Intelligent judgment and identification method for overlying strata fracture line position of rock burst working face based on micro-seismic monitoring

Abstract

The invention provides a rock burst working face overlying strata fracture line position intelligent identification method for micro-seismic monitoring, and relates to the technical field of rock burst. According to the method, after a working face is normally stoped, the inclination length of the whole working face is selected as a statistical interval, and the unit length is determined; then determining a fitting curve of the overlying strata fracture line of the working face, and micro-seismic event cumulative frequency and cumulative energy of the micro-seismic event in the tendency dimension in the pushing and mining process of the working face, and further determining a micro-seismic cumulative frequency curve and a cumulative energy curve of the tendency dimension of the working face; establishing a working face overlying strata fracture line position prediction model; and finally, predicting a new working face overlying strata fracture line position by adopting the working face overlying strata fracture line position prediction model. According to the method, the overlying strata fracture height of the working face is combined with the frequency and energy of a micro-seismic signal, and the overlying strata fracture height of the working face is quantitatively represented, so that the position of the overlying strata fracture line of the rock burst working face is determined.

Description

Technical Field

[0001] The present invention relates to the field of rock burst technology, and in particular to a method for intelligently identifying the position of a rock burst fault line on a rock burst working face through microseismic monitoring. Background Art

[0002] After the working face of a coal mine is mined, the rock strata within a certain height range of the roof are unloaded and broken and moved in the vertical direction, causing the overlying rock strata to break and move and the surface to subside. The degree of damage and the distribution of cracks gradually decrease from bottom to top, showing the characteristics of gradient damage, which will lead to damage to buildings in the surface subsidence area, and can easily cause safety problems and property losses. Therefore, determining the location of the fault line is an urgent issue to be solved in today's green mining and ecological protection of coal mines, and safe mining of coal seams with impact ground pressure.

[0003] At present, there are few methods for determining the location of the fault line of the working face of a mine, mainly theoretical analysis, numerical simulation calculation and on-site drilling monitoring. However, these methods have the following problems, resulting in poor results in actual engineering applications: (1) Theoretical analysis and numerical simulation calculation methods often simplify the site a lot, which is quite different from the actual complex geological conditions and mechanical properties and cannot correspond well to the actual situation; (2) On-site drilling methods take samples at limited locations, with incomplete information, high cost, and a lot of time, manpower and equipment consumption. It is also dangerous. At the same time, in deep coal mines, underground drilling detection will be limited by environmental factors such as stratum lithology, ground temperature, groundwater, etc., and the high cost makes the underground drilling method unsuitable for deep mines.

[0004] At present, all rock burst mines in my country are equipped with microseismic monitoring systems, which can effectively monitor microseismic events that characterize rock fractures during the mining process. However, there is currently no intelligent method for determining the location of the fault line of the working face based on microseismic monitoring technology. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide an intelligent identification method for the position of the overburden fracture line of the rock burst working face monitored by microseismic monitoring in view of the deficiencies of the above-mentioned prior art, and to determine the position of the fracture line of the working face by using a large amount of microseismic signal data.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: an intelligent identification method of the overburden fracture line position of the rock burst working face by microseismic monitoring, comprising the following steps:

[0007] Step 1: Determine the "unit length" of the working surface inclination dimension;

[0008] After the working face is mined normally, the inclination length of the entire working face is selected as the statistical interval L, and the "unit length" Δl is determined;

[0009] Step 2: Determine the fitting curve of the overburden fracture line of the working face;

[0010] Step 2.1: Establish a coordinate system for the working surface;

[0011] Take the lower left corner of the working face dip dimension section along the advancing direction as the origin, establish a rectangular coordinate system, the horizontal axis represents the dip length of the working face, and the vertical axis represents the overburden height of the working face;

[0012] Step 2.2: Select the horizontal coordinate position of the midpoint of each "unit length" and determine the height of the overburden fracture at the working face at that point by drilling;

[0013] If the statistical interval has N “unit lengths”, then N+1 drilling points on the fault line are obtained. The coordinates of the drilling points are expressed as (x n ,y n ), where x n Indicates the horizontal coordinate of the nth drilling point, y n Indicates the ordinate of the nth drilling point;

[0014] Step 2.3: Determine the parameterized characterization formula of the overburden fracture line of the working face through cubic spline interpolation;

[0015] Step 2.3.1: Take N “unit lengths” as interpolation intervals, where the nth interpolation interval is represented as [x n ,x n+1 ], taking N+1 drilling points as data points for cubic spline interpolation;

[0016] Step 2.3.2: Fit the curve of the overburden fracture line of the working face;

[0017] For each interpolation interval, a cubic polynomial of the overburden fault line is fitted. The cubic polynomial of the overburden fault line fitted in the nth interpolation interval is expressed as:

[0018] S n (x) = a n x 3 +b n x 2 +c n x+d n

[0019] In the formula, S n (x) represents the cubic polynomial fitted to the nth interpolation interval; a n , b n 、c n d n represents the coefficient to be determined; x represents the horizontal coordinate of the inclination dimension of the working face;

[0020] Step 2.3.3: Solve for coefficient an , b n 、c n d n , construct the overall equation group of cubic interpolation curve of overburden fault line, and obtain the parameterized expression of cubic interpolation curve of overburden fault line;

[0021] Step 2.3.3.1: Determine the boundary conditions of the interpolation curve of the overburden fracture line of the working face;

[0022] The boundary condition of the interpolation curve of the overburden fracture line of the working face is that the second-order derivative value of the interpolation curve of the overburden fracture line of the working face at the two end points of the statistical interval is zero, which is expressed as:

[0023] S″(x 0 )=0

[0024] S″(x n )=0

[0025] In the formula, S″(x 0 ) is the cubic polynomial S(x) of the position of the overburden fracture line at the working face at the drilling point x 0 The second-order derivative value at S″(x n ) is the cubic polynomial S(x) of the position of the overburden fracture line at the working face at the drilling point x n The second derivative value at ;

[0026] Step 2.3.3.2: Determine the interpolation conditions of the interpolation curve of the overburden fracture line of the working face;

[0027] The interpolation condition of the interpolation curve of the overburden fracture line of the working face is that the function value of the interpolation curve of the overburden fracture line of the working face at the two end points of the interpolation interval is the ordinate corresponding to the abscissa of the point on the fracture line, expressed as:

[0028] S n (x n )=y n

[0029] S n (x n+1 )=y n+1

[0030] In the formula, S n (x n ) The cubic polynomial of the nth interpolation interval at the drilling point x n The function value at S n (x n+1 ) The cubic polynomial of the nth interpolation interval at the drilling point x n+1 The function value at ;

[0031] Step 2.3.3.3: Construct the equation set of the interpolation curve of the overburden fracture line of the working face in each interpolation interval;

[0032] The boundary conditions and interpolation conditions of the interpolation curve of the overburden fracture line of the working face are brought into the cubic polynomial expression of the overburden fracture line fitted in the interpolation interval, and the linear equation group of the interpolation curve of the overburden fracture line of the working face in each interpolation interval is obtained, which is expressed as:

[0033]

[0034] Step 2.3.3.4: Construct the overall equation system of the cubic interpolation curve of the overburden fault line;

[0035] The equations of each interpolation interval are combined into an overall equation system to obtain the parameterized expression of the cubic interpolation curve of the overburden fault line:

[0036]

[0037] Step 3: Determine the cumulative frequency and cumulative energy of microseismic events in the dip dimension during the mining process of the working face;

[0038] Step 3.1: Establish the coordinate system of the cumulative frequency and cumulative energy of microseismic events;

[0039] Taking the lower left corner of the working face dip dimension section along the advancing direction as the origin, a rectangular coordinate system is established, the horizontal axis represents the dip length of the working face, and the vertical axis represents the cumulative frequency and cumulative energy of microseismic events on the working face;

[0040] Step 3.2: Determine the cumulative frequency and cumulative energy of microseisms in each "unit length" within the statistical interval L, and use N "unit lengths" as the interpolation interval to determine the cumulative frequency and cumulative energy of microseisms in the interpolation interval;

[0041] During the mining process of the working face, determine the cumulative frequency f of microseismic events in each "unit length" Δl within the statistical interval L i and the accumulated energy E i , taking N “unit lengths” as interpolation intervals, the cumulative frequency of microseismic events in the nth interpolation interval is expressed as f n , the cumulative energy of microseismic events is expressed as E n , expressed in coordinate form as (x n ,f n )、(x n ,E n );

[0042] Step 4: Determine the microseismic cumulative frequency curve and cumulative energy curve of the working face inclination dimension;

[0043] Step 4.1: Fit the cumulative frequency curve of microseismicity in the working face;

[0044] Step 4.1.1: Fit a cubic polynomial of the cumulative frequency of microseisms for each interpolation interval. The cubic polynomial of the cumulative frequency of microseisms fitted in the nth interpolation interval is expressed as:

[0045]

[0046] in, is the cubic polynomial of the cumulative frequency of microseismic events in the nth interpolation interval; is the coefficient to be determined; x is the horizontal coordinate of the overburden fault line;

[0047] Step 4.1.2: Solve for the coefficients The overall equation group of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is constructed, and the parameterized expression of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is obtained;

[0048] Step 4.1.2.1: Determine the boundary conditions of the interpolation curve of the cumulative frequency of microseismic inclination of the working face;

[0049] The boundary condition of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is that the second-order derivative value of the interpolation curve of the cumulative frequency of microseismic inclination of the working face at the two end points of the statistical interval is zero, which can be expressed as:

[0050] S f ″ (x 0 )=0

[0051] S f″ (x n )=0

[0052] In the formula, S f″ (x 0 ) is the cumulative frequency interpolation curve of microseismicity in the working face at point x 0 The second derivative value at S f″ (x n ) is the cumulative frequency interpolation curve of microseismicity in the working face at point x n The second derivative value at ;

[0053] Step 4.1.2.2: Determine the interpolation conditions of the interpolation curve of the cumulative frequency of microseismicity of the working face inclination;

[0054] The interpolation condition of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is that the function value of the interpolation curve of the cumulative frequency of microseismic inclination of the working face at the two end points of the interpolation interval is the ordinate corresponding to the abscissa of the point on the fault line, expressed as:

[0055]

[0056] In the formula, The cumulative frequency interpolation curve of microseismic inclination of the working face in the nth interpolation interval isn The function value at The cumulative frequency interpolation curve of microseismic inclination of the working face in the nth interpolation interval is n+1 The function value at ;

[0057] Step 4.1.2.3: Construct the interpolation curve equation group of the cumulative frequency of microseismic inclination of the working face in each interpolation interval;

[0058] Substituting the boundary conditions and interpolation conditions of the interpolation curve of the cumulative frequency of microseismic inclination of the working face into the cubic polynomial expression of the interpolation curve of the cumulative frequency of microseismic inclination of the working face, the linear equation group of the interpolation curve of the cumulative frequency of microseismic inclination of the working face in each interpolation interval is obtained, which is expressed as:

[0059]

[0060] Step 4.1.2.4: Construct the overall equation group of the interpolation curve of the cumulative frequency of microseismicity of the working face;

[0061] The equations of each interpolation interval of the interpolation curve of the cumulative frequency of microseismic inclination of the working face are combined into an overall equation system, and the parameterized expression of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is obtained:

[0062]

[0063] Step 4.2: Fit the cumulative frequency energy of microseismic inclination of the working face;

[0064] Step 4.2.1: Fit a cubic polynomial of microseismic cumulative energy for each interpolation interval. The cubic polynomial of microseismic cumulative energy fitted in the nth interpolation interval is expressed as:

[0065]

[0066] in, is the cubic polynomial of microseismic cumulative energy in the nth interpolation interval; is the coefficient to be determined; x is the horizontal coordinate of the overburden fault line;

[0067] Step 4.2.2: Solve for the coefficients The overall equation group of the interpolation curve of the cumulative energy of microseismic inclination of the working face is constructed, and the parameterized expression of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is obtained;

[0068] Step 4.2.2.1: Determine the boundary conditions of the interpolation curve of microseismic cumulative energy of the working face;

[0069] The boundary condition of the interpolation curve of the cumulative energy of microseismic energy of the working face is that the second-order derivative value of the interpolation curve of the cumulative energy of microseismic energy of the working face at the two end points of the statistical interval is zero, which can be expressed as:

[0070] S E″ (x 0 )=0

[0071] S E″ (x n )=0

[0072] In the formula, S E″ (x 0 ) is the interpolation curve of the cumulative energy of microseismic inclination of the working face at the drilling point x 0 The second derivative value at S E″ (x n ) is the interpolation curve of the cumulative energy of microseismic inclination of the working face at the drilling point x n The second derivative value at ;

[0073] Step 4.2.2.2: Determine the interpolation conditions of the interpolation curve of microseismic cumulative energy of the working face inclination;

[0074] The interpolation condition of the working face inclination microseismic cumulative energy interpolation curve is that the function value of the working face inclination microseismic cumulative energy interpolation curve at the two end points of the interpolation interval is the ordinate corresponding to the abscissa of the point on the fault line, expressed as:

[0075]

[0076] In the formula, The interpolation curve of the microseismic cumulative energy of the working face in the nth interpolation interval is n The function value at The interpolation curve of the microseismic cumulative energy of the working face in the nth interpolation interval is n+1 The function value at ;

[0077] Step 4.2.2.3: Construct the equation group of the interpolation curve of microseismic cumulative energy of the working face inclination;

[0078] Substituting the boundary conditions and interpolation conditions of the interpolation curve of microseismic cumulative energy of the working face inclination into the cubic polynomial expression of the interpolation curve of microseismic cumulative energy of the working face inclination, the linear equation group of the interpolation curve of microseismic cumulative energy of the working face inclination in each interpolation interval is obtained, which is expressed as:

[0079]

[0080] Step 4.2.2.4: Construct the overall equation set of the interpolation curve of the cumulative frequency of microseismicity of the working face;

[0081] The equations of each interpolation interval are combined into an overall equation system to obtain the parameterized expression of the cumulative frequency interpolation curve of microseismic inclination of the working face:

[0082]

[0083] Step 5: Establish a prediction model for the position of the overburden fracture line of the working face;

[0084] Step 5.1: Determine the training samples;

[0085] In the statistical interval L, M drilling points are selected as training samples. The horizontal coordinate of the training sample is represented by x m , m = 1, 2, ..., M, and the corresponding working face overburden fracture height is expressed as S(x m ), the cumulative frequency of microseismic events is expressed as S f (x m ), the microseismic cumulative energy is expressed as S E (x m );

[0086] Step 5.2: Design a neural network model architecture for predicting the location of the overburden fracture line of the working face;

[0087] Step 5.2.1: Determine the basic structure of the neural network for predicting the position of the overburden fracture line of the working face;

[0088] Step 5.2.1.1: Determine the input layer size and feature vector of the neural network for predicting the position of the overburden fracture line of the working face;

[0089] The feature vector input to the neural network input layer is expressed as:

[0090] X m =[S f (x m ),S E (x m ),Δx m ,w,h,D]

[0091]

[0092] In the formula, represents the input layer of the neural network; X m represents the feature vector of the mth training sample; Δx m represents the distance from the mth training sample to the midpoint of the working surface inclination dimension; w represents the width of the working surface; h represents the height of the working surface; D represents the buried depth of the working surface;

[0093] Step 5.2.1.2: Design the intermediate layer structure of the neural network for predicting the position of the overburden fracture line of the working face;

[0094] Step 5.2.1.2.1: Determine the number of intermediate layers of the neural network;

[0095] Assume that the neural network has a total of L layers, and the middle layer includes L-1 layers;

[0096] Step 5.2.1.2.2: Determine the number of neurons in each layer of the neural network;

[0097] According to the "empirical formula", the number of neurons in each layer of the middle layer is determined by the following calculation formula:

[0098]

[0099] Among them, N h l Represents the number of neurons in the lth layer of the neural network; N i l N represents the number of input neurons in the lth layer of the neural network; o l represents the number of neurons in the output layer of the neural network layer l; M represents the number of training samples; β represents any variable value that can be taken by itself, usually ranging from 2 to 10;

[0100] Step 5.2.1.2.3: Determine the weights and biases of the intermediate layers of the neural network;

[0101] In a neural network, the middle layer calculates the process from input to output through forward propagation. The output of each layer is the result of a linear transformation of the output of the previous layer and then the transformation of the activation function. The formula is expressed as:

[0102]

[0103] Where W l represents the weight matrix of the lth layer of the neural network; b l Represents the bias vector of the lth layer of the neural network; Represents the output result of the mth training sample after the l-1th layer of the neural network; Represents the activation output of the mth training sample in the l-1th layer of the neural network The result obtained after linear transformation; express The result obtained after transformation by activation function σ;

[0104] Step 5.2.1.3: Determine the size of the neural network output layer;

[0105] The output layer of the neural network is the predicted overburden fracture height of the working face corresponding to the mth “unit length” of the working face, expressed as:

[0106]

[0107] In the formula, H m It indicates the height of the overburden fracture of the working face corresponding to the mth “unit length” of the predicted working face;

[0108] Step 5.2.2: Initialize the neural network model parameters;

[0109] Step 5.2.2.1: Calculate the loss function value;

[0110] Select mean square error as the loss function, and the calculation formula is:

[0111]

[0112] In the formula, Loss represents the loss value;

[0113] Step 5.2.2.2: Calculate the gradient;

[0114]

[0115] in, Represents the loss function Loss for the weight matrix W of the l-th layer neural network l The gradient of Represents the loss function Loss for the l-th layer neural network bias vector b l The gradient of

[0116] Step 5.2.2.3: Iteratively update the neural network model parameters;

[0117] Use the gradient descent method to update the weight matrix and bias vector. The calculation formula is:

[0118]

[0119] Among them, t represents the number of iterations, α represents the correction coefficient, which is used to control the update of the weight matrix W of the l-th layer neural network. l and the bias vector b of the lth layer neural network l the step size in the process;

[0120] Step 5.2.2.4: Repeat steps 5.2.2.2 and 5.2.2.3 until the condition for stopping iteration is met, as shown in the following formula:

[0121]

[0122] In the formula, express The infinite norm of ; express The infinite norm of 1 , ε 2 Indicates the set threshold;

[0123] Step 6: Use the working face overburden fracture line position prediction model to predict the overburden fracture line position of the new working face;

[0124] Step 6.1: Divide the inclination dimension of the working face to be predicted into Q “unit lengths”, where the coordinates of the midpoint of the qth unit length are represented by x q ;

[0125] Step 6.2: Collect working face data;

[0126] Step 6.2.1: Determine the physical information of the working surface;

[0127] Determine the inclined length, height, and buried depth of the working face, represented by x pred ,h pred , D pred ;

[0128] Determine the distance Δx between the measuring point and the midpoint of the working surface inclination length m pred ;

[0129] Step 6.2.2: Collect microseismic data during the mining process of the working face;

[0130] The microseismic monitoring equipment is used to collect the microseismic signals during the mining process of the working face, and the microseismic cumulative frequency and microseismic cumulative energy of the qth “unit length” of the dip dimension during the mining process of the working face are determined. express;

[0131] Step 6.3: Use the working face overburden fracture line position prediction model to predict the fracture line position of the working face to be predicted;

[0132] The data collected in step 6.2 are input into the neural network model of the working face overburden fracture line position prediction model. The input data is expressed as:

[0133]

[0134] in, is the input data of the mth sample;

[0135] The fracture line position of the working face to be predicted is obtained as shown in the following formula:

[0136]

[0137] In the formula, Represents the fracture height corresponding to the midpoint of the qth "unit length".

[0138] The beneficial effect of adopting the above-mentioned technical scheme is that the method for intelligently identifying the position of the overburden fracture line of the rock burst working face monitored by microseismic monitoring provided by the present invention proposes a data-driven method, which uses a large amount of microseismic signal data to solve the problem of determining the position of the working face fracture line that is difficult to solve by analytical methods, and combines the height of the overburden fracture of the working face with the frequency and energy of the microseismic signal to quantitatively characterize the height of the overburden fracture of the working face, thereby determining the position of the overburden fracture line of the rock burst working face, and providing scientific and powerful guidance for actual engineering. BRIEF DESCRIPTION OF THE DRAWINGS

[0139] Figure 1 A flow chart of a method for intelligently identifying the position of overburden fracture lines in a rock burst working face monitored by microseismic monitoring according to an embodiment of the present invention;

[0140] Figure 2 A schematic diagram of overburden fracture of a working face provided by an embodiment of the present invention;

[0141] Figure 3 A fitting curve diagram of the overburden fracture line of the working face provided by an embodiment of the present invention;

[0142] Figure 4 A fitting curve diagram of the cumulative frequency of microseismic events on a working surface provided by an embodiment of the present invention;

[0143] Figure 5 A fitting curve diagram of the microseismic cumulative energy of a working face provided by an embodiment of the present invention;

[0144] Figure 6 A schematic diagram of the structure of a working face overburden fracture line position prediction model provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0145] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0146] This embodiment is aimed at the N115 working face of a certain mine. The length of the working face in the dip dimension is 200m, the height is 4m, 350m of the working face has been mined, impact has occurred, and the mine has installed a microseismic monitoring system. The method for intelligent identification of the overburden fracture line of the impact ground pressure working face using microseismic monitoring of the present invention is used to determine the position of the fracture line of the overburden strata on the working face.

[0147] In this embodiment, the method for intelligently identifying the position of the overburden fracture line of the rock burst working face monitored by microseismic monitoring is as follows: Figure 1 , 2 As shown, the following steps are included:

[0148] Step 1: Determine the "unit length" of the working surface inclination dimension;

[0149] After the working face is mined normally, the inclination length of the entire working face is selected as the statistical interval L, and the "unit length" Δl is determined;

[0150] In this embodiment, the dip length of the entire working face is selected as the statistical interval L of 200m, and the "unit length" Δl is determined to be 20m;

[0151] Step 2: Determine the fitting curve of the overburden fracture line of the working face;

[0152] Step 2.1: Establish a coordinate system for the working surface;

[0153] Take the lower left corner of the working face dip dimension section along the advancing direction as the origin, establish a rectangular coordinate system, the horizontal axis represents the dip length of the working face, and the vertical axis represents the overburden height of the working face;

[0154] Step 2.2: Select the horizontal coordinate position of the midpoint of each "unit length" and determine the height of the overburden fracture at the working face at that point by drilling;

[0155] If the statistical interval has N “unit lengths”, then N+1 drilling points on the fault line are obtained. The coordinates of the drilling points are expressed as (x n ,y n ), where x n Indicates the horizontal coordinate of the nth drilling point, y n Indicates the ordinate of the nth drilling point;

[0156] In this embodiment, the statistical interval is set to have a total of 10 "unit lengths", and 11 drilling points on the fault line are obtained, and the coordinates of the drilling points are (0,0), (20,8), (40,17), (60,27), (80,38), (100,37), (120,39), (140,29), (160,18), (180,7), (200,0);

[0157] Step 2.3: Determine the parameterized characterization formula of the overburden fracture line of the working face through cubic spline interpolation;

[0158] Step 2.3.1: Take N “unit lengths” as interpolation intervals, where the nth interpolation interval is represented as [x n ,x n+1 ], taking N+1 drilling points as data points for cubic spline interpolation;

[0159] Step 2.3.2: Fit the curve of the overburden fracture line of the working face;

[0160] For each interpolation interval, a cubic polynomial of the overburden fault line is fitted. The cubic polynomial of the overburden fault line fitted in the nth interpolation interval is expressed as:

[0161] S n (x) = a n x 3 +b n x 2 +c n x+d n

[0162] In the formula, S n (x) represents the cubic polynomial fitted to the nth interpolation interval; a n 、b n 、c n d n represents the coefficient to be determined; x represents the horizontal coordinate of the inclination dimension of the working face;

[0163] Step 2.3.3: Solve for coefficient a n 、b n 、c n d n , construct the overall equation group of cubic interpolation curve of overburden fault line, and obtain the parameterized expression of cubic interpolation curve of overburden fault line;

[0164] Step 2.3.3.1: Determine the boundary conditions of the interpolation curve of the overburden fracture line of the working face;

[0165] The boundary condition of the interpolation curve of the overburden fracture line of the working face is that the second-order derivative value of the interpolation curve of the overburden fracture line of the working face at the two end points of the statistical interval is zero, which is expressed as:

[0166] S″(x 0 )=0

[0167] S″(x n )=0

[0168] In the formula, S″(x 0 ) is the cubic polynomial S(x) of the position of the overburden fracture line at the working face at the drilling point x 0 The second-order derivative value at S″(x n ) is the cubic polynomial S(x) of the position of the overburden fracture line at the working face at the drilling point x n The second derivative value at ;

[0169] Step 2.3.3.2: Determine the interpolation conditions of the interpolation curve of the overburden fracture line of the working face;

[0170] The interpolation condition of the interpolation curve of the overburden rock fracture line of the working face is that the function value of the interpolation curve of the overburden rock fracture line of the working face at the two end points of the interpolation interval is the ordinate corresponding to the abscissa of the point on the fracture line, expressed as:

[0171] S n (x n )=yn

[0172] S n (x n+1 )=y n+1

[0173] In the formula, S n (x n ) The cubic polynomial of the nth interpolation interval at the drilling point x n The function value at S n (x n+1 ) The cubic polynomial of the nth interpolation interval at the drilling point x n+1 The function value at ;

[0174] Step 2.3.3.3: Construct the equation set of the interpolation curve of the overburden fracture line of the working face in each interpolation interval;

[0175] The boundary conditions and interpolation conditions of the interpolation curve of the overburden fracture line of the working face are brought into the cubic polynomial expression of the overburden fracture line fitted in the interpolation interval, and the linear equation group of the interpolation curve of the overburden fracture line of the working face in each interpolation interval is obtained, which is expressed as:

[0176]

[0177] Step 2.3.3.4: Construct the overall equation system of the cubic interpolation curve of the overburden fault line;

[0178] The equations of each interpolation interval are combined into an overall equation system to obtain the parameterized expression of the cubic interpolation curve of the overburden fault line:

[0179]

[0180] In this embodiment, the overburden fracture of the working face is as follows: Figure 2 As shown in the figure, the constructed working face overburden fracture line fitting curve is as follows: Figure 3 shown.

[0181] This embodiment selects the seventh microseismic cumulative frequency interpolation interval as an example for detailed calculation and explanation. The interpolation interval is [x 7 ,x 8 ], let the cubic polynomial of the interpolation interval be

[0182] S 3 (x) = a 3 x 3 +b 3 x 2 +c 3 x+d 3

[0183] Let's solve the coefficient a 3 、b 3、c 3 d 3 ,

[0184] Solving the boundary conditions, the second-order derivatives at both ends of the interpolation interval are zero, that is,

[0185] S″(x 0 )=0

[0186] S″(x 11 )=0

[0187] That is, S″(x 0 )=6a 3 x 0 +2b 3 =0, S″(x 11 )=6a 3 x 11 +2b 3 =0

[0188] The interpolation condition in the third interpolation interval is

[0189] S 3 (x 3 )=y 3

[0190] S 3 (x 4 )=y 4

[0191] That is S 3 (x 3 )=a 3 x 3 3 +b 3 x 3 2 +c 3 x 3 +d 3 =y 3 , S 3 (x 4 )=a 3 x 4 3 +b 3 x 4 2 +c 3 x 4 +d 3 =y 4 ;

[0192] By analogy, the calculated breakline interpolation curve is:

[0193]

[0194] Step 3: Determine the cumulative frequency and cumulative energy of microseismic events in the dip dimension during the mining process of the working face;

[0195] Step 3.1: Establish the coordinate system of the cumulative frequency and cumulative energy of microseismic events;

[0196] Taking the lower left corner of the working face dip dimension section along the advancing direction as the origin, a rectangular coordinate system is established, the horizontal axis represents the dip length of the working face, and the vertical axis represents the cumulative frequency and cumulative energy of microseismic events on the working face;

[0197] Step 3.2: Determine the cumulative frequency and cumulative energy of microseisms in each "unit length" within the statistical interval L, and use N "unit lengths" as the interpolation interval to determine the cumulative frequency and cumulative energy of microseisms in the interpolation interval;

[0198] During the mining process of the working face, determine the cumulative frequency f of microseismic events in each "unit length" Δl = 20m in the statistical interval L i and the accumulated energy E i , taking N “unit lengths” as interpolation intervals, the cumulative frequency of microseismic events in the nth interpolation interval is expressed as f n , the cumulative energy of microseismic events is expressed as E n , expressed in coordinate form as (x n ,f n )、(x n ,E n );

[0199] In this embodiment, the cumulative frequencies of microseismic events are (0, 30), (20, 57), (40, 112), (60, 99), (80, 98), (100, 105), (120, 102), (140, 98), (160, 111), (180, 49), and (200, 27);

[0200] The cumulative energies of microseismic events are (0,2205.69), (20,26996.22), (40,61845.86), (60,76674.30), (80,229916.79), (100,108252.23), (120,126769.05), (140,127858.20), (160,147339.96), (180,162280.65), (200,14338.40);

[0201] Step 4: Determine the microseismic cumulative frequency curve and cumulative energy curve of the working face inclination dimension;

[0202] Step 4.1: Fit the cumulative frequency curve of microseismicity in the working face;

[0203] Step 4.1.1: Fit a cubic polynomial of the cumulative frequency of microseisms for each interpolation interval. The cubic polynomial of the cumulative frequency of microseisms fitted in the nth interpolation interval is expressed as:

[0204]

[0205] in, is the cubic polynomial of the cumulative frequency of microseismic events in the nth interpolation interval; is the coefficient to be determined; x is the horizontal coordinate of the overburden fault line;

[0206] Step 4.1.2: Solve for the coefficients The overall equation group of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is constructed, and the parameterized expression of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is obtained;

[0207] Step 4.1.2.1: Determine the boundary conditions of the interpolation curve of the cumulative frequency of microseismic inclination of the working face;

[0208] The boundary condition of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is that the second-order derivative value of the interpolation curve of the cumulative frequency of microseismic inclination of the working face at the two end points of the statistical interval is zero, which can be expressed as:

[0209] S f″ (x 0 )=0

[0210] S f″ (x n )=0

[0211] In the formula, S f″ (x 0 ) is the cumulative frequency interpolation curve of microseismicity in the working face at point x 0 The second derivative value at S f″ (x n ) is the cumulative frequency interpolation curve of microseismicity in the working face at point x n The second derivative value at ;

[0212] Step 4.1.2.2: Determine the interpolation conditions of the interpolation curve of the cumulative frequency of microseismicity of the working face inclination;

[0213] The interpolation condition of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is that the function value of the interpolation curve of the cumulative frequency of microseismic inclination of the working face at the two end points of the interpolation interval is the ordinate corresponding to the abscissa of the point on the fault line, expressed as:

[0214]

[0215] In the formula, The cumulative frequency interpolation curve of microseismic inclination of the working face in the nth interpolation interval isn The function value at The cumulative frequency interpolation curve of microseismic inclination of the working face in the nth interpolation interval is n+1 The function value at ;

[0216] Step 4.1.2.3: Construct the interpolation curve equation group of the cumulative frequency of microseismic inclination of the working face in each interpolation interval;

[0217] Substituting the boundary conditions and interpolation conditions of the interpolation curve of the cumulative frequency of microseismic inclination of the working face into the cubic polynomial expression of the interpolation curve of the cumulative frequency of microseismic inclination of the working face, the linear equation group of the interpolation curve of the cumulative frequency of microseismic inclination of the working face in each interpolation interval is obtained, which is expressed as:

[0218]

[0219] Step 4.1.2.4: Construct the overall equation group of the interpolation curve of the cumulative frequency of microseismicity of the working face;

[0220] The equations of each interpolation interval of the interpolation curve of the cumulative frequency of microseismic inclination of the working face are combined into an overall equation system, and the parameterized expression of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is obtained:

[0221]

[0222] In this embodiment, the interpolation curve of the cumulative frequency of microseismic inclination of the working face is expressed as:

[0223]

[0224] Step 4.2: Fit the cumulative frequency energy of microseismic inclination of the working face;

[0225] Step 4.2.1: Fit a cubic polynomial of microseismic cumulative energy for each interpolation interval. The cubic polynomial of microseismic cumulative energy fitted in the nth interpolation interval is expressed as:

[0226]

[0227] in, is the cubic polynomial of microseismic cumulative energy in the nth interpolation interval; is the coefficient to be determined; x is the horizontal coordinate of the overburden fault line;

[0228] Step 4.2.2: Solve for the coefficients The overall equation group of the interpolation curve of the cumulative energy of microseismic inclination of the working face is constructed, and the parameterized expression of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is obtained;

[0229] Step 4.2.2.1: Determine the boundary conditions of the interpolation curve of microseismic cumulative energy of the working face;

[0230] The boundary condition of the interpolation curve of the cumulative energy of microseismic energy of the working face is that the second-order derivative value of the interpolation curve of the cumulative energy of microseismic energy of the working face at the two end points of the statistical interval is zero, which can be expressed as:

[0231] S E″ (x 0 )=0

[0232] S E″ (x n )=0

[0233] In the formula, S E″ (x 0 ) is the interpolation curve of the cumulative energy of microseismic inclination of the working face at the drilling point x 0 The second derivative value at S E″ (x n ) is the interpolation curve of the cumulative energy of microseismic inclination of the working face at the drilling point x n The second derivative value at ;

[0234] Step 4.2.2.2: Determine the interpolation conditions of the interpolation curve of microseismic cumulative energy of the working face inclination;

[0235] The interpolation condition of the working face inclination microseismic cumulative energy interpolation curve is that the function value of the working face inclination microseismic cumulative energy interpolation curve at the two end points of the interpolation interval is the ordinate corresponding to the abscissa of the point on the fault line, expressed as:

[0236]

[0237] In the formula, The interpolation curve of the microseismic cumulative energy of the working face in the nth interpolation interval is n The function value at The interpolation curve of the microseismic cumulative energy of the working face in the nth interpolation interval is n+1 The function value at ;

[0238] Step 4.2.2.3: Construct the equation group of the interpolation curve of microseismic cumulative energy of the working face inclination;

[0239] Substituting the boundary conditions and interpolation conditions of the interpolation curve of microseismic cumulative energy of the working face inclination into the cubic polynomial expression of the interpolation curve of microseismic cumulative energy of the working face inclination, the linear equation group of the interpolation curve of microseismic cumulative energy of the working face inclination in each interpolation interval is obtained, which is expressed as:

[0240]

[0241] Step 4.2.2.4: Construct the overall equation set of the interpolation curve of the cumulative frequency of microseismicity of the working face;

[0242] The equations of each interpolation interval are combined into an overall equation system to obtain the parameterized expression of the cumulative frequency interpolation curve of microseismic inclination of the working face:

[0243]

[0244] In this embodiment, the expression for the interpolation curve of the cumulative frequency of microseismic inclination of the working face is obtained as follows:

[0245]

[0246] The fitting curve of the cumulative frequency of microseismic inclination of the working face is constructed as follows: Figure 4 As shown, the cumulative energy fitting curve is as follows Figure 5 As shown;

[0247] Step 5: Establish a prediction model for the position of the overburden fracture line of the working face;

[0248] Step 5.1: Determine the training samples;

[0249] In the statistical interval L, M drilling points are selected as training samples. The horizontal coordinate of the training sample is represented by x m , m = 1, 2, ..., M, and the corresponding working face overburden fracture height is expressed as S(x m ), the cumulative frequency of microseismic events is expressed as S f (x m ), the microseismic cumulative energy is expressed as S E (x m );

[0250] In this embodiment, 100 points are selected as training samples in the statistical interval L=200m, as shown in Table 1:

[0251] Table 1 Training sample points

[0252]

[0253]

[0254]

[0255] Step 5.2: Design a neural network model architecture for predicting the location of the overburden fracture line of the working face;

[0256] Step 5.2.1: Determine the basic structure of the neural network for predicting the position of the overburden fracture line of the working face;

[0257] Step 5.2.1.1: Determine the input layer size and feature vector of the neural network for predicting the position of the overburden fracture line of the working face;

[0258] In this embodiment, the size of the neural network input layer is 6, and the input feature vector is expressed as:

[0259] X m =[S f (x m ),S E (x m ),Δx m ,w,h,D]

[0260]

[0261] In the formula, represents the input layer of the neural network; X m represents the feature vector of the mth training sample; Δx m represents the distance from the mth training sample to the midpoint of the working surface inclination dimension; w represents the width of the working surface; h represents the height of the working surface; D represents the buried depth of the working surface;

[0262] Take the 10th training sample as an example:

[0263]

[0264] Step 5.2.1.2: Design the intermediate layer structure of the neural network for predicting the position of the overburden fracture line of the working face;

[0265] Step 5.2.1.2.1: Determine the number of intermediate layers of the neural network;

[0266] Assume that the neural network has a total of L layers, and the middle layer includes L-1 layers;

[0267] In this embodiment, it is assumed that the neural network has 4 layers, and the middle layer includes 3 layers;

[0268] Step 5.2.1.2.2: Determine the number of neurons in each layer of the neural network;

[0269] According to the "empirical formula", the number of neurons in each layer of the middle layer is determined by the following calculation formula:

[0270]

[0271] Among them, N h l Represents the number of neurons in the lth layer of the neural network; N i l N represents the number of input neurons in the lth layer of the neural network; o l represents the number of neurons in the output layer of the neural network layer l; M represents the number of training samples; β represents any variable value that can be taken by itself, usually ranging from 2 to 10;

[0272] In this embodiment, the number of neurons in the first layer is calculated to be 2, the number of neurons in the second layer is 3, and the number of neurons in the third layer is 3; β is 5;

[0273] Step 5.2.1.2.3: Determine the weights and biases of the intermediate layers of the neural network;

[0274] In a neural network, the middle layer calculates the process from input to output through forward propagation. The output of each layer is the result of a linear transformation of the output of the previous layer and then the transformation of the activation function. The formula is expressed as:

[0275]

[0276] Where W l represents the weight matrix of the lth layer of the neural network; b l Represents the bias vector of the lth layer of the neural network; Represents the output result of the mth training sample after the l-1th layer of the neural network; Represents the activation function of the mth training sample in the l-1th layer of the neural network The result obtained after linear transformation; express The result obtained after transformation by activation function σ;

[0277] Step 5.2.1.3: Determine the size of the neural network output layer:

[0278] The output layer of the neural network is the predicted overburden fracture height of the working face corresponding to the mth “unit length” of the working face, expressed as:

[0279]

[0280] In the formula, H m It indicates the height of the overburden fracture of the working face corresponding to the mth “unit length” of the predicted working face;

[0281] In this embodiment, the neural network structure for predicting the position of the overburden fracture line of the working face is as follows: Figure 6 shown.

[0282] Step 5.2.2: Initialize the neural network model parameters;

[0283] In this embodiment, the initialized weight matrix is:

[0284]

[0285] W 4 =[-0.1801 0.2394 -0.1503]

[0286] b1 =[-0.1145 -0.3952]

[0287] b 2 =[0.2532 -0.2307 -0.3764]

[0288] b 3 =[0.4225 0.0416 0.3694]

[0289] b 4 =[-0.5364]

[0290] Step 5.2.2.1: Calculate the loss function value;

[0291] Select mean square error as the loss function, and the calculation formula is:

[0292]

[0293] In the formula, Loss represents the loss value;

[0294] Step 5.2.2.2: Calculate the gradient;

[0295]

[0296] in, Represents the loss function Loss for the weight matrix W of the l-th layer neural network l The gradient of Represents the loss function Loss for the l-th layer neural network bias vector b l The gradient of

[0297] Step 5.2.2.3: Iteratively update the neural network model parameters;

[0298] Use the gradient descent method to update the weight matrix and bias vector. The calculation formula is:

[0299]

[0300] Among them, t represents the number of iterations, α represents the correction coefficient, which is used to control the update of the weight matrix W of the l-th layer neural network. l and the bias vector b of the lth layer neural network l The step size in the process, in this embodiment, α is set to 0.01;

[0301] Step 5.2.2.4: Repeat steps 5.2.2.2 and 5.2.2.3 until the condition for stopping iteration is met, as shown in the following formula:

[0302]

[0303] In the formula, express The infinite norm of ; express The infinite norm of 1 , ε 2 Indicates the set threshold;

[0304] In this embodiment, set v 1 =0.02,ε 2 = 0.02, the number of iterations is t = 20000, and the calculated weight matrix and bias vector are:

[0305]

[0306] Step 6: Use the working face overburden fracture line position prediction model to predict the overburden fracture line position of the new working face;

[0307] Step 6.1: Divide the inclination dimension of the working face to be predicted into Q “unit lengths”, where the coordinates of the midpoint of the qth unit length are represented by x q ;

[0308] Step 6.2: Collect working face data;

[0309] Step 6.2.1: Determine the physical information of the working surface;

[0310] Determine the inclined length, height, and buried depth of the working face, represented by x pred ,h pred , D pred ;

[0311] Determine the distance Δx between the measuring point and the midpoint of the working surface inclination length m pred ;

[0312] Step 6.2.2: Collect microseismic data during the mining process of the working face;

[0313] The microseismic monitoring equipment is used to collect the microseismic signals during the mining process of the working face, and the microseismic cumulative frequency and microseismic cumulative energy of the qth “unit length” of the dip dimension during the mining process of the working face are determined. express;

[0314] Step 6.3: Use the working face overburden fracture line position prediction model to predict the fracture line position of the working face to be predicted;

[0315] The data collected in step 6.2 are input into the neural network model of the working face overburden fracture line position prediction model. The input data is expressed as:

[0316]

[0317] in, is the input data of the mth sample;

[0318] The fracture line position of the working face to be predicted is obtained as shown in the following formula:

[0319]

[0320] In the formula, Represents the fracture height corresponding to the midpoint of the qth "unit length".

[0321] In this embodiment, the inclination length x of the working surface is determined pred =200, height h pred =4, burial depth D pred =800, cumulative frequency of microseismic events Cumulative energy of microseismic events Distance Δx from the midpoint of the working surface inclination dimension m =2.31;

[0322] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. 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 or all of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope defined by the claims of the present invention.

Claims

1. An intelligent identification method for the position of overburden fracture lines in rock burst working faces using microseismic monitoring, characterized in that: The following steps are involved: Step 1: Determine the "unit length" of the working surface inclination dimension; After the working face is mined normally, the inclination length of the entire working face is selected as the statistical interval L, and the "unit length" Δl is determined; Step 2: Determine the fitting curve of the overburden fracture line of the working face; Step 3: Determine the cumulative frequency and cumulative energy of microseismic events in the dip dimension during the mining process of the working face; Step 4: Determine the microseismic cumulative frequency curve and cumulative energy curve of the working face inclination dimension; Step 5: Establish a prediction model for the position of the overburden fracture line of the working face; The neural network for predicting the position of the overburden fault line of the working face has a total of L layers, and the input feature vector includes the distance from the training sample to the midpoint of the working face dip dimension, the width of the working face, the height of the working face, the buried depth of the working face, and the microseismic cumulative frequency and microseismic cumulative energy of microseismic events in the dip dimension during the working face mining process; The middle layer includes L-1 layer, and the number of neurons in each layer of the middle layer is determined according to the "empirical formula"; In a neural network, the middle layer calculates the process from input to output through forward propagation. The output of each layer is the result of a linear transformation of the output of the previous layer and then the transformation through an activation function. The output layer of the neural network is the predicted overburden fracture height of the working face corresponding to the mth "unit length" of the working face; The neural network selects mean square error as a loss function, calculates the gradient of the loss function to the weight matrix and the bias vector, and iteratively updates the weight matrix and the bias vector; Step 6: Use the working face overburden fracture line position prediction model to predict the position of the new working face overburden fracture line.

2. The method for intelligently identifying the position of the overburden fracture line of the rock burst working face by microseismic monitoring according to claim 1 is characterized by: The specific method of step 2 is: Step 2.1: Establish a coordinate system for the working surface; Take the lower left corner of the working face dip dimension section along the advancing direction as the origin, establish a rectangular coordinate system, the horizontal axis represents the dip length of the working face, and the vertical axis represents the overburden height of the working face; Step 2.2: Select the horizontal coordinate position of the midpoint of each "unit length" and determine the height of the overburden fracture at that point of the working face by drilling; If the statistical interval has N "unit lengths", then N+1 drilling points on the fault line are obtained. The coordinates of the drilling points are expressed as (x n ,y n ), where x n Indicates the horizontal coordinate of the nth drilling point, y n Indicates the ordinate of the nth drilling point; Step 2.3: Determine the parameterized characterization formula of the overburden fault line of the working face through cubic spline interpolation.

3. The method for intelligently identifying the position of the overburden fracture line of the rock burst working face by microseismic monitoring according to claim 1 is characterized by: The specific method of step 2.3 is: Step 2.3.1: Take N "unit lengths" as interpolation intervals, where the nth interpolation interval is represented as [x n ,x n+1 ], taking N+1 drilling points as data points for cubic spline interpolation; Step 2.3.2: Fit the curve of the overburden fracture line of the working face; For each interpolation interval, a cubic polynomial of the overburden fault line is fitted. The cubic polynomial of the overburden fault line fitted in the nth interpolation interval is expressed as: S n (x)=a n x 3 +b n x 2 +c n x+d n In the formula, S n (x) represents the cubic polynomial fitted to the nth interpolation interval; a n , b n 、c n ,d n represents the coefficient to be determined; x represents the horizontal coordinate of the inclination dimension of the working face; Step 2.3.3: Solve for coefficient a n , b n 、c n ,d n , the overall equation group of cubic interpolation curve of overburden fault line is constructed, and the parameterized expression of cubic interpolation curve of overburden fault line is obtained.

4. The method for intelligently identifying the position of the overburden fracture line of the rock burst working face by microseismic monitoring according to claim 3 is characterized by: The specific method of step 2.3.3 is: Step 2.3.3.1: Determine the boundary conditions of the interpolation curve of the overburden fracture line of the working face; The boundary condition of the interpolation curve of the overburden fracture line of the working face is that the second-order derivative value of the interpolation curve of the overburden fracture line of the working face at the two end points of the statistical interval is zero; Step 2.3.3.2: Determine the interpolation conditions of the interpolation curve of the overburden fracture line of the working face; The interpolation condition of the interpolation curve of the overburden fracture line of the working face is that the function value of the interpolation curve of the overburden fracture line of the working face at the two end points of the interpolation interval is the ordinate corresponding to the abscissa of the point on the fracture line; Step 2.3.3.3: Construct the equation set of the interpolation curve of the overburden fracture line of the working face in each interpolation interval; Substitute the boundary conditions and interpolation conditions of the interpolation curve of the overburden fracture line of the working face into the cubic polynomial expression of the overburden fracture line fitted in the interpolation interval to obtain the linear equation group of the interpolation curve of the overburden fracture line of the working face in each interpolation interval; Step 2.3.3.4: Construct the overall equation system of the cubic interpolation curve of the overburden fault line; The equations of each interpolation interval are combined into an overall equation system to obtain the parameterized expression of the cubic interpolation curve of the overburden fault line.

5. The method for intelligently identifying the position of the overburden fracture line of the rock burst working face by microseismic monitoring according to claim 4 is characterized by: The specific method of step 3 is: Step 3.1: Establish the coordinate system of the cumulative frequency and cumulative energy of microseismic events; Taking the lower left corner of the working face dip dimension section along the advancing direction as the origin, a rectangular coordinate system is established, the horizontal axis represents the dip length of the working face, and the vertical axis represents the cumulative frequency and cumulative energy of microseismic events on the working face; Step 3.2: Determine the cumulative frequency and cumulative energy of microseisms in each "unit length" within the statistical interval L, and use N "unit lengths" as the interpolation interval to determine the cumulative frequency and cumulative energy of microseisms in the interpolation interval; During the mining process of the working face, the cumulative frequency f of microseismic events in each "unit length" Δl within the statistical interval L is determined. i and the accumulated energy E i , taking N "unit lengths" as interpolation intervals, the cumulative frequency of microseismic events in the nth interpolation interval is expressed as f n , the cumulative energy of microseismic events is expressed as E n , expressed in coordinate form as (x n ,f n )、(x n ,E n ).

6. The method for intelligently identifying the position of the overburden fracture line of the rock burst working face by microseismic monitoring according to claim 5 is characterized by: The specific method of step 4 is: Step 4.1: Fit the cumulative frequency curve of microseismicity in the working face; Step 4.1.1: Fit a cubic polynomial of the cumulative frequency of microseisms for each interpolation interval. The cubic polynomial of the cumulative frequency of microseisms fitted in the nth interpolation interval is expressed as: in, is the cubic polynomial of the cumulative frequency of microseismic events in the nth interpolation interval; is the coefficient to be determined; x is the horizontal coordinate of the overburden fault line; Step 4.1.2: Solve for the coefficients The overall equation group of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is constructed, and the parameterized expression of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is obtained; Step 4.2: Fit the cumulative frequency energy of microseismic inclination of the working face; Step 4.2.1: Fit a cubic polynomial of microseismic cumulative energy for each interpolation interval. The cubic polynomial of microseismic cumulative energy fitted in the nth interpolation interval is expressed as: in, is the cubic polynomial of microseismic cumulative energy in the nth interpolation interval; is the coefficient to be determined; x is the horizontal coordinate of the overburden fault line; Step 4.2.2: Solve for the coefficients The overall equation group of the interpolation curve of the cumulative energy of microseismic inclination of the working face is constructed, and the parameterized expression of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is obtained.

7. The method for intelligently identifying the position of the overburden fracture line of the rock burst working face by microseismic monitoring according to claim 6 is characterized by: The specific method of step 4.1.2 is: Step 4.1.2.1: Determine the boundary conditions of the interpolation curve of the cumulative frequency of microseismic inclination of the working face; The boundary condition of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is that the second-order derivative value of the interpolation curve of the cumulative frequency of microseismic inclination of the working face at the two end points of the statistical interval is zero; Step 4.1.2.2: Determine the interpolation conditions of the interpolation curve of the cumulative frequency of microseismicity of the working face inclination; The interpolation condition of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is that the function value of the interpolation curve of the cumulative frequency of microseismic inclination of the working face at the two end points of the interpolation interval is the ordinate corresponding to the abscissa of the point on the fault line; Step 4.1.2.3: Construct the interpolation curve equation group of the cumulative frequency of microseismic inclination of the working face in each interpolation interval; Substitute the boundary conditions and interpolation conditions of the interpolation curve of the cumulative frequency of microseismic inclination of the working face into the cubic polynomial expression of the interpolation curve of the cumulative frequency of microseismic inclination of the working face, and obtain the linear equation group of the interpolation curve of the cumulative frequency of microseismic inclination of the working face in each interpolation interval; Step 4.1.2.4: Construct the overall equation group of the interpolation curve of the cumulative frequency of microseismicity of the working face; The equation group of each interpolation interval of the interpolation curve of the cumulative frequency of microseismic inclination of the working face is combined into an overall equation group to obtain a parameterized expression of the interpolation curve of the cumulative frequency of microseismic inclination of the working face.

8. The method for intelligently identifying the position of the overburden fracture line of the rock burst working face by microseismic monitoring according to claim 7 is characterized by: The specific method of step 4.2.2 is: Step 4.2.2.1: Determine the boundary conditions of the interpolation curve of microseismic cumulative energy of the working face; The boundary condition of the interpolation curve of the cumulative energy of microseismic inclination of the working face is that the second-order derivative value of the interpolation curve of the cumulative energy of microseismic inclination of the working face at the two end points of the statistical interval is zero; Step 4.2.2.2: Determine the interpolation conditions of the interpolation curve of microseismic cumulative energy of the working face inclination; The interpolation condition of the interpolation curve of the microseismic cumulative energy of the working face inclination is that the function value of the interpolation curve of the microseismic cumulative energy of the working face inclination at the two end points of the interpolation interval is the ordinate corresponding to the abscissa of the point on the fault line; Step 4.2.2.3: Construct the equation group of the interpolation curve of microseismic cumulative energy of the working face inclination; Substitute the boundary conditions and interpolation conditions of the interpolation curve of microseismic cumulative energy of the working face inclination into the cubic polynomial expression of the interpolation curve of microseismic cumulative energy of the working face inclination, and obtain the linear equation group of the interpolation curve of microseismic cumulative energy of the working face inclination in each interpolation interval; Step 4.2.2.4: Construct the overall equation set of the interpolation curve of the cumulative frequency of microseismicity of the working face; The equations of each interpolation interval are combined into an overall equation group to obtain the parameterized expression of the cumulative frequency interpolation curve of microseismic inclination of the working face.

9. The method for intelligently identifying the position of overburden fracture lines in rock burst working faces monitored by microseismic monitoring according to claim 8 is characterized by: The specific method of step 6 is: Step 6.1: Divide the inclination dimension of the working face to be predicted into Q "unit lengths", where the coordinates of the midpoint of the qth unit length are represented by x q ; Step 6.2: Collect working face data; Step 6.2.1: Determine the physical information of the working surface; Determine the inclined length, height, and buried depth of the working face, represented by x pred ,h pred , D pred ; Determine the distance Δx between the measuring point and the midpoint of the working surface inclination length m pred ; Step 6.2.2: Collect microseismic data during the mining process of the working face; The microseismic monitoring equipment is used to collect the microseismic signals during the mining process of the working face, and the microseismic cumulative frequency and microseismic cumulative energy of the qth "unit length" of the dip dimension during the mining process of the working face are determined. express; Step 6.3: Use the working face overburden fracture line position prediction model to predict the fracture line position of the working face to be predicted; The data collected in step 6.2 is input into the neural network model of the working face overburden fracture line position prediction model to obtain the fracture line position of the working face to be predicted.

Images