Early warning method for the risk of instability of anchored rock mass under mining disturbance
By utilizing representative values of anchor cable stress, anchor bolt tray pressure, and roof delamination displacement during deep gold mining, a benchmark zone and response hysteresis time were established. This solved the problem of inaccurate early warning of rock mass instability in traditional monitoring methods, and enabled highly reliable early warning and continuous tracking of rock mass instability risks.
Patent Information
- Application Number
- CN202611621120.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-10-20
- Publication Date
- 2026-11-17
AI Technical Summary
During the mining of deep gold mines, traditional monitoring methods are difficult to accurately assess the risk of instability of anchored rock masses under mining disturbances, especially due to stress distribution changes caused by blasting or mechanical cutting and the incompleteness of monitoring data, making it difficult to effectively warn of irreversible deformation and cumulative damage of the rock mass.
By acquiring representative values of anchor cable stress, anchor plate pressure, and top plate delamination displacement, a bearing reference zone and displacement reference are established. The displacement retention increment is calculated using the deviation change and response hysteresis time. Combined with activity path records, continuous tracking and early warning of rock mass instability risk can be achieved.
It improves the sensitivity to rock mass instability risk and the reliability of early warning results, reduces the misjudgment rate and the risk of early warning interruption, and realizes the capture of the hysteresis characteristics of rock mass stress-displacement response and continuous tracking throughout its entire life cycle.
Smart Images

Figure CN123117844A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data processing technology, specifically to a method for early warning of the risk of instability in anchored rock mass under mining disturbance. Background Technology
[0002] In deep gold mining operations, blasting or mechanical cutting can alter the stress distribution of the surrounding rock, causing changes in anchor cable stress, anchor bolt tray pressure, and roof delamination displacement. Monitoring platforms use records before and after mining events to determine the support stress and surrounding rock deformation. Anchor cable stress and anchor bolt tray pressure are load-bearing parameters, which are different from roof delamination displacement. A reading of a load-bearing parameter close to the pre-event level cannot independently indicate that the roof delamination displacement has stopped increasing.
[0003] After mining disturbance, the load-bearing measurement readings may deviate from the pre-event level and then fall back, during which time the roof delamination displacement may still increase. If the first and last readings of the observation interval are used to replace the change process within the interval, the situation of deviation in the middle may be masked by the fall back at the end. Normal response lag, short-term fluctuations and existing residual displacements will also affect the comparison of displacement changes. Based solely on the last reading or a single change, it is difficult to determine whether the new displacement during the load-bearing fall back is still retained in subsequent observations.
[0004] During continuous mining, the short-term stability after an event may be interrupted by subsequent disturbances. If the stability assessment restarts, existing abnormal records may be cleared at the same time. Or, if the displacement reference is changed repeatedly during the event handling, the displacement changes that occurred previously may be masked by the current observation. Communication interruption can only indicate that the current data is incomplete. It is still difficult to confirm whether the displacement anomaly during the previous load-bearing decline has been resolved based solely on the stabilization of the final reading or the absence of new anomalies. Summary of the Invention
[0005] To address the aforementioned technical problems, this invention proposes a method for early warning of instability risk in anchored rock masses under mining disturbance, comprising: Obtain the anchor cable force, anchor bolt tray pressure, roof delamination displacement and mining event time of the same monitoring unit, determine the representative value of the measurement item according to the foundation sub-window, take the anchor cable force and anchor bolt tray pressure as the bearing measurement item, and take the roof delamination displacement measurement item representative value as the displacement representative value. The effective basic sub-window within the preset observation period before the event is used as the reference sub-window. The upper and lower limits of the corresponding load reference band are determined by the representative values of each load measurement item in the reference sub-window. The median of the displacement representative values of the reference sub-window is used as the displacement reference. The load reference band and the displacement reference remain fixed during the event processing. The observation interval is formed by consecutive effective basic sub-windows after the event. The deviation is calculated according to the out-of-band distance from the representative value of each load-bearing measure to the corresponding load-bearing reference band. The observation interval in which the net reduction of the difference between the maximum value of the deviation of at least one load-bearing measure and the end value is greater than the corresponding preset net reduction threshold, and the maximum reverse deviation of each load-bearing measure is less than or equal to the corresponding preset reverse deviation threshold, is recorded as the load-bearing regression interval. Within the load return interval, a comparison start point and an end holding interval are determined according to a preset response hysteresis duration and a preset holding duration. The comparison start point is earlier than or equal to the start point of the end holding interval. The displacement retention increment is obtained by subtracting the displacement representative value of the comparison start point from the average value of the displacement representative values within the end holding interval. The latest completed basic sub-window is recorded as the newest basic sub-window. When the newest basic sub-window is valid, the net displacement increment is obtained by subtracting the displacement reference from the displacement representative value of the newest basic window. When the displacement increment is greater than the preset increment threshold, additional judgment criteria are added or an activity path record is established. The activity path record remains active until the closing condition is met, and instruction information is sent based on the activity status of the activity path record and the comparison result between the representative value of the measurement item and the preset engineering limit value. The instruction information includes a status flag, displacement trend category and the net displacement increment.
[0006] Preferably, the representative value of the measurement item includes: forming a matching group from the valid measured data of each measurement item that has been time-calibrated and meets the preset synchronization tolerance; recording the basic sub-window with a number of matching groups reaching a preset number and time coverage meeting the preset requirements as a valid basic sub-window; and recording the basic sub-window with a number of matching groups less than the preset number as an invalid basic sub-window; taking the median of the observed values of each measurement item in the matching group within the valid basic sub-window to obtain the representative value of the measurement item; and the invalid basic sub-window interrupting the time continuity required for subsequent statistics.
[0007] Preferably, the load-bearing reference band includes: recording the effective basic sub-window within a preset observation period before the event as the reference sub-window, and using the median of the representative values of each load-bearing item in the reference sub-window as the corresponding baseline value; adding a preset measurement margin to the absolute values of the differences between the preset lower quantile value and the preset upper quantile value of the representative value of each load-bearing item in the reference sub-window and the corresponding baseline value, respectively, to obtain the lower tolerance and the upper tolerance; subtracting the lower tolerance from the baseline value and adding the upper tolerance, respectively, to obtain the lower limit and upper limit of the corresponding load-bearing reference band.
[0008] Preferably, the calculation of the maximum reverse bias includes: for each load-bearing measure, the deviation when the representative value of the load-bearing measure is within the corresponding load-bearing reference band is recorded as zero; the deviation of each basic sub-window in the observation interval is subtracted from the minimum deviation from the starting point of the observation interval to the basic sub-window to obtain the reverse bias, and the maximum value of each reverse bias is taken as the maximum reverse bias.
[0009] Preferably, the calculation of the displacement retention increment adopts the following windowing method: the first basic sub-window after delaying the preset response hysteresis time from the starting point of the load return interval is recorded as the comparison starting point; the consecutive effective basic sub-windows selected from the end of the load return interval and whose cumulative time reaches the preset retention time are recorded as the end retention interval.
[0010] Preferably, the status flag includes: taking the ratio of the difference between the representative values of two measurement items of a base sub-window that are adjacent in time and are both valid base sub-windows to the preset sub-window duration as the rate of change, and taking the absolute value of the difference as the fluctuation; when the absolute value of the rate of change and the fluctuation of each measurement item are within the corresponding limit and continue for a preset duration, the status flag is recorded as a stable state; when creating or adding the activity path record, the cumulative time of the stable state is reset.
[0011] Preferably, the displacement trend category includes: taking the latest basic sub-window as the end, selecting consecutive valid basic sub-windows whose cumulative duration after the event reaches a preset long-term duration, calculating the displacement range, the first-to-last difference, and the linear slope, wherein the first-to-last difference is the representative value of the end displacement of the consecutive valid basic sub-window minus the representative value of the first displacement; when the displacement range is less than or equal to a preset amplitude threshold and the absolute value of the linear slope v is less than or equal to a preset rate threshold, the displacement trend category is recorded as the first category; when the first-to-last difference is greater than the preset trend threshold and the linear slope is greater than the preset rate threshold, the displacement trend category is recorded as the second category; when the data conditions for the long-term interval are met, but the situation is outside the first and second categories, the displacement trend category is recorded as the third category; when the data is insufficient, the displacement trend category is recorded as pending determination, and all the above thresholds are positive numbers.
[0012] Preferably, the status update of the activity path record includes: retaining the activity path record when the status flag changes; for the activity path record, when the status flag is the stable state, the representative values of each load-bearing item in the latest basic sub-window are all within the corresponding load-bearing reference zone, the net displacement increment is within the preset displacement range, and the representative values of each item are all within the corresponding preset engineering limit, if the displacement trend category is the first category, then the activity path record is closed and saved; if the displacement trend category is the third category and the absolute value of the linear slope is less than or equal to the preset rate threshold, then the activity path record is recorded as observation level; in other cases, the activity path record is recorded as continuous tracking level.
[0013] The present invention has the following beneficial effects: 1. By calculating the deviation through a fixed bearing reference zone and jointly verifying the net reduction and maximum reverse increase of the deviation, the interference of local stress oscillation and false fall is filtered out. At the same time, the bearing change and the top plate delamination displacement are limited to the same observation interval for comparison, which provides a highly reliable time window for the correlation analysis of rock mass instability risk and reduces the misjudgment rate caused by time period misalignment or noise interference.
[0014] 2. Due to the hysteresis effect between the anchored rock mass bearing pullback and the roof delamination deformation, and the fact that the monitoring data is easily affected by constant bias and terminal transient noise, a preset response hysteresis time is introduced within the bearing pullback interval to determine the comparison starting point. The displacement increment is calculated by maintaining the average value of the interval at the end. This can adapt to the hysteresis characteristics of the rock mass stress-displacement response, capture the irreversible delamination deformation of the roof during the bearing pullback, and further reduce the risk of constant zero-point drift of the sensor and system bias interference through difference calculation, smooth the terminal transient data noise, and improve the sensitivity to hidden instability risks.
[0015] 3. To address the issue that rock mass stress and deformation often exhibit short-term false stability or brief data interruptions throughout the entire mining disturbance cycle, leading to premature deactivation of alarms by traditional early warning systems, this paper proposes a new approach. By decoupling the update of status flags from the activity status of path records and maintaining a fixed displacement benchmark during event processing, existing activity path records are continuously retained and used in the judgment until strict closure conditions are met. This reduces the risk of early warning interruptions caused by short-term stress stabilization, local data fluctuations, or brief data gaps. Furthermore, it enables continuous tracking of cumulative damage and irreversible deformation of rock mass under mining disturbance throughout its entire lifecycle, improving the continuity and reliability of early warning results. Attached Figure Description
[0016] Figure 1 The flowchart illustrates the method for early warning of instability risk of anchored rock mass under mining disturbance in this invention. Detailed Implementation
[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. 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.
[0018] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.
[0019] Step 1: Obtain the measurement data and determine the representative value of each basic sub-window.
[0020] In this implementation, the monitoring and processing unit processes the anchor cable force data, anchor bolt tray pressure data, roof delamination displacement data, and mining event records within the same monitoring unit. The monitoring and processing unit obtains the observed values through the anchor cable force gauge, anchor bolt tray pressure box, and roof delamination displacement gauge, and obtains the mining event time through the mining operation log or the event record interface connected to the mining equipment.
[0021] Mining events include operational events such as coal mining advance and tunneling operations recorded in the mining operation log. The mining event time is the actual occurrence time of the corresponding operational event, and the record entry time and data reception time are saved separately. The mining event record shall include at least the monitoring unit number, event number, event type, actual occurrence time and record source.
[0022] The anchor cable stress data, anchor bolt tray pressure data, and roof slab delamination displacement data are respectively recorded as measurement items, and the measurement item categories are as follows: The values are taken as anchor cable force, anchor bolt tray pressure, or top plate delamination displacement; anchor cable force is expressed in kilonewtons, anchor bolt tray pressure in megapascals, top plate delamination displacement in millimeters, and time in seconds.
[0023] The three measurement items are connected to the same clock acquisition device. When using separate acquisition terminals, each acquisition terminal calibrates the sampling time according to the unified time source. The unified time source can be provided by the ground time synchronization device and transmitted to the acquisition terminal via a wired network. The monitoring and processing unit records the clock deviation, calibration time and synchronization status of each acquisition terminal, and corrects the original sampling time based on the calibration record.
[0024] The preset synchronization tolerance is denoted as The unit is seconds. The difference between the maximum and minimum sampling times of the three measured items after calibration is less than or equal to... When the time is right, it is considered to meet the preset synchronization tolerance; observations with abnormal synchronization status or clock deviation exceeding the configuration range are recorded as time verification abnormalities and are not included in the matching group statistics.
[0025] The monitoring and processing unit checks the validity of the observations in chronological order. The checks include the validity of the timestamp, communication verification results, sensor operating status, measurement range, and calibration records. Observations that fail to verify communication, are in a faulty state, or are outside the effective measurement range of the sensor are recorded as invalid observations, and the corresponding reasons are saved.
[0026] The sensor's measurement range and preset engineering limits are configured separately. When an observed value is within the sensor's effective measurement range but exceeds the preset engineering limit, the observed value is retained and an engineering limit exceedance check is performed. When an observed value exceeds the statistical range of normal historical data, its validity is verified in conjunction with equipment status and range checks. It is not only invalidated based on statistical range alone; range saturation, communication interruption, and sensor malfunction are respectively marked as equipment anomalies and sent along with indication information. Observations confirmed as valid through the above checks are recorded as valid measured observations, and the sampling data composed of valid measured observations are recorded as valid measured data.
[0027] Each set of valid measured data for each measurement item that has been time-calibrated and meets the preset synchronization tolerance is designated as a matching group. Specifically, during matching, the sampling time of the measurement item with the lowest configured sampling frequency is used as the benchmark. From the remaining two measurement items, valid measured observations with the smallest time difference that have not yet participated in matching are selected. If there are identical time differences, the observation with the earlier sampling time is selected. The matching group is determined when the maximum time difference between the three observations meets the synchronization tolerance requirement. Each original observation participates in one matching group.
[0028] According to the preset sub-window duration The measurement data is divided into basic sub-windows arranged chronologically. The unit is seconds; the first... The starting time of each basic sub-window is recorded as follows: The time interval corresponding to this basic sub-window Adjacent basic sub-windows satisfy .
[0029] When all observations of a matching group fall within the same basic sub-window, the matching group participates in the statistics of that basic sub-window; when a matching group crosses the boundary of a basic sub-window, its original record is retained, and the matching group as a whole is excluded from the representative value statistics of the corresponding basic sub-window.
[0030] Let the minimum number of matching groups required for each basic sub-window be denoted as . , It is a positive integer; the number of matching groups reaches And basic sub-windows whose time coverage meets the preset requirements are counted as valid basic sub-windows; those below are counted as invalid. The base subwindow is recorded as an invalid base subwindow.
[0031] Within each valid basic sub-window, the median of the observations for the same item in each matching group is taken to obtain the representative value of the item. ,in, This indicates the measurement category, with values including anchor cable stress, anchor bolt tray pressure, or top slab delamination displacement. Indicates the basic sub-window number. These are the test items In the The representative values of the measurement items within each basic sub-window are in the same unit as the corresponding measurement item; when When determining the top slab delamination displacement, the representative value of the measured item for the top slab delamination displacement is... This is denoted as the representative value of displacement, and uniformly denoted as... ,Right now ( (This refers to the top slab delamination displacement, in millimeters).
[0032] An invalid basic sub-window disrupts the time continuity required for subsequent statistics. The regression interval, terminal hold interval, stable state judgment interval, and preset long-term interval are all limited to the same continuous valid basic sub-window segment. The two valid basic sub-windows before and after the invalid basic sub-window are assigned to different segments. After an invalid basic sub-window appears, the starting sequence number of the continuous valid basic window segment is redefined from the first valid basic sub-window that follows it.
[0033] For short-term missing segments in the displayed curve, after obtaining the effective measured observations at both ends of the missing segment, linear interpolation can be performed based on the observations at both ends and their sampling times. The upper limit of the interpolation segment length is preset in the sampling configuration. The interpolated data is marked as supplementary data and only participates in display and equipment diagnosis. The load-bearing reference zone, displacement reference, load-bearing regression interval, displacement retention increment, stable state, and displacement trend category are all calculated using effective measured observations.
[0034] After a missing test or equipment malfunction ends, the continuous valid basic sub-windows will be re-accumulated starting from the basic sub-window where valid measured data is obtained again; each judgment in the subsequent steps will be restored after the corresponding time requirement is met.
[0035] The preset sub-window duration is determined based on the sampling frequency and the number of matching groups required for each window. The lowest configured sampling frequency among the three test items is denoted as... Check if the configuration meets the requirements. The actual number of matching groups is still verified window by window. Within the candidate sub-window durations that meet the quantity condition, the representative value of each measurement item is calculated for the same batch of normal historical data. The change range of the representative value of each measurement item under different candidate durations is compared, and the processing time limit of the monitoring and processing unit is combined to determine the final value. .
[0036] In one configuration of this embodiment, the sampling frequency for each measurement item is once per second. , , The values mentioned above are empirical values for implementation configuration examples. They should be determined based on the data acquisition equipment, synchronization verification results, and on-site data during deployment, and can be adjusted by the implementer according to the specific implementation scenario.
[0037] Before a sensor is put into use, it must be calibrated. The calibration record must include at least the monitoring unit number, sensor number, measurement item name, calibration date, calibration point, indication error, correction factor, measurement margin, calibration validity period, and record version. The measurement margin is determined by the maximum absolute residual after calibration and the reading resolution, and is stored as a positive number.
[0038] Sensor maintenance checks the working status, communication status, time synchronization status, and indication drift according to a preset cycle. In this embodiment, the complete calibration cycle is configured to be six months, which is an empirical value and can be adjusted by the implementer according to the implementation scenario. If the validity period of equipment calibration is shorter than this cycle, calibration is arranged according to the earlier expiration time. When sensor replacement, indication drift exceeding the limit, or equipment failure occurs, calibration or repair is carried out in advance. Sensors that cannot be calibrated online are replaced and calibrated under the maintenance arrangement.
[0039] When calibration or replacement causes changes in the reading scale, first verify the conversion relationship between the old and new data; before the verification of the same reading scale is completed, the corresponding time period is recorded as incomparable data, the existing path record is retained and the automatic shutdown judgment of the corresponding event is paused.
[0040] Step 2: Establish the load-bearing reference zone and displacement reference based on the pre-event reference sub-window and keep them fixed during event handling.
[0041] This step uses the mining event time from the mining event record. Based on this, it represents the actual time of occurrence of the mining event, in seconds; the monitoring and processing unit is selected based on the overall location... The effective base sub-window within is used as the reference sub-window. The preset observation duration is in seconds, with an empirical value of 600 seconds, which can be adjusted by the implementer according to the specific implementation scenario.
[0042] Let the set of indexes of the reference sub-windows be denoted as The minimum number of baseline subwindows required to establish a baseline is denoted as . The number of baseline sub-windows has reached When the time verification, equipment status and calibration records of the corresponding observations all meet the validity conditions specified in step one, the load-bearing reference zone and displacement reference are calculated.
[0043] Establish load-bearing reference zones for the two load-bearing test items, and take the load-bearing test items from each reference sub-window. The median of the representative values of the test items is used as the baseline value. ,Right now Its unit is the same as the corresponding load-bearing measurement item.
[0044] Calculate the load-bearing items in each reference sub-window separately. The preset lower quantile of the representative value of the test item and preset upper quantile value lower quantile level With upper quantile level satisfy This embodiment can be adopted and This value is based on experience and can be adjusted by the implementer according to the specific implementation scenario.
[0045] The same calculation rule is used for all quantiles involved, and the samples are arranged in ascending order according to their sorting position. Linear interpolation values at ... The corresponding quantiles, where, The corresponding sample size; sorting position. When the value is an integer, the sample value at that position is taken.
[0046] Bearing test items The preset measurement margin is denoted as It is read from the calibration record, and the unit is the same as the corresponding load-bearing test item. ; Calculate the lower tolerance separately and upper tolerance .
[0047] Bearing reference zone lower limit Take the baseline value minus the lower tolerance, and the upper limit. Take the baseline value plus the upper tolerance, i.e. , The bearing reference zone is a closed interval. When the representative value of a test item is equal to the lower limit or the upper limit, it is considered to be within the bearing reference zone.
[0048] The median of the displacement representative values in each reference sub-window is taken as the displacement reference. ,Right now The unit is millimeters; the bearing reference zone and displacement reference are stored together with the monitoring unit number, mining event number, reference sub-window number, calibration record version and parameter version, and remain fixed during the processing of the mining event.
[0049] If the number of reference sub-windows is insufficient, the calibration record is invalid, or there is an unresolved time verification anomaly in the reference data before the event, the event will be recorded as reference unevaluable. At this time, the engineering limit value check of the valid measurement items will be retained, the path judgment relying on the bearing reference zone and displacement reference will be suspended, and the reference unevaluable mark will be carried in the indication information.
[0050] Preset observation duration Determined based on the number of baseline sub-windows and changes in data before the event, first by... Cumulative satisfaction The baseline sub-window is then checked to see if the changes in the baseline values calculated by the baseline sub-windows before the two adjacent events are within the range corresponding to the preset measurement margin. If the check fails, the observation time is increased or the baseline is marked as pending review. The time period in the baseline sub-window that has been confirmed to be continuously affected by the previous mining event is removed from the new independent event baseline statistics.
[0051] In one set of configurations in this embodiment, take , ;use Furthermore, when the event time is aligned with the boundary of the basic sub-window, the complete observation interval contains sixty basic sub-windows, and the number of effective basic sub-windows actually participating in the benchmark statistics is verified one by one.
[0052] For multiple mining events occurring consecutively within the same monitoring unit, the preset event merging duration is recorded as follows: The time interval between adjacent mining events is less than or equal to When an event is classified into the same event group, subsequent mining events are also classified into the same event group if the current event group still has activity path records or is still in the process of recovery and confirmation.
[0053] The event group establishes the bearing reference zone and displacement reference using the reference sub-window before the earliest mining event. When subsequent mining events are added to the event group, the frozen reference and existing path records are retained, the event number and the actual occurrence time are added, and the continuous effective base sub-windows after the event are re-accumulated. The steady state candidate count is cleared to zero, and the displacement trend category is marked as pending determination until the long-term interval requirement is met again.
[0054] For duplicate records with the same event number, only the record content is checked; event processing is not initiated repeatedly. Mining events in different monitoring units are processed separately. When all activity path records within an event group have been closed, the current measurement item meets the subsequent closure conditions, and the time since the most recent mining event within the group reaches [a certain threshold], [further action is taken]. When that happens, the recovery confirmation for that event group ends.
[0055] After an event group ends, mining events that occur will create a new event group, and a new baseline sub-window will be selected for the corresponding event. If there are not enough baseline sub-windows for the new event, the event will be treated as not being evaluated.
[0056] Each preset duration, threshold, engineering limit, and displacement range is saved according to the monitoring unit and parameter version. Normal historical events are jointly confirmed by mining operation records, on-site handling records, and review records, and are not only selected based on the automatic closing records of this method. Parameter verification data and review data are separated according to event number. After the parameters are determined, they are recalculated using another set of event data, and false triggering, missed judgment, and processing delays are recorded.
[0057] The preset displacement range used for engineering limit values and closure path records is configured according to the on-site support design and verification records, and is saved separately from the bearing reference zone obtained based on historical statistics.
[0058] Step 3: Determine the load return interval based on the deviation of the load measurement items.
[0059] Following a mining event, the monitoring and processing unit records the continuous valid basic sub-windows from the event time up to the latest basic sub-window as the observation interval, with each measurement item using the same basic sub-window number range; this embodiment uses a preset number of observation sub-windows. Scroll through the observation intervals. It is a positive integer, and its value satisfies the requirement that the length of the comparison start and end intervals be maintained.
[0060] Record the starting number of the current consecutive valid base sub-window segment as Record the latest basic sub-window number as The cumulative number reached At that time, the starting number of the observation interval according to Determine and satisfy If the quantity is insufficient, the load-bearing regression judgment will be marked as pending. After an invalid foundation sub-window appears, it will be re-determined from the first valid foundation sub-window thereafter. When a new mining event is added to an event group, the first complete and valid base subwindow since that event is redefined. .
[0061] Calculate bearing capacity measurement items In the Deviation of each basic sub-window Its unit is the same as the corresponding load-bearing measurement item; when At that time, take ,when At that time, take ,when At that time, take .
[0062] Bearing measurement items within the observation interval The maximum deviation is denoted as The deviation of the end base sub-window is Subtract the final deviation from the maximum deviation to obtain the net reduction. Its unit is the same as the corresponding load-bearing measurement item, and it is calculated according to the following formula.
[0063]
[0064] For the observation interval with the sequence number The base sub-window will start from the starting number. The minimum deviation from the base sub-window is denoted as ,in The basic sub-window number within this range; Calculate the reverse bias increment, and record the maximum value among the reverse bias increments as the maximum reverse bias increment. Calculate using the following formula.
[0065]
[0066] The maximum reverse bias check measures the increase in deviation within the observation interval relative to the previously reached minimum deviation. All biased segments within the interval participate in the calculation under the same rule.
[0067] The monitoring and processing unit uses sequential recursive calculation, starting with the initial basic sub-window. , Calculate the subsequent basic sub-windows in sequence. and After traversing to the last basic sub-window, retrieve... The maximum deviation is updated in the same traversal.
[0068] For a given observation interval, the above calculation performs one sequential traversal, with a time complexity of O(log n). In addition to the already saved observation sequences, the additional space required for computation is... When the starting point of the observation interval changes with the rolling process, the traversal is re-executed for the new interval, and the minimum value statistics are limited to the current interval.
[0069] Setting load-bearing test items Preset net reduction threshold and preset reverse bias threshold , It is a positive number. Both are non-negative, and their units are the same as the corresponding load-bearing measurement item; at least one load-bearing measurement item satisfies... And each load-bearing test item satisfies When this period is reached, the observation interval is recorded as the bearing regression interval.
[0070] If the observation interval includes the segment of increased deviation after a mining event, and thus exceeds the preset reverse deviation threshold, the current load regression judgment is recorded as invalid; after the subsequent basic sub-window arrives, the rolling interval check continues, and event processing continues.
[0071] Preset net reduction threshold Based on valid historical data from the same monitoring unit without mining disturbance, the net reduction in these historical intervals is calculated using the same length as the actual observation interval, and its preset upper quantile value is taken plus twice the measurement margin. The larger one is selected as the candidate value.
[0072] Preset reverse bias threshold Based on the normal load-bearing decline process verified on-site, the maximum reverse deviation was calculated according to the same observation interval selection rules, and its preset upper quantile value was taken. The larger one is selected as the candidate value.
[0073] In this embodiment, the 99th percentile value can be used as the upper quantile value. The candidate value is determined as the threshold of the corresponding parameter version after independent historical event recalculation. When the observation interval length, sampling configuration or sensor combination changes, the corresponding threshold is re-verified.
[0074] The load-bearing capacity regression interval is determined by using both the net decrease and the maximum reverse increase in skewness. This is to identify a substantial decline and limit secondary surges, and to avoid misjudging temporary declines after multiple disturbances as load-bearing capacity regression.
[0075] Step 4: Calculate the displacement increment and add judgment criteria or establish an activity path record when the increment exceeds the preset threshold.
[0076] Once the load regression interval is established, the monitoring and processing unit reads the preset response hysteresis duration. Keep for the preset duration ; This indicates the time delay between the starting point of the comparison and the starting point of the regression interval. This indicates the duration required to hold the endpoint, both in seconds.
[0077] The number of basic subwindows corresponding to the response hysteresis duration is denoted as . The number of basic sub-windows required to maintain the end interval is denoted as ;in This embodiment sets The actual delay time and hold time used are respectively the number of integer sub-windows mentioned above and... The product is saved with the parameter version.
[0078] Self-supporting regression interval starting point Delay Each basic sub-window is used as the starting point for comparison, and its sequence number is [number missing]. ; End of self-supporting regression interval reverse selection A series of consecutive valid base sub-windows are used as the end-hold intervals, with their starting sequence number being... The set of serial numbers is denoted as .
[0079] The comparison between the starting and ending points must satisfy the following conditions: The corresponding observation interval length condition is: The comparison start point does not exist, the end-hold interval length is insufficient, or the relevant basic sub-window is invalid. When the displacement increment is calculated, it is recorded as a variable to be calculated.
[0080] The displacement representative value of the comparison starting point is The end of the interval is kept in the sequence number. The displacement representation value of the base sub-window The displacement retention increment is obtained by subtracting the displacement representative value at the comparison starting point from the average value of all representative displacement values within the end retention interval. Its unit is millimeters, and it is calculated using the following formula:
[0081] Displacement retains increment It can be positive, zero, or negative. When data conditions are insufficient, the mark to be calculated is retained. Existing displacement data and path records are saved separately.
[0082] The preset response hysteresis duration is determined by pairing records of load changes and displacement responses in normal historical events. For historical events with identifiable displacement responses, the time when the load measurement item first leaves its reference zone and the first time when the displacement increment relative to the corresponding reference starting point exceeds twice the displacement measurement margin for three consecutive basic sub-windows are recorded. The positive part of the difference between the subsequent time and the previous time is taken as the hysteresis sample, and the preset upper quantile value of the hysteresis sample is taken, and then... Round up to determine the hysteresis duration of the candidate response.
[0083] The measurement margin of the displacement measurement item is denoted as The unit is millimeters. Historical events without identifiable displacement responses are retained as event records. Hysteresis duration statistics only use samples with paired response times.
[0084] The preset hold duration is determined based on the duration of short-term displacement drop and recovery in normal historical events. The average displacement at the end and the displacement retention increment are recalculated according to the candidate hold duration, and the results are checked with independent event data. The response hysteresis duration, hold duration and the number of observation sub-windows are jointly verified, and the parameters are written into the parameter record after the aforementioned interval length conditions are met.
[0085] Set a preset incremental threshold Its unit is millimeters and Following the same comparison start and end interval rules as in this step, calculate the displacement increment for the load-bearing regression interval that has been established in normal historical events, and take its preset upper quantile value. The larger of the values is used as the candidate threshold, which is then determined by recalculation of independent events.
[0086] when Establish an active path record and save the monitoring unit number, event group number, baseline version, parameter version, start and end sequence number of the load return interval, net reduction and maximum reverse deviation of the two load measurement items, comparison start point, end holding interval, displacement retention increment and establishment time.
[0087] If there are already activity path records for the same event group, and the subsequent rolling interval meets the establishment conditions again, the judgment criteria for this instance will be added and the last judgment time will be updated, while the judgment criteria for the first establishment will be retained; for duplicate processing of the same group of monitoring data and the same interval number, only existing records will be checked.
[0088] when At this time, the displacement increment of this interval is saved according to the calculation result, and the previously established activity path records are retained and processed according to subsequent steps five and six.
[0089] The activity status and monitoring level of the path record are saved separately. The monitoring level of a newly created path record is recorded as the continuous tracking level. The activity status indicates that the record is still participating in subsequent judgments, and the monitoring level indicates the prompt level corresponding to the current record. Both are updated separately.
[0090] Existing activity path records are retained during the data interruption, and the start and end times of this data interruption are marked.
[0091] The displacement retention increment setting includes response hysteresis and end-hold duration to accommodate the lag in anchored rock mass deformation, avoid short-term fluctuations in the initial stage of load return, and confirm whether the displacement has been truly retained.
[0092] Step 5: Determine the status flags based on the changes in the measurement items and retain the existing path records.
[0093] The monitoring and processing unit checks the rate of change and fluctuation of each measurement item during the basic sub-window update process, and performs a thorough examination of the time-continuous and valid sub-windows. The and the first Each basic sub-window calculates the difference between representative values of the test items. The rate of change is obtained from this. and fluctuations The rate of change is measured in units per second for the corresponding measurement item, and the unit of fluctuation is the same as for the corresponding measurement item. After a measurement is missed or an equipment malfunction occurs, the rate of change calculation resumes after two consecutive valid baseline sub-windows are obtained.
[0094] Set test items Rate limit With fluctuation limits Check them separately and Under a fixed sub-window duration, both checks can also be performed. Joint verification is performed, and the rate of change and fluctuation are calculated from the same adjacent difference, with the corresponding units and configuration limits retained respectively.
[0095] Set the preset time required for steady state determination. The unit is seconds. The number of adjacent basic sub-window pairs that continuously pass all rate checks and fluctuation checks is recorded as follows: For each newly added pair of valid basic sub-windows that are adjacent in time and have passed all checks, Add one more time to satisfy At that time, the status flag will be set. This is denoted as a stable state.
[0096] The number of basic sub-windows that pass the inspection consecutively is At that time, the corresponding number of valid basic sub-windows is If the current basic sub-window is valid but the rate of change or fluctuation of any measurement item exceeds the corresponding limit, it will... Record it as pending confirmation, and Reset to zero.
[0097] All test items passed the inspection, but the cumulative duration has not yet reached the target. hour, Remain in a pending confirmation state; if the current base sub-window is invalid, This is recorded as a data shortage state, and... Reset to zero.
[0098] Rate limits and fluctuation limits are determined based on normal historical data verified on-site, for the same... The absolute value and fluctuation of the rate of change are calculated for each time-adjacent and valid sub-window, and the preset upper quantile values are used for each. The rate limit is simultaneously checked against the differential range corresponding to the measurement margin. The fluctuation limit was checked simultaneously. The larger of the corresponding upper quantile value and the measurement margin range is used as the candidate value, and the measurement margin of the displacement measurement item is taken as... .
[0099] The duration for determining a stable state is determined based on the duration of disturbances in normal historical events and the stability confirmation records, along with the limits for each measurement item. In the current event group processing, the rate limit, fluctuation limit, and bearing reference band width are kept constant and configured separately.
[0100] Introducing re-trigger count and manual review threshold , Let be a non-negative integer, representing the number of times a stable candidate process that has already begun is cancelled due to changes in test items exceeding the limit. It is a positive integer; when there are already consecutively passed basic sub-window pairs, and subsequent valid basic sub-windows cause the candidate process to be canceled due to excessive rate of change or fluctuation, it will be... Add one more time.
[0101] The same consecutive out-of-limit segment is counted as only one re-trigger; data interruption counts due to data loss or equipment failure are recorded separately. When the time comes, send a manual review request containing the corresponding event group, the over-limit test item, the time of occurrence, the number of re-triggers, and the existing path record number, and retain the existing path record.
[0102] status flags Update the activity status of the path record separately. When the status flag changes to a stable state, the existing path record is still judged according to the closing condition in step six. When the status flag changes to a pending confirmation state or a data insufficient state, the existing path record continues to be retained.
[0103] When a new mining event occurs within an event group, stable candidate processes are re-accumulated, and previous re-trigger records and path records of that event group are retained; when the duration is insufficient, the measurement item exceeds the limit again, or the data is interrupted, existing path records remain active.
[0104] Resetting stability confirmation when creating or adding activity path records is to prevent the previously accumulated stability duration from being mistakenly closed for subsequent newly added risk records, and to ensure that the stability status truly reflects the data changes after the addition of the basis.
[0105] Step Six: Determine the displacement trend category and send an instruction message based on the path record and the comparison results with the engineering limit.
[0106] The monitoring and processing unit updates the status indication information after each basic sub-window is completed, and records the latest completed basic sub-window as the newest basic sub-window, with its sequence number as follows: When the latest basic sub-window is valid, its displacement value is used as the representative value. Subtract displacement reference The net displacement increment is obtained. The unit is millimeters.
[0107] When the latest basic sub-window is invalid, the current net displacement increment is recorded as pending determination, the historical net displacement increment along with its corresponding time is retained, and recorded separately from the current judgment result.
[0108] Set preset duration The unit is seconds, and the corresponding number of basic child windows is This embodiment is set up and the actual long-term interval duration will be used. Write the parameter record.
[0109] Select in reverse, starting with the latest basic sub-window. A preset long-term interval is formed by a series of consecutive valid basic sub-windows, with its starting sequence number being... The entire interval is located after the mining event time. When using event group processing, it must also be located after the most recent mining event within the group and within the current continuous valid basic sub-window segment corresponding to that event.
[0110] The displacement range is obtained by subtracting the minimum value from the maximum value of the representative displacement over a long-term interval. Calculate the representative value of the end displacement and subtract the representative value of the beginning displacement to obtain the difference between the beginning and end displacements. Both units are millimeters.
[0111] Starting from the base sub-window For independent variable and displacement representative value As the dependent variable, the linear slope is obtained through least squares fitting. The average value of the starting time of the basic sub-window within the long-term interval is denoted as... The average value of the displacement representative value is denoted as ,according to Calculate the slope, with the unit being millimeters per second.
[0112] Set preset amplitude threshold Preset trend threshold and preset rate threshold The first two are in millimeters, and the latter is in millimeters per second; all three parameters are positive numbers.
[0113] when and When the displacement trend category is defined as the first category, then the displacement trend category is defined as the first category. and When the displacement trend category is defined, it is recorded as the second category.
[0114] If the data conditions for a long-term interval are met, but the situation falls outside the first and second categories, the displacement trend category will be recorded as the third category. This includes situations such as cyclical changes, declines, and situations where some indicators reach the threshold while others do not. If the number of long-term intervals is insufficient, the interval crosses an invalid basic sub-window, or contains data prior to a new mining event, the displacement trend category will be recorded as pending determination. The determination of the third category is based on the premise that the long-term interval meets the aforementioned data conditions.
[0115] The preset long-term duration is determined based on the duration of disturbances and displacement recovery observation records of normal historical events, and is verified together with the preset stable state judgment duration. Using the same long-term interval length, the displacement range, the absolute value of the first and last difference, and the absolute value of the linear slope are calculated for the normal historical data confirmed on site, and the corresponding preset upper quantile value is taken as the threshold candidate value.
[0116] The thresholds corresponding to the displacement range and the first-to-last difference are simultaneously checked against a displacement measurement margin of twice the normal value. Take the corresponding upper quantile value and The larger of the two values is used as the candidate value; the upper bound of the measurement margin corresponding to the linear slope is denoted as . ,according to calculate.
[0117] The preset rate threshold uses the preset upper quantile value of the absolute value of the historical linear slope and... The larger of the values is selected as the candidate value and determined after independent event recalculation; the upper bound of the measurement margin is calculated using the absolute value of the window-by-window displacement error. The configuration of constraints is a prerequisite; data exceeding this prerequisite is processed according to the calibration and quality rules in step one.
[0118] Compare the representative values of each measurement item in the latest basic sub-window with the corresponding preset engineering limits. For measurement items using upper and lower limits, record the lower and upper limits of the engineering limit as follows: and ;when When the engineering limit comparison result of the test item is within the limit, it is recorded as exceeding the limit if it is lower than the lower limit or higher than the upper limit. Test items using unilateral engineering limit values are checked according to the configured unilateral boundary.
[0119] If any valid measurement item exceeds the limit, the result of exceeding the limit is retained; if there is a missing measurement item and all other valid measurement items are within the limit, the overall engineering limit comparison result is recorded as pending determination.
[0120] The presence or absence of project limit checks and activity path records are handled separately. When an activity path record exists, the second instruction information carries the project limit comparison result. In other cases, the first instruction information carries the project limit comparison result. Project limit exceedance results that are not subject to load regression judgment and path record establishment conditions are also sent.
[0121] For the original observations that have passed the equipment status and range verification, a point-by-point engineering limit check can be set up in parallel with the basic sub-window processing; the limit-breaking results of the original observations and the limit-breaking results of the representative values of the measurement items are respectively marked with the data source and the time of occurrence, and are saved separately.
[0122] Set preset displacement range ,in and These are the lower and upper limits of the net displacement increment allowed for closing path recording, respectively, in millimeters. This range is determined based on on-site support requirements and verification records, and remains fixed during incident handling.
[0123] Perform a close check on each activity path record, and when the status flag is set... In a stable state, the representative values of the two load-bearing items in the latest basic sub-window are both within the corresponding load-bearing reference zone, and the displacement trend category is Category 1. When all representative values of the measured items in the latest basic sub-window are within the corresponding preset engineering limits, close the path record.
[0124] To close activity path records, short-term stability, normal load capacity, acceptable displacement, and stable long-term trends must be met simultaneously to avoid mistakenly closing risk records due to short-term stabilization or local fluctuations.
[0125] In this embodiment, the basis for establishing a newly established path record is first saved, and the closing judgment is performed starting from the next basic sub-window. When closing, the closing time, the basic sub-window number on which the closing is based, the status flag, the representative values of the two load-bearing measurement items, the displacement trend category, the net displacement increment, and the comparison results of the engineering limit value are saved.
[0126] When all four conditions regarding the steady state, the location of the load-bearing measurement item, the net displacement increment, and the engineering limit are met, the displacement trend category is the third category, and When this happens, the monitoring level of the activity path record is recorded as observation level; observation level path records continue to participate in the update and closure judgment of each basic sub-window, and the activity status remains unchanged.
[0127] When the displacement trend category is Category 2, the displacement trend category is to be determined, the data is insufficient, the net displacement increment exceeds the preset displacement range, any load-bearing measurement item leaves the corresponding load-bearing reference zone, the status flag exits the stable state, or any measurement item exceeds the limit, the monitoring level of the corresponding activity path record will be recorded as continuous tracking level, and the specific reason will be recorded; other activity path records that are not subject to closing conditions and observation conditions will also be recorded as continuous tracking level.
[0128] Information on engineering overruns and equipment malfunctions will be retained as separate notification items and will continue to be included when the monitoring level is adjusted. Path records that have been closed will be transferred to the history record, and activity record statistics will only include path records that are still active.
[0129] After a record is closed, the starting point of the observation interval used to identify a new path must be later than the latest base sub-window when the record was closed. When the new observation interval meets the path establishment conditions again, a new path record is established and associated with the previously closed record number.
[0130] The status indication information includes at least the monitoring unit number, event group number, latest basic sub-window time, path record number and activity status, monitoring level, status flag, displacement trend category, net displacement increment, engineering limit comparison result, and data quality mark.
[0131] When an activity path record exists, a second indication message containing the corresponding status indication information is sent; otherwise, a first indication message is sent, along with the current engineering limit comparison result and data quality mark. When the conditions for calculating displacement trend and net displacement increment are met, the first indication message also carries the corresponding results.
[0132] The first indication information corresponds to the case where the current activity path record is empty. When the displacement trend category is the second category, the project exceeds the limit, the benchmark cannot be evaluated, or the data is insufficient, the corresponding mark is retained respectively, and the recovery conditions are confirmed separately according to the aforementioned inspection results.
[0133] Each update is executed in the following order: data quality verification, update of representative values of measurement items, check of engineering limits, check of path establishment conditions, calculation of stable state and displacement trend, determination of path classification or closure, and sending of indication information. The indication information adopts the path activity status after the update is completed, and the intermediate status during the path establishment and closure process is saved separately in the processing record.
[0134] The path records are stored using independent record numbers. The record content includes monitoring unit number, event group number, baseline version, parameter version, basis for initial establishment, basis for subsequent updates, activity status, monitoring level, data interruption record, manual review record, basis for closure, and update time.
[0135] Different monitoring units maintain their own data queues and processing statuses. Updates within the same monitoring unit and event group are executed sequentially according to the basic sub-window number. Path record updates are committed using a transaction method. When the same basic sub-window data is received repeatedly, the data version is checked, and the same version is processed only once.
[0136] The path records can be queried by monitoring unit, event group, activity status, monitoring level, and time range, and the number of activity records, observation-level records, continuous tracking-level records, and data interruption times can be counted.
[0137] In this embodiment, the above process is performed once after each basic sub-window is completed, so that... For example, the maximum waiting time for late data after the basic sub-window is completed is configured as follows: The processing time limit for this state calculation is configured as follows: The instruction message sending time limit is configured as follows: .
[0138] If the measurement data and event logs arrive within the agreed time limit and the required continuous window conditions are met, the configured time limit for sampling to send the corresponding indication information will be determined according to... Verification; the cumulative observation time required for response lag, end-point maintenance, stable state confirmation, and long-term intervals is counted separately. The processing device performs concurrent load acceptance according to the number of configured monitoring units. If the processing time limit is exceeded, an overtime mark is recorded and an equipment operation prompt is sent.
[0139] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make several improvements and substitutions without departing from the technical principles of the present invention, and these improvements and substitutions should also be considered within the scope of protection of the present invention.
Claims
1. A method for early warning of instability risk in anchored rock mass under mining disturbance, characterized in that, include: Obtain the anchor cable force, anchor bolt tray pressure, roof delamination displacement and mining event time of the same monitoring unit, determine the representative value of the measurement item according to the foundation sub-window, take the anchor cable force and anchor bolt tray pressure as the bearing measurement item, and take the roof delamination displacement measurement item representative value as the displacement representative value. The effective basic sub-window within the preset observation period before the event is used as the reference sub-window. The upper and lower limits of the corresponding load reference band are determined by the representative values of each load measurement item in the reference sub-window. The median of the displacement representative values of the reference sub-window is used as the displacement reference. The load reference band and the displacement reference remain fixed during the event processing. The observation interval is formed by consecutive effective basic sub-windows after the event. The deviation is calculated according to the out-of-band distance from the representative value of each load-bearing measure to the corresponding load-bearing reference band. The observation interval in which the net reduction of the difference between the maximum value of the deviation of at least one load-bearing measure and the end value is greater than the corresponding preset net reduction threshold, and the maximum reverse deviation of each load-bearing measure is less than or equal to the corresponding preset reverse deviation threshold, is recorded as the load-bearing regression interval. Within the load return interval, a comparison start point and an end hold interval are determined according to a preset response hysteresis duration and a preset hold duration, wherein the comparison start point is earlier than or equal to the start point of the end hold interval; The displacement retention increment is obtained by subtracting the displacement representative value of the comparison starting point from the average value of the displacement representative values within the end retention interval. The latest completed basic sub-window is recorded as the newest basic sub-window. When the newest basic sub-window is valid, the net displacement increment is obtained by subtracting the displacement reference from the displacement representative value of the newest basic window. When the displacement increment is greater than the preset increment threshold, additional judgment criteria are added or an activity path record is established. The activity path record remains active until the closing condition is met, and instruction information is sent based on the activity status of the activity path record and the comparison result between the representative value of the measurement item and the preset engineering limit value. The instruction information includes a status flag, displacement trend category and the net displacement increment.
2. The method for early warning of instability risk of anchored rock mass under mining disturbance as described in claim 1, characterized in that, The representative values of the test items include: The effective measured data of each measurement item that has been time-calibrated and meets the preset synchronization tolerance are grouped into matching groups. The basic sub-window with the number of matching groups in the same basic sub-window reaching the preset number and the time coverage meeting the preset requirements is recorded as a valid basic sub-window. The basic sub-window with the number of matching groups less than the preset number is recorded as an invalid basic window. The median of the observations in the matching group within the effective basic sub-window for each measurement item is taken to obtain the representative value of the measurement item; the invalid basic sub-window interrupts the time continuity required for subsequent statistics.
3. The method for early warning of instability risk of anchored rock mass under mining disturbance as described in claim 1, characterized in that, The bearing reference band includes: The effective basic sub-window within the preset observation period before the event is recorded as the reference sub-window, and the median of the representative values of each of the load-bearing test items in the reference sub-window is the corresponding baseline value. The absolute values of the differences between the preset lower quantile value and the preset upper quantile value of the representative value of each load-bearing test item in the reference sub-window and the corresponding baseline value are respectively added to a preset measurement margin to obtain the lower tolerance and the upper tolerance. The lower limit and the upper limit of the corresponding load-bearing reference band are obtained by subtracting the lower tolerance from the baseline value and adding the upper tolerance.
4. The method for early warning of instability risk of anchored rock mass under mining disturbance as described in claim 1, characterized in that, The calculation of the maximum reverse bias includes: For each load-bearing test item, the deviation when the representative value of the load-bearing test item is within the corresponding load-bearing reference band is recorded as zero; The deviation of each basic sub-window within the observation interval is subtracted from the minimum deviation from the starting point of the observation interval to that basic sub-window to obtain the reverse bias increment. The maximum value of each reverse bias increment is taken as the maximum reverse bias increment.
5. The method for early warning of instability risk of anchored rock mass under mining disturbance according to claim 1, characterized in that, The displacement retention increment is calculated using the following windowing method: The first basic sub-window after delaying the preset response hysteresis time from the starting point of the bearing regression interval is recorded as the comparison starting point; The consecutive effective basic sub-windows selected from the end of the bearing regression interval and whose cumulative duration reaches the preset holding duration are denoted as the end holding interval.
6. The method for early warning of instability risk of anchored rock mass under mining disturbance according to claim 1, characterized in that, The status flags include: The ratio of the difference between the representative values of the measured items of two adjacent basic sub-windows that are both effective basic sub-windows to the preset sub-window duration is used as the rate of change, and the absolute value of the difference is used as the fluctuation. When the absolute value of the rate of change and the fluctuation of each measurement item are within the corresponding limit and continue for a preset duration, the state flag is recorded as a stable state; when creating or adding the activity path record, the cumulative time of the stable state is reset.
7. The method for early warning of instability risk of anchored rock mass under mining disturbance according to claim 6, characterized in that, The displacement trend categories include: Using the latest basic sub-window as the end, select consecutive effective basic sub-windows whose cumulative duration after the event reaches a preset long duration, and calculate the displacement range, the first-to-last difference, and the linear slope. The first-to-last difference is the representative value of the end displacement of the consecutive effective basic sub-window minus the representative value of the first displacement. When the displacement range is less than or equal to a preset amplitude threshold and the absolute value of the linear slope is less than or equal to a preset rate threshold, the displacement trend category is recorded as the first category; when the difference between the first and last values is greater than a preset trend threshold and the linear slope is greater than the preset rate threshold, the displacement trend category is recorded as the second category; when the data conditions for the long-term interval are met, but the situation falls outside the first and second categories, the displacement trend category is recorded as the third category; when the data is insufficient, the displacement trend category is recorded as pending determination. All of the above thresholds are positive numbers.
8. The method for early warning of instability risk of anchored rock mass under mining disturbance according to claim 7, characterized in that, The status update of the activity path record includes: When the status flag changes, the activity path record is retained; for the activity path record, when the status flag is the stable state, the representative values of each load-bearing item in the latest basic sub-window are all within the corresponding load-bearing reference zone, the net displacement increment is within the preset displacement range, and the representative values of each item are all within the corresponding preset engineering limit, if the displacement trend category is the first category, the activity path record is closed and saved; if the displacement trend category is the third category and the absolute value of the linear slope is less than or equal to the preset rate threshold, the activity path record is recorded as observation level. The activity path records in other cases are denoted as continuous tracking level.
