Mine dust concentration fiber sensor on-line monitoring system and method

Abstract

This invention discloses an online monitoring system and method for mine dust concentration using fiber optic sensing. It includes a sensing fiber with a dust-repellent coating, an optical signal modulation and acquisition unit, a signal processing unit, and a self-cleaning unit. Pulsed light is injected into the fiber using an OTDR to collect backscattered signals, which are then processed and demodulated to simultaneously determine the dust concentration and light attenuation rate along the path. The core of the system lies in using a joint judgment logic to control cleaning: the self-cleaning unit is only triggered to perform targeted cleaning of a specific section when both the light signal attenuation rate and the retrieved dust concentration value continuously exceed a set threshold. The cleaning cycle can be dynamically adjusted based on the historical dust concentration in the roadway. Integrating temperature compensation, two-dimensional distribution inversion, and dynamic adjustment of the cleaning cycle, it achieves accurate and reliable online monitoring and intelligent fixed-point maintenance of mine dust, effectively avoiding the ineffective consumption of cleaning resources while ensuring monitoring accuracy.

Description

Technical Field

[0001] This invention belongs to the field of mine safety monitoring technology, and in particular relates to an online monitoring system and method for mine dust concentration using fiber optic sensing. Background Technology

[0002] Mine dust is a major hazard in coal mine production, endangering workers' health and posing an explosion risk. Currently, sensors are commonly used to measure dust concentration, but these methods suffer from limitations such as being limited to fixed measurement points, high intrinsic safety requirements for circuitry, and probes being easily covered by dust, leading to decreased sensitivity and high maintenance costs. Distributed fiber optic sensing technology (such as OTDR) holds great promise for mine roadway monitoring due to its inherent safety, interference resistance, and long-distance continuous monitoring capabilities. However, directly applying it to high-concentration dust environments faces severe long-term reliability challenges: the exposed surface of the sensing fiber readily and continuously adsorbs dust particles. This not only alters the light scattering characteristics of the fiber surface, distorting the concentration inversion model, but also causes additional non-targeted attenuation of the optical signal, resulting in measurement drift or even remote signal failure.

[0003] Current technologies lack a distributed fiber optic dust concentration monitoring method that can adapt to the high-concentration dust environment of mines and maintain stable and reliable monitoring performance over a long period without frequent manual intervention. How to fundamentally suppress dust adhesion on the surface of the sensing fiber and automatically detect and recover from performance degradation caused by contamination during monitoring is a key challenge that must be overcome to move this technology from theory to practical engineering application. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides an online monitoring system and method for mine dust concentration using fiber optic sensing, aiming to solve the long-term reliability problem of sensing fiber optics in high-dust environments and achieve accurate monitoring across the entire cross-section, over long distances, and with intelligence.

[0005] The embodiments of the present invention disclose the following technical solutions.

[0006] The online monitoring system for mine dust concentration using fiber optic sensors includes:

[0007] Sensing fiber optic cable, optical signal modulation and acquisition unit, signal processing unit, and self-cleaning unit;

[0008] The sensing optical fiber is laid along the area to be monitored, and its surface is provided with a dust-repellent functional layer.

[0009] The optical signal modulation and acquisition unit is used to inject pulsed optical signals into the sensing optical fiber and receive backscattered light generated by the sensing optical fiber.

[0010] The signal processing unit is used to process the optical signal returned by the sensing fiber to demodulate the attenuation rate of the backscattered optical signal and the real-time inversion of the dust concentration value.

[0011] The self-cleaning unit is used to trigger directional cleaning of a specific section of the sensing fiber only when the attenuation rate continuously exceeds a first threshold and the real-time inversion dust concentration value at the corresponding location continuously exceeds a second threshold.

[0012] Furthermore, the signal processing unit is also used to fuse the demodulated multi-location dust concentration data with the tunnel spatial model to generate a two-dimensional dust concentration distribution heat map of the tunnel cross section.

[0013] Furthermore, the signal processing unit of the present invention also integrates a trained neural network model for analyzing the time-domain sequence of the backscattered signal and outputting a state classification of the sensing optical fiber, wherein the state classification includes at least normal, dust contamination, and mechanical damage.

[0014] Furthermore, the system also includes an alarm unit that outputs an alarm for abnormal fiber contamination or functional layer failure when the self-cleaning unit continuously triggers cleaning of the same section within a predetermined number of times, or when the signal attenuation rate does not recover to the normal range after cleaning.

[0015] Furthermore, the system also includes a data interface unit for uploading concentration values, decay rate indicators, cleaning logs, and heat map data to the mine integrated monitoring platform, and can be configured to automatically activate ventilation or spray dust suppression equipment when the concentration continues to exceed the limit.

[0016] Furthermore, the optical signal modulation and acquisition unit includes an optical time domain reflectometer (OTDR), and the cleaning trigger position of the self-cleaning unit is precisely located based on the time domain analysis results of the OTDR.

[0017] Furthermore, the self-cleaning unit of the present invention includes a compressed air source, a pipeline, a solenoid valve, and a directional nozzle; the nozzle is configured to be installed at a key position along the sensing optical fiber for blowing onto a specific section of the sensing optical fiber.

[0018] Furthermore, in this invention, the term "abnormal attenuation rate" refers to the following: the attenuation rate of the backscattered light signal monitored by the signal processing unit continuously exceeds a first set threshold; the term "confirmation based on real-time inverted dust concentration" refers to the following: when the real-time inverted dust concentration value continuously exceeds a second set threshold, a cleaning command is confirmed to be triggered.

[0019] Furthermore, the first threshold and the second threshold can be set differently according to the different regions of the location (tunneling roadway, coal mining face, return airway, transport roadway).

[0020] Furthermore, the cleaning trigger cycle of the self-cleaning unit is dynamically adjusted based on historical dust concentration data of the monitored area. The dynamically adjusted cleaning cycle... Based on the average dust concentration in the tunnel Adjustments are made, and the adjustment formula is as follows: ,in , These are constants calibrated based on the on-site environment.

[0021] Furthermore, the dust-repellent functional layer of this invention is a composite coating comprising a SiO2 matrix and low surface energy nanoparticles; preferably, the SiO2 matrix is ​​SiO2 aerogel, and the low surface energy nanoparticles are polytetrafluoroethylene nanoparticles. Even further, the dry weight ratio of the SiO2 aerogel to the polytetrafluoroethylene nanoparticles is 88:12; the thickness of the dust-repellent functional layer is 5~12 μm, and its refractive index is lower than that of the cladding refractive index of the sensing optical fiber.

[0022] Furthermore, the system also includes a temperature compensation unit, which is used to perform real-time temperature compensation on the Mie scattering intensity obtained by the optical signal modulation and acquisition unit by acquiring and calculating the Raman scattering signal in the sensing optical fiber.

[0023] This invention also provides a method for online monitoring of mine dust concentration using fiber optic sensing of the above-mentioned monitoring system, comprising the following steps:

[0024] a. Injecting light pulses into the sensing optical fiber coated with a dust-repellent functional layer;

[0025] b. Collect and analyze the backscattered light signal to obtain the dust concentration value at the monitoring point and the signal attenuation rate of the sensing fiber itself;

[0026] c. Real-time acquisition of the attenuation rate of the backscattered light signal of the sensing fiber and the real-time inversion dust concentration value at that location;

[0027] d. Determine whether the decay rate continuously exceeds a first threshold and whether the concentration value continuously exceeds a second threshold;

[0028] e. When the conditions in step d are met simultaneously, a cleaning command is generated to control the cleaning actuator at the corresponding position to perform directional cleaning on a specific section of the sensing optical fiber;

[0029] f. Dynamically adjust the cleaning cycle of the cleaning system based on the average dust concentration in the tunnel.

[0030] The beneficial effects of this invention are:

[0031] This invention provides an online monitoring system and method for fiber optic dust concentration in mines. By coating the surface of the sensing fiber with a SiO2 / PTFE composite dust-repellent coating, the probability of dust adhesion is reduced from both physical and chemical perspectives. Simultaneously, combined with a self-cleaning unit, it solves the problems of severe signal drift and frequent maintenance required by traditional fiber optic sensors in harsh dusty environments. The invention creatively introduces "backscattered light signal attenuation rate" as a self-diagnostic indicator, decoupling "fiber contamination" from "changes in environmental dust concentration" at the signal level. Combined with a neural network model, it effectively distinguishes between dust contamination and mechanical damage, reducing the system's false alarm rate. A dual-threshold collaborative judgment logic of "self-diagnostic indicator (attenuation rate) + environmental concentration" is established, triggering cleaning only when the fiber is indeed contaminated and the environment is harsh, avoiding energy waste or untimely cleaning caused by fixed-period cleaning. Furthermore, the cleaning cycle can be dynamically adjusted based on historical dust load, significantly reducing compressed air consumption and operation and maintenance costs. By synchronously calculating the Raman scattering signal and performing real-time temperature compensation, the influence of underground temperature fluctuations on the Mie scattering intensity is eliminated; OTDR technology can be used to perform full-section, continuous monitoring of long-distance roadways without monitoring blind spots, making it particularly suitable for complex environments such as mines that are flammable, explosive, and have high dust levels. Attached Figure Description

[0032] Figure 1 This is a system flowchart of the online monitoring system and method for mine dust concentration using fiber optic sensing proposed in this invention.

[0033] Figure 2 This is a logic diagram of the dual threshold judgment of the online monitoring system and method for mine dust concentration using fiber optic sensing proposed in this invention. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the following description, in conjunction with the accompanying drawings and embodiments, will further clarify them.

[0035] The present invention will be described in further detail. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0036] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0037] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature; in the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0038] Example 1

[0039] This embodiment provides an online monitoring system for coal mine dust concentration using fiber optic sensing. The system mainly consists of a fiber optic sensing unit, an optical signal modulation and acquisition unit, a signal processing and control unit, and a self-cleaning execution unit.

[0040] Specifically, the fiber optic sensing unit uses 62.5 / 125μm standard communication-grade multimode fiber as the sensing medium. To maximize the sensing range, the multimode fiber is laid in a zigzag pattern along the roof of the mine roadway, and simultaneously arranged in a zigzag pattern along both sides of the mine roadway at a height of 1.5 to 2 meters above the ground. Furthermore, a straight fiber is laid near the floor on both sides of the mine roadway. All fibers are fixed with dedicated clamps to cover the cross-section of the roadway.

[0041] Furthermore, to address the dust adhesion problem, the outer surface of the sensing optical fiber is coated with a dust-repellent functional layer. The preparation method of this dust-repellent functional layer is as follows: First, SiO2 aerogel is prepared using tetraethyl orthosilicate (TEOS) as a precursor via a sol-gel method. Then, the SiO2 aerogel is composited with polytetrafluoroethylene (PTFE) nanoparticles with an average particle size of 200 nm at a dry weight ratio of 88:12 and dispersed in an ethanol solvent to form a suspension. Finally, the multimode optical fiber is impregnated and pulled through the suspension at a constant speed of 10 cm / min and cured at 120°C for 30 minutes. The final composite coating has a thickness of 8 ± 1 μm, and its refractive index is lower than that of the optical fiber cladding. Utilizing the porous structure of the aerogel and the low surface energy of PTFE, dust accumulation is effectively prevented.

[0042] Specifically, the optical signal modulation and acquisition unit is integrated into a standard industrial chassis, and its core components include a pulsed laser emitter and an optical time-domain reflectometer (OTDR). The laser emitter uses a DFB laser with a center wavelength of 1550nm, which, after modulation by the OTDR, generates a probe light pulse with a pulse width of 10ns and a repetition frequency of 1kHz. This light pulse is injected into the sensing fiber through a 2x1 fiber coupler, and the resulting backscattered light returns along the original path and is received by the OTDR.

[0043] Specifically, the signal processing and control unit not only processes sensor signals but also integrates an intelligent diagnostic module, an alarm unit, and a data interface unit. The data interface unit (such as an Ethernet port or a 4 / 5G module) is used to upload the processed dust concentration distribution heatmap, self-diagnostic indicators, cleaning event logs, and system status information to the mine's integrated monitoring platform in real time. The alarm unit is configured to automatically output audible and visual alarms and remote alerts for "abnormal fiber optic contamination" or "dust removal functional layer failure" when the self-cleaning execution unit triggers cleaning of the same fiber optic segment more than three times consecutively within one hour, or when the signal attenuation rate of that segment fails to recover to the baseline level (e.g., below 5%) within two minutes after cleaning.

[0044] Specifically, the self-cleaning actuator includes a compressed air source, a piping network, a solenoid valve, and a directional nozzle. The nozzle is pre-installed upstream of a key location along the sensing optical fiber, with the nozzle orifice aligned with a specific section of the fiber. The system controls the opening and closing of the solenoid valve via a PLC, using compressed air at a pressure of 0.5 MPa to directionally blow clean the contaminated optical fiber.

[0045] Furthermore, the system can receive instructions from the monitoring platform through the data interface unit or, according to the built-in logic, automatically send a linkage start command to the ventilation fan or spray dust suppression equipment in the area when the dust concentration in a certain area exceeds the safety threshold (e.g., 50 mg / m³) for 3 consecutive minutes, and record the concentration and attenuation rate data before and after linkage for dust suppression effect analysis.

[0046] Example 2

[0047] This embodiment, based on the hardware of Embodiment 1, details the operational logic of the signal processing unit. The signal processing and control unit is deployed on an industrial-grade edge computing gateway.

[0048] Specifically, the monitoring process is as follows:

[0049] Signal preprocessing: The gateway receives the digital signal acquired by the OTDR and performs wavelet transform processing on it. Preferably, a db4 wavelet basis is used, decomposed into 5 levels to filter out high-frequency noise and baseline drift, and extract the clean scattering signal curve.

[0050] Concentration inversion model: The Mie scattering component is separated from the preprocessed signal, and its intensity is mapped to the dust concentration as shown in the following formula: Where is the background noise intensity, is the intrinsic attenuation constant of the optical fiber, is the length of the optical fiber, and is the Mie scattering coefficient calibrated experimentally.

[0051] Temperature Compensation: Further, to eliminate the influence of downhole temperature, the system simultaneously acquires Raman scattering signals to calculate the temperature distribution. The Mie scattering intensity is corrected using the following formula to obtain the normalized intensity: Where is the temperature coefficient, with a value of -0.0023 / ℃, and is the calibration temperature of 25℃.

[0052] Furthermore, to distinguish the specific causes of signal anomalies (such as dust contamination and mechanical damage), the edge computing gateway not only runs the aforementioned inversion algorithm but also deploys a lightweight convolutional neural network model 24. The input of this model is a 1024-point backscattered signal time-domain sequence; the hidden layer includes a one-dimensional convolutional layer (16 kernel size, 4 stride) for extracting local features, and an LSTM layer (64 units) for learning time-series dependencies; the output layer is a Softmax layer, which outputs the probabilities of three categories: "normal", "dust contamination", and "fiber optic mechanical damage".

[0053] Furthermore, the edge computing gateway integrates the calculated dust concentration data along the tunnel with a pre-established three-dimensional spatial model of the tunnel, generating a two-dimensional dust concentration distribution heat map of the tunnel cross-section using an interpolation algorithm, and displaying it visually. The analysis results of the neural network model (such as "mechanical damage") will be overlaid on the heat map as an independent layer to achieve precise location of the fault point.

[0054] Example 3

[0055] This embodiment focuses on illustrating the system's self-maintenance logic, which is implemented through collaboration between the PLC and the edge computing gateway.

[0056] Specifically, the cleaning trigger uses a "dual threshold collaborative judgment" logic:

[0057] The system calculates the backscattered signal attenuation rate of each segment of the sensing fiber in real time as a self-diagnostic indicator. The area is designated as a high-risk section and cleaning is triggered only when the following two conditions are met:

[0058] The signal attenuation rate of a certain optical fiber exceeds the first threshold (set to 15% in this embodiment) for 5 consecutive minutes.

[0059] The average real-time dust concentration in the area where this fiber optic cable is located is simultaneously greater than the second threshold (set to 30 mg / m³ in this embodiment) for 5 consecutive minutes.

[0060] Once triggered, the PLC controls the upstream solenoid valve at the corresponding position to open, and the jetting continues for 10 seconds.

[0061] Furthermore, to save energy and reduce consumption, the cleaning cycle (in hours) of the cleaning system is not fixed, but dynamically adjusted based on the average dust concentration (mg / m³) in the tunnel. The adjustment formula is as follows:

[0062]

[0063] In this embodiment, constant A is set to 24, and constant B is set to -0.05. For example, when the average dust concentration in the tunnel is 10 mg / m³, the system calculates a cleaning cycle of approximately 14.5 hours; when it rises to 20 mg / m³, the cleaning cycle automatically shortens to 8.8 hours. This dynamic adjustment mechanism, while ensuring the cleanliness of the optical fiber, can save approximately 40% of compressed air consumption compared to a fixed-cycle cleaning strategy.

[0064] Furthermore, the self-maintenance logic includes an effect verification step. After each cleaning cycle and purging (10 seconds), the system pauses for 30 seconds and then re-monitors the signal attenuation rate of that section. If the attenuation rate does not drop to the preset normal range (e.g., below 8%) within the next 2 minutes, the system determines that the cleaning effect is poor, immediately records a "cleaning failure" event, and can trigger another cleaning cycle according to a preset strategy (e.g., at 1-minute intervals). If the attenuation rate still does not meet the standard after two consecutive cleaning cycles, the alarm unit is escalated to output a maintenance request.

[0065] Furthermore, the first threshold (15%) and the second threshold (30 mg / m³) are default values. The system administrator can make differentiated settings on the management interface according to different areas of the roadway (excavation roadway, coal mining face, return air roadway, and transport roadway) to achieve precise control.

Claims

1. A fiber optic sensing online monitoring system for mine dust concentration, characterized in that, include: The sensing optical fiber has a dust-repellent layer on its surface; The optical signal modulation and acquisition unit is used to inject pulsed optical signals into the sensing optical fiber and receive backscattered light. The signal processing unit is used to process the optical signal returned by the sensing fiber to demodulate the attenuation rate of the backscattered light signal and the real-time inverted dust concentration value. The self-cleaning unit is used to trigger directional cleaning of a specific segment of the sensing fiber only when the attenuation rate continuously exceeds a first threshold and the real-time inversion dust concentration value at the corresponding location continuously exceeds a second threshold.

2. The system according to claim 1, characterized in that, The optical signal modulation and acquisition unit includes an optical time domain reflectometer (OTDR), and the cleaning trigger position of the self-cleaning unit is located based on the time domain analysis results of the backscattered light by the OTDR.

3. The system according to claim 1, characterized in that, The signal processing unit is also used to solve the Raman scattering signal in the backscattered light to obtain the temperature distribution along the sensing fiber and to perform temperature compensation on the real-time inverted dust concentration value.

4. The system according to claim 1, characterized in that, The self-cleaning unit includes a compressed air source, piping, a solenoid valve, and a directional nozzle; the nozzle is configured to align with a specific segment of the sensing optical fiber.

5. The system according to claim 1, characterized in that, The term "abnormal attenuation rate" refers to the attenuation rate continuously exceeding a first set threshold.

6. The system according to claim 1, characterized in that, The confirmation based on real-time inversion dust concentration means that when the real-time inversion dust concentration value continuously exceeds the second set threshold, cleaning is confirmed to be triggered.

7. The system according to claim 1, characterized in that, The first threshold and the second threshold can be set differently according to the different regions of the location (tunneling roadway, coal mining face, return airway, transport roadway).

8. The system according to claim 1, characterized in that, The cleaning trigger cycle of the self-cleaning unit is dynamically adjusted based on historical dust concentration data of the monitored area.

9. The system according to claim 1, characterized in that, The system is configured such that during the directional cleaning performed by the self-cleaning unit, the optical signal modulation and acquisition unit and the signal processing unit continue to operate.

10. The apparatus according to claim 8, characterized in that, The dynamically adjusted cleaning cycle T is based on the average dust concentration C in the roadway. avg Dynamic adjustment, the adjustment formula is: , where A and B are constants calibrated based on the on-site environment.

11. The system according to claim 1, characterized in that, The signal processing unit is also used to fuse the demodulated multi-location dust concentration data with the tunnel spatial model to generate a two-dimensional dust concentration distribution heat map of the tunnel cross section.

12. The system according to claim 1, characterized in that, It also includes a temperature compensation unit, which is used to perform real-time temperature compensation of the Mie scattering intensity by calculating the Raman scattering signal in the sensing fiber.

13. The system according to claim 1, characterized in that, The system also includes an alarm unit. When the self-cleaning unit continuously triggers cleaning of the same section within a predetermined number of times, or when the attenuation rate does not recover to the normal range after cleaning, the alarm unit outputs an alarm for abnormal fiber contamination or functional layer failure.

14. The system according to claim 1 or 10, characterized in that, The system is configured to: after the self-cleaning unit performs targeted cleaning, re-monitor the attenuation rate of the section; if the attenuation rate does not drop to the expected level within a predetermined time, then trigger cleaning again or generate a maintenance alarm.

15. The system according to claim 1, characterized in that, The system also includes a data interface unit for uploading the real-time inverted dust concentration value, the attenuation rate self-diagnostic index, the cleaning event log, and the two-dimensional dust concentration distribution heat map to the mine integrated monitoring platform.

16. The system according to claim 1 or 15, characterized in that, The system is configured to automatically activate mine ventilation or spray dust suppression equipment when the real-time inverted dust concentration value continuously exceeds the safety threshold, and record the concentration and attenuation rate data before and after the activation event for effect analysis.

17. The system according to claim 1, characterized in that, The refractive index of the dust-repellent functional layer is lower than that of the refractive index of the sensing fiber cladding.

18. The system according to claim 1, characterized in that, The dust-repellent functional layer is a composite coating containing a SiO2 matrix and low surface energy nanoparticles.

19. The system according to claim 18, characterized in that, The SiO2 matrix is ​​SiO2 aerogel, and the low surface energy nanoparticles are polytetrafluoroethylene nanoparticles.

20. The system according to claim 19, characterized in that, The dry weight ratio of the SiO2 aerogel to the polytetrafluoroethylene nanoparticles is 88:

12.

21. The system according to claim 20, characterized in that, The thickness of the dust-repellent functional layer is 5~12μm, and its refractive index is lower than that of the cladding refractive index of the sensing optical fiber.

22. The system according to claim 21, characterized in that, The sensing fiber is a standard communication-grade multimode fiber.

23. The system according to claim 22, characterized in that, The sensing optical fiber is laid in a non-linear path on the roof and / or sidewalls of the tunnel to cover the tunnel cross section.

24. The system according to claim 23, characterized in that, The sensing optical fiber is laid in a zigzag pattern on the roof and sides of the tunnel.

25. A method for preparing a dust-repellent coating for the apparatus of claim 1, characterized in that, include: Provides a composite suspension of SiO2 aerogel and polytetrafluoroethylene nanoparticles; The optical fiber is vertically immersed and passed through the suspension at a constant speed to form a liquid film on the surface; The optical fiber with the liquid film is heat-treated to form a solidified composite coating.

26. The method according to claim 25, characterized in that, The constant speed is 8~12 cm / min, the heat treatment temperature is 110~130°C, and the time is 25~35 minutes.

27. A method for diagnosing fiber optic sensing signals in the system of claim 1, characterized in that, include: Obtain the time-domain sequence of the backscattered signal; The time-domain sequence is analyzed using a trained neural network model; The state classification of the sensing fiber is output, and the state classification includes at least normal, dust contamination, and mechanical damage.

28. A method for online monitoring of mine dust concentration using fiber optic sensing, employing the system described in claim 1, characterized in that, include: Inject light pulses into a sensing optical fiber coated with a dust-repellent functional layer; The backscattered light signal was collected and analyzed to obtain the dust concentration value and signal attenuation rate. Based on the dust concentration value and signal attenuation rate, the cleaning system is controlled to operate according to the joint judgment logic. The cleaning cycle of the cleaning system is dynamically adjusted according to the average dust concentration in the roadway.

29. An intelligent cleaning control method for fiber optic sensing devices in mines, characterized in that, include: Real-time acquisition of the attenuation rate of the backscattered light signal of the sensing fiber and the real-time inversion dust concentration value at that location; Determine whether the decay rate continuously exceeds a first threshold and whether the concentration value continuously exceeds a second threshold; When all of the above conditions are met, a cleaning instruction is generated; According to the cleaning command, the cleaning actuator at the corresponding position is controlled to perform directional cleaning on a specific section of the sensing optical fiber.

30. The method according to claim 29, characterized in that, The cleaning process involves blowing compressed air.

31. The apparatus according to claim 30, characterized in that, The directional cleaning is performed by blowing compressed air at a pressure of 0.5 to 0.8 MPa, with each blowing session lasting 5 to 15 seconds.

Images