An inversion imaging method for roof failure monitoring based on optical fiber strain

Abstract

The application discloses an electrical method roof failure monitoring inversion imaging method based on optical fiber strain, electrode sensors and optical fiber sensors are arranged in a hole at the same time, the electrode sensors are used to acquire apparent resistivity data in a target monitoring area, the optical fiber sensors are used to acquire strain data of each electrode sensor and surrounding rock in real time, and the coupling quality of each electrode sensor and rock mass is evaluated in combination with a set threshold value, data weight factors of each electrode sensor are obtained according to the evaluation result, so that a data weighting matrix of all apparent resistivity is obtained, the data weighting matrix in the electrical method inversion is dynamically adjusted adaptively, the influence of unreliable data caused by poor electrode coupling on the inversion imaging result is reduced, and finally the accuracy and reliability of the electrical method inversion result are effectively improved. The method can effectively make up for the interference influence of the electrical method monitoring under complex working conditions, and provides strong data support for accurate monitoring and imaging of the roof failure of a mining working face.

Description

Technical Field

[0001] This invention belongs to the field of underground engineering monitoring technology, specifically a fiber optic strain-based electrical method for roof damage monitoring and inversion imaging. Background Technology

[0002] In mining operations, real-time monitoring of roof damage is crucial for production safety and efficient resource extraction. Currently, monitoring typically relies on sensor deployment within exploratory boreholes. Electrical resistivity tomography (ERT) electrodes are widely used to monitor changes in roof electrical parameters. However, during mining operations, exploratory boreholes are affected by the development of the roof's "two zones" (collapse zone and fracture zone), which can lead to borehole wall rupture or even collapse. This reduces the electrode coupling quality within the borehole, affecting the accuracy of monitoring data and causing non-convergence or distortion in ERT inversion imaging results. Fiber optic sensors, with their high sensitivity and anti-interference capabilities, can monitor pressure and strain changes within exploratory boreholes in real time, thereby analyzing the development of the roof's "two zones." However, their monitoring range is limited to within the exploratory borehole and cannot comprehensively reflect overall roof changes.

[0003] Although electrical resistivity tomography (ERT) and fiber optic technologies each have significant advantages, existing research has largely focused on single applications or joint interpretation of data from collaborative monitoring. Previous studies have shown a relationship between the excitation current in ERT and rock mass strain, but the electrode coupling conditions and strain data have not yet been effectively combined to improve the accuracy of ERT inversion. Furthermore, the reliability of ERT monitoring data is largely affected by the quality of electrode coupling.

[0004] Therefore, the research direction of this invention is to provide a new method that can acquire strain information at the electrode in real time, dynamically evaluate the electrode coupling quality, and then adjust the data weighting matrix in the electrical inversion process to ultimately improve the reliability and stability of the inversion results. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides an electrical resistivity tomography (OR) method for monitoring and inverting roof damage based on fiber optic strain. By acquiring strain information at the electrodes in real time through optical fiber, dynamically evaluating the electrode coupling quality, and then adjusting the data weighting matrix in the OR process, the reliability and stability of the inversion results are ultimately improved, providing strong data support for accurate imaging of roof damage under complex working conditions.

[0006] To achieve the above objectives, the technical solution adopted by this invention is: an electrical method for monitoring and inverting roof damage based on fiber optic strain, the specific steps of which are as follows:

[0007] A. Deployment of Electrode Sensors and Fiber Optic Sensors: Before mining begins in the coal seam, exploratory boreholes are drilled inclined upwards from the coal seam face. Multiple electrode sensors and fiber optic sensors are deployed within these boreholes. All electrode sensors are connected to an electrical resistivity meter outside the borehole, ensuring good coupling between each sensor and the rock mass for real-time monitoring of roof rock mass damage. Fiber optic sensors are deployed along the borehole axis, with each electrode sensor in close contact with a fiber optic sensor at a different location. These sensors are connected to a fiber optic demodulator outside the borehole for monitoring roof rock mass deformation at each sensor location. Multiple detection time points are then pre-set during the mining process.

[0008] B. Obtaining initial state data of the rock mass: Before the coal seam working face is mined, the electrical resistivity instrument is started to control each electrode sensor to perform an electrical resistivity detection on the roof rock mass to obtain the apparent resistivity background value. At the same time, the fiber optic demodulator is started to control the fiber optic sensor to perform a fiber optic strain measurement on the roof rock mass to obtain the strain background value at each electrode sensor before mining.

[0009] C. Collecting rock mass data during the mining process: When mining begins, at the first detection time point, the electrical resistivity meter and fiber optic demodulator are activated simultaneously to acquire the apparent resistivity data and strain data at each electrode sensor at that detection time point. Then, the absolute value of the difference between the strain data at each electrode sensor and the strain background value at each electrode sensor in step B is taken to obtain the strain difference value at each electrode sensor at that detection time point.

[0010] D. Determine the coupling quality between each electrode sensor and the rock mass during the mining process: Set a threshold for the strain difference value. and maximum allowable strain difference If the strain difference at a certain electrode sensor is less than or equal to a threshold value... If the strain difference at a certain electrode sensor is greater than the strain difference threshold, it is determined that the electrode sensor is well coupled with the rock mass; And less than the maximum allowable strain difference. If the strain difference at a certain electrode sensor is greater than or equal to the maximum allowable strain difference, then the electrode sensor is considered to have poor coupling with the rock mass. If the strain difference is not found, the electrode sensor is determined to be completely faulty. Based on the above criteria, the strain difference values ​​at each electrode sensor in step C are analyzed, and the data weighting factor for each electrode sensor at this detection time point is determined using the following formula:

[0011]

[0012] in The threshold value for strain difference; This represents the maximum permissible strain difference. For electrode sensor i, the data weighting factor; The strain difference value of electrode sensor i;

[0013] E. Apparent Resistivity Inversion: Using the data weighting factors of each electrode sensor from step D. Calculate the data weighting factor for each apparent resistivity data point at the current detection time point in step C. Since each apparent resistivity data point is associated with four electrodes (i.e., power supply electrodes A and B, and measurement electrodes M and N), the weighting factor for each apparent resistivity data point is calculated using the following formula:

[0014]

[0015] The data weighting factor for measuring electrode N;

[0016] The data weighting factors for all apparent resistivity data at this detection time point are calculated using the above formula and combined into a data weight matrix Wd for electrical resistivity inversion. The data weighting matrix Wd is a diagonal matrix, and its diagonal elements are the data weighting factors for each apparent resistivity data point, as shown in the following form:

[0017]

[0018] Where N is the total number of apparent resistivity data;

[0019] Then, the aforementioned data weighting matrix is ​​applied to the objective function of the electrical resistivity inversion, and the data is optimized by weighting. The weighted least squares method is then used for inversion to adjust the reliability of the apparent resistivity data and improve the stability of the electrical resistivity inversion. The specific formula is as follows:

[0020]

[0021] in The apparent resistivity prediction data obtained from the inversion model m; For regularization parameters; Weighting matrix for the model; For reference model;

[0022] The final image of the top plate obtained by electrical inversion at the time point of this detection was obtained;

[0023] F. Continuous monitoring of roof damage: During the mining process, steps C to E are repeated at each detection time point to obtain roof images obtained by electrical inversion at each detection time point, and the roof damage during the mining process is continuously monitored.

[0024] Furthermore, in step A, multiple electrode sensors are arranged at equal intervals inside the probe hole, with a distance of 1m between two adjacent electrode sensors. The fiber optic sensors are then tightly connected to each electrode sensor at different positions using adhesive tape to ensure that the strain data collected by the fiber optic sensors accurately reflects the coupling of the electrodes.

[0025] Furthermore, in step A, the angle between the exploration hole and the mining direction is 30° to 45°; the length of the exploration hole is 100 meters, and the diameter of the borehole is 100 millimeters; after the electrode sensors and fiber optic sensors are installed, the exploration hole is grouted with bentonite slurry to ensure good coupling between each electrode sensor and the fiber optic sensor and the rock mass of the borehole wall.

[0026] Furthermore, in step A, each of the deployed electrode sensors is numbered sequentially from the deepest point of the probe hole outwards, and the position corresponding to each electrode sensor number is marked on the fiber strain curve collected by the fiber optic sensor according to the position of each electrode sensor, which is used for subsequent collection of strain data at each electrode sensor.

[0027] Furthermore, steps B and C, the electrical detection, employ the Wenner polarity method.

[0028] Furthermore, the threshold and maximum allowable strain difference in step D are obtained as follows: first, the top rock mass of the working face to be monitored is sampled, and the sample is subjected to strain test in the laboratory, thereby calibrating and obtaining the threshold and maximum allowable strain difference corresponding to this detection.

[0029] Compared with existing technologies, this invention drills upward-sloping exploration holes within the longwall face, simultaneously deploying electrode sensors and fiber optic sensors within these holes. The electrode sensors acquire apparent resistivity data within the target monitoring area, while the fiber optic sensors acquire real-time strain data between each electrode sensor and the surrounding rock. The coupling quality between each electrode sensor and the rock mass is evaluated using set thresholds. Based on the evaluation results, data weighting factors for each electrode sensor are obtained, resulting in a combined data weighting matrix Wd for all apparent resistivity data. This data weighting matrix is ​​then introduced into the electrical resistivity inversion process. Furthermore, the coupling quality of each electrode sensor is evaluated and the data weighting matrix Wd is updated at each detection time point. By adaptively and dynamically adjusting the data weighting matrix in the electrical resistivity inversion, the impact of unreliable data due to poor electrode coupling on the inversion imaging results is reduced, achieving high-resolution electrical resistivity imaging of roof damage. Ultimately, this effectively improves the accuracy and reliability of the electrical resistivity inversion results. This method can effectively compensate for interference in electrical resistivity monitoring under complex working conditions, providing strong data support for the accurate monitoring and imaging of roof damage in longwall faces. Attached Figure Description

[0030] Figure 1 This is a schematic diagram of the overall process of the present invention;

[0031] Figure 2 This is a schematic diagram of the layout of the electrode sensor and the fiber optic sensor in this invention. Detailed Implementation

[0032] The present invention will be further described below.

[0033] like Figure 1 As shown, the specific steps of this invention are as follows:

[0034] A. Deploying electrode sensors and fiber optic sensors: such as Figure 2 As shown, before the coal seam working face is mined, an exploration hole is drilled inclined upwards from the coal seam working face, and multiple electrode sensors and fiber optic sensors are installed in the hole. All electrode sensors are connected to an electrical resistivity meter outside the exploration hole, and the coupling quality between each electrode sensor and the rock mass is ensured for real-time monitoring of the damage to the roof rock mass. The distance between two adjacent electrode sensors is 1m, and the fiber optic sensors are tightly connected to each electrode sensor at different positions using tape to ensure that the strain data collected by the fiber optic sensors can accurately reflect the coupling of the electrodes. Fiber optic sensors are deployed along the axis of the exploration borehole, with each electrode sensor in close contact with a fiber optic sensor at a different location. The fiber optic sensors are connected to a fiber optic demodulator outside the borehole to monitor the deformation of the roof rock mass at each electrode sensor location. The angle between the exploration borehole and the mining direction is 30° to 45°. The borehole is 100 meters long and 100 millimeters in diameter. After the electrode and fiber optic sensors are deployed, the borehole is grouted with bentonite slurry to ensure good coupling between the electrode and fiber optic sensors and the rock mass of the borehole wall. Each electrode sensor is sequentially numbered from the deepest point of the borehole outwards, and its position is marked on the fiber optic strain curve acquired by the fiber optic sensor according to its location. This marking is used for subsequent acquisition of strain data at each electrode sensor location. Multiple detection time points are then pre-set during the mining process.

[0035] B. Obtaining initial state data of the rock mass: Before the coal seam working face is mined, the electrical resistivity instrument is started and the Wenner electrode running mode is used to control each electrode sensor to perform an electrical resistivity detection on the roof rock mass to obtain the apparent resistivity background value. At the same time, the fiber optic demodulator is started to control the fiber optic sensor to perform a fiber optic strain measurement on the roof rock mass to obtain the strain background value at each electrode sensor before mining.

[0036] C. Collecting rock mass data during the mining process: When mining begins, at the first detection time point, the electrical resistivity meter and fiber optic demodulator are activated simultaneously to acquire the apparent resistivity data and strain data at each electrode sensor at that detection time point. Then, the absolute value of the difference between the strain data at each electrode sensor and the strain background value at each electrode sensor in step B is taken to obtain the strain difference value at each electrode sensor at that detection time point.

[0037] D. Determine the coupling quality between each electrode sensor and the rock mass during the mining process: Set a threshold for the strain difference value. and maximum allowable strain difference The specific process is as follows: First, samples are taken from the roof rock mass of the working face to be monitored, and then strain tests are conducted on the samples in the laboratory to calibrate and determine the threshold and maximum allowable strain difference value corresponding to this detection. If the strain difference value at a certain electrode sensor is less than or equal to the threshold strain difference value... If the electrode sensor is well coupled with the rock mass, the data weighting factor is 1; if the strain difference at a certain electrode sensor is greater than the strain difference threshold... And less than the maximum allowable strain difference. If the strain difference at a certain electrode sensor is greater than or equal to the maximum allowable strain difference, then the electrode sensor is considered to have poor coupling with the rock mass. If the electrode sensor fails completely, its data weighting factor is 0. Based on the above criteria, the strain difference values ​​at each electrode sensor in step C are analyzed, and the data weighting factor for each electrode sensor at this detection time point is determined using the following formula:

[0038]

[0039] in The threshold value for strain difference; This represents the maximum permissible strain difference. For electrode sensor i, the data weighting factor; The strain difference value of electrode sensor i;

[0040] E. Apparent Resistivity Inversion: Using the data weighting factors of each electrode sensor from step D. Calculate the data weighting factor for each apparent resistivity data point at the current detection time point in step C. Since each apparent resistivity data point is associated with four electrodes (i.e., power supply electrodes A and B, and measurement electrodes M and N), the weighting factor for each apparent resistivity data point is calculated using the following formula:

[0041]

[0042] The data weighting factor for measuring electrode N;

[0043] The data weighting factors for all apparent resistivity data at this detection time point are calculated using the above formula and combined into a data weight matrix Wd for electrical resistivity inversion. The data weighting matrix Wd is a diagonal matrix, and its diagonal elements are the data weighting factors for each apparent resistivity data point, as shown in the following form:

[0044]

[0045] Where N is the total number of apparent resistivity data;

[0046] Then, the aforementioned data weighting matrix is ​​applied to the objective function of the electrical resistivity inversion, and the data is optimized by weighting. The weighted least squares method is then used for inversion to adjust the reliability of the apparent resistivity data and improve the stability of the electrical resistivity inversion. The specific formula is as follows:

[0047]

[0048] in The apparent resistivity prediction data obtained from the inversion model m; For regularization parameters; Weighting matrix for the model; For reference model;

[0049] The final image of the top plate obtained by electrical inversion at the time point of this detection was obtained;

[0050] F. Continuous monitoring of roof damage: During the mining process, steps C to E are repeated at each detection time point to obtain roof images obtained by electrical inversion at each detection time point, and the roof damage during the mining process is continuously monitored.

[0051] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An optical fiber strain-based electrical method roof failure monitoring inversion imaging method, characterized in that, The specific steps are: A, layout of electrode sensor and optical fiber sensor: before mining in the coal seam working face, a probe hole is drilled upwardly from the coal seam working face, and a plurality of electrode sensors and optical fiber sensors are arranged in the hole, the plurality of electrode sensors are connected with the electrical method instrument outside the probe hole, and the coupling quality of each electrode sensor with the rock mass is ensured, which is used for real-time monitoring of the damage of the roof rock mass; the optical fiber sensors are arranged along the axis of the probe hole, and each electrode sensor is in close contact with the optical fiber sensors at different positions, the optical fiber sensors are connected with the optical fiber demodulator outside the probe hole, which is used for monitoring the deformation of the roof rock mass at the position of each electrode sensor; then a plurality of detection time points in the mining process are set in advance; B, obtaining the initial state data of the rock mass: before mining in the coal seam working face, the electrical method instrument is started to control each electrode sensor to perform electrical method detection on the roof rock mass to obtain the apparent resistivity background value, and the optical fiber demodulator is started to control the optical fiber sensor to perform optical fiber strain measurement on the roof rock mass to obtain the strain background value of each electrode sensor before mining; C, collecting rock mass data in the mining process: start mining in the working face, and when the first detection time point is reached in the mining process, the electrical method instrument and the optical fiber demodulator are started at the same time to obtain the apparent resistivity data collected at the detection time point and the strain data at each electrode sensor, respectively; then the strain data collected at each electrode sensor is subtracted from the strain background value of each electrode sensor in step B to obtain the strain difference value of each electrode sensor at the detection time point; D. Determine the coupling quality of each electrode sensor and rock mass in the mining process: set the threshold value of strain difference and the maximum allowable strain difference ; If the strain difference at a certain electrode sensor is less than or equal to a threshold value of the strain difference , it is determined that the electrode sensor is well coupled with the rock mass; if the strain difference at a certain electrode sensor is greater than the threshold value of the strain difference and less than a maximum allowable strain difference , it is determined that the electrode sensor is poorly coupled with the rock mass; if the strain difference at a certain electrode sensor is greater than or equal to the maximum allowable strain difference , it is determined that the electrode sensor is completely failed; the strain differences at the electrode sensors in step C are analyzed according to the above standards, and the data weight factors of the electrode sensors at the detection time point are determined by using the following formula: wherein is a threshold value for the strain difference; is a maximum allowed strain difference; is a data weight factor for electrode sensor i; is a strain difference for electrode sensor i; E. Resistivity inversion: Using the data weighting factors for each electrode sensor in step D The data weighting factors for each apparent resistivity data at the current sounding time point in step C are calculated Since each apparent resistivity data is associated with four electrodes, the data weighting factors for each apparent resistivity data are calculated according to the following equation: The data weight factor of all apparent resistivity data at this detection time point is calculated by the above formula, which is combined into a data weight matrix Wd for electrical method inversion. The data weight matrix Wd is a diagonal matrix, and the diagonal elements are the data weight factors of each apparent resistivity data. The specific form is as follows: Where N is the total number of apparent resistivity data; Then the above data weight matrix is applied to the objective function of electrical method inversion to optimize and solve the data, and the weighted least squares method is used for inversion. The specific formula is as follows: wherein apparent resistivity prediction data computed for the inversion model m; is a regularization parameter; is a model weighting matrix; is a reference model; Finally, the roof imaging obtained by electrical method inversion at this detection time point is obtained; F, continuously monitor the roof damage: in the mining process of the working face, each detection time point is reached, and steps C to E are repeated to obtain the roof imaging obtained by electrical method inversion at each detection time point, and the roof damage in the mining process is continuously monitored.

2. The method according to claim 1, wherein, The plurality of electrode sensors in step A are equally spaced in the probe hole, the distance between two adjacent electrode sensors is 1m, and the different positions of the optical fiber sensor are in close contact with each electrode sensor by using adhesive tape, so as to ensure that the strain data collected by the optical fiber sensor can truly reflect the coupling condition of the electrode.

3. The fiber optic strain-based electrical method roof failure monitoring inversion imaging method of claim 1, wherein, The angle between the probe hole and the mining direction in step A is 30° to 45°; the length of the probe hole is 100 meters, and the diameter of the borehole is 100 millimeters; the probe hole is grouted with bentonite slurry after the electrode sensor and the optical fiber sensor are arranged, so as to ensure that each electrode sensor and the optical fiber sensor are well coupled with the rock mass of the hole wall.

4. The fiber optic strain-based electrical method roof failure monitoring inversion imaging method of claim 1, wherein, The step A numbers each electrode sensor from the deepest part of the exploration hole outwardly in sequence, and marks the position corresponding to the number of each electrode sensor in the optical fiber strain curve collected by the optical fiber sensor according to the position of each electrode sensor, for collecting strain data at each electrode sensor subsequently.

5. The fiber optic strain-based electrical method roof failure monitoring inversion imaging method of claim 1, wherein, The steps B and C adopt Wenner running electrode method.

6. The fiber optic strain-based electrical method roof failure monitoring inversion imaging method of claim 1, wherein, The threshold value of strain difference and the maximum allowable strain difference in the step D are obtained in the following manner: sampling the roof rock mass of the required monitoring working face, and performing strain experiment on the sample in the laboratory, so as to calibrate the threshold value of strain difference and the maximum allowable strain difference corresponding to the present detection.

Images