Mine water disaster early warning method based on geoelectric field parameter change

Abstract

The application relates to the technical field of mine water disaster monitoring, in particular to a mine water disaster early warning method based on geoelectric field parameter change; the method comprises the following steps: acquiring multi-dimensional geoelectric field parameters in real time and synchronously, performing time sequence feature extraction and fusion calculation on the multi-dimensional geoelectric field parameters, and obtaining a global early warning index; generating a primary early warning signal based on the global early warning index; constructing a three-dimensional geological model based on existing geological data, and generating a high-risk area according to an iterative growth rule; determining the early warning level of the primary early warning signal based on the high-risk area and early warning thresholds at all levels, and performing corresponding early warning response; the application realizes efficient early warning and hazard source positioning of mine water disasters, reduces the false alarm rate and the missed alarm rate of water disasters, and improves the reliability of early warning and the pertinence of emergency response.

Description

Technical Field

[0001] This application relates to the field of mine water hazard monitoring technology, specifically to a mine water hazard early warning method based on changes in geoelectric field parameters. Background Technology

[0002] As coal resource development continues to extend deeper and mining areas expand into more complex geological regions, mine hydrogeological conditions are becoming increasingly complex, leading to a rise in the probability and potential severity of water-related accidents. Therefore, constructing an efficient, accurate, and real-time mine water hazard early warning system has become a core requirement for ensuring safe coal mine production and a key direction for technological breakthroughs in the industry.

[0003] Traditional mine water hazard early warning methods mainly rely on manual hydrological observation and drilling exploration, which have inherent defects such as information lag, low spatial resolution, and difficulty in capturing early signs of water inrush.

[0004] In recent years, the geoelectric field method has been introduced into the field of mine water hazard early warning due to its sensitivity to groundwater and ability to achieve large-scale real-time monitoring. However, existing technical solutions have certain limitations: on the one hand, they rely on a single parameter for threshold judgment, which is easily affected by the complex electromagnetic environment and equipment noise underground, resulting in high false alarm and false negative rates, affecting the reliability of early warning; on the other hand, they cannot accurately locate water hazard sources spatially, resulting in water prevention and control measures that lack specificity, are largely haphazard, and have a low level of intelligence.

[0005] For example, Chinese patent CN108104876B discloses a real-time graded early warning method and system for water hazards based on mine electrical resistivity monitoring. This method uses three-dimensional resistivity inversion to automatically and in real-time process the monitoring data, and automatically grades and judges the risk of water inrush at the working face based on the real-time changes in the resistivity of the coal-bearing strata obtained from the inversion, thus realizing the automation and intelligence of water hazard early warning at the working face.

[0006] The existing technologies mentioned above all suffer from the problems proposed in this background: relying on a single parameter for risk assessment leads to high false alarm and false alarm rates; and failing to accurately locate water hazard sources in three-dimensional space results in water hazard prevention measures lacking specificity and being largely haphazard.

[0007] The information disclosed in this background section is intended only to enhance the understanding of the overall background of this application and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0008] The purpose of this application is to provide a mine water hazard early warning method based on changes in geoelectric field parameters. By real-time monitoring of multiple parameters of the geoelectric field, the method can achieve accurate early warning and hazard source location of mine water hazards, thereby improving the reliability and accuracy of mine water hazard early warning.

[0009] To achieve the above objectives, this application provides the following technical solution:

[0010] In a first aspect, this application provides a mine water hazard early warning method based on changes in geoelectric field parameters, including:

[0011] Multidimensional geoelectric field parameters are acquired in real time; time-series features are extracted from the multidimensional geoelectric field parameters to obtain feature indices; a global early warning index is calculated based on the feature indices, and a primary early warning signal is generated based on the global early warning index.

[0012] Collect existing geological data from the mining company and construct a three-dimensional geological model based on the data;

[0013] In response to the primary warning signal, a high-risk area is generated based on the characteristic indicators and the three-dimensional geological model, according to a preset iterative growth rule.

[0014] Calculate the spatial volume, expansion rate, and internal early warning index of the high-risk area; determine the early warning level of the primary early warning signal based on preset early warning thresholds at each level, and perform corresponding early warning responses.

[0015] As a preferred embodiment of the mine water hazard early warning method based on changes in geoelectric field parameters described in this application, the specific steps for generating high-risk areas are as follows:

[0016] Traverse the characteristic indicators of each monitoring electrode point in the monitoring area and mark the monitoring electrode points whose natural potential change index and resistivity change index are both greater than the preset risk threshold as high-risk seed points.

[0017] Initialize a high-risk region with each high-risk seed point as the core;

[0018] Based on the preset iterative growth rules, iterative growth is performed on any high-risk area;

[0019] If all non-uniform cell grids surrounding any high-risk region no longer satisfy the iterative growth rule, then stop the iterative growth and obtain the corresponding high-risk region.

[0020] As a preferred embodiment of the mine water hazard early warning method based on changes in geoelectric field parameters described in this application, the iterative growth rule includes data similarity rules, spatial continuity rules, and geological guidance rules.

[0021] A non-uniform cell grid that simultaneously satisfies the data similarity rule, spatial continuity rule, and geological orientation rule is added to the high-risk area, and the regional extent and average feature matrix of the high-risk area are updated.

[0022] The spatial continuity rule is as follows: if any non-uniform cell grid outside any high-risk area is coplanar, shares an edge, or shares an angle with the corresponding high-risk area in three-dimensional space, then the non-uniform cell grid is determined to satisfy the spatial continuity rule.

[0023] As a preferred embodiment of the mine water hazard early warning method based on changes in geoelectric field parameters described in this application, the data similarity rules are as follows:

[0024] Calculate the arithmetic mean of the same characteristic index of all monitoring electrode points within each non-uniform cell grid; construct the characteristic matrix of the non-uniform cell grid based on the arithmetic mean;

[0025] Calculate the average feature matrix of all monitoring electrode points within each high-risk area; use the average feature matrix as the feature matrix of the corresponding high-risk area.

[0026] Calculate the Euclidean distance between the feature matrices of all non-uniform cell grids surrounding the high-risk area and the feature matrix of the corresponding high-risk area; if the Euclidean distance is less than a preset distance threshold, then the non-uniform cell grids are determined to satisfy the data similarity rule.

[0027] As a preferred embodiment of the mine water hazard early warning method based on changes in geoelectric field parameters described in this application, the geological guidance rules are specifically as follows:

[0028] Based on a three-dimensional geological model, the orientation of any non-uniform cell grid outside a high-risk area relative to the corresponding high-risk area is obtained.

[0029] If the non-uniform cell grid is located in the high-risk area along the direction of the stratum dip, then the non-uniform cell grid is determined to satisfy the geological guidance rule, and the distance threshold in the data similarity rule is multiplied by the preset first relaxation coefficient to obtain the distance threshold corresponding to the non-uniform cell grid.

[0030] If the non-uniform cell grid is located in the direction of the fault or fracture zone within the preset area outside the high-risk area, then the non-uniform cell grid is determined to satisfy the geological guidance rule, and the distance threshold in the data similarity rule is multiplied by the preset second relaxation coefficient to obtain the distance threshold corresponding to the non-uniform cell grid.

[0031] If there is an aquitard boundary or intact rock layer in the three-dimensional geological model between the non-uniform cell grid and the high-risk area, then the non-uniform cell grid is determined not to meet the geological guidance rule.

[0032] As a preferred embodiment of the mine water hazard early warning method based on changes in geoelectric field parameters described in this application, the primary early warning signal is generated as follows:

[0033] The characteristic indicators of each monitoring electrode point are extracted, including: spontaneous potential change index, excitation current trend index and resistivity change index;

[0034] Set the weight coefficients for each characteristic index; normalize the natural potential mutation index, excitation current trend index and resistivity change index of each monitoring electrode point respectively, and then perform a weighted summation according to the weight coefficients to obtain the fusion early warning index of each monitoring electrode point.

[0035] Calculate the arithmetic mean of the fusion warning index of all monitoring electrode points in the area to be monitored to obtain the global warning index; set a global warning threshold, and generate a primary warning signal when the global warning index is greater than the global warning threshold.

[0036] As a preferred embodiment of the mine water hazard early warning method based on changes in geoelectric field parameters described in this application, the step of extracting the resistivity change index specifically includes:

[0037] The resistivity of all monitoring electrode points at any sampling time is inverted to obtain the three-dimensional resistivity distribution of the area to be monitored at that sampling time.

[0038] The gradient of the three-dimensional resistivity distribution at sampling time t is calculated to obtain the spatial rate of change of resistivity at the corresponding monitoring electrode point.

[0039] The initial monitoring time is used as the reference time, and the resistivity at the reference time is used as the reference resistivity;

[0040] Calculate the ratio of the resistivity at any sampling time t to the reference resistivity to obtain the resistivity change rate over time at the corresponding monitoring electrode point;

[0041] The resistivity change index is obtained by weighted summing of the spatial rate of change of resistivity and the temporal rate of change of resistivity.

[0042] As a preferred embodiment of the mine water hazard early warning method based on changes in geoelectric field parameters described in this application, the calculation steps for the three-dimensional resistivity distribution state include:

[0043] A three-dimensional spatial coordinate system is established with the area to be monitored as the target; the origin of the three-dimensional spatial coordinate system is any vertex of the area to be monitored, and the directions of the X-axis, Y-axis, and Z-axis are consistent with the directions of the world coordinate system.

[0044] The three-dimensional space formed by the area to be monitored in the three-dimensional spatial coordinate system is divided into non-uniform cell grids, and the resistivity of each non-uniform cell grid is set as the unknown parameter to be solved.

[0045] A forward modeling algorithm is constructed using the unknown parameters as independent variables to calculate the resistivity of each non-uniform cell grid, and the Poisson equation for the steady current field is solved using the finite element method or the finite difference method.

[0046] Define the objective function of the inversion algorithm and solve the objective function using a linearized iterative algorithm to obtain the three-dimensional resistivity distribution of the area to be monitored.

[0047] As a preferred embodiment of the mine water hazard early warning method based on changes in geoelectric field parameters described in this application, the early warning levels are divided into yellow, orange, and red warnings from low to high; the determination method for the warning level and early warning response of the primary early warning signal is as follows:

[0048] Set warning thresholds at various levels; the warning thresholds at various levels include spatial volume critical value, expansion rate critical value, and internal warning index threshold.

[0049] For a primary warning signal, if there is no high-risk area within the monitoring area, the primary warning signal will be determined as a yellow warning, and attention will be paid to the global warning index.

[0050] If there is at least one high-risk area within the monitoring area, then each high-risk area is traversed, and a warning judgment is made according to the preset warning judgment rules.

[0051] As a preferred embodiment of the mine water hazard early warning method based on changes in geoelectric field parameters described in this application, the early warning judgment rule is as follows:

[0052] If the spatial volume of the high-risk area is greater than the spatial volume threshold, or the expansion rate of the high-risk area is greater than the expansion rate threshold, or the internal early warning index of the high-risk area is greater than the internal early warning index threshold, then the corresponding high-risk area will be marked as an emergency.

[0053] If there are no high-risk areas marked as an emergency within the monitored area, the primary warning signal will be determined as an orange warning, and the high-risk areas not marked as an emergency will be marked in the three-dimensional geological model.

[0054] If at least one high-risk area is marked as an emergency, the primary warning signal will be determined as a red warning and an emergency audible and visual alarm will be triggered; the high-risk areas marked as emergency and the high-risk areas not marked as emergency will be marked with different symbols in the three-dimensional geological model.

[0055] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0056] By acquiring multi-dimensional geoelectric field parameters such as natural potential, excitation current, and resistivity in real time, and performing time-series feature extraction and weighted fusion calculation, a global early warning index is generated. This reduces false alarms and missed alarms caused by the susceptibility of single parameters to interference, and improves the sensitivity and reliability of early warning. By constructing a three-dimensional geological model and dynamically generating high-risk areas based on data similarity, spatial continuity, and geological guidance rules, the precise location of water hazard sources in three-dimensional space is achieved. By calculating the spatial volume, expansion rate, and internal early warning index of high-risk areas, and performing graded early warning and response based on preset early warning thresholds at each level, dynamic assessment of water hazard risk and targeted emergency response are achieved, improving the practicality and emergency efficiency of water hazard early warning. Attached Figure Description

[0057] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0058] Figure 1 A flowchart of the mine water hazard early warning method based on changes in geoelectric field parameters provided in this application;

[0059] Figure 2 A flowchart for calculating the resistivity change index provided in this application;

[0060] Figure 3 A flowchart illustrating the calculation of the three-dimensional resistivity distribution state for the mine water hazard early warning method based on changes in geoelectric field parameters provided in this application;

[0061] Figure 4 A flowchart illustrating the generation of high-risk areas for the mine water hazard early warning method based on changes in geoelectric field parameters provided in this application. Detailed Implementation

[0062] The technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments and specific features in the embodiments are detailed descriptions of the technical solution of this application, rather than limitations thereof. In the absence of conflict, the embodiments and technical features in the embodiments can be combined with each other.

[0063] like Figure 1 As shown in the figure, this embodiment introduces a mine water hazard early warning method based on changes in geoelectric field parameters, including:

[0064] Multidimensional ground electric field parameters are acquired in real time and preprocessed; the multidimensional ground electric field parameters include natural potential, excitation current and resistivity.

[0065] In this embodiment, multiple electrodes are installed in the underground monitoring area according to a preset layout scheme to construct a three-dimensional geoelectric field monitoring system covering the monitoring area; the monitoring area is set as a high-frequency area where water hazards occur, including but not limited to the mining face, roadway roof, and floor; the layout scheme includes: arranging the monitoring electrodes in the roadway roof, floor, or monitoring boreholes, and arranging the common measuring electrode and the infinity power supply electrode in an area with good grounding coupling and no stray current interference;

[0066] Optionally, a mine-use network parallel electrical resistivity meter is used to control all channels to synchronously acquire the instantaneous values ​​of the natural potential, excitation current of the power supply circuit, and power supply potential of each monitoring electrode point with preset monitoring parameters, thereby obtaining raw geoelectric field data, and calculating resistivity based on the excitation current and power supply potential; the multi-channel parallel acquisition capability of the mine-use network parallel electrical resistivity meter can ensure that multi-dimensional geoelectric field parameters are acquired simultaneously at different monitoring electrode points; the natural potential refers to the electric field potential generated by the natural polarization effect of the underground medium in the mine under the condition that no excitation current is applied by the measuring device, and its physical essence is the external manifestation of charge migration and accumulation in the underground medium of the mine;

[0067] The resistivity can be used to reflect the average conductivity of the underground medium in the mine. Its value is significantly positively correlated with the water content of the medium. Therefore, the resistivity can directly reflect the change in the water content of the underground medium in the mine.

[0068] The original geoelectric field data is preprocessed to obtain time-varying natural potential sequences, excitation current sequences, and resistivity sequences.

[0069] Specifically, the preprocessing includes: using zero-phase-shift band-stop filtering to eliminate power frequency interference and harmonic interference in the original geoelectric field data and resistivity; zero-phase-shift band-stop filtering technology can filter out power frequency interference and harmonic interference while avoiding signal phase distortion caused by traditional filtering, so as to ensure the authenticity of the geoelectric field parameter variation trend; normalizing the filtered multidimensional geoelectric field parameters to suppress baseline drift and ensure data stability.

[0070] Feature indicators are obtained by extracting time-series features from the multidimensional geoelectric field parameters. The normalized feature indicators are then weighted and summed to obtain a global early warning index. These feature indicators include the spontaneous potential mutation index, the excitation current trend index, and the resistivity change index. These feature indicators reflect the precursors of water hazards from different dimensions, avoiding the limitations of single indicators. By extracting and weighting the time-series features of the multidimensional geoelectric field parameters, comprehensive information on precursors of water hazards can be captured, improving early warning sensitivity and reducing false alarm rates. For example, the spontaneous potential mutation index can reflect seepage activity and identify dominant water flow channels; the excitation current trend index can reflect changes in the overall water content of the medium; and the resistivity change index can quantify the evolution trend of the water-bearing area.

[0071] Specifically, spontaneous potential is extremely sensitive to groundwater seepage. When new fissures form in a rock mass or water seeps in, the electrochemical field of the rock mass changes, leading to rapid and significant abrupt changes in spontaneous potential. The spontaneous potential abrupt change index is a quantitative characterization of the intensity and rate of change in spontaneous potential caused by groundwater seepage. The extraction method of the spontaneous potential abrupt change index is as follows:

[0072] Calculate the average natural potential of any monitoring electrode point during a preset time period, such as the power supply intermittent period in the past day, and use the average value as the background value of the corresponding monitoring electrode point.

[0073] The continuous sampling time is divided into different cumulative time periods with a preset time length; the preset time length is determined according to the monitoring requirements, for example, every 12 hours is a cumulative time period; the average value of all natural potential sampling values ​​in each cumulative time period is calculated, and the average value is used as the cumulative representative value;

[0074] Calculate the cumulative time period for each monitoring electrode point. The difference between the cumulative representative value and the corresponding background value is used, and the absolute value of the difference is taken as the monitoring electrode point in the cumulative time period. The cumulative change in natural potential; where i is a positive integer, representing the i-th cumulative time period; the cumulative change in natural potential reflects the strength of the sudden change in natural potential;

[0075] When i is greater than 1, calculate the cumulative time period for any monitoring electrode point. The cumulative representative value and the corresponding monitoring electrode point in the previous cumulative time period The difference between the cumulative representative values ​​is used, and the ratio of the absolute value of the difference to the length of the two cumulative time periods is taken as the value of the corresponding monitoring electrode point in the cumulative time period. The cumulative rate of change of natural potential; the cumulative rate of change of natural potential reflects the average rate of change of natural potential over a continuous cumulative time period;

[0076] For any cumulative time period The cumulative representative values ​​of each monitoring electrode point are mapped to the corresponding three-dimensional spatial coordinates of the monitoring electrode point, and a continuous three-dimensional natural potential distribution model is generated by the Kriging interpolation algorithm. The Kriging interpolation algorithm is an interpolation method based on regional variable theory, which can fully consider the spatial correlation of data, so that the generated natural potential distribution is continuous and smooth, avoiding abrupt artifacts caused by traditional interpolation algorithms.

[0077] Gradient calculation is performed on the three-dimensional natural potential distribution model to obtain the natural potential spatial gradient; the direction of the gradient points to the direction of the fastest increase in natural potential, and the magnitude represents the drastic change in natural potential; the magnitude of the gradient is taken as the natural potential spatial gradient.

[0078] The cumulative change amplitude, cumulative change rate, and spatial gradient of the natural potential at the same monitoring electrode point are weighted and summed to obtain the natural potential mutation index of the corresponding monitoring electrode point; the weighting coefficients of the cumulative change amplitude, cumulative change rate, and spatial gradient of the natural potential can be adjusted according to the actual situation of the mine.

[0079] When water intrudes into the monitored area, the overall conductivity of the rock mass increases, causing the excitation current to show a continuous and trending increase. The excitation current trend index is used to capture the changing trend of the excitation current and reflect the overall change in the water content of the medium. By performing trend analysis on the excitation current, the interference of single-point current fluctuations can be avoided, and the reliability of water hazard identification can be improved.

[0080] The method for extracting the excitation current trend index is as follows:

[0081] A time window is set, and a linear fitting method is used to perform trend analysis on the excitation current sequence of each monitoring electrode point, and the slope of the change of the excitation current of each monitoring electrode point in the most recent time window is calculated; the slope of the change of the excitation current of each monitoring electrode point is taken as the excitation current time change rate.

[0082] Within any time window, based on the excitation current of each monitoring electrode point, a continuous excitation current distribution model is generated using the Kriging interpolation algorithm, and gradient calculation is performed on the excitation current distribution model to obtain the spatial rate of change of the excitation current.

[0083] The excitation current trend index is obtained by weighted summing of the excitation current time change rate and the excitation current spatial change rate. The weighting coefficients of the excitation current time change rate and the excitation current spatial change rate can be adjusted according to the actual situation of the mine. For example, the weighting coefficient of the excitation current time change rate can be set to 0.6 and the weighting coefficient of the excitation current spatial change rate can be set to 0.4.

[0084] Resistivity is the most direct parameter characterizing the water-bearing capacity of rock masses. Compared with the problem that single-point resistivity is easily affected by local interference, the volume average resistivity of the water-bearing anomaly area can better reflect the overall evolution trend of the water-bearing anomaly area; the resistivity change index is used to quantify the overall evolution trend of the water-bearing anomaly area.

[0085] like Figure 2 As shown, the extraction steps for the resistivity change index specifically include:

[0086] For the resistivity of all monitoring electrode points at any sampling time t, resistivity tomography is used for inversion calculation to obtain the three-dimensional resistivity distribution of the downhole monitoring area at sampling time t; such as Figure 3 As shown, the calculation method for the three-dimensional resistivity distribution state is as follows:

[0087] A three-dimensional spatial coordinate system is established with the area to be monitored as the target, so that the three-dimensional resistivity distribution is highly consistent with the actual geological environment, avoiding the inversion deviation of resistivity distribution due to geometric errors; the origin of the three-dimensional spatial coordinate system is any vertex of the area to be monitored, and the directions of the X-axis, Y-axis and Z-axis are consistent with the directions of the world coordinate system.

[0088] The three-dimensional space formed by the area to be monitored in the three-dimensional spatial coordinate system is divided into non-uniform unit grids, and the resistivity of each non-uniform unit grid is set as the unknown parameter to be solved. The size of the non-uniform unit grid is determined according to the distribution of the monitoring electrode points. In areas where the monitoring electrode points are dense, the size of the non-uniform grid is set to a grid of 5 meters by 5 meters, and in areas where the monitoring electrode points are sparse, the size of the non-uniform grid is set to a grid of 10 meters by 10 meters.

[0089] A forward modeling algorithm is constructed using the unknown parameters as independent variables to calculate the resistivity of each non-uniform cell grid, and the Poisson equation for the steady current field is solved using the finite element method or the finite difference method.

[0090] The objective function of the inversion algorithm is defined, and the objective function is solved using a linearized iterative algorithm to obtain the three-dimensional resistivity distribution of the area to be monitored. The objective function includes a data fitting term and a model constraint term. The data fitting term is used to minimize the error between resistivity and resistivity. The model constraint term is used to ensure that the resistivity is within a reasonable geological range and to avoid outliers.

[0091] In a three-dimensional spatial coordinate system, gradient calculation is performed on the three-dimensional resistivity distribution state to obtain the resistivity spatial change rate at each monitoring electrode point; the resistivity spatial change rate reflects the degree of non-uniformity of resistivity in space; the specific method of gradient calculation is as follows: the gradient vector of each non-uniform cell grid at the corresponding monitoring electrode point is calculated by using the numerical difference method, and the magnitude of the gradient vector is used as the resistivity spatial change rate at the corresponding monitoring electrode point.

[0092] Using the initial monitoring time as a reference time and the resistivity at the reference time as a reference resistivity, the ratio of the resistivity at any sampling time t to the reference resistivity is calculated to obtain the resistivity change rate over time at the corresponding monitoring electrode point.

[0093] The resistivity change index is obtained by weighted summation of the spatial change rate and the temporal change rate of resistivity. The weighting coefficients of the spatial change rate and the temporal change rate of resistivity can be adjusted according to the actual situation of the mine. For example, the weighting coefficient of the spatial change rate of resistivity can be set to 0.6, and the weighting coefficient of the temporal change rate of resistivity can be set to 0.4.

[0094] The natural potential mutation index, excitation current trend index, and resistivity change index are normalized, and the fusion early warning index is calculated based on the normalized characteristic indicators using a weighted summation method.

[0095] Optionally, the weight coefficients of the feature indicators reflect the priority of each feature indicator's contribution and can be obtained by training with historical data or simulated experimental data; the weight coefficients of the feature indicators can also adopt empirical values ​​set by experts; for example, the weight of the resistivity change index is set to 0.5, the weight of the natural potential change index is set to 0.3, and the weight of the excitation current trend index is set to 0.2.

[0096] The arithmetic mean of the fusion early warning indices of all monitoring electrode points within the monitoring area is calculated to obtain the global early warning index. A global early warning threshold is set, such as 0.65. When the global early warning index is greater than the global early warning threshold, a primary early warning signal is generated. By weighted fusion of feature indicators, collaborative early warning of multi-dimensional geoelectric field parameters is achieved, avoiding misjudgments caused by single parameters and improving the reliability of early warning.

[0097] Existing geological data from the mining site is collected, and a three-dimensional geological model is constructed based on this data. The geological data includes geological profiles, borehole data and their columnar sections, well logging data, geophysical data, and geological experimental results. The method for constructing the three-dimensional geological model is as follows:

[0098] The borehole data is standardized to extract the depth, lithology, lithological interface depth, fault location, and thickness of aquifer and impermeable layer for each borehole. The coordinate system of all geological exploration data is unified to ensure the consistency of spatial location of various data.

[0099] Based on the stratigraphic sequence in the borehole columnar section, the standard stratigraphic layers from oldest to newest are determined; based on the geological profile and the stratigraphic interface elevations of adjacent boreholes, the stratigraphic interface elevations of the un-drilled area are calculated using an interpolation algorithm, and a continuous three-dimensional stratigraphic framework is constructed.

[0100] For example, the Kriging interpolation algorithm is used to calculate the elevation of the stratigraphic interface in the un-drilled area; the variation function model of the Kriging interpolation algorithm is a spherical model, and its parameters include: the nugget value is set to 0.05 square meters, the sill value is set to 1.2 square meters, and the range is set to 150 meters; an anisotropy ratio is introduced in the Kriging interpolation process, wherein the anisotropy ratio along the stratigraphic strike is set to 1.5, and the anisotropy ratio along the stratigraphic dip is set to 1.0; the search radius of the Kriging interpolation is set to 225 meters, and the number of known boreholes participating in the calculation for each interpolation point is not less than 5;

[0101] For faults and fracture zones, based on the fault breakpoint coordinates in the borehole data, the fault strike and dip angle in the geological profile, and the structural extension characteristics in the geophysical data, the three-dimensional spatial morphology of the faults and fracture zones is constructed, and their strike, dip, dip angle and fault displacement are marked.

[0102] For the collapse column, a three-dimensional boundary model of the collapse column is generated by surface fitting based on the column boundary coordinates and morphological characteristics in the borehole data; for the aquifer and the impermeable layer, the spatial distribution range of the aquifer and the impermeable layer is delineated in the three-dimensional stratigraphic framework, and their permeability and water-bearing capacity are marked, in combination with the stratigraphic lithology and geological test results.

[0103] For example, a non-uniform rational B-spline surface fitting is used to generate the three-dimensional boundary model of the collapse column; the control point grid density of the non-uniform rational B-spline surface fitting is: no less than 8 control points per ring along the transverse profile of the collapse column, and no less than 1 control point per meter in the longitudinal profile; the order of the non-uniform rational B-spline surface fitting is set to 3 in both the U and V directions, the minimum node spacing is 0.5 meters, and the fitting tolerance is 0.1 meters; for the spatial delineation of the aquifer and the impermeable layer, a permeability coefficient threshold is set according to the geological test results, such as 1.0 × 10⁻⁶.-5 m / s and 1.0×10 -8 m / s; with a permeability coefficient greater than or equal to 1.0 × 10 m / s -5 Areas with a permeability coefficient of m / s or less are identified as aquifers, and areas with a permeability coefficient less than or equal to 1.0 × 10⁻⁶ m / s are classified as aquifers. -8 The area with a speed of m / s is identified as a waterproof layer;

[0104] The three-dimensional geological model is obtained by spatially integrating the three-dimensional stratigraphic framework, the three-dimensional spatial morphology of faults and fracture zones, the three-dimensional boundary model of collapse columns, and the spatial distribution of delineated aquifers and impermeable layers. Through the three-dimensional geological model, the geological structure and monitoring data are deeply integrated, which not only improves the geological rationality of the growth in high-risk areas and makes the growth in high-risk areas more consistent with actual hydrogeological conditions, but also improves the accuracy of locating water hazard sources.

[0105] In response to the initial warning signal, based on the aforementioned characteristic indicators and the three-dimensional geological model, a high-risk area is generated through regional growth.

[0106] like Figure 4 As shown, the specific steps for generating high-risk areas are as follows:

[0107] The characteristic indicators of each monitoring electrode point in the monitoring area are traversed, and the monitoring electrode points whose natural potential change index and resistivity change index are both greater than the risk threshold are marked as high-risk seed points; the risk threshold is greater than the global early warning threshold, which is used to screen monitoring electrode points with high early warning confidence and avoid noise interference.

[0108] Initialize a high-risk region centered on each high-risk seed point;

[0109] Based on the iterative growth rule, each high-risk region is iteratively grown until all non-uniform cell meshes around the high-risk region no longer satisfy the iterative growth rule.

[0110] The iterative growth rules include data similarity rules, spatial continuity rules, and geological guidance rules;

[0111] The data similarity rules are determined as follows:

[0112] Calculate the arithmetic mean of the same characteristic index of all monitoring electrode points in each non-uniform cell grid, and construct the characteristic matrix of the corresponding non-uniform cell grid based on these arithmetic means;

[0113] Specifically, the feature matrix is ​​as follows: for any non-uniform cell grid, calculate the arithmetic mean of the natural potential change index, the arithmetic mean of the excitation current trend index, and the arithmetic mean of the resistivity change index of all monitoring electrode points in the non-uniform cell grid; arrange the arithmetic mean of the natural potential change index, the arithmetic mean of the excitation current trend index, and the arithmetic mean of the resistivity change index in order, and the resulting one-dimensional vector is used as the feature matrix of the non-uniform cell grid.

[0114] Calculate the average feature matrix of all monitoring electrode points within any high-risk area; the average feature matrix is ​​calculated as follows: calculate the average value of the same feature index of all monitoring electrode points within the high-risk area, and arrange these average values ​​in order to obtain a one-dimensional vector, which is used as the average feature matrix of all monitoring electrode points within the corresponding high-risk area; use the average feature matrix as the feature matrix of the high-risk area.

[0115] For any non-uniform cell grid outside the high-risk area, calculate the Euclidean distance between its feature matrix and the feature matrix of the corresponding high-risk area;

[0116] If the Euclidean distance is less than a preset distance threshold, the corresponding non-uniform cell grid is determined to be similar to the high-risk area in terms of electrical characteristics, and the non-uniform cell grid satisfies the data similarity rule.

[0117] The spatial continuity rule is determined as follows: if any non-uniform cell grid on the periphery of any high-risk area is directly connected to the corresponding high-risk area in three-dimensional space, then the corresponding non-uniform cell grid is determined to satisfy the spatial continuity rule; the spatial continuity rule ensures the physical continuity of the high-risk area and ensures that the spatial morphology of the high-risk area is consistent with that of the actual water-bearing anomaly area.

[0118] In this embodiment, if any non-uniform cell grid on the periphery of any high-risk area is coplanar, shares an edge, or shares an angle with the corresponding high-risk area in three-dimensional space, then the non-uniform cell grid is determined to satisfy the spatial continuity rule.

[0119] The determination method for the geological guidance rule is as follows:

[0120] Based on a three-dimensional geological model, for any high-risk area, the orientation of any non-uniform cell grid outside the high-risk area relative to the high-risk area is obtained.

[0121] If the non-uniform cell grid is located in the high-risk area along the direction of the stratum dip, then the non-uniform cell grid is determined to satisfy the geological guidance rule. At the same time, the distance threshold in the data similarity rule is multiplied by a preset first relaxation coefficient to obtain the distance threshold corresponding to the non-uniform cell grid, allowing the high-risk area to grow in the direction of the non-uniform cell grid; the first relaxation coefficient is greater than 1, such as 1.2.

[0122] If the non-uniform cell grid is located in the direction of the fault or fracture zone within a preset area surrounding the high-risk area, then the non-uniform cell grid is determined to satisfy the geological guidance rule. Simultaneously, the distance threshold in the data similarity rule is multiplied by a preset second relaxation coefficient to obtain the distance threshold corresponding to the non-uniform cell grid, allowing the high-risk area to grow in the direction of the non-uniform cell grid. The second relaxation coefficient is greater than 1, such as 1.5. The first and second relaxation coefficients make the growth of the high-risk area more closely conform to the geological laws of groundwater migration.

[0123] If there is an aquitard boundary or intact rock layer in the three-dimensional geological model between the non-uniform cell grid and the high-risk area, it is determined that the non-uniform cell grid does not meet the geological guidance rule, and the high-risk area is prohibited from growing in the direction of the non-uniform cell grid.

[0124] Non-uniform cell grids that simultaneously satisfy data similarity rules, spatial continuity rules, and geological guidance rules are incorporated into the high-risk area, and the regional range and feature matrix of the high-risk area are updated. Through the iterative growth rules, intelligent expansion of the high-risk area is achieved, which helps to accurately capture the spatial distribution of water hazard sources.

[0125] Calculate the spatial volume, expansion rate, and internal early warning index of the high-risk area; determine the early warning level of the primary early warning signal based on preset early warning thresholds at each level, and set the corresponding early warning response;

[0126] The spatial volume of the high-risk area is calculated as follows: for any high-risk area, obtain all non-uniform cell meshes that constitute the high-risk area; the spatial volume of the corresponding high-risk area is obtained by accumulating the volumes of all non-uniform cell meshes.

[0127] The expansion rate is calculated as follows: the difference between the spatial volume of the same high-risk area at any sampling time and the previous sampling time is calculated, and the ratio of the absolute value of the difference to the sampling time interval is determined as the expansion rate of the high-risk area at the corresponding time.

[0128] The internal early warning index is obtained by calculating the arithmetic mean of the fusion early warning indices of all monitoring electrode points within the high-risk area, and using the arithmetic mean as the internal early warning index of the high-risk area.

[0129] The warning levels are divided into yellow warning, orange warning and red warning, from low to high.

[0130] The methods for determining the warning level and warning response are as follows:

[0131] Set warning thresholds at various levels; the warning thresholds at various levels include spatial volume critical value, expansion rate critical value, and internal warning index threshold.

[0132] Optionally, the critical value of the spatial volume is set according to the critical water inrush volume of the working face; for example, the critical water inrush volume of the working face is 200m³. 3 If the water inrush rate is / h, and the inrush water may fill the goaf or roadway within 30 minutes, then the critical value for the space volume is set at 100m³. 3 The critical expansion rate is determined based on historical data or simulation experiments on the evolution rate of water hazards; for example, by analyzing statistical data from historical water inrush cases in mining areas, the average rate of expansion of the pre-water inrush anomaly zone during the accelerated expansion stage is approximately 50 m³ / s. 3 / h, then the critical value for the spread rate is set to 50m. 3 / h; The internal early warning index threshold is greater than the global early warning threshold, such as 0.7;

[0133] For a primary warning signal, if there is no high-risk area within the monitoring area, the primary warning signal will be determined as a yellow warning, and technical personnel will be prompted to pay attention to the global warning index.

[0134] If at least one high-risk area exists within the monitored area, then each high-risk area is traversed, and a warning judgment is made according to the preset warning judgment rules; the warning judgment rules are as follows:

[0135] If the spatial volume of the high-risk area is greater than the spatial volume threshold, the corresponding high-risk area will be marked as an emergency.

[0136] If the expansion rate of the high-risk area is greater than the expansion rate threshold, the corresponding high-risk area will be marked as an emergency.

[0137] If the internal early warning index of the high-risk area is greater than the internal early warning index threshold, the corresponding high-risk area will be marked as an emergency.

[0138] If there are no high-risk areas marked as an emergency within the monitored area, the primary warning signal will be determined as an orange warning, and all high-risk areas will be marked in the three-dimensional geological model for technicians to determine the location of the hazard source.

[0139] If at least one high-risk area is marked as an emergency, the primary warning signal is determined to be a red warning and an emergency audible and visual alarm is triggered. At the same time, the high-risk areas marked as emergency and those not marked as emergency are marked in the three-dimensional geological model with different symbols. Through graded warning, real-time assessment and targeted response to water hazard risks are realized, improving the level of intelligence in mine safety management.

[0140] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0141] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of protection of this application, and these forms are all within the protection scope of this application.

Claims

1. A mine water hazard early warning method based on changes in geoelectric field parameters, characterized in that, Includes the following steps: Real-time synchronous acquisition of multi-dimensional geoelectric field parameters; The multidimensional geoelectric field parameters are subjected to time-series feature extraction to obtain feature indices; a global early warning index is calculated based on the feature indices, and a primary early warning signal is generated based on the global early warning index. Collect existing geological data from the mining company and construct a three-dimensional geological model based on the data; In response to the primary warning signal, a high-risk area is generated based on the characteristic indicators and the three-dimensional geological model, according to a preset iterative growth rule. Calculate the spatial volume, expansion rate, and internal early warning index of the high-risk area; The warning level of the primary warning signal is determined based on preset warning thresholds at each level, and a corresponding warning response is initiated.

2. The mine water hazard early warning method based on changes in geoelectric field parameters as described in claim 1, characterized in that, The specific steps for generating high-risk areas are as follows: Traverse the characteristic indicators of each monitoring electrode point in the monitoring area and mark the monitoring electrode points whose natural potential change index and resistivity change index are both greater than the preset risk threshold as high-risk seed points. Initialize a high-risk region with each high-risk seed point as the core; Based on the preset iterative growth rules, iterative growth is performed on any high-risk area; If all non-uniform cell grids surrounding any high-risk region no longer satisfy the iterative growth rule, then the iterative growth stops, and the corresponding high-risk region is obtained.

3. The mine water hazard early warning method based on changes in geoelectric field parameters as described in claim 2, characterized in that, The iterative growth rules include data similarity rules, spatial continuity rules, and geological guidance rules; A non-uniform cell grid that simultaneously satisfies the data similarity rule, spatial continuity rule, and geological orientation rule is added to the high-risk area, and the regional extent and average feature matrix of the high-risk area are updated. The spatial continuity rule is as follows: if any non-uniform cell grid outside any high-risk area is coplanar, shares an edge, or shares an angle with the corresponding high-risk area in three-dimensional space, then the non-uniform cell grid is determined to satisfy the spatial continuity rule.

4. The mine water hazard early warning method based on changes in geoelectric field parameters as described in claim 3, characterized in that, The data similarity rules are as follows: Calculate the arithmetic mean of the same characteristic index of all monitoring electrode points within each non-uniform cell grid; construct the characteristic matrix of the non-uniform cell grid based on the arithmetic mean; Calculate the average feature matrix of all monitoring electrode points in each high-risk area; The average feature matrix is ​​used as the feature matrix of the corresponding high-risk area; Calculate the Euclidean distance between the feature matrices of all non-uniform cell grids surrounding the high-risk area and the feature matrix of the corresponding high-risk area; if the Euclidean distance is less than a preset distance threshold, then the non-uniform cell grids are determined to satisfy the data similarity rule.

5. The mine water hazard early warning method based on changes in geoelectric field parameters as described in claim 4, characterized in that, The specific geological guidance rules are as follows: Based on a three-dimensional geological model, the orientation of any non-uniform cell grid outside a high-risk area relative to the corresponding high-risk area is obtained. If the non-uniform cell grid is located in the high-risk area along the direction of the stratum dip, then the non-uniform cell grid is determined to satisfy the geological guidance rule, and the distance threshold in the data similarity rule is multiplied by the preset first relaxation coefficient to obtain the distance threshold corresponding to the non-uniform cell grid. If the non-uniform cell grid is located in the direction of the fault or fracture zone within a preset area outside the high-risk area, then the non-uniform cell grid is determined to satisfy the geological guidance rule, and the distance threshold in the data similarity rule is multiplied by the preset second relaxation coefficient to obtain the distance threshold corresponding to the non-uniform cell grid. If there is an aquitard boundary or intact rock layer in the three-dimensional geological model between the non-uniform cell grid and the high-risk area, then the non-uniform cell grid is determined not to meet the geological guidance rule.

6. The mine water hazard early warning method based on changes in geoelectric field parameters as described in claim 5, characterized in that, The primary warning signal is generated in the following way: The characteristic indicators of each monitoring electrode point are extracted, including: spontaneous potential change index, excitation current trend index and resistivity change index; Set the weight coefficients for each characteristic index; normalize the natural potential mutation index, excitation current trend index and resistivity change index of each monitoring electrode point respectively, and then perform a weighted summation according to the weight coefficients to obtain the fusion early warning index of each monitoring electrode point. Calculate the arithmetic mean of the fusion warning index of all monitoring electrode points in the area to be monitored to obtain the global warning index; set a global warning threshold, and generate a primary warning signal when the global warning index is greater than the global warning threshold.

7. The mine water hazard early warning method based on changes in geoelectric field parameters as described in claim 6, characterized in that, The extraction steps for the resistivity change index specifically include: The resistivity of all monitoring electrode points at any sampling time is inverted to obtain the three-dimensional resistivity distribution of the area to be monitored at that sampling time. The gradient of the three-dimensional resistivity distribution at sampling time t is calculated to obtain the spatial rate of change of resistivity at the corresponding monitoring electrode point. The initial monitoring time is used as the reference time, and the resistivity at the reference time is used as the reference resistivity; Calculate the ratio of the resistivity at any sampling time t to the reference resistivity to obtain the resistivity change rate over time at the corresponding monitoring electrode point; The resistivity change index is obtained by weighted summing of the spatial rate of change of resistivity and the temporal rate of change of resistivity.

8. The mine water hazard early warning method based on changes in geoelectric field parameters as described in claim 7, characterized in that, The calculation steps for the three-dimensional resistivity distribution state include: A three-dimensional spatial coordinate system is established with the area to be monitored as the target; the origin of the three-dimensional spatial coordinate system is any vertex of the area to be monitored, and the directions of the X-axis, Y-axis, and Z-axis are consistent with the directions of the world coordinate system. The three-dimensional space formed by the area to be monitored in the three-dimensional spatial coordinate system is divided into non-uniform cell grids, and the resistivity of each non-uniform cell grid is set as the unknown parameter to be solved. A forward modeling algorithm is constructed using the unknown parameters as independent variables to calculate the resistivity of each non-uniform cell grid, and the Poisson equation for the steady current field is solved using the finite element method or the finite difference method. Define the objective function of the inversion algorithm and solve the objective function using a linearized iterative algorithm to obtain the three-dimensional resistivity distribution of the area to be monitored.

9. The mine water hazard early warning method based on changes in geoelectric field parameters as described in claim 8, characterized in that, The warning levels are divided into yellow, orange, and red warnings, from low to high; the determination method for the warning level and warning response of the primary warning signal is as follows: Set warning thresholds at various levels; the warning thresholds at various levels include spatial volume critical value, expansion rate critical value, and internal warning index threshold. For a primary warning signal, if there is no high-risk area within the monitoring area, the primary warning signal will be determined as a yellow warning, and attention will be paid to the global warning index. If there is at least one high-risk area within the monitoring area, then each high-risk area is traversed, and a warning judgment is made according to the preset warning judgment rules.

10. The mine water hazard early warning method based on changes in geoelectric field parameters as described in claim 9, characterized in that, The early warning judgment rules are as follows: If the spatial volume of the high-risk area is greater than the spatial volume threshold, or the expansion rate of the high-risk area is greater than the expansion rate threshold, or the internal early warning index of the high-risk area is greater than the internal early warning index threshold, then the corresponding high-risk area will be marked as an emergency. If there are no high-risk areas marked as an emergency within the monitored area, the primary warning signal will be determined as an orange warning, and the high-risk areas will be marked in the three-dimensional geological model. If at least one high-risk area is marked as an emergency, the primary warning signal will be determined as a red warning and an emergency audible and visual alarm will be triggered; the high-risk areas marked as emergency and the high-risk areas not marked as emergency will be marked with different symbols in the three-dimensional geological model.

Images