional positioning module combines optical fiber spatial coordinates, a three-dimensional GIS model of the tunnel, time difference of arrival (TDOA), and amplitude attenuation model to transform the one-dimensional distance positioning of the armored single-mode sensing fiber into three-dimensional spatial coordinates (X / Y / Z) underground. The positioning error is ≤5m, and it supports simultaneous positioning and trajectory tracking of multiple trapped personnel. This represents a technological breakthrough from "one-dimensional distance" to "three-dimensional precise positioning," and is used to solve the pain points of low positioning accuracy and inability to provide spatial coordinates in traditional radar and audio detectors.

[0013] In one or more embodiments of the present invention, the status assessment module constructs a life status level assessment system by analyzing the breathing rate, heart rhythm and micro-motion intensity of the trapped personnel in real time, automatically determines the three core states of the trapped personnel: survival, coma, and serious injury, and outputs real-time vital sign curves and status warning information, realizing the upgrade from "survival status" to "status assessment", which fills the technical gap that traditional systems cannot assess the life status of trapped personnel.

[0014] In one or more embodiments of the present invention, the visualization command module interfaces with the underground three-dimensional tunnel model to display the location, number, vital signs and signal strength of trapped personnel in real time. It supports real-time backtracking of vital data and abnormal sound and light alarms, and can be linked with the mine monitoring system and emergency command platform to achieve data sharing. At the same time, it automatically plans the optimal rescue route based on the positioning results to guide on-site rescue operations and improve search and rescue efficiency, thereby breaking through the limitations of fragmented and passive rescue guidance in traditional systems.

[0015] In one or more embodiments of the present invention, the armored single-mode sensing fiber adopts an armored single-mode communication fiber with a diameter of 9 / 125μm. During the laying process, a dual redundancy design is adopted, and an emergency interface is reserved. Through the standardized construction process of "trench excavation - coupling agent filling - fiber embedding - pressure compaction - loss testing", the fiber and the coal and rock mass are tightly acoustically bonded. This is used to solve the technical problems of low coupling efficiency and severe signal attenuation in traditional fiber laying, and ensure that the fiber can still transmit signals stably after a disaster. Compared with traditional wired sensors and wireless detection equipment, it has stronger resistance to damage and environmental adaptability.

[0016] In one or more embodiments of the present invention, the workflow of the system includes the following steps:

[0017] After the disaster, the ground-based DAS demodulation host is activated and connected to the pre-embedded or temporarily supplemented armored single-mode sensing optical fiber.

[0018] The system collects background noise from the well and establishes a noise model to complete baseline calibration; the DAS host continuously scans the entire length of the optical fiber, collects vibration / acoustic data, and uploads it to the host computer.

[0019] The presence of trapped personnel is determined by identifying noise and extracting vital signs using recognition algorithms.

[0020] The system combines positioning algorithms to calculate 3D coordinates and marks them on the 3D model of the tunnel; assesses the life status and triggers an early warning; plans a rescue route based on the positioning results and tracks status changes in real time; the above process realizes full automation of the "deployment-calibration-scanning-identification-positioning-assessment-rescue" process without the need for close-range manual operation, thereby avoiding rescue personnel from entering dangerous areas and improving rescue safety.

[0021] In one or more embodiments of the present invention, the DAS demodulation host adopts a combination of frequency division multiplexing sequence synchronous detection and differential defolding algorithm to solve the large strain phase folding problem and restore the large strain range; at the same time, it reconstructs high sampling rate and distortion-free signal data through time-series interleaving processing to ensure accurate capture and restoration of weak life signals at the pm level, thereby overcoming the limitations of traditional DAS technology in its ability to extract weak signals.

[0022] A method for detecting and locating the vital signs of trapped personnel in underground mines based on distributed fiber optic acoustic sensing includes the following steps:

[0023] S1. Deploy armored single-mode sensing optical fiber, using permanent pre-embedding or temporary patching methods, fix it in the target area downhole, and achieve tight coupling with the surrounding rock through a coupling agent to complete the optical fiber loss test and debugging.

[0024] S2. Start the ground DAS demodulation host and data preprocessing unit, inject narrow linewidth pulsed laser into the armored single-mode sensing fiber, collect the background noise in the well and establish a noise model to complete the noise reduction parameter calibration.

[0025] The S3 and DAS demodulation host continuously acquires backscattered Rayleigh light signals along the entire length of the optical fiber. After completing the optical-to-electric conversion, the vibration / acoustic data is transmitted to the data preprocessing unit to filter out environmental noise and enhance the signal.

[0026] S4, the host computer life signal recognition module extracts life signal features such as breathing, heartbeat, micro-movement, and voice through a fusion algorithm, distinguishes human signals from interference signals, determines the existence of trapped personnel, and distinguishes signals from multiple personnel through cluster analysis.

[0027] The S5 three-dimensional positioning module combines fiber optic coordinates, a three-dimensional model of the tunnel, and a TDOA and amplitude attenuation joint positioning algorithm to calculate the three-dimensional spatial coordinates of the trapped personnel and achieve precise positioning.

[0028] S6. The status assessment module analyzes vital signs parameters, determines the life status of trapped personnel, and outputs early warning information.

[0029] S7, the visual command module displays location results and life status, plans the optimal rescue route, links with the emergency command platform, guides on-site rescue, and tracks changes in the status of trapped personnel in real time until the rescue is completed.

[0030] Compared with the prior art, the beneficial effects of the present invention are:

[0031] (1) This life status detection and positioning system breaks through the coverage limitations of traditional detection equipment, realizes long-distance and full-area continuous detection in the well, and has strong adaptability to disaster environment. It adopts a passive armored sensing fiber optic dual deployment method, which does not require underground power supply or electronic nodes. It can work stably in disaster scenarios such as high dust, high humidity, and collapse, solving the pain points of traditional wired / wireless detection equipment being easily damaged after disaster and having limited coverage, and greatly expanding the coverage of life detection.

[0032] (2) This life status detection and positioning system can accurately identify life signals and distinguish between multiple people. Through STFT, CNN-LSTM fusion algorithm and K-means clustering analysis, it can accurately extract four core life signals, namely breathing, heartbeat, micro-movement and voice, from a strong noise environment. The recognition accuracy is ≥95%. It can effectively distinguish the superimposed signals of multiple trapped people and overcome the defects of traditional systems that can only judge "whether there are people" and cannot distinguish the signal type and the confusion of multiple people.

[0033] (3) This life status detection and positioning system achieves a breakthrough from "one-dimensional distance" to "three-dimensional precise positioning". Combining TDOA positioning, amplitude attenuation correction and three-dimensional GIS model of the tunnel, it transforms fiber optic one-dimensional positioning into three-dimensional spatial coordinates in the mine. The positioning error is ≤5m. It supports simultaneous positioning and trajectory tracking of multiple personnel, solves the problem of low positioning accuracy and inability to provide spatial coordinates of traditional radar and audio detectors, and provides accurate basis for rescue route planning.

[0034] (4) This life status detection and positioning system fills the technical gap in the traditional system for life status assessment. By quantitatively extracting vital signs parameters such as respiratory rate, heart rhythm, and micro-motion intensity, it constructs a life status level assessment system, automatically determines the three states of trapped personnel: survival, coma, and serious injury, and outputs graded warnings. It realizes the upgrade from "whether there is survival" to "quantitative assessment of status", provides a scientific basis for the allocation of rescue priorities and the formulation of plans, and significantly improves rescue efficiency and the survival rate of trapped personnel. Attached Figure Description

[0035] Figure 1 This is a flowchart illustrating the overall architecture of the present invention. Detailed Implementation

[0036] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings, but it should be understood that the scope of protection of the present invention is not limited to the specific embodiments.

[0037] like Figure 1 As shown, it illustrates a distributed fiber optic acoustic sensing-based system for detecting and locating the vital signs of trapped personnel in underground mines, according to an embodiment of the present invention, comprising:

[0038] The underground sensing layer includes armored single-mode sensing optical fibers. The sensing optical fibers adopt a dual deployment method combining permanent pre-embedding and temporary patching. During permanent pre-embedding, they are laid along the entire length of underground roadways, working faces, refuge chambers, and key pipelines. During temporary patching, they are quickly lowered and pulled through boreholes and rescue channels. The sensing optical fibers are fixed to the roadway wall or anchor bolts and achieve tight acoustic bonding with the underground surrounding rock through a coupling agent. It does not require underground power supply or electronic nodes, has intrinsic safety and explosion-proof characteristics, and can work stably in underground high dust, high humidity, high temperature and high pressure and disaster collapse environments, greatly improving the coupling efficiency and transmission signal-to-noise ratio of weak life signals.

[0039] The ground demodulation and acquisition layer includes a DAS demodulation host and a data preprocessing unit. The DAS demodulation host is based on the phase-sensitive optical time-domain reflectometry (φ-OTDR) principle and incorporates a narrow-linewidth laser, pulse modulator, optical coupler, photodetector, and high-speed data acquisition unit. It can inject narrow-linewidth pulsed laser into the armored single-mode sensing fiber, acquire the Rayleigh scattered light signal after the fiber is in place, and complete the optical-to-electrical signal conversion, outputting continuous vibration / acoustic wave time-domain and frequency-domain data along the entire length of the fiber. The data preprocessing unit uses adaptive noise reduction and wavelet threshold denoising algorithms to filter out background environmental noise such as downhole fans, water flow, and surrounding rock vibration, achieving preliminary enhancement of life signals and anomaly screening.

[0040] The host computer analysis and application layer includes a life signal recognition module, a three-dimensional positioning module, a status assessment module, and a visualization command module. These modules work together to complete life signal extraction, location calculation, status assessment, and rescue guidance, respectively.

[0041] The downhole sensing layer, the surface demodulation and acquisition layer, and the host computer analysis and application layer are sequentially connected.

[0042] The key parameters of the DAS demodulation host meet the following requirements: detection distance ≥ 20km, spatial resolution ≤ 1m, sampling rate ≥ 5kHz, phase demodulation accuracy < 1mrad, capture full-band signals from 1Hz to 20kHz, covering all types of life signals including human breathing (0.1~0.5Hz), heartbeat (0.8~2Hz), micro-motion (1~10Hz), and voice / knocking for help (200Hz~5kHz), which is used to overcome the technical limitations of traditional detection equipment in short distance and narrow frequency band.

[0043] The life signal recognition module employs a fusion algorithm combining time-frequency analysis (STFT, wavelet transform), deep learning (CNN, LSTM), and pattern matching to construct a locally trained life signal recognition model. This model can accurately extract four core life signals—breathing, heartbeat, micro-movements, and voice calls for help—from noisy environments. It can effectively distinguish human life signals from underground mechanical interference and environmental noise. Simultaneously, through multi-feature fusion and clustering analysis techniques, it solves the technical challenge of signal superposition and confusion among multiple trapped individuals, achieving an accuracy rate of ≥95%. This overcomes the shortcomings of traditional systems that can only determine "whether there are people" but cannot distinguish signal types.

[0044] The 3D positioning module combines fiber optic spatial coordinates, a 3D GIS model of the tunnel, time difference of arrival (TDOA), and amplitude attenuation model to transform the one-dimensional distance positioning of the armored single-mode sensing fiber into three-dimensional spatial coordinates (X / Y / Z) underground. The positioning error is ≤5m, and it supports simultaneous positioning and trajectory tracking of multiple trapped personnel. This represents a technological breakthrough from "one-dimensional distance" to "precise three-dimensional positioning," addressing the pain points of low positioning accuracy and inability to provide spatial coordinates in traditional radar and audio detectors.

[0045] The status assessment module constructs a life status level assessment system by analyzing the breathing rate, heart rhythm and micro-motion intensity of trapped personnel in real time. It automatically determines the three core states of trapped personnel: survival, coma, and serious injury, and outputs real-time vital sign curves and status warning information. This upgrade from "survival status" to "status assessment" fills the technical gap that traditional systems cannot assess the life status of trapped personnel.

[0046] The visualization command module interfaces with the underground 3D tunnel model to display the location, number, vital signs, and signal strength of trapped personnel in real time. It supports real-time backtracking of vital data and abnormal audible and visual alarms. It can also be linked with the mine monitoring system and emergency command platform to achieve data sharing. At the same time, it automatically plans the optimal rescue route based on the positioning results to guide on-site rescue operations and improve search and rescue efficiency, thus overcoming the limitations of traditional fragmented and passive rescue guidance.

[0047] The armored single-mode sensing fiber uses 9 / 125μm diameter armored single-mode communication fiber. It adopts a dual-redundancy design during laying and reserves an emergency interface. Through a standardized construction process of "trench excavation - couplant filling - fiber embedding - pressure compaction - loss testing", it achieves tight acoustic bonding between the fiber and the coal and rock mass. This solves the technical problems of low coupling efficiency and severe signal attenuation in traditional fiber laying, ensuring that the fiber can still transmit signals stably after a disaster. Compared with traditional wired sensors and wireless detection equipment, it has stronger resistance to damage and environmental adaptability.

[0048] The system's workflow includes the following steps:

[0049] After the disaster, the ground-based DAS demodulation host is activated and connected to the pre-embedded or temporarily supplemented armored single-mode sensing optical fiber.

[0050] The system collects background noise from the well and establishes a noise model to complete baseline calibration; the DAS host continuously scans the entire length of the optical fiber, collects vibration / acoustic data, and uploads it to the host computer.

[0051] The presence of trapped personnel is determined by identifying noise and extracting vital signs using recognition algorithms.

[0052] The system combines positioning algorithms to calculate 3D coordinates and marks them on the 3D model of the tunnel; assesses the life status and triggers an early warning; plans a rescue route based on the positioning results and tracks status changes in real time; the above process realizes full automation of the "deployment-calibration-scanning-identification-positioning-assessment-rescue" process without the need for close-range manual operation, thereby avoiding rescue personnel from entering dangerous areas and improving rescue safety.

[0053] The DAS demodulation host uses a combination of frequency division multiplexing sequence synchronous detection and differential defolding algorithm to solve the problem of phase folding in large strain and restore large strain range. At the same time, it reconstructs high sampling rate and distortion-free signal data through time-series interleaving processing to ensure accurate capture and restoration of weak life signals at the pm level, thus overcoming the limitations of traditional DAS technology in extracting weak signals.

[0054] A method for detecting and locating the vital signs of trapped personnel in underground mines based on distributed fiber optic acoustic sensing includes the following steps:

[0055] S1. Deploy armored single-mode sensing optical fiber, using permanent pre-embedding or temporary patching methods, fix it in the target area downhole, and achieve tight coupling with the surrounding rock through a coupling agent to complete the optical fiber loss test and debugging.

[0056] S2. Start the ground DAS demodulation host and data preprocessing unit, inject narrow linewidth pulsed laser into the armored single-mode sensing fiber, collect the background noise in the well and establish a noise model to complete the noise reduction parameter calibration.

[0057] The S3 and DAS demodulation host continuously acquires backscattered Rayleigh light signals along the entire length of the optical fiber. After completing the optical-to-electric conversion, the vibration / acoustic data is transmitted to the data preprocessing unit to filter out environmental noise and enhance the signal.

[0058] S4, the host computer life signal recognition module extracts life signal features such as breathing, heartbeat, micro-movement, and voice through a fusion algorithm, distinguishes human signals from interference signals, determines the existence of trapped personnel, and distinguishes signals from multiple personnel through cluster analysis.

[0059] The S5 three-dimensional positioning module combines fiber optic coordinates, a three-dimensional model of the tunnel, and a TDOA and amplitude attenuation joint positioning algorithm to calculate the three-dimensional spatial coordinates of the trapped personnel and achieve precise positioning.

[0060] S6. The status assessment module analyzes vital signs parameters, determines the life status of trapped personnel, and outputs early warning information.

[0061] S7, the visual command module displays location results and life status, plans the optimal rescue route, links with the emergency command platform, guides on-site rescue, and tracks changes in the status of trapped personnel in real time until the rescue is completed.

[0062] In this embodiment, the status assessment module extracts and analyzes three core vital sign parameters of the trapped personnel in real time: respiratory rate, heart rhythm, and micromotion intensity. It constructs a life status level assessment system, automatically outputs real-time vital sign curves, and triggers status warning information at different levels. This upgrades the system from "whether the person is alive or not" to "quantitative assessment of life status," filling the technical gap that traditional detection systems cannot assess the physiological status of trapped personnel.

[0063] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0064] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A system for detecting and locating the vital signs of trapped personnel in underground mines based on distributed fiber optic acoustic sensing, characterized in that, include: The underground sensing layer includes armored single-mode sensing optical fiber. The sensing optical fiber adopts a dual deployment method combining permanent pre-embedding and temporary patching. The sensing optical fiber is fixed to the roadway wall or anchor bolt and achieves tight acoustic bonding with the underground surrounding rock through a coupling agent. The ground demodulation and acquisition layer includes a DAS demodulation host and a data preprocessing unit. The DAS demodulation host is based on the phase-sensitive optical time-domain reflectometry (φ-OTDR) principle and incorporates a narrow-linewidth laser, pulse modulator, optical coupler, photodetector, and high-speed data acquisition unit. The data preprocessing unit employs adaptive noise reduction and wavelet threshold denoising algorithms to filter out background environmental noise such as downhole fans, water flow, and surrounding rock vibration, achieving preliminary enhancement and anomaly screening of vital signals. The host computer analysis and application layer includes a life signal recognition module, a three-dimensional positioning module, a status assessment module, and a visualization command module. Each module works in concert to complete life signal extraction, positioning calculation, status assessment, and rescue guidance, respectively. The downhole sensing layer, the surface demodulation and acquisition layer, and the host computer analysis and application layer are sequentially connected by signals.

2. The underground personnel trapped life status detection and positioning system based on distributed fiber optic acoustic sensing according to claim 1, characterized in that, The key parameters of the DAS demodulation host meet the following requirements: detection distance ≥ 20km, spatial resolution ≤ 1m, sampling rate ≥ 5kHz, phase demodulation accuracy < 1mrad, capture full-band signals from 1Hz to 20kHz, covering all types of life signals including human breathing (0.1~0.5Hz), heartbeat (0.8~2Hz), micro-motion (1~10Hz), and voice / knocking for help (200Hz~5kHz).

3. The underground personnel trapped life status detection and positioning system based on distributed fiber optic acoustic sensing according to claim 2, characterized in that, The vital signal recognition module employs a fusion algorithm combining time-frequency analysis (STFT, wavelet transform), deep learning (CNN, LSTM), and pattern matching to construct a locally trained vital signal recognition model. Simultaneously, it utilizes multi-feature fusion and clustering analysis techniques to overcome the shortcomings of traditional systems that can only determine "whether there are people" and cannot distinguish signal types.

4. The underground personnel trapped life status detection and positioning system based on distributed fiber optic acoustic sensing according to claim 3, characterized in that, The three-dimensional positioning module combines optical fiber spatial coordinates, a three-dimensional GIS model of the tunnel, time difference of arrival (TDOA), and amplitude attenuation model to convert the one-dimensional distance positioning of the armored single-mode sensing fiber into three-dimensional spatial coordinates (X / Y / Z) in the mine.

5. The underground personnel trapped life status detection and positioning system based on distributed fiber optic acoustic sensing according to claim 4, characterized in that, The status assessment module constructs a life status level assessment system by analyzing the breathing rate, heart rhythm and micro-motion intensity of the trapped personnel in real time.

6. The underground personnel trapped life status detection and positioning system based on distributed fiber optic acoustic sensing according to claim 5, characterized in that, The visualization command module interfaces with the underground 3D tunnel model to display the location, number, vital signs, and signal strength of trapped personnel in real time. It supports real-time backtracking of vital data and abnormal audible and visual alarms. It can also link with the mine monitoring system and emergency command platform to achieve data sharing. At the same time, it automatically plans the optimal rescue route based on the positioning results to guide on-site rescue operations and improve search and rescue efficiency.

7. The underground personnel trapped life status detection and positioning system based on distributed fiber optic acoustic sensing according to claim 6, characterized in that, The armored single-mode sensing fiber uses a 9 / 125μm diameter armored single-mode communication fiber, and adopts a dual-redundancy design during laying with a reserved emergency interface. Through a standardized construction process of "trench excavation - coupling agent filling - fiber embedding - pressure compaction - loss testing", the fiber and the coal and rock mass are tightly acoustically bonded.

8. The underground personnel trapped life status detection and positioning system based on distributed fiber optic acoustic sensing according to claim 7, characterized in that, The system's workflow includes the following steps: After the disaster, the ground-based DAS demodulation host is activated and connected to the pre-embedded or temporarily supplemented armored single-mode sensing fiber optic cable. The system collects background noise from the well and establishes a noise model to complete baseline calibration; the DAS host continuously scans the entire length of the optical fiber, collects vibration / acoustic data, and uploads it to the host computer. The presence of trapped personnel is determined by identifying noise and extracting vital signs using recognition algorithms. The system combines positioning algorithms to calculate three-dimensional coordinates, which are then marked on the three-dimensional model of the tunnel; assesses the vital status and triggers early warnings; plans rescue routes based on the positioning results, and tracks status changes in real time.

9. The underground personnel trapped life status detection and positioning system based on distributed fiber optic acoustic sensing according to claim 8, characterized in that, The DAS demodulation host uses a combination of frequency division multiplexing sequence synchronous detection and differential defolding algorithm, and reconstructs high sampling rate, distortion-free signal data through time-series interleaving processing to ensure accurate capture and restoration of weak life signals at the pm level.

10. A method for detecting and locating the vital signs of trapped personnel in underground mines based on distributed fiber optic acoustic sensing, characterized in that, Includes the following steps: S1. Deploy armored single-mode sensing optical fiber, using permanent pre-embedding or temporary patching, and fix it in the target area downhole. Achieve tight coupling with the surrounding rock through a coupling agent, and complete the optical fiber loss test and debugging. S2. Start the ground DAS demodulation host and data preprocessing unit, inject narrow linewidth pulsed laser into the armored single-mode sensing fiber, collect the background noise in the well and establish a noise model to complete the noise reduction parameter calibration. The S3 and DAS demodulation host continuously acquires backscattered Rayleigh light signals along the entire length of the optical fiber. After completing the optical-to-electric conversion, the vibration / acoustic data is transmitted to the data preprocessing unit to filter out environmental noise and enhance the signal. S4, the host computer life signal recognition module extracts life signal features such as breathing, heartbeat, micro-movement, and voice through a fusion algorithm, distinguishes human signals from interference signals, determines the existence of trapped personnel, and distinguishes signals from multiple personnel through cluster analysis. The S5 three-dimensional positioning module combines fiber optic coordinates, a three-dimensional model of the tunnel, and a TDOA and amplitude attenuation joint positioning algorithm to calculate the three-dimensional spatial coordinates of the trapped personnel and achieve precise positioning. S6. The status assessment module analyzes vital signs parameters, determines the life status of trapped personnel, and outputs early warning information. S7, the visual command module displays location results and life status, plans the optimal rescue route, links with the emergency command platform, guides on-site rescue, and tracks changes in the status of trapped personnel in real time until the rescue is completed.

Images