A method for monitoring the relative gravity of surface subsidence during coal mining

Abstract

This invention discloses a method for monitoring the relative gravity of surface subsidence during coal mining. It involves deploying quantum gravity reference stations around the subsidence area, arranging relative gravimeters in a grid pattern within the basin, and arranging fiber optic gravity sensor chains in the underground roadway. Simultaneously, a UAV equipped with a cold atom gravity gradiometer performs periodic scans. Based on real-time data, a gravity field-mass migration model is established, correlating coal extraction volume, changes in rock density, and surface gravity anomalies. Wavelet multi-scale decomposition is used to eliminate geological background interference, and hydrological correction is applied, incorporating water level and porosity to optimize gravity variation values. When the corrected gravity variation exceeds a threshold, the azimuth of rock fracture is calculated using the gravity gradient tensor. Equivalent extraction thickness is derived based on gravity variation inversion, and the surface subsidence morphology is dynamically predicted using a subsidence basin model. A state vector is constructed by combining gravity, gradient, and deformation data, and multi-source fusion prediction is achieved through a filtering algorithm.

Description

Technical Field

[0001] This invention relates to a method for monitoring surface subsidence, and more particularly to a method for monitoring the relative gravity of surface subsidence during coal mining. Background Technology

[0002] In the coal mining sector, surface subsidence monitoring is a crucial link in ensuring safe production and environmental remediation in mining areas. Traditional surface subsidence monitoring often employs single-point measurement methods, such as GPS positioning, leveling instruments, or synthetic aperture radar (InSAR). While these methods can acquire deformation data for a certain area, they generally suffer from limitations such as single monitoring dimensions and discontinuous spatial coverage. For example, GPS and leveling instruments rely on fixed measuring points, making it difficult to capture subtle deformation gradients in subsidence basins; InSAR, on the other hand, is limited by weather and topographical conditions, preventing high-frequency dynamic monitoring. Consequently, the real-time and refined monitoring capabilities for surface subsidence during coal mining are insufficient.

[0003] Current technologies face significant technical bottlenecks in monitoring changes in the surface gravity field caused by coal mining. Traditional gravity monitoring methods can only deploy instruments at a single point on the surface, failing to construct an integrated three-dimensional monitoring network that combines ground, air, and well monitoring, and thus making it difficult to comprehensively reflect the three-dimensional dynamic process of rock mass migration under the influence of mining. Furthermore, interference from the geological background field (such as deep tectonic movements) and changes in hydrological conditions (such as groundwater level fluctuations) severely affect the accuracy of gravity monitoring data. Existing data processing methods lack effective multi-scale interference removal mechanisms, resulting in large errors in monitoring results and an inability to accurately infer the subsidence evolution pattern.

[0004] Furthermore, existing subsidence prediction models are mostly based on empirical formulas or simplified physical models, failing to fully integrate the intrinsic relationship between changes in the gravity field and the mass migration of mining-induced rock masses. This makes it difficult to achieve dynamic inversion of equivalent mining thickness and accurate prediction of the spatial distribution of subsidence basins. Especially when facing coal seam mining under complex geological conditions, traditional methods cannot integrate multi-source monitoring data (such as gravity change values, gravity gradient components, and external deformation data), resulting in insufficient timeliness of subsidence early warning and inaccurate calculation of fracture azimuth angles, failing to meet the demand for high-precision, real-time monitoring in coal mine safety production. Summary of the Invention

[0005] This invention overcomes the shortcomings of the prior art and provides a method for monitoring the relative gravity of surface subsidence during coal mining.

[0006] To achieve the above objectives, the technical solution adopted by this invention is: a method for monitoring the relative gravity of surface subsidence during coal mining, comprising the following steps:

[0007] S1. Construct an integrated monitoring network of ground, air, and well; including setting up quantum gravity reference stations around the subsidence area, deploying relative gravimeters in the subsidence basin with a grid density of 100~500m*100~500m, arranging fiber optic grating gravity sensing chains in the downhole working face, and using UAVs equipped with cold atom gravity gradiometers for periodic regional scanning.

[0008] S2. Based on the real-time monitoring data from step S1, establish a gravity field-mass transfer theoretical model, with the following formula:

[0009] ;

[0010] In the formula, ρ represents the change in Earth's surface gravity; G is the gravitational constant. Density of coal; d represents the extracted volume; d represents the average depth. Spatial coordinates The density variation function of the rock strata at the location; r is the distance from the mass element to the measuring point;

[0011] S3. Wavelet multi-scale decomposition is used to eliminate the interference of the geological background field in the surface gravity variation values. The formula expression is:

[0012] ;

[0013] In the formula, The actual surface gravity variation value is represented by N; the total number of factors in the geological background field is represented by N; and the factor of a single geological background field is represented by k. This is the scaling factor; These are wavelet basis functions; This is the wavelength threshold.

[0014] In a preferred embodiment of the present invention, in step S1, the arrangement of the fiber grating gravity sensing chain satisfies the spatial relationship:

[0015] ;

[0016] In the formula, H represents the sensor spacing; H represents the average depth.

[0017] In a preferred embodiment of the present invention, in step S1, when the UAV scans, the scanning path satisfies the spatial sampling theorem:

[0018] ;

[0019] In the formula, Minimum phase velocity; This is the highest deformation frequency.

[0020] In a preferred embodiment of the present invention, after performing step S3, hydrological disturbance correction is further included:

[0021]

[0022] In the formula, This is the corrected value for the change in surface gravity. For correction factors; This refers to the change in water level. The porosity of the rock stratum is 0.15 to 0.35.

[0023] In a preferred embodiment of the present invention, when the corrected surface gravity change value meets the warning threshold, a fracture warning is triggered, and the warning is issued using a formula. The rupture azimuth angle was calculated; where, The azimuth of the rock strata fracture; The mixed derivative components of the gravity gradient tensor in the yz direction; The xz-direction mixed derivative components of the gravity gradient tensor.

[0024] In a preferred embodiment of the present invention, the fiber optic grating gravity sensing chain uses a temperature compensation model to calculate the gravity sensitivity, and the formula is as follows:

[0025] ;

[0026] In the formula, g is the gravity sensitivity; This is the gravity sensitivity coefficient; Here, T is the temperature coupling coefficient; T is the downhole temperature. For gravity-sensitive fiber wavelengths, The reference fiber wavelength.

[0027] In a preferred embodiment of the present invention, the method further includes inverting the equivalent mining thickness based on gravity changes for subsidence dynamic prediction; based on the absolute value of the maximum gravity change obtained from monitoring, combined with the gravitational constant, coal density and average mining depth data, the equivalent mining thickness of the coal seam is inverted through the equivalent mining thickness calculation formula; then the equivalent mining thickness is substituted into the subsidence basin prediction model, and the spatial distribution of the subsidence basin is described by an exponential function with the equivalent mining thickness as the independent variable, using the maximum subsidence value as the amplitude, wherein the exponential term includes two independent attenuation terms along the strike and dip of the working face, and the attenuation coefficient is determined by the ratio of the mining depth to the tangent of the main influence angle in the corresponding direction.

[0028] In a preferred embodiment of the present invention, a three-dimensional state vector including subsidence, subsidence gradient and time-varying derivative is established, and time series prediction is achieved through state transition matrix; at the same time, an observation equation is constructed, and gravity change value, vertical component of gravity gradient and external deformation monitoring data are used as observation input, and a filtering algorithm including system noise and observation noise is used to achieve multi-source data fusion.

[0029] In a preferred embodiment of the present invention, when constructing the integrated monitoring network of the sky, earth, and well in step S1, the relative gravimeter adopts an adaptive grid densification algorithm to dynamically adjust the deployment density. When a local gravity anomaly gradient is detected to exceed a preset threshold, the densification module is automatically triggered to add a micro gravity sensor in the abnormal area to form a dynamic densification monitoring network. The density of the densified grid is 1.5 to 3 times that of the original grid density, and the spacing between the densified sensors satisfies the critical spacing formula in the rock stratum movement angle theory.

[0030] In a preferred embodiment of the present invention, a dynamic correction coefficient is introduced into the equivalent mining thickness calculation formula. The dynamic correction coefficient is dynamically adjusted based on real-time monitored rock stress change data through a BP neural network model. The dynamic correction achieves adaptive improvement of mining thickness inversion accuracy in areas with abrupt changes in geological conditions.

[0031] The corrected formula is ,in, The base correction factor; This is a stress-sensitive correction amount; This represents the real-time stress variation value; The stress reference threshold; The stress ratio is a hyperbolic tangent function. Mapping to the [-1,1] interval achieves a smooth transition of stress changes.

[0032] This invention addresses the shortcomings of the prior art and has the following beneficial effects:

[0033] (1) A comprehensive, all-encompassing three-dimensional monitoring network was constructed by deploying quantum gravity reference stations around the subsidence area, deploying relative gravimeters with a grid density of 100-500 meters within the subsidence basin, arranging fiber optic grating sensor chains in the downhole working face, and using UAVs equipped with gravity gradiometers for periodic scanning. The spacing of the downhole sensor chains was dynamically adjusted according to the average mining depth to ensure that rock fracture signals could be captured even during deep mining. At the same time, the interference of downhole temperature on monitoring accuracy was offset by dual-fiber differential measurement. The UAV scanning path was designed based on deformation propagation characteristics to avoid spatial aliasing of monitoring data. This overcame the limitations of traditional single-point monitoring, controlled the gravity field monitoring error within ±1 microgal, and greatly improved the data acquisition efficiency compared with traditional methods, realizing dynamic monitoring of the entire space from the surface to the downhole.

[0034] (2) A gravity field theoretical model considering coal extraction quality loss and overburden density redistribution was established. The dynamic process of mass migration within the rock mass was accurately characterized by three-dimensional spatial integration, which reduced the calculation error of gravity changes under complex geological conditions compared with the traditional simplified model. Wavelet multi-scale decomposition technology was used to eliminate background interference such as deep tectonic movements, and hydrological correction was performed by combining water level changes and rock porosity parameters. This effectively eliminated the influence of geological and hydrological factors on the monitoring data, improved the signal-to-noise ratio of the true gravity change value, and provided high-precision data support for subsidence analysis.

[0035] (3) Based on gravity monitoring data, the equivalent mining thickness of the coal seam is inverted and substituted into the subsidence basin prediction model. The spatial distribution of the subsidence area can be accurately described by the exponential function. By constructing a three-dimensional state vector containing subsidence amount, change gradient and time-varying characteristics, and integrating gravity field data and external deformation monitoring information, the early warning of rock fracture is realized. At the same time, the rock fracture azimuth can be calculated by gravity gradient tensor analysis, which significantly improves the real-time early warning capability of coal mine subsidence disaster. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a flowchart of a preferred embodiment of the monitoring method of the present invention. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein. Therefore, the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0040] like Figure 1 As shown, a method for monitoring the relative gravity of surface subsidence during coal mining includes the following steps:

[0041] S1. Construct an integrated monitoring network of ground, air, and well; this includes setting up quantum gravity reference stations around the subsidence area, deploying relative gravimeters in the subsidence basin at a grid density of 100~500m*100~500m, arranging fiber optic grating gravity sensing chains in the downhole working face, and using UAVs equipped with cold atom gravity gradiometers for periodic regional scanning.

[0042] In step S1, the arrangement of the fiber Bragg grating gravity sensing chain satisfies the spatial relationship:

[0043] ;

[0044] In the formula, H represents the sensor spacing; H represents the average depth. Based on the rock strata movement angle theory, the sensor spacing... Dynamic correlation with average depth H: The greater the mining depth, the wider the range of rock deformation transmission, requiring an increase in monitoring spacing; however, when H>250, a limit is imposed. =50m, ensuring that critical layer fracture signals can still be captured during deep mining.

[0045] The fiber optic grating gravity sensing chain uses a temperature-compensated model to calculate gravity sensitivity, and its formula is as follows:

[0046] ;

[0047] In the formula, g is the gravity sensitivity; This is the gravity sensitivity coefficient; Here, T is the temperature coupling coefficient; T is the downhole temperature. For gravity-sensitive fiber wavelengths, The reference fiber wavelength is used. Downhole temperature variations cause fiber wavelength drift, leading to distortion in gravity sensitivity. This model utilizes coaxial dual-fiber differential measurement: For gravity-sensitive fiber wavelengths, Reference fiber wavelength; is the temperature coupling coefficient; T is the downhole temperature; real-time cancellation of thermal noise.

[0048] In step S1, when the UAV scans, the scanning path satisfies the spatial sampling theorem:

[0049] ;

[0050] In the formula, Minimum phase velocity; This represents the highest deformation frequency. To avoid spatial aliasing, the band spacing must be designed based on the deformation propagation characteristics. This theorem ensures that the sampling interval can resolve the smallest deformation wavelength.

[0051] Step S1 aims to establish a full-space three-dimensional monitoring system. The quantum gravity reference station provides a high-precision absolute reference (±0.5μGal) to eliminate instrument system errors; the gridded relative gravimeter enables continuous monitoring of the surface gravity field, and the 100~500m grid density balances cost and spatial resolution; the downhole fiber optic sensor chain directly senses the mass migration of mining-induced rock strata; and the UAV-borne gradient instrument obtains the gravity gradient tensor through periodic scanning.

[0052] The combined use of ground-air-well and gravity-well systems covers the entire area affected by mining activities, solving the problem of blind spots in traditional single-point monitoring and greatly improving the efficiency of gravity data acquisition.

[0053] To further clarify, the quantum gravity reference station is deployed on stable bedrock outside the subsidence area. Employing the principle of cold atom interferometry, it provides an absolute gravity reference with an accuracy of ±0.5 μGal. Its function is not only to eliminate instrument system errors but also to observe changes in the background gravity field caused by tectonic movements in the separated region (such as slow crustal uplift) over a long period.

[0054] Gridded relative gravimeters are deployed on the surface of the subsidence basin at a grid density of 100–500 m. The grid density is dynamically adjusted according to the mining intensity. In high-intensity mining areas with an advance speed >5 m / d, a 100 m × 100 m grid is used to capture details of rapid deformation; in areas with weak deformation, a 500 m × 500 m grid is used to reduce monitoring costs.

[0055] The instrument used is a high-precision spring gravimeter (such as CG-6), which is used to synchronously observe and transmit data to the central processing platform daily.

[0056] S2. Based on the real-time monitoring data from step S1, establish a gravity field-mass transfer theoretical model, with the following formula:

[0057] ;

[0058] In the formula, ρ represents the change in Earth's surface gravity; G is the gravitational constant. Density of coal; d represents the extracted volume; d represents the average depth. Spatial coordinates The density variation function of the rock strata at a given location; r is the distance from the mass element to the measuring point. The model is based on the law of conservation of mass. The first term describes the mass loss (negative effect) caused by coal mining, and the second term characterizes the density redistribution (positive / negative effect) during the overburden fracturing-compaction process. Breaking away from the traditional point mass approximation, it introduces a triple integral to accurately characterize the spatial heterogeneity within the rock mass.

[0059] S3. Wavelet multi-scale decomposition is used to eliminate the interference of the geological background field in the surface gravity variation values. The formula expression is:

[0060] ;

[0061] In the formula, The actual surface gravity variation value is represented by N; the total number of factors in the geological background field is represented by N; and the factor of a single geological background field is represented by k. This is the scaling factor; These are wavelet basis functions; The wavelength threshold is used to separate background interference such as deep tectonic motion (wavelength > 5km).

[0062] After performing step S3, hydrological disturbance correction is also included:

[0063]

[0064] In the formula, This is the corrected value for the change in surface gravity. For correction factors; This refers to the change in water level. The porosity of the rock stratum is 0.15 to 0.35.

[0065] When the corrected surface gravity change value meets the warning threshold, a fault warning is triggered, and the warning is issued via the formula. The rupture azimuth angle was calculated; where, The azimuth of the rock strata fracture; The mixed derivative components of the gravity gradient tensor in the yz direction; The xz-direction mixed derivative components of the gravity gradient tensor.

[0066] It also includes using gravity variation inversion to calculate equivalent mining thickness for subsidence dynamic prediction; based on the absolute value of the maximum gravity variation obtained from monitoring, combined with the gravitational constant, coal density and average mining depth data, the equivalent mining thickness of the coal seam is inverted through the equivalent mining thickness calculation formula; then the equivalent mining thickness is substituted into the subsidence basin prediction model, and the maximum subsidence value is used as the amplitude, and an exponential function with equivalent mining thickness as the independent variable is used to describe the spatial distribution of the subsidence basin, where the exponential term includes two independent attenuation terms along the strike and dip of the working face, and the attenuation coefficient is determined by the ratio of the mining depth to the tangent of the main influence angle in the corresponding direction.

[0067] A three-dimensional state vector containing subsidence, subsidence gradient, and time-varying derivative is established, and time series prediction is achieved through the state transition matrix. At the same time, an observation equation is constructed, taking gravity change value, vertical component of gravity gradient, and external deformation monitoring data as observation inputs. A filtering algorithm containing system noise and observation noise is used to achieve multi-source data fusion.

[0068] In another embodiment, when constructing the integrated ground-air-well monitoring network in step S1, the adaptive grid densification algorithm is implemented through the following process: First, the surface relative gravimeter grid density is initialized to a 200m×200m baseline grid; when a local gravity anomaly gradient exceeds the 5μGal / m² threshold in real time, the system automatically triggers the densification module. The densification strategy uses a quadtree segmentation algorithm to refine the grid in the anomaly area into 100m×100m or 50m×50m sub-grids, resulting in a grid density 1.5-3 times that of the original grid. During the densification process, the spacing of the UAV scanning flight path is simultaneously adjusted to half of the densified grid to ensure that the spatial sampling rate meets the Nyquist criterion. For example, in a mining depth of 300m, when a precursor signal of rock strata fracturing is detected, the fiber optic grating sensor chain spacing is dynamically reduced from 60m to 30m, forming a three-dimensional monitoring network and achieving an upgrade in monitoring dimensions from "point-line-surface".

[0069] In the equivalent thickness inversion in step S7, the BP neural network model adopts a three-layer structure: the input layer contains real-time stress. , benchmark threshold The model incorporates eight characteristic parameters, including rock porosity; the hidden layer has 16 neurons, with the tanh function chosen as the activation function to achieve a smooth mapping from -1 to 1; the output layer outputs dynamic correction coefficients. The model was trained using 2000 sets of historical monitoring data from 2018 to 2024, with weights optimized through backpropagation and a regularization coefficient of 0.01 set to prevent overfitting. Correction formula... In Calibration was performed using a stress-sensitive test when... When =1, The maximum value of 0.3m corresponds to the equivalent mining thickness correction for a mining area with a mining depth of 500m. This correction coefficient, along with the vertical component of the gravity gradient and temperature compensation data, is input into a Kalman filter to improve the subsidence prediction accuracy to the ±0.1m level after multi-source data fusion.

[0070] The dynamic, encrypted monitoring network ensures the capture of minute deformation signals in areas with abrupt geological changes by adjusting sensor deployment density in real time. The dynamic correction coefficient, through a stress-gravity coupling mechanism, eliminates the impact of hydrological interference and stress fluctuations on the mining thickness inversion. The synergistic effect of these two components gives the system "adaptive and self-correcting" capabilities, reducing the timeliness of subsidence prediction from the traditional 72 hours to 2 hours under complex geological conditions such as deep mining and fault activation, providing more accurate decision support for safe coal mine production. In practice, the system automatically evaluates the monitoring network density and the effectiveness of the correction coefficient every 6 hours, continuously optimizing monitoring parameters through a closed-loop feedback mechanism, forming a dynamically evolving technical system.

[0071] Based on the preferred embodiments of the present invention described above, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A method of monitoring the relative gravity of surface subsidence during coal mining, characterised in that, Comprise the following steps: S1, construct the integration monitoring network of heaven and earth well; Wherein Comprise, outside the subsidence area, quantum gravity reference station is laid out, in the subsidence basin, relative gravimeter is deployed according to grid density 100 ~ 500 m * 100 ~ 500 m, fiber grating gravity sensing chain is arranged in the working face along the groove, and cold atom gravity gradiometer is carried by unmanned aerial vehicle to carry out periodic regional scanning; S2, based on the data monitored in real time in step S1, a gravity field-mass migration theoretical model is established, and the formula expression is: ； wherein G is the gravitational constant; is the density of coal; is the volume of extraction; d is the average depth; is the spatial coordinate is the density variation function of the rock formation at the point r is the distance of the mass element to the measuring point; S3, wavelet multiscale decomposition is used to eliminate the interference of geological background field in surface gravity change value, and the formula expression is: ； In the formula, is a real ground gravity change value; N is the total number of factors of the geological background field; k is a single geological background field factor; is a scale coefficient; is a wavelet base function; is a wavelength threshold value.

2. The method for monitoring the relative gravity of surface subsidence in the process of coal mining according to claim 1, characterized in that: In step S1, the laying of the fiber grating gravity sensing chain satisfies the spatial relationship formula: ； In the formula, is the sensor spacing; H is the average depth.

3. The method for monitoring the relative gravity of surface subsidence in the process of coal mining according to claim 1, characterized in that: In step S1, when the unmanned aerial vehicle scans, the scanning path satisfies the spatial sampling theorem: ； wherein is the minimum phase velocity; is the maximum deformation frequency.

4. The method for monitoring the relative gravity of surface subsidence in the process of coal mining according to claim 1, characterized in that: After step S3 is executed, hydrological interference correction is further included: In the formula, is the corrected ground gravity change value; is the correction coefficient; is the water level change amount; is the rock porosity, and the value is 0.15-0.

35.

5. A method of monitoring the relative gravity of ground subsidence during coal mining according to claim 4, characterised in that: When the corrected ground gravity change value meets the early warning threshold, a fracture early warning is triggered, and a fracture azimuth is calculated by the formula ; wherein, is the rock fracture azimuth; is the y-z direction mixed derivative component of the gravity gradient tensor; is the x-z direction mixed derivative component of the gravity gradient tensor.

6. A method of monitoring the relative gravity of surface subsidence during coal mining according to claim 2, characterized in that: The fiber grating gravity sensing chain uses a temperature compensation model to calculate gravity sensitivity, and the formula expression is: ； where g is the gravity sensitivity; is the gravity sensitivity coefficient; is the temperature coupling coefficient; T is the downhole temperature; is the gravity sensitive fiber wavelength, is the reference fiber wavelength.

7. The method for monitoring the relative gravity of surface subsidence in the process of coal mining according to claim 1, characterized in that: Further include based on gravity change inversion equivalent mining thickness, for subsidence dynamic prediction;Based on the maximum gravity change value absolute value obtained by monitoring, combined with the universal gravitation constant, coal density and average mining depth data, the equivalent mining thickness of coal seam is inverted through the equivalent mining thickness calculation formula;Further, the equivalent mining thickness is substituted into the subsidence basin prediction model, and the maximum subsidence value is taken as the amplitude, and the spatial distribution form of the subsidence basin is described by using the exponential function with the equivalent mining thickness as the independent variable, wherein the exponential term includes two independent attenuation terms along the working face trend and tendency, and the attenuation coefficients are determined by the ratio of mining depth to the tangent value of corresponding direction main influence angle.

8. The method for monitoring the relative gravity of surface subsidence in the process of coal mining according to claim 1, characterized in that: A three-dimensional state vector including subsidence, subsidence gradient and time-varying derivative is established, and time series prediction is realized through state transition matrix;Meanwhile, the observation equation is constructed, the gravity change value, gravity gradient vertical component and external deformation monitoring data are used as observation input, and the filtering algorithm containing system noise and observation noise is used to realize multi-source data fusion.

9. The method for monitoring the relative gravity of surface subsidence in the process of coal mining according to claim 1, characterized in that: In step S1, when the integration monitoring network of heaven and earth well is constructed, the relative gravimeter uses an adaptive grid encryption algorithm to dynamically adjust the layout density, when a local gravity anomaly gradient exceeds a preset threshold, an encryption module is automatically triggered to add a micro gravity sensor in the abnormal area, and a dynamic encryption monitoring network is formed, wherein the encryption grid density is 1.5 ~ 3 times of the original grid density, and the spacing between the sensors after encryption satisfies the critical spacing formula in the rock stratum movement angle theory.

10. The method for monitoring the relative gravity of surface subsidence in the process of coal mining according to claim 7, characterized in that: The dynamic correction coefficient is introduced in the equivalent mining thickness calculation formula, the dynamic correction coefficient is dynamically adjusted based on the real-time monitored rock stratum stress change data through the BP neural network model, and the adaptive improvement of mining thickness inversion accuracy in the geological condition mutation area is realized through dynamic correction.

Images