Tunnel rockburst position early warning method based on deeply-buried complex geological environment

Abstract

The invention belongs to the technical field of tunnel rockburst early warning, and particularly discloses a tunnel rockburst position early warning method based on a deeply-buried complex geological environment, and the method comprises the steps: preliminarily screening out an effective micro-seismic event from a collected micro-seismic signal in a delimited risk rockburst section through amplitude triggering and spatial positioning; secondly, two-dimensional features of energy and spatial position are introduced into each effective micro-seismic event to realize refined screening of the rock burst suspicion event, micro-seismic background interference is inhibited to the greatest extent, the false alarm rate is reduced, and meanwhile, when risk judgment is performed by using the screened rock burst suspicion event, the risk judgment accuracy is improved. Event statistical characteristics and a time sequence evolution rule are fused to construct a joint criterion, the missed judgment or misjudgment rate of a real rockburst precursor signal is greatly reduced, finally, when it is judged that a rockburst suspicion exists, a rockburst suspicion event magnetic concentration distribution interval is determined through clustering analysis, and rockburst early warning information containing a specific axial position range is output. And targeted prevention and control measures can be taken.

Description

Technical Field

[0001] This invention belongs to the field of tunnel rockburst early warning technology, and specifically discloses a method for early warning of tunnel rockburst location based on deep-buried complex geological environments. Background Technology

[0002] Rockburst is a common geological hazard during the excavation of deep-buried, high-stress hard rock tunnels. It is characterized by its suddenness and destructive power. Therefore, monitoring and early warning of rockburst is crucial to ensuring the safety of the project. Among them, microseismic monitoring, by capturing the elastic wave signals released by rock fractures, can invert the damage evolution process inside the surrounding rock and has become one of the main means of rockburst early warning.

[0003] Existing technologies include rockburst early warning methods based on microseismic monitoring. For example, Chinese Invention Patent Publication No. CN119507980A proposes a tunnel rockburst early warning method based on the spatial concentration of microseismic activity. By delineating a fixed area, the method comprehensively calculates the incubation time, cumulative energy, and spatial distance of microseismic events to obtain a comprehensive index of "concentration degree," which is then mapped to the rockburst probability through a piecewise function. The early warning result is the risk level and probability.

[0004] However, the technical solution has the following limitations in implementation: First, the core criterion of the above solution, "convergence", is the static composite value of microseismic events in terms of spatial and temporal parameters. Rockburst precursors are often accompanied by the process of micro-fracture events evolving from random to orderly and dense. Ignoring the temporal trend analysis of microseismic events can easily lead to the omission or misjudgment of real rockburst precursor signals.

[0005] Secondly, in deep-buried and complex geological environments, vibrations from construction machinery can generate background microseismic events. The comprehensive indicators used in the above scheme do not identify these microseismic events, which could easily include this background noise in the risk assessment, leading to an increased false alarm rate.

[0006] Third, the warning results output by the above scheme are a risk level and probability, which is a regional overall assessment, but cannot pinpoint the specific location of the risk in the tunnel, which is not conducive to taking targeted prevention and control measures. Summary of the Invention

[0007] To solve the above-mentioned technical problems, or at least partially solve them, the present invention provides a method for early warning of rockburst location in tunnels under complex geological environments.

[0008] The objective of this invention can be achieved through the following technical solution: a method for early warning of rockburst location in tunnels under complex geological conditions, including: deploying a microseismic monitoring network within a designated potential rockburst section.

[0009] Microseismic signals are collected using a microseismic monitoring network, and the collected microseismic signals are filtered by amplitude triggering and spatial positioning to identify valid microseismic events.

[0010] Energy and spatial location features are extracted from each valid microseismic event and compared with the limiting conditions obtained from the statistical analysis of historical rockburst precursor features to screen out suspected rockburst events.

[0011] Within a preset monitoring time window, the frequency percentage of suspected rockburst events is calculated, and the changing trend of the event intervals is analyzed.

[0012] A joint criterion is constructed based on the changing trends of the frequency ratio and the interval between events to determine whether there is a rockburst risk in a potential rockburst segment.

[0013] When a rockburst risk is determined, cluster analysis is used to identify the concentrated distribution range of suspected rockburst events, and rockburst early warning information containing the specific axial location range is output.

[0014] Combining all the above technical solutions, the positive effects of this invention are as follows: 1. This invention first screens out effective microseismic events by amplitude triggering and spatial positioning of the collected microseismic signals within the defined potential rockburst segment. Then, it introduces dual-dimensional features of energy and spatial location for each effective microseismic event to achieve refined identification of suspected rockburst events, thereby maximally suppressing microseismic background interference and helping to reduce the false alarm rate.

[0015] 2. When using identified rockburst suspected events for risk assessment, this invention comprehensively considers the proportion of occurrence within a preset analysis time window and the changing trend of the time interval between adjacent events, and integrates event statistical characteristics and temporal evolution laws to construct a joint criterion. This not only reduces the rate of missed or false judgment of real rockburst precursor signals, but also improves the timeliness and physical interpretability of rockburst early warning.

[0016] 3. When a rockburst is suspected, this invention further determines the concentrated distribution range of the suspected rockburst event through cluster analysis, and outputs rockburst early warning information containing the specific axial position range. This improves the early warning result from the traditional regional risk level to a location early warning with clear spatial orientation, which is conducive to taking targeted prevention and control measures and enhances the engineering practicality of the early warning information. Attached Figure Description

[0017] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0018] Figure 1 This is a diagram illustrating the implementation steps of the method of the present invention.

[0019] Figure 2 This is a flowchart illustrating the implementation of randomness testing for a sequence of time intervals between adjacent events in this invention.

[0020] Figure 3 This is a flowchart illustrating the implementation of trend analysis on the time interval sequence of adjacent events in this invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] See Figure 1 As shown, the present invention proposes a method for early warning of rockburst location in tunnels under complex geological conditions, including: S1, deploying a microseismic monitoring network within the designated potential rockburst section.

[0023] In tunnel engineering, due to the characteristics of large span and long length, implementing microseismic monitoring along the entire line would result in a huge amount of data, high monitoring costs, and low monitoring efficiency. By defining potential rockburst sections to limit the scope of microseismic monitoring, monitoring resources can be deployed on demand, thereby improving monitoring efficiency.

[0024] Specifically, potential rockburst sections refer to tunnel sections with geological characteristics prone to rockbursts. For example, they can be delineated based on geological assessment results. Before tunnel excavation, a geological survey is usually conducted to generate a geological assessment report. This report includes a rockburst risk analysis of the entire tunnel and clearly identifies high-risk sections as potential rockburst sections.

[0025] As a preferred embodiment, after delineating the potential rockburst section, the microseismic monitoring network is deployed as follows: Since rockbursts usually occur near the tunnel excavation face, especially in high-stress hard rock sections within a certain range ahead of the tunnel axis, exhibiting obvious spatial concentration, when deploying the microseismic monitoring network, detector arrays are uniformly deployed along the tunnel axis within the potential rockburst section to form a continuous microseismic monitoring network, ensuring full spatial coverage of precursory microseismic events of rockbursts. Each array contains multiple detectors.

[0026] The aforementioned detector was chosen because it can sense the elastic waves released when the surrounding rock fractures and convert the vibration signal into an electrical signal that can be recorded and analyzed. It is the sensing unit for microseismic monitoring to achieve precursor identification and signal acquisition.

[0027] Furthermore, an array structure composed of multiple detectors can calculate the spatial location of a microseismic event by taking advantage of the time difference (i.e., travel time difference) between the different detectors receiving the same microseismic signal and combining this with the known propagation velocity of elastic waves in the surrounding rock medium.

[0028] S2. Microseismic signals are collected using a microseismic monitoring network, and valid microseismic events are identified by filtering the collected microseismic signals through amplitude triggering and spatial positioning.

[0029] Considering that drilling, blasting, and other operations during tunnel construction continuously generate a large number of vibration signals, creating microseismic background interference, directly treating all signals collected by the microseismic monitoring network as valid microseismic events could easily lead to the introduction of a large amount of false information in subsequent rockburst suspicion analysis. Therefore, this invention uses amplitude triggering and spatial positioning to jointly identify valid microseismic events from the collected microseismic signals, effectively eliminating microseismic background interference.

[0030] In a specific embodiment, the identification of valid microseismic events is carried out as follows: S21. Since the amplitude of the microseismic signal reflects the strength of the energy released by the rock mass fracture to a certain extent, the elastic waves generated by the actual rock mass fracture usually have high energy, and their signal amplitude is significantly greater than the background interference. At this time, the event triggering condition is set based on the amplitude of the microseismic signal. The signal amplitude triggering threshold can be obtained by collecting the background noise signal of the geophone during the period when there is no obvious microseismic event, and calculating the root mean square value of its amplitude. Then, 3 to 5 times the root mean square value of the noise amplitude is taken as the signal amplitude triggering threshold. When the amplitude of the microseismic signal collected by the geophones distributed in different locations in the microseismic monitoring network triggers the threshold successively within the set time window (e.g., 50ms), it indicates that the signal has consistency across multiple stations and a certain energy intensity, and the possibility of background interference can be basically ruled out. At this time, it is recorded as a potential microseismic event.

[0031] S22. After initially eliminating background interference, in order to further confirm the authenticity of potential microseismic events, spatial location screening is also required. Since the microseismic signal will propagate from the rupture point to each detector at different times, for each potential microseismic event, the time difference of signal arrival is recorded by multiple detectors that triggered the event. Geometric positioning methods, such as travel time inversion, are used to calculate the spatial location of the event.

[0032] Applying to the above operations, the process of calculating the spatial location of the event using travel time inversion is as follows: S221. Establish a spatial rectangular coordinate system for positioning: take the center point of the tunnel entrance as the origin, the tunnel axis direction as the X-axis (positive direction pointing to the excavation direction), the vertical upward direction as the Z-axis, and the Y-axis as determined by the right-hand rule (perpendicular to the tunnel axis and horizontal to the right), and establish a spatial rectangular coordinate system in this way.

[0033] S222. Data Preparation: Under the established coordinate system, obtain the known spatial coordinates of the detector that triggered the potential microseismic event. and the time difference of arrival of the recorded signals That is, the observed travel time, in which Indicates the detector number.

[0034] S223. Model Establishment: Assume the spatial coordinates of the earthquake source to be solved are... The time of the earthquake was The P-wave velocity in the surrounding rock is ,in This can be obtained through wave speed testing.

[0035] S224. Constructing the objective function: For each detector, its theoretical timeout can be calculated as follows: The goal of positioning is to find an optimal set of... This minimizes the sum of squares of the differences between the observed and theoretical times of all triggered detectors, i.e., minimizes the objective function. , This indicates the number of detectors that trigger potential microseismic events.

[0036] S225. Iterative solution: The least squares iterative algorithm is used to solve the above nonlinear objective function. When the number of iterations or the residual meets the requirements, the final spatial location of the seismic source is output.

[0037] S23. Compare the spatial location of the event obtained from the solution with the known spatial range of the tunnel construction area. If the event location falls within this range, classify it as an artificial interference event and remove it. Confirm the remaining events as valid microseismic events.

[0038] S3. Extract energy and spatial location features from each valid microseismic event, compare them with the limiting conditions obtained from the statistical analysis of historical rockburst precursor feature data, and screen out suspected rockburst events.

[0039] Although the S2 step filters the raw microseismic signals by amplitude and spatial consistency to identify physically significant valid microseismic events, not all valid microseismic events are related to the rockburst incubation process. This is because the transition from microseismic activity to rockburst involves the continuous accumulation of energy and the evolution of microseismic events migrating towards the excavation face. Directly using all valid microseismic events for rockburst suspicion assessment may still result in the inclusion of a large number of non-rockburst precursor microseismic events, leading to an increased risk of misjudgment.

[0040] To address the aforementioned issues, this invention employs energy and spatial location characteristics to perform a secondary, refined screening of identified effective microseismic events, thereby identifying suspected events with rockburst precursor attributes and aiming to enhance the targeting of subsequent early warnings.

[0041] The following is a description of a specific embodiment: S31. For each valid microseismic event, the trigger time is taken as the starting time, and a time window containing the complete waveform signal is extracted. The square of the waveform signal amplitude within the time window is integrated over time to obtain the relative microseismic energy, which is used as an energy feature to characterize the relative magnitude of the surrounding rock fracture intensity.

[0042] S32. The spatial position of each effective microseismic event calculated by geometric positioning is orthogonally projected onto the tunnel axis to determine its projection point on the axial direction. The straight-line distance from the projection point along the tunnel axis to the current excavation face is calculated and recorded as the axial distance. This serves as a spatial position feature, characterizing the relative positional relationship between the microseismic event and the tunnel excavation front.

[0043] S33. Statistically analyze the relative energy values ​​of effective microseismic events recorded within a certain period (e.g., 6–24 hours) before each actual rockburst, and plot a frequency histogram. The energy range corresponding to the highest frequency group in the frequency histogram reflects the range where the energy distribution of microseismic events is most concentrated during the rockburst gestation stage, representing the typical energy level of precursory microseismic activity. Take the center value of this energy range as the lower limit of energy, which not only excludes low-energy background noise events but also retains rupture signals with precursor significance, and can characterize the lower limit energy level of microseismic activity during the historical rockburst gestation stage.

[0044] S34. Summarize the axial distance between the actual location of historical rockburst events and the tunnel face to form a distance sample set. Use the minimum value of this sample set as the near-end distance limit and the maximum value as the far-end distance limit. Then, combine the near-end distance limit and the far-end distance limit to form a high-incidence spatial range of rockbursts along the tunnel axis.

[0045] S35. If the energy characteristics of a valid microseismic event are greater than the lower limit of energy and its spatial location characteristics fall within the high-incidence spatial range of rockbursts, then the valid microseismic event shall be regarded as a suspected rockburst event.

[0046] In the above implementation process, only a lower limit threshold is set for energy characteristics because the higher the energy of a rockburst precursor event, the more dangerous it is. Setting a lower limit can exclude low-energy background events and avoid missing high-energy precursors due to setting an upper limit. Spatial location characteristics need to be limited to near and far limits because rockbursts have a clear spatial occurrence window. Only when a microseismic event is located within a certain distance in front of the tunnel face does it meet the mechanical conditions for rockburst gestation.

[0047] S4. Within the preset monitoring time window, calculate the percentage of occurrences of suspected rockburst events and analyze the changing trend of the event intervals.

[0048] After screening out suspected rockburst events through step S3, considering that a single suspected rockburst event may be sporadic or isolated, it is easy to make a misjudgment if the rockburst risk is determined based on individual events alone. Therefore, this invention further introduces two indicators: the proportion of the frequency of suspected rockburst events within a preset monitoring time window and the changing trend of the time interval between adjacent events, in order to comprehensively characterize the activity level and accelerated evolution characteristics of microseismic activity.

[0049] As one way to implement the present invention, the specific implementation of step S4 is as follows: S41. Within a preset monitoring time window, count the number of suspected rockburst events and the number of effective microseismic events. Use the ratio of the number of suspected rockburst events to the number of effective microseismic events as the occurrence frequency ratio. When the occurrence frequency ratio is higher, it indicates that the proportion of high-energy, near-face microseismic events in the current microseismic activity is higher, the rock mass fracturing activity tends to be more active, and it indicates that the risk of rockburst is increasing.

[0050] S42. Within the preset analysis time window, the selected suspected rockburst events are arranged in ascending order of their occurrence time to form an event time series.

[0051] S43. Calculate the time interval between adjacent events in the event time series to form an adjacent event time interval sequence. This time interval sequence reflects the evolution characteristics of microseismic activity in the time dimension. When the time interval shortens, it indicates that rockburst suspected events are occurring more and more frequently and the fracturing of the surrounding rock is accelerating. When the time interval increases, it indicates that the occurrence of rockburst suspected events is in a decaying state. Therefore, this sequence is the basis for assessing whether rockburst suspected events show temporal evolution.

[0052] S44. Considering that the analysis of shortening and increasing time intervals in step S43 only has physical meaning under the premise that the event time series shows a trend change pattern, but in actual monitoring, the occurrence of microseismic events may be affected by a variety of factors, and their time distribution sometimes shows irregular random fluctuations. If trend analysis is directly performed on random sequences, it is easy to lead to invalid analysis. Based on this, it is necessary to first test the randomness of the time interval sequences of adjacent events. Only when the test results show no randomness can trend analysis be carried out.

[0053] See Figure 2 As shown, in one optional implementation, the randomness test proceeds as follows: S441, each time interval in the sequence of adjacent event time intervals is compared with the previous interval, and a direction sign sequence is generated based on the increase / decrease relationship, for example... The + sign indicates that the current time interval is greater than the previous time interval, and the - sign indicates that the current time interval is less than the previous time interval. If the current time interval is the same as the previous time interval, the previous sign is used.

[0054] S442. In the direction symbol sequence, symbols that change continuously in the same direction are divided into a run. Each run reflects a unidirectional evolution process in which the time interval between microseismic events continues to increase or decrease. The total number of runs is counted.

[0055] S443. When each symbol in the direction symbol sequence is the opposite of its preceding symbol, the number of runs reaches its maximum. Summarizing the run counts yields the theoretical maximum number of runs, reflecting the most frequent and unsustainable stochastic limit state of microseismic activity time interval changes under a given number of events. For example, if the sequence of adjacent event time intervals contains m time intervals, since the first time interval has no preceding term, the direction symbols are calculated starting from the second interval, generating a total of m-1 direction symbols. In this case, if the symbols completely alternate, each symbol constitutes an independent run, therefore the theoretical maximum number of runs is m-1.

[0056] S444. Calculate the ratio of the total number of runs to the theoretical maximum number of runs, denoted as run density. This reflects the frequency of sequence changes. If the ratio is greater than or equal to the allowable density, it indicates that the actual number of runs is close to the theoretical maximum, the event time interval changes frequently and has no continuous directionality, and the microseismic activity exhibits highly discrete and trendless characteristics in time series. In this case, it is possible to test the randomness of the time interval sequence of adjacent events. Conversely, it indicates the existence of a long period of change in the same direction, and the microseismic activity exhibits non-randomness.

[0057] In the above-described embodiments, the allowable density can be set to 0.7, which corresponds to the case where the number of runs in the direction symbol sequence reaches more than half of the theoretical maximum value. When construction interference is strong, background microseismic events are frequent, and the proportion of random components in the signal increases. The value can be appropriately increased, for example, adjusted to 0.8. When construction interference is weak and the surrounding rock is in a relatively static and stable state, the microseismic signal mainly reflects the rock mass's own fracture behavior. The value can be appropriately decreased, for example, adjusted to 0.6, to improve the sensitivity of rockburst risk identification.

[0058] See Figure 3 As shown in S45, if the test result indicates non-randomness, then a trend analysis is performed on the time interval sequence of adjacent events. The specific analysis is as follows: S451, for each run in the time interval sequence of adjacent events, determine whether it is an increasing run or a decreasing run based on the sign of the same direction. Specifically, when the sign of the same direction of the run is +, the run is an increasing run; when the sign of the same direction of the run is -, the run is a decreasing run.

[0059] S452. Count the number of same-direction symbols in each run, and use that as the duration of the run.

[0060] S453. Summarize the number of increased runs and decreased runs, and calculate the average duration of all increased runs and the average duration of all decreased runs respectively.

[0061] S454. If the number of decreasing runs is greater than the number of increasing runs, or the average duration of decreasing runs is greater than the average duration of increasing runs, it indicates that the interval between suspected rockburst events tends to shorten and the frequency of microseismic activity increases. In this case, the time interval sequence of adjacent events is determined to be decreasing; otherwise, it is determined to be non-decreasing.

[0062] S5. Based on the changing trends of the frequency ratio and the interval between events, construct a joint criterion to determine whether there is a rockburst risk in a potential rockburst segment.

[0063] The specific criteria are as follows: when the following conditions are met simultaneously, the potential rockburst segment is determined to have a rockburst risk; otherwise, it is determined that there is no rockburst risk.

[0064] a) The frequency of suspected rockburst events is greater than the critical value, where the critical value reflects the lower limit of the proportion of suspected rockburst events in the overall microseismic activity. For example, it can be taken as more than half, i.e., 60%.

[0065] Condition a) Characterize the relative activity of suspected rockburst events to ensure that risk assessment is based on a sufficiently dense set of precursor signals.

[0066] b) The time interval sequence of adjacent events is determined to be non-random.

[0067] Condition b) is used to confirm the temporal evolution characteristics of suspected rockburst events, rather than accidental clustering caused by background noise, and is a prerequisite for identifying real rockburst precursors.

[0068] c) The trend of the time interval sequence analysis of adjacent events is decreasing.

[0069] Condition c) reflects the accelerated evolution behavior of the surrounding rock fracturing process.

[0070] The above three conditions construct a multi-evidence fusion criterion from three dimensions: event concentration (a), evolution regularity (b), and trend directionality (c). Only when microseismic activity is sufficiently frequent, non-random, and continuously accelerating will a rockburst risk warning be triggered, thereby improving the scientific rationality of the judgment.

[0071] S6. When a rockburst risk is determined, cluster analysis is used to determine the concentrated distribution range of suspected rockburst events, and rockburst early warning information containing the specific axial location range is output.

[0072] In a specific embodiment, rockburst early warning information is generated as follows: the spatial locations of all suspected rockburst events within the current monitoring time window are determined to form a risk location dataset.

[0073] Spatial clustering is performed on the risk location dataset to form risk location clusters.

[0074] Considering that rockburst fracture zones are typically located within a certain range ahead of the tunnel face and are distributed in a ring or sheet pattern along the tunnel axis, cluster analysis focuses on the spatial distribution characteristics along the tunnel axis when determining the concentrated distribution range of suspected rockburst events.

[0075] The specific spatial clustering is implemented as follows: First, the spatial location of each suspected rockburst event is orthogonally projected onto the tunnel axis to obtain its projection point on the tunnel axis. The engineering station number corresponding to the projection point is the axial mileage value of the event, forming an axial mileage sequence.

[0076] Subsequently, clustering based on neighborhood distance (e.g., setting a cluster radius of 5-10m) was used to group suspected rockburst events in spatial proximity into risk location clusters.

[0077] For each cluster of risk locations, calculate its minimum and maximum values ​​along the tunnel axis to form a continuous range of axial locations.

[0078] By designating each continuous axial position range as a specific risk location and using it as rockburst early warning information, we can provide the site with a risk section identification with clear spatial orientation, which is conducive to taking targeted prevention and control measures.

[0079] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0080] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0081] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0082] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations 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. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0083] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A tunnel rockburst position early warning method based on deep buried complex geological environment, characterized in that, The method comprises the following steps: deploying a microseismic monitoring network in a potential rockburst section; collecting microseismic signals by using the microseismic monitoring network, and screening the collected microseismic signals by using amplitude triggering and spatial positioning to identify effective microseismic events; extracting energy features and spatial position features of each effective microseismic event, comparing the energy features and the spatial position features with limited conditions obtained by statistical analysis of historical rockburst precursor feature data, and screening out rockburst suspicious events; calculating the proportion of the occurrence frequency of the rockburst suspicious events in a preset monitoring time window, and analyzing the change trend of the interval between the events; constructing a joint criterion based on the proportion of the occurrence frequency and the change trend of the interval between the events to determine whether the potential rockburst section has a rockburst risk; when it is determined that there is a rockburst risk, determining the concentrated distribution interval of the spatial positions of the rockburst suspicious events by cluster analysis, and outputting rockburst warning information containing a specific axial position range. 2.The tunnel rockburst position early warning method based on deep-buried complex geological environment according to claim 1, wherein: The microseismic monitoring network deployed in the potential rockburst section comprises the following steps: deploying a detector array along the tunnel axis direction in the potential rockburst section to form a microseismic monitoring network, and each array comprises a plurality of detectors. 3.The tunnel rockburst position early warning method based on deep complex geological environment according to claim 1, characterized in that: The effective microseismic events are identified by the following implementation process: setting an event triggering condition based on the amplitude of the microseismic signal, and when the amplitudes of the microseismic signals collected by the detectors distributed at different positions in the microseismic monitoring network successively trigger the threshold value within a set time window, a potential microseismic event is recorded; for each potential microseismic event, the time difference of the signals recorded by the multiple detectors triggering the event is used to calculate the spatial position of the event occurrence by geometric positioning; the calculated spatial position of the event occurrence is compared with the known spatial range of the tunnel construction area, and if the event position falls within the range, it is classified as an artificial interference event and is excluded, and the remaining events are confirmed as effective microseismic events. 4.The method for early warning of rock burst position of a tunnel based on deep-buried complex geological environment according to claim 1, characterized in that: The energy features and spatial position features of each effective microseismic event are extracted by the following steps: for each effective microseismic event, the triggering time is taken as the starting time, a time window containing the complete waveform signal is intercepted, the time integral of the waveform signal amplitude square in the time window is calculated to obtain the relative microseismic energy as the energy feature; the spatial position of each effective microseismic event calculated by geometric positioning is projected onto the tunnel axis, and the axial distance from the projection point to the current excavation tunnel face along the tunnel axis is calculated as the spatial position feature. 5.The method for early warning of rock burst position of a tunnel based on deep-buried complex geological environment according to claim 4, characterized in that: The rockburst suspicious events are screened by the following process: statistically analyzing the relative energy values of the effective microseismic events recorded in a certain time period before the occurrence of the real rockburst, drawing a frequency histogram, and taking the center value of the energy interval corresponding to the highest frequency group as the energy lower limit value; collecting the axial distances between the actual positions of the historical rockburst events and the tunnel tunnel face to form a distance sample set, and taking the minimum value of the sample set as the near-end limit value of the distance and the maximum value as the far-end limit value of the distance; the near-end limit value and the far-end limit value of the distance form a rockburst high-occurrence spatial interval along the tunnel axis; if the energy feature of a certain effective microseismic event is greater than the energy lower limit value and the spatial position feature falls within the rockburst high-occurrence spatial interval, the effective microseismic event is taken as a rockburst suspicious event. 6.The method for early warning of rock burst position of a tunnel based on deep-buried complex geological environment according to claim 1, characterized in that: The occurrence frequency proportion of the rock burst suspicious event is calculated, and a change trend of an event occurrence interval is analyzed as follows: In a preset monitoring time window, the number of rock burst suspicious events and the number of effective microseismic events are counted, and the proportion of the number of rock burst suspicious events to the number of effective microseismic events is taken as the occurrence frequency proportion; In the preset monitoring time window, the rock burst suspicious events screened out are arranged in ascending order according to occurrence time to form an event time sequence; The time intervals between adjacent events in the event time sequence are calculated to form an adjacent event time interval sequence; Randomness of the adjacent event time interval sequence is tested; If the test result shows non-randomness, trend analysis is performed on the adjacent event time interval sequence, and the trend is classified into a decreasing trend and a non-decreasing trend according to the analysis result. 7.The method for early warning of rock burst position of a tunnel based on deep-buried complex geological environment according to claim 6, characterized in that: The randomness of the adjacent event time interval sequence is tested as follows: Each time interval in the adjacent event time interval sequence is compared with a previous interval, and a direction symbol sequence is generated according to the increase / decrease relationship; In the direction symbol sequence, consecutive same-direction symbols are divided into a run, and the total number of runs is counted; When each symbol in the direction symbol sequence is opposite to the previous symbol, the number of runs reaches a maximum value, and the total number of runs is taken as a theoretically maximum run number; The ratio of the total number of runs to the theoretically maximum run number is calculated and is denoted as a run density; if the run density is greater than or equal to an allowable density, it is determined that the adjacent event time interval sequence has randomness, otherwise, it is determined that the adjacent event time interval sequence has non-randomness. 8.The method for early warning of rock burst position of a tunnel based on deep-buried complex geological environment according to claim 6, characterized in that: The trend analysis of the adjacent event time interval sequence is performed as follows: Each run divided from the adjacent event time interval sequence is determined to belong to an increase run or a decrease run according to the same-direction symbol; The number of same-direction symbols contained in each run is counted as the duration of the run; The number of increase runs and the number of decrease runs are counted, and the average duration of all increase runs and the average duration of all decrease runs are calculated; If the number of decrease runs is greater than the number of increase runs or the average duration of decrease runs is greater than the average duration of increase runs, it is determined that the adjacent event time interval sequence has a decreasing trend, otherwise, it is determined that the adjacent event time interval sequence has a non-decreasing trend. 9.The method for early warning of rock burst position of a tunnel based on deep-buried complex geological environment according to claim 8, characterized in that: The following implementation process is used to determine whether a potential rock burst section has a rock burst risk: When the following conditions are met simultaneously, it is determined that the potential rock burst section has a rock burst risk; otherwise, it is determined that there is no rock burst risk; a) The occurrence frequency proportion of the rock burst suspicious event is greater than a critical value; b) The adjacent event time interval sequence is determined to have non-randomness; c) The trend analysis of the adjacent event time interval sequence shows a decreasing trend. 10.The method for early warning of rock burst position of a tunnel based on deep-buried complex geological environment according to claim 1, characterized in that: The rock burst early warning information is output as follows: The spatial positions of all rock burst suspicious events in a current monitoring time window are located to form a risk position data set; The risk position data set is spatially clustered to form a risk position cluster; For each risk position cluster, the minimum value and the maximum value of the axial position along the tunnel axis are calculated to form a continuous axial position range; Each continuous axial position range is taken as a specific risk position as the rock burst early warning information.

Images