A method for detecting potential fluid distribution in underlying rock strata of coal seam floor

Abstract

The present invention relates to fluid distribution exploration, and specifically to a method for exploring the potential fluid distribution in the rock strata beneath the coal seam floor. A microseismic monitoring system is constructed at the coal mining face to monitor and collect seismic wave signals in the mine in real time during coal seam mining. A machine learning microseismic event detection method based on artificial intelligence is used to detect microseismic events caused by rock fractures in the rock strata beneath the coal seam floor. The longitudinal and transverse wave velocity ratios at the locations where the microseismic event clusters are located are determined based on the arrival time differences of the microseismic event clusters. The Poisson's ratio at the rock fracture is obtained based on the longitudinal and transverse wave velocity ratios at the locations where the microseismic event clusters are located. The potential fluid distribution in the rock strata beneath the coal seam floor is explored based on the distribution of Poisson's ratios at different rock fractures. The technical solution provided by the present invention can effectively overcome the defect of the prior art that the dynamic spatial distribution characteristics of potential water bodies in the rock strata beneath the coal seam floor cannot be accurately explored.

Description

Technical Field

[0001] The present invention relates to fluid distribution exploration, and in particular to a method for exploring potential fluid distribution in rock strata beneath a coal seam floor. Background Art

[0002] Coal is my country's primary energy source and a key cornerstone for the healthy and stable development of the national economy. With the increasing depth, intensity, speed, and scale of coal mining, the problem of water inrush from the coal seam floor is becoming increasingly serious. The essence of coal seam floor water inrush is that the underlying aquifer (primarily Carboniferous and Ordovician limestone aquifers) is a confined water-rich area, and its overlying rock strata contain a primary lifting zone of a certain height. During coal seam mining, the underlying rock mass fractures, forming a fracture zone. This fracture zone develops dynamically in time and space. The formation of a fracture zone within the underlying rock strata causes the primary lifting zone above the aquifer to gradually extend upward under the combined action of tensile secondary stress and confined water. During this process, confined water continuously invades and redistributes along the fracture zone. When the lifting zone connects with the fracture zone of the coal seam floor, the confined water enters the coal seam floor, causing coal seam floor water inrush. When the amount of water bursting from the coal seam floor is large, it will cause great harm to the personnel and production equipment in the mine.

[0003] Therefore, to prevent and control water inrush accidents from coal seam floors and implement timely preventive measures, it is necessary to detect and understand the dynamic spatial distribution characteristics of potential water bodies in the underlying rock strata during coal mining. Currently, the main methods for detecting potential water bodies in the underlying rock strata of coal seam floors include hydrogeological parameter observation, transient electromagnetic method, direct current method, and seismic exploration.

[0004] 1) Hydrogeological parameter observation method

[0005] The hydrogeological parameter observation method is based on the theory of groundwater runoff. By drilling holes in the mine to pump water and observing the changes in geological parameters such as water volume and water level, the distribution of potential water bodies in the rock strata beneath the coal seam floor can be explored.

[0006] 2) Transient electromagnetic method

[0007] The transient electromagnetic method (TEM) is based on differences in the electrical properties of rock. When the rock contains water, its apparent resistivity is low. This method uses an ungrounded return line (magnetic source) or a grounded electrode (line source) to transmit a pulsed magnetic field underground. During the pauses between pulses, the coil or grounded electrode receives the secondary eddy current field. By inverting and interpreting the secondary field curve, the distribution of the underground geoelectric structure is detected. This method is sensitive to low-resistance structures. Formations containing water exhibit relatively low resistance compared to those without water. Therefore, it can be used to explore the distribution of potential water bodies in the rock strata beneath the coal seam floor.

[0008] 3) Direct current method

[0009] Direct current (DC) electrical exploration is based on differences in rock electrical properties. When a rock mass contains water, its apparent resistivity is low. This method, a form of full-space electrical exploration, uses full-space field theory to study the variations in stratum electrical properties at depth, thereby obtaining various geological information about the depth of the strata. Water-bearing bodies exhibit relatively low resistivity.

[0010] This method involves placing electrodes on the floors of two trenches at the working face, with one side emitting an artificially induced electric field and the other receiving it. Once both trenches are fully receiving and transmitting, a full-space three-dimensional apparent resistivity inversion is performed. Based on abnormal changes in the apparent resistivity of the coal seam floor, the water content of the formation is determined. Currently, direct current electrical methods primarily rely on parallel electrical methods and audio-frequency electroscopy. By obtaining formation apparent resistivity characteristics, the distribution of potential water bodies in the underlying rock layers of the coal seam floor can be explored.

[0011] 4) Seismic exploration

[0012] Seismic exploration uses explosives as the active source of seismic waves and uses differences in rock impedance as the exploration basis. Based on seismic wave propagation theory, seismic inversion is performed using seismic reflection waves to obtain geophysical parameters related to formation water content. This includes post-stack seismic inversion and pre-stack seismic inversion. Post-stack seismic inversion obtains formation porosity data by inverting acoustic impedance (AI) and analyzing the relationship between acoustic impedance and formation porosity. It is generally believed that formations with higher porosity contain more water than those with lower porosity, thus exploring the distribution of potential water bodies in the rock layers underlying the coal seam floor.

[0013] Compared to post-stack seismic inversion, pre-stack seismic inversion uses common center point gathers or common reflection bins to analyze how the reflection wave amplitude varies with offset (or angle of incidence). This allows for the estimation of elastic parameters on both sides of the lithologic interface, such as the pseudo-Poisson's ratio, derived by adding the AVO intercept attribute and the gradient attribute. When water is present in the rock formation, the pseudo-Poisson's ratio increases. This allows for the investigation of potential water distribution in the rock formations beneath the coal seam floor.

[0014] Disadvantages of existing technology:

[0015] 1) Hydrogeological parameter observation method

[0016] This method, which involves drilling holes for pumping water, is very expensive and, due to economic constraints, cannot be widely used in coal mining areas. Furthermore, the method has a limited exploration area and is only reliable near the observation point, making it difficult to detect the distribution of potential water bodies in the underlying rock layers of the coal seam floor farther away from the observation point.

[0017] 2) Transient electromagnetic method, direct current method

[0018] a. Construction in underground coal mine tunnels is based on the electrical property differences of rocks. This is greatly affected by interference from production equipment in the tunnels, which can easily lead to false detection. That is, areas with iron artifacts appear as low-resistance features during the inversion imaging process, marking low-water-rich rock layers as high-water-rich rock layers, resulting in low detection accuracy.

[0019] b. This method is generally a static exploration method, which makes it difficult to effectively obtain the dynamic spatial distribution characteristics of potential water bodies in the underlying rock strata of the coal seam floor during coal seam mining, greatly reducing the timeliness and accuracy of coal seam floor water inrush disaster prediction;

[0020] c. The exploration area is limited and is only reliable near the observation point (two drifts of the working face) and within the effective area covered by the geophysical observation. It is difficult to explore the distribution of potential water bodies in the underlying rock strata of the coal seam floor that is far away from the observation point.

[0021] 3) Seismic exploration method

[0022] a. This method is constructed on the surface of a coal mine, using explosives as the active source of seismic waves. This method is difficult to construct in mountainous areas with complex conditions.

[0023] b. When there is a coal seam goaf above the target layer, it will seriously affect the propagation path of the seismic reflection wave, resulting in low accuracy of the inversion data and difficulty in effectively discovering the potential distribution of water bodies in the rock layer beneath the coal seam floor;

[0024] c. This method is implemented before the coal mine is put into production. It is a static exploration and is difficult to effectively obtain the dynamic spatial distribution characteristics of potential water bodies in the underlying rock strata of the coal seam floor during coal seam mining. This greatly reduces the timeliness and accuracy of coal seam floor water inrush disaster prediction;

[0025] d. This method has the following prerequisites: there are differences in wave impedance at the interface of underground strata, and the method is affected by many factors during its use, such as seismic wavelets, stratum velocity, incident and transmission angles, and structure. Summary of the Invention

[0026] (1) Technical problems solved

[0027] In response to the above-mentioned shortcomings of the existing technology, the present invention provides a method for exploring the potential fluid distribution in the rock strata beneath the coal seam floor, which can effectively overcome the defect of the existing technology that it cannot accurately explore the dynamic spatial distribution characteristics of potential water bodies in the rock strata beneath the coal seam floor.

[0028] (2) Technical solution

[0029] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0030] A method for detecting potential fluid distribution in the underlying rock strata of a coal seam floor comprises the following steps:

[0031] S1. Build a microseismic monitoring system at the coal mining face to monitor and collect seismic wave signals in the mine in real time during coal seam mining;

[0032] S2. Utilize an artificial intelligence-based machine learning microseismic event detection method to detect microseismic events caused by rock mass fractures in the underlying rock strata of the coal seam floor;

[0033] S3. determining the longitudinal and transverse wave velocity ratios at the location of the microseismic event cluster based on the arrival time differences of the microseismic event cluster;

[0034] S4. Based on the longitudinal and transverse wave velocity ratios at the locations of the microseismic event clusters, the Poisson's ratio at the rock mass fracture is obtained;

[0035] S5. Based on the distribution of Poisson's ratio at different rock fracture sites, explore the potential fluid distribution in the underlying rock strata of the coal seam floor.

[0036] Preferably, after detecting the microseismic event caused by rock mass fracture in the underlying rock layer of the coal seam floor in S2, the method includes:

[0037] Preprocess the seismic wave signal, decompose it in the time-frequency domain and remove noise to obtain effective microseismic signals generated by the underlying rock mass fracture, including:

[0038] S21. Perform time-frequency domain analysis on the collected continuous seismic wave signal, and perform multi-scale analysis after wavelet decomposition, as shown in the following formula:

[0039] W0d=W0f+ε*W0z

[0040] Where W0 is the wavelet transform operator, d is the input waveform, which can be decomposed into vectors d1, d2, ...d N , f represents the real waveform vector f1, f2, ...f N , z represents a Gaussian random vector z1, z2, ... z N , ε is the standard deviation of the additional noise signal;

[0041] S22, perform threshold quantization processing on the high frequency coefficients in each scale using the following formula:

[0042]

[0043] Among them, t N is the selected threshold, N is the total number of wavelet coefficients at the corresponding scale, and the expression of the threshold processing process is represents threshold processing;

[0044] S23, use the following formula to perform wavelet reconstruction to obtain the reconstructed waveform data f * :

[0045]

[0046] And perform inverse transformation W0 -1 , to complete the recovery of the time-frequency domain and obtain the original seismic wave signal after denoising in the time domain.

[0047] Preferably, determining the longitudinal and transverse wave velocity ratios at the location of the microseismic event cluster based on the arrival time differences of the microseismic event cluster in S3 includes:

[0048] S31. In the microseismic signal generated by the rupture of the underlying rock mass, the P and S wave phases are picked according to the first arrival times of the P and S waves;

[0049] S32, spatially locating the microseismic event using a collapse grid search positioning method, and simultaneously calculating the onset time of the microseismic event caused by the rock mass fracture in the underlying rock layer of the coal seam floor;

[0050] S33. Calculate the longitudinal and transverse wave velocity ratios at the location of the microseismic event cluster.

[0051] Preferably, calculating the longitudinal and transverse wave velocity ratios at the location of the microseismic event cluster in S33 includes:

[0052] S331, P-wave propagation velocity V of microseismic events P , shear wave propagation velocity V S They are expressed in the following formulas:

[0053]

[0054] S332. Calculate the longitudinal and transverse wave velocity ratios at the location of the microseismic event cluster using the following formula:

[0055]

[0056] Among them, L is the spatial distance between the earthquake source at the microfracture of the rock stratum beneath the coal seam floor and the microseismic sensor; T is the time of occurrence of the microseismic event, which can be obtained by spatial positioning calculation of the microseismic event; T0 is the first arrival time of the P wave in the microseismic signal, which can be obtained by P wave phase picking; T1 is the first arrival time of the S wave in the microseismic signal, which can be obtained by S wave phase picking.

[0057] Preferably, in S4, the Poisson's ratio at the rock mass fracture is obtained based on the longitudinal and transverse wave velocity ratios at the location of the microseismic event cluster, including:

[0058] During coal seam mining, the following formula is used to calculate the dynamic Poisson's ratio σ of the underlying rock layer of the coal seam floor:

[0059]

[0060] in, and Then the dynamic Poisson's ratio σ can be expressed as:

[0061]

[0062] Where ρ is the velocity of the underground rock medium, and λ and μ are the Lame coefficients.

[0063] Preferably, in S5, the potential fluid distribution of the underlying rock layer of the coal seam floor is explored based on the distribution of Poisson's ratios at different rock mass fractures, including:

[0064] S51. Before coal seam mining, rock samples are drilled in two slots of the coal mining face to conduct rock physics experiments. The longitudinal and shear wave propagation velocities of different rock layers below the coal seam floor are measured, and the static Poisson's ratio σ' of different rock layers is calculated.

[0065] S52. Based on the spatial location of the microseismic event caused by the rock mass rupture in the underlying rock strata of the coal seam floor, compare the dynamic Poisson's ratio σ of the rock mass at that location with the static Poisson's ratio σ' of the same rock stratum at that depth. When the dynamic Poisson's ratio σ is greater than the static Poisson's ratio σ', determine the water content of the rock stratum at that location.

[0066] S53. During coal seam mining, the water content of the rock layer is determined based on the changes in the dynamic Poisson's ratio σ.

[0067] S54. Obtain the dynamic Poisson's ratio σ of all rock fractures in the exploration area, and perform three-dimensional spatial analysis based on the spatial location of the microseismic events to dynamically explore the potential fluid distribution in the rock layer beneath the coal seam floor.

[0068] Preferably, the microseismic monitoring system includes a mine-use flameproof and intrinsically safe microseismic monitoring substation, the mine-use flameproof and intrinsically safe microseismic monitoring substation is connected to a mine-use intrinsically safe microseismic sensor, the mine-use flameproof and intrinsically safe microseismic monitoring substation is connected to a microseismic monitoring server via an industrial-grade optoelectronic switch, the microseismic monitoring server is connected to a data processing client, and the data processing client is connected to a printer;

[0069] The mine-used flameproof and intrinsically safe microseismic monitoring substation is connected to an industrial-grade clock signal converter, and the industrial-grade clock signal converter is connected to a GPS antenna via a GPS time signal synchronizer.

[0070] (3) Beneficial effects

[0071] Compared with the existing technology, the present invention provides a method for detecting the potential fluid distribution in the rock layer beneath the coal seam floor. By collecting the microseismic signal generated by the fracture of the rock mass beneath the coal seam floor during coal seam mining, the Poisson's ratio parameter of the rock mass is obtained by processing and analyzing the signal, and then the distribution of potential water bodies in the rock layer beneath the coal seam floor is detected. Compared with the existing technology, the present application uses a passive seismic source (rock fracture) to detect the distribution of potential water bodies in the rock layer beneath the coal seam floor. It is not affected by the underground production equipment of the coal mine, the surface topography and the overlying goaf of the rock layer. It can accurately detect the dynamic spatial distribution characteristics of the potential water bodies in the rock layer beneath the coal seam floor, and the detection area is wide, which can cover the entire mine, providing a new technical method for coal seam floor water inrush warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0073] Figure 1 It is a schematic diagram of the process of the present invention;

[0074] Figure 2 Schematic diagram of the microseismic monitoring system of the present invention;

[0075] Figure 3 Schematic diagram of the design of microseismic monitoring points in the microseismic monitoring system of the present invention;

[0076] Figure 4 Schematic diagram of the installation of microseismic sensors at microseismic monitoring points in the microseismic monitoring system of the present invention;

[0077] Figure 5 This is a waveform diagram of the environmental background noise recorded by the microseismic monitoring system of the present invention;

[0078] Figure 6 Schematic diagram of the structure of the convolutional neural network CNN in the present invention;

[0079] Figure 7 The waveform diagram of the microseismic signal generated by the rupture of the underlying rock mass detected by the convolutional neural network CNN in the present invention in the continuous recording;

[0080] Figure 8 The seismic wave signal waveform and denoising results in the present invention are shown in Figure 1, where (a), (b), and (c) represent the E, N, and Z components, respectively.

[0081] Figure 9Schematic diagram of picking up P and S wave phases in a microseismic signal waveform diagram in the present invention;

[0082] Figure 10 Schematic diagram of spatial positioning of microseismic events using the collapsed grid search positioning method in the present invention;

[0083] Figure 11 Schematic diagram of the spatial distribution of microseismic events obtained by using the collapsed grid search positioning method in the present invention;

[0084] Figure 12 This is a comparison of the longitudinal wave propagation velocity and the shear wave propagation velocity before and after the rock mass is saturated with water;

[0085] Figure 13 Comparison of Poisson's ratio before and after rock mass is saturated with water. DETAILED DESCRIPTION

[0086] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0087] A method for detecting the potential fluid distribution in the underlying rock layer of the coal seam floor, such as Figure 1 As shown in Figure 1, ① a microseismic monitoring system is built at the coal mining face to monitor and collect seismic wave signals in the mine during coal seam mining in real time.

[0088] By constructing a high-sensitivity, low-cost, low-noise microseismic monitoring system at the coal mining face, the seismic wave signals in the mine during coal mining are monitored and collected in real time, including underground production activity signals and microseismic signals caused by rock fractures induced by mining disturbances in the underlying rock strata beneath the coal seam floor.

[0089] like Figure 2 As shown, the microseismic monitoring system includes a mine flameproof and intrinsically safe microseismic monitoring substation 1, which is connected to a mine intrinsically safe microseismic sensor 2. The mine flameproof and intrinsically safe microseismic monitoring substation 1 is connected to a microseismic monitoring server 5 via an industrial-grade optoelectronic switch 3. The microseismic monitoring server 5 is connected to a data processing client 6, which is connected to a printer 7.

[0090] The mine flameproof and intrinsically safe microseismic monitoring substation 1 is connected to an industrial-grade clock signal converter 4, which is connected to a GPS antenna 9 via a GPS time signal synchronizer 8. The technical parameters of the microseismic monitoring system are shown in the following table:

[0091] Table 1 Technical parameters of microseismic monitoring system

[0092]

[0093] In order to monitor and collect seismic wave signals in the mine in real time during coal seam mining, it is necessary to deploy microseismic monitoring points on both sides of the coal mining face:

[0094] The distance between adjacent microseismic monitoring points in the same tunnel is about 100-110m. The microseismic monitoring points closest to the cut-hole in the two trenches are designed to be about 80m and 130m away from the cut-hole, forming an envelope coverage in space. Figure 3 As shown;

[0095] The microseismic sensor is installed by drilling, and the bottom of the hole is in the bedrock below the coal seam floor (the vertical hole depth is generally about 4m), and cement mortar is used as the sensor coupling agent. Figure 4 shown.

[0096] The voltage value of the ambient background noise waveform recorded by the microseismic monitoring system should be between 10 -5 V level, such as Figure 5 shown.

[0097] ② Use machine learning microseismic event detection methods based on artificial intelligence to detect microseismic events caused by rock fractures in the underlying rock strata of the coal seam floor.

[0098] In the technical solution of this application, a convolutional neural network CNN (such as Figure 6 (As shown in Figure 2), this technology detects microseismic events caused by rock fractures in the underlying strata during coal mining. Microseismic event detection can be considered a form of image classification, using a convolutional neural network (CNN) to classify a continuous waveform into noise and microseismic events.

[0099] Among them, the convolutional neural network CNN requires a large amount of labeled data: including microseismic events and noise, so that the neural network can automatically extract the corresponding features according to the labels. After a large amount of model training, a converged model can be obtained. When used for other continuous waveforms, the model will give a probability of being a microseismic event. If the probability is high, it can be regarded as a microseismic event, and if the probability is low, it is noise. The microseismic signal generated by the rupture of the underlying rock mass detected by the convolutional neural network CNN in the continuous recording is as follows: Figure 7 shown.

[0100] ③ Preprocess the seismic wave signal, decompose the seismic wave signal in the time-frequency domain and remove noise to obtain effective microseismic signals generated by the underlying rock mass fracture, including:

[0101] S21. Perform time-frequency domain analysis on the collected continuous seismic wave signal, and perform multi-scale analysis after wavelet decomposition, as shown in the following formula:

[0102] W0d=W0f+ε*W0z

[0103] Where W0 is the wavelet transform operator, d is the input waveform, which can be decomposed into vectors d1, d2, ...d N , f represents the real waveform vector f1, f2, ...f N , z represents a Gaussian random vector z1, z2, ... z N , ε is the standard deviation of the additional noise signal;

[0104] S22, perform threshold quantization processing on the high frequency coefficients in each scale using the following formula:

[0105]

[0106] Among them, t N is the selected threshold, N is the total number of wavelet coefficients at the corresponding scale, and the expression of the threshold processing process is represents threshold processing;

[0107] S23, use the following formula to perform wavelet reconstruction to obtain the reconstructed waveform data f * :

[0108]

[0109] And perform inverse transformation W0 -1 , to complete the recovery of the time-frequency domain and obtain the original seismic wave signal after denoising in the time domain.

[0110] When selecting the threshold function in S22, there are four forms to choose from: a fixed threshold Sqtwolog, a threshold minimaxi determined by the extreme value of the minimum mean square error, a threshold rigsure estimated by the minimum risk, and a heursure threshold. This application selects the heursure threshold form.

[0111] During coal mining, the seismic wave signals collected and monitored in real time by the microseismic monitoring system underground contain random noise, such as strong energy disturbances, noise generated by poor coupling between the microseismic sensor and the formation, and environmental background noise. Therefore, the seismic wave signals collected by the microseismic monitoring system must first be preprocessed. This involves decomposing the seismic wave signals in the time-frequency domain and removing noise to obtain effective microseismic signals generated by the underlying rock mass fractures.

[0112] Seismic wave signals and denoising results are as follows Figure 8As shown in Figure 3, (a), (b), and (c) represent the E, N, and Z components, respectively. After wavelet threshold denoising, the signal-to-noise ratio of the continuous microseismic signal is greatly improved, and the P-wave and S-wave phases are clearer, ensuring the subsequent accurate picking of the P and S-wave phases.

[0113] ④ According to the arrival time difference of the microseismic event cluster, determine the longitudinal and transverse wave velocity ratios at the location of the microseismic event cluster, including:

[0114] S31. In the microseismic signal generated by the rupture of the underlying rock mass, the P and S wave phases are picked according to the first arrival times of the P and S waves;

[0115] S32, spatially locating the microseismic event using a collapse grid search positioning method, and simultaneously calculating the onset time of the microseismic event caused by the rock mass fracture in the underlying rock layer of the coal seam floor;

[0116] S33. Calculate the longitudinal and transverse wave velocity ratios at the location of the microseismic event cluster.

[0117] The calculation of the longitudinal and transverse wave velocity ratios at the location of the microseismic event cluster includes:

[0118] S331, P-wave propagation velocity V of microseismic events P , shear wave propagation velocity V S They are expressed in the following formulas:

[0119]

[0120] S332. Calculate the longitudinal and transverse wave velocity ratios at the location of the microseismic event cluster using the following formula:

[0121]

[0122] Among them, L is the spatial distance between the earthquake source at the microfracture of the rock stratum beneath the coal seam floor and the microseismic sensor; T is the time of occurrence of the microseismic event, which can be obtained by spatial positioning calculation of the microseismic event; T0 is the first arrival time of the P wave in the microseismic signal, which can be obtained by P wave phase picking; T1 is the first arrival time of the S wave in the microseismic signal, which can be obtained by S wave phase picking.

[0123] During coal mining, the microseismic signals generated by the fracture of the underlying rock mass of the coal seam floor include P waves and S waves. In the time series, the P wave arrives first and the S wave arrives later. The P and S wave phases can be picked up based on their first arrival times, such as Figure 9 shown.

[0124] The collapsed grid search positioning method first divides the model space into regular grids, then performs a quick search on a coarse grid in the model space to obtain the error distribution, and then performs an encrypted search on the global minimum area based on the Gaussian error distribution assumption. The global minimum value is obtained through multiple encrypted searches, such as Figure 10 The spatial distribution of microseismic events obtained by using the collapsed grid search positioning method is shown in Figure 11 As shown (the balls in the figure represent the locations of microseismic events, and the colors represent the magnitudes of microseismic events).

[0125] ⑤ Based on the longitudinal and transverse wave velocity ratios at the location of the microseismic event cluster, the Poisson's ratio at the rock mass rupture is obtained, including:

[0126] During coal seam mining, the following formula is used to calculate the dynamic Poisson's ratio σ of the underlying rock layer of the coal seam floor:

[0127]

[0128] in, and Then the dynamic Poisson's ratio σ can be expressed as:

[0129]

[0130] Where ρ is the velocity of the underground rock medium, and λ and μ (shear modulus) are the Lame coefficients.

[0131] Poisson's ratio is a parameter that describes elastic objects and is sensitive to formation lithology and the fluids it contains. In 1976, A.R. Gregory discovered through experiments that when the porosity of a formation exceeds 25%, the Poisson's ratio difference between water-saturated and gas-saturated formations is very significant. He therefore proposed that the Poisson's ratio could be used to identify fluid properties.

[0132] The Poisson's ratio characteristics of rocks were experimentally studied under simulated formation conditions. The rock types used in the experiment primarily consisted of medium-fine-grained lithic feldspathic sandstone, fine-grained lithic quartz sandstone, fine-grained lithic sandstone, medium-grained feldspathic quartz sandstone, and medium-grained carbonate-bearing lithic quartz sandstone, with minor amounts of silty mudstone and mudstone. The results showed that the Poisson's ratio before water saturation was primarily distributed between 0.1 and 0.2, while after water saturation, the Poisson's ratio was primarily concentrated between 0.2 and 0.3, with an increase generally exceeding 30%.

[0133] From the perspective of acoustic experiments, the Poisson's ratio increases significantly after saturation of water, mainly because the longitudinal wave propagation velocity increases, while the transverse wave propagation velocity remains basically unchanged, resulting in an increase in the longitudinal and transverse wave velocity ratio, such as Figure 12 From the perspective of rock mechanics, water media may cause the weakening of rock structure and the enhancement of plastic deformation, leading to an increase in Poisson's ratio, such as Figure 13 shown.

[0134] ⑥ Based on the distribution of Poisson's ratio at different rock mass fractures, explore the potential fluid distribution of the underlying rock strata under the coal seam floor, including:

[0135] S51. Before coal seam mining, rock samples are drilled in two slots of the coal mining face to conduct rock physics experiments. The longitudinal and shear wave propagation velocities of different rock layers below the coal seam floor are measured, and the static Poisson's ratio σ' of different rock layers is calculated.

[0136] S52. Based on the spatial location of the microseismic event caused by the rock mass rupture in the underlying rock strata of the coal seam floor, compare the dynamic Poisson's ratio σ of the rock mass at that location with the static Poisson's ratio σ' of the same rock stratum at that depth. When the dynamic Poisson's ratio σ is greater than the static Poisson's ratio σ', determine the water content of the rock stratum at that location.

[0137] S53. During coal seam mining, the water content of the rock layer is determined based on the changes in the dynamic Poisson's ratio σ.

[0138] S54. Obtain the dynamic Poisson's ratio σ of all rock fractures in the exploration area, and perform three-dimensional spatial analysis based on the spatial location of the microseismic events to dynamically explore the potential fluid distribution in the rock layer beneath the coal seam floor.

[0139] Before coal seam mining, cores are drilled and subjected to acoustic wave and other tests in the laboratory. The cores tested primarily refer to rocks in the aquiclude and aquifer beneath the coal seam floor, typically mudstone, interbedded sand and mudstone, sandy mudstone, and limestone. These tests determine the longitudinal and shear wave propagation velocities at different depths and for different lithologies below the coal seam floor, and calculate the static Poisson's ratio, σ', of the rock formation at that location.

[0140] During coal mining, the underlying rock strata undergo a dynamic destruction process, including rock mass fracturing, the development of fissure channels, water ingress, continued rock mass fracturing, and the continued development of fissure channels. During this process, the water content of the rock strata can be determined based on the changes in the dynamic Poisson's ratio, σ.

[0141] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for detecting potential fluid distribution in the underlying rock strata of a coal seam floor, characterized by: The following steps are involved: S1. Build a microseismic monitoring system at the coal mining face to monitor and collect seismic wave signals in the mine in real time during coal seam mining; S2. Utilize an artificial intelligence-based machine learning microseismic event detection method to detect microseismic events caused by rock mass fractures in the underlying rock strata of the coal seam floor; S3. determining the longitudinal and transverse wave velocity ratios at the location of the microseismic event cluster based on the arrival time differences of the microseismic event cluster; S4. Based on the longitudinal and transverse wave velocity ratios at the locations of the microseismic event clusters, the Poisson's ratio at the rock mass fracture is obtained; S5. Based on the distribution of Poisson's ratio at different rock mass fractures, explore the potential fluid distribution in the underlying rock strata of the coal seam floor; In S5, the distribution of Poisson's ratio at different rock mass fractures is used to explore the potential fluid distribution in the underlying rock strata of the coal seam floor, including: S51. Before coal seam mining, rock samples are drilled in two slots of the coal mining face to conduct rock physics experiments. The longitudinal and shear wave propagation velocities of different rock layers below the coal seam floor are measured, and the static Poisson's ratio σ' of different rock layers is calculated. S52. Based on the spatial location of the microseismic event caused by the rock mass rupture in the underlying rock strata of the coal seam floor, compare the dynamic Poisson's ratio σ of the rock mass at that location with the static Poisson's ratio σ' of the same rock stratum at that depth. When the dynamic Poisson's ratio σ is greater than the static Poisson's ratio σ', determine the water content of the rock stratum at that location. S53. During coal seam mining, the water content of the rock layer is determined based on the changes in the dynamic Poisson's ratio σ. S54. Obtain the dynamic Poisson's ratio σ of all rock fractures in the exploration area, and perform three-dimensional spatial analysis based on the spatial location of the microseismic events to dynamically explore the potential fluid distribution in the rock layer beneath the coal seam floor.

2. The method for detecting potential fluid distribution in the underlying rock strata of a coal seam floor according to claim 1, characterized in that: After detecting the microseismic events caused by rock mass fracture in the underlying rock layer of the coal seam floor in S2, it includes: Preprocess the seismic wave signal, decompose it in the time-frequency domain and remove noise to obtain effective microseismic signals generated by the underlying rock mass fracture, including: S21. Perform time-frequency domain analysis on the collected continuous seismic wave signal, and perform multi-scale analysis after wavelet decomposition, as shown in the following formula: W0d=W0f+ε*W0z Among them, W0 is the wavelet transform operator, d is the input waveform, which can be decomposed into vectors d1, d2, ...d N , f represents the real waveform vector f1, f2, ...f N , z represents a Gaussian random vector z1, z2, ... z N , ε is the standard deviation of the additional noise signal; S22, perform threshold quantization processing on the high frequency coefficients in each scale using the following formula: Among them, t N is the selected threshold, N is the total number of wavelet coefficients at the corresponding scale, and the expression of the threshold processing process is represents threshold processing; S23, use the following formula to perform wavelet reconstruction to obtain the reconstructed waveform data f * : And perform inverse transformation W0 -1 , to complete the recovery of the time-frequency domain and obtain the original seismic wave signal after denoising in the time domain.

3. The method for detecting potential fluid distribution in the underlying rock strata of a coal seam floor according to claim 2, characterized in that: In S3, the longitudinal and transverse wave velocity ratios at the location of the microseismic event cluster are determined based on the arrival time difference of the microseismic event cluster, including: S31. In the microseismic signal generated by the rupture of the underlying rock mass, the P and S wave phases are picked according to the first arrival times of the P and S waves; S32, spatially locating the microseismic event using a collapse grid search positioning method, and simultaneously calculating the onset time of the microseismic event caused by the rock mass fracture in the underlying rock layer of the coal seam floor; S33. Calculate the longitudinal and transverse wave velocity ratios at the location of the microseismic event cluster.

4. The method for detecting potential fluid distribution in the underlying rock strata of a coal seam floor according to claim 3, characterized in that: The longitudinal and transverse wave velocity ratios at the location of the microseismic event cluster are calculated in S33, including: S331, P-wave propagation velocity V of microseismic events P , shear wave propagation velocity V S They are expressed in the following formulas: S332. Calculate the longitudinal and transverse wave velocity ratios at the location of the microseismic event cluster using the following formula: Among them, L is the spatial distance between the earthquake source at the microfracture of the rock stratum beneath the coal seam floor and the microseismic sensor; T is the time of occurrence of the microseismic event, which can be obtained by spatial positioning calculation of the microseismic event; T0 is the first arrival time of the P wave in the microseismic signal, which can be obtained by P wave phase picking; T1 is the first arrival time of the S wave in the microseismic signal, which can be obtained by S wave phase picking.

5. The method for detecting potential fluid distribution in the underlying rock strata of a coal seam floor according to claim 4, characterized in that: In S4, the Poisson's ratio at the rock mass rupture is obtained based on the longitudinal and transverse wave velocity ratios at the location of the microseismic event cluster, including: During coal seam mining, the following formula is used to calculate the dynamic Poisson's ratio σ of the underlying rock layer of the coal seam floor: in, and Then the dynamic Poisson's ratio σ can be expressed as: Where ρ is the velocity of the underground rock medium, and λ and μ are the Lame coefficients.

6. The method for detecting potential fluid distribution in the underlying rock strata of a coal seam floor according to any one of claims 1 to 5, characterized in that: The microseismic monitoring system comprises a mine flameproof and intrinsically safe microseismic monitoring substation (1), the mine flameproof and intrinsically safe microseismic monitoring substation (1) is connected to a mine intrinsically safe microseismic sensor (2), the mine flameproof and intrinsically safe microseismic monitoring substation (1) is connected to a microseismic monitoring server (5) via an industrial-grade optoelectronic switch (3), the microseismic monitoring server (5) is connected to a data processing client (6), and the data processing client (6) is connected to a printer (7); The mine flameproof and intrinsically safe microseismic monitoring substation (1) is connected to an industrial-grade clock signal converter (4), and the industrial-grade clock signal converter (4) is connected to a GPS antenna (9) via a GPS time signal synchronizer (8).

Images