Deep hard rock tunnel fault detection method

Abstract

The invention provides a deeply-buried hard rock tunnel fault detection method, which is used for gradually processing a correlation matrix formed by reflected wave theoretical arrival time, window data and different comprehensive evaluation indexes on the basis that a micro-seismic monitoring system acquires rock fracture signals of the day. Searching three parameter indexes of a corresponding fault structure model through particle swarm search processing, thereby effectively utilizing information of arrival time of rock fracture signal reflection waves, and completing inversion of fault information by utilizing arrival time residual errors in consideration of processing speed and processing precision. And safe, stable, accurate and efficient deep-buried hard rock tunnel engineering is supported and promoted by accurate data.

Description

Technical Field

[0001] The present application relates to the field of geology, and specifically to a method for detecting faults in deep-buried hard rock tunnels. Background Art

[0002] Bad geological bodies and bad structural planes are one of the main control factors for deep engineering disasters. The cognitive effect seriously affects the disaster warning and prevention effect. Among them, there are fault disasters such as rock bursts, which occur as a result of the combined action of excavation disturbance and the development of fault fractures. Using the velocity structure difference to identify faults has gradually become a consensus in the geophysical exploration field.

[0003] On this basis, the microseismic monitoring technology is an effective technology for monitoring and warning rock mass failures, which is crucial for obtaining early warning information on rock fractures to predict the risk of rock bursts. In previous research cases on microseismic monitoring of deep hard rock engineering, most of them can predict the rock burst risk level and failure type of potential failure areas based on the evolution law, energy size, and moment tensor characteristics of the spatial and temporal distribution of rock fracture signals.

[0004] However, the information on the arrival time of the reflected wave of the rock fracture signal has not been fully utilized. Although the source location error of the rock fracture theoretically affects it, it is possible to enhance the detection result by combining a large number of rock fracture signals.

[0005] Therefore, how to effectively extract and utilize the arrival time information of the reflected wave of these large amounts of real-time updated rock fracture signals to establish a method for advanced fault detection suitable for deep-buried hard rock tunnels is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0006] The present application provides a method for detecting faults in deep-buried hard rock tunnels. Based on the rock fracture signals collected by the microseismic monitoring system on the same day, it gradually processes the correlation matrix composed of the theoretical arrival time of the reflected wave, window data, and different comprehensive evaluation indicators, and then searches for the three major parameter indicators of the corresponding fault structure model through particle swarm search processing. In this way, the arrival time information of the reflected wave of the rock fracture signal is effectively utilized, and the fault information is inversed using the arrival time residual while taking into account the processing speed and processing accuracy, so as to promote the safe, stable, accurate, and efficient deep-buried hard rock tunnel project with accurate data support.

[0007] The present application provides a method for detecting faults in deep-buried hard rock tunnels, and the method includes: Collect the rock fracture signals on the same day through a microseismic monitoring system pre-deployed at the engineering site of the deep-buried hard rock tunnel; Locate the source of the rock fracture signal, and obtain the two-dimensional position and waveform of the corresponding source as non-linear source parameters; Based on the non-linear seismic source parameters and the coordinates of the microseismic sensors in the microseismic monitoring system, the theoretical arrival time of the reflected wave is calculated through the time-distance curve model; Taking the theoretical arrival time of the reflected wave as the window center of the Hamming window, adjacent data before and after are extracted from the rock fracture signal as window data; For each data point in the window data, the corresponding comprehensive evaluation index considering the relative signal-to-noise ratio of the window function is calculated to form a correlation matrix composed of different comprehensive evaluation indexes; The first root-mean-square velocity in the homogeneous medium with the initial state set on the same day, the first shortest distance from the seismic source with the largest abscissa to the fault interface, and the first included angle between the fault interface and the tunnel axis are used as the starting input values for the particle swarm search process, and the corresponding particle swarm search process is carried out to search for the target root-mean-square velocity, target shortest distance, and target included angle with the maximum comprehensive evaluation index, so as to realize the update of the corresponding fault structure model for the deep-buried hard rock tunnel.

[0008] From the above content, it can be concluded that the present application has the following beneficial effects: Under the goal of advanced detection of faults in deep-buried hard rock tunnels, based on the rock fracture signals collected by the microseismic monitoring system on the same day, the present application gradually processes the theoretical arrival time of the reflected wave, window data, and the correlation matrix composed of different comprehensive evaluation indexes, and then searches for the three major parameter indexes of the corresponding fault structure model through the particle swarm search process, thereby effectively utilizing the information of the arrival time of the reflected wave of the rock fracture signal, and completing the inversion of fault information using the arrival time residual while taking into account the processing speed and processing accuracy, so as to promote the safe, stable, accurate and efficient deep-buried hard rock tunnel project with accurate data support. Description of the Drawings

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative efforts.

[0010] Figure 1 It is a schematic flow chart of a method for detecting faults in deep-buried hard rock tunnels of the present application. Detailed Embodiments

[0011] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0012] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments described herein can be implemented in an order different from that shown or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or modules does not necessarily have to be limited to those steps or modules clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products, or devices. The naming or numbering of steps that appear in the present application does not mean that the steps in the method flow must be executed in the time / logical sequence indicated by the naming or numbering. The named or numbered process steps can be changed in the execution order according to the technical purpose to be achieved, as long as the same or similar technical effects can be achieved.

[0013] The division of modules in the present application is a logical division. In actual implementation, there may be other division methods. For example, multiple modules can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed coupling, direct coupling, or communication connection between each other can be through some interfaces. The indirect coupling or communication connection between modules can be in an electrical or other similar form, which is not limited in the present application. And the modules or sub-modules described as separate components may or may not be physically separated, may or may not be physical modules, or may be distributed to multiple circuit modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of the present application.

[0014] Next, the method for detecting faults in deep-buried hard rock tunnels provided by the present application will be introduced.

[0015] First, refer to Figure 1 , Figure 1 which shows a schematic flow diagram of a method for detecting faults in deep-buried hard rock tunnels provided by the present application. The method for detecting faults in deep-buried hard rock tunnels provided by the present application specifically may include the following steps S101 to step S106: Step S101, collect the rock fracture signals of the day through a microseismic monitoring system pre-deployed at the engineering site of a deep-buried hard rock tunnel. It can be understood that microseismic monitoring technology is an effective technology for monitoring and warning rock mass failure. It is widely regarded as an important means to obtain precursor information of rock fractures and predict rockburst risks. It has been widely used in deep rock engineering. In high-buried areas, the number of microfracture events is large. The spatial distribution of these passive sources has been used to evaluate the rockburst risk level in the monitored area. However, it should be noted that it has not been used for fault detection because there is still a lack of suitable methods for calculating and picking up the arrival time of reflected waves of rock fracture signals for non-linear observation arrays.

[0016] In this case, the present application proposes a reflected wave propagation path model and theoretical arrival time calculation method based on layer inversion theory, a cross-correlation arrival time fitting method considering the signal-to-noise ratio information of rock fractures, and an inversion method based on particle swarm intelligence optimization theory, all of which are reflected in the following scheme introduction content.

[0017] Correspondingly, the present application needs to deploy a corresponding microseismic monitoring system at the engineering site of a deep-buried hard rock tunnel where fault advance detection is required, so as to collect the rock fragmentation signals required for the processing of the present application scheme through multiple microseismic sensors involved in the system.

[0018] It should be noted that the rock fragmentation signals collected here have a time span of 1 day (24 hours). Corresponding to the fault advance detection to be achieved by the present application scheme, the processing is carried out in units of 1 day time granularity. In this way, in practical applications, data support for updating the fault situation of deep-buried hard rock tunnel projects can be provided every day.

[0019] Of course, it can be understood that under the design of the present application scheme, the 1-day time span / processing unit can also be refined into smaller time spans, such as 12 hours, 8 hours, etc. After replacing with a smaller time span, the scheme processing remains unchanged.

[0020] Specifically, for the microseismic monitoring system involved here, in addition to including microseismic sensors, it can also include system structures such as data transmission links, which can be flexibly adjusted according to the actual situation in practical applications. The core is to obtain the corresponding rock fragmentation signals through the deployed microseismic sensors.

[0021] Specifically, as an exemplary embodiment, the microseismic monitoring system may be configured with a microseismic sensor array (i.e., an array composed of different microseismic sensors). Microseismic sensors are deployed at three monitoring section positions in the deep-buried hard rock tunnel, and are specifically deployed in the manner of 3 microseismic sensors, 2 microseismic sensors, and 3 microseismic sensors in sequence (i.e., the cross-section distribution method of the number of sensors of 3-2-3, a total of 8 unidirectional sensors), covering the area 40m - 120m behind the tunnel face, and transmitted back to the background through optical fiber transmission or cloud upload communication methods.

[0022] Among them, the microseismic sensors can be specifically deployed at the top and / or side walls of the deep-buried hard rock tunnel at the monitoring section position, and are symmetrically arranged based on the cross-section, and the cross-section is usually perpendicular to the tunnel direction. In the case of covering the area 40m - 120m behind the tunnel face, the rock fragmentation signals required by this application can be conveniently and effectively collected.

[0023] It can be understood that on the data transmission link, devices in the form / name of data acquisition units or collectors can be configured to facilitate centralized management of microseismic sensors and also facilitate the transmission of the rock fragmentation signals collected by the microseismic sensors.

[0024] The background can be the background equipment involved in specific application scenarios such as project departments, laboratories, or cloud platforms.

[0025] In addition, for microseismic sensors and the rock fragmentation signals they collect, they themselves belong to the category of existing technologies in this field, so no specific elaboration will be made here.

[0026] In addition, it can be understood that the rock series signals collected here are for detecting the fault structure. In addition to the natural seismic sources that occur on-site during the development of the deep-buried hard rock project, there may also be some seismic sources that may appear due to engineering reasons, or in a sense, are "active" seismic sources. For the latter, obviously, it can have the characteristics of a pre-planned schedule along with engineering operations, and the relative detonation time can be determined. Thus, compared with exploring the signals that appear along with the seismic source from the rock fragmentation signals, the rock fragmentation signals can be collected specifically according to the determined detonation time or the seismic source with a controllable detonation time.

[0027] Correspondingly, as an exemplary embodiment, by deploying a microseismic monitoring system in advance at the engineering site of the deep-buried hard rock tunnel, collecting the rock rupture signals of the day may specifically include: For the seismic source with a controllable detonation time, by deploying a microseismic monitoring system in advance at the engineering site of the deep-buried hard rock tunnel, collecting the rock rupture signals of the day.

[0028] It can be understood that the rock fragmentation signals collected under this setting are more targeted. In terms of signal collection work, significantly streamlined data collection costs can be incurred to obtain highly accurate and effective rock fragmentation signals.

[0029] In addition, after obtaining the rock fragmentation signals in the initial stage, corresponding preprocessing can also be involved to eliminate invalid signals and abnormal signals. Moreover, the signal quality can be further enhanced in terms of standardizing the signal length, etc. Correspondingly, for the above embodiments, obviously, the required preprocessing costs can be further reduced.

[0030] Step S102, perform source location on the rock fracture signal, and take the obtained two-dimensional position and waveform corresponding to the source as non-linear source parameters; It can be understood that after the rock fragmentation signals containing various information are collected by the microseismic sensors, corresponding source location processing can be carried out to determine the corresponding source and configure it as a two-dimensional position (x i , y i ) in a two-dimensional coordinate system. In addition, after determining the source and its corresponding signal content, its specific waveform can also be determined. Thus, the information in these two aspects can be used as non-linear source parameters to lay a foundation for subsequent data processing.

[0031] Among them, it should be understood that the source location processing involved here itself belongs to the scope of the existing technology in this field. Of course, in actual applications, according to actual application requirements, technical means further optimized and improved on the basis of the existing technology or even novel self-developed technical means can also be introduced (the same applies to the existing technology mentioned in other aspects of this application).

[0032] Specifically, as an exemplary embodiment, the source location of the rock fracture signal here may specifically include: After calibrating the rock fracture signal according to the direct P-wave and direct S-wave in the signal, perform source location through the time difference location method.

[0033] It can be understood that the rock fragmentation signals accompanied by the source may have corresponding signal characteristics in terms of the direct P-wave and direct S-wave. Therefore, compared with directly performing source location, first calibrating according to the direct P-wave and direct S-wave to determine the signal range of the specific source location can further reduce the data processing volume and also promote a higher-precision source location effect in the subsequent process.

[0034] In addition, it can also be seen in the embodiments herein that for seismic source positioning, the Time Difference (TD) method can be specifically used to achieve it. Of course, in addition to the TD method, in practical applications, other types of seismic source positioning methods can also be used, which can be flexibly adjusted according to actual needs.

[0035] Step S103, based on the nonlinear seismic source parameters and the coordinates of the microseismic sensors in the microseismic monitoring system, calculate the theoretical arrival time of the reflected wave through the travel-time curve model. After obtaining the nonlinear seismic source parameters (the two-dimensional position and waveform of the seismic source), the coordinates of the microseismic sensors in the corresponding microseismic monitoring system can be combined as the model input parameters, and through the travel-time curve model (time-distance curve model) designed in this application, the theoretical arrival time of the reflected wave can be calculated for subsequent use in combination with the indirect processing of the actual arrival time of the reflected wave.

[0036] Furthermore, as an exemplary embodiment, based on the nonlinear seismic source parameters and the coordinates of the microseismic sensors in the microseismic monitoring system, calculating the theoretical arrival time of the reflected wave through the travel-time curve model may specifically include: Based on the nonlinear seismic source parameters and the coordinates of the microseismic sensors in the microseismic monitoring system, in combination with the given third root mean square velocity, the third shortest distance, and the third included angle between the fault interface and the tunnel axis, calculate the theoretical arrival time of the reflected wave through the travel-time curve model. For the travel-time curve model, when y i ≤y1, there is: y i ≥y1, there is: Where, corresponding to the seismic source i, t i is the theoretical arrival time of the reflected wave, S i is the distance between the microseismic sensor and the seismic source i, v ori is the velocity set during the initial positioning of the microseismic monitoring system, h1 is the third shortest distance from the seismic source with the largest abscissa to the fault interface, H is the depth of the velocity layer, a is the third included angle between the fault interface and the tunnel axis, b is the angle between the connection line between the farthest seismic source M i+1 and the other seismic sources and the vertical axis direction in the xy plane, x i is the abscissa in the two-dimensional position of the seismic source, y i is the ordinate in the two-dimensional position of the seismic source, x1 is the abscissa of the microseismic sensor, y1 is the ordinate of the microseismic sensor, v is the third root mean square velocity in the homogeneous medium, and a′ is the angle between the connection line between the seismic source and the microseismic sensor and the vertical direction.

[0037] It should be understood that for the waveform in the non-linear source parameters, which is a time series and involves the application of timestamps, it can provide the time points related to different types of parameters and other information.

[0038] Next, a specific expansion and explanation will be given to the travel-time curve model specifically designed in the present application above.

[0039] For the fault medium involved in vertical construction or the vertical tunnel direction, the travel-time curve model can be expressed by the following formula: Among them, for the source M i , t is the arrival time of the reflected wave, H is the depth of the velocity layer, v0 represents the root-mean-square velocity value in the homogeneous medium, X i is the horizontal distance between the source and the microseismic sensor, t0 is the travel time of the self-excitation-reception path, h i is the shortest distance from the source i to the fault interface.

[0040] Among them, v0, which needs to be obtained in the subsequent inversion process, can be converted into the laminar velocity field using the Dix formula for fault identification.

[0041] For the fault medium involved in horizontal construction or the horizontal tunnel direction, the travel-time curve model of the source in the tunnel observation system can be derived by the same method and can be specifically expressed by the following formula: Among them, corresponding to the source i, corresponding to the xy platform (or the x-y plane), t i is the arrival time of the reflected wave, S i is the distance between the microseismic sensor and the source i, h i is the shortest distance from the source i to the fault interface, defined as half of the self-excitation-reception travel path, a is the angle between the fault interface and the tunnel axis, and a' is the angle between the connection line between the source i and the microseismic sensor and the vertical direction.

[0042] At the same time, for the self-excitation-reception propagation path corresponding to each non-linear source in the tunnel, taking the propagation path corresponding to the source closest to the surface of the structure as a reference, the paths corresponding to other sources can be expressed by the following formula and involve two cases of a and b: y Among them, corresponding to the source i, h i is the shortest distance from the source i to the fault interface, h1 is the third shortest distance from the source with the largest abscissa to the fault interface, a is the angle between the fault interface and the tunnel axis, and b is the farthest source Mi+1 The angle between the connecting line between the source and the other sources and the vertical axis direction in the xy plane, x i Is the abscissa in the two-dimensional position of the source, y i Is the ordinate in the two-dimensional position of the source, x1 is the abscissa of the microseismic sensor, y1 is the ordinate of the microseismic sensor, and b can be expressed by the following formula and involves two cases of a and b: Among them, corresponding to the source i, t i Is the theoretical arrival time of the reflected wave, S i Is the distance between the microseismic sensor and the source i, v ori Is the velocity set during the initial positioning of the microseismic monitoring system, h1 is the third shortest distance from the source with the largest abscissa to the fault interface, H is the depth of the velocity layer, a is the third angle between the fault interface and the tunnel axis, and b is the angle between the connecting line between the farthest source M i+1 And the other sources and the vertical axis direction in the xy plane, x i Is the abscissa in the two-dimensional position of the source, y i Is the ordinate in the two-dimensional position of the source, x1 is the abscissa of the microseismic sensor, y1 is the ordinate of the microseismic sensor, v is the third root mean square velocity in the homogeneous medium, and a′ is the angle between the connecting line between the source and the microseismic sensor and the vertical direction.

[0043] In this way, through the above brief formula deduction logic, the construction of the non-linear source time-distance curve model of the layer velocity model is completed, which is closely related to the establishment of the root mean square velocity field. After establishing the time-distance curve model, the corresponding theoretical arrival time of the reflected wave can be calculated according to the given parameters.

[0044] Step S104, taking the theoretical arrival time of the reflected wave as the window center of the Hamming window, extracting the adjacent data before and after from the rock fracture signal as the window data; It can be understood that for disordered microseismic sources, the lack of a linear coherence axis makes it complicated to directly observe the arrival time of the reflected wave, difficult to directly determine the actual arrival time of the reflected wave, and thus impossible to construct a least squares solution between the theoretical calculation and the actual arrival times of the reflected wave.

[0045] In this case, the present application considers establishing an objective function for the similarity between the theoretical arrival time of the reflected wave and the signal within the specified window instead of the least squares solution between the two, so as to determine the corresponding fault structure based on the cross-correlation arrival time fitting method considering the signal-to-noise ratio information of rock fractures and the inversion method based on the particle swarm intelligence optimization theory.

[0046] After obtaining the arrival time of the reflected wave theory, the present application can window the previous rock fracture signal and use the Hamming window to extract the window data adjacent to the arrival time of the reflected wave theory before and after (the arrival time of the reflected wave theory is located at the center point) for data processing. Among them, the window data can be denoted as M windows 。

[0047] Step S105: For each data point in the window data, calculate the corresponding comprehensive evaluation index considering the relative signal-to-noise ratio of the window function to form a correlation matrix composed of different comprehensive evaluation indexes; It can be understood that within the window data M windows contains a large number of data points within the time range corresponding to the window size. For this, corresponding to the subsequent particle swarm search processing, it is necessary to calculate the comprehensive evaluation index of each data point respectively to form a correlation matrix composed of different comprehensive evaluation indexes.

[0048] For the comprehensive evaluation index involved here, it can be denoted as K. Specifically, for the widely used methods of evaluating similarity information, they can be divided into the stacking method and the similarity coefficient method. For the same seismic source conditions, the present application specifically considers that the energy and signal-to-noise ratio of different rock fracture signals vary greatly, and proposes an evaluation index considering the relative signal-to-noise ratio of the window function.

[0049] Specifically, as an exemplary embodiment, the calculation formula of the comprehensive evaluation index can be: ]> Where K is the comprehensive evaluation index, M is the window data length, N is the number of sensors, S is the relative signal-to-noise ratio, S i is the relative signal-to-noise ratio of the current window data, S i ′ is the relative signal-to-noise ratio of the next window data, S max is the maximum relative signal-to-noise ratio in all window data, M windows is the current window data, σ 2 is the variance operator, n is the variance of the set of all window data, f is the amplitude value, and the subscripts i and j are used to distinguish the time series of different windows.

[0050] It can be understood that in the correlation matrix [K] composed of different comprehensive evaluation indexes, the corresponding parameters with the maximum index (i.e., the maximum K value), namely the second root mean square velocity (v max1 ), the second shortest distance (h max1 ), and the second included angle (a max1 ), correspond to the target to be inverted in the particle search processing, and are therefore used as one of the input data for the subsequent particle search processing.

[0051] In step S106, the first root mean square velocity in the homogeneous medium in the initial state set on that day, the first shortest distance from the earthquake source with the largest horizontal coordinate to the fault interface, and the first angle between the fault interface and the tunnel axis, and the second root mean square velocity, second shortest distance, and second angle corresponding to the maximum index in the correlation matrix are used as the starting input values of the particle swarm search process. The corresponding particle swarm search process is carried out to search for the target root mean square velocity, target shortest distance, and target angle that obtain the maximum comprehensive evaluation index, so as to update the corresponding fault structure model of the deep hard rock tunnel.

[0052] It is understandable that the traditional method of analyzing the RMS velocity field of the travel time curve usually relies on a simple grid search, which may be limited in the multi-extreme value inversion in practice. In this regard, the present application takes into account the advantages of the particle swarm algorithm in the nonlinear update search direction, and specifically proposes a fault parameter inversion method based on particle swarm optimization. This method combines the prior information of the initial grid search to further refine the search of the initial inversion results, thereby optimizing the inversion of the fault information in two steps.

[0053] As for the model input, it can be understood that, on the one hand, it involves the first root mean square velocity (v in ), the shortest distance from the earthquake source with the largest horizontal coordinate to the fault interface (h in ), the first angle between the fault interface and the tunnel axis (a in ) If it is the first day of the project, it needs to be preset. If it is the nth day of the project, the parameters obtained by inversion on the previous day are used. On the other hand, it also involves the second root mean square velocity (v max1 ), the second shortest distance (h max1 ) and the second angle (a max1 ).

[0054] The goal of the particle swarm search process is to search as close as possible to the state of the maximum index in the correlation matrix [K] composed of different comprehensive evaluation indicators through inversion, or to search as close as possible to the second root mean square velocity (v max1 ), the second shortest distance (h max1 ) and the second angle (a max1 ) the optimal value P t+1 , that is, the target root mean square velocity (v max2 ), the shortest distance to the target (h max2 ) and target angle (a max2 ), the three are the three major parameter indicators of the fault structure model.

[0055] The target RMS velocity (vmax2 ), corresponding to the root mean square velocity model. In the case where the processing link here is used to update a velocity field model, further, combined with the target shortest distance (h max2 ) and the target angle (a max2 ), the fault structure can be better reflected from the overall level, and additional content can be added to the velocity field model to form a fault structure model reflecting the current fault situation, thus achieving the design goal of using the arrival time residual to invert the fault information.

[0056] Among them, the fault structure model can be denoted as F(v, h, a).

[0057] In addition, it can be understood that although the particle swarm search processing itself can adopt existing algorithms, corresponding specific parameters need to be configured for the search object.

[0058] That is to say, it can be understood that the particle swarm search processing may also involve other aspects of input data, such as the search range.

[0059] In this regard, as an exemplary embodiment, the search range of the particle swarm search processing may specifically include the range formed by adding and subtracting the preset velocity value search step from the second root mean square velocity value (i.e., v max1 ±N V , [v max1 -N V , v max1 +N V ), the range formed by adding and subtracting the preset shortest distance search step from the second shortest distance (i.e., h max1 ±N h , [h max1 -N h , h max1 +N h ), and the range formed by adding and subtracting the preset angle search step from the second angle (i.e., a max1 ±N a , [a max1 -N a , a max1 +N a ). And this also means that three aspects of search steps are preset in advance, namely N V , N h , and N a ; In addition, the configuration parameters of the particle swarm search processing may also include the number of particles N p , the current flight speed v t , the position P t , the flight range P bound , the historical optimal value P of a single particle best , the historical optimal value G of the populationbest The inertia weight w, the cognitive constant c1, the social constant c2, the random value r1 between [0, 1], and the random value r2 between [0, 1].

[0060] In addition, it can be understood that, considering the complex and cumbersome characteristics of the processing logic of the particle swarm search process itself and that it is existing content, no further detailed description of the specific search process will be given here.

[0061] Regarding the fault structure model F(v, h, a), it can be understood that this is a data model in this field that can be used to reflect the relevant fault structure from three dimensions, rather than the structure model that is conventionally understood and directly indicates the three-dimensional coordinates of each part of the fault structure.

[0062] In addition, it can also involve specific data application processes such as local storage, off-site storage, result output, result display, outputting a prompt for completing the search, or other aspects of data analysis and processing.

[0063] It can be understood that in terms of the data application process of the results, it can be flexibly adjusted according to the pre-configured and real-time configured data application strategies / rules.

[0064] Finally, for the above solution content, generally speaking, under the goal of advanced detection of faults in deep-buried hard rock tunnels, based on the rock fracture signals collected by the microseismic monitoring system on the day of acquisition, this application gradually processes the correlation matrix composed of the arrival time of reflected waves, window data, and different comprehensive evaluation indicators, and then searches for the three major parameter indicators of the corresponding fault structure model through the particle swarm search process. In this way, the information of the arrival time of the reflected waves of the rock fracture signals is effectively utilized, and the inversion of fault information using the arrival time residual is completed while taking into account the processing speed and processing accuracy, so as to promote the safe, stable, accurate and efficient deep-buried hard rock tunnel project with accurate data support.

[0065] The above has introduced the method for detecting faults in deep-buried hard rock tunnels provided by this application in detail. Specific examples are used in this article to elaborate on the principle and implementation manner of this application. The description of the above embodiments is only used to help understand the core idea of this application; at the same time, for those skilled in the art, based on the idea of this application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to this application.

Claims

1. A method for detecting faults in deep-buried hard rock tunnels, characterized in that, The method includes: Collecting the rock fracture signals of the day through a microseismic monitoring system pre-deployed at the engineering site of a deep-buried hard rock tunnel; Performing hypocenter location on the rock fracture signals, and taking the obtained two-dimensional position and waveform of the corresponding hypocenter as non-linear hypocenter parameters; Based on the non-linear hypocenter parameters and the coordinates of the microseismic sensors of the microseismic monitoring system, calculating the theoretical arrival time of the reflected wave through a travel-time curve model; Taking the theoretical arrival time of the reflected wave as the window center of a Hamming window, and extracting the adjacent data before and after from the rock fracture signals as window data; For each data point in the window data, calculating the corresponding comprehensive evaluation index considering the relative signal-to-noise ratio of the window function, and forming a correlation matrix composed of different comprehensive evaluation indexes; Taking the first root-mean-square velocity in a homogeneous medium with the initial state set on the day, the first shortest distance from the hypocenter with the largest abscissa to the fault interface, and the first included angle between the fault interface and the tunnel axis, and the second root-mean-square velocity, the second shortest distance, and the second included angle corresponding to the largest index in the correlation matrix as the starting input values for particle swarm search processing, and carrying out the corresponding particle swarm search processing to search for the target root-mean-square velocity, the target shortest distance, and the target included angle that maximize the comprehensive evaluation index, so as to realize the update of the corresponding fault structure model of the deep-buried hard rock tunnel.

2. The method according to claim 1, characterized in that, The microseismic monitoring system is configured with a microseismic sensor array. Microseismic sensors are deployed at three monitoring section positions of the deep-buried hard rock tunnel, and are specifically deployed in the manner of 3 microseismic sensors, 2 microseismic sensors, and 3 microseismic sensors in sequence, covering the area 40m - 120m behind the tunnel face, and transmitted back to the background through optical fiber transmission or cloud upload communication methods.

3. The method according to claim 1, characterized in that, The collecting the rock fracture signals of the day through a microseismic monitoring system pre-deployed at the engineering site of a deep-buried hard rock tunnel includes: For a hypocenter with controllable detonation time, collecting the rock fracture signals of the day through the microseismic monitoring system pre-deployed at the engineering site of the deep-buried hard rock tunnel.

4. The method according to claim 1, wherein The performing hypocenter location on the rock fracture signals includes: After calibrating the rock fracture signals according to the direct P-wave and direct S-wave in the signals, performing the hypocenter location through a time difference location method.

5. The method according to claim 1, wherein The calculating the theoretical arrival time of the reflected wave through a travel-time curve model based on the non-linear hypocenter parameters and the coordinates of the microseismic sensors of the microseismic monitoring system includes: Based on the non-linear hypocenter parameters and the coordinates of the microseismic sensors of the microseismic monitoring system, combining the given third root-mean-square velocity, the third shortest distance, and the third included angle between the fault interface and the tunnel axis, and calculating the theoretical arrival time of the reflected wave through the travel-time curve model; For the travel-time curve model, when y i ≤ y1, we have: y i When ≥ y1, there is: Among them, corresponding to the seismic source i, t i is the theoretical arrival time of the reflected wave, S i is the distance between the microseismic sensor and the seismic source i, v ori is the velocity set during the initial positioning of the microseismic monitoring system, h1 is the third shortest distance, H is the depth of the velocity layer, a is the third included angle, b is the connection line between the farthest seismic source M i+1 and the remaining seismic sources and the angle between the vertical axis direction in the xy plane, x i is the abscissa in the two-dimensional position of the seismic source, y i is the ordinate in the two-dimensional position of the seismic source, x1 is the abscissa of the microseismic sensor, y1 is the ordinate of the microseismic sensor, v is the third root mean square velocity, a′ is the angle between the connection line between the seismic source and the microseismic sensor and the vertical direction.

6. The method according to claim 5, wherein The calculation formula of the comprehensive evaluation index is: Among them, K is the comprehensive evaluation index, M is the window data length, N is the number of sensors, S is the relative signal-to-noise ratio, S i is the relative signal-to-noise ratio of the current window data, S i ′ is the relative signal-to-noise ratio of the next window data, S max is the maximum relative signal-to-noise ratio among all window data, M windows is the current window data, σ 2 is the variance operator, n is the variance of the set of all window data, f is the amplitude value, and the subscripts i and j are used to distinguish the time series of different windows.

7. The method according to claim 1, characterized in that, The search range of the particle swarm search process includes the range formed by adding and subtracting a preset velocity value search step from the second root mean square velocity value, the range formed by adding and subtracting a preset shortest distance search step from the second shortest distance, and the range formed by adding and subtracting a preset angle search step from the second angle; The configuration parameters for the particle swarm search process further include the number of particles N p , the current flying speed v t , the position P t , the flying range P bound , the historical optimal value P of a single particle best , the historical optimal value G of the population best , the inertia weight w, the cognitive constant c1, the social constant c2, the random value r1 between [0, 1] and the random value r2 between [0, 1].

Images