Evaluation method for prevention and control of mine water inrush disasters based on intelligent identification and early warning of fault water

Abstract

The present invention discloses a mine water inrush disaster prevention and control evaluation method based on intelligent identification and early warning of fault water, which relates to the technical field of mine water inrush disaster prevention and control. The method includes: first installing the required system, then analyzing and selecting the main parameters, weighting each parameter by entropy weight method, and calculating the weight of each parameter; importing the weight of each parameter into the PSO-SVM model of the extended Toth model to form an extended Toth flow field model, namely the ETFM model, which is further extended to the coal mine floor water inrush characteristics, and defining the comprehensive water pressure field distribution P t Finally, the corresponding water inrush risk warning value is derived based on game theory. By introducing a multidimensional fusion model, optimization algorithms, real-time data acquisition, and intelligent control, this invention addresses technical issues inherent in traditional methods, such as single monitoring parameters, insufficient intelligence in the warning model, low system integration, and passive prevention and control measures.

Description

Technical Field

[0001] The present invention relates to the technical field of mine water inrush disaster prevention and control, and in particular to a mine water inrush disaster prevention and control evaluation method based on intelligent identification and early warning of fault water. Background Art

[0002] As coal mining continues to increase in depth, floor water inrush is a common occurrence during the mining process. Water inrush caused by the activation of the mining floor and fault structures is the primary form of mine floor water disaster. Fault water inrush not only impacts coal mine safety and production efficiency, but also seriously threatens the lives of workers. Therefore, effective fault water inrush warning is crucial for ensuring safe coal mine production.

[0003] The primary cause of water inrush from coal mine floor faults is coal mine confined water, the most common and direct source of water inrush from coal mine floor faults. While the faults themselves contain no water, the combined effects of mining unloading disturbances and the rise of confined water activate the floor faults, creating fissures. Pressure from the confined water causes the water to rise along these fissures, leading to water inrush from the floor faults. Currently, common methods for controlling and preventing floor water inrush include pumping to reduce pressure, grouting, and retaining coal pillars. None of these methods can eliminate the probability of water inrush from the faults. Therefore, other methods, combined with corresponding early warning technologies, are needed to implement timely protective measures before a water inrush occurs, thereby minimizing losses.

[0004] At present, the existing technologies for the research reports on the early warning of floor fault water inrush are as follows:

[0005] Application No. 202411224450.9 discloses a mine roof and floor water inrush monitoring and prediction system and method. It proposes the basic framework of the mine prediction system, but does not provide a corresponding calculation model for how to combine historical data with the current water level dynamics. Application No. 202410998013.6 discloses a coal seam floor water inrush hazard assessment method based on the fracturing lifting mechanism. It uses the floor rupture pressure and reopening pressure laws to evaluate the hazard of the floor. However, floor water inrush is affected by many factors, and the factors considered are not comprehensive. Application No. 200910119379.7 discloses a mine water inrush disaster monitoring and early warning system and its control method. It connects the data transmission subsystem and the data and alarm release subsystem through a computer network to achieve mine water inrush disaster monitoring and early warning. It mainly focuses on the overall system architecture and data transmission method, and does not involve specific monitoring parameters and early warning models. Application No. 201922260113.6 discloses an underground coal mine water inrush warning and monitoring device, focusing on its structural design and self-cleaning function to ensure the long-term effectiveness of the sensor. Application No. 201810406872.6 discloses an intelligent early warning system and method for coal mine water hazards. This system uses big data analysis and artificial intelligence algorithms to provide intelligent early warnings for coal mine water hazards. This system emphasizes the application of the Internet of Things and big data technologies, focusing on comprehensive data collection and intelligent analysis.

[0006] Although the above-mentioned existing technologies can provide early warning for water inrush from floor faults, they still have the following technical problems:

[0007] The singleness of monitoring parameters, lack of intelligent early warning models, lack of real-time and dynamic features, poor system integration and coordination, and lack of initiative in prevention and control measures.

[0008] This shows that the prior art needs to be further improved. Summary of the Invention

[0009] The purpose of the present invention is to provide a mine water inrush disaster prevention and control evaluation method based on intelligent identification and early warning of fault water. By introducing a multidimensional fusion model, optimization algorithm, real-time data acquisition and intelligent control, it solves the technical problems existing in traditional methods such as single monitoring parameters, insufficient intelligence of early warning models, low system integration, and passive prevention and control measures.

[0010] In order to achieve the above object, the present invention adopts the following technical solutions:

[0011] A mine water inrush disaster prevention and control evaluation method based on intelligent identification and early warning of fault water includes the following steps:

[0012] (1) Install the required system

[0013] The system includes a bottom plate pressure water monitoring system, a multidimensional model early warning system and an intelligent drainage system. The bottom plate pressure water monitoring system includes a bottom plate pressure water monitor, a wired communication module and a downhole server. The bottom plate pressure water monitor includes a housing and a water level sensor, a water pressure sensor and a temperature sensor located inside the housing. The wired communication module is located on the top of the bottom plate pressure water monitor. Holes are drilled into the fault at a certain interval and different levels in the bottom plate, and the bottom plate pressure water monitor is installed in the resulting boreholes. The downhole server is used to collect data obtained by each sensor and send it to a filtering unit through a signal amplifier connected to the downhole server. After passing through the filtering unit, the collected data is sent to the multidimensional model early warning system located on the ground. The multidimensional model early warning system is connected to the intelligent drainage system located on the ground.

[0014] (2) After analysis, the main parameters of geological conditions are selected: height of water-conducting fracture zone, thickness of key layer, thickness of aquifer, thickness of impermeable layer, thickness ratio of brittle-plastic rock, fault strength, fault intersection, density of pinch-out points, water pressure of aquifer, water temperature, geological structure characteristics and rock mechanics. The weight of each parameter is calculated by entropy weight method to obtain the weight of each parameter;

[0015] (3) The weight of each parameter in step (2) is imported into the PSO-SVM model of the extended Toth model to form the extended Toth flow field model, namely the ETFM model, which is further extended to the water inrush characteristics of the coal mine floor and defines the comprehensive water pressure field distribution P t , as shown in formula (1):

[0016]

[0017] In formula (1): ρgH w The product of is the hydrostatic pressure; γT w The product of water temperature and water dynamics is the correction term; G i is the comprehensive factor of geological conditions; β i is the influence weight of each parameter of geological conditions; i, n is the number of main parameters affecting floor water inrush, i = 1, 2, 3, ..., n;

[0018] PSO is used to optimize the penalty parameter C of the SVM and the parameter γ of the radial basis kernel function (RBF). The SVM is trained and the risk level is divided according to the set threshold. C is used to balance the relationship between model complexity and error terms. γ is the parameter of the radial basis kernel function (RBF), which determines the influence range of the support vector in high-dimensional space.

[0019] (4) According to game theory, the corresponding water inrush risk warning value is obtained, as shown in formula (2):

[0020]

[0021] In formula (2): P t * is the optimal value of water inrush risk probability; P t is the probability of water inrush risk; U is the utility function;

[0022] Through formula (2), we can get: when P t <P t * When P t ≥P t * When the risk is high.

[0023] (5) The intelligent drainage system executes different drainage and control instructions according to the instruction level issued by the multi-dimensional model early warning system, and then determines the drainage volume.

[0024] The above-mentioned mine water inrush disaster prevention and control evaluation method based on intelligent identification and early warning of fault water, the shell is made of waterproof material, the water level sensor is fixed at the bottom of the shell, the water pressure sensor is fixed at the middle of the shell, and the temperature sensor is located at the top of the shell. The water level sensor and water pressure sensor are in full contact with the water body.

[0025] The above-mentioned mine water inrush disaster prevention and control evaluation method based on fault water intelligent identification and early warning, the height of the water-conducting fracture zone H f The calculation of is shown in formula (3):

[0026]

[0027] In formula (3): H s is the mining height; M is the measured value of the mine pressure; M0 is the standard mine pressure; k f , n is the empirical coefficient based on geological characteristics;

[0028] Critical layer thickness H k The calculation of is shown in formula (4):

[0029]

[0030] In formula (4): h i is the thickness of the i-th rock layer; ρ i is the density of the i-th rock layer; θ i is the inclination of the rock formation;

[0031] Aquifer thickness H w The calculation of is shown in formula (5):

[0032] H w =n w ×D w ×φ(5);

[0033] In formula (5): n w is the number of aquifer layers; D w is the thickness of a single aquifer layer; φ is the porosity of the aquifer;

[0034] Thickness of water barrier H s The calculation of is shown in formula (6):

[0035]

[0036] In formula (6): h j is the thickness of the jth aquiclude; k j is the permeability coefficient of the jth rock layer.

[0037] The above-mentioned mine water inrush disaster prevention and control evaluation method based on fault water intelligent identification and early warning, brittle plastic rock thickness ratio R ip The calculation of is shown in formula (7):

[0038]

[0039] In formula (7): H implic is the total thickness of the brittle rock layer; H p,min is the total thickness of the plastic rock layer;

[0040] Fault strength F s The calculation of is shown in formula (8):

[0041] F s =τ+σ n ×tan(φ f )(8);

[0042] In formula (8): τ is the shear strength on the fault plane; σ n is the normal stress; φ f is the fault friction angle;

[0043] Fault intersection point P c The calculation of is shown in formula (9):

[0044]

[0045] In formula (9): N croos is the number of intersections; A arcea is the area of ​​the region;

[0046] Aquifer P w The calculation of is shown in formula (10):

[0047] P w =ρ w ×g×H w (10);

[0048] In formula (10): w is the water density; g is the acceleration due to gravity; H w is the thickness of the aquifer;

[0049] Water temperature T w The calculation of is shown in formula (11):

[0050] T w =T0+ΔT×exp(-αH s )(11);

[0051] In formula (11): T w is the initial water temperature; ΔT is the temperature change amplitude; α is the heat transfer coefficient; H s is the thickness of the aquiclude.

[0052] The calculation of geological structure characteristics G is shown in formula (12):

[0053]

[0054] In formula (12): i is the weight of the i-th geological feature; E i Score the i-th geological feature;

[0055] Rock mechanical parameters R m The calculation of is shown in formula (13):

[0056]

[0057] In formula (13): c is the compressive strength of rock; E is the elastic modulus; ν is Poisson's ratio.

[0058] The core expression of the ETFM model for the mine water inrush disaster prevention and control evaluation method based on intelligent identification and early warning of fault water is shown in formula (14):

[0059]

[0060] In formula (14), φ(x, z) is the hydraulic potential energy of groundwater at any position (x, z); φ0 is the hydraulic potential of groundwater in the initial static state; ρ is the density of water; g is the acceleration due to gravity; μ reflects the viscosity coefficient of the friction force in the fluid; K x To describe the horizontal flow capacity of groundwater; K z is the vertical flow capacity of groundwater; x is the horizontal coordinate of the groundwater flow field; z is the vertical coordinate of the groundwater flow field.

[0061] The above-mentioned mine water inrush disaster prevention and control evaluation method based on intelligent identification and early warning of fault water, the intelligent drainage system is connected to the drainage device through a control module. When the intelligent drainage system receives the instruction level issued by the multidimensional model early warning system, the control module activates the drainage device and starts the drainage pump in the drainage device.

[0062] The above-mentioned method for preventing and controlling water inrush disasters in mines based on intelligent identification and early warning of fault water, the temperature sensor is connected to the external data cable through a waterproof connector and fixed to the housing through a fixer, the water level sensor uses laser for monitoring; the water pressure sensor is composed of a semiconductor pressure sensitive element, a housing and a protective component.

[0063] The above-mentioned mine water inrush disaster prevention and control evaluation method based on intelligent identification and early warning of fault water is also provided with a grouting isolation cap on the top of the bottom plate pressure water monitor to prevent slurry from affecting the bottom plate pressure water monitor.

[0064] In the above-mentioned mine water inrush disaster prevention and control evaluation method based on intelligent identification and early warning of fault water, the risk level in step three is: low risk: y<T1, where y is the risk value. In the water inrush risk prediction, y represents the water inrush risk level of a certain sample, which is calculated according to the input features through the training model; medium risk: T1<y≤T2; high risk: y≥T2; where T1 and T2 are thresholds for dividing risk levels, which are determined from training data or historical water inrush records.

[0065] Compared with the prior art, the present invention brings the following beneficial technical effects:

[0066] (1) The present invention solves the technical problems existing in traditional methods, such as single monitoring parameters, insufficient intelligence of early warning models, low system integration, and passive prevention and control measures.

[0067] (2) The present invention proposes a mine water inrush disaster prevention and control evaluation method based on intelligent identification and early warning of fault water. By intelligently monitoring the bottom plate pressurized water and combining it with a multi-dimensional model early warning system to evaluate the specific situation of the bottom plate pressurized water, the corresponding risk level is obtained, thereby effectively preventing and warning of bottom plate fault water inrush.

[0068] (3) In the method of the present invention, the bottom plate pressure water monitoring system includes a bottom plate pressure water monitor, a wired communication module and a downhole server. The bottom plate pressure water monitor includes a shell and a water level sensor, a water pressure sensor and a temperature sensor located in the shell. After the collected data is input into the downhole server, the data is transmitted to the multidimensional model early warning system located on the ground through the downhole server. The multidimensional model early warning system calculates and classifies the data.

[0069] (3) In the method of the present invention, the intelligent drainage system is combined with the multidimensional model early warning system. After the multidimensional model early warning system issues an instruction level, the intelligent drainage system is connected to the drainage device through the control module. When the multidimensional model early warning system issues an instruction to start the drainage device, the control module will immediately activate the drainage device, so that the drainage pump starts working. The drainage degree of the drainage device is determined according to the multidimensional model early warning system and the actual situation of the mine. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] The present invention will be further described below with reference to the accompanying drawings:

[0071] Figure 1 Schematic diagram of the arrangement of the systems required in the mine water inrush disaster prevention and control evaluation method based on intelligent identification and early warning of fault water in the present invention;

[0072] Figure 2 This is a detailed diagram of the borehole and bottom plate pressure water monitor of the present invention;

[0073] Figure 3 This is a schematic structural diagram of the bottom plate pressure water monitor of the present invention;

[0074] Figure 4 It is a detailed diagram of part of the structure of the bottom plate pressure water monitor of the present invention;

[0075] Figure 5 Schematic diagram of the layout of the multi-dimensional model early warning system and the intelligent drainage system of the present invention;

[0076] Figure 6 This is a flow chart of the evaluation method of the present invention.

[0077] In the picture:

[0078] 1. Drain pipe, 2. Hydraulic support, 3. Drilling equipment placement and grouting drilling, 4. Bottom plate pressure water monitor line pipeline, 5. Detailed structure of drilling and bottom plate pressure water monitor, 6. Bottom plate pressure water monitor, 7. Drilling grouting wall, 8. Grouting isolation cap, 9. Wired communication module, 10. Fixer, 11. Temperature sensor, 12. Water pressure sensor, 13. Fiber optic information transmission line, 14. Internal wire pipe, 15. Power cord, 16. Water level sensor, 17. Semiconductor pressure sensitive element, 18. Housing and protective components, 19. Intelligent drainage system, 20. Multi-dimensional model early warning system, 21. Filter unit, 22. Signal amplifier, 23. Downhole server. DETAILED DESCRIPTION

[0079] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application.

[0080] Combine Figures 1 to 5As shown, the system required by the present invention includes a bottom plate pressurized water monitoring system, a multi-dimensional model early warning system and an intelligent drainage system. The bottom plate pressurized water monitoring system includes a bottom plate pressurized water monitor 6, a wired communication module 9 and a downhole server 23. The bottom plate pressurized water monitor 6 includes a shell and a water level sensor 16, a water pressure sensor 12, a temperature sensor 11, an optical fiber information transmission line 13, an internal wire pipe 14 and a power line 15 located inside the shell. The wired communication module 9 is located at the top of the bottom plate pressurized water monitor 6, and the wired communication module 9 is used to send and receive information; the water pressure sensor 12 is composed of a semiconductor pressure sensitive element 17, a shell and a protective component 18 to protect internal components from the external environment.

[0081] like Figure 1 As shown, in Figure 1 The diagram shows a drainage pipeline 1, hydraulic supports 2, drilling equipment placement and grouting boreholes 3, floor plate pressure water monitor circuitry 4, and detailed structures of the boreholes and floor plate pressure water monitors 5. Holes are drilled into the fault at regular intervals and in different stages in the floor plate, and floor plate pressure water monitors 6 are installed in the resulting boreholes. A downhole server 23 collects data from each sensor and transmits it to a filter unit 21 via a signal amplifier 22 connected to the downhole server. After filtering, the collected data is sent to a multidimensional model early warning system 20 located on the surface. The multidimensional model early warning system is connected to an intelligent drainage system 19 located on the surface.

[0082] The housing is made of waterproof material, offering excellent water and pressure resistance. It is cylindrical in shape, approximately 50mm in diameter. The height is adjustable, but generally not less than 1m, to ensure adequate coverage for hydrogeological parameter monitoring. The water level sensor is fixed to the bottom of the housing and constructed of stainless steel to prevent corrosion from prolonged immersion in water, ensuring accurate and long-term water level measurement. The water pressure sensor is fixed in the middle of the housing, ensuring full contact with the water while protecting it from direct impact from the water flow. The temperature sensor is located at the top of the housing and connects to the external data cable via a waterproof connector to ensure stable data transmission.

[0083] The intelligent drain system is connected to the drainage device through the control module. When the intelligent drain system receives the instruction level issued by the multidimensional model early warning system, the control module activates the drainage device and starts the drainage pump in the drainage device.

[0084] The temperature sensor is connected to the external data line through a waterproof connector and fixed to the housing through a holder 10. The water level sensor uses laser for monitoring; the water pressure sensor is composed of a semiconductor pressure sensitive element, a housing and a protective component.

[0085] Preferably, a grouting isolation cap 8 is provided on the top of the bottom plate pressure water monitor to prevent the slurry from affecting the bottom plate pressure water monitor. The grouting isolation cap 8 is used to prevent the slurry from affecting the operation of the bottom plate pressure water monitor 6 and can be fixed in the slurry. The drilled grouting wall 7 is made of waterproof and corrosion-resistant Babbitt alloy material.

[0086] The coal seam floor working face is supported by a hydraulic support 2, and a drainage pipe 1 and drilling equipment are placed and a grouting hole 3 is drilled toward the fault on the ground. The borehole diameter is about 60 mm. The borehole is drilled to the fault and arranged in steps at a certain interval according to the actual situation on site. A bottom plate pressure water monitor 6 is arranged at the bottom of the equipment placement and grouting borehole 3 to monitor and collect information such as water level, water pressure and water temperature.

[0087] The data collected by the bottom plate pressure water monitor 6 is stored in the downhole server 23, transmitted to the signal amplifier 22 to strengthen information transmission to prevent information loss, and then the signal is denoised by the filter unit 21, and finally transmitted back to the multi-dimensional fusion model early warning system 20 for the final water inrush probability assessment. The specific assessment process is as follows: Figure 6 As shown, the specific steps are:

[0088] Step 1: Analyze through the surface big data information processing system, select the height of the water-conducting fracture zone, the thickness of the key layer, the thickness of the aquifer, the thickness of the impermeable layer, the thickness ratio of brittle-plastic rock, the strength of the fault, the fault intersection, the density of the pinch-out point, the water pressure of the aquifer, the water temperature, the geological structure characteristics, and the rock mechanics parameters, and calculate the weight of the selected data using the entropy weight method.

[0089] (1) Calculation of main parameters affecting bottom plate water inrush

[0090] Height of water-conducting fracture zone H f The calculation of is shown in formula (1):

[0091]

[0092] In formula (1): H s is the mining height; M is the measured value of the mine pressure; M0 is the standard mine pressure; k f , n is the empirical coefficient based on geological characteristics;

[0093] Critical layer thickness H k The calculation of is shown in formula (2):

[0094]

[0095] In formula (2): h i is the thickness of the i-th rock layer; ρ i is the density of the i-th rock layer; θ i is the inclination of the rock formation;

[0096] Aquifer thickness H w The calculation of is shown in formula (3):

[0097] H w =n w ×D w ×φ(3);

[0098] In formula (3): n w is the number of aquifer layers; D w is the thickness of a single aquifer layer; φ is the porosity of the aquifer;

[0099] Thickness of water barrier H s The calculation of is shown in formula (4):

[0100]

[0101] In formula (4): h j is the thickness of the jth aquiclude; k j is the permeability coefficient of the jth rock layer.

[0102] Brittle-plastic rock thickness ratio R ip The calculation of is shown in formula (5):

[0103]

[0104] In formula (5): H implic is the total thickness of the brittle rock layer; H p,min is the total thickness of the plastic rock layer;

[0105] Fault strength F s The calculation of is shown in formula (6):

[0106] F s =τ+σ n ×tan(φ f )(6);

[0107] In formula (6): τ is the shear strength on the fault plane; σ n is the normal stress; φ f is the fault friction angle;

[0108] Fault intersection point P c The calculation of is shown in formula (7):

[0109]

[0110] In formula (7): N croos is the number of intersections; A arcea is the area of ​​the region;

[0111] Aquifer P w The calculation of is shown in formula (8):

[0112] P w =ρ w ×g×H w (8);

[0113] In formula (8): w is the water density; g is the acceleration of gravity; Hw is the thickness of the aquifer;

[0114] Water temperature T w The calculation of is shown in formula (9):

[0115] T w =T0+ΔT×exp(-αH s )(9);

[0116] In formula (9): T w is the initial water temperature; ΔT is the temperature change amplitude; α is the heat transfer coefficient; H s is the thickness of the aquiclude. The calculation of the geological structure characteristic G is shown in formula (10):

[0117]

[0118] In formula (10): i is the weight of the i-th geological feature; E i Score the i-th geological feature;

[0119] Rock mechanical parameters R m The calculation of is shown in formula (11):

[0120]

[0121] In formula (11): c is the compressive strength of rock; E is the elastic modulus; ν is Poisson's ratio.

[0122] (2) Data standardization

[0123] The data normalization formula is:

[0124]

[0125] In formula (12): r ij is the standard quantization value, x ij is the jth eigenvalue of the i-th sample.

[0126] (3) Calculating information entropy

[0127] According to the standardized data r ij , calculate the probability distribution p of each parameter ij :

[0128]

[0129] (4) Calculate weight

[0130] Determine the weight w of each parameter according to the entropy value j :

[0131]

[0132] This weight w j It will be used to construct the comprehensive evaluation function.

[0133] Step 2: To improve the accuracy of water inrush prediction, the parameters such as water pressure distribution and geological conditions in the extended Toth flow field model are introduced into the particle swarm optimization support vector machine (PSO-SVM model to form a new water inrush prediction model.

[0134] Extended Toth Flow Model (ETFM):

[0135] The aquifer water pressure field distribution φ(x,z) and related parameters of geological conditions are extracted as characteristic inputs, and the core expression of the extended Toth model is:

[0136]

[0137] Extended to the characteristics of water inrush from coal mine floors, the comprehensive water pressure field distribution P is defined t :

[0138]

[0139] Where: ρgH w is the hydrostatic pressure, MPa; γT w is the correction term of water temperature on hydrodynamics; G i Comprehensive factors of geological conditions (such as fault intersection density, height of water-conducting fracture zones, etc.); β i The influence weight of geological factors.

[0140] (1) SVM model construction

[0141] SVM basic expression:

[0142] Input eigenvalues ​​X={x1,x2,...,x n}, the goal is to build a decision function:

[0143]

[0144] Where: K(x,x i ) is the kernel function, used for high-dimensional space mapping; α i is the support vector weight; b is the bias.

[0145] Kernel function selection, choose radial basis kernel function (RBF):

[0146]

[0147] Where: C is the penalty coefficient (control coefficient); γ is the kernel function width.

[0148] (2) PSO optimization of SVM parameters

[0149] Particle swarm optimization (PSO) is used to optimize the SVM hyperparameters C and γ, and the position update formula is:

[0150]

[0151]

[0152] Where: is the particle velocity; is the particle position; is the individual's best historical position; g t is the global historical optimal position; w, c1, c2 are weights and acceleration factors; r1, r2 are random numbers.

[0153] (3) PSO-SVM parameter optimization steps

[0154] ① Initialize the particle swarm and randomly set C and γ;

[0155] ②Calculate the fitness function of SVM (such as cross-validation accuracy);

[0156] ③ Update particle velocity and position to find the optimal C and γ;

[0157] ④Output the optimal parameters and build the final SVM model.

[0158] (4) New prediction model

[0159] The extended Toth flow field model and POS-SVM are input, and the complete water inrush prediction model includes the following steps:

[0160] Input the feature data calculated by the extended Toth model, perform data preprocessing, use PSO to optimize the SVM hyperparameters C and γ, train the SVM, and divide the risk level according to the set threshold:

[0161] Low risk: y<T; medium risk: y<T1; high risk: y≥T2.

[0162] (5) Complete formula framework

[0163] ① Extended Toth flow field model:

[0164]

[0165] ②SVM decision function:

[0166]

[0167] ③PSO optimization formula:

[0168]

[0169]

[0170] Step 3: Optimize the water inrush limit value using game theory

[0171] Nash equilibrium solution:

[0172]

[0173] Low risk: High risk:

[0174] Step 4: The intelligent drainage system performs different drainage and control instructions based on the command level issued by the multidimensional model early warning system. The intelligent drainage system is connected to the drainage device through the control module. When the multidimensional model early warning system issues an instruction to start the drainage device, the control module will immediately activate the drainage device and start the drainage pump. The drainage level (i.e., drainage volume) of the drainage device is determined by the multidimensional model early warning system and the actual situation of the mine. The prevention and control measures execution system plays a vital role. When the multidimensional model early warning system detects a potential risk of water inrush and issues a corresponding instruction, the system will respond quickly and implement the corresponding prevention and control measures.

[0175] The intelligent drainage system mainly includes the reception of multi-dimensional model early warning system instructions and the issuance of drainage signals, the connection and control of drainage devices, and the determination of drainage volume;

[0176] The system receives warning instructions from a multi-dimensional model warning system, which then issues drainage signals. These instructions are generated based on a comprehensive analysis of mine geological conditions, hydrological parameters, historical data, and current monitoring data. These drainage instructions are often specific and explicit operational commands, including but not limited to activating drainage systems, reinforcing support structures, and evacuating personnel.

[0177] Drainage systems are a crucial component of mine prevention and control measures, responsible for promptly draining accumulated water from the mine and reducing the risk of water inrush. The multidimensional model early warning system is connected to the drainage system via a control module, enabling remote control. When the multidimensional model early warning system issues a command to start the drainage system, the control module immediately activates the drainage system and begins pumping. The drainage level (i.e., drainage volume) of the drainage system is determined based on the warning value and the actual conditions of the mine.

[0178] Among them, the determination of displacement is a process of comprehensive consideration of multiple factors. The following are some of the main considerations:

[0179] ① Warning value: The warning value is set based on factors such as mine geological conditions and historical data. It reflects the current risk of water inrush. When the warning value exceeds the set threshold, the drainage system needs to be activated, and the drainage volume is determined based on the warning value.

[0180] ② Mine conditions: Mine conditions include the depth, extent, and rate of water accumulation. These factors influence the determination of drainage volume. For example, the deeper the water, the greater the volume of water that needs to be removed; the faster the water accumulation, the stronger the drainage system needs to be.

[0181] ③ Drainage device performance: Different models of drainage devices have different drainage capacities and efficiencies. When selecting a drainage device, you need to consider its performance parameters to ensure that it can meet your drainage needs.

[0182] Based on the above factors, the following steps can be used to determine the water discharge:

[0183] ① Monitoring and early warning values: Real-time monitoring of the mine’s hydrological parameters. When the early warning value exceeds the set threshold, the drainage device is triggered.

[0184] ②Evaluate the actual situation of the mine: Evaluate the actual situation of the mine based on parameters such as water accumulation depth, water accumulation range, and water accumulation speed.

[0185] ③Select drainage device: Select appropriate drainage device according to the actual situation of the mine and the performance parameters of the drainage device.

[0186] ④ Set the drainage volume: According to the warning value and the actual situation of the mine, set a reasonable drainage volume to ensure that the accumulated water can be discharged in a timely and effective manner.

[0187] It should be noted that determining the drainage volume is a dynamic process that requires continuous adjustment and optimization based on actual conditions. At the same time, regular inspection and maintenance of the drainage system is also required to ensure its normal operation and stable performance.

[0188] Parts not described in the present invention can be implemented by referring to the existing technology.

[0189] Those skilled in the art should recognize that the above embodiments are merely intended to illustrate the present application and are not intended to limit the present application. Any appropriate changes and modifications to the above embodiments should fall within the scope of protection of the claims of the present application as long as they are within the spirit of the present application.

Claims

1. A mine water inrush disaster prevention and control evaluation method based on intelligent identification and early warning of fault water, characterized in that: The following steps are involved: (1) Install the required system The system includes a bottom plate pressure water monitoring system, a multidimensional model early warning system and an intelligent drainage system. The bottom plate pressure water monitoring system includes a bottom plate pressure water monitor, a wired communication module and a downhole server. The bottom plate pressure water monitor includes a housing and a water level sensor, a water pressure sensor and a temperature sensor located inside the housing. The wired communication module is located on the top of the bottom plate pressure water monitor. Holes are drilled into the fault at a certain interval and different levels in the bottom plate, and the bottom plate pressure water monitor is installed in the resulting boreholes. The downhole server is used to collect data obtained by each sensor and send it to a filtering unit through a signal amplifier connected to the downhole server. The collected data is sent to the multidimensional model early warning system located on the ground through the filtering unit. The multidimensional model early warning system is connected to the intelligent drainage system located on the ground. (2) After analysis, the main parameters of geological conditions are selected: height of water-conducting fracture zone, thickness of key layer, thickness of aquifer, thickness of impermeable layer, thickness ratio of brittle-plastic rock, fault strength, fault intersection, density of pinch-out points, water pressure of aquifer, water temperature, geological structure characteristics and rock mechanics. The weight of each parameter is calculated by entropy weight method to obtain the weight of each parameter; (3) The weight of each parameter in step (2) is imported into the PSO-SVM model of the extended Toth model to form the extended Toth flow field model, namely the ETFM model, which is further extended to the water inrush characteristics of the coal mine floor and defines the comprehensive water pressure field distribution P t , as shown in formula (1): In formula (1): ρgH w The product of is the hydrostatic pressure; γT w The product of water temperature and water dynamics is the correction term; G i is a comprehensive factor of geological conditions; β i is the influence weight of each parameter of geological conditions; i, n is the number of main parameters affecting floor water inrush, i = 1, 2, 3, ..., n; PSO is used to optimize the penalty parameter C of SVM and the parameter γ of radial basis kernel function RBF, train SVM, and divide the risk level according to the set threshold; (4) According to game theory, the corresponding water inrush risk warning value is obtained, as shown in formula (2): In formula (2): P t * is the optimal value of water inrush risk probability; P t is the probability of water inrush risk; U is the utility function; Through formula (2), we can get: when P t <P t * When P t ≥P t * When the risk is high; (5) The intelligent drainage system executes different drainage and control instructions according to the instruction level issued by the multi-dimensional model early warning system, and then determines the drainage volume.

2. A mine water inrush disaster prevention and control evaluation method based on intelligent identification and early warning of fault water according to claim 1, characterized in that: The shell is made of waterproof material, the water level sensor is fixed at the bottom of the shell, the water pressure sensor is fixed at the middle of the shell, and the temperature sensor is located at the top of the shell. The water level sensor and water pressure sensor are in full contact with the water body.

3. The method for preventing and controlling water inrush disasters in mines based on intelligent identification and early warning of fault water according to claim 1 is characterized in that: Height of water-conducting fracture zone H f The calculation of is shown in formula (3): In formula (3): H s is the mining height; M is the measured value of the mine pressure; M0 is the standard mine pressure; k f , n is the empirical coefficient based on geological characteristics; Critical layer thickness H k The calculation of is shown in formula (4): In formula (4): h i is the thickness of the i-th rock layer; ρ i is the density of the i-th rock layer; θ i is the inclination of the rock formation; Aquifer thickness H w The calculation of is shown in formula (5): H w =n w ×D w ×φ (5); In formula (5): n w is the number of aquifer layers; D w is the thickness of a single aquifer layer; φ is the porosity of the aquifer; Thickness of water barrier H s The calculation of is shown in formula (6): In formula (6): h j is the thickness of the jth aquiclude; k j is the permeability coefficient of the jth rock layer.

4. The method for preventing and controlling water inrush disasters in mines based on intelligent identification and early warning of fault water according to claim 3 is characterized by: Brittle-plastic rock thickness ratio R ip The calculation of is shown in formula (7): In formula (7): H implic is the total thickness of the brittle rock layer; H p,min is the total thickness of the plastic rock layer; Fault strength F s The calculation of is shown in formula (8): F s =t+s n ×tan(φ f ) (8); In formula (8): τ is the shear strength on the fault plane; σ n is the normal stress; φ f is the fault friction angle; Fault intersection point P c The calculation of is shown in formula (9): In formula (9): N croos is the number of intersections; A arcea is the area of ​​the region; Aquifer P w The calculation of is shown in formula (10): P w =ρ w ×g×H w (10); In formula (10): w is the water density; g is the acceleration of gravity; Hw is the thickness of the aquifer; Water temperature T w The calculation of is shown in formula (11): T w =T0+ΔT×exp(-αH s ) (11); In formula (11): T w is the initial water temperature; ΔT is the temperature change amplitude; α is the heat transfer coefficient; H s is the thickness of the aquiclude; the calculation of the geological structure characteristic G is shown in formula (12): In formula (12): i is the weight of the i-th geological feature; E i Score the i-th geological feature; Rock mechanical parameters R m The calculation of is shown in formula (13): In formula (13): c is the compressive strength of rock; E is the elastic modulus; ν is Poisson's ratio.

5. The method for preventing and controlling water inrush disasters in mines based on intelligent identification and early warning of fault water according to claim 1 is characterized in that: The core expression of the ETFM model is shown in formula (14): In formula (14), φ(x, z) is the hydraulic potential energy of groundwater at any position (x, z); φ0 is the hydraulic potential of groundwater in the initial static state; ρ is the density of water; g is the acceleration due to gravity; μ reflects the viscosity coefficient of the friction force in the fluid; K x To describe the horizontal flow capacity of groundwater; K z is the vertical flow capacity of groundwater; x is the horizontal coordinate of the groundwater flow field; z is the vertical coordinate of the groundwater flow field.

6. The method for preventing and controlling water inrush disasters in mines based on intelligent identification and early warning of fault water according to claim 1, characterized in that: The intelligent drain system is connected to the drainage device through the control module. When the intelligent drain system receives the instruction level issued by the multidimensional model early warning system, the control module activates the drainage device and starts the drainage pump in the drainage device.

7. The method for preventing and controlling water inrush disasters in mines based on intelligent identification and early warning of fault water according to claim 2, characterized in that: The temperature sensor is connected to the external data line through a waterproof connector and fixed to the housing through a fixer. The water level sensor uses laser for monitoring; the water pressure sensor consists of a semiconductor pressure sensitive element, a housing and a protective component.

8. The method for preventing and controlling water inrush disasters in mines based on intelligent identification and early warning of fault water according to claim 1, characterized in that: A grouting isolation cap is also provided on the top of the bottom plate pressure water monitor to prevent slurry from affecting the bottom plate pressure water monitor.

9. The method for preventing and controlling water inrush disasters in mines based on intelligent identification and early warning of fault water according to claim 1, characterized in that: The risk levels in step three are: low risk: y<T1, where y is the risk value; medium risk: T1<y≤T2; high risk: y≥T2; where y represents the water inrush risk level of a sample, which is calculated based on the input features through the training model; T1 and T2 are thresholds used to divide the risk levels, which are determined from training data or historical water inrush records.

Images