An intelligent self-monitoring anchor rod system and method suitable for deep large-span high-stress mining roadway

Abstract

The application discloses an intelligent self-monitoring anchor rod system and method suitable for deep large-span high-stress mining roadway, and belongs to the technical field of artificial intelligence and data processing. The system comprises a data acquisition module arranged on an anchor rod body and used for acquiring multi-source monitoring data representing a stress state and / or a deformation state of the anchor rod in real time; a data processing module in communication connection with the data acquisition module, used for receiving the multi-source monitoring data, and performing preprocessing, time series modeling, feature extraction and state evolution analysis on the multi-source monitoring data to construct an anchor rod state model; and a state evaluation module in communication connection with the data processing module, used for performing hierarchical determination on a current working state of the anchor rod according to the anchor rod state model and outputting. The application can improve the systematicness, continuity and intelligent level of the anchor rod working state monitoring while realizing the anchor rod supporting function, and can provide a more reliable technical means for deep high-stress roadway supporting safety management and decision-making.

Description

Technical Field

[0001] This invention relates to the fields of artificial intelligence and data processing technology, and more specifically to an intelligent self-monitoring anchor bolt system and method suitable for deep, large-span, high-stress mining roadways. Background Technology

[0002] As coal mines and underground engineering continue to advance into deeper areas, mining roadways are increasingly characterized by greater depth, longer spans, and higher levels of ground stress. Under the combined effects of the high-stress environment at depth and mining disturbances, the surrounding rock of the roadways deforms significantly, exhibiting complex failure modes. The anchor bolt support structure endures high loads and repeated disturbances over long periods, posing a severe challenge to its safety and stability.

[0003] Currently, rock bolts are one of the most widely used support components in mining roadways, primarily providing active support force to constrain surrounding rock deformation and improve overall rock stability. Existing rock bolt designs often focus on load-bearing capacity and structural strength, making it difficult to obtain real-time data on their operational status during service. When rock bolts experience abnormal stress or when slippage or failure occurs at the anchoring interface, there is often a lack of effective real-time sensing methods, easily leading to support failure, which is characterized by its concealment and suddenness.

[0004] To understand the deformation and failure characteristics of the surrounding rock in roadways, engineering projects often employ methods such as surface displacement monitoring, delamination monitoring, or stress monitoring. However, these monitoring devices are typically installed separately from the bolt support system, making it difficult for the monitoring information to directly reflect the synergistic stress relationship between the bolts and the surrounding rock. Furthermore, the monitoring results suffer from insufficient representativeness and delayed response. Under complex mining conditions, a single monitoring parameter is insufficient to accurately characterize the stability of the surrounding rock and the anchoring status of the bolts. In addition, the surrounding rock in deep, long-span mining roadways often exhibits significant time-dependent deformation and nonlinear failure characteristics, with the bolt stress state constantly evolving over time and under mining conditions. Existing support and monitoring methods largely rely on empirical judgment and post-hoc analysis, lacking continuous, real-time monitoring methods for the working state of the bolts, making it difficult to provide reliable data support for support parameter optimization and safety early warning.

[0005] Therefore, how to ensure the load-bearing capacity of anchor bolts while achieving real-time monitoring of their stress and anchorage status, thereby improving the safety and intelligence of the support system in deep, high-stress mining roadways, has become an urgent technical problem to be solved. To this end, various excitation devices capable of generating different forms of impact loads have been developed, such as: Chinese patent CN201010618198.1 discloses a fiber-reinforced plastic smart anchor bolt that uses fiber optic gratings to transmit temperature and strain signals for long-term testing of the anchor bolt's environmental temperature and strain state. While this technology improves the ability to acquire anchor bolt stress information to some extent, it primarily focuses on monitoring a single sensor or a single physical quantity, lacking a systematic analysis of the overall working state of the anchor bolt and failing to accurately reflect the cooperative stress relationship between the anchored section and the free section.

[0006] Chinese patent CN201820691231.5 invented a multi-information intelligent advanced anchor bolt device. By integrating sensor devices into the anchor bolt, it attempts to realize the information acquisition function of the advanced anchor bolt for surrounding rock stability evaluation and safety early warning. It mainly focuses on the sensor layout and information acquisition body, and does not have a complete data transmission, real-time analysis and early warning system overall architecture.

[0007] Chinese patent CN202411526732.4 discloses an early warning method and threshold determination method for an intelligent anchor bolt system. This patent relates to an early warning method and threshold setting mechanism for an intelligent anchor bolt system. It achieves early warning by determining a monitoring threshold, but focuses on the "early warning method" itself and does not fully cover the comprehensive collection, analysis and system integration structure of anchor bolt monitoring data.

[0008] In summary, existing impact load excitation devices and technologies each have their own characteristics, but their main limitations are as follows: (1) Most technical solutions are limited to sensor embedding or data acquisition body, and have not formed a complete system architecture to support real-time analysis, early warning and visualization; (2) Lack of collaborative analysis and anchorage status determination mechanism for monitoring data between the free section and the anchorage section of the anchor bolt; (3) There is no publicly available technology that can fully integrate the perception layer, transmission layer, processing layer, early warning layer and display layer to make it a complete intelligent monitoring system for deep, high-stress, large-span mining roadways.

[0009] Therefore, how to provide an intelligent self-monitoring anchor bolt system and method suitable for deep, large-span, high-stress mining roadways is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0010] In view of this, the present invention provides an intelligent self-monitoring anchor bolt system and method suitable for deep, large-span, high-stress mining roadways. By integrating support and monitoring functions, the system enables real-time perception and analysis of the anchor bolt stress state and anchoring stability, thereby improving the safety and reliability of the roadway support system.

[0011] To achieve the above objectives, the present invention adopts the following technical solution: An intelligent self-monitoring bolt system suitable for deep, large-span, high-stress mining roadways includes: The data acquisition module is installed on the anchor bolt body and is used to acquire multi-source monitoring data that characterizes the stress state and / or deformation state of the anchor bolt in real time. The data processing module is communicatively connected to the data acquisition module and is used to receive the multi-source monitoring data and perform preprocessing, time series modeling, feature extraction and state evolution analysis on the multi-source monitoring data to construct an anchor bolt state model. The status assessment module is communicatively connected to the data processing module and is used to classify and output the current working status of the anchor bolt according to the anchor bolt status model.

[0012] Furthermore, the data acquisition module includes: Stress sensors and / or strain sensors are installed in the anchorage section of the anchor bolt; Strain sensors deployed in the free section of the anchor bolt; Displacement sensors are installed at the head of anchor bolts.

[0013] Furthermore, the data processing module includes: The multi-source monitoring data is subjected to sliding window weighted denoising, and Gaussian weights are used to smooth the data within the window; Outliers are identified and corrected based on the rate of change between adjacent time points; Consistency deviation is calculated for monitoring data from multiple similar sensors, and sensor weights are dynamically adjusted based on the deviation values. Different types of monitoring parameters are normalized by interval, and the normalization interval is dynamically adjusted according to the rate of parameter change. Time axis unification, linear interpolation synchronization, and time delay compensation are performed on monitoring data with different sampling frequencies.

[0014] Furthermore, the data processing module also includes

[0015] Construct the state vector of the anchor rod at time t, the state vector including the normalized stress of the anchorage section, the strain of the free section and the displacement of the rod head; The anchor bolt's comprehensive health index is calculated based on the state vector, and the anchor bolt's status is classified according to the comprehensive health index. The classification includes four levels: normal, attention, warning, and danger.

[0016] Furthermore, the state assessment module includes: The weights of each monitoring parameter in the calculation of the comprehensive health index are adaptively updated based on the rate of change of the monitoring data. The early warning threshold is adaptively adjusted based on the sliding window mean and standard deviation of the monitoring data. The sensitivity coefficient of the state discrimination model is dynamically adjusted based on the service time and / or cumulative bearing history of the anchor bolts.

[0017] Furthermore, it also includes: The data transmission module is used to upload the multi-source monitoring data and / or the early warning information to a cloud server or a ground monitoring center.

[0018] Furthermore, it also includes: The visualization module is used to display the working status of the anchor bolts, historical data curves, and early warning information in real time.

[0019] A system for an intelligent self-monitoring anchor bolt system suitable for deep, large-span, high-stress mining roadways includes: Acquire multi-source monitoring data of anchor bolts during their service life; The multi-source monitoring data is preprocessed, including noise reduction, outlier correction, and data normalization. Based on the processed monitoring data, a time series model of the working state of the anchor bolt is constructed, and state feature parameters are extracted. Based on the variation law of the state characteristic parameters and the multi-parameter coupling relationship, the working state of the anchor bolt is classified and output.

[0020] As can be seen from the above technical solution, compared with the prior art, this invention provides an intelligent self-monitoring anchor bolt system and method suitable for deep, large-span, high-stress mining roadways. By integrating the anchor bolt support structure with the monitoring function, the anchor bolt, while bearing the supporting function of the surrounding rock, can acquire its own working status and the surrounding rock environment information in real time, thereby realizing continuous monitoring of the anchor bolt's stress and deformation state, improving the completeness and reliability of the support component's state perception. This invention, by deploying multiple sensing units at key stress points of the anchor bolt, realizes multi-parameter acquisition of the anchor bolt's axial stress, deformation, and its evolution process. Combined with a data acquisition and processing module, the monitoring data is analyzed in real time, effectively avoiding the problem of insufficient information caused by relying on only a single parameter or local monitoring in the prior art. This helps to more accurately reflect the overall stress characteristics of the anchor bolt and the stability state of the surrounding rock. By constructing a monitoring data transmission and processing system, this invention can realize real-time uploading, centralized analysis, and long-term storage of the anchor bolt's working status, transforming anchor bolt monitoring from post-event detection to process monitoring and trend analysis, providing continuous and traceable data support for roadway surrounding rock stability assessment and support effect determination. This invention also enables the identification and graded early warning of abnormal stress states through comprehensive analysis of anchor bolt monitoring data. This allows for the early detection of support risks, facilitating timely implementation of corresponding support reinforcement or parameter adjustment measures by on-site management personnel. This reduces the safety risk of support failure in deep, large-span, high-stress mining roadways. Furthermore, the intelligent self-monitoring anchor bolt system described in this invention features a high degree of structural integration and is compatible with conventional anchor bolt construction techniques. It does not significantly impact existing roadway support processes, exhibiting excellent engineering adaptability and suitability for widespread application in complex geological conditions and long-term service environments.

[0021] In summary, while realizing the function of anchor bolt support, this invention significantly improves the systematicness, continuity and intelligence of anchor bolt working status monitoring, and can provide more reliable technical means for safety management and decision-making in deep high-stress roadway support. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of the system structure provided by the present invention; Figure 2 This is a schematic diagram of the method flow provided by the present invention. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] See Figure 1 This invention discloses an intelligent self-monitoring anchor bolt system suitable for deep, large-span, high-stress mining roadways, comprising: The data acquisition module is installed on the anchor bolt body and is used to acquire multi-source monitoring data that characterizes the stress state and / or deformation state of the anchor bolt in real time. The data processing module is communicatively connected to the data acquisition module and is used to receive the multi-source monitoring data and perform preprocessing, time series modeling, feature extraction and state evolution analysis on the multi-source monitoring data to construct an anchor bolt state model. The status assessment module is communicatively connected to the data processing module and is used to classify and output the current working status of the anchor bolt according to the anchor bolt status model.

[0026] In one specific embodiment, the data acquisition module includes: Stress sensors and / or strain sensors are installed in the anchorage section of the anchor bolt; Strain sensors deployed in the free section of the anchor bolt; Displacement sensors are installed at the head of anchor bolts.

[0027] In one specific embodiment, the data processing module includes: The multi-source monitoring data is subjected to sliding window weighted denoising, and Gaussian weights are used to smooth the data within the window; Outliers are identified and corrected based on the rate of change between adjacent time points; Consistency deviation is calculated for monitoring data from multiple similar sensors, and sensor weights are dynamically adjusted based on the deviation values. Different types of monitoring parameters are normalized by interval, and the normalization interval is dynamically adjusted according to the rate of parameter change. Time axis unification, linear interpolation synchronization, and time delay compensation are performed on monitoring data with different sampling frequencies.

[0028] Furthermore, the data processing module also includes

[0029] Construct the state vector of the anchor rod at time t, the state vector including the normalized stress of the anchorage section, the strain of the free section and the displacement of the rod head; The anchor bolt's comprehensive health index is calculated based on the state vector, and the anchor bolt's status is classified according to the comprehensive health index. The classification includes four levels: normal, attention, warning, and danger.

[0030] Furthermore, the state assessment module includes: The weights of each monitoring parameter in the calculation of the comprehensive health index are adaptively updated based on the rate of change of the monitoring data. The early warning threshold is adaptively adjusted based on the sliding window mean and standard deviation of the monitoring data. The sensitivity coefficient of the state discrimination model is dynamically adjusted based on the service time and / or cumulative bearing history of the anchor bolts.

[0031] Specifically, the mathematical methods for denoising multi-source monitoring data are as follows: Let the first k Each anchor rod at time A certain monitoring parameter was collected.

[0032] 1. Sliding window weighted noise reduction: A sliding window of length 2M+1 was applied to the original data. The denoised data is as follows:

[0033] The weighting coefficients satisfy:

[0034] In one specific embodiment, the weights are distributed using a Gaussian distribution:

[0035] in, k Number the anchor bolt (the kth anchor bolt); For the first i Each sampling time, i =1,2,…, N ; N This represents the total number of sampling points (time series length). x k ( t i ) is the first k anchor bolt at time t i The collected raw monitoring parameter values; 2M+1 is the total length of the sliding window (odd number); M is the half length of the window, that is, M points extending to the left and right from the center of the window; For the first k anchor bolt at time t i Monitoring parameter values ​​after noise reduction and smoothing; jThis is a relative index within the window, with a range of values. j = -M, -(M - 1), …, 0, …, M - 1, M ; w j For the first in the window j The weight coefficients at each position satisfy the following conditions: The sum of all weights is 1 (normalization); 0 >w ±1 > >w ±M The weight decreases as the distance from the center of the window increases; r `r` is the bandwidth parameter of the Gaussian kernel, controlling the rate of weight decay (the smaller `r` is, the faster the weight decays, and the "sharper" the window); `m` is the summation index, used to iterate through all points within the window. -M, -(M - 1), …, 0, …, M - 1, M ; is the Gaussian kernel function, used to calculate unnormalized weights; This is a normalization factor that ensures the sum of all weights is 1.

[0036] 2. Outlier Detection and Correction: Define the rate of change between adjacent time points:

[0037] in, Indicates the first k The anchor bolt at the first i Each sampling time The rate of change between adjacent time points; When the following conditions are met:

[0038] in If the rate of change is determined based on the anchor bolt material and design parameters, then this point is considered an outlier.

[0039] Outlier correction:

[0040] 3. Multi-sensor consistency constraints: Assuming the same anchor bolt is installed A number of similar sensors, with a noise reduction value of [value missing]. Calculate the consistency deviation:

[0041] like:

[0042] The weight of this sensor data in subsequent analysis is then reduced to:

[0043] in, n The total number of similar sensors deployed for the same anchor bolt; j For the first j Each sensor is numbered; For the first k The first anchor bolt j The noise-reduced monitoring values ​​of each sensor; For the first k The first anchor bolt m The noise-reduced monitoring values ​​of each sensor; for n The average value of the monitored values ​​after noise reduction of each sensor; d j For the first j Consistency deviation of individual sensors; This is the consistency deviation threshold, used to determine whether sensor data is abnormal; α j For the first j The weight of each sensor in subsequent analysis; η This is the weight attenuation coefficient, used to reduce the weight of abnormal sensors.

[0044] In one specific embodiment, the multi-parameter normalization mathematical method is as follows: The anchor monitoring parameters include the stress data σ of the anchorage section, the strain data ε of the free section, and the displacement data u of the rod head.

[0045] 1. Interval normalization: For any parameter Normalization

[0046] in, The minimum and maximum values ​​of historical data are determined by design limits or historical statistics.

[0047] 2. Adaptive dynamic range adjustment: When the rate of change is monitored to meet the following:

[0048] Then dynamically scale the normalized interval:

[0049] in, For parameters p(t) Rate of change over time; γ p This is the threshold for the rate of change of parameters, used to trigger dynamic range adjustment; β The range scaling factor satisfies 0 <β <0.5; p′ max This is the dynamically adjusted upper limit of the interval; p′ min This is the dynamically adjusted lower limit of the interval.

[0050] 3. Parameter weighted normalization: Set weights for different parameters ,satisfy:

[0051] in, For anchorage section stress data σ Normalized weights; For free segment strain data ε Normalized weights; For pole head displacement data u Normalized weights.

[0052] In one specific embodiment, the mathematical method for multi-source data time synchronization is as follows: Suppose the sampling time series of different sensors are as follows:

[0053] 1. Unified timeline: Establish a unified timeline:

[0054] in, t 0 This is the starting point of the timeline; Δt To standardize the time step; L This represents the total number of time points on the timeline. T This is the unified set of timelines.

[0055] 2. Interpolation synchronization: For unaligned data, linear interpolation is used:

[0056] in, s Number the sensors; t For the target interpolation time; The time is the left endpoint of the interpolation interval; The time is the right endpoint of the interpolation interval; For the first s One sensor in t The interpolation result at time step.

[0057] 3. Delay compensation: Let the communication delay be... Correct timestamp:

[0058] In one specific embodiment, the anchor bolt state model construction method

[0059] 1. Multi-parameter state vector: Constructing anchors at any time State vector:

[0060] 2. Comprehensive Status Indicators: Define the comprehensive health index of anchor bolts:

[0061] in, m The number of parameters in the state vector (here) m =3); w i For the first i The weights of each state parameter; X i (t) For the first i Each state parameter in t The normalized value at time; H(t) For anchor bolts in t Comprehensive health indicators at all times.

[0062] 3. State determination:

[0063] Specifically, dynamic adjustment methods include: 1. Weights are dynamically updated: The weights are adaptively adjusted based on the rate of change.

[0064] in, w i (t) For the first i Each state parameter in t Adaptive weights at any given time; X i (t) For the first i Each state parameter in t The normalized value at time; m This represents the total number of state parameters.

[0065] 2. Adaptive adjustment of early warning threshold:

[0066] in, The mean within the sliding window. For comprehensive health indicators H(t) standard deviation , The threshold adjustment coefficient satisfies... .

[0067] 3. Sensitivity adjustment: Define the model sensitivity coefficient:

[0068] The final state discrimination index is:

[0069] in, For comprehensive health indicators H(t) Rate of change over time; S(t) These are the model sensitivity coefficients; This serves as a criterion for determining the final state.

[0070] In one specific embodiment, it further includes: The data transmission module is used to upload the multi-source monitoring data and / or the early warning information to a cloud server or a ground monitoring center.

[0071] In one specific embodiment, it further includes: The visualization module is used to display the working status of the anchor bolts, historical data curves, and early warning information in real time.

[0072] See Figure 2 A system for intelligent self-monitoring anchor bolts suitable for deep, large-span, high-stress mining roadways, comprising: Acquire multi-source monitoring data of anchor bolts during their service life; The multi-source monitoring data is preprocessed, including noise reduction, outlier correction, and data normalization. Based on the processed monitoring data, a time series model of the working state of the anchor bolt is constructed, and state feature parameters are extracted. Based on the variation law of the state characteristic parameters and the multi-parameter coupling relationship, the working state of the anchor bolt is classified and output.

[0073] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0074] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An intelligent self-monitoring anchor bolt system suitable for deep, large-span, high-stress mining roadways, characterized in that, include: The data acquisition module is installed on the anchor bolt body and is used to acquire multi-source monitoring data that characterizes the stress state and / or deformation state of the anchor bolt in real time. The data processing module is communicatively connected to the data acquisition module and is used to receive the multi-source monitoring data and perform preprocessing, time series modeling, feature extraction and state evolution analysis on the multi-source monitoring data to construct an anchor bolt state model. The status assessment module is communicatively connected to the data processing module and is used to classify and output the current working status of the anchor bolt according to the anchor bolt status model.

2. The intelligent self-monitoring anchor bolt system suitable for deep, large-span, high-stress mining roadways according to claim 1, characterized in that, The data acquisition module includes: Stress sensors and / or strain sensors are installed in the anchorage section of the anchor bolt; Strain sensors deployed in the free section of the anchor bolt; Displacement sensors are installed at the head of anchor bolts.

3. The intelligent self-monitoring anchor bolt system suitable for deep, large-span, high-stress mining roadways according to claim 1, characterized in that, The data processing module includes: The multi-source monitoring data is subjected to sliding window weighted denoising, and Gaussian weights are used to smooth the data within the window; Outliers are identified and corrected based on the rate of change between adjacent time points; Consistency deviation is calculated for monitoring data from multiple similar sensors, and sensor weights are dynamically adjusted based on the deviation values. Different types of monitoring parameters are normalized by interval, and the normalization interval is dynamically adjusted according to the rate of parameter change. Time axis unification, linear interpolation synchronization, and time delay compensation are performed on monitoring data with different sampling frequencies.

4. The intelligent self-monitoring anchor bolt system suitable for deep, large-span, high-stress mining roadways according to claim 3, characterized in that, The data processing module also includes Construct the state vector of the anchor rod at time t, the state vector including the normalized stress of the anchorage section, the strain of the free section and the displacement of the rod head; The anchor bolt's comprehensive health index is calculated based on the state vector, and the anchor bolt's status is classified according to the comprehensive health index. The classification includes four levels: normal, attention, warning, and danger.

5. The intelligent self-monitoring anchor bolt system suitable for deep, large-span, high-stress mining roadways according to claim 4, characterized in that, The status assessment module includes: The weights of each monitoring parameter in the calculation of the comprehensive health index are adaptively updated based on the rate of change of the monitoring data. The early warning threshold is adaptively adjusted based on the sliding window mean and standard deviation of the monitoring data. The sensitivity coefficient of the state discrimination model is dynamically adjusted based on the service time and / or cumulative bearing history of the anchor bolts.

6. The intelligent self-monitoring anchor bolt system suitable for deep, large-span, high-stress mining roadways according to claim 1, characterized in that, Also includes: The data transmission module is used to upload the multi-source monitoring data and / or the early warning information to a cloud server or a ground monitoring center.

7. The intelligent self-monitoring anchor bolt system suitable for deep, large-span, high-stress mining roadways according to claim 1, characterized in that, Also includes: The visualization module is used to display the working status of the anchor bolts, historical data curves, and early warning information in real time.

8. A system utilizing the intelligent self-monitoring anchor bolt system as described in any one of claims 1 to 7, applicable to deep, large-span, high-stress mining roadways, characterized in that, include: Acquire multi-source monitoring data of anchor bolts during their service life; The multi-source monitoring data is preprocessed, including noise reduction, outlier correction, and data normalization. Based on the processed monitoring data, a time series model of the working state of the anchor bolt is constructed, and state feature parameters are extracted. Based on the variation law of the state characteristic parameters and the multi-parameter coupling relationship, the working state of the anchor bolt is classified and output.

Images