A mine tunnel damage state monitoring method, device, equipment and medium

Abstract

This application discloses a method, device, equipment, and medium for monitoring the damage state of mine roadways, relating to the field of mine seismic assessment technology. The method includes: determining the peak particle velocity (PPV) of each roadway surface at different sources and distances; determining an initial value for the attenuation fitting coefficient based on the PPV of each roadway surface; determining a PPV threshold based on the initial value; determining a target support mechanical model using the PPV threshold and a pre-constructed support mechanical model; determining several support parameter vectors based on the target support mechanical model; training an initial backpropagation neural network using each support parameter vector, the degree of damage, and the PPV of the roadway surface to obtain a target backpropagation neural network; and determining the degree of damage to the roadway based on the target backpropagation neural network, the current support parameter vector, and the current PPV of the roadway surface. This achieves accurate assessment of mine roadway damage.

Description

Technical Field

[0001] This invention relates to the field of mine seismic assessment technology, and in particular to a method, device, equipment and medium for monitoring the damage status of mine roadways. Background Technology

[0002] Existing technologies related to mine seismic and roadway damage assessment are mainly divided into three independently implemented schemes, with no data interaction or linkage between them: 1. Sensor monitoring solution: The three-dimensional accelerometer module is used to collect roadway vibration signals. After noise reduction and analog-to-digital conversion by the signal processing circuit board, the signals are uploaded to the host computer. This only realizes the acquisition and preliminary display of mine vibration acceleration signals. The sensor is fixed by the housing connection slot. The module is equipped with multiple sets of sensing units, each set is equipped with multiple parallel accelerometers, which can detect acceleration signals in three-dimensional orthogonal directions.

[0003] 2. Numerical simulation scheme: A three-dimensional geological model is built using numerical simulation software. A special damage unit is embedded in the model to simulate the fracture of the rock layer interface. Preset mine seismic waveform sample data is input to deduce the rock layer damage evolution process. The degree of roadway damage is qualitatively assessed only through the single mapping relationship between peak particle velocity and mine seismic intensity.

[0004] 3. Macro-level survey plan: The survey path for mine seismic damage is planned manually, and survey points are set up at fixed intervals. Descriptive data on the damage morphology of the roadways at the survey point locations are collected. Combined with the subjective judgment of on-site personnel and qualitative information such as the distribution characteristics of roadway cracks, the intensity and damage level of the mine seismic event are comprehensively determined.

[0005] However, these solutions have the following core flaws and cannot complement each other: 1. Sensor monitoring solution: It only focuses on vibration signal acquisition and primary processing, without combining the roadway support type to carry out vibration resistance characteristic analysis, and cannot establish a quantitative correlation between support parameters and mine seismic damage. It is completely lacking in the ability to support the optimization of support solutions.

[0006] 2. Numerical simulation scheme: It lacks dynamic calibration and fusion of real-time on-site monitoring data, and relies solely on preset seismic samples for simulation. It is difficult to adapt to the complex and ever-changing geological and mining conditions underground, and it does not include a support structure response analysis module. The evaluation results are detached from the actual on-site situation and have extremely low practicality.

[0007] 3. Macro-level survey scheme: The purely manual observation mode results in low data collection efficiency and strong subjectivity. It fails to achieve three-dimensional coupling of monitoring data, simulation data, and field survey data, and the assessment results are inaccurate and lack real-time performance, which cannot meet the needs of rapid emergency response after a mine tremor.

[0008] Therefore, how to overcome the problems of fragmented multi-scale analysis, insufficient quantification of support response, low accuracy of multi-source data fusion, and poor dynamic adaptability in existing mine seismic assessment technologies is a problem that needs to be considered at present. Summary of the Invention

[0009] In view of this, the purpose of this invention is to provide a method, device, equipment, and medium for monitoring the damage state of mine roadways. This invention overcomes the problems of fragmented multi-scale analysis, insufficient quantification of support response, low accuracy of multi-source data fusion, and poor dynamic adaptability in existing mine seismic assessment technologies. It breaks through the limitations of single-scale analysis, achieving full-process quantification of seismic source generation—energy propagation—support response—roadway damage, significantly improving the comprehensiveness and accuracy of the assessment. The specific solution is as follows: In a first aspect, this application discloses a method for monitoring the damage state of mine roadways, including: Based on the pre-constructed energy propagation model and the source energy release determination model, the peak particle velocity of each roadway surface at different source release energies and distances is determined. Based on the peak particle velocity of each roadway surface, the initial value of the attenuation fitting coefficient is determined. Based on the initial value, the peak particle velocity threshold of the roadway surface at the start of roadway damage is determined. The target support mechanical model is determined by the peak particle velocity threshold on the roadway surface and the pre-constructed support mechanical model. Based on the target support mechanical model, several support parameter vectors are determined. The initial backpropagation neural network is trained using each of the support parameter vectors, the degree of damage, and the peak particle velocity on the roadway surface to obtain the target backpropagation neural network. The degree of damage to the roadway is determined based on the target backpropagation neural network, the current support parameter vector, and the current peak particle velocity on the roadway surface.

[0010] Optionally, before determining the peak particle velocity at different roadway surfaces and distances based on the pre-built energy propagation model and the source energy determination model, the method further includes: The stress in the coal and rock mass measured in the field is determined by stress gauges, and the physical and mechanical parameters of the rock mass are determined based on indoor rock mechanics tests. The dip angle, fracture propagation rate, and fracture length are determined based on historical mine seismic records. The physical and mechanical parameters of the rock mass include the friction coefficient of the weak surface of the rock mass. A model for determining the energy released from the seismic source is determined based on the formula for determining the energy released from the seismic source; the formula for determining the energy released from the seismic source is: ; in, The energy released by the seismic source is σ, the stress of the coal and rock mass is μ, the friction coefficient of the weak surface of the rock mass is α, the dip angle of the fracture surface is V, the fracture propagation rate is L, and the length of the fracture surface is L. The model for determining the energy released from the seismic source is based on three-dimensional finite element or discrete element methods to determine the digital twin of the rock mass; The spatial coordinates of the seismic source, the energy level released by the seismic source, and the rupture range are determined using the digital twin based on the Mohr-Coulomb criterion and the maximum tensile stress criterion. Based on borehole and geological exploration data, a three-dimensional geological body including faults and rock strata boundaries was established; The initial energy absorption coefficient of each unit in the three-dimensional geological body is determined based on the lithology; The peak particle velocity on the surface of the target tunnel is determined by the current three-dimensional geological body based on the spatial coordinates of the seismic source, the energy level released by the seismic source, the rupture range, and the energy absorption coefficient. The peak particle velocity of the target roadway surface is compared with the actual peak particle velocity of the roadway surface measured by downhole sensors; Based on the corresponding comparison results, the initial energy absorption coefficient of the three-dimensional geological body is adjusted to obtain the adjusted three-dimensional geological body. The adjusted three-dimensional geological body is then determined as the current three-dimensional geological body, and the process jumps back to the step of determining the peak particle velocity of the target tunnel surface based on the spatial coordinates of the seismic source, the energy level released by the seismic source, the rupture range, and the energy absorption coefficient of the current three-dimensional geological body. When the difference between the peak particle velocity of the target tunnel surface and the actual peak particle velocity of the tunnel surface meets the preset conditions, the current three-dimensional geological body is determined as the energy propagation model.

[0011] Optionally, the determination of peak particle velocities on the surface of each roadway at different distances and based on a pre-built energy propagation model and a source energy determination model includes: Several sets of earthquake source release energies are determined based on the model for determining the energy released from the earthquake source. Based on a pre-built energy propagation model, the peak particle velocity on the roadway surface at different distances is determined according to the energy released by each seismic source.

[0012] Optionally, determining the initial value of the attenuation fitting coefficient based on the peak particle velocity at the surface of each of the roadways includes: Based on the peak particle velocity at the surface of each of the aforementioned roadways, a multivariate nonlinear regression is performed on the formula for determining the peak particle velocity at the roadway surface to determine the initial value of the attenuation fitting coefficient; the formula for determining the peak particle velocity at the roadway surface is: ; Wherein, PPV is the peak particle velocity on the surface of the tunnel; The energy released from the seismic source; R is the propagation distance from the seismic source to the measuring point in the tunnel; λ is the rock mass wave impedance; k, a, b, and c are attenuation fitting coefficients; d is the rock stratum interface correction coefficient; Accordingly, determining the peak particle velocity threshold of the roadway surface at the time of roadway damage initiation based on the initial value includes: The peak particle velocity at the roadway surface when sandstone begins to crack and / or mudstone begins to detach was determined as the peak particle velocity threshold at the roadway surface through numerical simulation and field observation.

[0013] Optionally, before determining the target support mechanical model using the peak particle velocity threshold on the roadway surface and the pre-built support mechanical model, the method further includes: An initial support mechanical model is constructed based on the formulas for determining the ultimate vibration resistance of the support structure and the ultimate bending moment of the support; the formula for determining the ultimate vibration resistance of the support structure is: ; in, To support the ultimate vibration resistance of the structure, For anchor bolt and anchor cable prestress, , , These represent the elastic modulus, cross-sectional area, and anchorage length of the rod, respectively, with ΔL representing the vibration displacement. For the stiffness of the metal mesh, The area of ​​the net body subjected to force; The formula for determining the ultimate bending moment of the support is: ; in, The ultimate bending moment of the support. For the yield strength of the support steel, Where D is the cross-sectional modulus of the support frame, and D is the actual erection spacing. To design standard spacing.

[0014] Optionally, determining the target support mechanical model using the peak particle velocity threshold on the roadway surface and a pre-built support mechanical model includes: The ultimate vibration resistance of the target support structure or the ultimate bending moment of the target support is determined by multiplying the peak particle velocity threshold on the roadway surface with a preset conversion coefficient. The target vibration displacement and spacing difference are determined by the initial support mechanical model based on the ultimate vibration resistance of the target support structure or the ultimate bending moment of the target support. The support deformation and structural stress are determined based on the target vibration displacement and the spacing difference. The support deformation and structural stress are compared with the measured support deformation and measured structural stress, respectively. The initial support mechanical model is adjusted according to the comparison results to obtain the target support mechanical model.

[0015] Optionally, the step of training the initial backpropagation neural network using each of the support parameter vectors, the degree of damage, and the peak particle velocity on the roadway surface to obtain the target backpropagation neural network includes: The support parameter vector and the peak particle velocity on the roadway surface are determined as inputs, and the degree of damage is determined as output. The initial backpropagation neural network is trained based on the inputs and outputs to obtain the target backpropagation neural network. The support parameter vector includes the anchor bolt spacing and the anchor bolt and anchor cable prestress.

[0016] Secondly, this application discloses a mine roadway damage monitoring device, comprising: The threshold determination module is used to determine the peak particle velocity of each roadway surface at different distances and different sources of energy release based on a pre-built energy propagation model and a source energy release determination model. Based on the peak particle velocity of each roadway surface, the module determines the initial value of the attenuation fitting coefficient and determines the peak particle velocity threshold of the roadway surface when roadway damage is initiated based on the initial value. The model determination module is used to determine the target support mechanical model by using the peak particle velocity threshold of the roadway surface and the pre-constructed support mechanical model; The training module is used to determine several support parameter vectors based on the target support mechanical model, and to train the initial backpropagation neural network using each of the support parameter vectors, the degree of damage, and the peak particle velocity on the roadway surface, so as to obtain the target backpropagation neural network. The damage degree determination module is used to determine the damage degree of the roadway based on the target backpropagation neural network, the current support parameter vector, and the current peak particle velocity on the roadway surface.

[0017] Thirdly, this application discloses an electronic device, including: Memory, used to store computer programs; A processor for executing computer programs to implement the steps of the aforementioned method for monitoring the damage status of mine roadways.

[0018] Fourthly, this application discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the aforementioned method for monitoring the damage status of mine roadways.

[0019] This application first determines the peak particle velocity of each roadway surface at different distances and with different energy releases from earthquake sources based on a pre-constructed energy propagation model and a source energy determination model. Based on these peak particle velocities, an initial value for the attenuation fitting coefficient is determined, and a threshold for the peak particle velocity at the initiation of roadway damage is determined according to this initial value. A target support mechanics model is then determined using the peak particle velocity threshold and a pre-constructed support mechanics model. Several support parameter vectors are determined based on the target support mechanics model. An initial backpropagation neural network is trained using these support parameter vectors, the degree of damage, and the peak particle velocities on the roadway surface to obtain a target backpropagation neural network. Finally, the degree of roadway damage is determined based on the target backpropagation neural network, the current support parameter vector, and the current peak particle velocity on the roadway surface. Therefore, this application, through deep coupling of multi-scale models and data interoperability, breaks through the limitations of single-scale analysis, achieving full-process quantification of earthquake source occurrence—energy propagation—support response—roadway damage, significantly improving the comprehensiveness and accuracy of the assessment. It can be widely adapted to various coal mine underground roadways for mine seismic safety monitoring systems, especially for roadway stability assessment projects in deep high-stress mines and mines with rock burst hazards, providing precise technical support for roadway safety assurance in complex geological and mining environments. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0021] Figure 1 This is a flowchart of a method for monitoring the damage status of mine roadways disclosed in this application; Figure 2 This is a schematic diagram of the structure of a mine roadway damage monitoring device disclosed in this application; Figure 3 This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] The core shortcomings of existing technical solutions are as follows, and they cannot complement each other: Sensor monitoring solution: It only focuses on vibration signal acquisition and primary processing, without combining it with roadway support type for vibration resistance characteristic analysis. It cannot establish a quantitative correlation between support parameters and mine-induced seismic damage, and completely lacks the support capability for support scheme optimization. Numerical simulation solution: It lacks dynamic calibration and fusion of real-time on-site monitoring data, relying solely on preset mine-induced seismic samples for simulation. This makes it difficult to adapt to the complex and variable geological and mining conditions underground, and it does not include a support structure response analysis module. The evaluation results are detached from actual on-site conditions, resulting in extremely low practicality. Macroscopic survey solution: The purely manual observation mode leads to low data acquisition efficiency and strong subjectivity. It fails to achieve three-dimensional coupling of monitoring data, simulation data, and on-site survey data, resulting in poor accuracy and insufficient real-time performance of the evaluation results, failing to meet the needs of rapid emergency response after a mine-induced seismic event. To address the aforementioned technical problems, this application discloses a method, device, equipment, and medium for monitoring the damage state of mine roadways. This method overcomes the difficulties of existing mine seismic assessment technologies, such as fragmented multi-scale analysis, insufficient quantification of support response, low accuracy of multi-source data fusion, and poor dynamic adaptability. It breaks through the limitations of single-scale analysis, realizes full-process quantification from seismic source occurrence to energy propagation, support response, and roadway damage, and significantly improves the comprehensiveness and accuracy of the assessment.

[0024] See Figure 1 As shown in the figure, an embodiment of the present invention discloses a method for monitoring the damage state of mine roadways, including: Step S11: Based on the pre-constructed energy propagation model and the source energy release determination model, determine the peak particle velocity of each roadway surface at different distances and the energy released by different sources. Based on the peak particle velocity of each roadway surface, determine the initial value of the attenuation fitting coefficient. Based on the initial value, determine the peak particle velocity threshold of the roadway surface when roadway damage is initiated.

[0025] In this embodiment, the stress of the coal and rock mass measured in the field is determined by a stress gauge, the physical and mechanical parameters of the rock mass are determined based on indoor rock mechanics tests, and the dip angle, fracture propagation rate, and fracture length are determined based on historical mine seismic records. The physical and mechanical parameters of the rock mass include the friction coefficient of the weak surface of the rock mass. A model for determining the energy released from the seismic source is determined based on a formula for determining the energy released from the seismic source. The formula for determining the energy released from the seismic source is as follows: ; in, The seismic source release energy is defined by σ, the stress in the coal and rock mass, μ, the friction coefficient of the weak surface of the rock mass, α, the dip angle of the fracture surface, V, the fracture propagation velocity, and L, the length of the fracture surface extension. A digital twin of the rock mass is determined based on the seismic source release energy determination model using three-dimensional finite element or discrete element methods. The spatial coordinates of the seismic source, the energy level of the seismic source release, and the fracture extent are determined using the digital twin according to the Mohr-Coulomb criterion and the maximum tensile stress criterion. A three-dimensional geological body containing faults and strata boundaries is established based on borehole and geological exploration data. The initial energy absorption coefficient of each unit in the three-dimensional geological body is set according to the lithology. The target tunnel is determined using the current three-dimensional geological body based on the spatial coordinates of the seismic source, the energy level of the seismic source release, the fracture extent, and the energy absorption coefficient. Peak particle velocity on the surface of the target roadway; compare the peak particle velocity on the target roadway surface with the actual peak particle velocity on the roadway surface measured by downhole sensors; adjust the initial energy absorption coefficient of the three-dimensional geological body according to the comparison results to obtain the adjusted three-dimensional geological body; determine the adjusted three-dimensional geological body as the current three-dimensional geological body, and jump back to the step of determining the peak particle velocity on the target roadway surface based on the spatial coordinates of the seismic source, the energy level released by the seismic source, the rupture range, and the energy absorption coefficient of the current three-dimensional geological body; when the difference between the peak particle velocity on the target roadway surface and the actual peak particle velocity on the roadway surface meets the preset conditions, the current three-dimensional geological body is determined as the energy propagation model.

[0026] Specifically, the source-scale dynamic model (source energy release determination model) is a multi-parameter coupled dynamic model based on the rock mass fracture mechanism, and its core simplified expression is: ; In the formula: The energy released by the seismic source is represented by σ, the stress of the coal and rock mass, μ, the friction coefficient of the weak surface of the rock mass, α, the dip angle of the fracture surface, V, the fracture propagation velocity, and L, the length of the fracture surface extension. The three-dimensional finite element / discrete element method is used as the numerical calculation carrier, and the judgment conditions of shear fracture (Mohr-Coulomb criterion) and tensile fracture (maximum tensile stress criterion) are embedded. The core parameters such as the spatial coordinates (x, y, z) of the seismic source, energy level, and fracture range are output.

[0027] Specific construction steps: ① Parameter acquisition through field measurement: In-situ stress parameters are measured using borehole stress gauges, and physical and mechanical parameters of the rock mass are determined through indoor rock mechanics tests. Combined with historical mine seismic records, the dominant regional rupture mechanism and rupture surface dip angle are determined; ② Numerical framework construction: A three-dimensional geological grid is constructed based on geological exploration data. A fine grid of 0.5m×0.5m is used in the core influence area of ​​the seismic source (within a radius of 50m), and the grid size is gradually enlarged in the outer area to balance computational accuracy and efficiency; ③ Model calibration and verification: Input the field measured parameters, select 3-5 typical historical mine seismic events, calibrate sensitive parameters such as rupture propagation velocity, control the simulation error to ≤10%, and simultaneously build a two-way data interaction interface with the energy propagation model to realize real-time push of seismic source parameters.

[0028] The energy propagation path model (energy propagation model) is a three-dimensional geological-energy attenuation coupling model embedded with viscoelastic damage units. Its core objective is to establish a quantitative correlation between source energy, propagation distance, surrounding rock lithology, and peak point velocity (PPV) on the tunnel surface. By simulating the propagation and attenuation of energy in the strata through damage units, it accurately extracts PPV distribution data at key points in the tunnel.

[0029] Specific construction steps: ① Basic geological modeling: Based on borehole and geological profile data, construct a three-dimensional geological model containing lithological zones, faults, and fracture zones, and refine the mesh at rock strata interfaces and fracture development zones; ② Damage unit embedding: Embed viscoelastic damage units inside the model and set initial energy absorption coefficients according to different lithologies; ③ Attenuation law fitting: Input energy parameters output by the seismic source model to simulate the propagation and attenuation process of seismic energy along different paths; ④ Dynamic calibration and optimization: Input real-time measured PPV data from roadway sensors, iteratively correct damage unit parameters and attenuation coefficients to ensure that the deviation between simulated and measured values ​​is ≤8%, providing standardized energy input indicators for subsequent coupled analysis.

[0030] In addition, the roadway response model is a dual-module coupled model of "real-time sensor monitoring + support structure mechanical analysis". The monitoring module adopts a three-dimensional acceleration sensor module (each group contains 3 parallel accelerometers, sampling frequency 1000Hz, range 0-200m / s²) to collect the axial, horizontal radial and vertical radial vibration signals of the roadway in real time. The support mechanics module builds dedicated mechanical analysis models for the two mainstream underground support forms, namely anchor wire mesh and U-shaped support.

[0031] Specific construction steps: ① Standardized sensor deployment: Sensors are deployed at 5m intervals, with cross-point placement on the top slab and both sides, to build a real-time data acquisition and transmission terminal; ② Support mechanics modeling: Based on the actual support type on site, input the prestress of anchor bolts and cables, support spacing, cross-sectional parameters, etc., to build a corresponding support mechanics model; ③ Response correlation calibration: Simultaneously collect vibration signals, support deformation, and rod / support stress data to construct a correlation matrix between vibration response and structural stress; ④ Dynamic parameter correction: Combine with on-site macroscopic damage survey data to correct model interface parameters and yield threshold, control the quantification error to ≤10%, and achieve accurate connection between energy propagation results and support response.

[0032] In this embodiment, several sets of source release energies are determined based on the source release energy determination model; the peak particle velocity at different distances on the roadway surface is determined based on the pre-constructed energy propagation model according to the source release energy. Then, a multivariate nonlinear regression is performed on the roadway surface peak particle velocity determination formula based on each of the aforementioned roadway surface peak particle velocities to determine the initial value of the attenuation fitting coefficient; the roadway surface peak particle velocity determination formula is: ; Wherein, PPV is the peak particle velocity on the surface of the tunnel; The energy released by the seismic source; R is the propagation distance from the seismic source to the measuring point in the roadway; λ is the rock mass wave impedance; k, a, b, and c are attenuation fitting coefficients; d is the rock stratum interface correction coefficient; correspondingly, the determination of the peak particle velocity threshold of the roadway surface when roadway damage is initiated based on the initial value includes: determining the peak particle velocity of the roadway surface when sandstone begins to crack and / or mudstone begins to detach as the peak particle velocity threshold of the roadway surface through numerical simulation and field observation.

[0033] Specific construction steps: ① Sample data acquisition: Extract more than 50 sets of simulated PPV data with different source energy and propagation distance from the energy propagation model, and simultaneously measure the rock mass wave impedance and rock layer interface distribution at the corresponding locations; ② Initial model fitting: Use a multivariate nonlinear regression algorithm to fit and determine the initial values ​​of each coefficient; ③ Iterative calibration and optimization: Connect the PPV data measured by field sensors, and use the least squares method to iteratively correct the fitting coefficients until the simulation and measurement error is ≤8% and then stop the calibration; ④ Damage threshold determination: Combine the numerical simulation of the surrounding rock damage initiation value and the field measured critical value of crack appearance to determine the damage initiation PPV threshold for different lithologies such as sandstone, mudstone, and coal seam.

[0034] Step S12: Determine the target support mechanical model by using the peak particle velocity threshold of the roadway surface and the pre-constructed support mechanical model.

[0035] In this embodiment, an initial support mechanical model is constructed based on the formula for determining the ultimate vibration resistance of the support structure and the formula for determining the ultimate bending moment of the support; the formula for determining the ultimate vibration resistance of the support structure is: ; in, To support the ultimate vibration resistance of the structure, For anchor bolt and anchor cable prestress, , , These represent the elastic modulus, cross-sectional area, and anchorage length of the rod, respectively, with ΔL representing the vibration displacement. For the stiffness of the metal mesh, The area of ​​the net body subjected to force; The formula for determining the ultimate bending moment of the support is: ; in, The ultimate bending moment of the support. For the yield strength of the support steel, Where D is the cross-sectional modulus of the support frame, and D is the actual erection spacing. To design standard spacing.

[0036] Specifically, the anchor-mesh-cable support adopts a mechanical model of "rod-surrounding rock bonding-mesh collaborative load bearing," with the core expression being: ; in, To support the ultimate vibration resistance of the structure, For anchor bolt and anchor cable prestress, , , These are the elastic modulus, cross-sectional area, and anchorage length of the rod, respectively. This is the vibration displacement. For the stiffness of the metal mesh, The area of ​​the mesh subjected to force is set; bonding damage is initiated when the interface shear stress is greater than 2.5 MPa.

[0037] U-shaped support: Adopting a "beam-column rigid connection-surrounding rock contact" mechanical model, the core expression is: ; in, The ultimate bending moment of the support. For the yield strength of the support steel, Where D is the cross-sectional modulus of the support frame, and D is the actual erection spacing. To design the standard spacing, a plastic damage model was used to simulate the mechanical behavior of the stent after yielding.

[0038] Then, the ultimate vibration resistance of the target support structure or the ultimate bending moment of the target support is determined based on the product of the peak particle velocity threshold on the roadway surface and the preset conversion coefficient. The target vibration displacement and spacing difference are determined using the initial support mechanical model based on the ultimate vibration resistance of the target support structure or the ultimate bending moment of the target support. The support deformation and structural stress are determined based on the target vibration displacement and the spacing difference. The support deformation and structural stress are compared with the measured support deformation and measured structural stress, respectively. The initial support mechanical model is adjusted according to the comparison results to obtain the target support mechanical model. Specific construction steps: ① Calibrate the basic mechanical parameters of the model through indoor pull-out and bending tests; ② Collect vibration response, support deformation, and structural stress data simultaneously on-site; ③ Build a mechanical model corresponding to the support type and complete parameter calibration; ④ Verify the quantitative law between support parameters and structural response to provide a basis for determining the degree of damage.

[0039] Step S13: Based on the target support mechanical model, determine several support parameter vectors, and train the initial backpropagation neural network using each support parameter vector, the degree of damage, and the peak particle velocity on the roadway surface to obtain the target backpropagation neural network.

[0040] In this embodiment, several support parameter vectors are determined based on the target support mechanical model. The support parameter vectors and the peak particle velocity on the roadway surface are determined as inputs, and the degree of damage is determined as the output. The initial backpropagation neural network is trained based on the inputs and outputs to obtain the target backpropagation neural network. The support parameter vectors include the anchor bolt spacing and the anchor bolt and anchor cable prestress.

[0041] In this embodiment, a surface fitting model is constructed using "support design parameters - vibration response parameters - damage coefficient" as the three-dimensional coordinate axes: D = f(P, R); In the formula, D is the damage coefficient, P is the support parameter vector, and R is the vibration response parameter vector. The damage coefficient is uniformly divided into three levels: 0-0.3 (minor damage), 0.3-0.6 (moderate damage), and 0.6-1.0 (severe damage).

[0042] Specific construction steps: ① Sample library construction: Collect more than 300 sets of "support parameters-vibration response-on-site damage degree" samples. The damage degree is determined by a combination of macroscopic investigation and ultrasonic flaw detection; ② Model training: Use the BP neural network algorithm to train and fit the sample data to generate a three-dimensional correlation model; ③ Pattern verification: Quantitatively analyze the control pattern of parameters such as prestress and support spacing on the damage coefficient, directly supporting the optimization design of support scheme.

[0043] Specifically, historical data is collected. More than 300 samples are needed, each containing three things: the support parameters at the time (e.g., anchor spacing, prestress magnitude); the vibration response parameters at the time (PPV, frequency); and the actual degree of damage (determined through manual sketching and ultrasonic flaw detection). Model training: A backpropagation neural network (BP neural network) is built in the computer, using the first two items (support parameters, vibration parameters) as input and the degree of damage (damage coefficient D) as output, to generate a three-dimensional surface fitting model. Finally, the model is validated: it is checked whether the trained model conforms to common sense: for example, whether the trend of "the greater the prestress, the smaller the damage coefficient" is correct. Specifically, a closed-loop correction model of "BP neural network + sliding window iteration" is used, with the core correction formula being... , (In the formula: t is the number of iterations, Here, P is the indicator weight adjustment amount, P is the numerical model parameter vector, and η is the learning rate. (To assess the error gradient). Specific construction steps: ① Historical sample database construction: Collect mining seismic monitoring, simulation, and survey data from the past 3 years to construct 500 standardized samples; ② Neural network construction: Construct a 3-layer BP network, with the input layer matching 12 evaluation indicators + 5 core model parameters, the hidden layer having 10 neurons, and the output layer being the comprehensive evaluation index; ③ Sliding window iteration: Set 30 sets of data as the sliding window, and calculate the evaluation error for every 10 new sets of field verification data. When the error > 10%, parameter and weight correction is initiated; ④ Convergence determination: After 3 consecutive iterations, convergence is determined when the evaluation error ≤ 8%, ensuring the model adapts to different geological conditions and mining scenarios.

[0044] Step S14: Determine the degree of damage to the roadway based on the target backpropagation neural network, the current support parameter vector, and the current peak particle velocity on the roadway surface.

[0045] In this embodiment, the current support parameter vector and the current peak particle velocity on the roadway surface are input into the target backpropagation neural network to obtain the degree of damage to the roadway.

[0046] In addition, adopting a framework of "hierarchical indicator system + combined weighting + graded quantitative judgment", the comprehensive evaluation index calculation formula is as follows: ; In the formula, S is the comprehensive evaluation index, and W... i Let X be the weight of the i-th indicator combination. i This is the standardized value of the indicator.

[0047] Specific construction steps: ① Three-level indicator system construction: The target layer is the tunnel damage risk level; the criterion layer is divided into four categories: source characteristics, energy propagation, support response, and macroscopic damage; the indicator layer is refined into 12 quantifiable core indicators; ② Combined weight calculation: The subjective weight of the Analytic Hierarchy Process (AHP) and the objective weight of the entropy weight method are integrated. The formula for the combined weight is W. i =0.6W 主观 +0.4W 客观 ① Ensure consistency of judgment matrix CR < 0.1; ② Data standardization: Use the range method to standardize positive and negative indicators respectively to eliminate the influence of dimensions; ③ Level threshold determination: Use K-means clustering algorithm to cluster historical assessment data to determine the damage level threshold, which is completely matched with the damage coefficient classification of the three-dimensional correlation model and has no logical conflict.

[0048] This approach, with its deeply coupled, multi-scale three-level model and interconnected data, overcomes the limitations of single-scale analysis, enabling the quantification of the entire process from earthquake source occurrence to energy propagation, support response, and tunnel failure, significantly improving the comprehensiveness and accuracy of the assessment. Establishing a quantitative correlation between support parameters and the degree of damage allows for precise control of key parameters such as prestress and support spacing, improving the economic efficiency of the solution while ensuring support safety. Iterative correction of model parameters and weights through machine learning eliminates the reliance on pre-set samples in traditional simulations, adapting to various mining scenarios such as deep, high-stress, and complex geological conditions. Real-time monitoring, rapid assessment, and visualized early warning linkage enable rapid location of damaged areas and accurate determination of their severity, shortening emergency response time and improving disaster management efficiency.

[0049] In summary, this application first determines the peak particle velocity of each roadway surface at different distances and with different energy releases from earthquake sources based on a pre-constructed energy propagation model and a source energy determination model. Based on these peak particle velocities, an initial value for the attenuation fitting coefficient is determined, and a threshold for the peak particle velocity at the start of roadway damage is determined according to this initial value. A target support mechanics model is then determined using the peak particle velocity threshold and a pre-constructed support mechanics model. Several support parameter vectors are determined based on the target support mechanics model. An initial backpropagation neural network is trained using these support parameter vectors, the degree of damage, and the peak particle velocity of the roadway surface to obtain a target backpropagation neural network. Finally, the degree of roadway damage is determined based on the target backpropagation neural network, the current support parameter vector, and the current peak particle velocity of the roadway surface. Therefore, this application, through deep coupling of multi-scale models and data interoperability, breaks through the limitations of single-scale analysis, achieving full-process quantification of earthquake source occurrence—energy propagation—support response—roadway damage, significantly improving the comprehensiveness and accuracy of the assessment. It can be widely adapted to various coal mine underground roadways for mine seismic safety monitoring systems, especially for roadway stability assessment projects in deep high-stress mines and mines with rock burst hazards, providing precise technical support for roadway safety assurance in complex geological and mining environments.

[0050] Meanwhile, this application adopts a standardized process of "three-dimensional benchmarking + error quantification + judgment correction" to ensure that the verification dimensions correspond one-to-one with the core parameters of the model, eliminating verification deviations. Specific verification details are as follows: ① Source parameter benchmarking: The relative error between the sensor-inverted source energy and the simulated source energy is ≤12%; ② Energy propagation benchmarking: The error between the simulated PPV value and the measured PPV value is ≤8% in sandstone areas and ≤10% in mudstone areas; ③ Damage degree benchmarking: The consistency between the simulated damage coefficient and the macroscopic survey damage level is ≥90%.

[0051] Specific implementation steps: Within one hour of the occurrence of the mine earthquake event, three types of data are collected simultaneously: sensor monitoring, numerical simulation, and macroscopic survey; the deviation of each dimension is calculated using the relative error formula; if a single error exceeds the standard, the corresponding model parameters are corrected by backtracking; if all three types of indicators meet the standard, the assessment result is deemed reliable.

[0052] Furthermore, visualization is achieved using a multi-layer overlay and dynamic interactive mode based on professional mining platforms such as Surfer, Matlab, or 3DMine.

[0053] Specific implementation steps: ① Data preprocessing: Spatiotemporally align the three types of data, unify the mine coordinate system, and remove abnormal and invalid data; ② Basic layer construction: Draw geological background layers, roadway outline layers, and sensor deployment point layers; ③ Core result visualization: Use the Kriging spatial interpolation algorithm to generate seismic intensity contour lines and roadway damage heat maps, with grading and color matching completely consistent with the assessment level, and mark the epicenter location and key damage points; ④ Interactive function development: Implement view zooming, data filtering, historical backtracking, and result export functions, link with the emergency decision-making system, highlight high-risk areas, and provide intuitive support for on-site disposal.

[0054] In addition, the following equivalent alternatives can be adopted for different mine equipment conditions and rock mass characteristics without affecting the core technology effect: 1. Numerical simulation method replacement: The core numerical calculation can be replaced by the discrete element method and the boundary element method, which can be adapted to special rock mass mechanical characteristics scenarios and can still achieve dynamic coupling calibration with monitoring data.

[0055] 2. Alternative weighting methods: Combinatorial weighting can use entropy weighting or principal component analysis to replace the analytic hierarchy process. This allows for automatic allocation of indicator weights through data-driven approaches, adapting to different data quality and volume scenarios.

[0056] 3. Algorithm replacement: Dynamic correction can use genetic algorithms or support vector machines to replace BP neural networks, so as to achieve adaptive correction of model parameters and balance evaluation accuracy and computational efficiency.

[0057] See Figure 2 As shown in the figure, an embodiment of the present invention discloses a mine roadway damage monitoring device, comprising: The threshold determination module 11 is used to determine the peak particle velocity of each roadway surface at different distances and different sources of energy release based on a pre-built energy propagation model and a source energy release determination model, determine the initial value of the attenuation fitting coefficient based on the peak particle velocity of each roadway surface, and determine the peak particle velocity threshold of the roadway surface when roadway damage is initiated based on the initial value. The model determination module 12 is used to determine the target support mechanical model by using the peak particle velocity threshold of the roadway surface and the pre-constructed support mechanical model. Training module 13 is used to determine several support parameter vectors based on the target support mechanical model, and to train the initial backpropagation neural network using each support parameter vector, the degree of damage, and the peak particle velocity on the roadway surface to obtain the target backpropagation neural network. The damage determination module 14 is used to determine the damage degree of the roadway based on the target backpropagation neural network, the current support parameter vector, and the current peak particle velocity on the roadway surface.

[0058] In summary, this application first determines the peak particle velocity of each roadway surface at different distances and with different energy releases from earthquake sources based on a pre-constructed energy propagation model and a source energy determination model. Based on these peak particle velocities, an initial value for the attenuation fitting coefficient is determined, and a threshold for the peak particle velocity at the start of roadway damage is determined according to this initial value. A target support mechanics model is then determined using the peak particle velocity threshold and a pre-constructed support mechanics model. Several support parameter vectors are determined based on the target support mechanics model. An initial backpropagation neural network is trained using these support parameter vectors, the degree of damage, and the peak particle velocity of the roadway surface to obtain a target backpropagation neural network. Finally, the degree of roadway damage is determined based on the target backpropagation neural network, the current support parameter vector, and the current peak particle velocity of the roadway surface. Therefore, this application, through deep coupling of multi-scale models and data interoperability, breaks through the limitations of single-scale analysis, achieving full-process quantification of earthquake source occurrence—energy propagation—support response—roadway damage, significantly improving the comprehensiveness and accuracy of the assessment. It can be widely adapted to various coal mine underground roadways for mine seismic safety monitoring systems, especially for roadway stability assessment projects in deep high-stress mines and mines with rock burst hazards, providing precise technical support for roadway safety assurance in complex geological and mining environments.

[0059] In some specific embodiments, the device can also be used to determine the stress of coal and rock mass measured in the field using a stress gauge, determine the physical and mechanical parameters of the rock mass based on indoor rock mechanics tests, and determine the dip angle, fracture propagation rate, and fracture length based on historical mine seismic records; the physical and mechanical parameters of the rock mass include the friction coefficient of the weak surface of the rock mass; and determine the source energy release determination model based on the source energy release determination formula; the source energy release determination formula is: ; in, The seismic source release energy is defined by σ, the stress in the coal and rock mass, μ, the friction coefficient of the weak surface of the rock mass, α, the dip angle of the fracture surface, V, the fracture propagation velocity, and L, the length of the fracture surface extension. A digital twin of the rock mass is determined based on the seismic source release energy determination model using three-dimensional finite element or discrete element methods. The spatial coordinates of the seismic source, the energy level of the seismic source release, and the fracture extent are determined using the digital twin according to the Mohr-Coulomb criterion and the maximum tensile stress criterion. A three-dimensional geological body containing faults and strata boundaries is established based on borehole and geological exploration data. The initial energy absorption coefficient of each unit in the three-dimensional geological body is set according to the lithology. The target tunnel is determined using the current three-dimensional geological body based on the spatial coordinates of the seismic source, the energy level of the seismic source release, the fracture extent, and the energy absorption coefficient. Peak particle velocity on the surface of the target roadway; compare the peak particle velocity on the target roadway surface with the actual peak particle velocity on the roadway surface measured by downhole sensors; adjust the initial energy absorption coefficient of the three-dimensional geological body according to the comparison results to obtain the adjusted three-dimensional geological body; determine the adjusted three-dimensional geological body as the current three-dimensional geological body, and jump back to the step of determining the peak particle velocity on the target roadway surface based on the spatial coordinates of the seismic source, the energy level released by the seismic source, the rupture range, and the energy absorption coefficient of the current three-dimensional geological body; when the difference between the peak particle velocity on the target roadway surface and the actual peak particle velocity on the roadway surface meets the preset conditions, the current three-dimensional geological body is determined as the energy propagation model.

[0060] In some specific embodiments, the threshold determination module 11 can be used to determine several sets of source release energies based on the source release energy determination model; and to determine the peak particle velocity of the roadway surface at different distances based on the source release energy according to the pre-constructed energy propagation model.

[0061] In some specific embodiments, the threshold determination module 11 can be used to perform a multivariate nonlinear regression on the peak particle velocity determination formula for each of the roadway surfaces based on the peak particle velocity of each roadway surface, in order to determine the initial value of the attenuation fitting coefficient; the peak particle velocity determination formula for the roadway surface is: ; Wherein, PPV is the peak particle velocity on the surface of the tunnel; The energy released by the seismic source is R; the propagation distance from the seismic source to the measuring point in the tunnel is λ; the rock mass wave impedance is k, a, b, and c are attenuation fitting coefficients; d is the rock layer interface correction coefficient; the peak particle velocity on the tunnel surface when sandstone begins to crack and / or mudstone begins to detach is determined as the peak particle velocity threshold on the tunnel surface through numerical simulation and field observation.

[0062] In some specific embodiments, the device can also be used to construct an initial support mechanical model based on the formula for determining the ultimate vibration resistance of the support structure and the formula for determining the ultimate bending moment of the support; the formula for determining the ultimate vibration resistance of the support structure is: ; in, To support the ultimate vibration resistance of the structure, For anchor bolt and anchor cable prestress, , , These represent the elastic modulus, cross-sectional area, and anchorage length of the rod, respectively, with ΔL representing the vibration displacement. For the stiffness of the metal mesh, The area of ​​the net body under stress; the formula for determining the ultimate bending moment of the support is: ; in, The ultimate bending moment of the support. For the yield strength of the support steel, Where D is the cross-sectional modulus of the support frame, and D is the actual erection spacing. To design standard spacing.

[0063] In some specific embodiments, the model determination module 12 can be used to determine the ultimate vibration resistance of the target support structure or the ultimate bending moment of the target support based on the product of the peak mass velocity threshold of the roadway surface and a preset conversion coefficient; determine the target vibration displacement and spacing difference based on the ultimate vibration resistance of the target support structure or the ultimate bending moment of the target support using the initial support mechanical model; determine the support deformation and structural stress based on the target vibration displacement and the spacing difference; compare the support deformation and structural stress with the measured support deformation and measured structural stress respectively, and adjust the initial support mechanical model according to the corresponding comparison results to obtain the target support mechanical model.

[0064] In some specific embodiments, the training module 13 can be used to determine the support parameter vector and the peak particle velocity on the roadway surface as inputs, and the degree of damage as outputs, and to train the initial backpropagation neural network based on the inputs and outputs to obtain the target backpropagation neural network; the support parameter vector includes the anchor bolt spacing and the anchor bolt and anchor cable prestress.

[0065] Furthermore, embodiments of this application also disclose an electronic device, Figure 3 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application.

[0066] Figure 3 This is a schematic diagram of the structure of an electronic device 20 provided in an embodiment of this application. Specifically, the electronic device 20 may include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the mine roadway damage monitoring method disclosed in any of the foregoing embodiments. Furthermore, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0067] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0068] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon can include operating system 221, computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0069] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the mine roadway damage monitoring method executed by the electronic device 20 as disclosed in any of the foregoing embodiments, the computer program 222 may further include computer programs capable of performing other specific tasks.

[0070] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned method for monitoring the damage state of mine roadways. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.

[0071] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0072] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0073] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0074] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0075] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for monitoring the damage state of mine roadways, characterized in that, include: Based on the pre-constructed energy propagation model and the source energy release determination model, the peak particle velocity of each roadway surface at different source release energies and distances is determined. Based on the peak particle velocity of each roadway surface, the initial value of the attenuation fitting coefficient is determined. Based on the initial value, the peak particle velocity threshold of the roadway surface at the start of roadway damage is determined. The target support mechanical model is determined by the peak particle velocity threshold on the roadway surface and the pre-constructed support mechanical model. Based on the target support mechanical model, several support parameter vectors are determined. The initial backpropagation neural network is trained using each of the support parameter vectors, the degree of damage, and the peak particle velocity on the roadway surface to obtain the target backpropagation neural network. The degree of damage to the roadway is determined based on the target backpropagation neural network, the current support parameter vector, and the current peak particle velocity on the roadway surface. The initial value for determining the attenuation fitting coefficient based on the peak particle velocity of each of the roadways includes: Based on the peak particle velocity at the surface of each of the aforementioned roadways, a multivariate nonlinear regression is performed on the formula for determining the peak particle velocity at the roadway surface to determine the initial value of the attenuation fitting coefficient; the formula for determining the peak particle velocity at the roadway surface is: ； Wherein, PPV is the peak particle velocity on the surface of the tunnel; The energy released from the seismic source; R is the propagation distance from the seismic source to the measuring point in the tunnel; λ is the rock mass wave impedance; k, a, b, and c are attenuation fitting coefficients; d is the rock stratum interface correction coefficient; Accordingly, determining the peak particle velocity threshold of the roadway surface at the time of roadway damage initiation based on the initial value includes: The peak particle velocity at the roadway surface when sandstone begins to crack and / or mudstone begins to detach was determined as the peak particle velocity threshold at the roadway surface through numerical simulation and field observation. Before determining the target support mechanical model using the peak particle velocity threshold on the roadway surface and the pre-built support mechanical model, the process also includes: An initial support mechanical model is constructed based on the formulas for determining the ultimate vibration resistance of the support structure and the ultimate bending moment of the support; the formula for determining the ultimate vibration resistance of the support structure is: ； in, To support the ultimate vibration resistance of the structure, For anchor bolt and anchor cable prestress, , , These represent the elastic modulus, cross-sectional area, and anchorage length of the rod, respectively, with ΔL representing the vibration displacement. For the stiffness of the metal mesh, The area of ​​the net body subjected to force; The formula for determining the ultimate bending moment of the support is: ； in, The ultimate bending moment of the support. For the yield strength of the support steel, Where D is the cross-sectional modulus of the support frame, and D is the actual erection spacing. To design standard spacing; The determination of the target support mechanical model through the peak particle velocity threshold on the roadway surface and the pre-constructed support mechanical model includes: The ultimate vibration resistance of the target support structure or the ultimate bending moment of the target support is determined by multiplying the peak particle velocity threshold on the roadway surface with a preset conversion coefficient. The target vibration displacement and spacing difference are determined by the initial support mechanical model based on the ultimate vibration resistance of the target support structure or the ultimate bending moment of the target support. The support deformation and structural stress are determined based on the target vibration displacement and the spacing difference. The support deformation and structural stress are compared with the measured support deformation and measured structural stress, respectively. The initial support mechanical model is adjusted according to the comparison results to obtain the target support mechanical model.

2. The method for monitoring the damage state of mine roadways according to claim 1, characterized in that, Before determining the peak particle velocity at different roadway surfaces at different distances based on the pre-built energy propagation model and the source energy determination model, the process also includes: The stress in the coal and rock mass measured in the field is determined by stress gauges, and the physical and mechanical parameters of the rock mass are determined based on indoor rock mechanics tests. The dip angle, fracture propagation rate, and fracture length are determined based on historical mine seismic records. The physical and mechanical parameters of the rock mass include the friction coefficient of the weak surface of the rock mass. A model for determining the energy released from the seismic source is determined based on the formula for determining the energy released from the seismic source; the formula for determining the energy released from the seismic source is: ； in, The energy released by the seismic source is σ, the stress of the coal and rock mass is μ, the friction coefficient of the weak surface of the rock mass is α, the dip angle of the fracture surface is V, the fracture propagation rate is L, and the length of the fracture surface is L. The model for determining the energy released from the seismic source is based on three-dimensional finite element or discrete element methods to determine the digital twin of the rock mass; The spatial coordinates of the seismic source, the energy level released by the seismic source, and the rupture range are determined using the digital twin based on the Mohr-Coulomb criterion and the maximum tensile stress criterion. Based on borehole and geological exploration data, a three-dimensional geological body including faults and rock strata boundaries was established; The initial energy absorption coefficient of each unit in the three-dimensional geological body is determined based on the lithology; The peak particle velocity on the surface of the target tunnel is determined by the current three-dimensional geological body based on the spatial coordinates of the seismic source, the energy level released by the seismic source, the rupture range, and the energy absorption coefficient. The peak particle velocity of the target roadway surface is compared with the actual peak particle velocity of the roadway surface measured by downhole sensors; Based on the corresponding comparison results, the initial energy absorption coefficient of the three-dimensional geological body is adjusted to obtain the adjusted three-dimensional geological body. The adjusted three-dimensional geological body is then determined as the current three-dimensional geological body, and the process jumps back to the step of determining the peak particle velocity of the target tunnel surface based on the spatial coordinates of the seismic source, the energy level released by the seismic source, the rupture range, and the energy absorption coefficient of the current three-dimensional geological body. When the difference between the peak particle velocity of the target tunnel surface and the actual peak particle velocity of the tunnel surface meets the preset conditions, the current three-dimensional geological body is determined as the energy propagation model.

3. The method for monitoring the damage state of mine roadways according to claim 1, characterized in that, The determination of peak particle velocities on the surface of each roadway at different distances based on the pre-built energy propagation model and the source energy release determination model includes: Several sets of earthquake source release energies are determined based on the model for determining the energy released from the earthquake source. Based on a pre-built energy propagation model, the peak particle velocity on the roadway surface at different distances is determined according to the energy released by each seismic source.

4. The method for monitoring the damage state of mine roadways according to any one of claims 1 to 3, characterized in that, The step of training the initial backpropagation neural network using the support parameter vectors, the degree of damage, and the peak particle velocity on the roadway surface to obtain the target backpropagation neural network includes: The support parameter vector and the peak particle velocity on the roadway surface are determined as inputs, and the degree of damage is determined as output. The initial backpropagation neural network is trained based on the inputs and outputs to obtain the target backpropagation neural network. The support parameter vector includes the anchor bolt spacing and the anchor bolt and anchor cable prestress.

5. A device for monitoring the damage status of mine roadways, characterized in that, include: The threshold determination module is used to determine the peak particle velocity of each roadway surface at different distances and different sources of energy release based on a pre-built energy propagation model and a source energy release determination model. Based on the peak particle velocity of each roadway surface, the module determines the initial value of the attenuation fitting coefficient and determines the peak particle velocity threshold of the roadway surface when roadway damage is initiated based on the initial value. The model determination module is used to determine the target support mechanical model by using the peak particle velocity threshold of the roadway surface and the pre-constructed support mechanical model; The training module is used to determine several support parameter vectors based on the target support mechanical model, and to train the initial backpropagation neural network using each of the support parameter vectors, the degree of damage, and the peak particle velocity on the roadway surface, so as to obtain the target backpropagation neural network. The damage degree determination module is used to determine the damage degree of the roadway based on the target backpropagation neural network, the current support parameter vector, and the current peak particle velocity on the roadway surface. The threshold determination module is used to perform a multivariate nonlinear regression on the peak particle velocity determination formula for each of the roadway surfaces based on the peak particle velocity of each roadway surface, in order to determine the initial value of the attenuation fitting coefficient; the peak particle velocity determination formula for the roadway surface is: ； Wherein, PPV is the peak particle velocity on the surface of the tunnel; The energy released by the seismic source; R is the propagation distance from the seismic source to the measuring point in the tunnel; λ is the rock mass wave impedance; k, a, b, and c are attenuation fitting coefficients; d is the rock layer interface correction coefficient; the peak particle velocity on the tunnel surface when sandstone begins to crack and / or mudstone begins to detach is determined as the peak particle velocity threshold on the tunnel surface through numerical simulation and field observation. The device is also used to construct an initial support mechanical model based on the formula for determining the ultimate vibration resistance of the support structure and the formula for determining the ultimate bending moment of the support; the formula for determining the ultimate vibration resistance of the support structure is: ； in, To support the ultimate vibration resistance of the structure, For anchor bolt and anchor cable prestress, , , These represent the elastic modulus, cross-sectional area, and anchorage length of the rod, respectively, with ΔL representing the vibration displacement. For the stiffness of the metal mesh, The area of ​​the net body subjected to force; The formula for determining the ultimate bending moment of the support is: ； in, The ultimate bending moment of the support. For the yield strength of the support steel, Where D is the cross-sectional modulus of the support frame, and D is the actual erection spacing. To design standard spacing; The model determination module is used to determine the ultimate vibration resistance of the target support structure or the ultimate bending moment of the target support based on the product of the peak mass velocity threshold on the roadway surface and a preset conversion coefficient; to determine the target vibration displacement and spacing difference based on the ultimate vibration resistance of the target support structure or the ultimate bending moment of the target support using the initial support mechanical model; to determine the support deformation and structural stress based on the target vibration displacement and the spacing difference; to compare the support deformation and structural stress with the measured support deformation and measured structural stress respectively, and to adjust the initial support mechanical model according to the comparison results to obtain the target support mechanical model.

6. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing a computer program to implement the steps of the method for monitoring the damage status of mine roadways as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, A computer program is stored on a computer-readable storage medium, and when executed by a processor, the computer program implements the steps of the method for monitoring the damage status of mine roadways as described in any one of claims 1 to 4.

Images