Rock mass damage prediction method and system based on microseismic data

Abstract

The rock mass damage prediction method and system based on microseismic data disclosed by the present application belong to the technical field of rock mass damage monitoring and prediction, and obtain the event initial motion direction, frequency band offset trajectory and energy release duration in the monitoring area; the event inter-direction consistency discrimination rule is constructed based on the event initial motion direction and the frequency band offset trajectory, and the event cluster set is screened; each event in the event cluster set is reorganized according to the occurrence sequence, and the transfer sequence representing the energy migration continuity is generated in combination with the energy release duration; the abnormal convergence section is identified according to the transfer sequence, the energy aggregation mutation index and the time shrinkage coefficient are extracted; the energy aggregation mutation index and the time shrinkage coefficient are coupled and analyzed to form the damage evolution critical criterion; when the damage evolution critical criterion is triggered, the rock mass local damage development trend and the instability prediction result are output in combination with the spatial distribution range of the event cluster; the present application can improve the accuracy of early identification and instability prediction of rock mass local damage.

Description

Technical Field

[0001] This invention relates to the field of rock mass damage monitoring and prediction technology, specifically to a method and system for predicting rock mass damage based on microseismic data. Background Technology

[0002] In deep underground engineering projects such as high-stress tunnels or local unloading zones of slopes, microseismic signals often exhibit low energy, strong noise, and multi-source superposition characteristics. Especially in the initiation stage of local microcracks, traditional methods based on single-event counting or energy statistics are difficult to capture the subtle evolution of early damage in a timely manner, which can easily lead to a lag in the identification of precursors to sudden instability. At the same time, the signal propagation paths between different monitoring points are complex and there are local obstructions, making it difficult to continuously characterize the spatial expansion process of damage, resulting in the difficulty in timely prediction of small-scale hidden damage. Summary of the Invention

[0003] The purpose of this invention is to provide a rock mass damage prediction method and system based on microseismic data to address the shortcomings of the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a rock mass damage prediction method based on microseismic data, comprising: The system acquires raw microseismic signals from multiple points within the monitoring area, performs adaptive filtering and event identification to obtain multi-point microseismic event signals, and extracts the initial motion direction, frequency band offset trajectory, and energy release duration of the events. Based on the initial direction of the event and the frequency band offset trajectory, a rule for determining the direction consistency between events is constructed, and a set of event clusters with the characteristic of direction convergence is formed by filtering. The events in the event cluster are reorganized in the order of occurrence, and combined with the duration of energy release, a transfer sequence characterizing the continuity of energy migration is generated. Based on the energy connectivity and time compression characteristics of adjacent events in the transmission sequence, abnormal convergence segments are identified, and energy aggregation mutation indices and time contraction coefficients are extracted. By coupling the energy accumulation mutation index with the time contraction coefficient, the critical state of microseismic activity transforming from dispersed to concentrated is determined, thus forming a critical criterion for damage evolution. When the critical criterion for damage evolution is triggered, the local damage development trend and instability prediction results of the rock mass are output, based on the spatial distribution range of the corresponding event cluster.

[0005] Preferably, the extraction of the initial motion direction of the event includes: performing multi-channel arrival time picking on the identified multi-point microseismic event signals to determine the initial motion arrival time sequence of each monitoring point; based on the initial motion arrival time sequence and combined with the spatial layout relationship of the monitoring points, calculating the time difference distribution of the event wavefront propagation and constructing the propagation time gradient feature; and comparing the propagation speed in different directions according to the propagation time gradient feature to determine the energy-priority propagation direction and generate the initial motion direction of the event.

[0006] Preferably, extracting the frequency band offset trajectory includes: performing segmented time-frequency analysis on the identified multi-point microseismic event signals to obtain a frequency band energy distribution sequence that changes over time; based on the frequency band energy distribution sequence, extracting the frequency band with the highest energy proportion and its change over time to form a dominant frequency evolution sequence; calculating the frequency change amplitude and direction between adjacent time points according to the dominant frequency evolution sequence to construct frequency band offset change features; and using the frequency band offset change features to correlate continuous time periods to generate a frequency band offset trajectory.

[0007] Preferably, extracting the energy release duration includes: performing envelope extraction processing on the identified multi-point microseismic event signals to obtain an energy envelope curve reflecting changes in signal intensity; based on the energy envelope curve, setting an energy threshold, determining the energy start-up point and attenuation termination point, and forming the boundary of the effective energy interval; calculating the corresponding time span according to the boundary of the effective energy interval and correcting it in conjunction with the energy attenuation rate to obtain an initial duration parameter; using the initial duration parameter to perform consistency screening on adjacent events, eliminating abnormally short or long-term events, and outputting the energy release duration.

[0008] Preferably, the construction of the inter-event direction consistency judgment rule includes: Based on the initial motion direction of adjacent microseismic event signals, the degree of directional deflection between successive events is calculated to form a continuous directional change sequence. Based on the continuous directional change sequence, combined with the frequency band offset trajectory of the corresponding microseismic event signal, the associated segment where the frequency band offset direction and the degree of directional deflection change synchronously are extracted to form a direction-frequency band coordinated change feature. Based on the aforementioned direction-frequency band coordinated change characteristics, three types of discrimination conditions are set: same-direction continuity, opposite-direction shift, and abrupt change, and a direction consistency discrimination rule between events is constructed. Using the direction consistency discrimination rule between events, the temporally adjacent microseismic event signals are compared one by one to filter out the associated events that meet the continuous propagation characteristics, and the direction consistency discrimination rule is output.

[0009] Preferably, the filtering process forms a set of event clusters with directional convergence characteristics, including: Based on the direction consistency discrimination rule, the temporally adjacent microseismic event signals are initially merged to form candidate event groups; For each candidate event group, the initial motion direction and frequency band offset trajectory of each microseismic event signal within the group are combined to calculate the degree of concentration of direction change and the degree of continuation of frequency band offset in the same direction, and generate direction convergence evaluation parameters. Based on the directional convergence evaluation parameters, the candidate event groups are screened, and event groups with continuous directional changes and coordinated evolution of frequency band offset are retained to form directional convergence event units. The event units that converge in direction are merged according to temporal continuity and spatial proximity to output an event cluster set.

[0010] Preferably, the generation of the transfer sequence characterizing the continuity of energy migration includes: Based on the event cluster set, the microseismic event signals are arranged in the order of their occurrence. Combined with the time interval relationship between adjacent microseismic event signals, a sequential transmission event chain is formed. For the sequential transmission event chain, the energy release duration corresponding to each microseismic event signal is extracted, and the duration connection changes between the microseismic event signals are compared to generate duration transmission characteristics. Based on the duration transmission characteristics, adjacent microseismic event signals with gradually extending or continuous durations are identified, and energy continuity segments are constructed. The energy continuity segments are then connected in series according to their occurrence order to output a transmission sequence. Based on the transmission sequence, the degree of energy connection and the change in the time interval between adjacent microseismic event signals are calculated to form an energy-time continuity feature sequence. According to the energy-time continuity feature sequence, adjacent microseismic event signal segments with continuously increasing energy connection and continuously shortening time intervals are identified and determined as abnormal convergence segments.

[0011] Preferably, the extraction of energy accumulation abrupt change index includes: based on the abnormal convergence segment, extracting the duration of energy release and the order of events corresponding to each microseismic event signal within the segment to form a segment energy succession sequence; comparing the duration increments and continuous trends between adjacent microseismic event signals to identify the abrupt change starting position where the duration changes from a stable continuity to rapid growth; based on the abrupt change starting position, calculating the cumulative difference and growth factor of the duration of adjacent microseismic event signals before and after the abrupt change to generate an energy accumulation abrupt change index; and extracting the time contraction coefficient includes: based on the abnormal convergence segment, extracting the time interval between adjacent microseismic event signals within the segment; comparing the shortening magnitude and continuous trend of adjacent time intervals before and after the abrupt change to identify the contraction starting position where the time interval changes from a slow decrease to rapid compression; calculating the cumulative reduction and compression ratio of the time interval between adjacent microseismic event signals before and after the contraction to generate a time contraction coefficient used to represent the degree of acceleration of the event triggering rhythm within the segment.

[0012] Preferably, the critical criteria for damage evolution include: based on the energy accumulation mutation index and the time contraction coefficient, extracting the corresponding change relationship between the two within the same abnormal convergence segment, identifying the overlapping segment where the energy accumulation mutation index continuously increases and the time contraction coefficient increases synchronously, and determining it as the concentrated transformation sensitive segment; The degree of enhanced coordination between the energy concentration mutation index and the time contraction coefficient is calculated to generate critical coupling characteristic values ​​that characterize the intensity of the transformation of microseismic activity from dispersion to concentration. Based on the critical coupling characteristic values ​​and preset judgment conditions, the state of abnormal convergence sections is judged. When the critical coupling characteristic values ​​reach the triggering requirements, the critical criterion for damage evolution is formed.

[0013] This invention also provides a rock mass damage prediction system based on microseismic data, comprising: Data acquisition module: acquires raw microseismic signals from multiple points within the monitoring area, performs adaptive filtering and event identification to obtain multi-point microseismic event signals, and extracts the initial motion direction, frequency band offset trajectory and energy release duration of the events; Event cluster filtering module: Based on the initial direction of the event and the frequency band offset trajectory, construct the direction consistency discrimination rule between events, and filter to form a set of event clusters with direction convergence characteristics; Transmission sequence construction module: Reorganizes each event in the event cluster set according to the order of occurrence, and combines the duration of energy release to generate a transmission sequence that characterizes the continuity of energy migration; Anomaly Convergence Identification Module: Based on the energy connectivity and time compression characteristics of adjacent events in the transmission sequence, it identifies abnormal convergence segments and extracts energy aggregation mutation indicators and time contraction coefficients. Critical state determination module: The energy accumulation mutation index is coupled with the time contraction coefficient to determine the critical state of microseismic activity from dispersion to concentration, and a critical criterion for damage evolution is formed. Instability prediction module: When the critical criterion for damage evolution is triggered, the module outputs the local damage development trend and instability prediction results of the rock mass, based on the spatial distribution range of the corresponding event cluster.

[0014] The technical effects and advantages provided by the present invention in the above technical solution are as follows: 1. This invention introduces a collaborative analysis of the initial propagation direction of events and the frequency band offset trajectory, constructs a directional consistency discrimination rule, and filters to form a set of event clusters with directional convergence characteristics, thereby achieving a joint characterization of the spatial expansion path and internal structural evolution of microseismic activity. Compared with existing technologies that rely solely on energy or event counts for statistical analysis, this invention, by identifying the continuity of the propagation direction of microseismic waves and the consistency of frequency component migration, can reflect the stable evolution characteristics of stress release paths during crack propagation. This effectively distinguishes random noise interference from the actual damage evolution process, solves the problem of difficulty in identifying weak damage stages, and significantly improves the reliability of early damage identification.

[0015] 2. This invention constructs a transmission sequence of energy migration continuity based on the duration of energy release, and extracts energy accumulation mutation indicators and time contraction coefficients by identifying anomalous convergence segments, thereby achieving a quantitative characterization of the transformation process of microseismic activity from dispersed to concentrated. This technique directly corresponds to the physical process of fractures within the rock mass from dispersed initiation to localized interconnection and accumulation. Compared to traditional methods that struggle to characterize the coupled evolution of energy concentration and event acceleration, this invention, through the synergistic determination of energy accumulation and time compression, can accurately capture the critical state of damage evolution. Simultaneously, by combining the spatial distribution range of event clusters, it constrains the direction and concentration of damage expansion, thereby directly improving the ability to identify precursors of local instability and achieving more targeted and predictive instability prediction. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0017] Figure 1 This is a flowchart of the rock mass damage prediction method based on microseismic data of the present invention.

[0018] Figure 2 This is a flowchart of the rock mass damage prediction system module based on microseismic data of the present invention.

[0019] Figure 3 The flowchart of the method for constructing inter-event direction consistency discrimination rules according to the present invention is shown. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, 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.

[0021] Example 1, please refer to Figure 1 As shown in this embodiment, the rock mass damage prediction method based on microseismic data includes: The system acquires raw microseismic signals from multiple points within the monitoring area, performs filtering and event identification to obtain multi-point microseismic event signals, and extracts the initial motion direction, frequency band offset trajectory, and energy release duration of the events.

[0022] By extracting key information representing the early damage evolution characteristics of rock masses from complex, multi-source, and noisy microseismic raw signals, filtering and event identification are used to separate effective signals from background interference. Furthermore, the initial motion direction, frequency band offset trajectory, and energy release duration of the event are obtained. This allows for the characterization of the intrinsic variation law of microseismic event signals from three different dimensions: propagation direction characteristics, frequency evolution behavior, and energy release process. This provides a continuous and discriminative basic feature support for subsequent construction of event correlation and identification of damage evolution trends.

[0023] First, multiple microseismic monitoring points are deployed at predetermined locations within the monitoring area. Each monitoring point continuously acquires rock mass vibration waveforms at a uniform sampling frequency, obtaining multi-point raw microseismic signals. Since these raw microseismic signals typically contain mechanical construction vibrations, environmental background disturbances, and random noise, the raw waveforms acquired by each monitoring point are first filtered. A bandpass filter matching the target fracture signal frequency band is preferred to remove irrelevant low-frequency interference components and high-frequency noise components, resulting in a pre-processed signal with improved signal-to-noise ratio. Subsequently, event identification is performed on the pre-processed signal. Based on sudden increases in waveform amplitude, short-term abnormal energy rises, and continuous signal changes, the start and end times of the microseismic event signals are determined, and multi-point microseismic event signals are extracted from the continuous waveforms. Furthermore, by combining the event results identified by multiple monitoring points at similar times, event signals belonging to the same rock mass response process are matched to establish a multi-point observation data set for the same microseismic event signal. This provides basic data for subsequent extraction of the event's initial motion direction, frequency band shift trajectory, and energy release duration.

[0024] The extraction of the initial direction of the event includes the following steps: For multi-point microseismic event signals after event identification, the initial fluctuation segment of the event in the waveform corresponding to each monitoring point is first extracted, and the arrival time of the initial fluctuation segment is then processed. The arrival time picking can be based on the first abrupt change in the waveform that significantly deviates from the background noise, thereby determining the initial arrival time of each monitoring point, and forming an initial arrival time sequence according to the monitoring point number or spatial location. Since the path length of the same microseismic event signal propagating to different monitoring points varies, the initial arrival times between different monitoring points usually have a temporal difference, which can reflect the propagation direction characteristics of the event wavefront.

[0025] After obtaining the initial arrival time sequence, the arrival time differences between adjacent monitoring points and between monitoring points in different directions are calculated by combining the actual spatial deployment positions of each monitoring point within the monitoring area. This yields the propagation time difference distribution of the event wavefront in different directions. Based on this, the propagation speed of the wavefront from near to far is analyzed according to the propagation time difference distribution, constructing a propagation time gradient feature. This propagation time gradient feature is used to characterize the spatial propagation bias of the wavefront. When the arrival time increases more continuously and the time difference changes more stably in a certain direction, it indicates that the event energy has a more obvious preferential propagation trend along that direction.

[0026] Based on the propagation time gradient characteristics, the propagation velocities in multiple candidate directions are compared, and the directions with relatively fast propagation velocities and continuous time gradient changes are selected as the energy-priority propagation directions of the microseismic event signal, generating the corresponding event initiation direction. The event initiation direction can be represented as azimuth information relative to a preset reference direction of the monitoring area, used to characterize the main expansion direction of the initial rupture disturbance of the event.

[0027] The extraction of frequency band offset trajectories includes: for multi-point microseismic event signals after event identification, firstly, they are continuously segmented according to a preset time window, dividing the multi-point microseismic event signals into multiple short-time analysis intervals that are connected end-to-end. Subsequently, time-frequency analysis is performed on each short-time analysis interval to obtain the energy distribution of different frequency bands within the corresponding interval, and these are arranged in chronological order to form a frequency band energy distribution sequence that changes over time. Since the frequency components of microseismic event signals usually undergo dynamic shifts during crack initiation, propagation, and local penetration, the frequency band energy distribution sequence can reflect the frequency evolution process within a single microseismic event signal.

[0028] After obtaining the frequency band energy distribution sequence, the frequency band with the highest energy proportion in each short-time analysis interval is further identified, and the corresponding characteristic frequencies of the frequency band are extracted to form a dominant frequency evolution sequence in chronological order. The dominant frequency evolution sequence is used to characterize the migration of the dominant frequency components of the microseismic event signal over time. When the fracture state inside the rock mass changes, the dominant frequency often exhibits a change characteristic of fluctuating from high to low, from low to high, or locally, thus providing a basis for identifying the frequency band shift trend.

[0029] Based on the dominant frequency evolution sequence, the frequency change amplitude and direction between adjacent time points are calculated to distinguish whether the dominant frequency shifts to higher or lower frequency bands, and frequency band shift characteristics are constructed accordingly. These frequency band shift characteristics reflect not only the magnitude of frequency migration between time points but also the continuity and consistency of the migration process. When the frequency change direction remains consistent across multiple adjacent short-time analysis intervals, and the change amplitude shows continuous accumulation, it indicates that a clear frequency band shift trend exists within the microseismic event signal.

[0030] After obtaining the frequency band shift characteristics, a correlation analysis is further performed on the direction and amplitude of frequency changes within continuous time periods. Multiple short-time analysis intervals with consistent change directions and temporal continuity are connected to form a frequency band shift trajectory reflecting the dominant frequency migration path. This frequency band shift trajectory can be used to characterize the frequency transfer mode of microseismic event signals during continuous evolution, reflecting the internal response changes of local rock mass damage from initiation to propagation, and providing a frequency characteristic basis for subsequent identification of energy migration continuity and determination of critical damage evolution states.

[0031] Extracting the duration of energy release involves the following steps: For multi-point microseismic event signals after event identification, envelope extraction processing is first performed on their waveforms to obtain an energy envelope curve that reflects the continuous change of signal intensity over time. This envelope extraction processing can be achieved by smoothly tracking the amplitude changes of the event signal, transforming the discrete fluctuations in the original waveform into a continuously changing curve. Since microseismic event signals typically exhibit a gradual increase in energy during occurrence, reaching a peak and then gradually decaying, the energy envelope curve can relatively intuitively characterize the complete energy release process of a single microseismic event signal.

[0032] After obtaining the energy envelope curve, an energy threshold is set based on the background noise level and the event signal strength. Using this energy threshold as a benchmark, the energy envelope curve is then used to determine its range. When the energy envelope curve first rises continuously and exceeds the energy threshold, the corresponding moment is determined as the energy starting point. When the energy envelope curve falls continuously after reaching its peak and drops below the energy threshold, the corresponding moment is determined as the attenuation termination point.

[0033] By determining the starting point of energy rise and the ending point of attenuation, the effective energy range boundary corresponding to a single microseismic event signal can be formed, thereby defining the time range in which the event actually participates in energy release.

[0034] Based on the boundary of the effective energy range, the time span between the initial energy rise point and the attenuation termination point is calculated to obtain the corresponding initial duration parameter. Considering that different microseismic event signals may have different attenuation rates in the tail attenuation phase, to improve the stability of the duration representation, the time span is further corrected by combining the rate of change of the energy envelope curve in the attenuation phase. When the energy at the tail of the event decays slowly, the corresponding duration is appropriately extended; when the energy at the tail of the event falls rapidly, the corresponding duration is maintained or appropriately compressed, thereby obtaining an initial duration parameter that better reflects the actual energy release process.

[0035] After obtaining the initial duration parameters, consistency screening is performed on microseismic event signals that are temporally adjacent or spatially close. The initial duration parameters of adjacent events are compared. If the duration of an event is significantly shorter or longer than that of its neighbors, and deviates from the overall variation pattern by more than a preset range, then the event is determined to be an abnormally short-duration event or an abnormally long-duration event. These abnormal events can be removed from subsequent analyses, thus avoiding duration distortion caused by noise interference, incomplete event truncation, or local abnormal fluctuations. Ultimately, a stable and reliable energy release duration is output, providing a temporal characteristic basis for subsequent construction of the transfer sequence and determination of energy migration continuity.

[0036] like Figure 3 As shown, based on the initial direction of the event and the frequency band offset trajectory, a directional consistency discrimination rule is constructed between events, and a set of event clusters with directional convergence characteristics is formed by filtering.

[0037] After obtaining the initial motion direction of each microseismic event signal, the adjacent microseismic event signals are first arranged in the order of their occurrence time, and the change in the directional angle between the previous and subsequent microseismic event signals is calculated one by one to characterize the degree of deflection of the adjacent microseismic event signals in the propagation direction.

[0038] The changes in the directional angles of multiple consecutive adjacent microseismic event signals are cascaded in chronological order to form a continuous directional change sequence. This sequence reflects the evolution of the propagation direction of microseismic activity. When the changes in the directional angles between adjacent microseismic event signals are consistently small and gradual, it indicates that the microseismic activity has strong continuous propagation characteristics in space. When the changes in the directional angles suddenly increase, it indicates that the propagation path may have turned or been disturbed.

[0039] After obtaining the continuous direction change sequence, the frequency band offset trajectory corresponding to each microseismic event signal is further retrieved, and the change direction of the frequency band offset trajectory is analyzed in correspondence with the degree of direction deflection in the continuous direction change sequence.

[0040] To determine whether the frequency band offset trajectory of a continuous multi-point microseismic event signal maintains a consistent migration direction, and whether this migration process matches the increase or decrease in the degree of directional deflection, when the direction of change of the frequency band offset trajectory and the degree of directional deflection show synchronous enhancement, synchronous weakening, or synchronous stability in continuous events, the corresponding time segment is extracted as the associated segment, and the directional-frequency band coordinated change feature is formed accordingly.

[0041] The direction-frequency band coordinated variation feature is used to characterize the coupling relationship between the propagation direction change and the internal frequency evolution of the microseismic event signal, thereby improving the accuracy of identifying continuous propagation behavior.

[0042] Based on the aforementioned direction-frequency band coordinated change characteristics, the correlation status between different microseismic event signals is classified and determined. When the initial motion direction of adjacent microseismic event signals changes little and the frequency band offset trajectory maintains continuous migration in the same direction, it is determined to be continuous in the same direction; when the initial motion direction of adjacent microseismic event signals changes significantly in the opposite direction and the frequency band offset trajectory shows a corresponding reverse shift, it is determined to be reverse offset; when the degree of deviation of the initial motion direction of adjacent microseismic event signals exceeds a preset range and the frequency band offset trajectory shows a significant jump in a short period of time, it is determined to be a sudden change.

[0043] Please see Figure 3 As shown, based on three discrimination conditions—continuation in the same direction, offset in the opposite direction, and abrupt change—a directional consistency discrimination rule is constructed between events to serve as the basis for screening continuously propagating events.

[0044] After establishing the directional consistency discrimination rule between the events, the microseismic event signals that are adjacent in time are further compared one by one. Events that meet the same direction continuity discrimination condition or the preset continuous propagation requirement are identified as related events, and events that meet the reverse offset or abrupt change discrimination condition are identified as discontinuous propagation events.

[0045] Using the above method, event relationships with stable propagation directions and consistent frequency evolution trends can be screened from continuous microseismic event signals. The final output of direction consistency discrimination results provides a discrimination basis for subsequent screening to form a set of event clusters with direction convergence characteristics.

[0046] After obtaining the direction consistency discrimination rule, the adjacent microseismic event signals with time intervals within a preset range are first compared according to the occurrence time sequence of the microseismic event signals. Then, based on the direction consistency discrimination rule, it is determined whether the adjacent microseismic event signals meet the continuous propagation requirement. For adjacent microseismic event signals that meet the continuous propagation requirement, they are initially merged to form candidate event groups.

[0047] The microseismic event signals in the candidate event group are continuous in their occurrence sequence and meet the preset consistency condition in their propagation direction changes, thus providing a basis for subsequent identification of directional convergence relationships.

[0048] After forming the candidate event groups, the initial motion direction and frequency band offset trajectory of all microseismic event signals in each candidate event group are further extracted, and statistical analysis is performed on both within the group.

[0049] By analyzing the magnitude of the initial motion direction changes of each microseismic event signal within the group, the degree of concentration of directional changes is determined; simultaneously, by analyzing the direction of frequency band offset trajectory changes and the continuity of each microseismic event signal within the group, the degree of unidirectional continuity of frequency band offset is determined.

[0050] The degree of concentration of the directional change and the degree of continuation of the frequency band shift in the same direction are comprehensively characterized to generate corresponding directional convergence evaluation parameters. These directional convergence evaluation parameters reflect the degree of consistency between the propagation direction and frequency evolution of microseismic event signals within a candidate event group.

[0051] Based on the aforementioned directional convergence evaluation parameters, it is necessary to screen each candidate event group.

[0052] When the directional change of a candidate event group is highly concentrated and the frequency band shift trajectory maintains cooperative evolution among consecutive events, the candidate event group is determined to have stable directional convergence characteristics and is retained to form directional convergence event units. When the directional dispersion of a candidate event group is large, or the frequency band shift trajectory has obvious interruptions or reverse jumps, it is eliminated. Through the above screening process, directional convergence event units with continuity in both propagation direction and frequency shift can be obtained.

[0053] After obtaining the directional convergence event units, the occurrence time and spatial distribution of the corresponding microseismic event signals of each directional convergence event unit are further combined to merge adjacent directional convergence event units. When two directional convergence event units are adjacent in time or the interval is less than a preset value, and meet the proximity condition in spatial location, they are merged into the same event cluster; directional convergence event units that do not meet the conditions are retained separately.

[0054] The final output event cluster set is used to represent the microseismic activity clusters with common propagation trends and co-evolutionary characteristics within the monitoring area, providing a basis for subsequent construction of energy migration continuity transfer sequences.

[0055] The events in the event cluster are reorganized in the order of occurrence, and combined with the duration of energy release, a transfer sequence representing the continuity of energy migration is generated.

[0056] After obtaining the event cluster set, the microseismic event signals contained within each event cluster are first rearranged according to their occurrence time sequence, and a continuity analysis is performed in conjunction with the time intervals between adjacent microseismic event signals. When the time interval between adjacent microseismic event signals is within a preset continuity range, it is determined that there is a temporal succession relationship between the subsequent microseismic event signal and the preceding microseismic event signal. Based on this, multiple microseismic event signals within the same event cluster that satisfy the temporal succession relationship are sequentially connected to form a sequential event chain.

[0057] The sequential event chain is used to characterize the continuous evolution of microseismic activity within an event cluster over time, providing a basis for subsequent identification of energy migration paths.

[0058] After forming the sequential event chain, the energy release duration corresponding to each microseismic event signal in the chain is further extracted, and the duration variation relationship between adjacent microseismic event signals is compared and analyzed.

[0059] The duration of energy release in a subsequent microseismic event signal is determined relative to the preceding microseismic event signal; it remains close, gradually lengthens, or significantly shortens. This process identifies the duration continuity between adjacent microseismic event signals. The duration continuity of multiple consecutive adjacent microseismic event signals is combined according to their occurrence sequence to generate duration transmission characteristics. These characteristics reflect the temporal stability of the energy release process within an event cluster. When the duration gradually extends or smoothly succeeds, it indicates strong energy migration continuity between microseismic activities.

[0060] Based on the duration transmission characteristics, adjacent microseismic event signals in a sequential event chain are identified and filtered. When the duration of energy release of multiple consecutive adjacent microseismic event signals gradually extends, or when the subsequent microseismic event signal forms a continuous succession within a short period of time after the previous microseismic event signal ends, it is determined that there is a continuous energy succession relationship between these microseismic event signals, and the corresponding segment is constructed as an energy succession segment. When the duration change between adjacent microseismic event signals shows a sudden drop, a sudden rise, or an interruption, and does not meet the preset succession conditions, the corresponding segment is removed from the energy succession segment.

[0061] Through the above processing, effective segments that better reflect the continuous energy migration process can be identified from the sequential event chain.

[0062] After obtaining the energy succession segments, they are connected in series according to the order in which they appear in the event cluster, while maintaining the connection between the segments, to generate a transmission sequence.

[0063] Among them, the transmission sequence is used to characterize the continuous process of energy transmission from the signal of the previous microseismic event to the signal of the next microseismic event within the event cluster. It can reflect the successive evolution characteristics of microseismic activity during the local damage propagation process of the rock mass, and provide a basis for subsequent identification of abnormal convergence sections and extraction of energy accumulation abrupt change indicators.

[0064] Based on the energy connectivity and time compression characteristics of adjacent events in the transmission sequence, abnormal convergence segments are identified, and energy aggregation mutation indices and time contraction coefficients are extracted.

[0065] After obtaining the transmission sequence, the energy release duration corresponding to each microseismic event signal is extracted one by one according to the order of the microseismic event signals in the transmission sequence. The energy connection degree between adjacent microseismic event signals is calculated by combining the connection relationship between the end time of the previous microseismic event signal and the start time of the next microseismic event signal.

[0066] Extract the time interval between adjacent microseismic events and record the changes in the time interval between multiple consecutive adjacent events sequentially.

[0067] The energy continuity of adjacent microseismic event signals is matched one-to-one with the changes in the time interval between occurrences to form an energy-time continuity feature sequence. This energy-time continuity feature sequence is used to synchronously characterize the continuous evolution of microseismic event signals in the transmission sequence during the energy release succession process and the triggering rhythm change process, thereby providing a basis for identifying local energy rapid concentration sections.

[0068] After forming the energy-time continuity feature sequence, segment identification is performed on multiple consecutive adjacent microseismic event signals along the extension direction of the transmission sequence. It is determined whether the energy continuity between adjacent microseismic event signals shows a continuously increasing trend, and simultaneously whether the corresponding occurrence time interval shows a continuously shortening trend. When multiple consecutive adjacent microseismic event signals simultaneously satisfy the condition of gradually increasing energy continuity and gradually compressing occurrence time intervals, it is determined that there are obvious local clustering evolution characteristics between these adjacent microseismic event signals, and the corresponding segment is identified as an abnormal convergence segment. When the energy continuity between any adjacent microseismic event signal is interrupted and decreases, or the occurrence time interval no longer shortens, the continuation determination of the current segment is terminated.

[0069] Identifying anomalous convergence segments in the transmission sequence, where energy shifts from dispersed continuity to localized concentration and enhancement, provides a basis for subsequent extraction of energy aggregation mutation indices and time contraction coefficients.

[0070] After identifying the anomalous convergence segment, the microseismic event signals within this segment are extracted sequentially, and the energy release duration corresponding to each microseismic event signal is retrieved. The energy release durations are arranged according to the chronological order of the microseismic event signals, forming a segment energy continuity sequence. This segment energy continuity sequence is used to characterize the sequential relationship between the microseismic event signals within the anomalous convergence segment in the energy release duration process, allowing the energy evolution state of microseismic activity within the segment to be presented in a continuous manner.

[0071] Since abnormal convergence zones typically correspond to the stage where local rock mass damage transitions from slow evolution to concentrated enhancement, further analysis of the energy continuity sequence of these zones can identify whether there are abrupt changes in energy accumulation.

[0072] After forming the energy succession sequence of the section, the duration of energy release of adjacent microseismic event signals is compared one by one, the duration increment between adjacent microseismic event signals is calculated, and the trend of the duration increment is analyzed along the occurrence sequence of the microseismic event signals.

[0073] When the duration increment between multiple consecutive adjacent microseismic event signals remains within a small range, the energy release process in that section can be considered to be in a stable continuous state. When the duration increment between subsequent adjacent microseismic event signals increases significantly and maintains an expanding trend in consecutive events, it indicates that the energy release process in the abnormal convergence section has changed from a stable continuous state to a rapid growth state.

[0074] Based on the changing trend, the location where the duration increment begins to significantly increase is identified, and this location is determined as the mutation initiation location. This mutation initiation location is used to characterize the starting node where a significant change in the energy accumulation state occurs, thus providing a boundary basis for subsequent quantification of the degree of energy concentration enhancement.

[0075] After determining the abrupt change initiation location, the adjacent microseismic event signals before and after the initiation location are divided into a pre-abrupt change stage and a post-abrupt change stage, respectively. For both the pre-abrupt change stage and the post-abrupt change stage, the cumulative duration of energy release for the corresponding microseismic event signals is statistically calculated, and the cumulative difference and growth factor of the post-abrupt change stage relative to the pre-abrupt change stage are calculated. The cumulative difference characterizes the overall enhancement magnitude of the energy release duration within the anomalous convergence zone, and the growth factor characterizes the transition intensity when energy release shifts from normal continuity to concentrated enhancement.

[0076] An energy accumulation abrupt change index is generated based on the cumulative difference and the growth factor to reflect the degree of change in the rapid accumulation of microseismic energy in local areas within the abnormal convergence zone. Therefore, the energy accumulation abrupt change index can not only characterize whether energy enhancement has occurred, but also characterize the strength differences in the enhancement process.

[0077] Meanwhile, within the anomalous convergence zone, the time intervals between adjacent microseismic event signals are further extracted and arranged sequentially according to their occurrence order. By comparing multiple consecutive adjacent time intervals one by one, the shortening amplitude between each adjacent time interval is calculated, and the continuous trend of the shortening amplitude is analyzed. When multiple consecutive adjacent time intervals show only a slight decrease, the triggering rhythm of the microseismic event signals can be considered to be in a relatively slow state; when the shortening amplitude of subsequent adjacent time intervals continuously increases, and multiple adjacent microseismic event signals appear densely in a short period of time, it indicates that the event triggering rhythm within the anomalous convergence zone has changed from slow convergence to rapid compression. Based on this, locations where the shortening trend of time intervals significantly intensifies are identified and determined as the contraction initiation locations.

[0078] After determining the contraction initiation position, the time intervals of adjacent microseismic event signals in the pre-contraction and post-contraction stages are further segmented and statistically analyzed, using the contraction initiation position as a boundary. The cumulative reduction and compression ratio of the time intervals before and after contraction are calculated respectively. The cumulative reduction characterizes the degree of tightening of the overall triggering interval of microseismic event signals within the anomalous convergence zone, while the compression ratio characterizes the magnitude of the change in the event triggering rhythm from scattered occurrence to short-duration concentrated occurrence. Furthermore, a time contraction coefficient is generated based on the cumulative reduction and the compression ratio to characterize the degree of acceleration in the triggering rhythm of microseismic event signals within the anomalous convergence zone.

[0079] By correlating the energy accumulation mutation index with the time contraction coefficient, the evolution of microseismic activity within the anomalous convergence zone can be jointly characterized. When the energy accumulation mutation index continuously increases and the time contraction coefficient increases synchronously, it indicates that the microseismic activity not only shows significant accumulation in energy release but also exhibits significant compression in the event triggering rhythm, suggesting that local rock mass damage has entered an accelerated evolution stage. When either parameter does not change significantly, it indicates that the concentrated evolution characteristics within the anomalous convergence zone are not yet sufficient. The joint analysis based on the energy accumulation mutation index and the time contraction coefficient can provide a reliable basis for subsequently determining the critical state of damage evolution and outputting local instability prediction results.

[0080] Taking the surrounding rock of a deep underground tunnel as an example, 12 microseismic monitoring points were set up in the monitoring area to continuously collect raw microseismic signals at a sampling frequency of 10kHz. During 48 hours of continuous monitoring, a total of 286 valid microseismic event signals were identified. Using the method of this invention, 7 event clusters were formed, one of which, located in the right shoulder corner of the tunnel, contained 21 consecutive microseismic event signals. A transmission sequence was constructed for this event cluster, and an abnormal convergence segment was identified, resulting in an abnormal convergence segment consisting of 8 consecutive microseismic event signals. The key parameters of this segment are shown in the table below. Table 1 Characteristic parameters of microseismic event signals within the anomalous convergence zone As shown in Table 1, the duration of energy release shows a continuous increasing trend, while the time interval between adjacent events shows a continuous shortening trend, indicating that microseismic activity is evolving from scattered triggering to short-duration concentrated triggering. Based on the above data, the cumulative difference in energy release duration is calculated to be 1.21 seconds, with an increase factor of 2.82; the cumulative reduction in time interval is 33 seconds, with a compression ratio of 78.6%. Based on this, an energy concentration mutation index and a time contraction coefficient are generated.

[0081] Furthermore, the above parameters were coupled and analyzed to form a critical criterion for damage evolution, and the results were compared with those of traditional methods. The results are as follows: Table 2 Comparison results between the method of the present invention and the traditional method According to Table 2 and the results of on-site observation, approximately 4.5 hours after the method of the present invention triggers the critical criterion for damage evolution, the region exhibits obvious spalling, with the crack width expanding from approximately 2 mm to approximately 6 mm, and the spatial location is highly consistent with the distribution range of the event cluster.

[0082] By coupling the energy concentration mutation index with the time contraction coefficient, the critical state of microseismic activity transforming from dispersed to concentrated is determined, thus forming a critical criterion for damage evolution.

[0083] After obtaining the energy accumulation mutation index and time contraction coefficient corresponding to the same anomalous convergence segment, the two are first matched according to the occurrence order of each microseismic event signal in the anomalous convergence segment, and a correlation relationship that changes synchronously with the evolution of the event is established.

[0084] The energy accumulation mutation indices and time contraction coefficients corresponding to each stage are arranged in a one-to-one correspondence to form a corresponding change relationship reflecting the coordinated change process of the two. This corresponding change relationship is used to characterize the synchronous evolution state between the energy concentration enhancement process and the event triggering rhythm compression process within the anomalous convergence segment, thereby providing a basis for identifying the key stage of the transformation of microseismic activity from dispersion to concentration.

[0085] After establishing the corresponding change relationship, the changing trends of the energy accumulation mutation index and the time contraction coefficient are further analyzed along the evolution direction of the abnormal convergence segment.

[0086] Determine whether the energy accumulation mutation index continues to rise within multiple consecutive adjacent stages, and simultaneously determine whether the corresponding time contraction coefficient continues to increase synchronously; when both show a synergistic enhancement trend within the same continuous segment, identify that segment as an overlapping segment and determine it as a concentrated conversion sensitive segment.

[0087] After identifying the sensitive section for concentrated transformation, a coordination analysis was further conducted on the energy accumulation abrupt change index and the time contraction coefficient within this sensitive section. Specifically, the enhancement magnitude, the order of enhancement, and the degree of synchronization between the two during continuous evolution were compared to determine the matching level between the energy accumulation enhancement process and the time compression process. When the increasing trend of the energy accumulation abrupt change index and the increasing trend of the time contraction coefficient are consistent within the continuous section, and there is no significant disconnect between their enhancement processes, it indicates that the evolutionary state of microseismic activity from dispersed to concentrated has a high degree of coordination.

[0088] Based on this, the degree of synergistic enhancement between the two within the concentrated transformation sensitive section is quantified, generating a critical coupling characteristic value. This critical coupling characteristic value is used to characterize the intensity level of microseismic activity transforming from a dispersed state to a concentrated state within the anomalous convergence section.

[0089] After obtaining the critical coupling characteristic value, it is further compared with the preset judgment condition to determine the current evolution state of the abnormal convergence segment.

[0090] When the critical coupling characteristic value is lower than the preset judgment condition, it indicates that although there is some energy accumulation and time contraction in the abnormal convergence section, it has not yet reached a significant level of concentrated transformation. When the critical coupling characteristic value reaches or exceeds the preset judgment condition, it is determined that the abnormal convergence section has entered the critical state of microseismic activity transforming from dispersed to concentrated, and a critical criterion for damage evolution is formed accordingly. The critical criterion for damage evolution can serve as an important basis for subsequently outputting the local damage development trend and instability prediction results of the rock mass.

[0091] The critical criteria for damage evolution are not based on a single energy change characteristic or a single time change characteristic, but rather on the joint analysis of energy accumulation abrupt change indicators and time contraction coefficients to identify critical states. Therefore, misjudgments can be avoided by relying solely on local energy anomalies or local time compression. By comprehensively utilizing the synergistic relationship between these two factors, the general microseismic active phase and the accelerated damage evolution phase can be more accurately distinguished, thereby improving the ability to identify precursors of local rock mass instability.

[0092] When the critical criterion for damage evolution is triggered, the local damage development trend and instability prediction results of the rock mass are output, based on the spatial distribution range of the corresponding event cluster.

[0093] When a critical criterion for damage evolution is triggered, the event cluster corresponding to that criterion is first identified, and the spatial distribution locations of each microseismic event signal within the event cluster are extracted. This allows for the determination of the spatial distribution range of the event cluster within the monitoring area. The current expansion state of local damage activity can be determined based on the spatial aggregation boundary, extension direction, and distribution density of the multi-point microseismic event signals within the event cluster. When the spatial distribution range of the event cluster continues to extend along its existing direction, it indicates that the local rock mass damage is developing in a directional trend; when the spatial distribution range of the event cluster expands from a local concentration to the periphery, it indicates that the local rock mass damage has the risk of further expansion. Based on the spatial distribution range and its changing characteristics of the event cluster, the development trend of local rock mass damage in the corresponding area can be output to characterize the direction, range, and concentration of damage expansion.

[0094] After obtaining the local damage development trend of the rock mass, and combining the concentrated transformation state of microseismic activity reflected by the critical damage evolution criterion, the corresponding area is assessed for instability risk. When the critical damage evolution criterion is triggered, and the spatial distribution range of the event cluster shows obvious concentration, continuous extension, or rapid expansion, the area is determined to have a high probability of local instability, and the corresponding instability prediction result is output. When the critical damage evolution criterion is triggered, but the spatial distribution range of the event cluster changes weakly or the expansion is not obvious, the area is determined to be in the accelerated damage evolution stage, and the corresponding early warning result is output. By combining the critical damage evolution criterion with the spatial distribution range of the event cluster, local damage to the rock mass can be determined simultaneously from two dimensions: evolution intensity and spatial expansion state, thereby improving the pertinence and reliability of the instability prediction results.

[0095] Example 2, please refer to Figure 2 As shown in this embodiment, the rock mass damage prediction system based on microseismic data includes: Data acquisition module: acquires raw microseismic signals from multiple points within the monitoring area, performs adaptive filtering and event identification, and extracts the initial motion direction, frequency band offset trajectory, and energy release duration of the event; Event cluster filtering module: Based on the initial direction of the event and the frequency band offset trajectory, construct the direction consistency discrimination rule between events, and filter to form a set of event clusters with direction convergence characteristics; Transmission sequence construction module: Reorganizes each event in the event cluster set according to the order of occurrence, and combines the duration of energy release to generate a transmission sequence that characterizes the continuity of energy migration; Anomaly Convergence Identification Module: Based on the energy connectivity and time compression characteristics of adjacent events in the transmission sequence, it identifies abnormal convergence segments and extracts energy aggregation mutation indicators and time contraction coefficients. Critical state determination module: The energy accumulation mutation index is coupled with the time contraction coefficient to determine the critical state of microseismic activity from dispersion to concentration, and a critical criterion for damage evolution is formed. Instability prediction module: When the critical criterion for damage evolution is triggered, the module outputs the local damage development trend and instability prediction results of the rock mass, based on the spatial distribution range of the corresponding event cluster.

[0096] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A rock mass damage prediction method based on microseismic data, characterized in that: include: The system acquires raw microseismic signals from multiple points within the monitoring area, performs adaptive filtering and event identification to obtain multi-point microseismic event signals, and extracts the initial motion direction, frequency band offset trajectory, and energy release duration of the events. Based on the initial direction of the event and the frequency band offset trajectory, a rule for determining the direction consistency between events is constructed, and a set of event clusters with direction convergence characteristics is formed by filtering. The events in the event cluster are reorganized in the order of occurrence, and combined with the duration of energy release, a transfer sequence characterizing the continuity of energy migration is generated. Based on the energy connectivity and time compression characteristics of adjacent events in the transmission sequence, abnormal convergence segments are identified, and energy aggregation mutation indices and time contraction coefficients are extracted. By coupling the energy accumulation mutation index with the time contraction coefficient, the critical state of microseismic activity transforming from dispersed to concentrated is determined, thus forming a critical criterion for damage evolution. When the critical criterion for damage evolution is triggered, the local damage development trend and instability prediction results of the rock mass are output, based on the spatial distribution range of the corresponding event cluster.

2. The rock mass damage prediction method based on microseismic data according to claim 1, characterized in that: The extraction of the initial motion direction of the event includes: performing multi-channel arrival time picking on the identified multi-point microseismic event signals to determine the initial motion arrival time sequence of each monitoring point; based on the initial motion arrival time sequence and combined with the spatial layout relationship of the monitoring points, calculating the time difference distribution of the event wavefront propagation and constructing the propagation time gradient feature; and comparing the propagation speed in different directions according to the propagation time gradient feature to determine the energy-priority propagation direction and generate the initial motion direction of the event.

3. The rock mass damage prediction method based on microseismic data according to claim 1, characterized in that: Extracting the frequency band offset trajectory includes: performing segmented time-frequency analysis on the identified multi-point microseismic event signals to obtain a frequency band energy distribution sequence that changes over time; based on the frequency band energy distribution sequence, extracting the frequency band with the highest energy proportion and its change over time to form a dominant frequency evolution sequence; calculating the frequency change amplitude and direction between adjacent time points based on the dominant frequency evolution sequence to construct frequency band offset change features; and using the frequency band offset change features to correlate continuous time periods to generate a frequency band offset trajectory.

4. The rock mass damage prediction method based on microseismic data according to claim 1, characterized in that: Extracting the duration of energy release includes: performing envelope extraction processing on the identified multi-point microseismic event signals to obtain an energy envelope curve reflecting changes in signal intensity; based on the energy envelope curve, setting an energy threshold, determining the energy start-up point and attenuation termination point, and forming the boundary of the effective energy interval; calculating the corresponding time span according to the boundary of the effective energy interval and correcting it in conjunction with the energy attenuation rate to obtain an initial duration parameter; using the initial duration parameter to perform consistency screening on adjacent events, eliminating abnormally short or long-duration events, and outputting the duration of energy release.

5. The rock mass damage prediction method based on microseismic data according to claim 1, characterized in that: The established rules for determining the directional consistency between construction events include: Based on the initial motion direction of adjacent microseismic event signals, the degree of directional deflection between successive events is calculated to form a continuous directional change sequence. Based on the continuous directional change sequence, combined with the frequency band offset trajectory of the corresponding microseismic event signal, the associated segment where the frequency band offset direction and the degree of directional deflection change synchronously are extracted to form a direction-frequency band coordinated change feature. Based on the aforementioned direction-frequency band coordinated change characteristics, three types of discrimination conditions are set: same-direction continuity, opposite-direction shift, and abrupt change, and a direction consistency discrimination rule between events is constructed. Using the direction consistency discrimination rule between events, the temporally adjacent microseismic event signals are compared one by one to filter out the associated events that meet the continuous propagation characteristics, and the direction consistency discrimination rule is output.

6. The rock mass damage prediction method based on microseismic data according to claim 1, characterized in that: The filtering process forms a set of event clusters with directional convergence characteristics, including: Based on the direction consistency discrimination rule, the signals of multiple temporally adjacent microseismic events are initially merged to form candidate event groups; For each candidate event group, the initial motion direction and frequency band offset trajectory of each microseismic event signal within the group are combined to calculate the degree of concentration of direction change and the degree of continuation of frequency band offset in the same direction, and generate direction convergence evaluation parameters. Based on the directional convergence evaluation parameters, the candidate event groups are screened, and event groups with continuous directional changes and coordinated evolution of frequency band offset are retained to form directional convergence event units. The event units that converge in direction are merged according to temporal continuity and spatial proximity to output an event cluster set.

7. The rock mass damage prediction method based on microseismic data according to claim 1, characterized in that: The generation of the transfer sequence characterizing the continuity of energy migration includes: Based on the event cluster set, the microseismic event signals are arranged in the order of their occurrence. Combined with the time interval relationship between adjacent microseismic event signals, a sequential transmission event chain is formed. For the sequential transmission event chain, the energy release duration corresponding to each microseismic event signal is extracted, and the duration connection changes between the microseismic event signals are compared to generate duration transmission characteristics. Based on the characteristics of duration transmission, we identify adjacent multi-point microseismic event signals with gradually extending or continuous durations, and construct energy succession segments; we then connect the energy succession segments in series according to their occurrence order to output the transmission sequence. Based on the transmission sequence, the energy continuity and occurrence time interval changes between adjacent microseismic event signals are calculated to form an energy-time continuity characteristic sequence. According to the energy-time continuity characteristic sequence, adjacent microseismic event signal segments with continuously increasing energy continuity and continuously shortening occurrence time intervals are identified as abnormal convergence segments.

8. The rock mass damage prediction method based on microseismic data according to claim 1, characterized in that: Extracting energy accumulation abrupt change indicators includes: based on the abnormal convergence segment, extracting the energy release duration and event sequence corresponding to each microseismic event signal within the segment to form a segment energy succession sequence; comparing the duration increments and continuous change trends between adjacent microseismic event signals to identify the abrupt change starting position where the duration changes from a stable succession to rapid growth; based on the abrupt change starting position, calculating the cumulative difference and growth factor of the duration of adjacent microseismic event signals before and after the abrupt change to generate energy accumulation abrupt change indicators; Extracting the time contraction coefficient includes: based on the abnormal convergence segment, extracting the occurrence time interval between adjacent microseismic event signals within the segment, comparing the shortening magnitude and continuous change trend of adjacent time intervals, identifying the contraction starting position where the time interval changes from slow decrease to rapid compression, calculating the cumulative reduction and compression ratio of the time interval between adjacent microseismic event signals before and after contraction, and generating a time contraction coefficient to represent the degree of acceleration of the event triggering rhythm within the segment.

9. The rock mass damage prediction method based on microseismic data according to claim 1, characterized in that: The critical criteria for damage evolution are formed, including: based on the energy accumulation mutation index and the time contraction coefficient, extracting the corresponding change relationship between the two within the same abnormal convergence segment, identifying the overlapping segment where the energy accumulation mutation index continues to rise and the time contraction coefficient increases synchronously, and determining it as the concentrated transformation sensitive segment; The degree of enhanced coordination between the energy concentration mutation index and the time contraction coefficient is calculated to generate critical coupling characteristic values ​​that characterize the intensity of the transformation of microseismic activity from dispersion to concentration. Based on the critical coupling characteristic values ​​and preset judgment conditions, the state of abnormal convergence sections is judged. When the critical coupling characteristic values ​​reach the triggering requirements, the critical criterion for damage evolution is formed.

10. A rock mass damage prediction system based on microseismic data, used to implement the rock mass damage prediction method based on microseismic data as described in any one of claims 1-9, characterized in that: include: Data acquisition module: acquires raw microseismic signals from multiple points within the monitoring area, performs adaptive filtering and event identification to obtain multi-point microseismic event signals, and extracts the initial motion direction, frequency band offset trajectory and energy release duration of the events; Event cluster filtering module: Based on the initial direction of the event and the frequency band offset trajectory, construct the direction consistency discrimination rule between events, and filter to form a set of event clusters with direction convergence characteristics; Transmission sequence construction module: Reorganizes each event in the event cluster set according to the order of occurrence, and combines the duration of energy release to generate a transmission sequence that characterizes the continuity of energy migration; Anomaly Convergence Identification Module: Based on the energy connectivity and time compression characteristics of adjacent events in the transmission sequence, it identifies abnormal convergence segments and extracts energy aggregation mutation indicators and time contraction coefficients. Critical state determination module: The energy accumulation mutation index is coupled with the time contraction coefficient to determine the critical state of microseismic activity from dispersion to concentration, and a critical criterion for damage evolution is formed. Instability prediction module: When the critical criterion for damage evolution is triggered, the module outputs the local damage development trend and instability prediction results of the rock mass, based on the spatial distribution range of the corresponding event cluster.

Images