Soft rock roof reinforcement control system and method based on multi-stage coupling regulation

Abstract

The application discloses a soft rock roof reinforcement control system and method based on multi-stage coupling regulation, and relates to the technical field of soft rock support. The method comprises the following steps: S1, collecting roof monitoring data in real time and performing data preprocessing; S2, extracting internal evolution characteristics and crack hysteresis activity characteristics of the roof, comprehensively evaluating internal stability of the roof, and constructing a crack hysteresis statistical feature set; S3, identifying a disturbance event, evaluating disturbance sensitivity of the roof by combining response changes of the roof monitoring data before and after the disturbance event, and determining a state of the roof; and S4, comprehensively judging a regulation level, generating a regulation instruction, and outputting a regulation strategy and a reinforcement suggestion. The method solves the problem that potential instability risks of a soft rock roof are difficult to identify in time after short-time apparent stability of the soft rock roof after reinforcement, and thus leads to early degradation of a support strategy and secondary instability.

Description

Technical Field

[0001] This invention relates to the field of soft rock support technology, specifically to a soft rock roof reinforcement control system and method based on multi-stage coupling adjustment. Background Technology

[0002] Soft rock roof reinforcement and support technologies are widely used in coal mining and underground engineering. Under complex geological conditions and multiple disturbances, the surrounding rock structure exhibits significant spatiotemporal evolution characteristics due to factors such as mining stress, hydrological conditions, and construction activities. With the expansion of engineering scale and the diversification of operating environments, roof stability control has gradually evolved from single support methods to multi-faceted collaborative regulation, encompassing multiple aspects such as grouting reinforcement, anchor bolt and cable support, and monitoring and evaluation. Through continuous observation and analysis of the mechanical response and structural change processes of the surrounding rock, related technologies are constantly evolving towards refinement and dynamism to meet the demands of safe and efficient operation in underground engineering.

[0003] For example, Chinese patent CN115749881B discloses a method for roof support in water-rich soft rock roadways, relating to the field of soft rock support. It solves the problem that temporary support is required during anchor drilling for rock bolt support, which is time-consuming and labor-intensive due to manual construction. Manual construction is also prone to misalignment of the temporary support device, leading to uneven support force. The proposed method for roof support in water-rich soft rock roadways includes the following steps: first, adjusting the support spacing and applying support to both sides of the soft rock roadway; then, connecting and self-locking the main support frame; next, anchor drilling; and finally, anchoring and repositioning. This method for roof support in water-rich soft rock roadways saves time and labor during support operations. Through centering and correction support, the coordination between the support frame and the main support frame ensures uniform support force and prevents accidents during anchor drilling.

[0004] For example, Chinese patent CN119825414B discloses a method for treating soft rock roofs using advanced steel rod support and negative pressure guided grouting reinforcement. This method falls under the field of coal mining, and its key technical points are: During tunnel excavation, this invention employs "steel rods + negative pressure guided grouting reinforcement" for advanced support. The negative pressure guided grouting improves the grouting effect on both micro and deep cracks, enhancing the stability of the soft rock roof and ensuring the integrity of the roof at the working face. Because of the use of negative pressure guided grouting, the spacing range of the steel rods can also be increased.

[0005] However, after reinforcement, soft rock roofs often exhibit apparent stability, such as short-term displacement convergence and stabilization of anchor bolt axial forces. This state may only be a temporary equilibrium, with internal cracks still slowly expanding or the load-bearing structure not yet fully formed. Subsequent mining disturbances, equipment vibrations, or environmental changes can cause the roof to rapidly destabilize again. Existing methods often rely on instantaneous indicators to assess stability, making it difficult to identify such "pseudo-stability" states in a timely manner. This can lead to premature reduction of support strength or cessation of reinforcement, thus creating potential safety hazards.

[0006] Therefore, in order to address the above problems, there is an urgent need for a soft rock roof reinforcement control system and method based on multi-stage coupling regulation. Summary of the Invention

[0007] Technical problems to be solved

[0008] To address the shortcomings of existing technologies, this invention provides a soft rock roof reinforcement control system and method based on multi-stage coupling adjustment, which solves the problem that soft rock roofs appear stable in the short term after reinforcement but the potential instability risk is difficult to identify in time, leading to premature downgrading of support strategies and secondary instability.

[0009] Technical solution

[0010] To achieve the above objectives, the present invention provides the following technical solution: a soft rock roof reinforcement and control method based on multi-stage coupled regulation, comprising the following steps: S1, real-time acquisition of roof monitoring data, and time alignment, anomaly removal, and normalization processing of the roof monitoring data; S2, based on the preprocessed roof monitoring data, extraction of roof internal evolution characteristics and fracture hysteresis activity characteristics, comprehensive evaluation of roof internal stability, and construction of a fracture hysteresis statistical feature set based on roof internal stability; S3, identification of disturbance events using roof monitoring data, and evaluation of roof disturbance sensitivity by combining the response changes of roof monitoring data before and after the disturbance event, and roof state determination by combining roof disturbance sensitivity and fracture hysteresis statistical feature set; S4, fusion of roof internal stability and roof disturbance sensitivity, comprehensive judgment of regulation level, generation of regulation instructions based on regulation level, and output of regulation strategy and reinforcement suggestions by combining regulation instructions and roof state.

[0011] Furthermore, the specific process of real-time acquisition of roof monitoring data, and the time alignment, anomaly removal, and normalization of the roof monitoring data, is as follows: Real-time acquisition of roof monitoring data includes: acquiring surface displacement values ​​and deep delamination displacement values ​​of the roof using roof displacement gauges, multi-point displacement gauges, and delamination meters; acquiring axial force values ​​of all monitored anchor bolts using anchor bolt force gauges; acquiring microseismic event records and acoustic emission signals using a microseismic monitoring system and acoustic emission sensors. The microseismic event records include the occurrence time and energy of the microseismic events, and are statistically analyzed within a sliding time window to obtain the number of microseismic events, total microseismic energy, and acoustic emission ringing count within the current time window; acquiring vibration acceleration signals of the area where the roof is located using vibration sensors; and processing the roof monitoring data according to a unified time reference. Timestamps are aligned, and resampling is performed according to a fixed sampling period. For the roof surface displacement value, deep delamination displacement value, and axial force value of each monitored anchor bolt, the median value and median absolute deviation of the corresponding data within the current time window are calculated. If the absolute difference between the sampled value and the median value exceeds the multiple threshold of the median absolute deviation, it is judged as abnormal data and removed. For missing data, linear interpolation is used to complete it. Min-max normalization is performed on the roof monitoring data. For the roof surface displacement value and the deep delamination displacement value, the current sampled value is subtracted from the previous sampled value and then divided by the sampling time interval to obtain the roof surface subsidence rate and the deep delamination change rate. The mean and standard deviation of the axial force values ​​of all monitored anchor bolts are calculated to obtain the mean and standard deviation of the anchor bolt axial force.

[0012] Furthermore, based on the preprocessed roof monitoring data, the specific process for extracting the internal evolution characteristics and fracture hysteresis activity characteristics of the roof and comprehensively evaluating the internal stability of the roof is as follows: Read the preprocessed roof monitoring data; divide the current deep delamination rate by the sum of the roof surface subsidence rate and the smallest positive number to obtain the deep-surface response difference ratio; perform hyperbolic tangent function calculation on the deep-surface response difference ratio to obtain the deep-surface response modulation term; multiply the number of microseismic events within the current time window by the total microseismic energy, and take the square root of the product to obtain the fracture energy frequency value; add the fracture energy frequency value to the acoustic emission ringing count, then add it to one, and take the natural logarithm to obtain the internal fracture activity intensity term; multiply the deep-surface response modulation term by the internal fracture activity intensity term to obtain the fracture hysteresis activity value.

[0013] Furthermore, the specific process of constructing a statistical feature set of crack hysteresis based on the internal stability of the roof is as follows: according to the sampling time sequence, the crack hysteresis activity values ​​corresponding to each sampling time are continuously stored to form a sequence of crack hysteresis activity values. Based on the sliding time window, the mean and standard deviation of the crack hysteresis activity values ​​are calculated to obtain the mean and standard deviation of the crack hysteresis activity values.

[0014] Further, the specific process of identifying a disturbance event using roof monitoring data is as follows: calculate the root mean square value of the vibration acceleration signal within a sliding time window to obtain a vibration intensity value, and compare the vibration intensity value with an intensity threshold; when the vibration intensity value changes from not exceeding the intensity threshold to exceeding the intensity threshold, and exceeds the intensity threshold at N consecutive sampling moments, the moment when the intensity threshold is exceeded for the first time is determined as the disturbance event moment; taking the disturbance event moment as the center, intercept a time series of length T forward to construct a pre-disturbance time window, and intercept a time series of the same length backward to construct a post-disturbance time window.

[0015] Further, the specific process of evaluating roof disturbance sensitivity by combining the response changes of roof monitoring data before and after the disturbance event is as follows: based on the pre-disturbance time window and the post-disturbance time window, calculate the respective mean values of the roof surface subsidence rate to obtain a pre-disturbance subsidence rate value and a post-disturbance subsidence rate value, calculate the respective mean values of the bolt axial force to obtain a pre-disturbance axial force base value and a post-disturbance axial force base value, and calculate the respective mean values of the number of microseismic events in the current time window to obtain a pre-disturbance microseismic value and a post-disturbance microseismic value; subtract the pre-disturbance subsidence rate value from the post-disturbance subsidence rate value, retain only the part greater than zero, and then divide the result by the sum of the pre-disturbance subsidence rate value and a minimal positive number to obtain the relative increase of the roof surface subsidence rate; subtract the pre-disturbance axial force base value from the post-disturbance axial force base value, retain only the part greater than zero, and then divide the result by the sum of the pre-disturbance axial force base value and a minimal positive number to obtain the relative increase of the bolt axial force; subtract the pre-disturbance microseismic value from the post-disturbance microseismic value, retain only the part greater than zero, and then divide the result by the sum of the pre-disturbance microseismic value and a minimal positive number to obtain the relative increase of the number of microseismic events; calculate the sum of squares of the relative increase of the roof surface subsidence rate, the relative increase of the bolt axial force and the relative increase of the number of microseismic events, and take the square root of the sum of squares to obtain a disturbance sensitive response value.

[0016] Further, the specific process of judging the roof state by combining the roof disturbance sensitivity and the fracture hysteresis statistical feature set is as follows: subtract the mean value of the fracture hysteresis activity from the current fracture hysteresis activity value, and divide the result by the sum of the standard deviation of the fracture hysteresis activity value and a minimal positive number to obtain a fracture hysteresis activity deviation; the disturbance sensitive response value is compared with the disturbance threshold S1, and the fracture hysteresis activity deviation is compared with the deviation threshold Z1, then the roof state label is output; when ≥S1, and ≥Z1, it is determined that the current roof is in a risk coupling response state; when ≥S1, and <Z1, it is determined that the current roof is in a disturbance sensitive response state; when <S1, and ≥Z1, it is determined that the current roof is in an internal fracture evolution state; when <S1, and When it is less than Z1, it is determined that the current roof is in a stable response state.

[0017] Further, by integrating the internal stability of the roof and the disturbance sensitivity of the roof, the specific process for comprehensively judging the adjustment level is as follows: divide the standard deviation of the current bolt axial force by the sum of the mean value of the bolt axial force and a very small positive number to obtain the support force dispersion coefficient, perform a hyperbolic tangent function operation on the support force dispersion coefficient and then add one to obtain a force balance amplification term; multiply the disturbance sensitive response value by the crack hysteresis activity value at the corresponding moment, add a very small positive number, and take the square root to obtain a coupling response value, add one to the coupling response value and then take the natural logarithm to obtain a comprehensive risk adjustment term; multiply the comprehensive risk adjustment term by the force balance amplification term to obtain a stage coupling adjustment judgment value.

[0018] Further, the specific process of generating an adjustment instruction according to the adjustment level, and outputting an adjustment strategy and reinforcement suggestion in combination with the adjustment instruction and the roof state is as follows: compare the stage coupling adjustment judgment value with an adjustment threshold: when the stage coupling adjustment judgment value is less than the adjustment threshold, output a conventional adjustment instruction; when the stage coupling adjustment judgment value is greater than or equal to the adjustment threshold, output an enhanced adjustment instruction; read the roof state label, when the roof state label is a stable response state, if the output is a conventional adjustment instruction, output a control instruction to the support control device to keep the current support parameters unchanged, and keep the sampling frequency of roof monitoring data unchanged; if the output is an enhanced adjustment instruction, output a control instruction to increase the sampling frequency of roof monitoring data and the identification frequency of disturbance events, and recheck the current roof state; when the roof state label is a risk coupling response state, if the output is a conventional adjustment instruction, output a control instruction to the support control device to keep the current set value of bolt pre-tightening force from decreasing, lock the support parameter adjustment interface, and simultaneously output a control instruction to increase the sampling frequency of roof monitoring data; if the output is an enhanced adjustment instruction, output parameter suggestions for increasing the number of arranged bolts and improving the local support density to the support control device, and output adjustment suggestions for performing grouting reinforcement on the corresponding area to the grouting equipment; when the roof state label is a disturbance sensitive response state, if the output is a conventional adjustment instruction, output a control instruction to increase the sampling frequency of roof monitoring data and the identification frequency of disturbance events; if the output is an enhanced adjustment instruction, output a control instruction to increase the sampling frequency of roof monitoring data, and output parameter suggestions for improving the local support stiffness and increasing the number of bolts to the support control device; when the roof state label is an internal crack evolution state, if the output is a conventional adjustment instruction, output a control instruction to the support control device to extend the holding time of the current support parameters and maintain the set value of bolt pre-tightening force from decreasing; if the output is an enhanced adjustment instruction, output parameter suggestions for increasing the number of bolts to the support control device, and output adjustment suggestions for performing grouting reinforcement to the grouting equipment.

[0019] The second aspect of this invention provides a soft rock roof reinforcement control system based on multi-stage coupled regulation, comprising: a monitoring data acquisition and processing module for real-time acquisition of roof monitoring data, performing time alignment, anomaly removal, and normalization on the roof monitoring data; a multi-stage response feature construction module for extracting internal roof evolution features and fracture hysteresis activity features based on preprocessed roof monitoring data, comprehensively evaluating the internal stability of the roof, and constructing a fracture hysteresis statistical feature set based on the internal stability of the roof; a pseudo-stability identification and evaluation module for identifying disturbance events using roof monitoring data, evaluating roof disturbance sensitivity by combining the response changes of roof monitoring data before and after the disturbance event, and determining the roof state by combining the roof disturbance sensitivity and the fracture hysteresis statistical feature set; and a multi-stage coupled regulation execution module for fusing the internal stability of the roof and the roof disturbance sensitivity, comprehensively judging the regulation level, generating regulation commands based on the regulation level, and outputting regulation strategies and reinforcement suggestions by combining the regulation commands and the roof state.

[0020] Beneficial effects

[0021] The present invention has the following beneficial effects: (1) This invention eliminates the differences in time scale and dimension of multi-source monitoring data such as roof displacement, anchor bolt axial force, micro-vibration and vibration by performing unified time alignment, anomaly removal and normalization processing on the multi-source monitoring data, thereby improving the consistency and stability of the monitoring data, thus providing a reliable data basis for subsequent roof condition analysis and improving the overall assessment accuracy.

[0022] (2) This invention constructs the internal evolution characteristics of the roof and the hysteresis activity characteristics of the fractures, and forms a set of statistical features of fracture hysteresis, thereby achieving a synergistic characterization of the difference between the "apparent convergence" and the "internal fracture evolution" of the roof. This can effectively characterize the hidden evolution process inside the soft rock roof and improve the ability to identify hidden unstable behaviors.

[0023] (3) This invention evaluates the sensitivity of the roof to disturbance by combining the response changes before and after the disturbance event, and makes a joint judgment with the crack hysteresis statistical feature set, so as to accurately distinguish the risk coupling response state, the disturbance sensitive response state and the internal crack evolution state, avoid the misjudgment problem caused by relying solely on instantaneous indicators, and improve the reliability of roof stability state identification.

[0024] (4) This invention integrates the internal evolution characteristics of the roof and the disturbance sensitivity to comprehensively judge the adjustment level, and combines the roof state output graded adjustment strategy and reinforcement suggestions to realize closed-loop control from state identification to adjustment execution. It can trigger adjustment measures in advance before the potential instability risk of the roof occurs, prevent premature reduction of support strength, and improve the safety and adaptability of the roof reinforcement process.

[0025] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0026] Figure 1 A flowchart of a soft rock roof reinforcement and control method based on multi-stage coupling regulation; Figure 2 This is a structural diagram of a soft rock roof reinforcement control system based on multi-stage coupling regulation. Figure 3 This is a schematic diagram illustrating the time-series evolution of the disturbance-sensitive response value; Figure 4 This is a schematic diagram comparing the components of the disturbance-sensitive response value. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. As those skilled in the art will understand, 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.

[0028] Please see Figures 1-4 This invention provides a technical solution: a method for reinforcing and controlling soft rock roofs based on multi-stage coupled regulation, such as... Figure 1 As shown, the process includes the following steps: S1, real-time acquisition of roof monitoring data, and time alignment, anomaly removal, and normalization of the roof monitoring data; S2, based on the preprocessed roof monitoring data, extraction of internal roof evolution characteristics and fracture hysteresis activity characteristics, comprehensive evaluation of internal roof stability, and construction of a fracture hysteresis statistical feature set based on internal roof stability; S3, identification of disturbance events using roof monitoring data, and evaluation of roof disturbance sensitivity by combining the response changes of roof monitoring data before and after the disturbance event, and determination of roof state by combining roof disturbance sensitivity and fracture hysteresis statistical feature set; S4, fusion of internal roof stability and roof disturbance sensitivity, comprehensive judgment of regulation level, generation of regulation instructions based on regulation level, and output of regulation strategy and reinforcement suggestions by combining regulation instructions and roof state.

[0029] Specifically, the real-time acquisition of roof monitoring data, and the subsequent time alignment, anomaly removal, and normalization of the roof monitoring data, are as follows: Real-time acquisition of roof monitoring data includes: collecting surface displacement values ​​and deep delamination displacement values ​​of the roof using roof displacement gauges, multi-point displacement gauges, and delamination meters. Roof displacement gauges are preferably deployed in key deformation areas of the roof. Multi-point displacement gauges are used to obtain displacement information at different depths, and delamination meters are used to monitor delamination changes between roof layer structures, thus forming a multi-level displacement data source; collecting axial force values ​​of all monitored anchor bolts using anchor bolt force gauges. Anchor bolt force gauges are evenly distributed along the support area to cover different stress sections, ensuring the spatial representativeness of the axial force data; collecting microseismic event records and acoustic emission signals using a microseismic monitoring system and acoustic emission sensors. Microseismic event records include the occurrence time and energy of microseismic events, and are statistically analyzed within a sliding time window to obtain the number of microseismic events, total microseismic energy, and acoustic emission ring count within the current time window. The window length is set according to actual monitoring needs, preferably between 10 and 300 seconds, and is updated in a fixed step size, preferably between 1 and 10 seconds, to ensure the timeliness and stability of the statistical results. The microseismic monitoring system is a monitoring device composed of multiple seismic detectors deployed in the roof and surrounding rock. It is used to collect elastic wave signals generated by microfractures in the rock mass, and to identify microseismic events based on signal triggering and waveform analysis. It outputs the occurrence time and energy of the microseismic events. The occurrence time of the microseismic events is recorded based on a unified time reference, and the event energy is calculated by the microseismic monitoring system based on the energy integration of the waveform signal to ensure the consistency and comparability of the event energy representation. The acoustic emission ringing count is obtained by counting the acoustic emission signals based on a preset amplitude threshold, which is used to characterize the frequency of microfracture activity. Vibration acceleration signals in the area where the roof is located are collected by vibration sensors. The vibration sensors are preferably triaxial accelerometers and are fixedly installed on the roof to obtain stable and reliable vibration response data. The roof monitoring data is timestamped according to a unified time base and resampled according to a fixed sampling period. The unified time base uses the system standard clock, and the resampling period is preferably 0.1 seconds to 1 second to achieve consistency of multi-source data in the time dimension. For roof surface displacement values, deep delamination displacement values, and axial force values ​​of each monitored anchor bolt, the median value and median absolute deviation of the corresponding data within the current time window are calculated. If the absolute difference between the sampled value and the median value exceeds the multiple threshold of the median absolute deviation, it is judged as abnormal data and removed. The multiple threshold is preferably 3 to 5 to achieve a balance between the sensitivity of abnormal removal and data integrity. For missing data, linear interpolation is used to complete it. The roof monitoring data is subjected to min-maximum normalization processing. The minimum and maximum values ​​are obtained based on the data statistics within the sliding time window to adapt to the changes in data scale at different stages.For the roof surface displacement and deep delamination displacement values, the current sampled value is subtracted from the previous sampled value, and then divided by the sampling time interval to obtain the roof surface subsidence rate and the deep delamination change rate. The sampling time interval is kept consistent with the resampling period to ensure consistency in rate calculation. The mean and standard deviation of all monitored anchor bolt axial force values ​​are calculated to obtain the mean and standard deviation of the anchor bolt axial force. Through these steps, standardized processing and unified representation of multi-source monitoring data are achieved, providing a reliable input data foundation for subsequent feature construction and condition assessment.

[0030] In this implementation plan, by uniformly collecting, aligning, removing anomalies, and normalizing multi-source monitoring data of the roof, consistent representation of different types of data in terms of time and scale is achieved, effectively eliminating noise interference and data bias, and improving the stability and reliability of the data. At the same time, by constructing statistical characteristics of displacement rate and anchor bolt force, the ability to characterize changes and stress state of the roof structure is enhanced, thereby providing an accurate, continuous, and comparable data foundation for subsequent feature analysis and risk assessment.

[0031] Specifically, based on the preprocessed roof monitoring data, the process of extracting the internal evolution characteristics and fracture hysteresis activity characteristics of the roof, and comprehensively evaluating the internal stability of the roof, is as follows: Read the preprocessed roof monitoring data, divide the current deep delamination rate by the sum of the roof surface subsidence rate and the minimum positive number to obtain the deep-to-surface response difference ratio. This ratio is dimensionless. Both the deep delamination rate and the roof surface subsidence rate are calculated based on a uniform sampling time interval, and the minimum positive number is preferably selected as [value missing]. arrive To avoid numerical instability when the denominator approaches zero, a hyperbolic tangent function is applied to the deep surface response difference ratio to obtain the deep surface response modulation term. The hyperbolic tangent function nonlinearly compresses the deep surface response difference ratio, stabilizing its value within a finite interval, thus avoiding excessive amplification of outliers in subsequent calculations, while maintaining the dimensionless nature of the deep surface response modulation term. The number of microseismic events within the current time window is multiplied by the total microseismic energy, and the square root of the product is taken to obtain the fracture energy frequency value. This value comprehensively characterizes the frequency and energy intensity of fracture activity. The fracture energy frequency value is added to the acoustic emission ringing count, then added to one, and the natural logarithm is taken to obtain the internal fracture activity intensity term. The number of microseismic events, total microseismic energy, and acoustic emission ringing count are all dimensionless data after normalization, therefore the calculation results are also dimensionless. The addition operation is used to avoid zero or negative values ​​in the logarithmic operation, thus ensuring the validity and continuity of the calculation results. Multiplying the deep surface response modulation term with the internal fracture activity intensity term yields the fracture hysteresis activity value. This value comprehensively reflects the coupling relationship between the apparent deformation response of the roof and the evolution of internal fractures, thereby achieving a quantitative characterization of the internal stability state of the roof. The fracture hysteresis activity value is a dimensionless comprehensive index that constrains the difference between deep delamination changes and surface subsidence responses through nonlinear modulation and couples it with the fracture activity level characterized by microseismic energy and acoustic emission activity. This index is used to characterize the hysteresis characteristics of the continued development of internal fractures under relatively stable surface conditions, thus reflecting the potential instability risk of the roof. The evolution of internal fractures in the roof typically precedes surface deformation, manifesting as an inconsistency between deep and surface responses, accompanied by enhanced microseismic and acoustic emission activity. Therefore, by characterizing the "deep-surface response difference" as the ratio of the deep delamination rate to the surface subsidence rate, the degree of asynchrony between the internal structure and external deformation can be depicted. Furthermore, the combination of the number and energy of microseismic events, as well as acoustic emission ringing counts, are introduced to characterize the frequency and intensity of fracture activity. Finally, through nonlinear compression and product coupling, the value is significantly increased only when "internal fracture activity and deep-surface response mismatch" coexist, thus effectively characterizing the hysteretic evolution characteristics of fractures.

[0032] The specific formula for the crack hysteresis activity value is as follows: ; In the formula, It represents the value of hysteresis activity in the fracture, which is used to comprehensively characterize the intensity of the continuous evolution of internal fractures in the roof under apparent convergence conditions, and to reflect the degree of inconsistency between deep evolution and surface response, thereby identifying "pseudo-stability" characteristics. This indicates the number of microseismic events, characterizing the frequency of fracture activity; the larger the value, the more active the fracture activity. It represents the total energy of microseismic events, characterizing the intensity level of fracture activity. The larger the value, the more severe the fracture propagation or damage. The value represents the acoustic emission ringing count, which characterizes the microscopic activity of microcrack friction and propagation. The larger the value, the more obvious the continuous evolution of internal microcracks. It represents the rate of change of deep delamination, characterizing the speed of deep evolution of the internal structure of the roof. The larger the value, the faster the internal structure changes. This represents the rate of subsidence of the roof surface, characterizing the macroscopic deformation response of the roof surface layer. The larger the value, the more obvious the layer deformation. To represent extremely small positive numbers, prevent the denominator from being zero, and ensure computational stability, the preferred value is [value missing]. arrive .

[0033] In this implementation scheme, by introducing a coupled modeling method of deep surface response difference and fracture activity intensity, the inconsistency between the delamination evolution inside the roof and the surface deformation response is quantitatively characterized. Combined with the comprehensive characterization of fracture activity frequency and intensity by microseismic and acoustic emission signals, the implicit evolution process inside the roof is effectively identified. At the same time, through nonlinear compression and product coupling mechanism, the results are significantly enhanced only when the internal fractures are active and the apparent response is lagging, thereby improving the ability to distinguish the risk coupling response state and potential instability risk, and enhancing the accuracy and stability of the assessment results.

[0034] Specifically, the process of constructing a statistical feature set of crack hysteresis based on the internal stability of the roof is as follows: according to the sampling time order, the crack hysteresis activity values ​​corresponding to each sampling time are continuously stored to form a crack hysteresis activity value sequence. The crack hysteresis activity value sequence is sorted and stored according to a uniform sampling time interval and corresponds one-to-one with the corresponding timestamp to ensure the temporal consistency of subsequent statistical analysis. Based on a sliding time window, the mean and standard deviation of fracture hysteresis activity values ​​are calculated to obtain the mean and standard deviation of fracture hysteresis activity values. The sliding time window length is preferably 10 to 300 seconds, and updates are performed with a fixed step size, preferably 1 to 10 seconds, to improve the responsiveness to short-term changes while ensuring the stability of statistical results. The mean is used to characterize the overall level of fracture hysteresis activity within the current time window, and the standard deviation is used to characterize the fluctuation of fracture hysteresis activity values, thus reflecting the stability and dispersion of the fracture evolution process. Furthermore, the mean and standard deviation of fracture hysteresis activity values ​​together constitute a fracture hysteresis statistical feature set, which is used for deviation calculation and risk assessment in the subsequent roof condition determination process.

[0035] In this implementation scheme, by storing the hysteresis activity values ​​of the fractures in a time series and extracting the mean and standard deviation based on a sliding time window, a stable characterization of the fracture evolution level and its fluctuation characteristics is achieved. This can effectively reflect the overall trend and dispersion of the evolution inside the roof, thereby providing a continuous and reliable statistical basis for subsequent deviation judgment and state identification, and improving the accuracy and robustness of roof stability assessment.

[0036] Specifically, the process of identifying disturbance events using roof monitoring data is as follows: The root mean square (RMS) value of the vibration acceleration signal within a sliding time window is calculated to obtain the vibration intensity value. The vibration acceleration signal is preferably a triaxial acceleration signal. The composite vibration acceleration amplitude is obtained by squaring, summing, and square-taking the acceleration signals in the three directions. The RMS value is then calculated within a sliding time window, preferably 10 to 60 seconds, to improve the characterization of multi-directional vibration responses. The vibration intensity value is compared with an intensity threshold, which is preferably determined based on the statistical distribution of vibration intensity values ​​under historical normal operating conditions to improve the adaptability and reliability of the judgment. When the vibration intensity value changes from not exceeding the intensity threshold to exceeding the intensity threshold, and exceeds the intensity threshold for N consecutive sampling times, a disturbance event is determined to have occurred, and the moment the intensity threshold is first exceeded is determined as the disturbance event moment. Here, N is a preset number of consecutive sampling points, preferably 3 to 10, to avoid misjudgments caused by single-point anomalies and instantaneous noise. Based on this, a consistency check is performed on the identified disturbance events: taking the time of the disturbance event as the center, within a check time window of length τ (preferably 1 to 10 seconds), the increment of the number of microseismic events ΔNms and the change in the roof surface settlement rate Δvs within the time window are calculated respectively; where ΔNms is the difference between the number of microseismic events within the time window and the number of microseismic events in the adjacent equal-length time window before the disturbance event, and Δvs is the difference between the average roof surface settlement rate within the time window and the average roof surface settlement rate in the adjacent equal-length time window before the disturbance event; when ΔNms ≥ N1 or Δvs ≥ V1, the disturbance event is confirmed as a valid disturbance event; when ΔNms < N1 and Δvs < V1, the disturbance event is judged as a non-structural disturbance and is removed; where N1 is the threshold for the change in the number of microseismic events and V1 is the threshold for the change in the roof surface settlement rate, both of which are preferably determined based on the statistical distribution of historical normal operation data. Centered on the moment of the disturbance event, time series of length T are truncated forward to construct the pre-disturbance time window, and time series of the same length are truncated backward to construct the post-disturbance time window; where T represents the truncated time length, preferably 10 to 120 seconds, and the pre-disturbance time window and the post-disturbance time window have the same length to ensure the symmetry and comparability of the before and after response comparison; the time series are truncated based on a uniform sampling period to ensure the consistency of various monitoring data in the time dimension.

[0037] In this implementation scheme, disturbance events are identified based on the root mean square value of vibration acceleration signals, and combined with a continuous threshold judgment mechanism, which effectively avoids misjudgment caused by instantaneous noise interference. At the same time, by constructing a symmetrical time window before and after the disturbance, comparable analysis of the changes in response before and after the disturbance is realized, thereby improving the accuracy and stability of disturbance event identification and providing a reliable time positioning basis for subsequent disturbance sensitivity assessment.

[0038] Specifically, the process for assessing the roof disturbance sensitivity by combining the response changes of roof monitoring data before and after the disturbance event is as follows: Based on the time window before and after the disturbance, the average value of the roof surface subsidence rate is calculated to obtain the pre-disturbance and post-disturbance subsidence velocity values. The average value of the anchor bolt axial force is also calculated to obtain the pre-disturbance and post-disturbance baseline axial force values, where the average anchor bolt axial force is the average of the axial force values ​​of all currently monitored anchor bolts. The average value of the number of microseismic events within the current time window is also calculated to obtain the pre-disturbance and post-disturbance microseismic values. The pre-disturbance and post-disturbance time windows have the same length and are both based on a unified sampling period for data extraction to ensure comparability of the data before and after the disturbance. The average value is calculated by taking the arithmetic mean of the data from all sampling points within the window to reduce the impact of instantaneous fluctuations on the results. The difference between the post-disturbance settlement velocity and the pre-disturbance settlement velocity, retaining only the positive portion, is then divided by the sum of the pre-disturbance settlement velocity and the smallest positive number to obtain the relative increase in the roof surface settlement rate. This relative increase in the roof surface settlement rate characterizes the amplification of the roof surface deformation response after the disturbance; the larger the value, the more significant the deformation of the roof surface under the disturbance. The difference between the post-disturbance axial force baseline value and the pre-disturbance axial force baseline value, retaining only the positive portion, is then divided by the sum of the pre-disturbance axial force baseline value and the smallest positive number to obtain the relative increase in the anchor bolt axial force. The relative increase is used to characterize the degree of stress change in the support structure after disturbance. The larger the value, the stronger the stress response of the support system, reflecting the enhanced interaction between the surrounding rock and the support. The difference between the post-disturbance microseismic value and the pre-disturbance microseismic value, retaining only the positive part, and then dividing by the sum of the pre-disturbance microseismic value and the minimum positive number, yields the relative increase in the number of microseismic events. This relative increase in the number of microseismic events characterizes the activity level of fractures after disturbance; the larger the value, the more frequent the internal fracture propagation and rupture activity. The minimum positive number is preferably taken as [value missing]. arrive To avoid numerical instability when the denominator approaches zero, only the part greater than zero is retained. This is achieved by comparing the calculation result with 0; if the result is greater than 0, it is retained, otherwise it is set to 0. The retention of only the part greater than zero is used to highlight the risk amplification effect corresponding to the enhanced response after the disturbance. This invention focuses on the process of the system response evolving from a stable state to an unstable state after the disturbance is triggered. The enhanced response is usually positively correlated with the risk of structural instability. For the case where the response decreases after the disturbance compared to before the disturbance, it is preferred to consider that the corresponding structure tends to be in a relaxed or recovered state, and it is not used as the basis for judging the enhanced disturbance sensitivity in this invention, thereby avoiding the mutual cancellation of different physical quantities in the direction of increase or decrease, which would interfere with the risk assessment results. Furthermore, this invention uses a multi-source response synthesis mechanism for evaluation. Even if a single indicator decreases, its impact can be reflected by the increase of other response components. The sum of squares of the relative increases in roof surface settlement rate, anchor bolt axial force, and the number of microseismic events was calculated, and the square root of the sum of squares was taken to obtain the disturbance-sensitive response value. The square root of the sum of squares was used to achieve comprehensive fusion of multi-source responses, ensuring that each component participates in the calculation on a uniform scale, thus obtaining a quantitative result reflecting the overall degree of roof's response to disturbance. Roof instability under disturbance typically manifests as a simultaneous increase in surface deformation, support stress, and internal crack activity. Therefore, the relative increases in settlement rate, axial force, and microseismic events were used to characterize the changes in these three types of responses, and the sum of squares was used for comprehensive fusion, ensuring that the value only increases significantly when multiple source responses increase simultaneously, thereby achieving effective quantification of roof disturbance sensitivity.

[0039] The specific formula for the disturbance-sensitive response value is as follows: ; In the formula, This represents the disturbance-sensitive response value, which is used to comprehensively characterize the amplification of the roof's displacement response, support stress response, and fracture activity response before and after disturbance, thereby reflecting the roof's sensitivity to external disturbances. This represents the settling velocity after disturbance, characterizing the average level of the roof surface settlement rate after disturbance. A larger value indicates more significant deformation of the roof surface after disturbance. It represents the pre-disturbance settlement velocity value, which characterizes the average level of the top plate surface settlement rate before disturbance and is used as a benchmark for the deformation response of the top plate surface. This represents the baseline value of axial force after disturbance, which characterizes the average level of the average axial force of the anchor bolt after disturbance. The larger the value, the stronger the support stress response after disturbance. It represents the baseline value of axial force before disturbance, characterizing the average level of the average axial force of the anchor bolt before disturbance, and is used as a benchmark for the support force response; This represents the post-disturbance microseismic value, which characterizes the average number of microseismic events after the disturbance. The larger the value, the more active the fracture activity after the disturbance. It represents the pre-disturbance microseismic value, which characterizes the average number of microseismic events before the disturbance and is used as a benchmark for fracture activity response; To represent extremely small positive numbers and prevent division by zero errors and abnormal amplification when the denominator is close to zero, the preferred value is [value missing]. arrive .

[0040] In this embodiment, Table 1 is a data table of disturbance-sensitive response values. Among them, the smallest positive number is... The table details the pre-disturbance settlement velocity, post-disturbance settlement velocity, pre-disturbance axial force baseline value, post-disturbance axial force baseline value, pre-disturbance micro-vibration value, post-disturbance micro-vibration value, and disturbance sensitivity response value corresponding to five disturbance events. Specifically, disturbance event 1 corresponds to a pre-disturbance settlement velocity of 0.18, a post-disturbance settlement velocity of 0.24, a pre-disturbance axial force baseline value of 82, a post-disturbance axial force baseline value of 88, a pre-disturbance micro-vibration value of 3, a post-disturbance micro-vibration value of 5, and a disturbance sensitivity response value of 0.7489; disturbance event 2 corresponds to a pre-disturbance settlement velocity of 0.22, a post-disturbance settlement velocity of 0.45, a pre-disturbance axial force baseline value of 79, a post-disturbance axial force baseline value of 108, a pre-disturbance micro-vibration value of 4, a post-disturbance micro-vibration value of 11, and a disturbance sensitivity response value of 2.0713; and disturbance event 3 corresponds to a pre-disturbance settlement velocity of 0.15, a post-disturbance settlement velocity of 0.16, and a pre-disturbance axial force baseline value of 90. The values ​​for disturbance events are as follows: 1. Post-disturbance axial force base value: 92; Pre-disturbance micro-vibration value: 2; Post-disturbance micro-vibration value: 2; Disturbance sensitivity response value: 0.0703. 2. The values ​​for disturbance events are as follows: 3. Pre-disturbance subsidence velocity: 0.30; Post-disturbance subsidence velocity: 0.62; Pre-disturbance axial force base value: 85; Post-disturbance axial force base value: 132; Pre-disturbance micro-vibration value: 5; Post-disturbance micro-vibration value: 14; Disturbance sensitivity response value: 2.1641. 3. The values ​​for disturbance events are as follows: 4. Pre-disturbance subsidence velocity: 0.12; Post-disturbance subsidence velocity: 0.28; Pre-disturbance axial force base value: 76; Post-disturbance axial force base value: 95; Pre-disturbance micro-vibration value: 1; Post-disturbance micro-vibration value: 4; Disturbance sensitivity response value: 3.2925.

[0041] Table 1. Disturbance Sensitivity Response Values

[0042] like Figure 3 The figure shows a time-series evolution diagram of the disturbance-sensitive response value. The disturbance event number is used as the horizontal axis and the disturbance-sensitive response value as the vertical axis, reflecting the trend of roof response changes under different disturbances. (See Table 1 and...) Figure 3It can be seen that the disturbance sensitivity response values ​​exhibit significant fluctuations among different events. The lowest disturbance sensitivity response value corresponds to disturbance event 3, indicating that the roof is not sensitive to disturbances during this stage. The disturbance sensitivity response values ​​for disturbance events 2, 4, and 5 increase significantly, especially reaching the maximum value for disturbance event 5, indicating an enhanced roof response to disturbances. This trend is consistent with the synchronous increase in settling velocity, axial force, and microseismic activity after the disturbance, demonstrating that the disturbance sensitivity response value can effectively and comprehensively characterize the amplification of the multi-source responses of the roof before and after the disturbance. Therefore, it can be used to determine whether the roof is in a low-sensitivity, transitional, or high-sensitivity state, providing a basis for subsequent risk assessment and adjustment decisions.

[0043] like Figure 4 The diagram shows a comparison of the components of the disturbance-sensitive response values. The radar coordinate system is constructed using three dimensions: the relative increase in the roof surface subsidence rate, the relative increase in the anchor bolt axial force, and the relative increase in the number of microseismic events. Different disturbance events are represented by independent closed curves, thus visually demonstrating the differences in the composition and relative intensity of each disturbance event across the three types of response components. (See Table 1 and...) Figure 4 It can be seen that the distribution patterns of each disturbance event are significantly different in the three dimensions. Disturbance event 5 stands out significantly in the direction of the relative increase in the number of microseismic events, forming the largest outward expansion area. Disturbance events 2 and 4 both show relatively balanced and large expansion in the three dimensions, while disturbance event 3 shows the most significant overall contraction, indicating that its sensitive response is the weakest. The degree of extension in each direction in the figure is consistent with the corresponding relative increase value, indicating that the magnitude of the disturbance sensitive response value comes from the comprehensive superposition of the three types of response components. Among them, the relative increase in the number of microseismic events contributes the most significantly to the highly sensitive response events. Therefore, it can intuitively reflect the differences in the response structure and the dominant influencing factors of different disturbance events, and further verify the ability of the disturbance sensitive response value to characterize the coupling characteristics of multi-source responses.

[0044] In this implementation plan, a multi-source response comparison mechanism before and after disturbance is constructed to uniformly quantify the changes in roof surface deformation, support stress, and crack activity in the form of relative increases. The synergistic fusion of multi-dimensional responses is achieved through the sum of squares and the square root, so that the results only increase significantly when the multi-source responses are synchronously enhanced, thereby effectively improving the ability to characterize disturbance sensitivity. At the same time, the stability and comparability of the assessment results are enhanced through window symmetry processing and anomaly suppression mechanisms, providing a reliable basis for roof condition determination and adjustment decisions.

[0045] Specifically, the process of determining the roof condition by combining roof disturbance sensitivity and fracture hysteresis statistical feature set is as follows: Subtract the mean value of fracture hysteresis activity from the current fracture hysteresis activity value, and divide by the sum of the standard deviation of the fracture hysteresis activity value and the minimum positive number to obtain the fracture hysteresis activity deviation; wherein, the mean value of fracture hysteresis activity and the standard deviation of fracture hysteresis activity value are calculated based on the sliding time window, and the minimum positive number is preferably selected as [value missing]. to , which is used to avoid numerical instability when the standard deviation is close to zero; the fracture hysteresis activity deviation is used to characterize the deviation degree of the current fracture hysteresis activity relative to the historical statistical level, and a larger value thereof indicates a more abnormal current fracture evolution degree. The disturbance sensitive response value is compared with the disturbance threshold S1, the fracture hysteresis activity deviation is compared with the deviation threshold Z1, and a roof state label is output; wherein, the disturbance threshold and the deviation threshold are preferably determined based on the statistical distribution of historical normal operation data, so as to improve the adaptability and reliability of the determination result. When ≥S1, and ≥Z1, it is determined that the current roof is in a risk coupling response state, which is used to characterize a state where both the external response and the internal fracture activity are significantly enhanced but no obvious instability has occurred. A large disturbance sensitive response value indicates that the roof has generated a strong response to external disturbances, and a large fracture hysteresis activity deviation indicates that the internal fracture evolution has significantly deviated from the normal level. The simultaneous satisfaction of both conditions indicates that the roof is in a high-risk evolution stage but has not been completely destabilized. When ≥S1, and <Z1, it is determined that the current roof is in a disturbance sensitive response state, which is used to characterize a state where the roof is sensitive to external disturbance responses but internal fractures have not developed significantly. A large disturbance sensitive response value indicates that external disturbances have caused obvious structural responses, while a small fracture hysteresis activity deviation indicates that internal fractures have not been synchronously enhanced. At this time, the roof is in a stage of high sensitivity to disturbances but no significant structural deterioration. When <S1, and ≥Z1, it is determined that the current roof is in an internal fracture evolution state, which is used to characterize a state with weak apparent response but continuous internal fracture evolution. A small disturbance sensitive response value indicates that the external macroscopic response is not obvious, while a large fracture hysteresis activity deviation indicates that internal fractures are still continuously developing, reflecting the typical hidden danger feature of "apparently stable but internally evolving". When <S1, and <Z1, it is determined that the current roof is in a stable response state, which is used to characterize the operation state where the roof as a whole is under low disturbance and the internal structure is stable. Both the disturbance sensitive response value and the fracture hysteresis activity deviation are at low levels, indicating that no abnormality occurs in either the external response or the internal evolution, and the roof is in a stable working interval.

[0046] In this implementation plan, a two-dimensional discrimination mechanism based on the disturbance-sensitive response value and the deviation of the fracture hysteresis activity is constructed to achieve the collaborative analysis of the external response and internal fracture evolution of the roof, so that different instability evolution stages can be effectively distinguished. At the same time, the threshold is adaptively determined by combining historical statistical distribution, which improves the stability and adaptability of the judgment results, thereby accurately identifying key states such as pseudo-stability, disturbance sensitivity and latent evolution, and enhancing the foresight and reliability of roof risk assessment.

[0047] Specifically, the process of comprehensively judging the adjustment level by integrating the internal stability of the roof and the roof disturbance sensitivity is as follows: The support force dispersion coefficient is obtained by dividing the current standard deviation of the anchor bolt axial force by the sum of the mean anchor bolt axial force and the minimum positive number. The standard deviation and mean anchor bolt axial force are calculated based on the axial force data of all monitored anchor bolts within the current time window, and the minimum positive number is preferably selected as [value missing]. arrive The coefficient of variation is used to avoid numerical instability when the denominator approaches zero. The support stress dispersion coefficient is used to characterize the uniformity of the stress distribution in the support system. The larger the value, the more uneven the stress on the anchor bolts, indicating the existence of localized stress concentration. After performing a hyperbolic tangent function operation on the support stress dispersion coefficient and adding one, a stress balance amplification term is obtained. The support stress dispersion coefficient is nonlinearly compressed through the hyperbolic tangent function to stabilize its value range within a finite interval. The addition operation ensures that the stress balance amplification term is always greater than zero, thus amplifying and modulating subsequent risk terms. The output range of the hyperbolic tangent function is (-1, 1), and after the addition, the value range of the stress balance amplification term is (0, 2), which is used to amplify and characterize the uneven stress in the support. The more uneven the support stress, the closer the amplification term is to the upper limit, thereby enhancing the sensitivity to unevenness risks. The coupled response value is obtained by multiplying the disturbance-sensitive response value with the corresponding fracture hysteresis activity value, adding a small positive number, and then taking the square root. This product characterizes the coupling strength between the external disturbance response and the internal fracture evolution; the coupled response value increases significantly only when both are simultaneously large. The square root treatment mitigates the product amplification effect and maintains numerical scale stability. The disturbance-sensitive response value is calculated based on the relative increase of multiple sources, while the fracture hysteresis activity value is obtained through processing using ratios, logarithms, and hyperbolic tangent functions. Both are dimensionless indices, ensuring that the product operation has a consistent dimensional basis and possesses comparability and physical rationality. The comprehensive risk adjustment term is obtained by adding the coupled response value to one and taking the natural logarithm. The addition of one avoids zero or negative values ​​in the logarithmic calculation, and the natural logarithm further compresses the numerical range, enhancing the stability and comparability of the results. Furthermore, the addition ensures that the input to the logarithmic function is always greater than 1, thus maintaining a non-negative value for the comprehensive risk adjustment term and preventing negative values ​​from interfering with subsequent adjustment judgments. Simultaneously, the logarithmic mapping compresses the changes in the coupled response value to a finite interval, achieving a smooth transition between different risk levels. Multiplying the comprehensive risk adjustment term by the force balance amplification term yields the staged coupled adjustment judgment value. This value comprehensively reflects the current internal evolution state of the roof, the degree of disturbance response, and the force distribution of the support, providing a quantitative basis for generating subsequent adjustment commands. The roof control demand depends not only on the coupling degree between the internal crack evolution and the external disturbance response, but also on the uniformity of the support stress distribution. Therefore, by multiplying and coupling the disturbance sensitive response value with the crack hysteresis activity value to characterize the comprehensive risk intensity, and by introducing the anchor bolt stress dispersion coefficient for modulation, the results are significantly amplified when "risk is enhanced and stress is uneven" coexist, thereby achieving effective quantification and classification of the control demand level.

[0048] The specific formula for the stage coupling adjustment judgment value is as follows: ; In the formula, The stage coupling adjustment judgment value is used to comprehensively characterize the degree of internal crack evolution, disturbance response intensity and support stress equilibrium state of the roof, thereby determining the overall risk level of the roof and serving as the adjustment decision criterion; The value represents the hysteresis activity of the fracture, which characterizes the degree of inconsistency between the continuous evolution of fractures inside the roof and the deep surface response. The larger the value, the more obvious the internal latent instability. This represents the disturbance-sensitive response value, which characterizes the degree of amplification of the roof's response before and after a disturbance. The larger the value, the more sensitive it is to disturbances and the more prone it is to instability. This represents the standard deviation of the anchor bolt axial force, which characterizes the degree of dispersion of the support force. The larger the value, the more uneven the force distribution. This represents the average axial force of the anchor bolts, characterizing the overall support stress level, and is used to normalize the degree of dispersion. To represent extremely small positive numbers and prevent numerical instability caused by zero values ​​in the denominator and radical terms, the preferred value is [value missing]. arrive .

[0049] In this implementation plan, a three-dimensional coupled evaluation mechanism of "disturbance response, crack evolution, and stress distribution" is constructed to uniformly quantify the risk intensity inside the roof and the unbalanced stress state of the support. Through nonlinear compression and modulation amplification, multi-factor synergistic integration is achieved, so that the adjustment value is significantly improved only when the risk increases and the stress imbalance occurs. This improves the pertinence and accuracy of the adjustment decision and enhances the reliability and adaptability of roof control.

[0050] Specifically, the process of generating adjustment instructions based on the adjustment level and outputting adjustment strategies and reinforcement suggestions in conjunction with the adjustment instructions and the roof condition is as follows: The stage coupling adjustment judgment value is compared with the adjustment threshold: when the stage coupling adjustment judgment value is less than the adjustment threshold, a regular adjustment instruction is output; when the stage coupling adjustment judgment value is greater than or equal to the adjustment threshold, an enhanced adjustment instruction is output. The regular adjustment instruction is used to maintain the stable operation of the current support system, while the enhanced adjustment instruction is used to trigger targeted reinforcement measures. The adjustment threshold is preferably determined based on the statistical distribution of stage coupling adjustment judgment values ​​in historical normal operation data to improve the adaptability and reliability of the adjustment instruction division. The roof status label is read. When the roof status label indicates a stable response state, if the output is a normal adjustment command, a control command is sent to the support control device to maintain the current support parameters unchanged, and the sampling frequency of the roof monitoring data remains unchanged. In this state, both disturbance response and fracture evolution are at low levels, requiring no additional intervention to avoid unnecessary system disturbances. If the output is an enhanced adjustment command, a control command is sent to increase the sampling frequency of the roof monitoring data and the disturbance event identification frequency, and the current roof status is reviewed. When the adjustment judgment value increases, there may be short-term anomalies and state change trends. By increasing the monitoring... The frequency of measurement and identification can verify whether the risk persists, thereby avoiding misjudgment. Among them, improving the sampling frequency of roof monitoring data is preferably achieved by shortening the data sampling period. The preferred methods to improve the frequency of disturbance event identification include shortening the sampling period of vibration acceleration signals, shortening the sliding time window length of vibration intensity values, and reducing the number of sampling points N required for continuous over-threshold determination, so as to improve the real-time performance of disturbance event identification and enhance the ability to track state changes. The verification process is preferably to repeatedly calculate and determine the disturbance sensitive response value and the crack hysteresis activity value within a preset time interval to confirm whether the state change persists. When the roof status label is in a risk-coupled response state, if the output is a normal adjustment command, a control command is sent to the support control device to maintain the current anchor preload setting value and lock the support parameter adjustment interface. At the same time, a control command is sent to increase the sampling frequency of roof monitoring data. In this state, both internal cracks and external responses are enhanced, but instability has not yet occurred. Reducing the support will exacerbate the risk, so it is necessary to maintain or even strengthen the constraints and continue monitoring. If the output is a strengthening adjustment command, a parameter suggestion is sent to the support control device to increase the number of anchor bolts and increase the local support density. An adjustment suggestion is sent to the grouting equipment to perform grouting reinforcement on the corresponding area. In this state, it may have entered a high-risk evolution stage. Increasing the support density and grouting can improve the integrity and bearing capacity of the surrounding rock, thereby inhibiting crack propagation. Among these, increasing the number of anchor bolts and increasing the local support density are preferably determined based on the current abnormal area range. The abnormal area and the corresponding grouting reinforcement area can be determined according to the spatial location of the sensors deployed for collecting roof monitoring data.When the roof status label is in the disturbance-sensitive response state, if the output is a normal adjustment command, it will output a control command to increase the sampling frequency of roof monitoring data and the frequency of disturbance event identification. This state indicates that it is highly sensitive to external disturbances. Although the internal cracks have not yet developed significantly, monitoring needs to be strengthened to prevent further amplification of the disturbance. If the output is an enhanced adjustment command, it will output a control command to increase the sampling frequency of roof monitoring data and output parameter suggestions to the support control device to increase the local support stiffness and the number of anchor bolts. This state indicates that the external response may have been significantly enhanced. By increasing the support stiffness, the structural deformation sensitivity can be reduced, and the disturbance amplification effect can be suppressed from the source. Among these, increasing the local support stiffness is preferably achieved by adjusting the anchor bolt preload. When the roof status label indicates internal fracture evolution, if the output is a conventional adjustment command, it sends a control command to the support control device to extend the current support parameter holding time and maintain the anchor bolt preload setting value. This state indicates that internal fractures are continuously evolving but the apparent response is weak, requiring extended support time to prevent further development of latent deterioration. If the output is a reinforcement adjustment command, it sends a parameter suggestion to the support control device to increase the number of anchor bolts and an adjustment suggestion to the grouting equipment to perform grouting reinforcement. This state indicates that internal fractures may have been developing continuously, and strengthening the support and grouting can improve rock mass integrity and inhibit the transformation of latent instability into overt instability. All adjustment suggestions are recorded and prompted through the control terminal and can be distributed through the industrial control bus to guide on-site implementation. Simultaneously, after implementing adjustment measures, roof monitoring data is continuously collected and various evaluation indicators are updated to form a closed-loop control process, thereby improving adaptive adjustment capability and long-term operational stability.

[0051] In this implementation plan, a hierarchical control mechanism based on adjustment judgment values ​​and roof status labels is constructed to achieve targeted matching of adjustment strategies under different risk states, enabling control measures to be dynamically adjusted according to the internal evolution of the roof and changes in external disturbances. At the same time, by combining monitoring frequency adjustment, support reinforcement and grouting reinforcement, and introducing a closed-loop feedback update mechanism, the accuracy and timeliness of adjustment decisions are effectively improved, thereby enhancing the safety, stability and adaptability of roof reinforcement control.

[0052] Reference Figure 2As shown, the second aspect of the present invention provides a soft rock roof reinforcement control system based on multi-stage coupled regulation, applied to the aforementioned soft rock roof reinforcement control method based on multi-stage coupled regulation, comprising: a monitoring data acquisition and processing module, used to acquire roof monitoring data in real time, and perform time alignment, anomaly removal, and normalization processing on the roof monitoring data; a multi-stage response feature construction module, used to extract roof internal evolution features and fracture hysteresis activity features based on preprocessed roof monitoring data, comprehensively evaluate roof internal stability, and construct a fracture hysteresis statistical feature set based on roof internal stability; a pseudo-stability identification and evaluation module, used to identify disturbance events using roof monitoring data, and evaluate roof disturbance sensitivity by combining the response changes of roof monitoring data before and after the disturbance event, and determine the roof state by combining the roof disturbance sensitivity and the fracture hysteresis statistical feature set; and a multi-stage coupled regulation execution module, used to fuse roof internal stability and roof disturbance sensitivity, comprehensively judge the regulation level, generate regulation commands based on the regulation level, and output regulation strategies and reinforcement suggestions by combining the regulation commands and the roof state.

[0053] In this implementation plan, a multi-module collaborative system consisting of data processing, feature extraction, state determination, and regulation execution is constructed to achieve closed-loop processing of roof monitoring data from acquisition to decision execution. It can comprehensively characterize the internal evolution and disturbance response characteristics of the roof, accurately identify pseudo-stability and potential risk states, and dynamically generate targeted control strategies based on the regulation level, thereby improving the accuracy, real-time performance, and adaptability of roof reinforcement control.

[0054] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0055] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. As those skilled in the art will understand, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for reinforcing and controlling soft rock roof based on multi-stage coupled regulation, characterized in that, Includes the following steps: S1, collects roof monitoring data in real time, and performs time alignment, anomaly removal and normalization processing on the roof monitoring data; S2, Based on the preprocessed roof monitoring data, extract the internal evolution characteristics and fracture hysteresis activity characteristics of the roof, comprehensively evaluate the internal stability of the roof, and construct a fracture hysteresis statistical feature set based on the internal stability of the roof; S3. Use roof monitoring data to identify disturbance events, and combine the response changes of roof monitoring data before and after the disturbance event to assess the roof disturbance sensitivity. Combine the roof disturbance sensitivity and the crack hysteresis statistical feature set to determine the roof condition. S4 integrates the internal stability of the roof and the roof disturbance sensitivity to comprehensively judge the adjustment level, generate adjustment commands based on the adjustment level, and output adjustment strategies and reinforcement suggestions in combination with the adjustment commands and the roof status.

2. The method for strengthening and controlling soft rock roof based on multi-stage coupling regulation according to claim 1, characterized in that, The specific process of real-time acquisition of roof monitoring data, including time alignment, anomaly removal, and normalization, is as follows: Real-time acquisition of roof monitoring data includes: acquiring surface displacement values ​​and deep delamination displacement values ​​of the roof using roof displacement gauges, multi-point displacement gauges, and delamination meters; acquiring axial force values ​​of all monitored anchor bolts using anchor bolt force gauges; acquiring microseismic event records and acoustic emission signals using a microseismic monitoring system and acoustic emission sensors. The microseismic event records include the occurrence time and energy of microseismic events, and are statistically analyzed within a sliding time window to obtain the number of microseismic events, total microseismic energy, and acoustic emission ring count within the current time window; and acquiring vibration acceleration signals of the area where the roof is located using vibration sensors. The roof monitoring data were timestamped according to a unified time base and resampled according to a fixed sampling period. For the roof surface displacement value, deep delamination displacement value, and axial force value of each monitored anchor bolt, the median value and median absolute deviation of the corresponding data within the current time window were calculated. If the absolute difference between the sampled value and the median value exceeded the multiple threshold of the median absolute deviation, it was judged as abnormal data and removed. For missing data, linear interpolation was used to complete it. The roof monitoring data were subjected to min-max normalization. For the roof surface displacement value and the deep delamination displacement value, the current sampled value was subtracted from the previous sampled value and then divided by the sampling time interval to obtain the roof surface subsidence rate and the deep delamination change rate. The mean and standard deviation of the axial force values ​​of all monitored anchor bolts were calculated to obtain the mean and standard deviation of the anchor bolt axial force.

3. The method for strengthening and controlling soft rock roof based on multi-stage coupling regulation according to claim 1, characterized in that, The specific process of extracting the internal evolution characteristics and fracture hysteresis activity characteristics of the roof based on the preprocessed roof monitoring data, and comprehensively evaluating the internal stability of the roof is as follows: Read the preprocessed roof monitoring data, divide the current deep delamination rate by the sum of the roof surface subsidence rate and the minimum positive number to obtain the deep-surface response difference ratio, and perform hyperbolic tangent function operation on the deep-surface response difference ratio to obtain the deep-surface response modulation term; Multiply the number of microseismic events within the current time window by the total energy of the microseismic events, and take the square root of the product to obtain the fracture energy frequency value. Add the fracture energy frequency value to the acoustic emission ringing count, then add it to one, and take the natural logarithm to obtain the internal fracture activity intensity term. Multiplying the deep surface response modulation term by the internal fracture activity intensity term yields the fracture hysteresis activity value.

4. The method for strengthening and controlling soft rock roof based on multi-stage coupling regulation according to claim 1, characterized in that, The specific process for constructing the statistical feature set of crack hysteresis based on the internal stability of the top plate is as follows: According to the sampling time sequence, the fracture hysteresis activity values ​​corresponding to each sampling time are continuously stored to form a fracture hysteresis activity value sequence. Based on the sliding time window, the mean and standard deviation of the fracture hysteresis activity values ​​are calculated to obtain the mean and standard deviation of the fracture hysteresis activity values.

5. The method for strengthening and controlling soft rock roof based on multi-stage coupling regulation according to claim 1, characterized in that, The specific process for identifying disturbance events using roof monitoring data is as follows: The root mean square value of the vibration acceleration signal within the sliding time window is calculated to obtain the vibration intensity value. The vibration intensity value is compared with the intensity threshold. When the vibration intensity value changes from not exceeding the intensity threshold to exceeding the intensity threshold, and exceeds the intensity threshold for N consecutive sampling times, the moment when the intensity threshold is first exceeded is determined as the disturbance event moment. Taking the disturbance event moment as the center, time series of length T are truncated forward to construct the pre-disturbance time window, and time series of the same length are truncated backward to construct the post-disturbance time window.

6. The method for strengthening and controlling soft rock roof based on multi-stage coupled regulation according to claim 1, characterized in that, The specific process for assessing the sensitivity of the roof to disturbance by combining the response changes of roof monitoring data before and after the disturbance event is as follows: Based on the time window before and after the disturbance, the mean value of the subsidence rate of the roof surface is calculated to obtain the subsidence velocity value before and after the disturbance. The mean value of the axial force of the anchor bolts is calculated to obtain the axial force base value before and after the disturbance. The mean value of the number of micro-seismic events in the current time window is calculated to obtain the micro-seismic value before and after the disturbance. The difference between the post-disturbance settling velocity and the pre-disturbance settling velocity is calculated, and only the part greater than zero is retained. This difference is then divided by the sum of the pre-disturbance settling velocity and the smallest positive number to obtain the relative increase in the rate of settlement of the top surface. The difference between the post-disturbance axial force base value and the pre-disturbance axial force base value is calculated, and only the part greater than zero is retained. This difference is then divided by the sum of the pre-disturbance axial force base value and the smallest positive number to obtain the relative increase in anchor bolt axial force. The difference between the post-disturbance microseismic value and the pre-disturbance microseismic value is calculated, and only the part greater than zero is retained. This difference is then divided by the sum of the pre-disturbance microseismic value and the smallest positive number to obtain the relative increase in the number of microseismic events. Calculate the sum of squares of the relative increase in the rate of subsidence of the roof surface, the relative increase in the axial force of the anchor bolts, and the relative increase in the number of microseismic events, and take the square root of the sum of squares to obtain the disturbance-sensitive response value.

7. The method for strengthening and controlling soft rock roof based on multi-stage coupling regulation according to claim 1, characterized in that, The specific process for determining the roof condition by combining the roof disturbance sensitivity and the statistical feature set of crack hysteresis is as follows: The deviation of the fracture hysteresis activity is obtained by subtracting the mean value of the fracture hysteresis activity from the current value of fracture hysteresis activity and dividing by the sum of the standard deviation of the fracture hysteresis activity value and the smallest positive number. Perturbation-sensitive response value Compared with the disturbance threshold S1, the deviation of the fracture hysteresis activity is calculated. Compare with the deviation threshold Z1 and output the top plate status label; When ≥S1, and ≥Z1, it is determined that the current roof is in a risk coupling response state; when ≥S1, and <Z1, it is determined that the current roof is in a disturbance sensitive response state; when <S1, and ≥Z1, it is determined that the current roof is in an internal fracture evolution state; when <S1, and <Z1, it is determined that the current roof is in a stable response state.

8. The method for strengthening and controlling soft rock roof based on multi-stage coupling regulation according to claim 1, characterized in that, The specific process for comprehensively judging the adjustment level by integrating the internal stability of the roof plate and the roof plate's disturbance sensitivity is as follows: Divide the current standard deviation of anchor bolt axial force by the sum of the mean and minimum positive values ​​of anchor bolt axial force to obtain the support force dispersion coefficient. After performing hyperbolic tangent function operation on the support force dispersion coefficient, add one to obtain the force balance amplification term. Multiply the disturbance-sensitive response value by the corresponding fracture hysteresis activity value, add a very small positive number, and take the square root to obtain the coupled response value. Add the coupled response value to one and take the natural logarithm to obtain the comprehensive risk adjustment term. Multiplying the comprehensive risk adjustment term by the force equilibrium amplification term yields the stage coupling adjustment judgment value.

9. The method for strengthening and controlling soft rock roof based on multi-stage coupling regulation according to claim 1, characterized in that, The specific process of generating adjustment commands based on the adjustment level, and combining the adjustment commands with the roof status to output adjustment strategies and reinforcement suggestions is as follows: The stage coupling adjustment judgment value is compared with the adjustment threshold: when the stage coupling adjustment judgment value is less than the adjustment threshold, a normal adjustment command is output; when the stage coupling adjustment judgment value is greater than or equal to the adjustment threshold, an enhanced adjustment command is output. Read the roof status label. When the roof status label is in a stable response state, if the output is a normal adjustment command, then output a control command to the support control device to keep the current support parameters unchanged and keep the sampling frequency of the roof monitoring data unchanged; if the output is an enhanced adjustment command, then output a control command to increase the sampling frequency of the roof monitoring data and the frequency of disturbance event identification, and verify the current roof status. When the roof status label is in the risk coupling response state, if the output is a normal adjustment command, a control command is output to the support control device to keep the current anchor preload setting value from decreasing, and the support parameter adjustment interface is locked. At the same time, a control command to increase the sampling frequency of roof monitoring data is output. If the output is a reinforcement adjustment command, a parameter suggestion to increase the number of anchor bolts and increase the local support density is output to the support control device, and an adjustment suggestion to perform grouting reinforcement on the corresponding area is output to the grouting equipment. When the roof status label is in the disturbance sensitive response state, if the output is a normal adjustment command, the output will be a control command to increase the sampling frequency of roof monitoring data and the frequency of disturbance event identification; if the output is an enhanced adjustment command, the output will be a control command to increase the sampling frequency of roof monitoring data, and parameter suggestions to increase the local support stiffness and increase the number of anchor bolts will be output to the support control device. When the top plate status label indicates internal crack evolution, if the output is a normal adjustment command, it will output a control command to the support control device to extend the current support parameter holding time and maintain the anchor bolt preload setting value from decreasing; if the output is a reinforcement adjustment command, it will output a parameter suggestion to the support control device to increase the number of anchor bolts, and output an adjustment suggestion to the grouting equipment to perform grouting reinforcement.

10. A soft rock roof reinforcement control system based on multi-stage coupled regulation, characterized in that, include: The monitoring data acquisition and processing module is used to acquire roof monitoring data in real time and perform time alignment, anomaly removal and normalization on the roof monitoring data; A multi-stage response feature construction module is used to extract the internal evolution features and fracture hysteresis activity features of the roof based on the preprocessed roof monitoring data, comprehensively evaluate the internal stability of the roof, and construct a fracture hysteresis statistical feature set based on the internal stability of the roof. The pseudo-stability identification and evaluation module is used to identify disturbance events using roof monitoring data, and to evaluate the roof disturbance sensitivity by combining the response changes of roof monitoring data before and after the disturbance event. The roof status is determined by combining the roof disturbance sensitivity and the crack hysteresis statistical feature set. The multi-stage coupled regulation execution module is used to integrate the internal stability of the roof and the roof disturbance sensitivity, comprehensively judge the regulation level, generate regulation commands based on the regulation level, and output regulation strategies and reinforcement suggestions in combination with the regulation commands and the roof status.

Images