Coal pillar-free self-formed roadway dynamic pressure area management and control method based on intelligent prediction and dynamic regulation

Abstract

The application discloses a coal pillar-free self-formed roadway dynamic pressure area management and control method based on intelligent prediction and dynamic regulation, belongs to the technical field of coal mining, and comprises the following steps: establishing a machine learning database, then performing feature correlation determination, selection and standardization processing on the data in the database to obtain a data set, then establishing and optimizing a dynamic pressure area stable distance prediction model based on the data set, screening to obtain a final model, adding target mine parameters to the final model test set to obtain a dynamic pressure area range prediction result, selecting and designing supporting equipment according to the prediction result, monitoring the stability of the dynamic pressure area roadway in real time, and dynamically adjusting a control strategy according to the monitoring result. The coal pillar-free self-formed roadway dynamic pressure area management and control method based on intelligent prediction and dynamic regulation can provide a scientific basis for dynamic pressure area support design and equipment selection, ensures the safety of workers, improves the monitoring efficiency, ensures the accuracy of monitoring data, and realizes effective management and control of the dynamic pressure area roadway.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of coal mining, and particularly relates to a coal-pillar-free self-formed roadway dynamic pressure zone management and control method based on intelligent prediction and dynamic regulation. BACKGROUND

[0002] The coal-pillar-free self-formed roadway technology effectively improves the stress environment of the roadway surrounding rock by cutting off the stress transmission between the roadway roof and the goaf roof through the roof directional cutting seam technology, and automatically forms a new roadway for the next working face using the mine pressure. This technology greatly improves the resource recovery rate and eliminates the surrounding rock stress concentration problem caused by the coal pillar, which plays a significant role in promoting the sustainable development of coal mining.

[0003] According to the connection relationship of the coal-pillar-free self-formed roadway process in time and space, the surrounding rock structure of the roadway connecting groove can be divided into three regions: the advanced coal body support region, the lagging working face dynamic pressure region, and the formed roadway stable region. Among them, the dynamic pressure region is located within a distance behind the working face and the hydraulic support, and due to the caving of the goaf roof and the movement of the overlying strata, the roadway in this region is in an unstable state. The stability control of the dynamic pressure region roadway is crucial to ensure the success of the coal-pillar-free self-formed roadway.

[0004] The traditional technology mainly relies on post-mining site mine pressure monitoring to determine the stability of the dynamic pressure region roadway, and lacks a corresponding pre-mining prediction method. This leads to the fact that the pre-mining temporary support design and equipment procurement selection mainly rely on engineering experience, and lack theoretical basis and scientific guidance. Since the site mine pressure monitoring of the traditional technology mostly relies on manual operation, it is greatly limited in safety, efficiency and accuracy, and cannot meet the requirements of modern high efficiency and high precision. And for the control of the dynamic pressure region roadway surrounding rock, the traditional technology often adopts a single support form, which is difficult to adapt to the complex and variable site environment requirements. Therefore, a new method is urgently needed. SUMMARY

[0005] The purpose of the present application is to provide a coal-pillar-free self-formed roadway dynamic pressure zone management and control method based on intelligent prediction and dynamic regulation, which can provide scientific basis for dynamic pressure region support design and equipment selection, ensure worker safety, improve monitoring efficiency, ensure monitoring data accuracy, and realize effective management and control of the dynamic pressure region roadway.

[0006] To achieve the above purpose, the present application provides a coal-pillar-free self-formed roadway dynamic pressure zone management and control method based on intelligent prediction and dynamic regulation, comprising the following steps:

[0007] Step 1, collect mine exploitation data, and establish a machine learning database;

[0008] Step 2, correlation determination and feature selection are performed on the data features in the database in step 1, and the data features are standardized to obtain a processed data set;

[0009] Step 3, using the data in the data set in step 2, a prediction model of the stable distance of the dynamic pressure zone is established, and the prediction model is optimized to obtain an evaluation model;

[0010] Step 4, the evaluation model in step 3 is screened to obtain a final model;

[0011] Step 5, before mining, the relevant parameters of the target mine are added to the test set of the final model in step 4 as samples to obtain a dynamic pressure zone range prediction result; according to the prediction result, the dynamic pressure zone supporting equipment selection and support design are performed;

[0012] Step 6, in the mining process, the stability of the dynamic pressure zone roadway is monitored in real time, and the control strategy is dynamically adjusted according to the monitoring result.

[0013] Preferably, in step 1, the machine learning database is established by collecting mine related data including mining depth, coal seam thickness, coal seam inclination, cutting height, cutting angle, roof rock hardness, roadway height, roadway width, working face inclination length, working face strike length, and dynamic pressure stable distance through mine investigation and literature summary.

[0014] Preferably, in step 2, the correlation determination and feature selection of the data features include using three-dimensional modeling software FLAC 3D Orthogonal simulation experiments are carried out to obtain parameters with high correlation with the dynamic pressure stable distance; the data features are standardized, including using Box-Cox transformation technology to standardize the feature engineering database.

[0015] Preferably, in step 3, the prediction model includes an artificial neural network model, which continuously reduces errors through iteration to meet the set learning goal; the optimization processing of the prediction model includes using particle swarm optimization algorithm and genetic optimization algorithm to optimize the prediction model.

[0016] Preferably, in step 4, the evaluation model in step 3 is screened, including using correlation coefficient, mean absolute error, mean absolute percentage error and root mean square error evaluation index to screen the model.

[0017] Preferably, step 5 further includes temporarily supporting the dynamic pressure zone roadway according to the prediction result.

[0018] Preferably, in step 6, the stability of the dynamic pressure zone roadway is monitored in real time, including using a fiber-optic intelligent monitoring device to monitor the stress and displacement of the dynamic pressure zone roadway in real time; and the dynamic adjustment of the control strategy includes timely adjustment of the dynamic pressure zone temporary support design according to the monitoring results.

[0019] A dynamic pressure zone management device for coal pillar-free self-formed roadway based on intelligent prediction and dynamic regulation, comprising:

[0020] A data collection module is configured to collect mine exploitation data and establish a machine learning database.

[0021] A data feature processing module is connected to the data collection module and configured to determine the correlation of data features in the database and select features, and to standardize the data features to obtain a processed data set.

[0022] A model training module is connected to the data feature processing module and configured to use data in the data set to establish a prediction model for the stable distance of the dynamic pressure zone, and to optimize the prediction model to obtain a to-be-evaluated model.

[0023] A model screening module is connected to the model training module and configured to screen the to-be-evaluated model to obtain a final model.

[0024] A result prediction module is connected to the model screening module and configured to add relevant parameters of a target mine as samples to a test set of the final model before exploitation to obtain a prediction result of the range of the dynamic pressure zone, and to select a dynamic pressure zone support device and design a support according to the prediction result.

[0025] A real-time monitoring module is connected to the result prediction module and configured to monitor the stability of the dynamic pressure zone roadway in real time during exploitation, and to dynamically adjust a control strategy according to the monitoring result.

[0026] Preferably, the result prediction module further comprises a temporary support design sub-module configured to design a temporary support for the dynamic pressure zone roadway according to the prediction result.

[0027] Preferably, the computer program, when executed by a processor, implements the steps of any of the above methods.

[0028] Therefore, the present application adopts the above-mentioned coal pillar-free self-formed roadway dynamic pressure zone management method based on intelligent prediction and dynamic regulation, and compared with the prior art, the present application has the following remarkable beneficial effects:

[0029] (1) The method adopted by the present application can predict the stable distance of the coal pillar-free self-formed roadway dynamic pressure zone before exploitation, and provide scientific guidance for dynamic pressure zone support design and equipment selection.

[0030] (2) The method adopted by the present application cancels the manual monitoring in the well by optical fiber intelligent monitoring, which can guarantee the safety of workers, improve the efficiency and ensure the monitoring accuracy;

[0031] (3) The method adopted by the present application can realize dynamic control of the roadway in the dynamic pressure area, and realize safe and efficient mining of the self-formed roadway without coal pillars.

[0032] The technical solutions of the present application will be further described in detail below by means of the drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0033] Figure 1 The present application is a partition diagram of the self-formed roadway without coal pillars based on the intelligent prediction and dynamic control method for the dynamic pressure area of the self-formed roadway without coal pillars;

[0034] Figure 2 The present application is an embodiment flowchart of the intelligent prediction and dynamic control method for the dynamic pressure area of the self-formed roadway without coal pillars;

[0035] Figure 3 The present application is a parameter diagram of the orthogonal simulation experiment method of the intelligent prediction and dynamic control method for the dynamic pressure area of the self-formed roadway without coal pillars;

[0036] Figure 4 The present application is a schematic diagram of the artificial neural network model of the intelligent prediction and dynamic control method for the dynamic pressure area of the self-formed roadway without coal pillars;

[0037] Figure 5 The present application is a flowchart of the learning process of the artificial neural network model of the intelligent prediction and dynamic control method for the dynamic pressure area of the self-formed roadway without coal pillars;

[0038] Figure 6 The present application is a PSO-ANN dynamic pressure stable distance prediction model of the intelligent prediction and dynamic control method for the dynamic pressure area of the self-formed roadway without coal pillars;

[0039] Figure 7 The present application is a GO-ANN dynamic pressure stable distance prediction model of the intelligent prediction and dynamic control method for the dynamic pressure area of the self-formed roadway without coal pillars;

[0040] Figure 8 The present application is a schematic diagram of the temporary support of the single prop and the pi-shaped beam of the intelligent prediction and dynamic control method for the dynamic pressure area of the self-formed roadway without coal pillars, wherein Figure 8 (a) in the above figure represents a sectional view of the temporary support of the single prop and the pi-shaped beam, Figure 8 (b) in the above figure represents an axial sectional view of the temporary support of the single prop and the pi-shaped beam, Figure 8 (c) in the above figure represents an axial plan view of the temporary support of the single prop and the pi-shaped beam;

[0041] Figure 9 This is a schematic diagram of the unit support for the temporary support of the dynamic pressure zone control method for self-forming roadways without coal pillars based on intelligent prediction and dynamic control according to the present invention. Figure 9 (a) in the diagram represents the cross-sectional view of the temporary support of the unit support. Figure 9 (b) in the diagram represents the axial cross-sectional view of the temporary support of the unit bracket. Figure 9 (c) in the figure represents the axial plan view of the temporary support of the unit bracket.

[0042] Figure Labels

[0043] 1. Mining face; 2. Hydraulic support for the working face; 3. Advance coal seam support zone; 4. Lagging working face dynamic pressure zone; 5. Roadway stabilization zone; 6. Roof cutting line; 7. Goaf; 8. Single prop; 9. π-shaped beam; 10. Unit support; 11. Continuing working face; 12. Constant resistance anchor cable; 13. Rockfill support. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Unless otherwise defined, the technical or scientific terms used in the present invention should have the ordinary meaning understood by those skilled in the art.

[0045] Example 1

[0046] like Figure 1 As shown, mining face 1 is the working area where coal mining is currently underway. A hydraulic support 2 is installed on one side of mining face 1 to support the roof and protect the safety of workers and equipment. Based on the temporal and spatial continuity of the pillarless self-forming roadway process, the surrounding rock structure of the roadway can be divided into three zones: the leading coal body support zone 3, the lagging working face dynamic pressure zone 4, and the roadway stabilization zone 5. The leading coal body support zone 3 is located in the unmined coal body area in front of mining face 1. The lagging working face dynamic pressure zone 4 is located behind mining face 1 and the hydraulic support 2, and the roadway stabilization zone 5 is located behind the lagging working face dynamic pressure zone 4. The roof cutting line 6 marks the location of the directional cut in the roof. The goaf 7 is located behind mining face 1 and is the area where coal mining has been completed.

[0047] like Figure 2 As shown, the implementation process of the present invention's method for controlling dynamic pressure zones in self-forming roadways without coal pillars, based on intelligent prediction and dynamic regulation, includes the following steps:

[0048] Step 1, through mine investigation and literature summary, collect multiple mine related data, including mining depth, coal seam thickness, coal seam dip angle, cutting height, cutting angle, roof rock hardness, roadway height, roadway width, working face inclination length, working face strike length, dynamic pressure stable distance, etc. Organize the collected data, screen out abnormal data, and establish a machine learning database.

[0049] Step 2, use three-dimensional modeling software FLAC 3D Carry out orthogonal simulation test, analyze the influence of mining depth, coal seam thickness, coal seam dip angle, cutting height, cutting angle, roof rock hardness, roadway height, roadway width, etc. on dynamic pressure stable distance. As shown in Figure 2 , select the factors related to dynamic pressure stable distance, carry out multi-factor (m) multi-level (n) orthogonal simulation experiment, carry out n m times simulation experiment, get the parameters related to dynamic pressure stable distance.

[0050] After determining the related parameters by simulation, standardize the data as input parameters. Use Box-Cox transformation technology to preprocess the input variable database. Box-Cox transformation is to normalize the same kind of samples, so that the dispersed data is concentrated in a small area, greatly improving the classification accuracy. Since the transformation is performed on the same kind of samples, the independence of the data can be guaranteed, and the distribution of each kind of data after transformation is closer to normality. The general form of Box-Cox transformation is:

[0051]

[0052] Where y is a continuous variable, λ is a transformation parameter, different λ corresponds to different transformation methods, the specific transformation method can be determined by solving λ value, the estimation method of λ value can use maximum likelihood estimation, y (λ) is the new data value obtained after applying Box-Cox transformation.

[0053] Step 3, establish a prediction model of dynamic pressure zone stable distance, use artificial neural network model (ANN), which includes input layer, hidden layer and output layer. As shown in Figure 3 , ξ k represents input unit, C j represents hidden unit, O i represents output unit. Where w jk and W ijFor connection weights, w = {W, w} represents all connection weights. Output unit indices i = 1, 2, ..., I; hidden unit indices j = 1, 2, ..., J; input unit indices k = 1, 2, ..., K. Additionally, u represents different input patterns; P represents the number of input patterns, u = 1, 2, ..., P; g1 and g2 correspond to the activation functions of the appropriately selected hidden and output layers, respectively. Given an input pattern u, the input to hidden unit j is:

[0054]

[0055] The output is:

[0056]

[0057] in, w is the output value of the j-th hidden unit for the u-th input pattern. jk The connection weights are from the input layer to hidden layer unit j. Given the value of the k-th input feature under the u-th input pattern, the input to output unit i is:

[0058]

[0059] in, W represents the input value of the i-th hidden unit for the u-th input pattern. ij The connection weights from hidden layer unit j to output layer unit i are the final output results. for:

[0060]

[0061] In the data processing, the forward propagation of the signal and the backward propagation of the error are iterative. During this process, the current weight w is modified to obtain a new weight Δw. The error is continuously reduced through iteration to meet the set learning objective. The flowchart of the BP neural network learning process is shown below. Figure 4 As shown.

[0062] Building upon this, Particle Swarm Optimization (PSO) and Genetic Optimization (GO) algorithms are introduced to optimize the model. The PSO algorithm dynamically adjusts the model based on the optimal solutions of individual particles and the global optimum, thereby improving candidate solutions. In the PSO algorithm, the solution to the problem is represented by particles. The set of n particles in D-dimensional space is X. In each iteration, the update formula is:

[0063]

[0064] Where z is the inertia weight; 1≤d≤D; 1≤h≤n; t is the number of iterations; v hd It is the particle's velocity, xhd is the position of the particle; c1 and c2 are learning factors, c1 = c2 = 2; r1 and r2 are random numbers uniformly distributed in [0, 1].

[0065] As shown in Figure 5 , the establishment process of the PSO-ANN model is as follows:

[0066] Step 301, determine the neural network structure;

[0067] Step 302, initialize the weights and thresholds of the ANN;

[0068] Step 303, establish the relationship between PSO and weights and thresholds, and initialize PSO;

[0069] Step 304, calculate the fitness value of the particle, and update the position and speed of each particle;

[0070] Step 305, calculate the error of PSO-ANN, if the error meets the requirements, proceed to the next step; if it does not meet the requirements, repeat step 304;

[0071] Step 306, output the optimization result, train the ANN model, and obtain the PSO-ANN structure model.

[0072] The GO algorithm has multiple search trajectories, which is easy to parallelize, further improving the efficiency of the algorithm. In the iteration process, adaptive crossover probability is used. In the calculation process, the crossover probability will automatically adjust with the value of the fitness function. The calculation formula is as follows:

[0073]

[0074] Where Y s is the adaptive crossover probability; f max is the value of the fitness function of the individual with the largest fitness in the population; f avg is the average value of the fitness function of each generation in the population; f is the value of the fitness function of the two individuals crossed, one of which is larger; Q1, Q2 are constants in the interval from 0 to 1.

[0075] The adaptive mutation probability Y m is calculated as follows:

[0076]

[0077] As shown in Figure 6 , the establishment process of the GO-ANN structure is as follows:

[0078] Step 307, establish the ANN structure;

[0079] Step 308, train the data using ANN to generate the initial population;

[0080] Step 309, initial encoding and preprocessing of GO algorithm;

[0081] Step 310, fitness calculation using different initial values;

[0082] Step 311, genetic operation processing;

[0083] Step 312, obtain optimal weight and threshold, if not optimal value, repeat steps 309 and 310;

[0084] Step 313, assign values to neural network, find global optimal solution, and establish GO-ANN structure model.

[0085] Step 4, evaluate the two optimization models, use correlation coefficient (R 2 ), mean absolute error (MAE), mean absolute percentage error (MAPE) and root mean square error (RMSE) to evaluate the reliability and accuracy of the model, and select the optimal prediction model.

[0086] Correlation coefficient (R 2 ) represents the proportion of variance explained by the model to the total variance, with a value range of 0 to 1, and R 2 closer to 1, the better the model. The calculation formula is as follows:

[0087]

[0088] Where n is the number of samples; y i is the true value; is the predicted value, is the average value of the true value.

[0089] Mean absolute error (MAE) is the average of the square of the prediction error, and the smaller the MAE, the better the model. The calculation formula is as follows:

[0090]

[0091] Mean absolute percentage error (MAPE) is the absolute error between the predicted value and the actual value relative to the actual value, and the smaller the MAPE, the better the model. The calculation formula is as follows:

[0092]

[0093] Root mean square error (RMSE) is the square root of the mean square error, and the smaller the RMSE, the better. The calculation formula is as follows:

[0094]

[0095] Step 5, before mining, the relevant parameters of the target mine are added to the test set of the prediction model as samples to obtain the prediction stable distance L of the dynamic pressure zone, and the target mine combines the stable distance L of the dynamic pressure zone output by the prediction model to design the temporary support.

[0096] As shown in Figure 8 The other side of the single prop 8 is provided with a gangue blocking support 13. The upper side of the single prop 8 is provided with a π-shaped beam 9 and a unit support 10. When the predicted dynamic pressure zone is stable within 200 m of the lagging working face, that is, 0≤L≤200 m, the single prop 8 and the π-shaped beam 9 are used for temporary support together;

[0097] As shown in Figure 9 When the predicted dynamic pressure zone is stable beyond 200 m of the lagging working face, that is, L>200 m, the unit support 10 is used for temporary support. The upper side of the unit support 10 is provided with a constant resistance anchor cable 12.

[0098] The constant resistance anchor cable 12 of the dynamic pressure zone roadway roof is reinforced, and the single prop 8 or the unit support 10 is selected for temporary support according to the stable distance of the dynamic pressure zone, and a U-shaped steel is arranged on the side of the dynamic pressure zone roadway goaf 7 for gangue blocking support 13, which can effectively prevent the gangue in the goaf 7 from rushing into the roadway and form a stable rubble roadway side structure with the gangue in the goaf 7.

[0099] Step 6, the optical fiber intelligent monitoring system is used to monitor and analyze the change of the roof separation amount, the convergence of the roof and floor, the stress of the constant resistance anchor cable 12, and the stress of the support equipment in the dynamic pressure zone in real time, and the temporary support design of the dynamic pressure zone roadway is adjusted in time according to the analysis results. If the monitoring data of the dynamic pressure zone continuously exceeds the predicted value, the support strength is improved to ensure safety; if the monitoring data of the dynamic pressure zone continuously falls below the predicted value, the support strength is gradually reduced to save resources.

[0100] The optical fiber intelligent monitoring system monitors the roof separation amount, and when it is monitored that the roof separation amount continuously increases, the number of anchor cables and grouting reinforcement can be increased at the position with large roof separation amount;

[0101] The optical fiber intelligent monitoring system monitors the convergence of the roof and floor, and when it is monitored that the convergence of the roof and floor continuously increases, the number of anchor cables and the single prop 8 can be increased at the position with large convergence of the roof and floor;

[0102] The optical fiber intelligent monitoring system monitors the stress of the constant resistance anchor cable 12, and when it is monitored that the stress of the constant resistance anchor cable 12 exceeds the preset stress threshold of the constant resistance anchor cable 12, the spacing between the constant resistance anchor cables 12 can be reduced, and the anchor cables can be supplemented;

[0103] The optical fiber intelligent monitoring system monitors the transverse pressure of the gravel bank, and when the transverse pressure of the gravel bank exceeds the preset transverse pressure threshold of the gravel bank, the U-shaped steel row distance can be reduced;

[0104] The optical fiber intelligent monitoring system monitors the stress of the single prop 8, and when the stress of the single prop 8 exceeds the preset stress threshold, the row distance of the single prop 8 can be reduced;

[0105] The optical fiber intelligent monitoring system monitors the stress of the unit support 10, and when the stress of the unit support 10 exceeds the preset stress threshold, the row distance of the unit support 10 can be reduced.

[0106] In addition, when the single prop 8 is used for temporary support, the roadway rock pressure appears violently, the roadway surrounding rock deforms seriously, and the constant resistance anchor cable 12 fails, the unit support 10 is used to replace the single prop 8 for temporary support. When the single support and the unit support 10 are switched, the single support is replaced along the strike of the roadway, and the length of each section is less than 10m, the single prop 8 far away from the working face is removed first, so as to avoid the secondary disturbance of the surrounding rock, and the anchor cable is temporarily added to assist the roof stability during the replacement process.

[0107] Therefore, the coal pillar free self-forming roadway dynamic pressure area control method based on intelligent prediction and dynamic regulation can provide a scientific basis for support design and equipment selection of the dynamic pressure area, improve the monitoring efficiency while ensuring the safety of workers, ensure the accuracy of monitoring data, and realize effective control of the dynamic pressure area roadway.

[0108] Finally, it should be noted that: the above examples are only used to illustrate the technical solutions of the present application and not to limit them, although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can still be modified or replaced by equivalents, and these modifications or equivalent replacements cannot make the modified technical solutions deviate from the spirit and scope of the technical solutions of the present application.

Claims

1. A method for controlling and managing a dynamic pressure zone without a coal pillar and self-forming a roadway based on intelligent prediction and dynamic regulation, characterized in that, The method comprises the following steps: Step 1, collecting mine exploitation data, establishing a machine learning database, specifically: Through mine investigation and literature summary, relevant data of the mine are collected, including mining depth, coal seam thickness, coal seam inclination, cutting height, cutting angle, roof rock hardness, roadway height, roadway width, working face inclination length, working face strike length, dynamic pressure stable distance; Step 2, using the three-dimensional modeling software FLAC3D to carry out orthogonal simulation experiment on the data characteristics in the database of step 1, obtaining parameters related to the dynamic pressure stable distance, and using Box-Cox transformation technology to standardize the screened feature engineering database, obtaining the processed data set; the Box-Cox transformation formula is: ； wherein is a continuous variable; is a transformation parameter, is determined by maximum likelihood estimation; Step 3, using the data in the data set in step 2, establishing a prediction model of the dynamic pressure zone stable distance, the prediction model is an artificial neural network model, which continuously reduces the error through iteration to meet the set learning target; and the prediction model is optimized by using particle swarm optimization algorithm and genetic optimization algorithm, obtaining the to-be-evaluated model; Step 4, screening the to-be-evaluated model in step 3, the indexes used for screening include: correlation coefficient, mean absolute error, mean absolute percentage error and root mean square error, and the final model is obtained after screening; Step 5, before mining, the relevant parameters of the target mine are added to the test set of the final model in step 4 as samples, and the dynamic pressure zone range prediction result is obtained; according to the prediction result, the dynamic pressure zone supporting equipment selection, support design and temporary support design of the dynamic pressure zone roadway are carried out, specifically: If the dynamic pressure zone stable distance L≤200m, single prop and π type beam temporary support is adopted; if L>200m, unit support temporary support is adopted; Step 6, in the mining process, the stability of the dynamic pressure zone roadway is monitored in real time by using the optical fiber intelligent monitoring device to monitor the stress and displacement of the dynamic pressure zone roadway, and the temporary support design of the dynamic pressure zone is adjusted in time according to the monitoring result, specifically: When the roof separation amount continues to increase, the number of anchor cables and grouting reinforcement are increased at the position with large roof separation amount; when the top and bottom plate moving amount continues to increase, the number of anchor cables and the single prop are increased at the position with large top and bottom plate moving amount; When the stress of the constant resistance anchor cable exceeds the limit, the row spacing is reduced and the anchor cable is supplemented; when the transverse pressure of the broken rock side exceeds the limit, the row spacing of the U-shaped steel is reduced; When the stress of the single prop exceeds the preset stress threshold, the row spacing of the single prop is reduced; when the stress of the unit support exceeds the preset stress threshold, the row spacing of the unit support is reduced; When the single prop is used for temporary support, the roadway rock pressure appears violently, the roadway surrounding rock deforms seriously, and the constant resistance anchor cable fails when the single prop is used for temporary support, the unit support is used to replace the single prop, the single support is replaced along the strike of the roadway in sections, each section is less than 10m, the single prop far away from the working face is removed first to avoid the secondary disturbance of the surrounding rock, and anchor cables are temporarily added to assist the roof stability during the replacement process.

2. A coal pillar-free self-lane dynamic pressure zone management device based on intelligent prediction and dynamic regulation, applied to the coal pillar-free self-lane dynamic pressure zone management method based on intelligent prediction and dynamic regulation in claim 1, characterized in that, It comprises: A data collection module is used to collect mine exploitation data and establish a machine learning database; The data feature processing module is connected with the data collection module and is configured to use three-dimensional modeling software FLAC3D to perform orthogonal simulation experiments on data features in the machine learning database, to obtain parameters related to the dynamic pressure stability distance, and to use Box-Cox transformation technology to perform standardization processing on the screened feature engineering database, to obtain a processed data set. The model training module is connected with the data feature processing module and is configured to use data in the data set to establish a prediction model of the dynamic pressure zone stability distance, the prediction model being an artificial neural network model, to iteratively reduce errors to meet a set learning goal, and to use a particle swarm optimization algorithm and a genetic optimization algorithm to perform optimization processing on the prediction model, to obtain an evaluation model. The model screening module is connected with the model training module and is configured to screen the evaluation model, to obtain a final model. The result prediction module is connected with the model screening module and is configured to add relevant parameters of a target mine as samples into a test set of the final model before mining, to obtain a dynamic pressure zone range prediction result, to perform dynamic pressure zone support equipment selection, support design, and temporary support design for a dynamic pressure zone roadway according to the prediction result. The real-time monitoring module is connected with the result prediction module and is configured to use a fiber-optic intelligent monitoring device to monitor stress and displacement of a dynamic pressure zone roadway in real time during mining, and to timely adjust a dynamic pressure zone temporary support design according to a monitoring result.

3. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by a processor to implement steps of the method in claim 1.

Images