Coal mine disaster data analysis and prediction system based on three-component micro-motion detection

Abstract

The invention provides a coal mine disaster data analysis and prediction system based on three-component micro-motion detection, and the system comprises three mutually vertical micro-motion sensors which are respectively used for collecting micro-motion signals in the horizontal north-south direction, the horizontal east-west direction and the vertical direction; the three mutually vertical micro-motion sensors are arranged in a to-be-detected area of a coal mine according to a preset grid in a measuring point manner; preprocessing the initial three-component micro-motion signal acquired by the micro-motion sensor to obtain a preprocessed three-component micro-motion signal; determining a micro-motion frequency dispersion curve according to the preprocessed three-component micro-motion signal; performing inversion calculation based on the micro-motion dispersion curve, and determining stratum velocities at different depths; performing speed difference analysis based on the formation speed to obtain the formation speed of the local area, and determining a potential disaster area based on the formation speed of the local area; disaster prediction is carried out based on coal mining historical data and the geological structure map, and potential disaster types are obtained. According to the invention, the accuracy and reliability of coal mine disaster prediction are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular, to a coal mine disaster data analysis and prediction system based on three-component microtremor detection. Background Art

[0002] At present, the common disaster prediction methods in the process of coal mine exploitation mostly rely on the monitoring of a single physical quantity. For example, the possibility of disasters is judged by monitoring parameters such as roadway stress and gas concentration. However, these methods have a significant drawback, that is, they cannot accurately locate the specific location where the disaster occurs. Due to the complex geological structure underground in coal mines, the monitoring of a single physical quantity is easily interfered, and the monitoring data can only reflect the overall state of a certain area, making it difficult to accurately lock the disaster source. When abnormal data appears, it is difficult for the staff to quickly judge the specific location of the disaster, seriously affecting the accuracy and reliability of coal mine disaster prediction. Therefore, once the prediction is incorrect, it may lead to serious safety accidents, threatening the lives of miners and the normal production and operation of coal mines. Summary of the Invention

[0003] The present invention provides a coal mine disaster data analysis and prediction system based on three-component microtremor detection to improve the accuracy and reliability of coal mine disaster prediction.

[0004] In a first aspect, the present invention provides a coal mine disaster data analysis and prediction system based on three-component microtremor detection. The system includes three mutually perpendicular microtremor sensors, which are respectively used to collect microtremor signals in the horizontal north-south direction, horizontal east-west direction, and vertical direction; and the three mutually perpendicular microtremor sensors are arranged at measurement points according to a preset grid in the coal mine area to be detected. The system includes:

[0005] A data analysis module, which is used to determine a filtering frequency band based on the coal mine geological conditions in the coal mine area to be detected, and preprocess the initial three-component microtremor signals collected by the microtremor sensors at the grid measurement points corresponding to the coal mine area to be detected based on the filtering frequency band to obtain preprocessed three-component microtremor signals;

[0006] A spectrum analysis module, which is used to perform spectrum analysis on the preprocessed three-component microtremor signals of each grid measurement point, and obtain the microtremor dispersion curves of each grid measurement point according to the microtremor signal energy distribution at different frequencies;

[0007] An inversion calculation module, which is used to perform inversion calculation on the microtremor dispersion curves of each grid measurement point based on the layered medium theory to determine the formation velocities at different depths of each grid measurement point; the layered medium theory simplifies the formation into a multi-layer homogeneous medium model;

[0008] A disaster analysis module, configured to perform velocity difference analysis based on the formation velocities of each grid measurement point, obtain the formation velocities of a local area, and determine a potential disaster area based on the formation velocities of the local area;

[0009] A disaster prediction module, configured to perform disaster prediction on the potential disaster area based on the coal mining historical data and the geological structure map, and obtain the potential disaster types of the potential disaster area.

[0010] In a second aspect, the present invention further provides a method for analyzing and predicting coal mine disaster data based on three-component microtremor detection, which is implemented based on the system for analyzing and predicting coal mine disaster data based on three-component microtremor detection according to any one of the first aspects. The method includes:

[0011] Determine a filtering frequency band based on the coal mine geological conditions of the coal mine area to be detected, and perform preprocessing on the initial three-component microtremor signals collected by the microtremor sensors in the grid measurement points corresponding to the coal mine area to be detected based on the filtering frequency band to obtain preprocessed three-component microtremor signals;

[0012] Perform spectral analysis on the preprocessed three-component microtremor signals of each grid measurement point, and obtain the microtremor dispersion curves of each grid measurement point according to the energy distribution of the microtremor signals at different frequencies;

[0013] Perform inversion calculation on the microtremor dispersion curves of each grid measurement point based on the layered medium theory to determine the formation velocities of each grid measurement point at different depths; the layered medium theory simplifies the formation into a multi-layer homogeneous medium model;

[0014] Perform velocity difference analysis based on the formation velocities of each grid measurement point, obtain the formation velocities of a local area, and determine a potential disaster area based on the formation velocities of the local area;

[0015] Perform disaster prediction on the potential disaster area based on the coal mining historical data and the geological structure map, and obtain the potential disaster types of the potential disaster area.

[0016] The present invention further provides an electronic device, including: a memory, configured to store a computer software program; a processor, configured to read and execute the computer software program, and further implement the method for analyzing and predicting coal mine disaster data based on three-component microtremor detection as described above.

[0017] The present invention further provides a non-transitory computer-readable storage medium, in which a computer software program is stored, and when the computer software program is executed by a processor, the method for analyzing and predicting coal mine disaster data based on three-component microtremor detection as described above is implemented.

[0018] The present invention also provides a computer program product, including a computer program which, when executed by a processor, implements the method for analyzing and predicting coal mine disasters based on three-component microtremor detection as described above.

[0019] In the coal mine disaster data analysis and prediction system based on three-component microtremor detection provided by the embodiments of the present invention, the multi-directional signals obtained by three-component microtremor detection contain more information about the changes in underground media compared with single physical quantity monitoring, and can more accurately reflect the changes in the physical properties of the disaster occurrence area. In velocity inversion, based on the layered medium theory, the microtremor dispersion curves of each grid measurement point are inversely calculated, and the underground velocity can be inversely calculated more precisely. Compared with the existing methods, the understanding of the underground structure is clearer, so that the abnormal area can be located more accurately. By quantifying the formation velocity of a local area through the velocity difference of the formation velocity, it is possible to quantitatively judge the potential disaster area, that is, the location where the disaster occurs, based on the formation velocity of the local area, avoiding the problem of inaccurate positioning caused by single and fuzzy data in the existing methods. Therefore, the location of coal mine disasters can be accurately located, and the accuracy and reliability of coal mine disaster prediction are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a schematic structural diagram of the coal mine disaster data analysis and prediction system based on three-component microtremor detection provided by the present invention;

[0021] Figure 2 is a schematic flow diagram of the method for analyzing and predicting coal mine disasters based on three-component microtremor detection provided by the present invention;

[0022] Figure 3 is a schematic diagram of an embodiment of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present invention.

[0024] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, "a plurality of" means two or more, unless otherwise specifically defined.

[0025] In the description of the present invention, the term "for example" is used to mean "serving as an example, illustration, or explanation". Any embodiment described as "for example" in the present invention is not necessarily construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the present invention. In the following description, details are set forth for purposes of explanation. It should be understood that those of ordinary skill in the art can recognize that the present invention can be implemented without the use of these specific details. In other instances, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but rather to be in line with the broadest scope consistent with the principles and features disclosed in the present invention.

[0026] Optionally, referring to Figure 1 , Figure 1 FIG. Figure 1 is a schematic structural diagram of a coal mine disaster data analysis and prediction system based on three-component microtremor detection provided by the present invention. The coal mine disaster data analysis and prediction system based on three-component microtremor detection includes three mutually perpendicular microtremor sensors, which are respectively used to collect microtremor signals in the horizontal north-south direction, horizontal east-west direction, and vertical direction; and the three mutually perpendicular microtremor sensors are arranged at measurement points according to a preset grid in the coal mine area to be detected. Further, the system in the embodiment of the present invention further includes a data analysis module, a spectrum analysis module, an inversion calculation module, a disaster analysis module, and a disaster prediction module.

[0027] Optionally, the data analysis module is used to determine a filtering frequency band based on the coal mine geological conditions in the coal mine area to be detected, and preprocess the initial three-component microtremor signals collected by the microtremor sensors at the grid measurement points corresponding to the coal mine area to be detected based on the filtering frequency band to obtain preprocessed three-component microtremor signals.

[0028] Specifically: The data analysis module analyzes the coal mine geological conditions in the coal mine area to be detected, where the geological conditions include, but are not limited to, information such as rock type, geological structure (faults, folds, etc.), coal seam depth and thickness. Further, the data analysis module determines the filtering frequency band of the coal mine area to be detected according to the geological conditions, as specifically described in steps 101 to 104. Under different geological conditions, the effective frequency components of the microtremor signals are different, and a suitable filtering frequency band can remove noise interference and retain the effective signals. Therefore, the data analysis module preprocesses the initial three-component microtremor signals collected by the microtremor sensors at the grid measurement points corresponding to the coal mine area to be detected according to the filtering frequency band, where the preprocessing process is to remove noise signals that do not match the filtering frequency band to obtain preprocessed three-component microtremor signals.

[0029] In one embodiment, in a coal mine area to be detected, the geological exploration report shows that there are many fault structures in this area. The rock types are mainly sandstone and shale, and the average burial depth of the coal seam is 500 meters. According to the above conditions, it is determined that the microtremor signals with frequencies in the range of 5 - 20 Hz can better reflect the formation structure information and are less affected by noise. Therefore, the filtering frequency band is 5 - 20 Hz. Among the grid measurement points arranged in this area, the microtremor sensor collects the initial three-component microtremor signals, and the initial three-component microtremor signals are filtered through a band-pass filter with a frequency range of 5 - 20 Hz to remove the high-frequency and low-frequency noise signals, obtaining the preprocessed three-component microtremor signals.

[0030] Optionally, the spectrum analysis module is used to perform spectrum analysis on the preprocessed three-component microtremor signals of each grid measurement point, and obtain the microtremor dispersion curves of each grid measurement point according to the energy distribution of the microtremor signals at different frequencies.

[0031] Specifically: For the preprocessed three-component microtremor signals of each grid measurement point, the spectrum analysis module uses the fast Fourier transform (FFT) method to convert the microtremor signals in the time domain to the frequency domain, and obtains the signal amplitudes at different frequencies. Calculate the energy distribution according to the amplitudes of the microtremor signals at different frequencies. For example, for each frequency point, calculate the energy of this frequency point through the square of the signal amplitude and the sampling interval.

[0032] Furthermore, the spectrum analysis module correlates the energy of each frequency point with the corresponding frequency, and draws the microtremor dispersion curves of each grid measurement point. Among them, the microtremor dispersion curves reflect the propagation characteristics of the microtremor signals at different frequencies, as specifically described in steps 201 to 205.

[0033] Continuing with the above coal mine area as an example, perform a fast Fourier transform on the preprocessed three-component microtremor signals of each grid measurement point, convert the signals from the time domain to the frequency domain, and obtain the signal amplitudes corresponding to each frequency point. For example, at one of the grid measurement points, after calculation, the signal amplitude is 0.5 at a frequency of 8 Hz. Calculate the energy of this frequency point as 0.0025 according to the energy calculation formula. And so on, calculate the energy of each frequency point, and then correlate the frequency and the energy, and draw the microtremor dispersion curve of this grid measurement point, showing the distribution of the microtremor signal energy with frequency at this measurement point.

[0034] Optionally, the inversion calculation module performs inversion calculation on the microtremor dispersion curves of each grid measurement point based on the layered medium theory, and determines the formation velocities at different depths of each grid measurement point; the layered medium theory simplifies the formation into a multi-layer homogeneous medium model.

[0035] Specifically: The inversion calculation module simplifies the strata into a multi-layer homogeneous medium model based on the layered medium theory. For the microtremor dispersion curves of each grid measurement point, an inversion calculation method is adopted. Among them, the inversion calculation is usually an iterative optimization process. First, a set of initial formation parameters (such as the thickness and velocity of each layer) are set, and the theoretical microtremor dispersion curve is calculated according to the layered medium theory. Then, the theoretical dispersion curve is compared with the actually measured microtremor dispersion curve, and the error between the two is calculated. Through optimization algorithms (such as genetic algorithms, particle swarm algorithms, etc.), the formation parameters are continuously adjusted to minimize the error between the theoretical dispersion curve and the actual dispersion curve. When the error meets certain accuracy requirements, the obtained formation parameters are the formation velocities of each grid measurement point at different depths, as specifically described in steps 301 to 304.

[0036] For example, at a certain grid measurement point in the above-mentioned coal mine, for example, the strata are divided into 3 layers, and the initial thicknesses of each layer are 100 meters, 150 meters, and 250 meters respectively, and the initial velocities are 1500 m / s, 2000 m / s, and 2500 m / s respectively. According to the layered medium theory, the theoretical microtremor dispersion curve is calculated and compared with the actually measured microtremor dispersion curve, and it is found that the error is relatively large. The genetic algorithm is used, and after multiple iterations, the thickness and velocity parameters of each layer are continuously adjusted. For example, after 10 iterations, when the thicknesses of each layer are 110 meters, 140 meters, and 250 meters respectively, and the velocities are 1400 m / s, 2100 m / s, and 2600 m / s respectively, the error between the theoretical dispersion curve and the actual dispersion curve meets the accuracy requirements, and the formation velocities of this grid measurement point at different depths are determined.

[0037] Optionally, the disaster analysis module performs velocity difference analysis based on the formation velocities of each grid measurement point to obtain the formation velocities of local areas, and determines potential disaster areas based on the formation velocities of local areas.

[0038] Specifically: The disaster analysis module analyzes the formation velocities of each grid measurement point, and determines the formation velocity characteristics of local areas by comparing the formation velocity differences between adjacent grid measurement points. A velocity difference threshold is set. When the formation velocity difference between adjacent grid measurement points exceeds this threshold, it is considered that there is a velocity anomaly in this area. The grid measurement points with velocity anomalies are subjected to cluster analysis to divide local areas.

[0039] Furthermore, the disaster analysis module judges the areas that may have disaster hidden dangers, that is, potential disaster areas, according to the formation velocity distribution of local areas, combined with the actual needs and potential risks of coal mine mining. For example, areas where the formation velocity suddenly decreases may have hidden dangers such as cavities and fracture zones, which are likely to cause disasters such as roof caving and water inrush, as specifically described in steps 401 to 403.

[0040] For example, in the above-mentioned coal mine area, the speed difference threshold is 200 m / s. When analyzing the formation speed of grid measurement points, it is found that the formation speed of 5 adjacent grid measurement points in a certain area suddenly drops from 2000 m / s to 1800 m / s, exceeding the set threshold. These 5 grid measurement points are clustered and determined as a local area. Further analyzing the formation speed distribution of this local area and combining with the coal mine mining situation, it is judged that there may be a risk of roof collapse in this area due to ineffective treatment of goafs, and it is determined as a potential disaster area.

[0041] Optionally, the disaster prediction module performs disaster prediction on the potential disaster area based on the coal mine mining historical data and the geological structure map to obtain the potential disaster types of the potential disaster area.

[0042] Specifically: The disaster prediction module collects coal mine mining historical data, including information such as the coal seams mined, mining time, mining methods, and records of disasters that have occurred. At the same time, it obtains a detailed geological structure map to understand the geological structure characteristics of the potential disaster area, such as the distribution, strike, and dip angle of faults.

[0043] Furthermore, the disaster prediction module combines the formation speed characteristics of the potential disaster area with the mining historical data and the geological structure map, and uses the disaster prediction model for analysis. For example, according to the change in the formation speed near the fault, combined with the record of water inrush disasters that occurred in the fault area in the mining history, it is predicted that a water inrush disaster may occur in this potential disaster area; according to the change in the formation speed after coal seam mining and the mechanical properties of the roof rock strata, it is predicted that a roof collapse disaster may occur, and the potential disaster types of the potential disaster area are determined.

[0044] For example, in the above-mentioned determined potential disaster area, by consulting the coal mine mining historical data, it is found that in the past mining process of the coal seams near this potential disaster area, water inrush accidents occurred near the fault. At the same time, from the geological structure map, it can be seen that there is a large fault passing through this potential disaster area, and the formation speed of this potential disaster area changes significantly near the fault. Combining this information and using the water inrush disaster prediction model for analysis, it is judged that there is a potential risk of water inrush in this potential disaster area, and the potential disaster type of this potential disaster area is determined as a water inrush disaster.

[0045] The multi-directional signals obtained by the three-component micro-motion detection in the embodiment of the present invention contain more information about the changes in the underground medium and can more accurately reflect the changes in the physical characteristics of the disaster-prone area. In the velocity inversion, the micro-motion dispersion curve of each grid measuring point is inverted and calculated based on the layered medium theory, which can more accurately invert the underground velocity, thereby more accurately locating the abnormal area. The stratum velocity of the local area is quantified by the velocity difference of the stratum velocity, so that the potential disaster area, that is, the location where the disaster occurs, can be quantitatively judged based on the stratum velocity of the local area. Therefore, the location of the coal mine disaster can be accurately located, thereby improving the accuracy and reliability of coal mine disaster prediction.

[0046] In one embodiment, steps 101 to 104 are described as follows:

[0047] Step 101: Analyze the rock type to obtain rock physical characteristic parameters; the rock physical characteristic parameters include rock density, elastic modulus and Poisson's ratio;

[0048] Optionally, the coal mine geological conditions in the embodiment of the present invention include rock type and stratum thickness. Therefore, the disaster prediction system analyzes the rock type to obtain rock physical characteristic parameters, wherein the rock physical characteristic parameters include rock density, elastic modulus and Poisson's ratio. The elastic modulus E can be calculated based on the longitudinal wave velocity v p , shear wave velocity v s and rock density ρ, using the formula Calculation: Poisson's ratio ν is calculated by the formula calculate.

[0049] In one embodiment, a rock sample is obtained by drilling a hole in a coal mine to be explored. The density meter is used to measure the sample mass to be 500 g and the volume to be 200 cm 3 , then the rock density ρ=500 / 200=2.5g / cm 3 The longitudinal wave velocity v is measured using ultrasonic measuring equipment. p =3500m / s, shear wave velocity v s =2000m / s, according to the above formula, the elastic modulus E=2500*2000 2 *(3*3500 2 -4×2000 2 ) / (3500 2 -2*2000 2 )≈1.87*10 10 Pa, Poisson's ratio ν=(3500 2 -2*2000 2 ) / 2(3500 2 -2000 2 )

[0050] ≈0.25。

[0051] Step 102: Based on the rock physical property parameters, combined with the elastic mechanics theory and the wave equation when the microseismic signal propagates in each layer of medium, determine the propagation speed of the microseismic signal in different strata.

[0052] Further, based on the elastic mechanics theory, the disaster prediction system regards the strata as elastic media. For the propagation of the microseismic signal in each layer of medium, it satisfies the wave equation (where u is the displacement vector, v is the wave speed, and t is the time). In a homogeneous isotropic elastic medium, the longitudinal wave speed The transverse wave speed where λ and μ are Lame constants, which can be calculated from the elastic modulus E and Poisson's ratio ν through the formulas

[0053] Calculate, and based on the above formulas and the rock physical property parameters, determine the propagation speed of the microseismic signal in different strata.

[0054] In an embodiment, in the above coal mine area, according to the rock density ρ = 2500 kg / m 3 , elastic modulus E = 1.87 * 10 10 Pa, Poisson's ratio ν = 0.25, calculate the Lame constants

[0055]

[0056] Then calculate the longitudinal wave speed The transverse wave speed That is, the propagation speed of the microseismic signal in this stratum is determined.

[0057] Step 103: Based on the stratum thickness and the propagation speed of the microseismic signal in different strata, determine the theoretical frequency range when the microseismic signal propagates in the coal mine area to be detected, and based on the vibration signals generated by the operation of mechanical equipment in the coal mine area to be detected, conduct noise frequency analysis to determine the interference signal frequency range.

[0058] Further, the disaster prediction system obtains the stratum thickness information of the coal mine area to be detected, combines the stratum thickness and the propagation speed of the microseismic signal in different strata, and uses the calculation formula (f is the frequency, v is the wave speed, h is the stratum thickness) to calculate the frequency when the microseismic signal propagates in different strata. Further, the disaster prediction system traverses all strata to obtain the theoretical frequency range when the microseismic signal propagates in the coal mine area to be detected. At the same time, the disaster prediction system collects the vibration signals generated by the operation of mechanical equipment in the coal mine area to be detected, and determines the interference signal frequency range through spectrum analysis methods such as Fourier transform.

[0059] In one embodiment, there are three layers of strata in the coal mine area to be detected, with thicknesses of h1 = 50m, h2 = 80m, and h3 = 120m respectively. According to the longitudinal wave velocity v p = 3500m / s and the shear wave velocity v s = 2000m / s, calculate the frequency corresponding to the first layer of strata

[0060] The frequency corresponding to the second layer of strata

[0061] The frequency corresponding to the third layer of strata The obtained theoretical frequency range is approximately 8.33Hz - 35Hz. Collect the vibration signals of the mechanical equipment running in the coal mine. After Fourier transform analysis, the interference signal frequency ranges are determined to be 0 - 5Hz and 30 - 40Hz.

[0062] Step 104, exclude the interference signal frequency range from the theoretical frequency range to obtain the filtering frequency band.

[0063] Furthermore, the disaster prediction system compares the theoretical frequency range and the interference signal frequency range. Remove the overlapping part from the theoretical frequency range to obtain the final filtering frequency band. Therefore, it can be understood that the filtering frequency band is determined by subtracting the interference signal frequency range set from the theoretical frequency range set through set operations.

[0064] In the above example, the theoretical frequency range is 8.33Hz - 35Hz, and the interference signal frequency ranges are 0 - 5Hz and 30 - 40Hz. After removing the overlapping part, the obtained filtering frequency band is 8.33Hz - 30Hz. Therefore, this frequency band is determined as the filtering frequency band for preprocessing the micro - motion signal.

[0065] The embodiment of the present invention scientifically determines the filtering frequency band based on the rock physical properties, stratum structure, and mechanical equipment interference situation in the coal mine area to be detected. From the calculation of rock physical property parameters, to the determination of the propagation speed of micro - motion signals in the stratum, and then to the analysis of the theoretical frequency range and the interference signal frequency range, the finally obtained filtering frequency band can accurately retain the effective micro - motion signal frequency components related to the stratum structure and effectively exclude the interference signal frequency components generated by mechanical equipment, etc. Therefore, in the subsequent preprocessing of micro - motion signals, using this filtering frequency band can improve the signal - to - noise ratio of micro - motion signals, provide a high - quality data basis for subsequent stratum structure analysis and disaster prediction based on micro - motion signals, and improve the accuracy and reliability of coal mine disaster prediction.

[0066] In one embodiment, the descriptions of steps 201 to 205 are as follows:

[0067] Step 201 : For each grid measuring point, the three-component micro-motion signals after preprocessing are combined according to the time series to construct a three-dimensional signal matrix;

[0068] Optionally, for each grid point, the disaster prediction system combines the pre-processed three-component micro-motion signals (x, y, z directions) of each grid point in time series to construct a three-dimensional signal matrix S ijk Where i represents the grid measurement point number, j represents the signal component dimension (j = 1, 2, 3 correspond to the x, y, and z directions respectively), and k represents the time sampling point. The three-dimensional signal matrix fully records the changes in the signal in different directions of each measurement point over time.

[0069] Step 202: Calculate the ratio of the signal difference to the signal sum between sampling points at different times based on the three-dimensional signal matrix to obtain a spatiotemporal index, and determine the frequency search interval for each grid point based on the spatiotemporal index; the spatiotemporal index represents the degree of signal variation in the temporal and spatial dimensions.

[0070] Furthermore, the disaster prediction system calculates the three-dimensional signal matrix S for each grid point. ijk , calculate the ratio of the signal difference to the signal sum between sampling points at different times, and obtain the spatiotemporal index C of each grid point i , where the spatiotemporal index characterizes the degree of change of the signal in the time and space dimensions, C i The larger the value, the more drastic the signal change. The specific formula of the spatiotemporal index is as follows:

[0071]

[0072] Where N is the total number of time sampling points.

[0073] Furthermore, the disaster prediction system determines the frequency search interval [f min,i ,f max,i ], the specific formula is:

[0074]

[0075] Where T is the sampling interval and □ is a minimum positive number. The larger the spatiotemporal index, the higher the lower limit of the frequency search interval. The upper limit is always determined based on the minimum positive number to cover the possible frequency range.

[0076] Step 203: Divide the frequency search interval into a plurality of frequency sub-bands of equal proportion. For each frequency sub-band, extract the signal component corresponding to the frequency sub-band from the three-dimensional signal matrix, perform a square sum of the signal components, and divide the result by the sub-band frequency width and the number of time sampling points to obtain the energy density.

[0077] Further, the disaster prediction system divides the frequency search interval [f min,i , f max,i into M frequency sub-bands B im (m = 1, 2, …, M) in equal proportion. Among them, the frequency range of each sub-band is Therefore, the frequency sub-band distribution can be adaptively determined according to the signal characteristics of different measuring points.

[0078] Further, for each frequency sub-band B im , the disaster prediction system extracts the signal components corresponding to the frequency sub-band from the three-dimensional signal matrix S ijk . After performing the sum of signal squares on the signal components and dividing by the sub-band frequency width and the number of time sampling points, the energy density is obtained. Among them, the energy density reflects the degree of energy concentration of the signal per unit frequency and time. The specific formula for the energy density is:

[0079]

[0080] Among them, E im represents the energy density of the i-th frequency sub-band, N represents the number of time sampling points, |S ijk | represents the signal components of the i-th frequency sub-band in the three-dimensional signal matrix, [f min , f max represents the frequency search interval, and M represents the number of frequency sub-bands.

[0081] Step 204: Combine the center frequency of each frequency sub-band with the corresponding energy density into discrete mapping points, and use cubic spline non-linear interpolation to fit the discrete mapping points to obtain the frequency-energy curve; during the interpolation process of the frequency-energy curve, it is continuous at each discrete point, and the first derivative and the second derivative are continuous;

[0082] Further, the disaster prediction system combines the center frequency f c,im of each frequency sub-band with the corresponding energy density E im to form discrete mapping points (f c,im , E im ), where

[0083]

[0084] Among them, the discrete mapping points initially establish the correspondence between frequency and energy.

[0085] Further, the disaster prediction system uses the cubic spline non-linear interpolation method to fit the discrete mapping points (f cim , E im ) to obtain the frequency-energy curve E i(f). During the interpolation process of the frequency-energy curve, ensure that the curve is continuous at each discrete point and the first and second derivatives are continuous, so as to obtain a curve that accurately reflects the change of signal energy with frequency.

[0086] Step 205: Analyze the frequency-energy curve, determine the frequency point with the maximum energy change rate and its corresponding energy value, and connect the frequency points in sequence to obtain the microseismic dispersion curve of each grid measurement point.

[0087] Furthermore, the disaster prediction system analyzes the frequency-energy curve E i (f), determines the frequency point with the maximum energy change rate and its corresponding energy value, and connects the frequency points in sequence to obtain the microseismic dispersion curve D i (f). The microseismic dispersion curve highlights the key characteristics of the signal energy changing with frequency.

[0088] The embodiments of the present invention fully consider the spatio-temporal characteristics and energy distribution of microseismic signals, screen the key frequency range through spatio-temporal indexes, avoid blind analysis of the full-band signals, calculate the energy density and interpolate and fit, can more accurately capture the change trend of signal energy with frequency, so that the finally obtained microseismic dispersion curve can accurately reflect the propagation characteristics of microseismic signals at different frequencies, provides reliable basic data for subsequent formation velocity inversion based on the dispersion curve, potential disaster area analysis, etc., and improves the accuracy and reliability of coal mine disaster prediction.

[0089] In one embodiment, the descriptions of steps 301 to 304 are as follows:

[0090] Step 301: For each grid measurement point, based on the longitudinal wave velocity, transverse wave velocity and medium density of each layer of medium, construct the wave equation of each layer of medium, and combine the continuity conditions of stress and displacement at the interface of adjacent two layers of medium based on the medium thickness of each layer of medium to determine the boundary condition equation;

[0091] Optionally, for each grid measurement point, the disaster prediction system regards the formation as a model composed of n layers of homogeneous media, and the thickness of each layer of medium is h i (i = 1, 2,..., n), according to the longitudinal wave velocity v p,i of each layer of medium, the transverse wave velocity v s,i and the medium density ρ i Combined with the theory of elastic mechanics, construct the wave equation of each layer of medium. Among them, the wave equation characterizes the propagation law of waves in each layer of homogeneous medium, and establish the wave equation for each layer of medium:

[0092] Among them, u p,i and u s,iare the longitudinal wave displacement and the transverse wave displacement in the i-th layer of the medium, respectively, is the Laplace operator, and t is the time.

[0093] Furthermore, at the interface between two adjacent layers of the medium, the disaster prediction system obtains the boundary condition equation according to the continuity conditions of stress and displacement. Taking the i-th layer and the (i + 1)-th layer of the medium as an example:

[0094] u p,i (z = h i ) = u p,i+1 (z = 0). u s,i (z = h i ) = u s,i+1 (z = 0).

[0095] σ z,i (z = h i ) = σ z,i+1 (z = 0). τ xz,i (z = h i ) = τ xz,i+1

[0096] (z = 0).

[0097] where z is the vertical coordinate, σ z is the vertical stress, τ xz is the shear stress.

[0098] For example, at a certain grid measurement point in the coal mine area to be detected as mentioned above, the stratum is divided into 3 layers. The thickness of the first layer of the medium h1 = 50m, the longitudinal wave velocity v p1 = 1500m / s, the transverse wave velocity v s1 = 800m / s, and the medium density ρ1 = 2000kg / m 3 ;

[0099] The thickness of the second layer of the medium h2 = 80m, the longitudinal wave velocity v p2 = 2000m / s, the transverse wave velocity v s2 = 1200m / s, and the medium density ρ2 = 2300kg / m 3 ;

[0100] The thickness of the third layer of the medium h3 = 100m, the longitudinal wave velocity v p3 = 2500m / s, the transverse wave velocity v s3 = 1500m / s, and the medium density ρ3 = 2500kg / m 3 . According to the above parameters, the wave equations of each layer of the medium are constructed. For example, the longitudinal wave equation of the first layer is

[0101]

[0102] At the interface between the first layer and the second layer, according to the displacement continuity condition, we can obtain (for example, the z-axis is perpendicular to the formation interface), according to the stress continuity condition, we can obtain And so on to construct the complete boundary condition equation.

[0103] Step 302: Perform Fourier Hankel transform on the wave equation and the boundary condition equation in the spatial domain to obtain the transformed wave equation and the transformed boundary condition equation;

[0104] Furthermore, perform Fourier Hankel transform on the wave equation and the boundary condition equation in the spatial domain. Let the transformed longitudinal wave displacement and transverse wave displacement be respectively and where k is the wave number and ω is the angular frequency. Therefore, the transformed wave equation and the transformed boundary condition equation are:

[0105]

[0106] Step 303: Construct a layer transfer matrix based on the transformed wave equation and the transformed boundary condition equation; the layer transfer matrix characterizes the transformation relationship between displacement and stress when a wave propagates from one layer of medium to another layer of medium;

[0107] Furthermore, for each layer of medium, define the layer transfer matrix T i , the layer transfer matrix characterizes the transformation relationship between displacement and stress when a wave propagates from one layer of medium to another layer of medium. By deriving the transformed wave equation and the transformed boundary condition equation, the expression of the layer transfer matrix is obtained:

[0108]

[0109] where are respectively the longitudinal wave number and the transverse wave number in the i-th layer of medium.

[0110] Step 304: Based on the layer transfer matrix of each grid measurement point and the microtremor dispersion curve, perform inversion calculation to determine the formation velocity at different depths of each grid measurement point.

[0111] Furthermore, the disaster prediction system performs inversion calculation according to the layer transfer matrix and the microtremor dispersion curve of each grid measurement point to determine the formation velocity at different depths of each grid measurement point, as specifically described in Steps 3041 to 3043.

[0112] The embodiment of the present invention comprehensively considers various characteristics of the formation medium and the complex law of wave propagation, can effectively process the measured microtremor signal data, accurately invert the formation velocity at different depths of each grid measurement point, provides key basic information for subsequent analysis of potential disaster areas and disaster prediction in coal mines, and improves the accuracy and reliability of coal mine disaster prediction.

[0113] In one embodiment, the descriptions of steps 3041 to 3043 are as follows:

[0114] Step 3041, for each grid measurement point, multiply each layer of the layer transfer matrix in sequence to obtain the overall transfer matrix from the ground surface to the deep formation; the overall transfer matrix characterizes the comprehensive influence of the entire multi-layer homogeneous medium model on wave propagation;

[0115] Optionally, for each grid measurement point, the disaster prediction system multiplies the layer transfer matrices in sequence according to the formation order. For example, if the formation of a certain grid measurement point is divided into n layers, and the layer transfer matrices of each layer are T1, T2,..., T n , then the calculation method of the overall transfer matrix T total is: T ttol = T n ×... × T2 × T1. The overall transfer matrix synthesizes the influence of each layer of medium on wave propagation, contains the physical property information of all media from the ground surface to the deep formation, and the propagation transformation relationship of waves in the entire multi-layer homogeneous medium model can be described through this matrix.

[0116] In a certain grid measurement point in the coal mine area to be detected above, the formation is divided into 3 layers, and the layer transfer matrix of the first layer has been obtained The layer transfer matrix of the second layer The layer transfer matrix of the third layer

[0117] Calculate the overall transfer matrix T total = T3 × T2 × T1 according to the matrix multiplication rule. For example, the calculation method of the element in the first row and the first column of T total is: and so on, calculate all the elements of T total to obtain the overall transfer matrix from the ground surface to the deep formation.

[0118] Step 3042, based on the overall transfer matrix, determine the theoretical phase velocity of the wave at different frequencies, and construct an objective function based on the theoretical phase velocity and the measured phase velocity of the wave on the microtremor dispersion curve at different frequencies;

[0119] Furthermore, the disaster prediction system calculates the theoretical phase velocity of the wave at different frequencies based on the overall transfer matrix. For a given frequency f, calculate the theoretical phase velocity v theo (f) through the relevant formula of the overall transfer matrix and the wave propagation characteristics. Generally, according to the theory of layered media, the theoretical phase velocity can be deduced from the relationship between the elements of the overall transfer matrix and the frequency.

[0120] Furthermore, the disaster prediction system obtains the measured phase velocity v of the wave at different frequencies from the microtremor dispersion curve meas(f), construct the objective function J to measure the difference between the theoretical phase velocity and the measured phase velocity. The calculation formula of the objective function is as follows: This formula quantifies the difference between the two by summing the squares of the relative errors of the theoretical and measured phase velocities at different frequencies. The goal is to minimize the value of this function through optimization.

[0121] At the above grid measurement points, select frequencies f1 = 5Hz, f2 = 10Hz, f3 = 15Hz, f4 = 20Hz. Based on the calculated overall transfer matrix T total , calculate the theoretical phase velocity at each frequency through relevant formulas. For example, calculate v theo (f1) = 1200m / s, v theo (f2) = 1800m / s, v theo (f3) = 2200m / s, v theo (f4) = 2500m / s. Obtain the measured phase velocity v from the microtremor dispersion curve of this grid measurement point mces (f1) = 1100m / s, v meas (f2) = 1700m / s, v meas (f3) = 2100m / s, v meas (f4) = 2400m / s. Calculate according to the objective function formula:

[0122]

[0123] Step 3043, optimize the objective function based on the particle swarm simulated annealing hybrid algorithm. When the objective function converges to meet the preset accuracy requirements, determine the formation velocity parameters corresponding to the output optimal solution as the formation velocities at different depths of each grid measurement point.

[0124] Furthermore, the disaster prediction system uses the particle swarm simulated annealing hybrid algorithm to optimize the objective function. In the particle swarm algorithm, each particle represents a set of formation velocity parameter solutions. The particles search for the optimal solution by updating their own positions in the solution space. The update formula is:

[0125]

[0126] Among them, and are the velocity and position (corresponding to the formation velocity parameters) of the i-th particle at the t-th iteration respectively. ω is the inertia weight, c1 and c2 are learning factors, r 1d , r 2d are random numbers between [0, 1], p id is the individual optimal position of the i-th particle, and g d is the global optimal position.

[0127] The simulated annealing algorithm is used to prevent the particle swarm algorithm from falling into local optima. In each iteration, a worse solution is accepted with a certain probability, and the acceptance probability formula is as follows: where J old and J new are the objective function values before and after the update respectively, T is the current temperature, and the temperature gradually decreases with the iteration.

[0128] The disaster prediction system iterates continuously. When the objective function converges to meet the preset accuracy requirement (such as the objective function value is less than a certain threshold ε), the formation velocity parameters corresponding to the optimal solution output at this time are determined as the formation velocities of the grid measurement points at different depths. In the above example, the preset accuracy requirement of the disaster prediction system is ε = 0.005, the particle swarm size is set to 30, the inertia weight ω linearly decreases from 0.9 to 0.4, the learning factors c1 = c2 = 2, the initial temperature T0 = 100, and the temperature decrease coefficient α = 0.99.

[0129] During the iteration process, the particles continuously update their positions and velocities, and the objective function value is calculated after each update. For example, at the 50th iteration, the objective function value J = 0.006, which does not meet the accuracy requirement. Continuing the iteration, at the 80th iteration, the objective function value J = 0.0048 < ε. At this time, the formation velocity parameters corresponding to the optimal solution output, such as the longitudinal wave velocity v p1 = 1150 m / s and the shear wave velocity v s1 = 650 m / s, etc., are determined as the formation velocities of the grid measurement points at different depths.

[0130] In the embodiment of the present invention, the layered medium theory is combined with the optimization algorithm. The overall transfer matrix comprehensively reflects the comprehensive effect of the formation structure on wave propagation, the objective function accurately measures the deviation between the theory and the actual situation, and the hybrid algorithm effectively avoids local optimal solutions and improves the global optimality of the inversion result. Therefore, the formation velocity parameters can be accurately inverted from the microtremor dispersion curve, providing high-precision data support for coal mine geological structure analysis and potential disaster prediction, and improving the accuracy and reliability of coal mine disaster prediction.

[0131] In one embodiment, the descriptions of steps 401 to 403 are as follows:

[0132] Step 401, for each local area, arrange the formation velocities of each grid measurement point according to the spatial position to construct a formation velocity spatial matrix, and determine the velocity gradient tensor according to the velocity gradients of each grid measurement point in three directions in the formation velocity spatial matrix;

[0133] Optionally, for each local area, the disaster prediction system arranges the formation velocities of all grid measurement points in this area according to their spatial positions. For example, if the grid measurement points in a local area are distributed in a three-dimensional space, with m, n, and p grid measurement points along the x, y, and z directions respectively, then a formation velocity spatial matrix V of size m×n×p is constructed, where the matrix element V ijk represents the formation velocity of the grid measurement point at the position of x = i, y = j, and z = k.

[0134] Furthermore, the disaster prediction system calculates the velocity gradients of each grid measurement point in three directions in the formation velocity spatial matrix. For a certain grid measurement point (i, j, k), the velocity gradient G x (i, j, k) is calculated by the finite difference method: (when 1 < i < m, the boundary points use one-sided differences). Similarly, the velocity gradients G y (i, j,) and G z (i, j, k) in the y and z directions can be obtained. The velocity gradient tensor is composed of the velocity gradients in these three directions:

[0135] where Cross gradients such as are obtained through further partial derivative calculations (finite difference approximations are used in actual calculations).

[0136] In a local area of a certain coal mine, for example, there are 5 grid measurement points along the x direction, 4 grid measurement points along the y direction, and 3 grid measurement points along the z direction. After the disaster prediction system obtains the formation velocities of each grid measurement point, a formation velocity spatial matrix V of 5×4×3 is constructed. For example, V 2,3,2 = 1800 m / s represents that the formation velocity of the grid measurement point at the position of x = 2, y = 3, and z = 2 is 1800 m / s. For the grid measurement point (3, 2, 2), its velocity gradient is calculated.

[0137] In the x direction, Δx = 10 m (for example, the grid spacing is 10 m), Similarly, the velocity gradients in the y and z directions are calculated, and then the velocity gradient tensor G(3, 2, 2) is obtained. The same calculation is performed for all grid measurement points in this local area to obtain their respective velocity gradient tensors.

[0138] Step 402: Based on the numerical magnitude relationship between the modulus value of the velocity gradient tensor of each grid measurement point and a preset threshold, determine the abnormal grid measurement points, and construct a topological network according to the spatial distance between two abnormal grid measurement points;

[0139] Further, for the velocity gradient tensor of each grid measurement point, the disaster prediction system calculates the modulus value of the velocity gradient tensor of each grid measurement point. The specific formula for the modulus value of the velocity gradient tensor is

[0140]

[0141] Further, the disaster prediction system compares the modulus value of the velocity gradient tensor of each grid measurement point with a preset threshold τ. If |G(i,j,k)| > τ, then this grid measurement point is determined as an abnormal grid measurement point.

[0142] Further, the disaster prediction system calculates the spatial distance (where (x1,y1,z1) and (x2,y2,z2) are the coordinates of two abnormal grid measurement points) between two abnormal grid measurement points. When d is less than the set connection distance threshold δ, a connection edge is established between these two abnormal grid measurement points to construct a topological network. Among them, the topological network takes the abnormal grid measurement points as nodes and the connection relationships that meet the distance conditions as edges, visually showing the spatial correlation of the abnormal area.

[0143] In the above local area, the preset threshold τ = 15m / (s·m) and the connection distance threshold δ = 30m are set. After calculating the modulus values of the velocity gradient tensors of each grid measurement point, it is found that |G(2,2,2)| = 18m / (s·m) > τ for the grid measurement point (2,2,2), and |G(3,3,2)| = 20m / (s·m) > τ for the grid measurement point (3,3,2). These two points are determined as abnormal grid measurement points. Calculate the spatial distance between the two points

[0144] (because the grid spacing is 10m), so a connection edge is established between these two abnormal grid measurement points. The same judgment and connection operations are performed on all abnormal grid measurement points to construct the topological network of this local area.

[0145] Step 403: Based on the number of nodes and connection density of each connected domain in the topological network, determine the key connected domain, and based on the key connected domain, determine the potential disaster area of each local area.

[0146] Further, the disaster prediction system analyzes the constructed topological network to identify the connected domains (i.e., the set of nodes that are interconnected through connection edges) therein. For each connected domain, calculate its number of nodes N and connection density (where E is the number of connection edges within the connected domain). The system sets the screening conditions for the key connected domain. For example, select the connected domain with the number of nodes greater than the threshold N th and the connection density greater than the threshold ρ th as the key connected domain.

[0147] Furthermore, the disaster prediction system determines the potential disaster area of each local area based on the key connected domain, as described in steps 4031 to 4033.

[0148] In the topological network constructed above, the disaster prediction system sets the node number threshold N th =5, connection density threshold ρ th =0.4. The analysis found that one of the connected domains contains 8 nodes and the number of connecting edges E = 20, so the connection density of the connected domain is

[0149] And the number of nodes 8>N th , so this connected domain is determined as the key connected domain.

[0150] The embodiment of the present invention starts from the spatial variation law of formation velocity and can comprehensively identify areas in coal mines with abnormal formation velocity and close correlation through topological analysis, effectively locating potential disaster areas, thereby improving the accuracy and reliability of potential disaster area identification, and thus improving the accuracy and reliability of coal mine disaster prediction.

[0151] In one embodiment, steps 4031 to 4033 are described as follows:

[0152] Step 4031: For each local area, calculate the velocity variation index of each key connected domain according to the formation velocity in each key connected domain;

[0153] Optionally, for each key connected domain in a local area, the disaster prediction system calculates the velocity variation index to measure the degree of change in the formation velocity in the area. Assume that the key connected domain contains n grid measurement points, and the formation velocity of each grid measurement point is v i (i=1,2,…,n), the speed variation index S is calculated using the coefficient of variation formula, that is, in is the average formation velocity of the grid measuring points in the key connected domain, is the standard deviation of the formation velocity. The velocity variation index can reflect the degree of dispersion of the formation velocity in the key connected area relative to the average value. The larger the value, the more drastic the change in formation velocity.

[0154] In a key connected area of a local area of a coal mine, there are 8 grid measuring points, and their formation velocities are

[0155] v1=1200m / s, v2=1500m / s, v3=1800m / s, v4=1000m / s, v5

[0156] =2000m / s, v6=1300m / s, v7=1600m / s, v8=1400m / s.

[0157] First, calculate the average value:

[0158]

[0159] Then, calculate the standard deviation: σ = 295.8. Finally, calculate the velocity variation index:

[0160]

[0161] Step 4032: Compare the velocity variation index of each key connected domain with the preset velocity variation threshold to obtain the comparison result of each key connected domain;

[0162] Furthermore, the disaster prediction system compares the calculated velocity variation index of each key connected domain with the preset velocity variation threshold θ. The preset velocity variation threshold θ is a measurement standard set in advance according to factors such as coal mine geological conditions and mining experience. The comparison results are divided into three cases: when the velocity variation index S > θ, it indicates that the degree of formation velocity change in this key connected domain is relatively large, and there is a high potential disaster risk; when S = θ, it means that the degree of formation velocity change is in a critical state; when S < θ, it indicates that the formation velocity change is relatively small, and the potential disaster risk is relatively low. The system records the comparison result of each key connected domain to provide a basis for subsequent classification of the potential disaster area level.

[0163] In the above-mentioned local area of the coal mine, the disaster prediction system sets the preset velocity variation threshold θ = 0.15 according to past experience and the geological conditions of this area. Compare the calculated velocity variation index S = 0.206 of the key connected domain with θ. Since 0.206 > 0.15, it can be known that the degree of formation velocity change in this key connected domain is relatively large, and the potential disaster risk is relatively high, and record this comparison result.

[0164] Step 4033: According to the comparison result of each key connected domain, divide each key connected domain into potential disaster areas of different levels; the boundary of each disaster area is obtained by interpolating and fitting the boundary nodes of each key connected domain.

[0165] Furthermore, the disaster prediction system divides each key connected domain into potential disaster areas of different levels according to its comparison result. For example, it can be divided into high-risk areas (the velocity variation index is much greater than the threshold), medium-risk areas (the velocity variation index is slightly greater than the threshold), and low-risk areas (the velocity variation index is less than the threshold). For each key connected domain, the disaster prediction system determines the boundary of the potential disaster area by interpolating and fitting the coordinates of its boundary nodes. Cubic spline interpolation and other methods can be used for interpolation and fitting. For example, the boundary node coordinates are (x j , y j , z j)(j = 1, 2, …, m), a continuous and smooth surface is constructed through the cubic spline interpolation algorithm to obtain the boundary of the potential disaster area, so as to intuitively display potential disaster areas of different levels.

[0166] Continuing with the above example of a local area of a coal mine, according to the comparison results, the key connected domains with a velocity variation index greater than 0.2 are classified as high-risk potential disaster areas, those with a velocity variation index between 0.15 and 0.2 are classified as medium-risk areas, and those less than 0.15 are classified as low-risk areas. For the key connected domain with a velocity variation index of 0.206 mentioned above, its boundary contains 6 nodes, and the coordinates are

[0167] (100, 200, 300), (120, 220, 310), (140, 240, 320), (160, 260, 330), (

[0168] 180, 280, 340),

[0169] (200, 300, 350).

[0170] The cubic spline interpolation algorithm is used to process these boundary nodes to construct a continuous and smooth surface and determine the boundary of this high-risk potential disaster area.

[0171] Based on the characteristics of formation velocity changes, through rigorous mathematical calculations and spatial analysis, the embodiments of the present invention can not only accurately identify potential disaster areas, but also further distinguish the disaster risk levels, improving the accuracy and reliability of coal mine disaster prediction.

[0172] Optionally, referring to Figure 2 , Figure 2 is a schematic flow chart of the coal mine disaster data analysis and prediction method based on three-component microtremor detection provided by the present invention. The coal mine disaster data analysis and prediction method based on three-component microtremor detection includes:

[0173] Step 10: Determine the filtering frequency band based on the coal mine geological conditions of the coal mine area to be detected, and preprocess the initial three-component microtremor signals collected by the microtremor sensors at the grid measurement points corresponding to the coal mine area to be detected based on the filtering frequency band to obtain the preprocessed three-component microtremor signals;

[0174] Step 20: Perform spectral analysis on the preprocessed three-component microtremor signals of each grid measurement point, and obtain the microtremor dispersion curves of each grid measurement point according to the microtremor signal energy distribution at different frequencies;

[0175] Step 30: Perform inversion calculation on the microtremor dispersion curves of each grid measurement point based on the layered medium theory to determine the formation velocities of each grid measurement point at different depths; the layered medium theory simplifies the formation into a multi-layer homogeneous medium model;

[0176] Step 40: Perform velocity difference analysis based on the formation velocities of each grid measurement point to obtain the formation velocity of a local area, and determine a potential disaster area based on the formation velocity of the local area.

[0177] Step 50: Perform disaster prediction on the potential disaster area based on the coal mining historical data and geological structure map to obtain the potential disaster types of the potential disaster area.

[0178] The multi-directional signals obtained by the three-component microtremor detection in the third embodiment of the present invention contain more information about the changes in the underground medium and can more accurately reflect the changes in the physical properties of the disaster occurrence area. In velocity inversion, the microtremor dispersion curves of each grid measurement point are inversely calculated based on the layered medium theory, which can more accurately invert the underground velocity, and thus can more precisely locate the abnormal area. The formation velocity of the local area is quantified by the velocity difference of the formation velocity, so the potential disaster area, that is, the location where the disaster occurs, can be quantitatively judged according to the formation velocity of the local area. Therefore, the location of the coal mine disaster can be accurately located, and the accuracy and reliability of the coal mine disaster prediction are improved.

[0179] Please refer to Figure 3 , Figure 3 , which is a schematic diagram of an embodiment of an electronic device provided by an embodiment of the present invention. As Figure 3 shown, an embodiment of the present invention provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored on the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, the following steps are implemented:

[0180] Determine a filtering frequency band based on the coal mine geological conditions of the coal mine area to be detected, and perform preprocessing on the initial three-component microtremor signals collected by the microtremor sensors at the grid measurement points corresponding to the coal mine area to be detected based on the filtering frequency band to obtain preprocessed three-component microtremor signals;

[0181] Perform spectral analysis on the preprocessed three-component microtremor signals of each grid measurement point, and obtain the microtremor dispersion curve of each grid measurement point according to the energy distribution of the microtremor signals at different frequencies;

[0182] Perform inversion calculation on the microtremor dispersion curves of each grid measurement point based on the layered medium theory to determine the formation velocities of each grid measurement point at different depths; the layered medium theory simplifies the formation into a multi-layer homogeneous medium model;

[0183] Perform velocity difference analysis based on the formation velocities of each grid measurement point to obtain the formation velocity of a local area, and determine a potential disaster area based on the formation velocity of the local area;

[0184] Based on the historical data of coal mine exploitation and geological structure maps, disaster prediction is carried out on potential disaster areas to obtain the potential disaster types of potential disaster areas.

[0185] It should be noted that in the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0186] Those skilled in the art should understand that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0187] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the flows and / or blocks in the flowchart and / or block diagram can also be implemented. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the specified functions in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0188] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implement the specified functions in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0189] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the specified functions in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0190] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made to these embodiments by those skilled in the art once they learn the basic inventive concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present invention.

[0191] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A coal mine disaster data analysis and prediction system based on three-component micro-motion detection, characterized in that The system includes three mutually perpendicular micro - motion sensors, which are respectively used to collect micro - motion signals in the horizontal north - south direction, horizontal east - west direction and vertical direction; And the three mutually perpendicular micro - motion sensors are arranged at measuring points according to a preset grid in the coal mine area to be detected; The system includes: A data analysis module, which is used to determine a filtering frequency band based on the coal mine geological conditions in the coal mine area to be detected, and pre - process the initial three - component micro - motion signals collected by the micro - motion sensors at the grid measuring points corresponding to the coal mine area to be detected based on the filtering frequency band, so as to obtain pre - processed three - component micro - motion signals; A spectrum analysis module, which is used to perform spectrum analysis on the pre - processed three - component micro - motion signals of each grid measuring point, and obtain the micro - motion dispersion curves of each grid measuring point according to the micro - motion signal energy distribution at different frequencies; An inversion calculation module, which is used to perform inversion calculation on the micro - motion dispersion curves of each grid measuring point based on the layered medium theory to determine the formation velocities at different depths of each grid measuring point; The layered medium theory simplifies the formation into a multi - layer homogeneous medium model; A disaster analysis module, which is used to perform velocity difference analysis based on the formation velocities of each grid measuring point to obtain the formation velocities of local areas, and determine potential disaster areas based on the formation velocities of local areas; A disaster prediction module, which is used to perform disaster prediction on the potential disaster areas based on the coal mine mining historical data and geological structure maps to obtain the potential disaster types of the potential disaster areas.

2. The coal mine disaster data analysis and prediction system based on three-component microseismic detection according to claim 1, characterized in that, The determination of potential disaster areas based on the formation velocities of local areas includes: For each local area, arrange the formation velocities of each grid measuring point according to the spatial position to construct a formation velocity spatial matrix, and determine the velocity gradient tensor according to the velocity gradients of each grid measuring point in three directions in the formation velocity spatial matrix; Based on the numerical magnitude relationship between the modulus value of the velocity gradient tensor of each grid measuring point and a preset threshold, determine the abnormal grid measuring points, and construct a topological network according to the spatial distance between two abnormal grid measuring points; Based on the number of nodes and connection density of each connected domain in the topological network, determine the key connected domain, and determine the potential disaster area of each local area based on the key connected domain.

3. The data analysis and prediction system for coal mine disasters based on three-component microseismic detection according to claim 2, characterized in that The determination of the potential disaster area of each local area based on the key connected domain includes: For each local area, calculate the velocity variation index of each key connected domain according to the formation velocities within each key connected domain; Compare the velocity variation index of each key connected domain with a preset velocity variation threshold to obtain the comparison result of each key connected domain; According to the comparison results of each key connected domain, divide each key connected domain into potential disaster areas of different levels; The boundary of each disaster area is obtained by interpolating and fitting the boundary nodes of each key connected domain.

4. The coal mine disaster data analysis and prediction system based on three-component microseismic detection according to claim 1, characterized in that The inversion calculation of the micro - motion dispersion curves of each grid measuring point based on the layered medium theory to determine the formation velocities at different depths of each grid measuring point includes: For each grid measurement point, based on the longitudinal wave velocity, transverse wave velocity, and medium density of each layer of the medium, the wave equation of each layer of the medium is constructed, and the continuity conditions of stress and displacement at the interface between adjacent layers of the medium are combined based on the medium thickness of each layer of the medium to determine the boundary condition equation; Perform Fourier Hankel transform on the wave equation and the boundary condition equation in the spatial domain to obtain the transformed wave equation and the transformed boundary condition equation; Construct a layer transfer matrix based on the transformed wave equation and the transformed boundary condition equation; the layer transfer matrix characterizes the transformation relationship between displacement and stress when a wave propagates from one layer of the medium to another layer of the medium; Based on the layer transfer matrix of each grid measurement point and the microtremor dispersion curve, perform inversion calculation to determine the formation velocity at different depths of each grid measurement point.

5. The coal mine disaster data analysis and prediction system based on three-component microseismic detection according to claim 4, characterized in that, The performing inversion calculation based on the layer transfer matrix of each grid measurement point and the microtremor dispersion curve to determine the formation velocity at different depths of each grid measurement point includes: For each grid measurement point, multiply each layer of the layer transfer matrix in sequence to obtain an overall transfer matrix from the ground surface to the deep formation; the overall transfer matrix characterizes the comprehensive influence of the entire multi-layer homogeneous medium model on wave propagation; Based on the overall transfer matrix, determine the theoretical phase velocity of the wave at different frequencies, and based on the theoretical phase velocity and the measured phase velocity of the wave on the microtremor dispersion curve at different frequencies, construct an objective function; Based on the particle swarm simulated annealing hybrid algorithm, optimize the objective function. When the objective function converges to meet the preset accuracy requirements, determine the formation velocity parameters corresponding to the optimal solution output as the formation velocity at different depths of each grid measurement point.

6. The data analysis and prediction system for coal mine disasters based on three-component micro-motion detection according to claim 1, wherein Perform spectral analysis on the preprocessed three-component microtremor signals of each grid measurement point, and obtain the microtremor dispersion curve of each grid measurement point according to the energy distribution of the microtremor signals at different frequencies, including: For each grid measurement point, combine the preprocessed three-component microtremor signals according to the time series to construct a three-dimensional signal matrix; Based on the three-dimensional signal matrix, calculate the ratio of the signal difference to the signal sum between different time sampling points to obtain a spatio-temporal index, and based on the spatio-temporal index, determine the frequency search interval of each grid measurement point; the spatio-temporal index characterizes the degree of change of the signal in the time and space dimensions; Divide the frequency search interval into multiple equally-proportioned frequency sub-bands. For each frequency sub-band, extract the signal components corresponding to the frequency sub-band from the three-dimensional signal matrix, square and sum the signals, and then divide by the sub-band frequency width and the number of time sampling points to obtain the energy density; Combine the center frequency of each frequency sub-band with the corresponding energy density into discrete mapping points, and use cubic spline non-linear interpolation to fit the discrete mapping points to obtain a frequency-energy curve; during the interpolation process of the frequency-energy curve, it is continuous at each discrete point and the first derivative and the second derivative are continuous; Analyze the frequency-energy curve, determine the frequency point with the largest energy change rate and its corresponding energy value, and connect the frequency points in sequence to obtain the microtremor dispersion curve of each grid measurement point; The calculation formula of the energy density is as follows: Among them, E im represents the energy density of the i-th frequency sub-band, N represents the number of time sampling points, and |S ijk | represents the signal component of the i-th frequency sub-band in the three-dimensional signal matrix, [f min , f max represents the frequency search interval, and M represents the number of frequency sub-bands.

7. The data analysis and prediction system for coal mine disasters based on three-component microseismic detection according to any one of claims 1 to 6, characterized in that, The coal mine geological conditions include rock type and formation thickness; Determining a filtering frequency band based on the coal mine geological conditions of the coal mine to be detected includes: Analyzing the rock type to obtain rock physical property parameters; the rock physical property parameters include rock density, elastic modulus, and Poisson's ratio; Based on the rock physical property parameters, combining the theory of elasticity and the wave equation when the microtremor signal propagates in each layer of medium, determining the propagation speed of the microtremor signal in different strata; Based on the formation thickness and the propagation speed of the microtremor signal in different strata, determining the theoretical frequency range when the microtremor signal propagates in the coal mine to be detected area, and performing noise frequency analysis on the vibration signals generated by the operation of mechanical equipment in the coal mine to be detected area to determine the interference signal frequency range; Excluding the interference signal frequency range from the theoretical frequency range to obtain the filtering frequency band.

8. A method for analyzing and predicting coal mine disaster data based on three-component microseismic detection, characterized in that, Implemented based on the coal mine disaster data analysis and prediction system based on three-component microtremor detection according to any one of claims 1 to 7, the method includes: Determining a filtering frequency band based on the coal mine geological conditions of the coal mine to be detected, and preprocessing the initial three-component microtremor signals collected by the microtremor sensors at the grid measurement points corresponding to the coal mine to be detected area based on the filtering frequency band to obtain preprocessed three-component microtremor signals; Performing spectral analysis on the preprocessed three-component microtremor signals of each grid measurement point, and obtaining the microtremor dispersion curves of each grid measurement point according to the microtremor signal energy distribution at different frequencies; Performing inversion calculation on the microtremor dispersion curves of each grid measurement point based on the layered medium theory to determine the formation velocities at different depths of each grid measurement point; the layered medium theory simplifies the formation into a multi-layer homogeneous medium model; Performing velocity difference analysis based on the formation velocities of each grid measurement point to obtain the formation velocities of the local area, and determining the potential disaster area based on the formation velocities of the local area; Performing disaster prediction on the potential disaster area based on the coal mine mining historical data and the geological structure map to obtain the potential disaster types of the potential disaster area.

9. An electronic device, characterized in that, The electronic device includes a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the method for analyzing and predicting coal mine disaster data based on three-component microtremor detection according to claim 8 when executing the program.

10. A non-transitory computer-readable storage medium storing a computer software program, characterized in that, The processor implements the method for analyzing and predicting coal mine disaster data based on three-component microtremor detection according to claim 8 when executing the program.

Images