Mine laneway anchor rod support state online monitoring and early warning system

Abstract

The invention relates to the technical field of coal mine safety monitoring and early warning, in particular to an online monitoring and early warning system for the supporting state of a mine roadway anchor rod. Comprising an anchor rod distributed monitoring device, a gas parameter monitoring device, a dynamic disturbance monitoring device, a data aggregation and time service device, a ground comprehensive analysis server and an early warning issuing terminal. The server maps the multi-source sequence with the unified timestamp to a roadway space grid unit, an inversion coefficient matrix and an observation vector are constructed in a grid, an anchoring interface segment equivalent shear stiffness parameter vector is solved, and a support abnormity score, a gas abnormity score, a dynamic abnormity score and a coupling risk index are calculated; and outputting coal and gas outburst coupling early-warning information when the coal and gas outburst coupling early-warning information meets the threshold value and meets the time-space consistency triggering rule. According to the method, on-line early warning is triggered according to time-space consistency through regular inversion in a grid and coupling reasoning under constraint.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of coal mine safety monitoring and early warning, and particularly relates to a mine roadway anchor rod support state online monitoring and early warning system. BACKGROUND

[0002] Coal and gas outburst is one of the dynamic disasters prone to occur in underground mining activities, which is related to factors such as gas occurrence conditions, geological structure and stress evolution caused by mining disturbance. In order to reduce the outburst risk, the existing mine usually arranges gas parameter monitoring devices, dynamic disturbance monitoring devices (such as microseismic or acoustic emission monitoring devices) and drilling while drilling detection means, monitors the gas anomaly and dynamic disturbance characteristics, and performs early warning according to the monitoring results.

[0003] For example, the prior art with publication number CN114810213A discloses a multi-source information fusion intelligent early warning method and device for coal and gas outburst, which performs advanced detection on gas geological anomaly bodies and calculates advanced detection characteristic indexes, simultaneously performs real-time monitoring based on microseismic signals and gas emission data and calculates real-time monitoring characteristic indexes, and combines with drilling while drilling measurement to calculate drilling while drilling detection characteristic indexes; on this basis, multi-source heterogeneous data are fused and analyzed to determine a comprehensive early warning danger index, and a corresponding early warning is output according to comparison of the early warning threshold and the comprehensive early warning danger index.

[0004] Although the above-mentioned prior art can perform fusion early warning by using multi-source information such as gas and dynamic disturbance, the early warning elements mainly focus on disaster-causing factors such as gas anomaly and dynamic disturbance, and lack online quantitative characterization of the anchor rod support state related to the stability of the roadway surrounding rock in the same scale as the outburst early warning. Especially under the action of mining disturbance and stress-gas coupling, the load transmission capacity of the anchor rod anchoring interface may present a segmented change along the length direction of the anchor rod, if the fusion early warning is only based on the gas parameters and dynamic disturbance parameters, it is difficult to online identify and quantify the change of the load transmission capacity of the anchoring interface, and it is also difficult to associate and identify the support state change with the gas anomaly and dynamic anomaly on a unified spatial scale and trigger, so that it is difficult to form a coupling early warning output facing the spatial position of the roadway.

[0005] Therefore, the prior art has the following technical problems: in the process of multi-source monitoring and early warning of coal and gas outburst, there is a lack of a technical solution capable of online inversion and quantification of the segmented change of the load transmission capacity of the anchoring interface based on anchor rod monitoring data, and coupling identification and triggering of the quantification results with gas anomaly information and dynamic anomaly information in a unified spatial grid, which leads to difficulty in establishing a clear spatio-temporal correspondence between the outburst early warning and the change of the anchor rod support state. SUMMARY

[0006] To overcome the aforementioned technical deficiencies, the present invention aims to provide an online monitoring and early warning system for the support status of anchor bolts in mine roadways. This invention utilizes the distributed strain data of anchor bolts within a roadway spatial grid to segment the anchoring interface into equivalent shear stiffness parameter vectors. Regularized inversion and calculation of support anomaly score based on it. Then, score points for abnormal gas levels. And abnormal scores In the weighted adjacency matrix Coupled reasoning is performed under constraints and early warning is triggered according to the spatiotemporal consistency rule, thereby realizing online coupled early warning of support-gas-power.

[0007] This invention discloses an online monitoring and early warning system for the status of anchor bolt support in mine roadways, comprising: Anchor bolt distributed monitoring device: The anchor bolt distributed monitoring device is deployed on multiple anchor bolts in the mine roadway to obtain the discrete strain measurement sequence and the corresponding temperature sequence of each anchor bolt along the length of the bolt. A gas parameter monitoring device is installed in the mine roadway to obtain at least two gas monitoring sequences from gas volume fraction, borehole gas pressure, and borehole gas flow rate. Dynamic disturbance monitoring device, which is installed in the corresponding area of ​​the mine roadway, is used to acquire dynamic monitoring sequences formed by microseismic event information or acoustic emission event information; The data aggregation and timing device is used to uniformly synchronize the discrete strain measurement sequence, temperature sequence, gas monitoring sequence and dynamic monitoring sequence and generate a unified timestamp. The ground-based integrated analysis server communicates with the data aggregation and timing device. The ground-based integrated analysis server includes: Spatiotemporal grid mapping unit is used to map each sequence after a unified timestamp to a preset lane space grid unit; The anchorage interface piecewise parameter inversion unit is used to construct the inversion coefficient matrix based on the discrete strain measurement sequence and temperature sequence within each roadway spatial grid cell. With observation vector And solve for the piecewise equivalent shear stiffness parameter vector of the anchorage interface. ,in satisfy:

[0008] in, for The transpose of the matrix, For a predefined positive definite regularized matrix, The regularization coefficient is positive; The support-gas-power constraint graph reasoning unit is configured to construct a weighted adjacency matrix with roadway space grid cells as nodes , and based on the anchor interface segmented equivalent shear stiffness parameter vector to calculate the support anomaly score , calculate the gas anomaly score based on the gas monitoring sequence , calculate the power anomaly score based on the power monitoring sequence , and calculate the grid-level coupling risk index , wherein

[0009] , wherein , , is a preset weight coefficient The hierarchical early warning unit is configured to output coal and gas outburst coupling early warning information when the support anomaly score , the gas anomaly score , the power anomaly score and the spatiotemporal consistency triggering rule are simultaneously satisfied in the same roadway space grid cell; wherein , , are preset threshold values; the spatiotemporal consistency triggering rule is that in the same roadway space grid cell, the support anomaly score , the gas anomaly score , the power anomaly score respectively continuously satisfy the condition that the number of samples corresponding to the threshold values is not less than a preset integer , and at least a preset integer of the roadway space grid cells spatially adjacent to the roadway space grid cell respectively satisfy the condition that the number of continuous samples corresponding to the threshold values is not less than in the same time window The early warning publishing terminal is configured to receive and publish the coal and gas outburst coupling early warning information and the roadway space grid cell identifier thereof.

[0010] Preferably, the anchor rod distributed monitoring device is configured to set at least three strain measurement lines in the same anchor rod cross section and form a three-channel strain measurement sequence, and the anchor interface segmented parameter inversion unit is configured to solve the three-channel strain measurement sequence into an axial equivalent strain sequence and a bending component sequence, and only use the axial equivalent strain sequence to construct the inversion coefficient matrix and the observation vector .

[0011] Preferably, the anchor interface segmented parameter inversion unit is configured to perform temperature compensation on the axial equivalent strain sequence, and the axial equivalent strain after temperature compensation satisfies:

[0012] wherein, is an axial equivalent strain measurement value, is a temperature measurement value, is a reference temperature, is a temperature compensation coefficient; and the temperature-compensated axial equivalent strain sequence is used to construct an inversion coefficient matrix and an observation vector .

[0013] Preferably, the anchorage interface segment parameter inversion unit introduces a discrete shear hysteresis constraint when constructing the inversion coefficient matrix and the observation vector , so that each sampling position satisfies:

[0014] wherein, is a discrete value of the axial force of the anchor rod converted from the temperature-compensated axial equivalent strain sequence, and , is the elastic modulus of the anchor rod, is the cross-sectional area of the anchor rod; is a discrete value of the axial displacement of the anchor rod obtained by integrating the temperature-compensated axial equivalent strain sequence along the length, and , is the distance between adjacent sampling positions; is a segment parameter vector of the equivalent shear stiffness parameters of the anchorage interface segment corresponding to the segment; and the inversion coefficient matrix and the observation vector are assembled from the coefficient terms of the above constraint at all sampling positions.

[0015] Preferably, the anchorage interface segment parameter inversion unit determines the segment number and segment boundary of the segment parameter vector of the equivalent shear stiffness parameters of the anchorage interface segment in a way of segmenting the support construction record, and the support construction record at least includes the start and end depth or start and end position record of the grouting segment, and the start and end depth or start and end position of each grouting segment is projected as an axial coordinate boundary along the axial direction of the anchor rod, so that the axial coordinate boundary is taken as a set of segment boundaries.

[0016] Preferably, the regular matrix adopts a second-order difference regular matrix and satisfies , wherein is a second-order difference operator matrix; and the anchorage interface segment parameter inversion unit performs a non-negative projection process on the segment parameter vector of the equivalent shear stiffness parameters of the anchorage interface segment solved, so that the vector each component of the components satisfies a non-negative constraint.

[0017] Preferably, the support-gas-dynamic constraint graph reasoning unit calculates a support abnormality score when the support abnormality score , , is obtained by integrating support abnormality sub-items according to preset weights, wherein: the support abnormality sub-items are weighted deviation amounts of the equivalent shear stiffness parameter vector of the anchoring interface segment relative to the reference vector , and satisfy:

[0018] wherein is a preset diagonal weight matrix, is the equivalent shear stiffness parameter vector of the anchoring interface segment under the reference working condition; the support abnormality sub-items are inversion residual norm ; the support abnormality sub-items are continuous coverage lengths of low stiffness zones, the low stiffness zones are a continuous segment set within the same roadway space grid element and satisfy , wherein is a preset proportionality coefficient, and are the first components of the vector and the vector , respectively; the support abnormality score satisfies:

[0019] wherein, , , are preset non-negative weight coefficients.

[0020] Preferably, the gas parameter monitoring device is at least used to obtain the borehole gas pressure and the borehole gas flow, and the support-gas-dynamic constraint graph reasoning unit calculates a gas abnormality score when the gas abnormality score at least includes a pressure gradient extreme value item of the borehole gas pressure along the borehole depth direction and a flow deviation item of the borehole gas flow relative to a reference flow , wherein the reference flow is a statistical reference value of the same roadway space grid element within a reference working condition or a sliding statistical window.

[0021] Preferably, the event information outputted by the power disturbance monitoring device at least includes event energy and event occurrence time, and the support-gas-power constraint graph reasoning unit calculates the power anomaly score At least includes the event energy cumulative density term and the event rate term within the unit time window; and the spatio-temporal grid mapping unit obtains the positioning uncertainty parameter corresponding to the event position information, and according to the positioning uncertainty parameter, the single event is probabilistically distributed to one or more roadway space grid units and then participates in the calculation of the power anomaly score .

[0022] Preferably, the edge weight of the weighted adjacency matrix is determined by the roadway spatial distance and the support construction time difference, and further introduces a geological structure dominant direction parameter, so that the edge weight of the adjacent roadway space grid units along the geological structure dominant direction is multiplied by a direction gain coefficient greater than 1, and the edge weight of the transversely adjacent roadway space grid units is multiplied by a direction gain coefficient equal to 1, wherein the geological structure dominant direction parameter is given by roadway geological data or tunneling measurement data.

[0023] Preferably, the support-gas-power constraint graph reasoning unit determines the coupling edge weight between the support anomaly score and the gas anomaly score using the time lag mutual information maximization rule, selects the time lag that maximizes the mutual information between the lag sequence of the support anomaly score and the current sequence of the gas anomaly score from a preset time lag set as the support-gas coupling time lag, and outputs the support-gas coupling time lag to the hierarchical early warning unit.

[0024] Preferably, the hierarchical early warning unit further satisfies the event chain timing constraint before outputting the coal and gas outburst coupling early warning information, and the event chain timing constraint includes: the time when the support anomaly score first reaches the support threshold , the time when the gas anomaly score first reaches the gas threshold , the time when the power anomaly score first reaches the power threshold , satisfies falls within a preset interval , and falls within a preset interval .

[0025] Preferably, the ground comprehensive analysis server further includes a digital twin consistency correction unit, which updates the twin parameter vector according to the mismatch vector of the measured strain vector and the twin predicted strain vector and the updated twin parameter vector is utilized to update the reference vector or the diagonal weight matrix The updated reference vector or the updated diagonal weight matrix is fed back to the support-gas-power constraint graph reasoning unit to participate in subsequent support anomaly scoring and the calculation of the grid-level coupling risk index.

[0026] Preferably, the digital twin consistency correction unit applies parameter upper and lower bound constraints when updating the twin parameter vector and adopts a Huber-type piecewise robust function to suppress the influence of the mismatch vector abnormal component, the Huber-type piecewise robust function satisfies:

[0027] wherein is a preset threshold, and the updated twin parameter vector is used for online updating of the reference vector or the diagonal weight matrix

[0028] Preferably, the system further comprises a linkage control interface unit for solving ventilation set value and drainage set value when outputting the coal and gas outburst coupling early warning information, wherein the linkage post-predicted risk index satisfies:

[0029] and aims to minimize the objective function under the constraints , , solves and ; wherein , is the reference set value, , is a preset sensitivity coefficient, is a preset risk upper limit, , , , is the set value boundary.

[0030] After the above technical solutions are adopted, compared with the prior art, the following beneficial effects are achieved: ​​​1. Realize the quantitative online characterization of support state, improve the pertinence of early warning: Through temperature compensation of anchor discrete strain measurement sequence and temperature sequence, and inversion of segmented equivalent shear stiffness parameter vector of anchoring interface under discrete shear hysteresis force transmission constraint, the support state is changed from "qualitative judgment" to calculable parameterized results, which is convenient for stable identification of local weak section on grid cell scale.

[0031] 2. Support-gas-dynamic multi-source coupling identification, reduce the risk of false judgment of single signal: The grid level coupling risk index is constructed by using support abnormal score, gas abnormal score and dynamic abnormal score, and the early warning is triggered by hierarchical threshold criterion, so that the early warning decision is constrained by support response, gas change and dynamic disturbance, which is beneficial to improve the reliability of discrimination in complex working conditions.

[0032] 3. Introduce the spatiotemporal consistency triggering rule to enhance the stability and noise resistance of early warning: By continuous sample number constraint and spatial adjacent grid consistency constraint, the influence of accidental peak, local noise or short-time missing data on early warning output is reduced, so that the early warning is more consistent with the spatiotemporal continuity characteristics of roadway risk evolution.

[0033] 4. Strengthen the causal consistency by event chain time sequence constraint to improve the credibility of early warning: Before outputting the coal and gas outburst coupling early warning information, further constraint the first threshold time of support abnormality, gas abnormality and dynamic abnormality to meet the preset time sequence interval, so that the early warning trigger is consistent with the evolution sequence of multi-source signals, which is beneficial to reduce the trigger situation that does not conform to the typical evolution process.

[0034] 5. Determine the support-gas coupling time delay based on time delay mutual information to improve the asynchronous signal fusion effect: The coupling time delay is determined by maximizing the mutual information in the preset time delay set, and is used for coupling edge weight / alignment constraint, so that the fusion of support response and gas response is more reasonable when there is propagation / lag, thereby improving the effectiveness of coupling reasoning.

[0035] 6. Digital twin consistency correction and online update to adapt to working condition drift and individual differences: The mismatch vector between measured strain vector and twin predicted strain vector is used to update the twin parameter vector, and the benchmark vector or diagonal weight matrix is updated online, so that the system can maintain adaptive consistency of parameter benchmark under conditions such as support material difference, construction difference and environmental change.

[0036] 7. Use Huber type robust mechanism and parameter upper and lower bound constraint to improve the robustness under abnormal data: The abnormal component influence of mismatch vector is suppressed by segmented robust function, and the parameter upper and lower bound constraint is applied to reduce the probability of model update instability caused by abnormal points, local faults or accidental interference.

[0037] 8. Probability distribution considering the uncertainty of the dynamic event location, improving the spatial consistency of the dynamic side index: After the event is probabilistically distributed to one or more roadway spatial grid cells according to the location uncertainty parameter, it participates in the dynamic anomaly score calculation, so that the contribution of the dynamic disturbance to the grid level risk is more consistent with the spatial diffusion characteristics under the condition of location error.

[0038] 9. Weighted adjacency matrix comprehensive distance, construction time difference and geological structure dominant direction, improve the expression ability of spatial reasoning: Through the edge weight, the spatial proximity, construction correlation and directional difference are reflected, so that the modeling of the propagation / correlation relationship between grids is closer to the actual geological and construction conditions of the roadway.

[0039] 10. Early warning-linkage control integration, supporting the constraint solution of ventilation and extraction set value: When outputting the early warning information, further solve the ventilation set value and the extraction set value, and optimize under the risk upper limit and set value boundary constraint, so that the early warning result can be directly converted into an executable control decision reference, improving the timeliness and operability of disposal. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 Schematic diagram of the axial distribution of the equivalent shear stiffness of the anchoring interface segment along the anchor rod; Figure 2 Schematic diagram of the variation curve of the borehole gas pressure with the borehole depth; Figure 3 Schematic diagram of the comparison of the risk index with time; Figure 4 Schematic diagram of the comparison of the score in grid cell A with the threshold value; Figure 5 Flowchart of the online monitoring and early warning system for the mine roadway anchor support state according to the present application. DETAILED DESCRIPTION

[0041] The advantages of the present application will be further described below in combination with the drawings and specific examples.

[0042] The exemplary embodiments will be described in detail hereinbelow with reference to the drawings. When the following description refers to the drawings, identical numbers on different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present disclosure. Instead, they are merely examples of apparatuses and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0043] The terminology used in the disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. As used in the description of the disclosure and the appended claims, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0044] It is to be understood that the terms first, second, third, etc. can be employed merely for distinguishing between information objects of different categories and quantities, and are not necessarily intended to denote temporal or chronological precedence of one information object over another, unless specifically so indicated by the context. It is to be understood that the terms "comprises", "comprising", "includes", "including" and "has" or "having", are used interchangeably and are meant to be open-ended terms that specific to the extent required by the specification and claims. It is to be understood that such terms are intended to be interpreted from the perspective of a person of ordinary skill in the art, and are not intended to be interpreted from the perspective of a person skilled in the art.

[0045] In the description of the present application, it should be understood that the terms "longitudinal", "lateral", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like refer to the orientation or positional relationship as shown in the drawings, which are for purposes of convenience and brevity in describing the present application and its attendant advantages, and are not intended to be limiting of the present application or any embodiment thereof, as described above and as claimed below. It is to be understood that the terms "including", "comprising", "consisting" and "consisting essentially of" are used interchangeably and are meant to be open ended terms that specific to the extent required by the specification and claims.

[0046] In the description of the present application, unless otherwise specified and limited, it should be understood that the terms "mounting", "connecting", "connection" should be interpreted broadly, for example, it can be mechanical connection or electrical connection, or internal communication of two elements, or direct connection, or indirect connection through intermediate medium, and the specific meaning of the above terms can be understood by the person skilled in the art according to the specific circumstances.

[0047] In the following description, the suffixes such as "module", "part" or "unit" used for an element are used only for convenience of explanation of the present application, and have no specific meaning by themselves. Therefore, "module" and "part" can be used interchangeably.

[0048] Referring to Figure 5As shown, this embodiment provides an online monitoring and early warning system for the support status of anchor bolts in mine roadways, including: a distributed anchor bolt monitoring device, which is deployed on multiple anchor bolts in the mine roadway to acquire discrete strain measurement sequences and corresponding temperature sequences for each anchor bolt along its length; a gas parameter monitoring device, which is installed in the mine roadway to acquire at least two gas monitoring sequences from gas volume fraction, borehole gas pressure, and borehole gas flow rate; and a dynamic disturbance monitoring device, which is installed in a corresponding area of ​​the mine roadway to acquire micro-vibrations. The system comprises: a dynamic monitoring sequence formed from event information or acoustic emission event information; a data aggregation and timing device for uniformly timing the discrete strain measurement sequence, temperature sequence, gas monitoring sequence, and dynamic monitoring sequence and generating a unified timestamp; and a ground-based integrated analysis server, which is communicatively connected to the data aggregation and timing device. The ground-based integrated analysis server includes: a spatiotemporal grid mapping unit for mapping each sequence after unified timestamp to a preset roadway spatial grid unit; and an anchoring interface segmented parameter inversion unit for constructing an inversion coefficient matrix based on the discrete strain measurement sequence and temperature sequence within each roadway spatial grid unit. With observation vector And solve for the piecewise equivalent shear stiffness parameter vector of the anchorage interface. ,in satisfy: ,in, for The transpose of the matrix, For a predefined positive definite regularized matrix, Positive regularization coefficients; support-gas-dynamic constraint graph reasoning unit, used to construct a weighted adjacency matrix with roadway spatial grid units as nodes. And based on the segmented equivalent shear stiffness parameter vector of the anchoring interface. Calculate support anomaly score Calculate gas anomaly score based on gas monitoring sequence Calculate dynamic anomaly score based on dynamic monitoring sequence And calculate the grid-level coupling risk index. ,in: ,in, , , Preset weighting coefficients; graded early warning units, used within the same roadway spatial grid unit, when simultaneously meeting the support anomaly score... Gas abnormality score abnormal power score And when the spatiotemporal consistency triggering rule is met, output coal and gas outburst coupling early warning information; among which , , is a preset threshold value; the spatiotemporal consistency triggering rule is that in the same roadway space grid unit, the support anomaly score , the gas anomaly score , and the dynamic anomaly score respectively satisfy the condition that the number of samples corresponding to the threshold value is not less than a preset integer , and at least a preset integer of the roadway space grid units spatially adjacent to the roadway space grid unit respectively satisfy the condition that the number of corresponding continuous samples is not less than in the same time window; the early warning issuing terminal is configured to receive and issue the coal and gas outburst coupling early warning information and the roadway space grid unit identifier. Specifically, the following is performed: The mine roadway bolting state online monitoring and early warning system of the embodiment is arranged in the two-side support section of a coal mining roadway and the front of a tunneling head, and the system includes an anchor rod distributed monitoring device, a gas parameter monitoring device, a dynamic disturbance monitoring device, a data aggregation and time service device, a ground comprehensive analysis server, and an early warning issuing terminal. The anchor rod distributed monitoring device is arranged on multiple anchor rods of the mine roadway, and is configured to obtain a discrete strain measurement sequence along the length direction of each anchor rod and a corresponding temperature sequence; the gas parameter monitoring device is arranged in the mine roadway, and is configured to obtain at least two gas monitoring sequences in a gas volume fraction, a borehole gas pressure, and a borehole gas flow; the dynamic disturbance monitoring device is arranged in a corresponding area of the mine roadway, and is configured to obtain a dynamic monitoring sequence formed by microseismic event information or acoustic emission event information; the data aggregation and time service device is configured to unify the time service of the discrete strain measurement sequence, the temperature sequence, the gas monitoring sequence, and the dynamic monitoring sequence and generate a unified time stamp; the ground comprehensive analysis server is in communication connection with the data aggregation and time service device, and the early warning issuing terminal is configured to receive and issue the coal and gas outburst coupling early warning information and the roadway space grid unit identifier.

[0049] For ease of instantiation, the roadway space is divided into preset roadway space grid units, and the grid unit size is 2 m x 2 m x 2 m. One of the roadway space grid units is selected as a grid unit A, and one anchor rod in the grid unit A is selected as an example anchor rod. The length of the example anchor rod is 2.4 m, the sampling interval along the axial direction of the anchor rod is , and the sampling position is . The sampling period is 1 s, the support-gas-dynamic statistical time window is 60 s, and the sliding statistical window is 10 min.

[0050] It should be noted that, for ease of description and understanding, the following technical names are defined by symbols in the embodiment: an axial equivalent strain measurement value is denoted as an axial equivalent strain measurement value , and a temperature-compensated axial equivalent strain is denoted as a temperature-compensated axial equivalent strain , the temperature measurement value is denoted as temperature measurement value , the reference temperature is denoted as reference temperature , the temperature compensation coefficient is denoted as temperature compensation coefficient . The anchor rod elastic modulus is denoted as anchor rod elastic modulus , the anchor rod cross-sectional area is denoted as anchor rod cross-sectional area , the anchor rod axial force dispersion value is denoted as anchor rod axial force dispersion value , the anchor rod axial displacement dispersion value is denoted as anchor rod axial displacement dispersion value .

[0051] The inversion coefficient matrix is denoted as inversion coefficient matrix , the observation vector is denoted as observation vector , the second-order difference operator matrix is denoted as second-order difference operator matrix , the regularization matrix is denoted as regularization matrix , the positive regularization coefficient is denoted as positive regularization coefficient . The anchor interface segment equivalent shear stiffness parameter vector is denoted as anchor interface segment equivalent shear stiffness parameter vector , the anchor interface segment equivalent shear stiffness parameter vector under the reference working condition is denoted as reference vector , the preset diagonal weight matrix is denoted as preset diagonal weight matrix .

[0052] The support abnormality sub-item is denoted as support abnormality sub-item , , The support abnormality score is denoted as support abnormality score The gas abnormality score is denoted as gas abnormality score The power abnormality score is denoted as power abnormality score The grid-level coupling risk index is denoted as grid-level coupling risk index . The weighted adjacency matrix is denoted as weighted adjacency matrix . The support threshold, the gas threshold, and the power threshold are respectively denoted as preset threshold , , . The continuous sample number threshold is denoted as preset integer The spatial consistent grid number threshold is denoted as preset integer . The proportion coefficient is denoted as preset proportion coefficient The support-gas coupling time lag is denoted as support-gas coupling time lag .

[0053] The twin parameter vector is denoted as twin parameter vector The Huber type segment robust function threshold is denoted as preset threshold The ventilation setting value is denoted as ventilation setting value , the extraction set value is recorded as extraction set value , the reference set value is recorded as reference set value , , the sensitivity coefficient is recorded as preset sensitivity coefficient , , the upper limit of risk is recorded as preset upper limit of risk , the set value boundary is recorded as , , , . The event chain timing interval end point is recorded as preset interval end point , , , .

[0054] In this embodiment, the three-channel strain will be described in detail as axial equivalent strain sequence and bending component sequence, as follows: The anchor rod distributed monitoring device is arranged at least three strain measuring lines in the same anchor rod section to form a three-channel strain measurement sequence. For example, the three strain measuring lines are arranged at equal intervals along the circumference, and the corresponding circumferential angle is approximately 、 、 , the three-channel strain measurement values are recorded as , , The ground comprehensive analysis server decomposes the three-channel strain measurement sequence to obtain the axial equivalent strain sequence and the bending component sequence.

[0055] The axial equivalent strain measurement value is obtained by three-channel averaging, which satisfies:

[0056] The bending component sequence describes the strain difference caused by the bending of the section by two independent bending components. To avoid introducing uncommon symbols, this embodiment gives a set of implementable calculation methods: define the first bending component measurement value as:

[0057] Define the second bending component measurement value as:

[0058] The first bending component measurement value and the second bending component measurement value jointly constitute a bending component sequence for representing the bending response caused by the uneven circumferential strain of the same section. In the embodiment, the anchoring interface segmented parameter inversion unit only uses the axial equivalent strain measurement value for subsequent temperature compensation and inversion coefficient matrix construction, and the bending component sequence is used for data quality identification, eccentric force working condition identification or display, and does not participate in the inversion coefficient matrix construction.

[0059] In the embodiment, the temperature compensation and example data will be described in detail as follows: The anchoring interface segmented parameter inversion unit performs temperature compensation on the axial equivalent strain measurement value, and the temperature-compensated axial equivalent strain satisfies:

[0060] In the embodiment, the reference temperature is , and the temperature compensation coefficient is . The original data and the temperature compensation results of the example anchor rod in a grid unit A at a certain sampling time are shown in Table 1 (the unit of “ ” is micro-strain).

[0061] Table 1: Original data and temperature compensation results of the example anchor rod (grid unit A, single time)

[0062] In the embodiment, the discrete shear hysteresis force transmission constraint, matrix assembly and inversion solution will be described in detail as follows: In the embodiment, the elastic modulus of the anchor rod is , and the cross-sectional area of the anchor rod is . The discrete value of the axial force of the anchor rod is calculated from the temperature-compensated axial equivalent strain:

[0063] The discrete value of the axial displacement of the anchor rod is obtained by integrating the temperature-compensated axial equivalent strain along the length:

[0064] Wherein . The discrete values of the axial force of the anchor rod and the axial displacement of the anchor rod can be obtained from Table 1 as shown in Table 2 (in Table 2, the axial force of the anchor rod is in kN, and the axial displacement of the anchor rod is in mm, and the calculation is converted according to SI units).

[0065] Table 2: Discrete values of the axial force of the anchor rod and the axial displacement of the anchor rod (grid unit A, single time)

[0066] The number of segments of the segmental equivalent shear stiffness parameter vector of the anchoring interface and the segment boundaries are determined by the way of segment constraint of the support construction record. The support construction record at least includes the start and end depth or start and end position record of the grouting segment, and the start and end depth or start and end position of each grouting segment is projected as an axial coordinate boundary along the anchor rod axis, so that the axial coordinate boundary is taken as a segment boundary set. The embodiment determines the segment boundary set as , forming three segments: 、 、 .

[0067] The anchoring interface segment parameter inversion unit introduces a discrete shear hysteresis force constraint when constructing the inversion coefficient matrix and the observation vector, so that each sampling position satisfies:

[0068] Wherein the segment parameter corresponding to the segmental equivalent shear stiffness parameter vector of the anchoring interface is denoted by . The observation term is defined as: In the embodiment, 6 constraint equations are formed, and is filled in the corresponding column in the inversion coefficient matrix according to the segment to which the sampling point belongs, and the observation term is filled in the observation vector, so that the assembly of the inversion coefficient matrix and the observation vector is completed, and an example is shown in Table 3.

[0069] Example assembly of inversion coefficient matrix A and observation vector b (grid unit A, single time) Table 3

[0070] To suppress the ill-conditioned inversion and meet the constraint of “positive definite regular matrix”, the embodiment adopts a second-order difference regular matrix and satisfies , wherein the second-order difference operator matrix adopts the construction of “second-order difference row + boundary constraint supplementary row”, so that the second-order difference operator matrix is full rank, thereby ensuring is a positive definite matrix. Taking the three segment parameters as an example, the second-order difference operator matrix can be taken as:

[0071] Wherein the first row is used for second-order difference smoothing constraint, and the last two rows are used for boundary endpoint constraint, so that is full rank. Thus, the regular matrix is obtained as:

[0072] The regular coefficient is taken as a positive regular coefficient . The segmental equivalent shear stiffness parameter vector of the anchoring interface satisfies:

[0073] The piecewise equivalent shear stiffness parameter vector of the anchorage interface of mesh element A is obtained by solving as follows:

[0074] The obtained segmented equivalent shear stiffness parameter vector of the anchorage interface is then subjected to nonnegative projection processing to ensure that each component of the vector satisfies the nonnegativity constraint. The segmented equivalent shear stiffness parameter vector of the anchorage interface under the reference working condition is obtained from the statistical analysis of the steady-state samples. In this embodiment, the reference vector is taken as:

[0075] The axial distribution curve of the equivalent shear stiffness of the anchorage interface is shown in the figure. Figure 1 As shown. Figure 1 The x-coordinate represents the axial position of the anchor bolt. (Unit: meters), representing the spatial position along the length of the anchor rod from its exposed end or reference end; the vertical axis represents the equivalent shear stiffness. (The unit is MN / m), which represents the equivalent stiffness of the anchorage interface that transmits shear force within the corresponding axial segment. Figure 1 China and Israel "baseline curve" "" indicates the corresponding axial values ​​of the segmented equivalent shear stiffness parameter vector of the anchorage interface under the reference working condition, used to characterize the interface bonding and force transmission capacity of the support structure when it is in a stable state; "inversion curve" "" indicates the corresponding axial values ​​of the segmented equivalent shear stiffness parameter vector of the anchorage interface obtained by assembling the inversion coefficient matrix and observation vector from the discrete strain measurement sequence and temperature sequence after temperature compensation and discrete shear hysteresis force transmission constraint, and then solving the problem. Figure 1 The inversion curve shows a significant decrease relative to the reference curve in local sections, reflecting a decrease in the equivalent shear stiffness of the anchorage interface in that section, characterizing a weakening of interface bonding or a trend of local instability. The degree of deviation between the reference curve and the inversion curve is used to calculate the support anomaly sub-item, and the low-value continuous coverage section of the inversion curve is used to determine the continuous coverage length of the low-stiffness zone, thus providing input for the support anomaly score and the grid-level coupling risk index.

[0076] In this embodiment, support anomaly scores will be assigned. A detailed description is provided below: When calculating the support anomaly score in the support-gas-dynamic constraint diagram reasoning unit, the support anomaly sub-items are first obtained, and then the support anomaly score is synthesized according to preset non-negative weight coefficients. The preset diagonal weight matrix is:

[0077] One of the anomaly sub-items of the support is the weighted deviation of the equivalent shear stiffness parameter vector of the anchorage interface segment relative to the reference vector, satisfying:

[0078] The second of the supporting abnormal sub-items is the inversion residual norm:

[0079] The third of the supporting abnormal sub-items is the low stiffness belt continuous coverage length. The low stiffness belt is a continuous segment set satisfying in the same roadway space grid cell, and the proportion coefficient is taken as . The segment length of the third embodiment is , and segment 2 and segment 3 satisfy the low stiffness criterion, so the low stiffness belt continuous coverage length is:

[0080] To ensure that the three sub-items can be linearly synthesized and meet the clarity requirement, the preset non-negative weight coefficient of the embodiment is set as a weight coefficient containing a scaling factor, so that the supporting abnormal score after synthesis is a dimensionless quantity. Specifically, the preset non-negative weight coefficient can be written as:

[0081]

[0082]

[0083] wherein , , are dimensionless non-negative coefficients and , , , are statistical reference values in the reference working condition or the sliding statistical window, respectively, for unifying the dimension of the sub-items. Thus, the supporting abnormal score satisfies the following synthesis form:

[0084] At the same time, it is equivalent to synthesis after normalization according to the statistical reference value in numerical implementation. The embodiment takes , , , and takes , , . Substituting the data of the embodiment, the supporting abnormal score is . The supporting threshold is taken as , so the grid cell A at this moment satisfies .

[0085] It should be noted that the gas abnormal score will be described in detail in the embodiment, as follows: The gas parameter monitoring device at least acquires the borehole gas pressure and the borehole gas flow. Taking a borehole associated with a grid unit A as an example, the borehole depth sampling point is taken , the borehole gas pressure is taken MPa. The pressure gradient discrete value is calculated by difference:

[0086] The pressure gradient extreme value term is taken:

[0087] The gas abnormality score obtained in this embodiment is . The borehole gas flow measurement value is taken , the reference flow is taken , wherein the reference flow is a statistical reference value of the same grid unit in a reference working condition or a sliding statistical window, and the flow deviation term is taken:

[0088] The coefficient , is taken, and the gas abnormality score is taken:

[0089] The gas abnormality score obtained is . The gas threshold is taken , and the condition is met. The borehole gas pressure curve is shown in FIG. 1, Figure 2 , Figure 2 The abscissa of FIG. 1 is the borehole depth (unit: meter), which represents the depth position in the direction from the borehole opening to the borehole bottom along the borehole axis; the ordinate is the borehole gas pressure (unit: MPa), which represents the measured gas pressure at the corresponding depth position. Figure 2 The trend of the borehole gas pressure value at different depths with respect to the depth is described by the broken line and the data points, the pressure gradient discrete value is obtained by dividing the pressure difference of adjacent depth points by the depth difference, and the pressure gradient extreme value in the statistical window is extracted as a component of the gas abnormality score. Figure 2 If the local depth section gas pressure increases significantly or the pressure gradient extreme value rises in the embodiment, it indicates that the coal body gas content state or the seepage state changes significantly, which provides gas evidence for the coal and gas outburst risk assessment; meanwhile, Figure 2 the pressure gradient extreme value term and the borehole gas flow deviation term corresponding to are used together to form the gas abnormality score, which is further used in the calculation of the grid level coupling risk index.

[0090] It should be noted that the power abnormality score will be described in detail in this embodiment, as follows: The event information output by the power disturbance monitoring device at least includes event energy and event occurrence time. Taking a 60s statistical window as an example, 3 events are identified, and the event energy is . The event energy cumulative density term and the event rate term are respectively:

[0091]

[0092] Wherein , . Taking the coefficient , , the power anomaly score is:

[0093] The power anomaly score is obtained . The power threshold is taken as , and the condition is met.

[0094] The spatial grid mapping unit obtains the positioning uncertainty parameter corresponding to the event position information, and then distributes a single event to one or more roadway spatial grid units according to the probability: in the grid unit and the adjacent grid unit set in which the event estimated position is located, the distance from the grid center to the event estimated position is used to construct an unnormalized weight, and the distribution probability is obtained after normalization., and the event energy and the event count are weighted and accumulated to the corresponding grid unit statistics according to the distribution probability, and participate in the power anomaly score calculation, so that the power anomaly score reflects the positioning uncertainty.

[0095] It should be noted that in the present embodiment, the weighted adjacency matrix and the grid-level coupling risk index will be described in detail.

[0096] The support-gas-power constraint graph reasoning unit constructs a weighted adjacency matrix with roadway spatial grid units as nodes. The edge weight is determined by the roadway spatial distance and the support construction time difference, and further introduces the geological structure dominant direction parameter, so that the edge weight of the adjacent grid units along the geological structure dominant direction is multiplied by a direction gain coefficient greater than 1, and the edge weight of the transverse adjacent grid units is multiplied by a direction gain coefficient equal to 1. Taking grid unit A and adjacent grid unit B as an example, taking the roadway spatial distance , the support construction time difference , the distance decay scale , the time decay scale , the direction gain coefficient , the edge weight can be taken as:

[0097] The weighted adjacency matrix is assembled from each edge weight.

[0098] The grid-level coupling risk index adopts linear synthesis:

[0099] wherein the preset weight coefficient takes , , , and is substituted into the embodiment , , .

[0100] It should be noted that the spatiotemporal consistency triggering rule and the hierarchical early warning will be described in detail in the embodiment, and specifically as follows: The hierarchical early warning unit outputs the coal and gas outburst coupling early warning information when simultaneously satisfying , , and the spatiotemporal consistency triggering rule in the same grid cell. The spatiotemporal consistency triggering rule is that in the same grid cell, the number of samples respectively continuously satisfying the corresponding threshold condition for the support abnormal score, the gas abnormal score and the dynamic abnormal score are all not less than a preset integer , and at least a preset integer of grid cells spatially adjacent to the grid cell respectively satisfy the condition that the number of continuous samples is not less than in the same time window.

[0101] The embodiment takes , . The continuous sampling score of the grid cell A is shown in Table 4, and the threshold value in Table 4 takes , , .

[0102] The grid cell A continuously sampled score sequence example table 4

[0103] As can be seen from Table 4, the grid cell A continuously satisfies the threshold condition at =2, 3, 4. The grid cell B and the grid cell C spatially adjacent to the grid cell A respectively satisfy the condition that the number of continuous samples is not less than in the same time window, thereby satisfying the spatial consistency grid number threshold . The score and threshold value curve comparison diagram can be referred to Figure 4 . Figure 4 The abscissa of the score and threshold value curve comparison diagram is the sampling serial number ​, represents the sample sequence number obtained by continuous sampling in the same roadway space grid unit under the fixed sampling period; the ordinate is the score value (dimensionless), which is used to represent the value of the support abnormal score, the gas abnormal score, and the dynamic abnormal score on the continuous sample. Figure 4 The curve "support abnormal score " represents the process of support side abnormality degree varying with sampling sequence number, which is synthesized by support abnormal sub-items; the curve "gas abnormal score " represents the process of gas side abnormality degree varying with sampling sequence number, which is composed of pressure gradient extreme value item and flow deviation item, etc.; the curve "dynamic abnormal score " represents the process of dynamic side abnormality degree varying with sampling sequence number, which is composed of event energy cumulative density item and event rate item, etc. Figure 4 In the figure, the horizontal dashed lines respectively represent support threshold , gas threshold , and dynamic threshold , which are used for threshold judgment with the corresponding score curves; when the three score curves in the same grid unit all continuously meet the respective threshold conditions and the number of continuous samples is not less than a preset integer , Figure 4 the time continuity condition of "continuous threshold satisfaction" is embodied; and when at least a preset integer of the spatially adjacent grid units also meet the corresponding continuous sample number conditions in the same time window, the spatiotemporal consistency triggering rule can be further met. Figure 4 Therefore, the threshold judgment basis and continuity constraint conditions for intuitively expressing the hierarchical early warning triggering are used to provide visual criterion support for the output of the coal and gas outburst coupling early warning information.

[0104] It should be noted that in the present embodiment, the support-gas coupling time lag and coupling edge weight will be described in detail as follows: The support-gas-dynamic constraint graph reasoning unit determines the coupling edge weight between the support abnormal score and the gas abnormal score by using the time lag mutual information maximization rule. Taking a preset time lag set as an example, the support abnormal score sequence and the gas abnormal score sequence in the statistical window are discretized and binned, and the joint probability and the marginal probability are counted, and the mutual information can be calculated according to the following formula:

[0105] The time lag that makes the mutual information take the maximum value is selected from the preset time lag set as the support-gas coupling time lag:

[0106] The calculation result of the present embodiment is shown in Table 6, and the support-gas coupling time lag is obtained as .

[0107] Instance calculation result of mutual information changing with time lag Table 6

[0108] To use the mutual information result as the coupling edge weight, the embodiment adopts an implementable normalization mapping method: taking the support-gas coupling edge weight as the normalized value of the maximum mutual information value relative to the preset upper bound, or the normalized value relative to the maximum value in the time lag set. Exemplarily, the support-gas coupling edge weight can be taken as:

[0109] Thus, the support-gas coupling time lag is obtained as and output to the hierarchical early warning unit for time sequence alignment and constraint.

[0110] It should be noted that the event chain time sequence constraint will be described in detail in the embodiment, as follows: The hierarchical early warning unit further satisfies the event chain time sequence constraint before outputting the coal and gas outburst coupling early warning information. The event chain time sequence constraint includes the time when the support abnormal score first reaches the support threshold, the time when the gas abnormal score first reaches the gas threshold, and the time when the power abnormal score first reaches the power threshold, satisfying:

[0111]

[0112] The embodiment takes , , , , and takes the example time as:

[0113]

[0114]

[0115] The support abnormal score falls into , and the gas abnormal score falls into , so that the event chain time sequence constraint is established.

[0116] In addition, to show the change of risk over time and the difference from the comparative scheme, the embodiment gives a schematic diagram as shown in Figure 3 , in which the existing technology in the background technology of scheme A is used to form a comprehensive danger index and perform threshold triggering comparison. Figure 3 The abscissa of the schematic diagram is time (unit: minute), represents the statistical time or the sliding window center time advancing with time in the continuous monitoring process; the vertical coordinate is the risk index (dimensionless), which is used to represent the coupling risk level of the roadway space grid unit in the same time window synthesized by the support abnormal score, the gas abnormal score and the dynamic abnormal score. Figure 3 In the present embodiment The "present embodiment" curve represents the change of the grid-level coupling risk index with time, which is linearly synthesized by the support abnormal score , the gas abnormal score , and the dynamic abnormal score with preset weight coefficients; the "comparison scheme AP" curve represents the change of the risk index with time obtained under the comparison scheme. Figure 3 The comparison of the two curves in the same coordinate system shows the change trend, peak occurrence period and persistence difference of the risk index, wherein the persistent increase, peak enhancement or persistent residence near the threshold value of the risk index can be used as the input of the grading warning trigger criterion, Figure 3 The comparison relationship shown is used to illustrate the difference in risk index output under the same monitoring condition and different risk assessment strategies, thereby supporting the risk quantization input required for the warning trigger logic, event chain timing constraint and linkage control decision.

[0117] It should be noted that in the present embodiment, the digital twin consistency correction and online updating K0 or Q are described in detail as follows: The ground comprehensive analysis server comprises a digital twin consistency correction unit. The digital twin consistency correction unit updates the twin parameter vector according to the mismatch vector of the measured strain vector and the twin predicted strain vector, and applies parameter upper and lower bound constraints to the twin parameter vector. In order to suppress the influence of abnormal components of the mismatch vector, the present embodiment adopts a Huber type segmented robust function:

[0118] wherein the preset threshold is 0.12. The digital twin consistency correction unit can update the twin parameter vector in an iterative manner with projection, and after updating, the twin parameter vector is used for online updating of the reference vector or the diagonal weight matrix.

[0119] For ease of understanding, the present embodiment gives two implementable updating methods: Method one: online updating of reference vector When the stable trigger condition is met, the statistical median vector of the segment equivalent shear stiffness parameter vector sample of the anchoring interface in the recent sliding window is obtained by taking the statistical median value of each segment, and the reference vector is updated by exponential smoothing:

[0120] wherein the smoothing coefficient The embodiment takes The stable trigger condition can be taken as: the twin parameter vector changes at a rate lower than a preset threshold within a continuous sampling time, and the embodiment takes .

[0121] Method two: online update of the diagonal weight matrix

[0122] Let the diagonal element of the diagonal weight matrix be When the stable trigger condition is met, the diagonal element is proportionally corrected according to the corresponding component of the twin parameter vector and projected to a preset range:

[0123] wherein is the correction gain, , and the lower and upper bounds of the diagonal element. The embodiment takes The online updated reference vector or the online updated diagonal weight matrix is returned to the support-gas-dynamic constraint graph reasoning unit to participate in subsequent support anomaly score and grid-level coupling risk index calculation.

[0124] It should be noted that the executable solving rules of the linkage control interface unit under boundary constraints are described in detail in the embodiment, as follows: The system further includes a linkage control interface unit for solving the ventilation set value and the extraction set value when outputting the coal and gas outburst coupling early warning information. The predicted risk index after linkage meets:

[0125] and the target function

[0126] is the minimum target, under the constraints

[0127]

[0128]

[0129] solving.

[0130] To make the solving with boundary constraints directly executable, the embodiment adopts the rule of “unconstrained solution-projection-correction”: 1) First, ignore the boundary constraints to solve the minimum norm increment , , and obtain the candidate solution:

[0131]

[0132] 2) Boundary projection is performed on the candidate solution to obtain a projected solution:

[0133]

[0134] 3) If the projected solution has satisfied , output , . If the projected solution does not satisfy , perform single-variable correction: If is located at the boundary (equal to or ), solve the extraction set value according to the boundary of the constraint equation while keeping unchanged:

[0135] and perform boundary projection on to obtain ; If is located at the boundary, solve the ventilation set value according to the boundary of the constraint equation while keeping unchanged:

[0136] and perform boundary projection on to obtain .

[0137] If still cannot be satisfied after single-variable correction, output the limit solution (for example, , ) that is feasible on the boundary, and carry a "constraint unsatisfiable flag" in the linkage control interface output field to prompt the scheduling or equipment capacity deficiency.

[0138] In addition, in the present embodiment, the abnormal data and missing data processing rules will be described in detail as follows: When the strain measurement of a sampling position of the anchor rod distributed monitoring device is missing or exceeds the preset abnormal threshold, the ground comprehensive analysis server uses linear interpolation of adjacent sampling positions or the last time retention strategy to complete the sampling position, and removes or reduces the weight of the constraint equation corresponding to the missing position when assembling the inversion coefficient matrix and the observation vector; when the gas parameter monitoring device has a short-time missing, the nearest effective value is retained and the missing section is marked, and the missing section does not participate in the probability estimation of the mutual information statistical window; when the dynamic disturbance monitoring device has a short-time disconnection, the dynamic abnormal score is treated as zero event in the disconnection section and triggers the system self-checking prompt, so that the system still has a stable calculation process under complex downhole working conditions.

[0139] It should be noted that the embodiments of the present application have better implementation, and do not limit the present application in any form. Any skilled person in the art can change or modify the above disclosed technical content into equivalent effective embodiments without departing from the technical scheme of the present application. Any modification or equivalent change and modification of the above embodiments according to the technical essence of the present application still belongs to the scope of the technical scheme of the present application.

Claims

1. An online monitoring and early warning system for the status of anchor bolt support in mine roadways, characterized in that, include: A distributed monitoring device for anchor bolts is installed on multiple anchor bolts in a mine roadway to acquire discrete strain measurement sequences and corresponding temperature sequences for each anchor bolt along its length. A gas parameter monitoring device is installed in the mine roadway to acquire at least two gas monitoring sequences among gas volume fraction, borehole gas pressure, and borehole gas flow rate. A dynamic disturbance monitoring device is installed in the corresponding area of ​​the mine roadway to acquire a dynamic monitoring sequence formed by microseismic event information or acoustic emission event information. A data aggregation and timing device is used to perform unified timing synchronization on the discrete strain measurement sequence, the temperature sequence, the gas monitoring sequence, and the power monitoring sequence, and generate a unified timestamp. A ground-based integrated analysis server, communicatively connected to the data aggregation and timing device, comprises: Spatiotemporal grid mapping unit is used to map each sequence after a unified timestamp to a preset lane space grid unit; The anchorage interface piecewise parameter inversion unit is used to construct the inversion coefficient matrix based on the discrete strain measurement sequence and temperature sequence within each roadway spatial grid cell. With observation vector And solve for the piecewise equivalent shear stiffness parameter vector of the anchorage interface. ,in satisfy: in, for The transpose of the matrix, For a predefined positive definite regularized matrix, The regularization coefficient is positive; Support-gas-dynamic constraint graph reasoning unit, used to construct a weighted adjacency matrix with roadway spatial grid units as nodes. And based on the segmented equivalent shear stiffness parameter vector of the anchoring interface. Calculate support anomaly score Calculate gas anomaly score based on gas monitoring sequence Calculate dynamic anomaly score based on dynamic monitoring sequence And calculate the grid-level coupling risk index. ,in: in, , , Preset weighting coefficients; A graded early warning unit is used within the same roadway spatial grid unit when both support anomaly scores are met. Gas abnormality score abnormal power score And when the spatiotemporal consistency triggering rule is met, output coal and gas outburst coupling early warning information; among which , , The preset threshold is used; the spatiotemporal consistency triggering rule is: within the same roadway spatial grid unit, the support anomaly score... Gas abnormality score abnormal power score The number of samples that consecutively meet the corresponding threshold conditions is not less than a preset integer. And at least a preset integer adjacent to the spatial grid unit of the alleyway. Each laneway spatial grid cell satisfies the requirement that the corresponding number of consecutive samples is not less than [a certain value] within the same time window. conditions; The early warning release terminal is used to receive and release early warning information on coal and gas outburst coupling and its roadway spatial grid unit identifier.

2. The online monitoring and early warning system for the status of mine roadway anchor bolt support according to claim 1, characterized in that, The anchor bolt distributed monitoring device sets at least three strain measurement lines circumferentially along the same anchor bolt cross-section to form a three-channel strain measurement sequence. The anchorage interface segmented parameter inversion unit solves the three-channel strain measurement sequence into an axial equivalent strain sequence and a bending component sequence, and only uses the axial equivalent strain sequence to construct the inversion coefficient matrix. With the observation vector .

3. The online monitoring and early warning system for the status of mine roadway anchor bolt support according to claim 2, characterized in that, The anchorage interface segmented parameter inversion unit performs temperature compensation on the axial equivalent strain sequence, and the temperature-compensated axial equivalent strain... satisfy: in, These are the axial equivalent strain measurements. This is a temperature measurement value. As the reference temperature, The temperature compensation coefficient is used; and the temperature-compensated axial equivalent strain sequence is used to construct the inversion coefficient matrix. With observation vector ; The segmented parameter inversion unit of the anchoring interface constructs the inversion coefficient matrix. With the observation vector Discrete shear hysteresis force transmission constraints are introduced at each sampling position to ensure that... satisfy: in, The discrete value of the anchor bolt axial force is obtained by converting the temperature-compensated axial equivalent strain sequence, and , The elastic modulus of the anchor bolt. This represents the cross-sectional area of ​​the anchor bolt. The discrete value of axial displacement is obtained by integrating the temperature-compensated axial equivalent strain sequence along its length, and , The distance between adjacent sampling positions; To segment the equivalent shear stiffness parameter vector of the anchorage interface The corresponding piecewise parameters; and the inversion coefficient matrix is ​​obtained by assembling the coefficient terms constrained at all sampling positions as described above. With the observation vector .

4. The online monitoring and early warning system for the status of mine roadway anchor bolt support according to claim 1, characterized in that, The anchorage interface segment parameter inversion unit determines the equivalent shear stiffness parameter vector of the anchorage interface segment using a segmented constraint method based on support construction records. The number of segments and segment boundaries, the support operation record includes at least the start and end depths or start and end positions of the grouting segments, and the start and end depths or start and end positions of each grouting segment are projected along the anchor rod axis as axial coordinate boundaries, thereby using the axial coordinate boundaries as the segment boundary set. The regular matrix Use a second-order difference regularized matrix and satisfy ,in It is a second-order difference operator matrix; and the anchorage interface piecewise parameter inversion unit solves for the equivalent shear stiffness parameter vector of the anchorage interface piecewise. Perform nonnegative projection processing to segment the anchorage interface into equivalent shear stiffness parameter vectors. Each component satisfies the nonnegativity constraint.

5. The online monitoring and early warning system for the status of mine roadway anchor bolt support according to claim 1, characterized in that, The support-gas-dynamic constraint diagram reasoning unit calculates the support anomaly score. At that time, the support anomaly sub-item is obtained first. , , Then, the support anomaly sub-items are combined according to preset weights to form the support anomaly score. ,in: The support anomaly sub-item The equivalent shear stiffness parameter vector of the anchoring interface is segmented. Relative reference vector The weighted deviation satisfies: in For the preset diagonal weight matrix, The segmented equivalent shear stiffness parameter vector of the anchorage interface under the reference working condition; The support anomaly sub-item For inversion of residual norm ; The support anomaly sub-item The continuous coverage length of the low-stiffness zone is defined as the length of the low-stiffness zone within the same roadway spatial grid cell that satisfies... A continuous set of segments, This is a preset proportional coefficient. and They are vectors with vector The Quantity; The support abnormality score satisfy: in, , , These are preset non-negative weighting coefficients.

6. The online monitoring and early warning system for the status of mine roadway anchor bolt support according to claim 1, characterized in that, The event information output by the dynamic disturbance monitoring device includes at least the event energy and the time of event occurrence, and the support-gas-dynamic constraint diagram reasoning unit calculates the dynamic anomaly score. The time frame includes at least an event energy accumulation density term and an event rate term within a unit time window; and the spatiotemporal grid mapping unit acquires the positioning uncertainty parameter corresponding to the event location information, and allocates a single event to one or more roadway spatial grid units according to probability based on the positioning uncertainty parameter before participating in the dynamic anomaly score. The calculation.

7. The online monitoring and early warning system for the status of mine roadway anchor bolt support according to claim 1, characterized in that, The weighted adjacency matrix The edge weights are determined by the spatial distance of the roadway and the time difference of the support construction. Furthermore, the dominant geological structure direction parameter is introduced, so that the edge weights of adjacent roadway spatial grid units along the dominant geological structure direction are multiplied by a directional gain coefficient greater than 1, while the edge weights of adjacent roadway spatial grid units in the transverse direction are multiplied by a directional gain coefficient equal to 1. The dominant geological structure direction parameter is given by the roadway geological data or tunneling measurement data.

8. The online monitoring and early warning system for the status of mine roadway anchor bolt support according to claim 1, characterized in that, Before outputting coal and gas outburst coupled early warning information, the graded early warning unit further satisfies event chain timing constraints, which include: the support anomaly score. First time reaching the support threshold The moment The gas anomaly score First time reaching the gas threshold The moment The aforementioned dynamic anomaly score First time reaching the power threshold The moment satisfy Falling into the preset range ,and Falling into the preset range .

9. The online monitoring and early warning system for the status of mine roadway anchor bolt support according to claim 5, characterized in that, The ground-based integrated analysis server also includes a digital twin consistency correction unit, which updates the twin parameter vector based on the mismatch vector between the measured strain vector and the twin predicted strain vector. and using the updated twin parameter vector For the reference vector Or the diagonal weight matrix Perform online updates and send the updated baseline vectors. Or the updated diagonal weight matrix The data is fed back to the support-gas-dynamic constraint diagram reasoning unit to participate in the subsequent support anomaly scoring. Risk index of coupling with the grid level The calculation.

10. The online monitoring and early warning system for the status of mine roadway anchor bolt support according to claim 9, characterized in that, The digital twin consistency correction unit updates the twin parameter vector. Parameter upper and lower bound constraints are applied, and a Huber-type piecewise robust function is used to suppress the influence of outlier components in the mismatch vector. satisfy: in, The preset threshold, and the updated twin parameter vector Used for the reference vector or diagonal weight matrix Online updates.

Images