Multi-physics field cooperative regulation and control method in coal mine rock burst prevention and control

Abstract

The invention provides a multi-physics field cooperative regulation and control method in coal mine rock burst prevention and control, and relates to the technical field of coal mine rock burst prevention and control, and the method comprises the steps: synchronously collecting stress field, fracture field and vibration wave field data in real time through a distributed optical fiber sensor network and a micro-seismic monitoring system, and constructing a dynamically updated three-dimensional geomechanical model; based on an adaptive Kalman filtering algorithm, noise reduction multi-source data are fused, a dynamic stress concentration factor and an energy accumulation critical index are extracted, and a rock burst risk probability model is established; a multi-objective optimization algorithm is adopted to generate graded regulation and control instructions of grouting reinforcement, mining speed adjustment and pressure relief drilling, and the graded regulation and control instructions are executed in real time; and iteratively optimizing the model weight by using a transfer learning algorithm in combination with a historical case library to form a closed-loop feedback adaptive prevention and control system. According to the method, the analysis precision of rock mass fracture evolution under complex geological conditions is improved through multi-physics field collaborative perception and dynamic modeling, and risk quantification and real-time regulation and control are realized through multi-algorithm coupling analysis.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of coal mine rock burst prevention and control, in particular to a multi-physical field collaborative regulation method in coal mine rock burst prevention and control. BACKGROUND

[0002] The current coal mine rock burst prevention and control field is facing a series of technical challenges, which seriously restricts the prevention and control effect and safety. First, the existing monitoring means mostly relies on single physical field data acquisition, such as only using stress sensors or microseismic monitoring to independently obtain information, which leads to a lack of comprehensive perception of the coupling of stress field, fracture field and vibration wave field under complex geological conditions, and it is difficult to construct a high-precision dynamic geomechanical model. Secondly, the traditional data processing method often uses fixed algorithm or single noise reduction technology, and the fusion ability of multi-source heterogeneous data is insufficient, and there are significant errors in feature parameter extraction, such as the quantization index of stress concentration degree and energy accumulation trend relying on empirical formula or simplified model, which cannot accurately reflect the nonlinear fracture characteristics of rock mass.

[0003] In addition, the existing regulation strategy is generated based on a static risk assessment model, and the generation cycle of the regulation instruction is long, and it relies on manual experience to adjust the parameters, which cannot respond to the dynamic changes of the physical field in the mining process in real time, leading to insufficient timeliness and accuracy of key measures such as grouting reinforcement and pressure relief drilling. At the same time, the optimization mechanism of the risk model is missing, and most models use fixed weights or static training of historical data, which cannot adapt to the differences of different mine geological conditions and mining technology, and the prediction ability of the model decreases significantly after long-term use.

[0004] More importantly, the existing technology does not fully solve the multi-parameter coupling analysis problem, such as the correlation between stress concentration factor and energy accumulation index is not effectively quantified, leading to the risk level division relying on a single threshold or subjective judgment, and the false alarm rate and the miss rate are high. In addition, the regulation effect feedback mechanism is not perfect, most systems lack closed-loop optimization capability, and cannot dynamically modify model parameters according to actual regulation results, making it difficult to form a continuously improved prevention and control system. These problems together make it difficult for the real-time, accuracy and adaptability of coal mine rock burst prevention and control to meet the needs of complex mining environment. SUMMARY

[0005] In order to solve the technical problems of single monitoring means leading to insufficient perception of multi-physical field coupling, poor adaptability of data fusion algorithm causing significant quantization error of feature parameters, static risk assessment model lacking timeliness relying on manual experience, risk prediction model lacking dynamic optimization mechanism leading to decreased adaptability, lack of multi-parameter coupling correlation causing inaccurate risk level division, and lack of regulation feedback loop making it difficult to realize continuous iteration optimization of the model, the present application provides a multi-physical field collaborative regulation method in coal mine rock burst prevention and control.

[0006] The technical scheme provided by the present application is as follows:

[0007] The multi-physical field synergistic regulation method in coal mine rock burst prevention and control provided by the present application comprises:

[0008] S1, multi-physical field dynamic monitoring: through a distributed optical fiber sensor network, real-time acquisition of stress field, fracture field and seismic wave field data in a coal mining area, combined with a microseismic monitoring system to obtain the spatio-temporal distribution characteristics of rock mass fracture events, a dynamically updated three-dimensional geomechanical model is constructed;

[0009] S2, multi-algorithm coupling analysis: based on the monitoring data of S1, an adaptive Kalman filter algorithm is used to fuse and denoise the multi-physical field data, extract key feature parameters, including dynamic stress concentration factor DSCF and energy accumulation critical index EACI, and establish an rock burst risk probability model;

[0010] S3, dynamic regulation strategy generation: according to the output result of the risk probability model of S2, a multi-objective optimization algorithm is used to generate hierarchical regulation instructions, including grouting reinforcement parameters, mining speed adjustment coefficient and pressure relief borehole layout scheme, and the regulation instructions are executed in real time through an Internet of Things terminal;

[0011] S4, feedback closed-loop optimization: based on the similarity matching of the physical field data after regulation and the historical case library, the risk probability model weight in S2 is iteratively optimized using a transfer learning algorithm to form a dynamically adaptive prevention and control decision system.

[0012] The technical scheme provided by the present application has at least the following beneficial effects:

[0013] (1) In the present application, through the joint deployment of a distributed optical fiber sensor network and a microseismic monitoring system, real-time synchronous monitoring of stress field, fracture field and seismic wave field is realized, combined with a dynamically updated three-dimensional geomechanical model construction technology, breaking through the limitations of single physical field perception, significantly improving the analysis capability of multi-field coupling under complex geological conditions, solving the problem of incomplete capture of rock mass fracture evolution law by traditional monitoring methods and insufficient model accuracy, and providing a high reliability data basis for rock burst risk analysis.

[0014] (2) In the present application, based on the coupling application of adaptive Kalman filter algorithm and multi-objective optimization algorithm, fusion and denoising of multi-source heterogeneous data and accurate extraction of feature parameters are realized, combined with the quantitative model of dynamic stress concentration factor and energy accumulation critical index, an rock burst risk probability model is established, through real-time generation of hierarchical regulation instructions and automatic execution, solving the problem of static model dependence on artificial experience and regulation lag, significantly improving the timeliness and accuracy of key measures such as grouting reinforcement and mining speed adjustment.

[0015] (3) In the present application, a feedback closed-loop optimization mechanism is constructed by using a transfer learning algorithm and a domain-adaptive network, the regulated physical field data are dynamically matched with the historical case library, the weight iterative optimization and parameter adaptive calibration of the risk model are realized, the adaptability bottleneck caused by the solidified weight of the traditional model is broken through, the problems of missing of multi-parameter coupling correlation and inaccurate risk grade division are solved, a continuously evolving prevention and control decision system is formed, and the stability of the prevention and control effect under different geological conditions and mining processes is comprehensively improved. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0017] Figure 1 The flowchart of the multi-physical field synergistic regulation method in the coal mine rock burst prevention provided by the embodiment of the present application is shown in the figure.

[0018] Figure 2 The flowchart of establishing a rock burst risk probability model in the multi-physical field synergistic regulation method in the coal mine rock burst prevention provided by the embodiment of the present application is shown in the figure.

[0019] Figure 3 The flowchart of determining the mining speed adjustment coefficient in the multi-physical field synergistic regulation method in the coal mine rock burst prevention provided by the embodiment of the present application is shown in the figure.

[0020] Figure 4 The flowchart of feedback closed-loop optimization in the multi-physical field synergistic regulation method in the coal mine rock burst prevention provided by the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0021] The technical solutions in the present application will be described below with reference to the drawings.

[0022] In the embodiments of the present application, the words such as "example", "for example" are used to represent an example, illustration or description. Any embodiment or design scheme described as "example" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present application, the meaning expressed by "and / or" can be both, or can be one of the two.

[0023] In the embodiments of the present application, "image" and "picture" can be used interchangeably, and it should be pointed out that their meanings are consistent when the distinction is not emphasized.

[0024] In the embodiments of the present application, sometimes the subscript such as W1 can be mistakenly used as a non-subscript such as W1, and their meanings are consistent when the distinction is not emphasized.

[0025] To make the technical problems, technical solutions and advantages to be solved by the present application clearer, the following will be described in detail with reference to the drawings and specific embodiments.

[0026] Reference is made to the accompanying drawings and specific embodiments of the present application. Figure 1 Fig. 1 shows a flowchart of a method for multi-physical field collaborative regulation in coal mine rock burst prevention and control according to an embodiment of the present application.

[0027] The present application provides a method for multi-physical field collaborative regulation in coal mine rock burst prevention and control, and the processing flow can include the following steps:

[0028] S1, multi-physical field dynamic monitoring: real-time acquisition of stress field, fracture field and seismic wave field data of the coal mining area by a distributed optical fiber sensor network, combination with a microseismic monitoring system to obtain the spatio-temporal distribution characteristics of rock mass fracture events, and construction of a dynamically updated three-dimensional geomechanical model.

[0029] Specifically, the stress field, fracture field and seismic wave field data of the coal mining area are collected in real time by a distributed optical fiber sensor network (Distributed Optical Fiber Sensor Network) at a sampling frequency of 10 times per second, wherein the optical fiber sensors are arranged in a grid along the roof, floor and two sides of the roadway, and the grid spacing is 5 meters. The microseismic monitoring system (Microseismic Monitoring System) synchronously records the spatio-temporal distribution characteristics of rock mass fracture events, including source location, energy release and fracture propagation direction, with a data sampling accuracy of millimeter level. Combined with the above data, a dynamically updated three-dimensional geomechanical model is constructed using finite element analysis software, with a model update frequency of once per minute, and real-time rendering is realized through a parallel computing cluster.

[0030] In one possible implementation, S1 further includes:

[0031] S101, obtaining the initial ground stress distribution of the mining area by a static ground stress measuring instrument.

[0032] Specifically, a static stress measurement device is used to drill test the mining area, the drilling depth is 50 meters, the interval is 20 meters, the initial stress distribution is obtained by hydraulic fracturing method, and the measurement data includes horizontal principal stress, vertical stress and shear stress component.

[0033] S102, using the reverse time migration algorithm to dynamically disturb the vibration wave field data, and reconstructing the time and space evolution law of the disturbed stress field.

[0034] Specifically, the reverse time migration algorithm is used to process the vibration wave field data, specifically including inputting the original waveform data into the GPU accelerated computing node, and reconstructing the time and space evolution law of the disturbed stress field by time reversal, and the inversion resolution reaches 0.1 meters.

[0035] S103, combining the acoustic emission monitoring data to jointly calibrate the inversion result of S102, and correcting the dynamic stress boundary condition of the three-dimensional geomechanical model.

[0036] Specifically, combined with the fracture signal collected by the acoustic emission monitoring system, the inversion result is jointly calibrated, the dynamic stress boundary condition of the three-dimensional geomechanical model is corrected, and the calibration error is controlled within ±2%.

[0037] S2, multi-algorithm coupling analysis: based on the monitoring data of S1, using the adaptive Kalman filter algorithm to fuse and denoise the multi-physical field data, extracting key feature parameters including dynamic stress concentration factor DSCF and energy accumulation critical index EACI, and establishing an rock burst risk probability model.

[0038] Specifically, first, the adaptive Kalman filter algorithm is used to fuse and denoise the multi-physical field data collected by S1, the filter window length is 30 seconds, and the noise covariance matrix is dynamically adjusted according to the real-time data. After denoising, the key feature parameters are extracted, including dynamic stress concentration factor DSCF (Dynamic Stress Concentration Factor) and energy accumulation critical index EACI (Energy Accumulation Critical Index).

[0039] As shown in Figure 2 , in one possible implementation, the calculation of the dynamic stress concentration factor DSCF is realized by the following steps:

[0040] S201、Based on the nonlinear constitutive equation of rock mass, the influence of the principal stress difference on local stress concentration is corrected.

[0041] Specifically, based on the nonlinear constitutive equation of rock mass, the nonlinear relationship between the principal stress difference σ1-σ3 and the local stress concentration is fitted through triaxial test data, and the test loading rate is 0.1 MPa / s.

[0042] S202、The dynamic stress concentration factor is calculated by the following formula:

[0043]

[0044] Wherein, σ1 is the maximum principal stress, σ3 is the minimum principal stress, E r is the dynamic elastic modulus of rock mass, ε d is the dynamic strain rate, τ max is the maximum shear stress, R c is the uniaxial compressive strength of rock mass.

[0045] Specifically, σ1 and σ3 are obtained in real time by a geostress measuring instrument, E r is calculated by the dynamic stress-strain curve of triaxial cyclic loading and unloading test, ε d is measured by a strain gauge with a sampling frequency of 1000 Hz, τ max is determined by the Mohr circle analysis method, R c is obtained by uniaxial compression test using a rock mechanics testing machine.

[0046] In one possible implementation, the calculation method of the dynamic elastic modulus E r in S202 is as follows:

[0047] S2021、The dynamic stress-strain curve of rock mass is obtained by triaxial cyclic loading and unloading test (loading rate 0.5 MPa / s, cycle number 50 times), and the test environment humidity is controlled below 60%.

[0048] S2022、The nonlinear variation characteristics of the dynamic elastic modulus are fitted by the piecewise linear interpolation method (Piecewise Linear Interpolation), and specifically the curve is divided into 10 linear segments, and the elastic modulus of each segment is calculated by the slope and then the weighted average value is taken.

[0049] In one possible implementation, the calculation of the energy accumulation critical index EACI is completed by the following steps:

[0050] S203、The random forest regression algorithm (Random Forest Regression Algorithm) is adopted, and the microseismic event energy release rate, crack propagation rate and mining disturbance frequency are input as characteristic parameters.

[0051] Specifically, the microseismic event energy release rate (calculated by the first derivative of the energy-time curve), the fracture propagation rate (inverted by the acoustic emission signal frequency change), and the mining disturbance frequency (based on the mining machinery operation log statistics).

[0052] S204, initialize the algorithm parameters of the random forest regression based on the rock mass failure process analysis model (Realistic Failure Process Analysis Model).

[0053] Specifically, the tree depth is set to 15 layers, and the minimum leaf node sample number is 5.

[0054] S205, calculate the energy accumulation critical index by the following formula:

[0055]

[0056] Wherein, Q e (t) is the time energy density, V r is the rock mass unit volume, p e (t) is the energy release rate gradient, K IC is the rock mass fracture toughness, a c is the critical crack length, and a is the geological structure correction coefficient.

[0057] Specifically, Q e (t) is calculated by the ratio of microseismic energy data and rock mass volume, V r is 1 cubic meter of rock mass unit, p e (t) is fitted by the second derivative of energy release amount to time, K IC is measured by three-point bending test, a c is determined by fracture propagation simulation.

[0058] In one possible implementation, the determination method of the geological structure correction coefficient a in S205 is:

[0059] S2051, extract the fault density (length of fault per square meter), rock layer dip angle (compass measurement), and joint development degree (fracture number statistics) parameters according to the geological exploration data;

[0060] S2052, quantize the geological structure complexity by principal component analysis (Principal Component Analysis) to generate a mapping relationship table of the correction coefficient a, and map to the a coefficient range of 0.5-1.5.

[0061] In one possible implementation, the rock burst risk probability model in S2 specifically includes:

[0062] S206, based on the coupling relationship between the dynamic stress concentration factor DSCF and the energy accumulation critical index EACI, the safe zone, the warning zone and the dangerous zone are divided, specifically, the safe zone, the warning zone and the dangerous zone are divided in the two-dimensional phase plane with DSCF = 1.5 and EACI = 120 as the threshold.

[0063] S207, the priority of the control strategy of each partition is optimized by using genetic algorithm (Genetic Algorithm), and a probabilistic risk level is generated, specifically, the population size is 100, the crossover probability is 0.8, the mutation probability is 0.05, and the probabilistic risk level is generated after 200 iterations.

[0064] S3, dynamic control strategy generation: according to the output result of the risk probability model of S2, multi-objective optimization algorithm (Multi-Objective Optimization Algorithm) is used to generate hierarchical control instructions, including grouting reinforcement parameters, mining speed adjustment coefficient and pressure relief borehole layout scheme, and the control instructions are executed in real time through Internet of Things terminal.

[0065] As shown in Figure 3 , the determination method of the mining speed adjustment coefficient is:

[0066] S301, the long short-term memory network (Long Short-Term Memory Network, LSTM) is used to predict the energy accumulation trend in the next 10 hours, the network input is the EACI sequence in the past 24 hours, and the hidden layer node number is 128.

[0067] S302, the LSTM prediction result and the EACI value in S2 are weighted and fused according to the weight coefficient 0.7:0.3, the weight coefficient is dynamically adjusted according to the historical rock burst event database, specifically, the weight coefficient is dynamically adjusted according to the rock burst occurrence frequency of similar working conditions in the historical database.

[0068] S303, the mining speed adjustment coefficient and the corresponding control priority are output, the control priority is specifically divided into emergency (adjustment range ± 30%), high (± 20%) and medium (± 10%) according to the risk level. The grouting reinforcement parameters are calculated through the porosity-pressure relationship model, and the pressure relief borehole layout scheme adopts honeycomb arrangement, the hole spacing is 3 meters, and the hole diameter is 50 millimeters.

[0069] S4, feedback closed loop optimization: based on the similarity matching of the physical field data after control and the historical case library, the risk probability model weight in S2 is iteratively optimized by using transfer learning algorithm, and a dynamic self-adaptive prevention and control decision system is formed.

[0070] As shown in Figure 4 , in one possible implementation, S4 specifically includes:

[0071] S401, the physical field characteristics of the current mining area are mapped to the high-dimensional space of the historical case library by using a domain-adversarial neural network (DANN), and the feature dimension is 256.

[0072] S402, the model generalization ability is optimized by maximizing the similarity measurement value (Cosine Similarity) between cases, and the similarity threshold is set to 0.85.

[0073] S403, when the DSCF decline rate is less than 5% or the EACI rise rate exceeds 8% after continuous regulation for three times, a multi-source information fusion mechanism is triggered to recalibrate the model parameters.

[0074] Specifically, the multi-source information fusion mechanism comprises:

[0075] S4031, the microseismic event energy sequence (time domain integral), stress field mutation characteristics (gradient exceeding 10MPa / m) and historical regulation effect data (success / failure label) are fused;

[0076] S4032, the posterior probability distribution of the model parameters is updated based on Markov Chain Monte Carlo Sampling (Markov Chain Monte Carlo Sampling), the sampling number is 10000 times, and the convergence tolerance is 0.001.

[0077] The technical scheme provided by the embodiment of the application has at least the following beneficial effects:

[0078] (1) In the present application, by joint deployment of a distributed optical fiber sensor network and a microseismic monitoring system, real-time synchronous monitoring of stress field, fracture field and seismic wave field is realized, combined with dynamic updating of three-dimensional geomechanical model construction technology, the limitations of single physical field perception are broken through, the analytical ability of multi-field coupling under complex geological conditions is significantly improved, and the problems of incomplete capture of rock mass rupture evolution law by traditional monitoring methods and insufficient model precision are solved, providing a high credibility data basis for rock burst risk analysis.

[0079] (2) In the present application, based on the coupling application of adaptive Kalman filtering algorithm and multi-objective optimization algorithm, the fusion and noise reduction of multi-source heterogeneous data and the accurate extraction of feature parameters are realized, combined with the quantitative model of dynamic stress concentration factor and energy accumulation critical index, the rock burst risk probability model is established, the hierarchical regulation instruction is generated in real time and automatically executed, the problems of static model dependence on artificial experience and regulation lag are solved, and the timeliness and accuracy of grouting reinforcement, mining speed adjustment and other key measures are significantly improved.

[0080] (3) In the present application, a feedback closed-loop optimization mechanism is constructed by using a transfer learning algorithm and a domain-adversarial adaptive network, the regulated physical field data are dynamically matched with a historical case library, the weight iterative optimization and parameter adaptive calibration of the risk model are realized, the adaptability bottleneck caused by the solidified weight of the traditional model is broken through, the problems of missing of multi-parameter coupling correlation and inaccuracy of risk grade division are solved, a continuously evolving prevention and control decision system is formed, and the stability of the prevention and control effect under different geological conditions and mining processes is comprehensively improved.

[0081] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0082] The following points need to be explained:

[0083] (1) The attached drawings of the embodiments of the present application only involve the structures involved in the embodiments of the present application, and other structures can be referred to the usual design.

[0084] (2) In order to be clear, the thickness of the layer or area is enlarged or reduced in the drawings used to describe the embodiments of the present application, that is, the drawings are not drawn according to the actual proportion. It can be understood that when an element such as a layer, a film, an area or a substrate is referred to as being located on or under another element, the element can be "directly" located on or under another element or there can be an intermediate element.

[0085] (3) In the case of no conflict, the embodiments of the present application and the features in the embodiments can be combined with each other to obtain new embodiments.

[0086] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A multi-physical field coordinated control method for preventing and controlling rock burst in coal mines, characterized in that: include: S1. Multi-physics field dynamic monitoring: A distributed fiber optic sensor network collects real-time stress field, fracture field, and vibration wave field data in the coal mining area. Combined with a microseismic monitoring system, the temporal and spatial distribution characteristics of rock mass failure events are obtained to construct a dynamically updated three-dimensional geomechanical model. S2. Multi-algorithm coupling analysis: Based on the monitoring data in S1, an adaptive Kalman filter algorithm is used to fuse and reduce noise on the multi-physics field data, extract key characteristic parameters, including the dynamic stress concentration factor DSCF and the energy accumulation critical index EACI, and establish a rock burst risk probability model; S3. Dynamic Control Strategy Generation: Based on the output of the risk probability model in S2, a multi-objective optimization algorithm is used to generate hierarchical control instructions, including grouting reinforcement parameters, mining speed adjustment coefficients, and pressure relief drilling layout plans. The control instructions are then executed in real time through the IoT terminal; S4, feedback closed-loop optimization: Based on the similarity matching between the physical field data after regulation and the historical case library, the risk probability model weights in S2 are iteratively optimized using the transfer learning algorithm to form a dynamic and adaptive prevention and control decision-making system.

2. The multi-physical field coordinated control method for coal mine rock burst prevention and control according to claim 1 is characterized in that: Said S1 specifically includes: S101, obtaining the initial ground stress distribution in the mining area using a static ground stress measuring instrument; S102, using a reverse time migration algorithm to perform dynamic disturbance stress inversion on the vibration wave field data, and reconstructing the spatiotemporal evolution law of the disturbance stress field; S103. Combined calibration of the inversion result of S102 is performed with the acoustic emission monitoring data to correct the dynamic stress boundary conditions of the three-dimensional geomechanical model.

3. The multi-physical field coordinated control method for coal mine rock burst prevention and control according to claim 1 is characterized in that: The S2 specifically includes: S201. Based on the nonlinear constitutive equation of rock mass, the influence of principal stress difference on local stress concentration is corrected; S202. Calculate the dynamic stress concentration factor using the following formula: Among them, σ1 is the maximum principal stress, σ3 is the minimum principal stress, E r is the dynamic elastic modulus of rock mass, ε d is the dynamic strain rate, τ max is the maximum shear stress, R c is the uniaxial compressive strength of rock mass.

4. The multi-physical field coordinated control method for coal mine rock burst prevention and control according to claim 3 is characterized in that: Said S2 further comprises: S203, using a random forest regression algorithm, inputting the microseismic event energy release rate, crack expansion rate, and mining disturbance frequency as characteristic parameters; S204. Initializing algorithm parameters of random forest regression based on the rock mass fracture process analysis model; S205. Calculate the critical energy accumulation index using the following formula: Among them, Q e (t) is the time domain energy density, V r is the rock mass unit volume, ρ e (t) is the energy release rate gradient, K IC is the fracture toughness of rock mass, a c is the critical crack length, and α is the geological structure correction coefficient.

5. The multi-physical field coordinated control method for coal mine rock burst prevention and control according to claim 1, characterized in that: Said S3 further comprises: The method for determining the mining speed adjustment coefficient includes: S301. Use long short-term memory network to predict the energy accumulation trend in the next 10 hours; S302, performing weighted fusion of the prediction result and the EACI value in S2, with the weight coefficient dynamically adjusted according to the historical rock burst event database; S303: Output the mining speed adjustment coefficient and the corresponding control priority.

6. The multi-physical field coordinated control method for coal mine rock burst prevention and control according to claim 1, characterized in that: The S4 specifically includes: S401, using a domain adversarial adaptive network to map the physical field characteristics of the current mining area to the high-dimensional space of the historical case library; S402, optimizing the model generalization ability by maximizing the similarity measure between cases; S403: When the DSCF decrease rate is less than 5% or the EACI increase rate exceeds 8% after three consecutive adjustments, the multi-source information fusion mechanism is triggered to recalibrate the model parameters.

7. The multi-physical field coordinated control method for coal mine rock burst prevention and control according to claim 4, characterized in that: The rock burst risk probability model in S2 specifically includes: S206. Based on the coupling relationship between the dynamic stress concentration factor DSCF and the energy accumulation critical index EACI, the safety zone, warning zone and danger zone are divided; S207. Use a genetic algorithm to optimize the control strategy priority of each partition and generate a probabilistic risk level.

8. The multi-physical field coordinated control method for coal mine rock burst prevention and control according to claim 3 is characterized in that: The dynamic elastic modulus E in S202 r The calculation methods include: S2021. Obtain dynamic stress-strain curve of rock mass through triaxial cyclic loading and unloading test; S2022. Use piecewise linear interpolation method to fit the nonlinear variation characteristics of dynamic elastic modulus.

9. The multi-physical field coordinated control method for coal mine rock burst prevention and control according to claim 4, characterized in that: The method for determining the geological structure correction coefficient α in S205 includes: S2051. Extracting fault density, rock layer dip and joint development degree parameters based on geological exploration data; S2052. Quantify the geological structure complexity through principal component analysis and generate a mapping relationship table of the correction coefficient α.

10. The multi-physical field coordinated control method for coal mine rock burst prevention and control according to claim 6, characterized in that: The multi-source information fusion mechanism in S403 includes: S4031. Integrate microseismic event energy sequence, stress field mutation characteristics and historical control effect data; S4032. Update the posterior probability distribution of model parameters based on Markov chain Monte Carlo sampling.

Images