Coal seam water inflow dynamic monitoring method and system based on intelligent sensing system

Abstract

The invention discloses a coal seam water inflow dynamic monitoring method and system based on an intelligent sensing system, and the method comprises the steps: collecting real-time pressure wave signal data, and carrying out the preprocessing of the real-time pressure wave signal data; calculating a time difference sequence among different sensors, and obtaining an initial three-dimensional coordinate of the water inflow source; calculating a statistical index of the deviation characteristic to obtain a deviation index value, and if the deviation index value exceeds a preset deviation threshold value, dynamically correcting the wave velocity to obtain a wave velocity distribution diagram; the three-dimensional coordinates of the water inflow source are obtained again; determining a potential pressure wave refraction path set; simulating a multi-path propagation scene to obtain optimized three-dimensional coordinates of the water source; updating the particle state; boundary coordinates of the water burst influence range are obtained, if the boundary coordinates exceed a preset boundary threshold value, a real-time alarm signal is generated, the alarm level is determined, and corresponding emergency measures are taken. According to the method, the water gushing source positioning precision and disaster response efficiency are remarkably improved, and coal seam roadway safety is effectively guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent sensor, and particularly relates to a coal seam water inrush dynamic monitoring method and system based on an intelligent sensing system. BACKGROUND

[0002] Coal seam water inrush refers to the phenomenon that underground water or surface water flows into a coal seam roadway or working face through a channel such as a rock fissure, a fault, a goaf, etc. during coal mining. Coal seam water inrush monitoring is a core link of coal mine safety production, and is directly related to the stable operation of a mine and personnel safety. Water inrush as a sudden geological disaster may cause roadway water accumulation, equipment damage and even casualties, so dynamic monitoring and accurate positioning thereof become the key to guaranteeing efficient mining of a coal mine.

[0003] However, current monitoring technologies mostly rely on single physical quantity detection such as water pressure or flow, and are difficult to comprehensively capture the complex dynamic characteristics of water inrush, and perform poorly in a complex geological environment. This makes it difficult for a coal mine to quickly and accurately determine the source and influence range of sudden water inrush, increasing the difficulty of emergency disposal. Existing water inrush monitoring methods can collect water quantity or pressure data through sensors, but these methods usually only focus on numerical changes, and lack accurate positioning of the specific location of a water inrush source. Especially in a coal seam, a non-homogeneous medium, the propagation of pressure waves caused by water inrush is affected by factors such as geological structure and fissure distribution, making it difficult to accurately determine the wave speed. This makes the positioning technology based on pressure waves prone to large errors in actual application, and difficult to meet the demand for high-precision positioning in a complex roadway environment, and slow in response. SUMMARY

[0005] The present application provides a coal seam water inrush dynamic monitoring method and system based on an intelligent sensing system, to solve the technical problems that the prior art is prone to large errors, difficult to meet the demand for high-precision positioning in a complex roadway environment, and slow in response.

[0006] In a first aspect, to solve the above technical problems, the present application provides a coal seam water inrush dynamic monitoring method based on an intelligent sensing system, comprising: S10, collecting real-time pressure wave signal data based on a multi-point sensor array, preprocessing the real-time pressure wave signal data, and obtaining a pressure wave time sequence; S20, calculating a time difference sequence between different sensors based on the pressure wave time sequence, and obtaining an initial three-dimensional coordinate of a water inrush source based on a time difference positioning algorithm; S30, obtaining a wave speed deviation feature from the initial three-dimensional coordinate of the water inrush source, calculating a statistical index of the deviation feature, obtaining a deviation index value, dynamically correcting the wave speed if the deviation index value exceeds a preset deviation threshold, and obtaining a wave speed distribution map; S40, updating input parameters of the time difference positioning algorithm based on the wave speed distribution map, and re-obtaining a three-dimensional coordinate of the water inrush source; S50, obtaining a fracture distribution information matched with a current coal seam based on a preset geological structure database, and determining a set of potential pressure wave refraction paths; S60, simulating a multi-path propagation scenario based on the set of potential pressure wave refraction paths, selecting a shortest path, optimizing the re-obtained three-dimensional coordinate of the water inrush source, and obtaining an optimized three-dimensional coordinate of the water inrush source; S70, iteratively tracking the optimized three-dimensional coordinate of the water inrush source based on a particle filtering algorithm, fusing continuous pressure waves to update a particle state, and obtaining a dynamic positioning result in real time; and S80, obtaining a boundary coordinate of a water inrush influence range based on the dynamic positioning result, generating a real-time alarm signal and determining an alarm level if the boundary coordinate exceeds a preset boundary threshold, and taking corresponding emergency measures.

[0007] Optionally, step S10 comprises: S101, collecting real-time pressure wave signal data based on a multi-point sensor array deployed in a coal seam roadway; S102, decomposing the pressure wave signal data using a wavelet transform algorithm to obtain detail coefficients; S103, applying soft threshold processing to the detail coefficients to filter out random peaks caused by high-frequency noise; S104, reconstructing the pressure wave signal data by inverse wavelet transform to obtain reconstructed pressure wave signal data; and S105, obtaining a pressure wave time sequence based on the reconstructed pressure wave signal data.

[0008] Optionally, step S20 comprises: S201, processing the pressure wave time sequence based on DolphinDB to obtain time differences between multiple sensors; S202, forming a time difference sequence matrix based on the time differences between multiple sensors; S203, obtaining an estimated three-dimensional coordinate of a water inrush source based on the time difference sequence matrix, a time difference positioning algorithm, and a preset coal seam medium model; and S204, adjusting the three-dimensional coordinate using a least squares method to obtain an initial three-dimensional coordinate of the water inrush source if the estimated three-dimensional coordinate exceeds a preset coordinate threshold.

[0009] Optionally, the step S30 comprises: S301, obtaining a wave velocity deviation feature from the initial three-dimensional coordinates of the water inrush source, calculating a statistical index of the deviation feature to obtain a deviation index value; S302, if the deviation index value exceeds a preset deviation threshold, dynamically correcting the wave velocity to obtain a corrected wave velocity; and S303, discretizing the corrected wave velocity according to spatial coordinates to obtain a wave velocity distribution map.

[0010] Optionally, the step S40 comprises: S401, correcting the wave velocity distribution map based on a geological simulation software to obtain a corrected wave velocity distribution map; S402, importing the corrected wave velocity distribution map into a time difference positioning algorithm system to automatically update input parameters for a time difference positioning algorithm to obtain an optimized time difference positioning algorithm, wherein the input parameters comprise sensor array coordinates and corresponding time difference measurement values; S403, simulating a sound wave propagation path based on a ray tracing algorithm to calculate an effective wave velocity to optimize an over-determined equation set of the time difference positioning algorithm; and S404, solving the over-determined equation set by using a nonlinear least squares optimization method to thereby redetermine the three-dimensional coordinates of the water inrush source.

[0011] Optionally, the step S50 comprises: S501, extracting, based on a database query instruction, fracture distribution information matched with a current coal seam from a geological structure database; S502, fusing the fracture distribution information into the corrected wave velocity distribution map to obtain a fused map; and S503, based on the fused map, simulating wave propagation from a source point at an initial velocity by using a ray tracing algorithm, calculating a refraction path by Snell's law, generating a plurality of paths based on a Monte Carlo simulation method, and screening a path set with an energy loss less than a preset loss threshold to thereby determine a potential pressure wave refraction path set.

[0012] Optionally, the step S60 comprises: S601, simulating a multi-path propagation scenario based on the potential pressure wave refraction path set by using a ray tracing algorithm; S602, judging a matching degree between the simulated multi-path propagation scenario and an actual time difference sequence to thereby select a shortest path; and S603, based on the shortest path, optimizing the redetermined three-dimensional coordinates of the water inrush source to obtain optimized three-dimensional coordinates of the water inrush source.

[0013] Optionally, the step S70 comprises: S701, initializing a particle set of the optimized three-dimensional coordinates of the water inrush source based on a particle filtering algorithm; S702, iteratively tracking the optimized three-dimensional coordinates of the water inrush source and calculating a weight of each particle in the particle set by using a Monte Carlo method; and S703, based on the weight of each particle, fusing continuous pressure wave signals to update a state of each particle, achieving dynamic positioning by the updated particle state, and obtaining a dynamic positioning result in real time.

[0014] Optionally, the step S80 comprises: S801, obtaining the water inrush amount based on the dynamic positioning result; S802, processing the water inrush amount based on an edge detection algorithm to obtain boundary coordinates of the water inrush influence range; S803, if the boundary coordinates exceed a preset boundary threshold, generating a real-time alarm signal and outputting a pulse signal through a preset water level sensor to form a discrete event sequence; S804, performing feature extraction on the discrete event sequence to determine an alarm level; and S805, based on the alarm level, taking corresponding emergency measures.

[0015] In a second aspect, the present application provides a coal seam water inrush amount dynamic monitoring system based on an intelligent sensing system, comprising: a data acquisition module for collecting real-time pressure wave signal data based on a multi-point sensor array deployed in a coal seam roadway, pre-processing the real-time pressure wave signal data, and obtaining a pressure wave time sequence; a coordinate estimation module for calculating a time difference sequence between different sensors based on the pressure wave time sequence, and obtaining initial three-dimensional coordinates of a water inrush source based on a time difference positioning algorithm; a dynamic correction module for obtaining a wave speed deviation feature from the initial three-dimensional coordinates of the water inrush source, calculating a statistical index of the deviation feature to obtain a deviation index value, and dynamically correcting the wave speed if the deviation index value exceeds a preset deviation threshold to obtain a wave speed distribution map; a parameter updating module for updating input parameters of the time difference positioning algorithm based on the wave speed distribution map to re-obtain three-dimensional coordinates of the water inrush source; a path determination module for obtaining crack distribution information matching a current coal seam based on a preset geological structure database to determine a set of potential pressure wave refraction paths; a coordinate optimization module for simulating a multi-path propagation scenario based on the set of potential pressure wave refraction paths, selecting a shortest path, and optimizing the re-obtained three-dimensional coordinates of the water inrush source to obtain optimized three-dimensional coordinates of the water inrush source; a state updating module for iteratively tracking the three-dimensional coordinates of the water inrush source at the end point of the shortest path based on a particle filtering algorithm, updating particle states by fusing continuous pressure wave signals, and obtaining a dynamic positioning result in real time; and an alarm generation module for obtaining boundary coordinates of a water inrush influence range based on the dynamic positioning result, generating a real-time alarm signal if the boundary coordinates exceed a preset boundary threshold, determining an alarm level, and taking corresponding emergency measures.

[0016] The one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: The application collects real-time pressure wave signal data through a multi-point sensor array arranged in a coal seam roadway, carries out signal preprocessing to filter out noise in the real-time pressure wave signal data, obtains purified pressure wave time sequence, and calculates the time difference sequence between different sensors, so as to preliminarily estimate the three-dimensional coordinates of the water gushing source; for wave speed deviation, dynamically correct the wave speed distribution, update the input parameters of the positioning algorithm to reduce the error; through the geological structure database, the crack distribution information matched with the current coal seam is obtained, the potential pressure wave refraction path set is determined, the multi-path propagation scene is simulated, the shortest path is selected for optimization, and the particle filtering algorithm is used for iterative tracking, so that dynamic positioning adjustment is realized, and finally the water gushing influence range is extracted and a real-time alarm is generated. Through multi-source data fusion and dynamic optimization, the water gushing source positioning accuracy and disaster response efficiency are significantly improved, and the safety of the coal seam roadway is effectively ensured. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is a flowchart of a coal seam water gushing amount dynamic monitoring method based on an intelligent sensing system provided by the embodiment of the application; Figure 2 is a structural schematic diagram of a coal seam water gushing amount dynamic monitoring system based on an intelligent sensing system provided by the embodiment of the application. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the application.

[0019] In the related description of the embodiment, the terms “include, contain, have” and the like are open terms, which are generally preferred to be understood as containing but not limited to; the term “at least one” is generally preferred to be understood as one or more, wherein “multiple” refers to two or more; the term “at least one of” or the like refers to any combination of these terms, including any combination of single or multiple, for example, “at least one of a, b or c”, or “at least one of a, b and c”, which can represent a, b, c, a-b (i.e. a and b), a-c, b-c, or a-b-c, wherein a, b, and c can be single or multiple; the symbol “A / B” is used to describe the selection relationship of the associated object, which generally represents the “or” relationship before and after.

[0020] In the following description of the present embodiments, the terms used in the present application are merely for the purpose of describing particular embodiments and are not intended to limit the present application. The singular forms "a," "an," and "the" used in the present application and its claims are intended to include plural forms as well, unless the context clearly indicates otherwise.

[0021] Those skilled in the art should understand that in the following description of the present embodiments, the order of the serial numbers does not mean the order of execution, and some or all steps can be executed in parallel or in sequence, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the present embodiments.

[0022] Those skilled in the art should understand that the numerical ranges in the present embodiments should be understood to also specifically disclose each intermediate value between the upper limit and the lower limit of the range. Each smaller range between any stated value or stated range of intermediate values and any other stated value or range of intermediate values is also included within the present application. The upper limit and the lower limit of these smaller ranges can be independently included or excluded from the range.

[0023] Unless otherwise specified, the technical / scientific terms used herein have the same meaning as generally understood by those skilled in the art to which the present application belongs. Although only preferred methods and materials are described herein, any methods and materials similar or equivalent to those described herein can also be used in the implementation or testing of the present application. All documents mentioned in the specification are incorporated by reference to disclose and describe the methods and / or materials related to the documents. In the event of any conflict between the content of the specification and any incorporated document, the content of the specification shall prevail.

[0024] The present embodiments provide a coal seam water gushing amount dynamic monitoring method based on an intelligent sensing system, which can be executed by a computer, referring to Figure 1 The method can include the following steps: S10, based on a multi-point sensor array, collecting real-time pressure wave signal data, pre-processing the real-time pressure wave signal data, and obtaining a pressure wave time sequence.

[0025] In the exemplary embodiments, step S10 can include the following steps: S101, based on a multi-point sensor array deployed in a coal seam roadway, collecting real-time pressure wave signal data; S102, using a wavelet transform algorithm to decompose the pressure wave signal data and obtain a detail coefficient; S103, applying soft threshold processing to the detail coefficient to filter out random peaks caused by high-frequency noise; S104, reconstructing the pressure wave signal data through inverse wavelet transform to obtain reconstructed pressure wave signal data; S105, acquire a pressure wave time sequence based on the reconstructed pressure wave signal data.

[0026] Specifically, the embodiment first acquires pressure wave signal data in real time through a multi-point sensor array deployed in the coal seam roadway, wherein the multi-point sensor array is composed of 8 pressure sensors, each pressure sensor acquires data at a sampling frequency of 500 Hz, the data is transmitted to a central server using a wireless network protocol (such as ZigBee) to realize low-delay transmission, and the server-side software automatically parses the original signal sequence, wherein each sequence has a length of 1024 sampling points and contains fluctuation information of pressure values ranging from 0.1 MPa to 5.0 MPa, so that the acquired data ensures real-time and integrity, providing a basis for subsequent processing. Then, signal preprocessing is performed to filter out noise interference, and a wavelet transform algorithm (such as Daubechies wavelet basis function db4) is used for 4-level decomposition to decompose the signal into approximation coefficients (low-frequency components) and detail coefficients (high-frequency components). Then, a general threshold value is calculated by the following formula: wherein σ is the noise standard deviation estimate value 0.2, and N is the sequence length 1024. Soft threshold processing is applied to the detail coefficients, that is, if the coefficient absolute value is less than λ, it is set to zero, otherwise the signed value of λ is subtracted. In this way, random peaks caused by high-frequency noise (such as roadway vibration interference) are filtered out. The signal is reconstructed by inverse wavelet transform, and the analysis process shows that the signal-to-noise ratio of the original signal is improved from 15 dB to 25 dB after processing, proving that the noise is effectively removed and the main frequency components of the pressure wave such as 10 Hz to 50 Hz are retained. The roadway pressure change characteristics. Finally, the purified pressure wave time sequence is obtained, for example, the output sequence is a float array column with a length of 1024, and each point value is, for example, 0.8 MPa. After smoothing, a continuous curve is formed, which is used for further analysis of the stability of the roadway. The logic forms a closed loop from data acquisition to preprocessing to sequence generation, wherein the threshold value selection based on noise estimation ensures the adaptability of the algorithm. If the noise standard deviation estimate value σ increases from 0.2 to 0.3, the corresponding threshold value should be adjusted to about 1.12 according to the general threshold value formula , wherein N is the sequence length 1024, and the corresponding threshold value should be adjusted to about 1.12 to maintain the filtering effect, thereby supporting a real-time early warning system such as automatically triggering an alarm when the pressure exceeds 3.0 MPa.

[0027] S20, calculate a time difference sequence between different sensors based on the pressure wave time sequence, and acquire an initial three-dimensional coordinate of the water inrush source based on a time difference positioning algorithm.

[0028] The time difference positioning algorithm is a technique for determining the position of a signal source by measuring the time difference of the signal arriving at different observation stations. The core principle is to construct a hyperbolic equation set using the time difference, and determine the target coordinate through the intersection point of the hyperbolic curves.

[0029] In the example embodiment, step S20 can include the following steps: S201, processing the pressure wave time series based on DolphinDB to obtain a plurality of inter-sensor time differences; S202, forming a time difference sequence matrix based on the plurality of inter-sensor time differences; S203, based on the time difference sequence matrix, using a time difference positioning algorithm and a preset coal seam medium model to obtain an estimated three-dimensional coordinate of the water gushing source; S204, if the estimated three-dimensional coordinate exceeds a preset coordinate threshold, adjusting the three-dimensional coordinate using a least squares method to obtain an initial three-dimensional coordinate of the water gushing source.

[0030] Specifically, the embodiment first processes the purified pressure wave time series using DolphinDB. For example, assume that the time series of four sensors A, B, C, and D are P_A(t), P_B(t), P_C(t), and P_D(t), respectively, with a sampling rate of 1000 Hz and a length of 10 seconds. The inter-sensor time difference is calculated by a cross-correlation algorithm, for example, the cross-correlation function of A and B sequences is calculated as follows:

[0031] In the formula, : the cross-correlation function of signals A(t) and B(t+τ), indicating the similarity of the two at time offset τ.

[0032] P_A(t): usually represents the value of signal A at time t.

[0033] P_B(t+τ): represents the value of signal B at time t+τ.

[0034] ∫: integral sign, indicating integration over time t.

[0035] dt: differential of time t.

[0036] τ: time offset, representing the time shift of signal B relative to signal A.

[0037] Find the peak value = 0.5 seconds, indicating that B is delayed by 0.5 seconds relative to A. Similarly, calculate = 0.7 seconds, = 1.2 seconds, = 0.2 seconds, = 0.7 seconds, =0.5 seconds, forming a time difference sequence matrix [0.5, 0.7, 1.2; 0.2, 0.7; 0.5]. Subsequently, a time difference positioning algorithm (such as Chan algorithm) is used in combination with a preset coal seam medium model, wherein the medium sound velocity v=3500 m / s, the sensor coordinates are preset as A(0, 0, 0), B(100, 0, 0), C(0, 100, 0), D(50, 50, 50), a hyperbolic equation set is constructed based on the time difference sequence, for example, for τ_{AB}, the following equation can be constructed: √[(x-100)^2+y^2+z^2]-√[x^2+y^2+z^2]=v*0.5=1750m, then the least squares method is used to solve the nonlinear equation set to adjust the three-dimensional coordinates, so that the initial three-dimensional coordinates of the water gushing source can be obtained.

[0038] S30, obtaining a wave speed deviation feature from the initial three-dimensional coordinates of the water gushing source, calculating a statistical index of the deviation feature to obtain a deviation index value, and if the deviation index value exceeds a preset deviation threshold, dynamically correcting the wave speed to obtain a wave speed distribution map.

[0039] In an exemplary embodiment, step S30 can include the following steps: S301, obtaining a wave speed deviation feature from the initial three-dimensional coordinates of the water gushing source, calculating a statistical index of the deviation feature to obtain a deviation index value; S302, if the deviation index value exceeds a preset deviation threshold, dynamically correcting the wave speed to obtain a corrected wave speed; S303, discretizing the corrected wave speed according to spatial coordinates to obtain a wave speed distribution map.

[0040] Specifically, the embodiment uses the following method to dynamically correct the wave speed: Basic wave speed preset: set the initial wave speed according to the line type and environmental parameters (temperature, humidity).

[0041] Error correction: correct the wave speed using the known target depth method, and take the average value (range <5%) through multi-point measurement.

[0042] Introduce error correction coefficient (such as constant bias compensation of M transformation), reduce systematic error.

[0043] Multi-factor linkage adjustment: dynamically update the wave speed value in combination with terrain changes, etc.

[0044] S40, updating the input parameters of the time difference positioning algorithm based on the wave speed distribution map to reacquire the three-dimensional coordinates of the water gushing source.

[0045] In an exemplary embodiment, step S40 can include the following steps: S401, correcting the wave velocity distribution map based on a geological simulation software to obtain a corrected wave velocity distribution map; S402, importing the corrected wave velocity distribution map into a time difference positioning algorithm system, automatically updating input parameters to a time difference positioning algorithm, obtaining an optimized time difference positioning algorithm, the input parameters including sensor array coordinates and corresponding time difference measurement values; S403, simulating sound wave propagation paths based on a ray tracing algorithm, calculating effective wave velocities to optimize the overdetermined equation set of the time difference positioning algorithm; S404, solving the overdetermined equation set by using a nonlinear least squares optimization method, thereby re-determining the three-dimensional coordinates of the water inrush source.

[0046] The overdetermined equation set refers to a linear system with more equations than unknowns (such as target coordinate parameters), which needs to be solved for the optimal approximate solution by optimization methods such as the least squares method. The overdetermined equation set in the time difference positioning algorithm is represented as a hyperbolic equation set.

[0047] Specifically, the wave velocity distribution map is corrected using a geological simulation software (such as Petrel). A three-dimensional grid model is generated by integrating drilling data (obtained through drilling engineering) and seismic wave test results (seismic waves generated by explosives, pneumatic sources, etc.). For example, the wave velocity of the rock layer is adjusted from the initial uniform value of 3000 m / s to a stratified distribution, where the upper layer of sandstone is corrected to 2500 m / s, the middle layer of granite is 4500 m / s, and the lower layer of shale is 3500 m / s. The kriging interpolation algorithm is used to calculate the wave velocity values of the grid points, ensuring that the resolution of the three-dimensional grid model reaches 1 m accuracy to reflect geological heterogeneity, thereby forming the corrected wave velocity distribution map. Then, this wave velocity distribution map is imported into the time difference positioning algorithm system, and the input parameters are automatically updated. The input parameters include sensor array coordinates such as (0, 0, 0), (200, 0, 0), (0, 200, 0), (0, 0, 200) and corresponding time difference measurement values 0.012 s, 0.018 s, 0.015 s. At the same time, the ray tracing algorithm is used to simulate the sound wave propagation path to calculate the effective wave velocity, for example, the average wave velocity of 3800 m / s is obtained by integrating the path from the source point to the sensor, thereby optimizing the overdetermined equation set of the time difference positioning algorithm. The algorithm then uses a nonlinear least squares optimization method (such as Levenberg-Marquardt) to iterate and re-calculate the three-dimensional coordinates of the water inrush source. The initial estimate is based on the uniform wave velocity to obtain the three-dimensional coordinates (120, 110, 80), and the residual sum of squares is 2.5. After 10 iterations and the integration of the corrected wave velocity, it converges to (118.5, 112.3, 79.2), which is the re-acquired three-dimensional coordinates of the water inrush source.

[0048] Furthermore, after recalculating the three-dimensional coordinates of the water source, the error analysis module compared the results before and after, showing that the wave velocity correction reduced the time difference error caused by the path deviation by 20%. Logically, this is related to the automatic triggering of alarm threshold adjustment in the business monitoring system. For example, when the error is less than 4m, the drainage pump is linked to optimize scheduling, ensuring that the overall positioning accuracy is improved and supporting the continuity of mine water inrush prevention and control decisions.

[0049] S50: Based on a pre-set geological structure database, obtain fracture distribution information that matches the current coal seam, and determine the set of potential pressure wave refraction paths.

[0050] In an exemplary embodiment, step S50 may include the following steps: S501, based on database query commands, extracts fracture distribution information matching the current coal seam from the geological structure database; S502, the fracture distribution information is fused into the corrected wave velocity distribution map to obtain the fused map; S503, based on the fused graph, a ray tracing algorithm is used to simulate the propagation of the wave from the source point with an initial velocity. The refraction path is calculated using Snell's law. Multiple paths are generated based on the Monte Carlo simulation method. The set of paths with energy loss less than a preset loss threshold is selected to determine the potential set of pressure wave refraction paths.

[0051] In this exemplary embodiment, fracture distribution information matching the current coal seam is first extracted from the geological structure database using a database query command, for example, using the following SQL query command: SELECT `fracture density`, `fracture direction` FROM `geological table` WHERE `coal seam horizon`='specified horizon' AND `similarity`>0.85; Assuming the current coal seam depth is 500 meters, the query results show a fracture density of 2.5 fractures per square meter, with an average direction of 45 degrees northeast. Next, the similarity score will be calculated using the cosine similarity algorithm: = (A • B) / (|A||B|), where A is the current coal seam vector [depth 500, thickness 10] and B is the database vector [depth 510, thickness 9.5], resulting in a similarity of 0.92, which is higher than the preset similarity threshold, thereby confirming the match and extracting the data to support subsequent fusion. Next, these fracture information is fused into the corrected wave velocity distribution map, for example, using a grid interpolation algorithm such as the Kriging method, the wave velocity map is updated, assuming that the wave velocity distribution map before correction is a 100x100 grid, and the wave velocity value of each grid point is 1500 m / s, the fracture influence factor is calculated when fusion: f = 1 - 0.2 * density, i.e. f = 1 - 0.2 * 2.5 = 0.5, then the new wave velocity = original wave velocity * f = 1500 * 0.5 = 750 m / s, next, based on the following formula, the error function E = ∑(observed wave velocity - predicted wave velocity)^2 is iteratively optimized to minimize, the grid value is updated by the gradient descent algorithm, ensuring that the overall variance of the fused map is reduced from the original 5.0 to 2.3, providing a more accurate wave velocity model to correlate pressure wave propagation services. Finally, based on the fused map, a set of potential pressure wave refraction paths is determined, for example, applying a ray tracing algorithm to simulate wave propagation from the source point (coordinates 0, 0) at an initial speed of 2000 m / s, and Snell's law is used to calculate the refraction path:

[0052] where: : the angle of incidence, i.e. the angle between the ray or wavefront and the normal.

[0053] v1: the propagation speed of the wave in medium 1.

[0054] : the angle of refraction, i.e. the angle between the ray or wavefront and the normal after entering medium 2.

[0055] v2: the propagation speed of the wave in medium 2.

[0056] where v1 = 1500 m / s, v2 = 750 m / s, = 30 degrees, then = arcsin((750 / 1500) * sin 30) = 14.48 degrees, a total of 1000 paths are generated, and a set of paths with energy loss less than 10% is selected, a total of 50 potential paths are obtained.

[0057] S60, based on the set of potential pressure wave refraction paths, simulate a multi-path propagation scenario, select the shortest path, and optimize the re-acquired water inrush source three-dimensional coordinates to obtain optimized water inrush source three-dimensional coordinates.

[0058] In an exemplary embodiment, step S60 can include the following steps: S601, simulate a multi-path propagation scenario based on a set of potential pressure wave refraction paths by a ray tracing algorithm; S602, determine the matching degree between the simulated multi-path propagation scenario and the actual time difference sequence, thereby selecting the shortest path; S603, based on the shortest path, optimize the re-acquired water inrush source three-dimensional coordinates to obtain optimized water inrush source three-dimensional coordinates.

[0059] Specifically, the embodiment simulates a multi-path propagation scenario based on a set of potential pressure wave refraction paths by a ray tracing algorithm, for example, assuming that the medium velocity model is a layered structure, where the upper layer velocity is 2000 m / s and the lower layer is 3000 m / s, 10 potential paths are generated, and the propagation time from the source point to the receiver is calculated for each path, such as path 1 time is 0.5 seconds, path 2 is 0.6 seconds, the wave front propagation is analyzed by using the finite difference method, and the refraction angle is calculated by Snell's law, that is, , is the incident angle, v1 is the propagation velocity of the wave in medium 1, thereby constructing the scene time sequence [0.5, 0.55, 0.6], this simulation process is related to geological exploration business, ensures that the path set covers possible hydrological fault influence, and forms a complete propagation model from the initial estimate of the source point coordinates (50, 100, 20) to the receiver array. Then determine the matching degree between the simulated scenario and the actual time difference sequence, for example, the actual sequence is [0.51, 0.56, 0.59], the matching degree is calculated by using the Pearson correlation coefficient algorithm, the formula is: , wherein, ρ: Pearson correlation coefficient.

[0060] xi and yi: respectively represent the i-th observation value of variable x and variable y.

[0061] μx and μy: respectively represent the mean of variable x and variable y.

[0062] and : respectively represent the standard deviation of variable x and variable y, the correlation coefficient ρ=0.95 is obtained, which is higher than the preset correlation threshold 0.9, then the shortest path is selected, such as path 3 corresponding to the minimum residual 0.01 seconds, this judgment logic chain is related to the real-time monitoring system, and the path selection is automatically iterated to optimize the accuracy, avoiding human intervention. Finally, the optimized water inrush source coordinates are obtained.

[0063] S70, based on the particle filtering algorithm, iteratively track the optimized water inrush source three-dimensional coordinates, update the particle state by fusing continuous pressure wave signals, and real-time acquire dynamic positioning results.

[0064] In the example embodiment, step S70 can include the following steps: S701, initializing a particle set of optimized water inrush source three-dimensional coordinates based on a particle filter algorithm; S702, iteratively tracking the optimized water inrush source three-dimensional coordinates, and calculating the weight of each particle in the particle set by using a Monte Carlo method; S703, updating the state of each particle based on the weight of each particle by fusing the continuous pressure wave signals, realizing dynamic positioning through the updated particle state, and obtaining the dynamic positioning result in real time.

[0065] Specifically, the example embodiment first initializes the particle set of water inrush source coordinates by a particle filter algorithm, for example, assuming that the initial coordinates are (x=10.5, y=20.3, z=5.0) meters, and 1000 particles are generated, each particle is randomly sampled in a Gaussian distribution, the mean value is the initial coordinates, and the variance is 2.0 meters to simulate the position uncertainty; then iteratively tracking the optimized water inrush source three-dimensional coordinates, and calculating the weight of each particle by using a Monte Carlo method, for example, in the first iteration, the input pressure wave signal intensity is 15.2 Pa, and the particle is resampled according to the Bayesian update formula P(x|z)=P(z|x)P(x) / P(z), wherein P(z|x) is a likelihood function, and an exponential model is used , wherein exp: exponential function.

[0066] z: current observation value of the random variable.

[0067] μ: mean value of the distribution, i.e. average value or center position of the data.

[0068] σ: standard deviation of the distribution, which measures the dispersion degree of the data around the mean value.

[0069] (z−μ): the difference between the observation value z and the mean value μ, i.e. deviation.

[0070] : standardized squared deviation.

[0071] : this is the exponential part in the normal distribution probability density function, and the negative sign ensures that the function value decreases when z is far away from μ, and =1.0 Pa, resulting in that the particles with high weight such as coordinates (10.6, 20.4, 5.1) are retained, and the particles with low weight are eliminated, so that the coordinates are optimized to (10.55, 20.35, 5.05) meters; then the continuous pressure wave signals are fused and updated, for example, the second signal intensity sequence [16.1, 15.9, 16.0] Pa is received, the algorithm updates the particle state by fusing the Kalman gain, the gain K=0.6, and the calculation is K=PH T(HPH T +R) -1 , where K: often referred to as the gain matrix or Kalman gain (in the context of Kalman filtering), it represents how to update the state estimate based on new observation data.

[0072] P: the prior estimate error covariance matrix, which describes the uncertainty of the state estimate without new observation data.

[0073] H: the observation matrix, which maps the state space to the observation space, i.e., describes how to get the observation variable from the state variable.

[0074] R: the observation noise covariance matrix, which describes the noise or uncertainty in the observation data.

[0075] H T : the transpose of the observation matrix H.

[0076] (HPHT+R) −1 : this is part of the information matrix or the inverse of the covariance matrix, used to calculate the gain matrix K.

[0077] P initially is [[4.0, 0, 0], [0, 4.0, 0], [0, 0, 4.0]], leading the particle set to converge to the new mean (10.7, 20.5, 5.2) meters, ensuring real-time adjustment when the water source dynamic changes such as water pressure fluctuations. Finally, dynamic positioning adjustment is achieved, for example, after the signal is fused to 17.0 Pa for the 3rd iteration, the algorithm resamples to generate a new particle set, outputs the adjusted coordinates (10.8, 20.6, 5.3) meters, and verifies through the thought chain: if the signal strength exceeds the threshold of 16.0 Pa, increase the particle density to 1500 to improve accuracy, otherwise maintain the original density, forming a closed-loop optimization to ensure that the positioning error is less than 1.0 meters.

[0078] S80, based on the dynamic positioning result, obtains the boundary coordinates of the water inrush influence range, generates a real-time alarm signal if the boundary coordinates exceed the preset boundary threshold, determines the alarm level, and takes corresponding emergency measures.

[0079] In an exemplary embodiment, step S80 can include the following steps: S801, obtaining the water inrush amount based on the dynamic positioning result; S802, processing the water inrush amount based on an edge detection algorithm to obtain the boundary coordinates of the water inrush influence range; S803, if the boundary coordinates exceed the preset boundary threshold, generating a real-time alarm signal and outputting a pulse signal through a preset water level sensor to form a discrete event sequence; S804, feature extraction is performed on the discrete event sequence to determine the alarm level; S805, based on the alarm level, take corresponding emergency measures.

[0080] Specifically, the real-time acquired water inflow images are processed using edge detection algorithms (such as the Canny algorithm). A threshold is set with a lower limit of 50 and an upper limit of 150. The gradient magnitude is calculated and non-maximum suppression is applied to extract the boundary of the water inflow influence range. For example, if the boundary pixel coordinates are detected as (x=120.5, y=200.3), the water depth change is simulated by combining a flood diffusion model such as the shallow water equation. Assuming the initial water level is 2.5 meters and the flow velocity is 1.2 meters / second, the diffusion range is iteratively calculated using the finite difference method, resulting in an elliptical region with a boundary polygon area of ​​approximately 1500 square meters, thereby obtaining the boundary coordinates of the water inflow influence range.

[0081] Furthermore, the alert levels are divided into yellow, blue, and red alerts, from low to high. The emergency measures for each of the three alert levels are explained below: Yellow alert: If the water inflow exceeds the normal value by 10% or the frequency is ≤3 times / hour, notify the on-duty personnel to strengthen monitoring; Blue alert: Water inflow exceeds normal value by 50%, notify all personnel to evacuate; Red alert: If the water inflow exceeds 100% or is accompanied by rock movement, the entire mine must be evacuated immediately.

[0082] In the above embodiments of this application, real-time pressure wave signal data is collected by a multi-point sensor array deployed in the coal seam roadway. Signal preprocessing is performed to filter out noise from the real-time pressure wave signal data, obtaining a purified pressure wave time sequence. The time difference sequence between different sensors is calculated to initially estimate the three-dimensional coordinates of the water inrush source. For wave velocity deviations, the wave velocity distribution is dynamically corrected, and the input parameters of the positioning algorithm are updated to reduce errors. Fracturation distribution information matching the current coal seam is obtained from a geological structure database to determine a set of potential pressure wave refraction paths. A multi-path propagation scenario is simulated, the shortest path is selected for optimization, and a particle filter algorithm is used for iterative tracking to achieve dynamic positioning adjustment. Finally, the impact range of the water inrush is extracted, and a real-time alarm is generated. This invention, through multi-source data fusion and dynamic optimization, significantly improves the positioning accuracy of water inrush sources and the efficiency of disaster response, effectively ensuring the safety of coal seam roadways.

[0083] Based on the above embodiments, this application also provides a dynamic monitoring system for coal seam water inflow based on an intelligent sensing system. Figure 2 This is a schematic diagram of a dynamic monitoring system for coal seam water inflow based on an intelligent sensing system, as described in an embodiment of the present invention. Figure 2 As shown, the coal seam water inflow dynamic monitoring system 200 based on an intelligent sensing system may include: The data acquisition module 210 is configured to acquire real-time pressure wave signal data based on a multi-point sensor array deployed in a coal seam tunnel, preprocess the real-time pressure wave signal data, and acquire a pressure wave time sequence; The coordinate estimation module 220 is configured to calculate a time difference sequence between different sensors based on the pressure wave time sequence, and acquire an initial three-dimensional coordinate of the water gushing source based on a time difference positioning algorithm. The dynamic correction module 230 is configured to acquire a wave speed deviation feature from the initial three-dimensional coordinate of the water gushing source, calculate a statistical index of the deviation feature to obtain a deviation index value, and perform dynamic correction on the wave speed if the deviation index value exceeds a preset deviation threshold to obtain a wave speed distribution map. The parameter updating module 240 is configured to update input parameters of the time difference positioning algorithm based on the wave speed distribution map, and reacquire the three-dimensional coordinate of the water gushing source. The path determination module 250 is configured to acquire crack distribution information matched with a current coal seam based on a preset geological structure database, and determine a set of potential pressure wave refraction paths. The coordinate optimization module 260 is configured to simulate a multi-path propagation scenario based on the set of potential pressure wave refraction paths, select a shortest path, and optimize the reacquired three-dimensional coordinate of the water gushing source to obtain an optimized three-dimensional coordinate of the water gushing source. The state updating module 270 is configured to iteratively track the three-dimensional coordinate of the water gushing source at the end point of the shortest path based on a particle filtering algorithm, update a particle state by fusing continuous pressure waves, and acquire a dynamic positioning result in real time. The alarm generation module 280 is configured to acquire boundary coordinates of a water gushing influence range based on the dynamic positioning result, generate a real-time alarm signal and determine an alarm level if the boundary coordinates exceed a preset boundary threshold, and take corresponding emergency measures.

[0084] In the description of the present application, it should be noted that the terms "first", "second", "third" are only for descriptive purposes and cannot be understood or implied as indicating or suggesting relative importance.

[0085] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, device and unit can refer to the corresponding processes in the foregoing method embodiments, which will not be described herein.

[0086] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other manners. The described device embodiments are merely schematic, for example, the division of the units is only a logical function division, and there can be another division manner in actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units, can be indirect couplings or communication connections through some interfaces, devices or units, and can be electric, mechanical or in other forms.

[0087] The units described as separated components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. In actual implementation, some or all of the units can be selected according to actual needs to achieve the purposes of the embodiments.

[0088] In addition, each functional unit in the embodiments of the present application can be integrated in one processing unit, or each unit can exist physically as separate units, or two or more units can be integrated in one unit.

[0089] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer readable storage medium of a processor. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art, or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present application. The foregoing storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk, and various media that can store program codes.

[0090] In the description of the present application, it should be noted that the terms "first", "second", "third" are only for the purpose of description, and cannot be understood as indicating or implying relative importance.

[0091] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0092] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other manners. The described device embodiments are merely schematic, for example, the division of the units is only a logical function division, and there can be another division manner in actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units, or the among different units, can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0093] The units described as separated components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purposes of the embodiments of the present application.

[0094] In addition, each functional unit in the embodiments of the present application can be integrated in one processing unit, or each unit can exist physically as a separate unit, or two or more units can be integrated in one unit.

[0095] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a non-volatile computer readable storage medium executable by a processor. Based on this understanding, the technical solutions of the present application or the parts of the technical solutions that make essential contributions to the prior art can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various program codes that can be stored in the medium.

[0096] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present application, and are not intended to limit the technical solutions of the present application. The protection scope of the present application is not limited thereto. Although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those skilled in the art that any person skilled in the art can make modifications or easy changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed by the present application, or can easily think of changes or equivalent replacements for some of the technical features. The modifications, changes or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0097] In addition, although the operations of the methods of the present application are described in a particular order in the drawings, this does not require or imply that the operations must be performed in that particular order, or that all of the illustrated operations must be performed to achieve the desired results. Additionally or alternatively, certain steps can be omitted, combined into a single step, and / or divided into multiple steps.

Claims

1. A method for dynamic monitoring of coal seam water inflow based on an intelligent sensing system, characterized in that, include: S10, real-time pressure wave signal data is collected based on a multi-point sensor array, and the real-time pressure wave signal data is preprocessed to obtain a pressure wave time sequence; S20, calculate the time difference sequence between different sensors based on the pressure wave time sequence, and obtain the initial three-dimensional coordinates of the water source based on the time difference positioning algorithm; S30, obtain wave velocity deviation characteristics from the initial three-dimensional coordinates of the water source, calculate the statistical index of the deviation characteristics, obtain the deviation index value, and if the deviation index value exceeds the preset deviation threshold, dynamically correct the wave velocity to obtain a wave velocity distribution map. S40, update the input parameters of the time difference positioning algorithm based on the wave velocity distribution map, and re-obtain the three-dimensional coordinates of the water source; S50: Based on a pre-set geological structure database, obtain fracture distribution information that matches the current coal seam and determine the set of potential pressure wave refraction paths; S60, based on the set of potential pressure wave refraction paths, simulates a multi-path propagation scenario, selects the shortest path, and optimizes the reacquired three-dimensional coordinates of the water source to obtain the optimized three-dimensional coordinates of the water source. The S70 uses a particle filtering algorithm to iteratively track the optimized three-dimensional coordinates of the water source, integrates continuous pressure wave signals to update the particle state, and obtains dynamic positioning results in real time. S80, based on the dynamic positioning results, obtains the boundary coordinates of the water inrush area. If the boundary coordinates exceed the preset boundary threshold, it generates a real-time alarm signal, determines the alarm level, and takes corresponding emergency measures.

2. The method according to claim 1, characterized in that, Step S10 includes: S101, based on a multi-point sensor array deployed in coal seam roadways, collects real-time pressure wave signal data; S102, The pressure wave signal data is decomposed using a wavelet transform algorithm to obtain detail coefficients; S103, apply soft thresholding to the detail coefficients to filter out random peaks caused by high-frequency noise; S104, the pressure wave signal data is reconstructed by inverse wavelet transform to obtain the reconstructed pressure wave signal data; S105, Based on the reconstructed pressure wave signal data, obtain the pressure wave time sequence.

3. The method according to claim 1, characterized in that, Step S20 includes: S201, Process the pressure wave time series based on DolphinDB to obtain the time difference between multiple sensors; S202, Based on the time difference between the multiple sensors, a time difference sequence matrix is ​​formed; S203, Based on the time difference sequence matrix, the estimated three-dimensional coordinates of the water inrush source are obtained by using a time difference positioning algorithm and a preset coal seam medium model; S204, if the estimated three-dimensional coordinates exceed the preset coordinate threshold, the three-dimensional coordinates are adjusted using the least squares method to obtain the initial three-dimensional coordinates of the water source.

4. The method according to claim 1, characterized in that, Step S30 includes: S301, Obtain wave velocity deviation characteristics from the initial three-dimensional coordinates of the water source, calculate the statistical index of the deviation characteristics, and obtain the deviation index value; S302, if the deviation index value exceeds the preset deviation threshold, the wave speed is dynamically corrected to obtain the corrected wave speed. S303 discretizes the corrected wave velocity according to spatial coordinates to obtain the wave velocity distribution map.

5. The method according to claim 1, characterized in that, Step S40 includes: S401, Based on geological simulation software, the wave velocity distribution map is corrected to obtain the corrected wave velocity distribution map; S402, import the corrected wave velocity distribution map into the time difference positioning algorithm system, automatically update the input parameters for the time difference positioning algorithm, and obtain the optimized time difference positioning algorithm. The input parameters include sensor array coordinates and corresponding time difference measurement values. S403 uses a ray tracing algorithm to simulate the sound wave propagation path and calculate the effective wave velocity in order to optimize the overdetermined equations of the time difference positioning algorithm. S404. The overdetermined equations are solved using a nonlinear least squares optimization method, thereby redetermining the three-dimensional coordinates of the water source.

6. The method according to claim 5, characterized in that, Step S50 includes: S501, based on database query commands, extracts fracture distribution information matching the current coal seam from the geological structure database; S502, the fracture distribution information is fused into the corrected wave velocity distribution map to obtain the fused map; S503, based on the fused graph, a ray tracing algorithm is used to simulate the propagation of the wave from the source point with an initial velocity. The refraction path is calculated using Snell's law. Multiple paths are generated based on the Monte Carlo simulation method. The set of paths with energy loss less than a preset loss threshold is selected to determine the potential set of pressure wave refraction paths.

7. The method according to claim 1, characterized in that, Step S60 includes: S601 uses a ray tracing algorithm to simulate multipath propagation scenarios based on a set of potential pressure wave refraction paths. S602, determine the matching degree between the simulated multipath propagation scenario and the actual time difference sequence, and select the shortest path; S603, Based on the nearest path, optimize the reacquired three-dimensional coordinates of the water source to obtain optimized three-dimensional coordinates of the water source.

8. The method according to claim 1, characterized in that, Step S70 includes: S701, a set of particles that initialize and optimize the three-dimensional coordinates of the water source based on the particle filter algorithm; S702, iteratively tracks the optimized three-dimensional coordinates of the water source and uses the Monte Carlo method to calculate the weight of each particle in the particle set; S703 updates the state of each particle by fusing continuous pressure wave signals based on the weight of each particle, achieves dynamic positioning through the updated particle state, and obtains the dynamic positioning results in real time.

9. The method according to claim 1, characterized in that, Step S80 includes: S801, obtains water inflow based on dynamic positioning results; S802, based on the edge detection algorithm, processes the water inflow volume and obtains the boundary coordinates of the water inflow influence range; S803: If the boundary coordinates exceed the preset boundary threshold, a real-time alarm signal is generated, and a pulse signal is output through the preset water level sensor to form a discrete event sequence. S804, Perform feature extraction on the discrete event sequence to determine the alarm level; S805, based on the alarm level, take corresponding emergency measures.

10. A dynamic monitoring system for coal seam water inflow based on an intelligent sensing system, characterized in that, include: The data acquisition module is used to collect real-time pressure wave signal data based on a multi-point sensor array deployed in the coal seam roadway, preprocess the real-time pressure wave signal data, and obtain the pressure wave time sequence. The coordinate estimation module is used to calculate the time difference sequence between different sensors based on the pressure wave time sequence, and to obtain the initial three-dimensional coordinates of the water source based on the time difference positioning algorithm. The dynamic correction module is used to obtain wave velocity deviation characteristics from the initial three-dimensional coordinates of the water source, calculate the statistical index of the deviation characteristics, obtain the deviation index value, and if the deviation index value exceeds the preset deviation threshold, dynamically correct the wave velocity to obtain a wave velocity distribution map. The parameter update module is used to update the input parameters of the time difference positioning algorithm based on the wave velocity distribution map, and to re-acquire the three-dimensional coordinates of the water source; The path determination module is used to obtain fracture distribution information that matches the current coal seam based on a preset geological structure database, and to determine the set of potential pressure wave refraction paths. The coordinate optimization module is used to simulate multi-path propagation scenarios based on a set of potential pressure wave refraction paths, select the shortest path, and optimize the reacquired three-dimensional coordinates of the water source to obtain the optimized three-dimensional coordinates of the water source. The state update module is used to iteratively track the three-dimensional coordinates of the water source at the shortest path endpoint based on the particle filtering algorithm, and to update the particle state by fusing continuous pressure wave signals and obtain dynamic positioning results in real time. The alarm generation module is used to obtain the boundary coordinates of the water inrush area based on the dynamic positioning results. If the boundary coordinates exceed the preset boundary threshold, a real-time alarm signal is generated, the alarm level is determined, and corresponding emergency measures are taken.

Images