A coal mine underground goaf water disaster prevention and monitoring and early warning method

Abstract

The application provides a coal mine underground goaf water disaster prevention and monitoring and early warning method, relates to the water disaster disaster prevention and control and intelligent monitoring technical field in coal mine engineering. The method comprises the following steps: obtaining physical parameters of a monitoring area of a goaf for reflecting a water disaster evolution process; constructing a three-dimensional hydrology and mechanics coupling inversion model of the goaf; inputting the physical parameters into the three-dimensional hydrology and mechanics coupling inversion model to obtain a water pressure change rate, a seepage flux change rate and a fracture expansion rate of the goaf output by the three-dimensional hydrology and mechanics coupling inversion model; calculating a water disaster disturbance risk value of the goaf according to the water pressure change rate, the seepage flux change rate, the fracture expansion rate and a preset water disaster disturbance risk function; and determining a risk level of the water disaster disturbance risk value and calling a corresponding early warning strategy according to a preset risk level division threshold. The technical scheme disclosed by the application improves the water disaster prediction and early warning capability.

Description

Technical Field

[0001] This application relates to the field of water hazard prevention and intelligent monitoring technology in coal mine engineering, specifically, to a method for water hazard prevention, monitoring and early warning in underground goaf areas of coal mines. Background Technology

[0002] During deep mining operations, as the working face advances and mining disturbances intensify, the problem of water accumulation in goaf areas becomes increasingly prominent, posing a key obstacle to the safety, continuity, and green transformation of deep resource development. Goaf areas are structurally fragmented and have complex hydraulic connections, making them highly susceptible to the formation of hidden high-pressure water bodies. Once a seepage path is established, it can easily trigger water inrushes and collapses within a short period, especially in mines equipped with water-retaining coal pillars or natural water storage spaces, where the risks are particularly severe.

[0003] Traditional methods of water hazard prevention and control mainly rely on localized drainage, static exploration, and experience-based prediction, lacking spatial awareness of the complex seepage structure in goaf areas and dynamic intervention methods. Furthermore, current monitoring systems are mostly deployed at points, with single parameters, unable to establish continuous water pressure fields or identify dynamic fracture connectivity structures; early warning mechanisms are generally based on fixed thresholds, failing to reflect the evolution trend of disturbances; and response measures lack intelligent linkage, failing to achieve rapid control and closed-loop regulation. These technological bottlenecks directly limit the predictability and proactive prevention and control capabilities of sudden water hazards, resulting in a persistently high risk of mine water inrush accidents. Summary of the Invention

[0004] The technical problem to be solved by this application is that the early warning accuracy and efficiency of existing water hazard prevention methods are low, and therefore a method for water hazard prevention and monitoring early warning in underground goaf areas of coal mines is provided.

[0005] This application provides a method for the prevention, monitoring, and early warning of water hazards in underground goaf areas of coal mines, including:

[0006] Physical parameters for reflecting the evolution of water hazards in the goaf monitoring area are obtained; wherein, the goaf monitoring area includes the water storage space and its adjacent area, the surrounding rock failure zone area, and the area of ​​fracture development and seepage path; the physical parameters include dynamic water pressure monitoring data, surrounding rock micro-deformation monitoring data, electrical anomaly monitoring data, and structural fracture monitoring data;

[0007] A three-dimensional hydrological and mechanical coupled inversion model for the goaf is constructed. The three-dimensional hydrological and mechanical coupled inversion model includes an unsteady Darcy flow control model for inverting the pore water pressure field and the seepage velocity field, and a fracture strain evolution model for describing the tensile and shear cracking processes induced by structural disturbance. The unsteady Darcy flow control model and the fracture strain evolution model achieve physical field linkage feedback through the permeability and strain coupling relationship.

[0008] The physical parameters are input into the three-dimensional hydrological and mechanical coupled inversion model to obtain the water pressure change rate, seepage flow rate, and fracture propagation rate of the goaf output by the three-dimensional hydrological and mechanical coupled inversion model.

[0009] The water hazard risk value of the goaf is calculated based on the water pressure change rate, the seepage flow rate change rate, the fracture propagation rate, and the preset water hazard disturbance risk function.

[0010] Based on the water hazard disturbance risk value and the preset risk level classification threshold, the risk level of the water hazard disturbance risk value is determined and the early warning strategy corresponding to the risk level is invoked.

[0011] Preferably, in the method for prevention, monitoring, and early warning of water hazards in underground goaf areas of coal mines, the construction of a three-dimensional hydrological and mechanical coupled inversion model of the goaf area includes an unsteady Darcy flow control model for inverting the pore water pressure field and seepage velocity field, and a fracture strain evolution model for describing the tensile and shear cracking processes induced by structural disturbances. The unsteady Darcy flow control model and the fracture strain evolution model achieve physical field linkage feedback through the permeability and strain coupling relationship, including:

[0012] Construct the unsteady Darcy flow control model:

[0013] ;

[0014] in, Indicates a time step. Represents the spatial coordinates of the center point of a small volume element. For small volume units, the seepage velocity vector is... For the equivalent permeability tensor, For small volume units, Moisture content per unit volume For the instantaneous water pressure field of a small volume unit, The density of water, Elevation;

[0015] Construct the fracture strain evolution model:

[0016] ;

[0017] in, The crack strain of the small volume unit is... The crack propagation diffusion coefficient is... For the Laplace operator of the strain field, For the crack source term of the small volume unit, For parameter calibration coefficients, To collect event frequency;

[0018] The relationship between permeability and strain is expressed as follows:

[0019] ;

[0020] in, This is the basic permeability field of the small volume unit. The sensitivity coefficient for enhancing permeability during fracture propagation;

[0021] The seepage velocity vector, hydraulic head, and instantaneous water pressure field are determined by the dynamic water pressure monitoring data; the fracture strain and fracture source term are determined by the surrounding rock micro-deformation monitoring data; the frequency of the collected events is determined by the structural rupture monitoring data; and the foundation permeability field is determined by the electrical anomaly monitoring data.

[0022] Preferably, in the method for prevention, monitoring, and early warning of water hazards in underground goaf areas of coal mines, the unsteady Darcy flow control model and the fracture strain evolution model are configured with the following boundary conditions:

[0023] Seepage field boundary conditions: Apply no-flow boundary conditions at the outer boundary of the water storage space. n is the normal vector outside the boundary; at the boundary of the surrounding rock failure zone, a constant water head or constant water pressure boundary condition is applied. , This refers to the real-time water pressure in the goaf area.

[0024] Structural disturbance field boundary conditions: Apply fixed displacement boundary conditions at the boundary of the stable region of the surrounding rock. At the boundary of the surrounding rock failure zone, apply free tensile crack propagation conditions. .

[0025] Preferably, the method for prevention, monitoring, and early warning of water hazards in underground goaf areas of coal mines, wherein the construction of the unsteady Darcy flow control model and the construction of the fracture strain evolution model further include:

[0026] If the dynamic water pressure monitoring data indicates that the water pressure fluctuation exceeds the static threshold, then the real-time collected dynamic water pressure monitoring data will be loaded into the boundary input of the unsteady Darcy seepage control model according to the time step, for use in... Perform dynamic updates;

[0027] If the structural rupture monitoring data represents the microseismic frequency If the set threshold is exceeded, the strain evolution model of the crack will be... Switch from a conventional linear growth mode to an exponential amplification mode; adjust the sensitivity coefficient of the penetration rate-strain coupling relationship. This characterizes the increased water conductivity of fractures;

[0028] If the monitoring data on micro-deformation of the surrounding rock indicates the presence of fracture structures within the goaf, then this is the fracture source term in the fracture strain evolution model. Give a higher initial value or weaken the crack propagation diffusion coefficient .

[0029] Preferably, in the method for prevention, monitoring, and early warning of water hazards in underground goaf areas of coal mines, the water hazard disturbance risk value of the goaf area is calculated based on the water pressure change rate, the seepage flow change rate, the fracture propagation rate, and a preset water hazard disturbance risk function:

[0030] Preset water hazard disturbance risk function This can be achieved using the following function:

[0031] ;

[0032] The water pressure change rate is... The rate of change of the permeation flow is... The crack propagation rate, , , These are the corresponding risk weight coefficients;

[0033] The preset water hazard disturbance risk function is normalized:

[0034] ;

[0035] For the same time step t The mean of the calculated results of the water hazard disturbance risk function corresponding to all small volume units. For the same time step t Standard deviation of the water hazard disturbance risk function calculation results for all small volume units;

[0036] The calculated result of the preset water hazard disturbance risk function after normalization is used as the water hazard disturbance risk value.

[0037] Preferably, in the method for prevention, monitoring, and early warning of water hazards in underground goaf areas of coal mines, the step of determining the risk level of the water hazard disturbance risk value based on the water hazard disturbance risk value and a preset risk level classification threshold, and then calling the early warning strategy corresponding to the risk level, is achieved by obtaining the risk level of the water hazard disturbance risk value in the following manner:

[0038] When the risk value of water-related disturbance exceeds the first threshold, it is classified as Level 1 risk.

[0039] When the second threshold is less than the water hazard disturbance risk value and less than the first threshold, it is classified as a level two risk.

[0040] When the third threshold is less than the water hazard disturbance risk value and less than or equal to the second threshold, it is a level three risk level.

[0041] When the risk value of water hazard disturbance is less than or equal to the third threshold, it is classified as a level four risk level, corresponding to a safe and stable zone.

[0042] Preferably, the method for preventing and monitoring water hazards in underground goaf areas of coal mines, wherein the risk level of the water hazard disturbance risk value is obtained further includes:

[0043] If continuous The small volume units are spatially adjacent and all belong to If the risk level is specified, then the aggregation area is calculated: , For the first The volume of a small volume unit, when At that time, it is determined that the aggregation region has formed. An effective early warning aggregation system for risk levels; for The minimum effective volume threshold corresponding to the risk level; the above One, two, three, or four;

[0044] Write the effective early warning aggregate into the dynamic risk index table: , Indicates the first i In time Identified as An effective early warning aggregation system for risk levels. for Index table of risk areas at risk level 1 for The threshold corresponding to the risk level is: To effectively predict the volume of polymers, The minimum effective warning aggregate volume determination threshold;

[0045] A dynamic risk map is generated in real time based on the dynamic risk index table.

[0046] Based on the risk level of each valid early warning aggregate in the dynamic risk map, the early warning strategy is invoked:

[0047] The early warning strategy for Level 1 risk is to implement a mandatory response;

[0048] The early warning strategy for a level 2 risk is to implement pre-set early warning and control measures;

[0049] The early warning strategy for Level 3 risk is to adjust the monitoring frequency and expand the monitoring scope;

[0050] The early warning strategy for Level 4 risk is to maintain routine monitoring.

[0051] Preferably, the method for prevention, monitoring, and early warning of water hazards in underground goaf areas of coal mines, after invoking the early warning strategy corresponding to the risk level, further includes:

[0052] For effective early warning aggregates at risk levels one and two, if their water hazard disturbance risk value is continuously... The change within a time step exceeds a set threshold. This will increase the monitoring intensity of the effective early warning aggregate.

[0053] Preferably, the method for prevention, monitoring, and early warning of water hazards in underground goaf areas of coal mines, wherein the step of invoking an early warning strategy based on the risk level of each effective early warning aggregate in the dynamic risk map includes:

[0054] Based on the i-th valid early warning aggregate Risk level generation trigger signal: , for The response priority corresponding to the risk level is used for the resource scheduling and sorting mechanism in the response queue, and the mapping rules are as follows:

[0055] ;

[0056] Based on the aforementioned response priority, a response execution criterion is constructed: if Then the i-th valid early warning aggregate will be The trigger signal is added to the response scheduling queue; The minimum response priority threshold is set.

[0057] According to each of the aforementioned trigger signals , call Response strategy function corresponding to risk level F k Output control parameter set: , The response instruction set corresponding to the risk level includes mandatory response, early warning and control measures, adjustment of monitoring frequency and expansion of monitoring scope and scope adjustment, and maintenance of routine monitoring.

[0058] Preferably, in the method for prevention, monitoring, and early warning of water hazards in underground goaf areas of coal mines, the step of acquiring physical parameters of the goaf monitoring area used to reflect the evolution process of water hazards includes:

[0059] All physical parameters have been normalized to be dimensionless.

[0060] The technical solution provided in this application has the following technical effects compared with the prior art:

[0061] The method for water hazard prevention and monitoring / early warning in underground goaf areas provided in this application collects multi-dimensional physical parameters for the goaf monitoring area, which includes the water storage space and its adjacent area, the surrounding rock failure zone, and the area with fracture development and seepage paths. The physical parameters include dynamic water pressure monitoring data, surrounding rock micro-deformation monitoring data, electrical anomaly monitoring data, and structural fracture monitoring data. A three-dimensional hydrological and mechanical coupled inversion model is constructed, including an unsteady Darcy seepage control model and a fracture strain evolution model, as well as the permeability and strain coupling relationship between the two models. This scheme calculates the water pressure change rate, seepage flow change rate, and fracture propagation rate of the goaf based on the physical parameters using the three-dimensional hydrological and mechanical coupled inversion model. Based on the water pressure change rate, seepage flow change rate, fracture propagation rate, and a preset water hazard disturbance risk function, the water hazard disturbance risk value of the goaf is calculated. Based on the water hazard disturbance risk value and a preset risk level classification threshold, the risk level of the water hazard disturbance risk value is determined, and the corresponding early warning strategy is invoked. This solution enables multi-physics field monitoring, three-dimensional hydrological and mechanical coupled inversion modeling, risk level classification and early warning strategy invocation and control, solving the problems of delayed response, unclear regional control and low prediction accuracy in traditional coal mine water hazard monitoring, fundamentally improving water hazard prediction and early warning capabilities and system response rate. Attached Figure Description

[0062] Figure 1 This is a flowchart of a method for preventing and monitoring water hazards in underground goaf areas of coal mines, as described in one embodiment of this application.

[0063] Figure 2 This is a flowchart illustrating how, according to one embodiment of this application, a risk level is determined based on the water hazard disturbance risk value, and a corresponding early warning strategy is invoked based on that risk level. Detailed Implementation

[0064] The specific embodiments of this application will be further described below with reference to the accompanying drawings.

[0065] It is readily understood that, based on the technical solution of this application, various structural and implementation methods can be interchanged by those skilled in the art without altering the essential spirit of this application. Therefore, the following detailed embodiments and accompanying drawings are merely illustrative examples of the technical solution of this application and should not be considered as the entirety of this application or as limitations or restrictions on the technical solution of the application.

[0066] The following embodiments of this application provide a method for the prevention and monitoring of water hazards in underground goaf areas of coal mines based on the coupling of multi-source sensing, three-dimensional recognition, dynamic early warning and automatic response. It establishes a whole-process prevention and control mechanism for underground goaf areas from "monitoring and identification" to "risk control", fundamentally improving the ability to predict and warn of water hazards and the system response rate.

[0067] This embodiment provides a method for the prevention, monitoring, and early warning of water hazards in underground goaf areas of coal mines, which is applied to a monitoring system equipped with an operating system, such as... Figure 1 As shown, it includes:

[0068] S100: Obtain physical parameters of the goaf monitoring area to reflect the evolution of water hazards; wherein, the goaf monitoring area includes the water storage space and its adjacent area, the surrounding rock failure zone area, and the area of ​​fracture development and seepage path; the physical parameters include water pressure dynamic monitoring data, surrounding rock micro-deformation monitoring data, electrical anomaly monitoring data, and structural fracture monitoring data.

[0069] In practical applications, the aforementioned physical parameters can be acquired through sensors installed in the monitoring area of ​​the goaf, or through methods such as borehole sampling. Specifically, multi-source heterogeneous sensors are deployed in the key structural zones, water storage areas, and fracture channels of the goaf to form a spatially continuous, physically multidimensional, and structured data acquisition network, which can acquire key parameters such as water pressure, strain, seepage, electrical properties, and microseismic signals in real time.

[0070] Based on the geological structure model and the distribution characteristics of water storage risk, the monitoring areas are mainly divided into the following three categories: water storage space and its adjacent areas, including representative old goaf areas and residual water accumulation zones in mined-out areas; surrounding rock failure zones, including coal pillar dam boundaries, fault intersections, and working face advancement influence zones; and areas with fracture development and seepage paths, i.e., potential high-pressure water inrush channels. Within these three monitoring areas, sensors are deployed according to the principles of functional priority, regional representativeness, and data complementarity. The deployment spacing is generally 10–50 m, which can be adjusted according to underground space conditions. The types of sensors deployed include:

[0071] The water pressure and water level monitoring unit collects real-time water pressure data from the water accumulation area in the goaf. With water level ;

[0072] Distributed fiber optic strain gauge arrays are installed at key locations in the surrounding rock and coal pillars to monitor total strain. Used to identify the crack opening region and correct the crack strain of the model. ;

[0073] The resistivity sensing module uses a multi-point switchable four-electrode method to measure resistivity. Inverting changes in water-rich bodies and their conduction channels;

[0074] Microseismic and acoustic emission identification devices are deployed in fault or stress concentration areas to collect event frequencies. With energy It is used to identify early signs of instability such as crack slip and local fracture.

[0075] In addition, regarding the permeability parameters of the surrounding rock, such as porosity Moisture content Basic penetration rate Accurate values ​​can be obtained through drilling sampling and periodic calibration via indoor testing.

[0076] Various sensors form a local wired network via RS485 bus or fiber optic transmission, supplemented by LoRa or ZigBee wireless relay nodes to form a hybrid network structure. All data is aggregated to the underground relay gateway and then uploaded to the ground monitoring system via industrial Ethernet or 4G / 5G channels for data visualization and remote control.

[0077] In practical applications, the monitoring system adopts a time synchronization mechanism. When acquiring physical parameters, it supports unified timing and structured processing of multi-source data, and also has local preprocessing and anomaly filtering functions to improve acquisition efficiency and anti-interference capabilities. The monitoring system of this application can achieve full-process, spatial, and continuous perception of the state evolution characteristics of water pressure changes in goaf storage units, structural disturbances during fracture conduction, the decreasing trend of resistivity in water conduction paths, and changes in microseismic signal frequency. This provides raw data support for subsequent three-dimensional hydrological and mechanical coupled inversion models.

[0078] S200: Construct a three-dimensional hydrological and mechanical coupled inversion model for the goaf. The three-dimensional hydrological and mechanical coupled inversion model includes an unsteady Darcy flow control model for inverting the pore water pressure field and seepage velocity field, and a fracture strain evolution model for describing the tensile and shear cracking processes induced by structural disturbance. The unsteady Darcy flow control model and the fracture strain evolution model achieve physical field linkage feedback through the permeability and strain coupling relationship.

[0079] In this step, based on the multi-source physical parameters obtained in step S100, a three-dimensional hydrological and mechanical coupled inversion model based on small volume units is established to invert the pore water pressure field and seepage velocity field, describe the tensile and shear cracking processes induced by structural disturbance, and achieve coupling through the coupling relationship between permeability and strain. That is, the three-dimensional hydrological and mechanical coupled inversion model can be used to dynamically reconstruct the seepage structure, water pressure evolution state and surrounding rock fracture propagation process inside the goaf.

[0080] S300: Input the physical parameters into the three-dimensional hydrological and mechanical coupled inversion model to obtain the water pressure change rate, seepage flow change rate and fracture propagation rate of the goaf output by the three-dimensional hydrological and mechanical coupled inversion model.

[0081] The three-dimensional hydrological and mechanical coupled inversion model uses the multi-source, real-time physical parameters from step S100 to simultaneously solve for seepage behavior and fracture evolution in the same three-dimensional physical field, which can yield the rate of change of water pressure, the rate of change of seepage flow, and the rate of fracture propagation.

[0082] S400: The water hazard risk value of the goaf is calculated based on the water pressure change rate, the seepage flow rate change rate, the fracture propagation rate, and the preset water hazard disturbance risk function.

[0083] After obtaining the dynamic output results of the 3D coupled model, the risk value of water hazard disturbance is calculated. The preset water hazard disturbance risk function is pre-generated and stored in the monitoring system.

[0084] S500: Based on the water hazard disturbance risk value and the preset risk level classification threshold, determine the risk level of the water hazard disturbance risk value and call the early warning strategy corresponding to the risk level.

[0085] In practice, the calculated water hazard disturbance risk value can be directly compared with the predefined risk level classification threshold to determine the risk level corresponding to the water hazard disturbance risk value, and the early warning strategy corresponding to the risk level can be directly invoked.

[0086] The above-described scheme in this embodiment collects multi-dimensional physical parameters for the goaf monitoring area, which includes the water storage space and its adjacent area, the surrounding rock failure zone, and the area with fracture development and seepage paths. The physical parameters include dynamic water pressure monitoring data, surrounding rock micro-deformation monitoring data, electrical anomaly monitoring data, and structural fracture monitoring data. A three-dimensional hydrological and mechanical coupled inversion model is constructed, including an unsteady Darcy seepage control model and a fracture strain evolution model, as well as the permeability and strain coupling relationship between the two models. This scheme calculates the water pressure change rate, seepage flow change rate, and fracture propagation rate of the goaf based on the physical parameters using the three-dimensional hydrological and mechanical coupled inversion model. Based on the water pressure change rate, seepage flow change rate, fracture propagation rate, and a preset water hazard disturbance risk function, the water hazard disturbance risk value of the goaf is calculated. Based on the water hazard disturbance risk value and a preset risk level classification threshold, the risk level of the water hazard disturbance risk value is determined, and the corresponding early warning strategy is invoked. This solution enables multi-physics field monitoring, three-dimensional hydrological and mechanical coupled inversion modeling, risk level classification and early warning strategy invocation and control, solving the problems of delayed response, unclear regional control and low prediction accuracy in traditional coal mine water hazard monitoring, fundamentally improving water hazard prediction and early warning capabilities and system response rate.

[0087] Preferably, in step S200, constructing a three-dimensional hydrological and mechanical coupled inversion model of the goaf includes:

[0088] S201: Construct the unsteady Darcy flow control model:

[0089] ;

[0090] Where t represents the time step, and x, y, and z represent the spatial coordinates of the center point of the small volume element. For small volume units, the seepage velocity vector is... The equivalent permeability tensor (m²) is determined by the fracture morphology, aperture, and connectivity. For small volume units, Moisture content per unit volume For the instantaneous water pressure field of a small volume unit, Let Z be the unit weight of water, and Z be the elevation, defined vertically in the model for calculating the gravitational potential energy term. The water head gradient is also included. The water head is obtained by using head difference or weighted least squares fitting between adjacent monitoring points to characterize the rate and direction of water head change in three-dimensional space. In this relationship, the measured water pressure in the goaf is used as the basis. Instantaneous water pressure field as a small volume unit The observations at the boundary or monitoring point are used to reconstruct the continuous water pressure distribution field in a certain three-dimensional space for each small volume unit through interpolation and smoothing.

[0091] S202: Construct the aforementioned crack strain evolution model:

[0092] ;

[0093] in, The crack strain of the small volume unit represents the relative deformation of the crack opening or shear dilation within the unit volume, which is obtained in real time through sensors such as fiber optic strain arrays. is the crack propagation diffusion coefficient, (m² / s), which reflects the ability of crack strain to propagate to neighboring elements; Let m² be the Laplace operator for the strain field, describing the spatial variation characteristics of crack propagation. The fracture source term of the small volume unit reflects the differences in fracture induction mechanisms at different spatial locations; For parameter calibration coefficients, This refers to the frequency of events collected.

[0094] The relationship between permeability and strain is expressed as follows:

[0095] ;

[0096] in, The basic permeability field of the small volume unit can be obtained using lithological and borehole data or estimated by inversion from the measured static resistivity. λ is the sensitivity coefficient for enhancing permeability through fracture propagation, which is generally taken as 0.5–2 and is obtained based on field experience.

[0097] Among them, the seepage velocity vector Water head Instantaneous water pressure field The water pressure dynamic monitoring data is used to determine the measured results from the field sensors; the fracture strain and fracture source term are determined by the surrounding rock micro-deformation monitoring data; the acquisition event frequency is determined by the structural rupture monitoring data; and the foundation permeability field is determined by the electrical anomaly monitoring data. In this embodiment, the partial differential equations in the above model are discretized using a spatiotemporal coupling method, constructing a small-volume element mesh (e.g., 1×1×1 m³) in three-dimensional space. The time intervals of the input physical parameters are discretized at equal intervals, and periodic coupling iterations are performed. During the calculation process, multiple physical parameters are input into the above model: Input is obtained after detection by the water pressure and water level monitoring unit. Input is detected by a distributed fiber optic strain array. Resistivity is detected by the resistivity sensing module The result is then calculated and input. The data is input after detection by a micro-vibration and acoustic emission identification device. The instantaneous water pressure field can be calculated using the above formula. seepage velocity field fracture evolution field and equivalent permeability field , This represents the rate of change of water pressure. The rate of change of permeation flow; For expansion rate.

[0098] More preferably, to avoid reflection interference or dissipation error in the internal solution of the model due to boundary condition distortion, the unsteady Darcy flow control model and the fracture strain evolution model are configured with the following boundary conditions:

[0099] Seepage field boundary conditions: Apply no-flow boundary conditions (Neumann conditions) at the outer boundary of the water storage space. This indicates that water does not seep through the boundary, and n is the outer normal vector of the boundary, that is, the unit direction vector perpendicular to the boundary surface in three-dimensional space; at the boundary of the surrounding rock failure zone, a constant water head or constant water pressure boundary condition (Dirichlet condition) is applied. , This refers to the real-time water pressure in the goaf area.

[0100] Structural disturbance field boundary conditions: Apply fixed displacement boundary conditions at the boundary of the stable region of the surrounding rock. This indicates that the cracks within the stable zone no longer propagate or deform; at the boundary of the surrounding rock failure zone, a free tensile crack propagation condition is applied. This indicates that the crack can extend freely along the stress gradient direction without external constraints.

[0101] By setting the above boundaries, the unsteady Darcy seepage control model and the fracture strain evolution model can realize the coupled response calculation of seepage-fracture field in the goaf, surrounding rock and fault intersection area, providing a physical boundary condition basis for subsequent risk identification.

[0102] Furthermore, to improve the accuracy and practicality of the results of the unsteady Darcy flow control model and the fracture strain evolution model, dynamic embedding of field measured data is introduced. Specifically, the construction of the unsteady Darcy flow control model in S201 and the construction of the fracture strain evolution model in S202 further include:

[0103] (1) Boundary value update: If the dynamic water pressure monitoring data indicates that the water pressure fluctuation exceeds the static threshold, the real-time collected dynamic water pressure monitoring data is loaded into the boundary input of the unsteady Darcy seepage control model according to the time step, for the purpose of updating the boundary value. Dynamic updates are performed. Specifically, water pressure is collected in real time using sensors deployed at the boundaries of the water storage space and the boundaries of the surrounding rock failure zone. With water level The monitoring system is configured with static thresholds. When the water pressure at the monitoring point When the change exceeds the static threshold, the system automatically triggers a boundary reload command to read real-time time series data for that region. The boundary conditions are then input into the unsteady Darcy flow control model at time steps, and the boundary conditions are updated before each solution iteration to achieve dynamic time updating of the boundary conditions of the unsteady Darcy flow control model.

[0104] (2) Crack activation condition correction: If the structural fracture monitoring data represents the microseismic frequency If the set threshold is exceeded, the strain evolution model of the crack will be... Switch from a conventional linear growth mode to an exponential amplification mode; adjust the sensitivity coefficient of the penetration rate-strain coupling relationship. This is to characterize the increased water conductivity of fractures. Specifically, event frequencies are collected in real time using deployed microseismic and acoustic emission identification devices. With energy The monitoring system has a preset micro-vibration trigger threshold. When the frequency of events collected at any monitoring point Multiple consecutive sampling periods exceeding the preset micro-vibration trigger threshold At that time, the monitoring system automatically determines that the area has entered the fracture activation stage and automatically performs the following operations: adjusting the fracture source term parameters... The process switches from a conventional linear growth model to an exponential growth model, correspondingly increasing the fracture growth rate; then the permeability update coefficient is adjusted. The simulation simulates a sudden increase in the water conductivity of fractures; relevant areas are marked as high-risk units for subsequent risk indicator calculation and early warning.

[0105] (3) Fault structure embedding: If the monitoring data of microdeformation of the surrounding rock indicates the existence of fracture structures in the goaf, then it is the fracture source term in the fracture strain evolution model. Give a higher initial value or weaken the crack propagation diffusion coefficient This is to simulate the preferred propagation path of the fracture.

[0106] Specifically, the above three embedding mechanisms (1)-(3) are uniformly managed by the system: physical parameters are transmitted to the monitoring system in real time; the monitoring system judges whether to trigger boundary value update, fracture activation condition correction or fault structure embedding operation according to the above-set logic; physical parameters and boundary conditions are automatically updated at each time step, forming a monitoring, calculation and correction operation mechanism.

[0107] In the above scheme, preferably, in step S400, the water hazard disturbance risk value of the goaf is calculated based on the water pressure change rate, the seepage flow change rate, the fracture propagation rate, and a preset water hazard disturbance risk function. The preset water hazard disturbance risk function is implemented through the following function:

[0108] ;

[0109] As mentioned above, The water pressure change rate is... The rate of change of the permeation flow is... The crack propagation rate, , , These are the corresponding risk weight coefficients, which are set based on experience with lithology, burial depth, and microseismic response sensitivity.

[0110] The preset water hazard disturbance risk function is normalized:

[0111] ;

[0112] This represents the average of the calculated water hazard disturbance risk functions for all small volumetric units at the same time step t. The standard deviation of the calculated results of the water hazard disturbance risk function for all small volume units at the same time step t is used; the calculated results of the preset water hazard disturbance risk function after normalization are used as the water hazard disturbance risk value. Normalization can avoid the imbalance of risk degree caused by different dimensions.

[0113] Due to the nonlinear superposition nature of the disturbance process, The statistical distribution of the value approximates a normal distribution, therefore the standard quantile method can be used for risk level classification. In step S500, the risk level of the water hazard disturbance risk value is determined based on the water hazard disturbance risk value and a preset risk level classification threshold, and the corresponding early warning strategy is invoked. Figure 2 As shown, it includes:

[0114] S501: Flood Warning Classification:

[0115] When the flood disturbance risk value is greater than the first threshold, it is classified as Level 1 risk, which is a high-intensity disturbance zone and requires immediate coordinated response. In practice, the first threshold can be selected as 2.5. When the second threshold is less than the flood disturbance risk value and less than or equal to the first threshold, it is classified as Level 2 risk, which is a strong disturbance warning zone. In practice, the second threshold can be selected as 1.5. When the third threshold is less than the flood disturbance risk value and less than or equal to the second threshold, it is classified as Level 3 risk, which is a medium-intensity disturbance warning zone. In practice, the third threshold can be selected as 0.5. When the flood disturbance risk value is less than or equal to the third threshold, it is classified as Level 4 risk, which corresponds to a safe and stable zone. The selection of thresholds in this scheme is based on the standard deviation stratification theory in statistical analysis. Under the assumption that the normalized index follows an approximately normal distribution, disturbances exceeding 2.5 are in the upper 0.6% of the sample distribution, representing extreme abnormal disturbances; data in the 1.5-2.5 range account for about 5% of the distribution, corresponding to a risk area that deviates significantly from the stable state but has not reached the level of sudden outbreak; 0.5 to 1.5 is a slight fluctuation range; and the area below 0.5 is the background noise area of ​​disturbances.

[0116] Further preferably, to improve the accuracy of early warning and suppress occasional false alarms, a spatial connectivity constraint mechanism is introduced. The method for obtaining the risk level of the water hazard disturbance risk value in the above scheme also includes:

[0117] S502: Effective Early Warning Aggregate Classification

[0118] If continuous The small volume units are spatially adjacent and all belong to If the risk level is specified, then the aggregation area is calculated: , For the first The volume of a small volume unit, when At that time, it is determined that the aggregation region has formed. An effective early warning aggregation system for risk levels; for The minimum effective volume threshold corresponding to the risk level; where, risk level It can be numbered one, two, three, or four.

[0119] S503: Store in dynamic risk index table

[0120] Write the effective early warning aggregate into the dynamic risk index table: , Indicates the i-th time Identified as An effective early warning aggregation system for risk levels. for Index table of risk areas at risk level 1 for The threshold corresponding to the risk level is: To effectively predict the volume of polymers, The minimum effective warning aggregate volume determination threshold;

[0121] S504: Generate a dynamic risk map in real time based on the dynamic risk index table;

[0122] S505: Invoke the early warning strategy according to the risk level of each effective early warning aggregate in the dynamic risk map: the early warning strategy for the first-level risk level is to execute a mandatory response; the early warning strategy for the second-level risk level is to implement pre-set early warning control measures; the early warning strategy for the third-level risk level is to adjust the monitoring frequency and expand the monitoring range; and the early warning strategy for the fourth-level risk level is to maintain routine monitoring.

[0123] Furthermore, in the above scheme, after invoking the early warning strategy corresponding to the risk level, step S500 also includes: for the effective early warning aggregates of the first-level and second-level risk levels, if their water hazard disturbance risk values ​​are continuously... The change within a time step exceeds a set threshold. This increases the monitoring intensity of the effective early warning aggregation. That is, it determines whether the conditions are met. In the formula, This represents the risk index at the current time step. This is the weighted average disturbance index for the same region at the previous time step. The threshold for disturbance change. This represents the number of consecutive determinations. When the above criteria are met, the area is marked as a monitoring enhancement unit. This automatically triggers an increase in the sampling frequency for that area, while simultaneously switching from single-parameter acquisition mode to multi-channel collaborative mode. The final risk level area is determined through... The index table is output in a structured format.

[0124] In the above scheme, step S504, which involves invoking the early warning strategy based on the risk level of each effective early warning aggregate in the dynamic risk map, includes:

[0125] S5041: Based on the i-th valid early warning aggregate Risk level generation trigger signal: , for The response priority corresponding to the risk level is used for the resource scheduling and sorting mechanism in the response queue, and the mapping rules are as follows:

[0126] ;

[0127] S5042: Based on the aforementioned response priority, construct a response execution criterion: If Then the i-th valid early warning aggregate will be The trigger signal is added to the response scheduling queue; This is the minimum response priority threshold. For example, setting it to 1 means that only areas with a risk level of 3 or higher will be subject to a response.

[0128] S5043: According to each of the aforementioned trigger signals , call The response strategy function F corresponding to the risk level k Output control parameter set: , This is a set of response instructions corresponding to the risk level, including mandatory response, early warning and control measures, adjusting monitoring frequency and expanding monitoring scope, and maintaining routine monitoring. Specifically, based on the risk level... Call the preset response strategy function F k Output a set of response control parameters . The response instruction set corresponds to the risk level and includes four types: mandatory response, early warning and control measures, adjustment of monitoring frequency and expansion of monitoring scope, and maintenance of routine monitoring. According to the risk level, Level 1 risk areas (which may be indicated in red) require mandatory response, including activation of the emergency drainage system, area closure and work suspension, and activation of the grouting and sealing system; Level 2 risk areas (which may be indicated in orange) implement early warning and control measures, including initiating localized limited sampling, real-time monitoring with doubled sampling frequency, and preparation for remote personnel evacuation; Level 3 risk areas (which may be indicated in yellow) adjust the monitoring frequency and expand the monitoring scope; and Level 4 risk areas (which may be indicated in green) maintain routine monitoring.

[0129] The above plan, after implementing early warning strategies for risk areas at all levels, also includes:

[0130] Within the set window period Δ T Internal to the original risk area Monitor the disturbance variables, recalculate the risk index, and conduct a new round of risk assessment. The recalculated risk index can be expressed as follows:

[0131] ;

[0132] To ensure that the modeling always closely matches the on-site evolution process, the system uses the measured disturbance data after the response as the initialization parameters for the unsteady Darcy flow control model and the fracture strain evolution model in the next round of calculation, and so on in a loop.

[0133] In the above scheme, in step S100, the physical parameters sent by the sensors in the goaf monitoring area for reflecting the evolution of water hazards are all normalized to be dimensionless parameters.

[0134] To achieve long-term online operation of the goaf water hazard prevention system and dynamic self-calibration of the risk identification model, it is necessary to uniformly manage and standardize the data streams from various sensors and the inversion of the 3D hydrological and mechanical coupled inversion model. Different types of data have significantly different dimensions (Pa, m, m / s, S / m, etc.). Normalization can prevent imbalances in model solution weights and reduced feature correlations, ensuring the computational stability of multidimensional coupled analysis. The normalization process proposed in this step converts the relevant data into a dimensionless parameter set. This aims to achieve cross-scale fusion, dynamic correction, and online optimization of multi-source information. Specifically, it involves converting various types of sensor data (including water pressure, water level, fiber optic strain, resistivity, microseismic frequency, and energy) into a standardized format to form a comparable input matrix. The specific steps include:

[0135] S101: Convert all physical parameters into a standardized small-volume unit space-time indexed structure, using the following pattern:

[0136] ;

[0137] in, For the first i Classified parameter index, representing the spatial unit With time step Standardized indices of physical parameters obtained from the above, including hydrological indices (such as water pressure). seepage velocity Head gradient Rock mechanical properties (such as fracture strain) diffusion coefficient ), media parameters (such as porosity) Equivalent penetration rate ) and disturbance risk indicators wait.

[0138] S102: All standardized indicators are normalized to zero mean, ensuring cross-dimensional comparability. The data standardization function is in the form of:

[0139] ;

[0140] This represents the mean of different standardized indicators;

[0141] S103: The system automatically records the input dataset during each calculation cycle. Control strategy function and response results This constitutes the training sample library S. T :

[0142] ;

[0143] in, Indicates the first The dataset of physical parameters input over a time period, including water pressure Head gradient Crack strain resistivity and disturbance risk indicators wait; The control strategy function executed by the system during this cycle includes risk classification response parameters, drainage and grouting control commands, monitoring frequency adjustment strategies, etc. The system feedback results after the strategy is implemented are used to quantify the strategy's effectiveness, mainly including the rate of change in risk level. Water pressure recovery rate and crack propagation attenuation rate Indicators such as...

[0144] The physical parameter dataset Composed of multiple types of physical parameters, forming a multi-dimensional input vector structure; the corresponding system feedback results It includes multiple feature indicators that reflect the response effect, thus forming a dynamic mapping relationship of multiple inputs and multiple outputs, which is used for model self-learning and policy optimization.

[0145] Training sample library Using time periods as an index, a rolling window approach is employed for dynamic updates, meaning new data is automatically written and old data is replaced chronologically. When significant changes in environmental conditions, seepage status, or risk levels are detected, the sample database automatically triggers weight updates and strategy correction procedures to adaptively adjust key parameters (such as risk assessment thresholds and response priorities).

[0146] The above approach continuously accumulates the "input-strategy-result" correspondence, enabling bidirectional updates of data and strategies, and providing data support for subsequent online optimization and intelligent evolution of the model. The sample library in this approach not only possesses real-time and continuous characteristics but also traceability and self-learning features, providing a long-term and stable data foundation for flood risk prediction and dynamic control.

[0147] As needed, the above technical solutions can be combined to achieve the best technical effect.

[0148] The above are merely the principles and preferred embodiments of this application. It should be noted that, for those skilled in the art, several other modifications can be made based on the principles of this application, and these modifications should also be considered within the scope of protection of this application.

Claims

1. A method for preventing and monitoring water disasters in a coal mine goaf, characterized in that, The method comprises: acquiring physical parameters of a goaf monitoring area for a water disaster evolution process; wherein the goaf monitoring area comprises a water storage space and its adjacent boundary area, a surrounding rock damage zone area, a fracture development and seepage path area; the physical parameters comprise water pressure dynamic monitoring data, surrounding rock micro-deformation monitoring data, electrical anomaly monitoring data and structural fracture monitoring data; constructing a three-dimensional hydro-mechanical coupling inversion model of the goaf, the three-dimensional hydro-mechanical coupling inversion model comprising a non-steady-state Darcy seepage control model for inverting a pore water pressure field and a seepage velocity field and a fracture strain evolution model for describing a tensile and shear fracture process induced by structural disturbance, the non-steady-state Darcy seepage control model and the fracture strain evolution model realizing physical field linkage feedback through a permeability and strain coupling relationship; inputting the physical parameters into the three-dimensional hydro-mechanical coupling inversion model to obtain a water pressure change rate, a seepage flux change rate and a fracture expansion rate of the goaf output by the three-dimensional hydro-mechanical coupling inversion model; calculating a water disaster disturbance risk value of the goaf according to the water pressure change rate, the seepage flux change rate, the fracture expansion rate and a preset water disaster disturbance risk function; determining a risk level of the water disaster disturbance risk value according to the water disaster disturbance risk value and a preset risk level division threshold value, and calling a corresponding early warning strategy of the risk level.

2. The coal mine goaf water disaster prevention and monitoring and early warning method according to claim 1, characterized in that, The method of constructing a three-dimensional hydro-mechanical coupling inversion model of the goaf, the three-dimensional hydro-mechanical coupling inversion model comprising a non-steady-state Darcy seepage control model for inverting a pore water pressure field and a seepage velocity field and a fracture strain evolution model for describing a tensile and shear fracture process induced by structural disturbance, the non-steady-state Darcy seepage control model and the fracture strain evolution model realizing physical field linkage feedback through a permeability and strain coupling relationship, comprises: constructing the non-steady-state Darcy seepage control model: ； wherein, denotes the time step, denotes the spatial coordinates of the center point of the small volume element, is the seepage velocity vector of the small volume element, is the equivalent permeability tensor, is the water head of the small volume element, is the water content per unit volume, is the instantaneous water pressure field of the small volume element, is the specific weight of water, is the elevation; constructing the fracture strain evolution model: ； wherein, is the crack strain of the small volume element, is the crack propagation diffusion coefficient, is the Laplacian of the strain field, is the crack source term of the small volume element, is a parameter calibration coefficient, is the acquisition event frequency; the permeability and strain coupling relationship is expressed as: ； wherein, is the base permeability field of the small volume element, is the sensitivity coefficient of the fracture propagation enhancing permeability; wherein a seepage velocity vector, a water head and an instantaneous water pressure field are determined by the water pressure dynamic monitoring data; a fracture strain and a fracture source term are determined by the surrounding rock micro-deformation monitoring data; an acquisition event frequency is determined by the structural fracture monitoring data; and a basic permeability field is determined by the electrical anomaly monitoring data.

3. The coal mine goaf water disaster prevention and monitoring and early warning method according to claim 2, characterized in that, The non-steady-state Darcy seepage control model and the fracture strain evolution model are configured with the following boundary conditions: Seepage field boundary condition: no-flow boundary condition is applied at the outer boundary of the water storage space , is the outer normal vector of the boundary; a constant water head or water pressure boundary condition is applied at the boundary of the surrounding rock failure zone area , is the real-time water pressure of the goaf Structural disturbance field boundary condition: fixed displacement boundary condition is applied at the boundary of the stable zone of the surrounding rock ; free tension crack propagation condition is applied at the boundary of the failure zone of the surrounding rock .

4. The coal mine goaf water disaster prevention and monitoring and early warning method according to claim 3, characterized in that, In the method of constructing the non-steady-state Darcy seepage control model and the fracture strain evolution model, further comprising: If the water pressure dynamic monitoring data represents that the water pressure fluctuation exceeds the static threshold value, the real-time collected water pressure dynamic monitoring data is loaded to the boundary input of the unsteady Darcy seepage control model according to time steps, for dynamic updating ; if the structural breakage monitoring data indicates microseismic frequencies exceeding a set threshold, switching the fracture strain evolution model from a conventional linear growth pattern to an exponential amplification pattern; adjusting a sensitivity coefficient in the permeability and strain coupling relationship to characterize an increase in fracture hydraulic conductivity; If the micro-deformation monitoring data of the surrounding rock indicates that there is a fracture structure in the goaf, a fracture source term in the fracture strain evolution model Assign a higher initial value or weaken the fracture propagation diffusion coefficient .

5. The coal mine goaf water disaster prevention and monitoring and early warning method according to claim 1, characterized in that, In the method of calculating a water disaster disturbance risk value of the goaf according to the water pressure change rate, the seepage flux change rate, the fracture expansion rate and a preset water disaster disturbance risk function, further comprising: Pre-set water damage disturbance risk function By the following function: ； is the water pressure change rate, is the seepage flux change rate, is the crack propagation rate, , , are the corresponding risk weight coefficients, respectively. normalizing the preset water disaster disturbance risk function: ； for the same time step t the mean of the results of the calculation of the water hazard disturbance risk function for all the small volume units, for the same time step t the standard deviation of the results of the calculation of the water hazard disturbance risk function for all the small volume units; taking the calculation result of the normalized preset water disaster disturbance risk function as the water disaster disturbance risk value.

6. The coal mine goaf water disaster prevention and monitoring and early warning method according to claim 5, characterized in that, In the method of determining a risk level of the water disaster disturbance risk value according to the water disaster disturbance risk value and a preset risk level division threshold value, and calling a corresponding early warning strategy of the risk level, the risk level of the water disaster disturbance risk value is determined by the following method: When the water hazard disturbance risk value is greater than the first threshold value, the risk level is level one; When the second threshold value is less than the water hazard disturbance risk value and the water hazard disturbance risk value is less than or equal to the first threshold value, the risk level is level two; When the third threshold value is less than the water hazard disturbance risk value and the water hazard disturbance risk value is less than or equal to the second threshold value, the risk level is level three; When the water hazard disturbance risk value is less than or equal to the third threshold value, the risk level is level four, corresponding to a safe and stable region.

7. The coal mine goaf water disaster prevention and monitoring and early warning method according to claim 6, characterized in that, The method for obtaining the risk level of the water hazard disturbance risk value further comprises: If consecutive small volume units are spatially adjacent and belong to the same risk level, calculate the aggregation area: , the volume of the small volume unit, when , determine that the aggregation area forms an effective early warning aggregation of the risk level; is the minimum effective volume threshold corresponding to the risk level of ; the above is one, two, three or four; writing the effective early warning aggregate into a dynamic risk index table: , representing the th effective early warning aggregate identified as a risk level of , , a risk region set index table of a risk level of , a threshold value corresponding to a risk level of , a volume of the effective early warning aggregate, a volume determination threshold value of the lowest effective early warning aggregate; generating a dynamic risk map in real time according to the dynamic risk index table; calling an early warning strategy according to the risk level of each effective early warning aggregate in the dynamic risk map; the early warning strategy for the level one risk level is to execute a forced response; the early warning strategy for the level two risk level is to implement a pre-set early warning control measure; the early warning strategy for the level three risk level is to adjust the monitoring frequency and expand the monitoring range; the early warning strategy for the level four risk level is to maintain regular monitoring.

8. The coal mine goaf water disaster prevention and monitoring and early warning method according to claim 7, characterized in that, After the early warning strategy corresponding to the risk level is called, the method further comprises: For the effective early warning aggregation body of the first risk level and the second risk level, if the change range of the water disaster disturbance risk value of the effective early warning aggregation body exceeds the set threshold value in continuous time steps, the monitoring strength of the effective early warning aggregation body is increased. ​ 9. The coal mine goaf water disaster prevention and monitoring and early warning method according to claim 7, characterized in that, the calling of the early warning strategy according to the risk level of each effective early warning aggregate in the dynamic risk map comprises: According to the first effective early warning aggregate risk level generation trigger signal: , For the response priority corresponding to the risk level, the resource scheduling sorting mechanism in the response queue is as follows: ； Based on the response priority, a response execution criterion is constructed: if the triggering signal of the first effective early warning aggregate is added to the response scheduling queue; is the minimum response priority threshold set.​ According to each of the trigger signals , call level risk response strategy function corresponding to the policy , output control parameter set: , corresponding to the response instruction set for risk level, including mandatory response, early warning control measures, adjustment of monitoring frequency and expansion of monitoring range and range adjustment, and maintenance of regular monitoring.

10. The coal mine goaf water disaster prevention and monitoring and early warning method according to any one of claims 1-9, characterized in that, in the acquisition of the physical parameters of the goaf monitoring area for reflecting the water hazard evolution process, all the physical parameters are normalized to be dimensionless parameters.

Images