Tunnel surrounding rock stability automatic monitoring and early warning method and system

Abstract

The invention relates to the technical field of geophysical exploration, and particularly discloses a tunnel surrounding rock stability automatic monitoring and early warning method and system. The method comprises the steps that surrounding rock microseismic signals, convergence displacement and osmotic pressure data are collected through a space-time synchronization array to generate a multi-source fusion data set; based on a dynamic constitutive model separation frequency dispersion effect, extracting correlation characteristics of a microseismic event rate and a displacement rate; generating an instability energy density index; dynamically switching impact or creep damage criteria according to the deformation behavior of the surrounding rock; inputting the indexes into corresponding criteria to generate a damage state result; and if the result exceeds the threshold value, verifying the spatial focusing property of the micro-seismic event cluster, and outputting an early warning instruction when the focusing degree exceeds the critical value. Through multi-source fusion and a dynamic criterion mechanism, the accuracy of tunnel surrounding rock instability early warning is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of geophysical exploration, and particularly to an automatic monitoring and early warning method and system for the stability of tunnel surrounding rock. Background Art

[0002] In the field of geophysical exploration, microseismic data analysis technology has long faced the following limitations:

[0003] Traditional methods rely on a single microseismic event rate or displacement rate parameter and fail to integrate the spatio-temporal correlation of multi-dimensional physical quantities such as seepage pressure and displacement. In a tunnel environment, interference such as construction vibration and equipment noise causes microseismic signal distortion. A single index is difficult to accurately characterize the evolution process of surrounding rock damage and is prone to missing the risk of local instability.

[0004] Existing static damage criteria (such as fixed energy thresholds) cannot adapt to the stage characteristics of the surrounding rock from progressive creep to sudden shock failure. Affected by the superposition of geological structures and construction disturbances, the deformation mode of the tunnel surrounding rock dynamically changes, and rigid criteria are difficult to respond in real time, resulting in poor warning timeliness and high false alarm rates. Summary of the Invention

[0005] The present invention provides an automatic monitoring and early warning method and system for the stability of tunnel surrounding rock, and its main purpose is to solve the problem of low warning accuracy caused by the inability of static damage criteria to dynamically adapt to the transformation of the surrounding rock deformation stage.

[0006] To achieve the above object, an automatic monitoring and early warning method for the stability of tunnel surrounding rock provided by the present invention includes:

[0007] S1. Collect microseismic signals, convergence displacements, and seepage pressure data of the geological body through a spatio-temporal synchronization array to generate a multi-source fusion data set;

[0008] S2. Extract the correlation features between the microseismic event rate and the displacement rate based on the multi-source fusion data set;

[0009] S3. Generate an instability energy density index of the geological body based on the correlation features;

[0010] S4. Dynamically switch the damage criterion according to the deformation behavior of the geological body: adopt an impact damage criterion in the sudden deformation stage and a creep damage criterion in the progressive deformation stage;

[0011] S5. Input the instability energy density index into the damage criterion corresponding to the deformation stage to generate a determination result of the damage state of the geological body;

[0012] S6. When the damage state determination result exceeds the stage threshold, trigger the spatial focusing verification of the microseismic event cluster. If the focusing degree of the microseismic event cluster within the preset spatio-temporal window is greater than the preset critical value, output a warning instruction; otherwise, return to step S1 to collect data again.

[0013] Optionally, the method for collecting microseismic signals, convergence displacements, and seepage pressure data of a geological body through a spatio-temporal synchronization array to generate a multi-source fusion data set includes:

[0014] Collecting microseismic signals of a geological body based on a distributed sensor array;

[0015] Measuring the convergence displacement of the geological body based on displacement monitoring equipment;

[0016] Obtaining seepage pressure data of the geological body based on a seepage pressure monitoring device;

[0017] Aligning the timestamps of the microseismic signals, the convergence displacements, and the seepage pressure data through a time synchronization protocol;

[0018] Fusing the aligned microseismic signals, the convergence displacements, and the seepage pressure data to generate a multi-source fusion data set.

[0019] Optionally, before extracting the correlation features between the microseismic event rate and the displacement rate based on the multi-source fusion data set, it further includes:

[0020] Separating the dispersion effect of the multi-source fusion data set based on a preset dynamic constitutive model, and the expression of the dynamic constitutive model is:

[0021]

[0022] Wherein, is the stress, is the elastic modulus, is the strain, is the viscosity coefficient characterizing the stress-strain relationship of the geological body, is the time identifier;

[0023] Processing the separated multi-source fusion data set by applying a decoupling algorithm, wherein the decoupling algorithm is:

[0024]

[0025] Wherein, is the signal frequency component, is the frequency-dependent time delay, is the complex transfer function, is the imaginary unit;

[0026] Extracting the microseismic event rate and the displacement rate from the decoupled multi-source fusion data set.

[0027] Optionally, the associated features of the extracted microseismic event rate and displacement rate include:

[0028] Calculate the covariance of the microseismic event rate and displacement rate for each sampling point;

[0029] Accumulate the covariance to form an associated feature calculation formula, and solve the correlation coefficient in the associated feature calculation formula by the least squares method. The correlation coefficient characterizes the associated feature of the coupling strength between the microseismic event rate and the displacement rate.

[0030] Optionally, generating the instability energy density index of the geological body based on the associated feature includes:

[0031] Substitute the associated feature into a preset energy calculation function to obtain the instability energy density index of the geological body, where the preset energy calculation function is:

[0032]

[0033] where, is the instability energy density index, is the associated feature, is the microseismic event rate, is the displacement rate, is the time identifier.

[0034] Optionally, dynamically switching the damage criterion according to the deformation behavior of the geological body includes:

[0035] Real-time monitor the deformation rate of the geological body;

[0036] When the mutation of the deformation rate exceeds a preset threshold, trigger the impact damage criterion;

[0037] When the mutation of the deformation rate continuously is lower than the preset threshold, trigger the creep damage criterion;

[0038] Load the damage criterion parameters based on the triggered impact damage criterion or creep damage criterion.

[0039] Optionally, loading the damage criterion parameters based on the triggered impact damage criterion or creep damage criterion includes:

[0040] Analyze the currently activated damage criterion type;

[0041] Call the mechanical parameter library matching the damage criterion type;

[0042] Map the instability energy density index to the mechanical parameter library to generate damage criterion parameters including the strength reduction coefficient.

[0043] Optionally, generating the determination result of the damage state of the geological body includes:

[0044] Inputting the instability energy density index into the currently activated damage criterion;

[0045] Processing the instability energy density index through the algorithm of the damage criterion to generate a stability evaluation value;

[0046] Comparing the stability evaluation value with a preset stability threshold, and outputting the determination result of the damage state representing the critical failure probability of the rock mass.

[0047] Optionally, verifying the spatial focusing of the triggered microseismic event clusters includes:

[0048] Identifying the microseismic event clusters within a preset spatio-temporal window, and calculating the three-dimensional geometric centroid of the microseismic event clusters;

[0049] Traversing the Euclidean distances from each event in the microseismic event clusters to the three-dimensional geometric centroid;

[0050] Taking the average value of the Euclidean distances as the focusing degree, and verifying the spatial focusing of the microseismic event clusters based on the focusing degree and a preset critical value.

[0051] To solve the above problems, the present invention also provides an automatic monitoring and early warning system for the stability of tunnel surrounding rock, and the system includes:

[0052] A data acquisition module, configured to collect microseismic signals, convergence displacements, and seepage pressure data of the geological body through a spatio-temporal synchronization array, and generate a multi-source fusion data set;

[0053] An associated feature extraction module, configured to extract the associated features between the microseismic event rate and the displacement rate based on the multi-source fusion data set;

[0054] An instability energy density index generation module, configured to generate the instability energy density index of the geological body based on the associated features;

[0055] A damage criterion adoption module, configured to dynamically switch the damage criterion according to the deformation behavior of the geological body: adopting an impact damage criterion in the sudden deformation stage and a creep damage criterion in the progressive deformation stage;

[0056] A damage state determination module, configured to input the instability energy density index into the damage criterion corresponding to the deformation stage, and generate the determination result of the damage state of the geological body;

[0057] An early warning instruction generation module is used to trigger the spatial focusing verification of the microseismic event cluster when the damage state determination result exceeds the stage threshold. If the focusing degree of the microseismic event cluster within the preset time-space window is greater than the preset critical value, an early warning instruction is output; otherwise, return to step S1 to collect data again.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] 1. Microseismic, displacement and seepage pressure data are collected through a time-space synchronous array. The dispersion effect is separated by combining a dynamic constitutive model to construct an anti-interference multi-source fusion data set. The sensor layout and decoupling algorithm are optimized for the long and narrow space characteristics of the tunnel to improve the microseismic event location accuracy. Based on covariance and the least square method, the correlation characteristics between the microseismic event rate and displacement rate are extracted to quantify the co-evolution law of surrounding rock damage, overcoming the limitations of a single parameter in a complex tunnel environment.

[0060] 2. The impact / creep damage criterion is switched in real time according to the sudden change behavior of the surrounding rock deformation rate, and the strength reduction coefficient is generated by associating with the mechanical parameter library. This mechanism accurately distinguishes the sudden collapse and slow rheological risks induced by tunnel construction, avoiding misjudgment by the static model. And through the instability energy density index and the spatial focusing verification of the microseismic event cluster (such as three-dimensional centroid distance analysis), the early warning sensitivity of local instability of the tunnel is improved on the premise of ensuring a low false alarm rate, forming a closed-loop control logic. Description of the Drawings

[0061] Figure 1 It is a schematic flow chart of the automatic monitoring and early warning method for tunnel surrounding rock stability provided by an embodiment of the present invention;

[0062] Figure 2 It is a functional module diagram of the automatic monitoring and early warning system for tunnel surrounding rock stability provided by an embodiment of the present invention;

[0063] The realization, functional characteristics and advantages of the object of the present invention will be further described in conjunction with the embodiments with reference to the drawings. Detailed Embodiments

[0064] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0065] An embodiment of the present application provides an automatic monitoring and early warning method for the stability of tunnel surrounding rock. The execution subject of the automatic monitoring and early warning method for the stability of tunnel surrounding rock includes, but is not limited to, at least one of electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the automatic monitoring and early warning method for the stability of tunnel surrounding rock can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to: a single server, a server cluster, a cloud server, or a cloud server cluster, etc. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, Content Delivery Network (CDN), and big data and artificial intelligence platforms.

[0066] Referring to Figure 1 As shown, it is a schematic flowchart of the automatic monitoring and early warning method for the stability of tunnel surrounding rock provided by an embodiment of the present invention. In this embodiment, the automatic monitoring and early warning method for the stability of tunnel surrounding rock includes:

[0067] S1. Collect microseismic signals, convergence displacements, and seepage pressure data of the geological body through a spatio-temporal synchronization array to generate a multi-source fusion data set.

[0068] In the embodiment of the present invention, the step of collecting microseismic signals, convergence displacements, and seepage pressure data of the geological body through a spatio-temporal synchronization array to generate a multi-source fusion data set includes:

[0069] Collect microseismic signals of the geological body based on a distributed sensor array;

[0070] Measure the convergence displacement of the geological body based on displacement monitoring equipment;

[0071] Obtain the seepage pressure data of the geological body based on a seepage pressure monitoring device;

[0072] Align the timestamps of the microseismic signals, the convergence displacements, and the seepage pressure data through a time synchronization protocol;

[0073] Fuse the aligned microseismic signals, the convergence displacements, and the seepage pressure data to generate a multi-source fusion data set.

[0074] Specifically, the spatio-temporal synchronization array refers to a three-dimensional monitoring network composed of a distributed sensor array, displacement monitoring equipment, and a seepage pressure monitoring device, which realizes the unified calibration of spatially distributed sensors and the time reference through a time synchronization protocol to ensure the spatio-temporal consistency of multi-source data.

[0075] Specifically, the distributed sensor array refers to multiple microseismic sensors arranged along the axial and radial directions of the tunnel, covering the surrounding rock area inside the tunnel in an array form, and used to capture the tiny vibration signals generated by internal fractures in the rock mass.

[0076] Specifically, the displacement monitoring device refers to a device based on laser ranging or fiber optic sensing technology, used to measure the convergence displacement on the surface or inside of the tunnel surrounding rock, that is, the relative change in spatial position caused by the deformation of the surrounding rock inside the tunnel.

[0077] Specifically, the seepage pressure monitoring device refers to a piezoresistive sensor installed in the boreholes of the tunnel surrounding rock, used to monitor the change in groundwater pore pressure and reflect the influence of the rock mass seepage characteristics and water pressure on stability.

[0078] Specifically, the time synchronization protocol refers to calibrating the clocks of each monitoring device through the Global Positioning System (GPS) or the Network Time Protocol (NTP) to ensure that the timestamp error of different types of data does not exceed 1 millisecond.

[0079] Specifically, the multi-source fusion data set refers to a structured data set containing multi-dimensional physical quantities formed by aligning microseismic signals, convergence displacements, and seepage pressure data along a unified time axis for subsequent comprehensive analysis.

[0080] Furthermore, the layout of the distributed sensor array includes: setting a monitoring section every 50 meters along the longitudinal direction of the tunnel, and installing 1 three-component acceleration sensor (sampling frequency 1000Hz, range ±5g) at the crown, waist, and foot of each section to form a spatial array coverage; the sensors are fixed to the surface of the tunnel surrounding rock through anchor bolts, and the coupling agent is epoxy resin to ensure the effective transmission of vibration signals.

[0081] Furthermore, the measurement of the convergence displacement includes: arranging 2 groups of laser rangefinders (accuracy ±0.01mm) at each monitoring section to respectively monitor the crown settlement and sidewall convergence of the tunnel, and the length of the ranging baseline is 1.2 times the diameter of the tunnel; the rangefinders are automatically triggered for measurement, the sampling interval is 10 minutes, and the data is transmitted to the edge computing unit through the RS485 bus.

[0082] Furthermore, the acquisition of seepage pressure data includes: burying vibrating wire piezometers (range 0-10MPa, accuracy ±0.1%FS) in the boreholes of the surrounding rock around the tunnel (depth 10-15 meters), and sealing the boreholes with cement mortar to ensure the true conduction of water pressure.

[0083] Furthermore, the piezometers are connected to the data acquisition instrument through cables, the sampling frequency is 1 time per hour, and the temperature compensation coefficient is automatically calibrated.

[0084] Furthermore, each monitoring device is built with a GPS module to receive satellite clock signals (UTC time), and the clock error is controlled within ±1 ms through a hardware synchronization circuit; the edge computing unit runs the NTP protocol to periodically calibrate the clocks of all devices to form a unified time reference.

[0085] Furthermore, after the microseismic signals are subjected to anti-aliasing filtering (cutoff frequency 500 Hz) and amplification (gain 40 dB), they are converted into digital signals (resolution 16 bits) by a 24-bit ADC, and the format is the SEGY standard seismic data format; the convergence displacement and seepage pressure data are converted into the JSON format, which includes fields such as timestamps, monitoring point IDs, and physical quantity values.

[0086] Furthermore, the fusion algorithm uses time window matching (window width 50 ms) to interpolate and align the three types of data according to timestamps to generate a three-dimensional matrix [timestamp × monitoring point × physical quantity type], which is stored in a distributed database.

[0087] Generally speaking, this step solves the problem of misjudgment of surrounding rock stability caused by single-parameter monitoring in the prior art through spatio-temporal synchronization arrays and multi-source data fusion. The specific manifestations are as follows: multi-source data collaboratively reflect the state of the surrounding rock, avoiding missed judgments caused by noise interference in a single signal (such as only microseismic signals); the time synchronization accuracy ensures the spatio-temporal consistency of the data, providing a reliable basis for subsequent energy analysis and damage criterion switching; the fusion data set covers multi-dimensional parameters of mechanics (microseismic), geometry (displacement), and hydraulics (seepage pressure), improving the comprehensiveness of monitoring.

[0088] Generally speaking, the multi-source fusion data set generated in S1 serves as the input basis for subsequent steps: the microseismic signals provide the original waveform data for extracting the microseismic event rate in S2; the convergence displacement data is used to calculate the displacement rate in S2; the seepage pressure data assists in the judgment of deformation behavior in S4, such as high seepage pressure may exacerbate rock mass creep; the unified timestamp ensures the accuracy of spatio-temporal window matching of microseismic event clusters in S6. Through this step, the conversion from physical signal acquisition to structured data is achieved, constituting the data cornerstone of the entire monitoring and early warning process.

[0089] S2. Extract the correlation features between the microseismic event rate and the displacement rate based on the multi-source fusion data set.

[0090] In the embodiment of the present invention, before extracting the correlation features between the microseismic event rate and the displacement rate based on the multi-source fusion data set, it further includes:

[0091] Dispersion effect separation is performed on the multi-source fusion data set based on a preset dynamic constitutive model, and the expression of the dynamic constitutive model is:

[0092]

[0093] Among them, is stress, is the elastic modulus, is strain, is the viscosity coefficient characterizing the stress-strain relationship of the geological body, is the time identifier;

[0094] Apply a decoupling algorithm to process the separated multi-source fusion data set. Among them, the decoupling algorithm is:

[0095]

[0096] Among them, is the signal frequency component, is the frequency-dependent time delay, is the complex transfer function, is the imaginary unit;

[0097] Extract the microseismic event rate and displacement rate from the decoupled multi-source fusion data set.

[0098] In the embodiments of the present invention, the correlation feature is calculated by the least squares method:

[0099]

[0100] Among them, is the number of sampling points, is the correlation feature, is the average value of the microseismic event rate of the geological body within the preset spatio-temporal window, is the average value of the displacement rate of the geological body within the preset spatio-temporal window, is at the microseismic event rate at the moment, is at the displacement rate at the moment, is the time identifier.

[0101] In the embodiments of the present invention, the extraction of the correlation feature between the microseismic event rate and the displacement rate includes:

[0102] Calculate the covariance between the microseismic event rate and the displacement rate for each sampling point;

[0103] Accumulate the covariance to form a correlation feature calculation formula, and solve the correlation coefficient in the correlation feature calculation formula by the least squares method. The correlation coefficient characterizes the correlation feature of the coupling strength between the microseismic event rate and the displacement rate.

[0104] Specifically, a dynamic constitutive model refers to a mathematical model used to describe the stress-strain relationship of geological bodies. Considering the coupling effect of elastic deformation and viscous deformation, the dispersion effect in microseismic signals can be separated through this model, reflecting the true mechanical state of rock masses.

[0105] Specifically, the dispersion effect refers to the phenomenon that when seismic waves propagate in geological bodies, different frequency components cause waveform broadening and phase shift due to wave velocity differences, which will affect the accuracy of microseismic event location and feature extraction.

[0106] Specifically, the decoupling algorithm refers to a signal processing algorithm that compensates for the dispersion effect through a complex transfer function, realizes phase correction of different frequency components, and restores the true waveform of microseismic signals.

[0107] Specifically, the complex transfer function refers to a complex function that characterizes the phase and amplitude changes of a signal in the frequency domain, and is used to correct the frequency-dependent time delay caused by dispersion, so that each frequency component of the signal arrives synchronously.

[0108] Specifically, the least squares method refers to a parameter estimation method that solves the correlation coefficient by minimizing the sum of the squares of the errors between the observed values and the model predicted values, and quantifies the linear correlation between the microseismic event rate and the displacement rate.

[0109] Specifically, the correlation coefficient refers to a dimensionless parameter calculated by the least squares method, which is used to characterize the coupling strength between microseismic activities and surrounding rock deformation. The larger the value, the higher the synergy between the two.

[0110] Furthermore, the dynamic constitutive model parameters and their determination methods include: The elastic modulus is determined according to the geological exploration report to determine the lithology (such as granite E = 60 GPa, sandstone E = 30 GPa), with a value range of 20 - 80 GPa, and calibrated through indoor triaxial compression tests; the viscosity coefficient is set based on the rheological characteristics of rock masses. Soft rocks (such as mudstone) take 10 6 -10 8 Pa・s, hard rocks (such as granite) take 10 4 -10 6 Pa・s, and is fitted through creep test data; the strain is calculated from the convergence displacement data generated by S1, and the formula is , where is the displacement change (unit: mm), is the monitoring baseline length (such as the tunnel diameter of 5 m).

[0111] Furthermore, the model application and dispersion separation include: Substitute the stress σ (inverted from microseismic signals), strain and strain rate in the multi-source fusion dataset into the dynamic constitutive model to separate the elastic component and the viscous component ; Perform frequency-domain transformation (FFT) on the separated microseismic signals to identify the phase shift characteristics of different frequency components.

[0112] Furthermore, the calibration of frequency-dependent time delay Δt(f) includes: Select two points with a known distance (such as 10 m) in the tunnel surrounding rock, excite an artificial seismic source (such as an electric spark seismic source), and collect the arrival times of different frequency components; Establish the mapping relationship with frequency , for example: at , at , and obtain by polynomial fitting, where are the fitting coefficients.

[0113] Furthermore, the calculation of the complex transfer function and signal correction includes: For each frequency component , calculate the complex transfer function , where is the imaginary unit; Apply to the dispersive signal by multiplying in the frequency domain, perform inverse Fourier transform (IFFT) after completing phase correction, and obtain the decoupled time-domain signal with the waveform distortion rate controlled within 5%.

[0114] Furthermore, the calculation of the microseismic event rate includes: Set a trigger threshold (3 times the background noise, such as 200 μV) for the decoupled microseismic signals, and use the STA / LTA algorithm (short-time window / long-time window ratio) to identify effective events; Take 1 hour as the time window and count the number of events within this window to obtain the microseismic event rate (unit: times / h) at

[0115] Furthermore, the calculation of the displacement rate includes: Based on the convergence displacement data of S1, take the displacement difference between adjacent time points (with an interval of 1 day), and calculate in days, with the unit of mm / day. For example, the displacement at , the displacement at , then in days.

[0116] Furthermore, the setting of the spatio-temporal window includes: Select 24-hour data, corresponding to n = 8640 sampling points (10 ms / point) as the time window; With the monitoring section as the center, extend 50 m along the tunnel axis to ensure the spatial correlation of the data as the spatial window.

[0117] Furthermore, mean calculation and covariance accumulation include: calculating the mean of microseismic event rate and displacement rate in the time and space window; calculating for each sampling point and The product of accumulating the numerator ; Calculate the denominator , which is the variance of the microseismic event rate.

[0118] Furthermore, the correlation coefficient solution includes: substituting into the least squares formula to obtain the correlation coefficient For example, if the numerator is 50 and the denominator is 2500, then , indicating that microseismic activity is weakly correlated with displacement changes.

[0119] In general, this step solves the problem of feature misjudgment caused by microseismic signal distortion in the existing technology by separating the dispersion effect and extracting the correlation features: the dynamic constitutive model and decoupling algorithm eliminate the dispersion effect, improve the accuracy of microseismic event positioning, and avoid event misjudgment due to waveform broadening; the correlation features quantify the synergy between microseismic activity and deformation, overcome the one-sidedness of single parameter analysis, and provide a multidimensional basis for the assessment of surrounding rock damage status; the least squares method ensures the objectivity of the correlation coefficient calculation, reduces the error caused by artificial threshold setting, and improves the reliability of the monitoring system.

[0120] In general, the decoupled microseismic signals provide accurate waveforms for the spatial focusing calculation of microseismic event clusters in S6, ensuring the accuracy of positioning coordinates. The correlation coefficient, as the core parameter of the instability energy density indicator in S3, is used to quantify the accumulation of rock mass damage energy. The microseismic event rate and displacement rate provide the data basis for the switching of dynamic damage criteria in S4. For example, when the correlation coefficient suddenly increases, the impact damage criterion is triggered.

[0121] S3. Generate an instability energy density index of the geological body based on the associated features.

[0122] In an embodiment of the present invention, generating the instability energy density index of the geological body based on the correlation feature includes:

[0123] Substituting the associated features into a preset energy calculation function, the instability energy density index of the geological body is obtained, wherein the preset energy calculation function is:

[0124]

[0125] in, is the instability energy density index, is the associated feature, is the microseismic event rate, is the displacement rate, It is a time stamp.

[0126] In detail, the instability energy density index after fitting with the preset energy calculation function is a dimensionless parameter. The instability energy density index obtained after fitting is compared with the instability energy density obtained by experiment. The instability energy density obtained by experiment has a dimensionless parameter. , which is used to characterize the rate of accumulation of destructive energy per unit volume of rock mass per hour.

[0127] For example, when a granite sample is subjected to uniaxial compression, when the instability energy density index reaches a certain reference value, the corresponding rock mass energy storage density is , thus determining the energy density unit of on-site monitoring as .

[0128] In detail, the instability energy density index refers to a physical quantity used to quantify the destructive energy accumulated over time within a unit volume of a geological body. It reflects the energy evolution state of the rock mass from damage to instability through the synergistic effect of coupled microseismic activity and displacement deformation.

[0129] In detail, the energy calculation function refers to a preset mathematical model that converts multi-source monitoring data into energy accumulation indicators through the product operation of correlation characteristics, microseismic event rate and displacement rate, revealing the speed and intensity of energy accumulation inside the rock mass.

[0130] In detail, the correlation feature refers to a dimensionless parameter obtained by the least squares method, which characterizes the degree of dynamic coupling between the microseismic event rate and the displacement rate. A larger value indicates a higher synergy of the mechanical behavior during the rock damage process.

[0131] In detail, the microseismic event rate refers to the number of microfracture events occurring inside a geological body per unit time. It is obtained by performing time-series statistics on the decoupled microseismic signals and reflects the activity of fracture expansion inside the rock mass.

[0132] In detail, the displacement rate refers to the time rate of change of the convergent displacement on the surface or inside of a geological body, which is calculated from the displacement monitoring data at adjacent time points and reflects the speed and trend of surrounding rock deformation.

[0133] Furthermore, the theoretical basis of the physical modeling logic of the energy calculation function is that the essence of rock instability is a process in which energy accumulation exceeds a critical value. The microseismic event rate reflects the frequency of energy release, the displacement rate reflects the speed of deformation energy accumulation, and the correlation characteristics characterize the coupling efficiency of the two.

[0134] Furthermore, the mathematical expression of the physical modeling of the energy calculation function is ,in: The correlation coefficient from S2, with a value range of [0,1], is verified to have a positive correlation with the energy coupling efficiency through triaxial compression tests of rock mass; The unit is times / h. Based on the microseismic signal after S2 decoupling, the STA / LTA triggering algorithm (threshold is 3 times the background noise) is used to count the number of events within a 1-hour window. The unit is mm / day. The displacement difference between adjacent days is calculated from the converged displacement data of S1.

[0135] Specifically, the displacement rate was converted from mm / day to mm / h to ensure consistency with the time unit (h) of the microseismic event rate; and a mapping relationship between the instability energy density index and the actual fracture energy was established through indoor rock sample fracture tests.

[0136] Furthermore, real-time data acquisition includes: Improve the real-time retrieval of calculation results from S2, such as the calculated value of the correlation coefficient in a certain period is 0.3 (medium coupling state); By reading the statistical results of microseismic events after S2 decoupling, for example, the number of events in the current hour is 20 times / h; By obtaining the displacement rate in the S1 fusion dataset, for example, the current daily displacement change is 0.6 mm / day, which is converted to 0.025 mm / h.

[0137] Further, we can get This value indicates that each cubic meter of rock mass accumulates 0.15 joules of destructive energy per hour. When this value approaches the critical value calibrated in the laboratory (such as 1J / m³·h), the rock mass enters an instability warning state.

[0138] Specifically, the critical value is determined through indoor rock sample fracture tests. For example, the critical energy density for instability of sandstone samples under triaxial compression is 0.8 J / m³·h. Adjustments are made based on geological conditions. For example, the threshold value decreases by 10% for every 100m increase in burial depth (to account for the influence of ground stress), and the threshold value for soft rock phases is reduced by 30% compared to hard rock phases (due to differences in rheological properties).

[0139] In detail, the corrected threshold is generated by the laboratory benchmark threshold, burial depth correction factor and lithology correction factor.

[0140] In general, this step solves the problem in existing technologies that a single indicator (such as only the number of microseismic events or displacement) cannot accurately reflect the energy accumulation of the rock mass through energy density calculation based on multi-parameter fusion: it integrates the synergistic effect of microseismic activity and deformation behavior to avoid misjudgment caused by noise interference of a single signal. For example, when the microseismic event rate increases but the displacement rate remains unchanged, the energy indicator can be identified as an invalid fracture; it quantifies the energy accumulation rate to provide a quantitative basis for dynamic early warning, overcoming the subjectivity of "qualitative judgment" in traditional methods. For example, the rate of increase of energy density can be used to distinguish between gradual destruction and sudden instability of the rock mass; it calibrates in conjunction with laboratory physical models to ensure the physical interpretability of the energy indicator and establish a direct connection between the field monitoring data and rock mechanics theory.

[0141] Generally speaking, when the instability energy density index exceeds the critical value and the rising rate changes suddenly, the impact damage criterion is triggered; if the energy density remains higher than the threshold but the rate is stable, the creep damage criterion is activated; the energy density index and the spatial focusing of microseismic event clusters (calculated by S6) form a multi-dimensional verification. For example, when the instability energy density index is greater than 0.8 J / m³·h and the focusing degree of the event cluster is less than 10 m, the instability risk is double-confirmed; the energy density calculation depends on the correlation characteristics of S2 and the original monitoring data of S1, forming a complete logical chain of "signal acquisition - feature extraction - energy analysis - early warning decision-making" to ensure the data traceability of each step.

[0142] S4. Dynamically switch the damage criterion according to the deformation behavior of the geological body: adopt the impact damage criterion in the sudden deformation stage and the creep damage criterion in the progressive deformation stage.

[0143] In the embodiment of the present invention, the dynamically switching the damage criterion according to the deformation behavior of the geological body includes:

[0144] Real-time monitor the deformation rate of the geological body;

[0145] When the deformation rate changes suddenly and exceeds the preset threshold, trigger the impact damage criterion;

[0146] When the sudden change of the deformation rate continues to be lower than the preset threshold, trigger the creep damage criterion;

[0147] Load the damage criterion parameters based on the triggered impact damage criterion or creep damage criterion.

[0148] Specifically, the loading of the damage criterion parameters based on the triggered impact damage criterion or creep damage criterion includes:

[0149] Analyze the type of the currently activated damage criterion;

[0150] Call the mechanical parameter library matching the type of the damage criterion;

[0151] Map the instability energy density index to the mechanical parameter library to generate damage criterion parameters including the strength reduction coefficient.

[0152] Specifically, the deformation behavior refers to the comprehensive manifestation of the morphological changes such as displacement and crack expansion of the geological body under the action of stress, which is characterized by parameters such as deformation rate and deformation mode, and reflects the evolution stage of the rock mass stability.

[0153] Specifically, the dynamically switching the damage criterion means automatically switching the applicable damage assessment standard according to the real-time deformation characteristics of the geological body, so that the criterion matches the current failure mode of the rock mass and improves the early warning accuracy.

[0154] Specifically, the impact damage criterion refers to the criterion for the sudden deformation stage, which is applicable to the scenario of rapid rock mass rupture, characterized by a sudden increase in the microseismic event rate and a sharp rise in energy density, and focuses on the instantaneous release of failure energy.

[0155] Specifically, the creep damage criterion refers to the criterion for the progressive deformation stage, which is applicable to the scenario of slow rock mass creep, characterized by a continuous increase in displacement rate and a steady accumulation of energy density, and focuses on the long-term accumulation of deformation energy.

[0156] Specifically, the sudden change in deformation rate means that the displacement rate of the geological body changes significantly within a short time, exceeding the normal fluctuation range, usually indicating the sudden instability of the internal structure of the rock mass.

[0157] Specifically, the preset threshold refers to the critical value of the deformation rate preset according to geological conditions and historical data, which is used to judge the type of deformation behavior and is divided into an impact threshold and a creep threshold.

[0158] Specifically, the mechanical parameter library refers to the database that stores the mechanical parameters corresponding to different damage criteria, including the strength reduction coefficient, elastic modulus attenuation factor, etc., which are related to parameters such as the lithology and buried depth of the geological body.

[0159] Specifically, the strength reduction coefficient is a dimensionless parameter used to quantify the attenuation degree of the rock mass strength with the development of damage. The smaller the value, the more the bearing capacity of the rock mass decreases and the higher the instability risk.

[0160] Furthermore, the real-time monitoring of the deformation rate includes:

[0161] Extracting the convergence displacement data from the multi-source fusion dataset generated by S1, calculating the displacement rate using a sliding time window (such as 1 day); performing Kalman filtering on the displacement rate sequence to eliminate the interference of monitoring noise, and setting the filtering parameters according to the rock mass type (for hard rock, the process noise covariance Q = 0.01, for soft rock, the process noise covariance Q = 0.05); calculating the time derivative of the deformation rate , which characterizes the degree of deformation acceleration, and the five-point central difference method can be used: , with a time step Δt = 1 hour to ensure real-time performance.

[0162] Furthermore, the preset threshold setting mechanism includes: the impact damage threshold mechanism and the creep damage threshold mechanism. Among them, the impact damage threshold mechanism means that for hard rock, it takes 0.1 mm / day², and for soft rock, it takes 0.5 mm / day², which is determined based on indoor rock sample impact tests. When exceeding this value, it is considered that there is a sudden change in the deformation rate; the creep damage threshold refers to taking 1 / 5 of the impact threshold, that is, 0.02 mm / day² for hard rock and 0.1 mm / day² for soft rock. When it is lower than this value for 48 hours, it is determined as progressive deformation.

[0163] Further, the criterion triggering logic includes: impact damage criterion triggering and creep damage criterion triggering. Among them, impact damage criterion triggering means that when is greater than the impact threshold, and the instability energy density index of S3 increases by more than 20% within 1 hour, the sudden failure warning logic is triggered; creep damage criterion triggering means that when is less than the creep threshold and lasts for 48 hours, and at the same time the microseismic event rate has no significant fluctuation (change amount < 10%), the progressive damage assessment process is activated.

[0164] Further, the system generates a criterion type identifier in real time (0 = impact criterion, 1 = creep criterion), stores it in the status register, and provides it for subsequent modules to call through the API interface.

[0165] Specifically, the mechanical parameter library is stored using key-value pairs. The key is {lithology + burial depth + criterion type}, and the value is the corresponding parameter set. For example: key: {granite + 500m + impact criterion}, value: {strength reduction coefficient 0.6, elastic modulus decay rate 0.3 / day, energy release index 1.5}; key: {mudstone + 200m + creep criterion}, value: {strength reduction coefficient 0.8, elastic modulus decay rate 0.1 / day, energy release index 1.2}.

[0166] Specifically, the instability energy density mapping includes:

[0167] Establish a mapping function between the instability energy density index and the strength reduction coefficient, and use piecewise linear interpolation, including:

[0168] Impact criterion: when , the strength reduction coefficient is 0.8; when , the coefficient is 0.5.

[0169] Creep criterion: when , the strength reduction coefficient is 0.9; when , the coefficient is 0.6.

[0170] Calculate the corresponding coefficient of the current in real time. For example , the interpolated strength reduction coefficient is 0.65.

[0171] Generally speaking, this step solves the problem in the prior art that the fixed criterion cannot adapt to the differences in the deformation stages of rock masses by dynamically switching the damage criterion: different criteria are adopted for sudden deformation and progressive deformation to avoid early warning lags caused by continuing to use the creep criterion in the case of impact failure scenarios, or false alarms caused by misusing the impact criterion during progressive deformation; the strength reduction coefficient and energy density are mapped in real time, enabling the criterion parameters to be dynamically adjusted according to the damage degree, improving the timeliness and accuracy of surrounding rock stability assessment; combining the change rate of the deformation rate and the energy index for double verification to reduce criterion mis-triggering caused by fluctuations in a single parameter, such as excluding sudden increases in the deformation rate caused by external interferences such as blasting construction.

[0172] Generally speaking, the instability energy density index, as one of the core parameters for criterion triggering, is also used to calculate the strength reduction coefficient, forming a logical chain of "energy accumulation - deformation behavior - criterion parameters"; the loaded damage criterion parameters (such as the strength reduction coefficient) are used for the weight calculation in the spatial focusing verification of microseismic event clusters in S6. For example, the critical value of the focusing degree is adjusted from 10m to 8m under the impact criterion, improving the early warning sensitivity; the deformation rate data is sourced from the displacement monitoring in S1, and the criterion switching results are stored in the database, providing a basis for the early warning decision-making in subsequent S5 to ensure the traceability of the entire process data.

[0173] S5. Input the instability energy density index into the damage criterion corresponding to the deformation stage to generate the determination result of the damage state of the geological body.

[0174] In the embodiment of the present invention, the generation of the determination result of the damage state of the geological body includes:

[0175] Input the instability energy density index into the currently activated damage criterion;

[0176] Process the instability energy density index through the algorithm of the damage criterion to generate a stability evaluation value;

[0177] Compare the stability evaluation value with a preset stability threshold and output the determination result of the damage state representing the critical failure probability of the rock mass.

[0178] Specifically, the instability energy density index refers to a physical quantity calculated by coupling the microseismic event rate, displacement rate, and their associated characteristics, which is used to quantify the rate of damage energy accumulation per unit volume of the geological body and reflect the evolution process of the rock mass from damage to instability.

[0179] Specifically, the deformation stage refers to different evolution states of the geological body under stress, including the sudden deformation stage (such as impact failure) and the progressive deformation stage (such as creep failure), which is dynamically determined by S4 according to the deformation rate.

[0180] Specifically, the damage criterion refers to the evaluation criteria established based on rock mechanics theory, which is divided into impact damage criterion and creep damage criterion, and is dynamically switched according to the deformation stage, and is used to convert the energy index into a stability evaluation value.

[0181] Specifically, the stability evaluation value refers to the quantitative value obtained after processing the instability energy density index through the damage criterion algorithm, which characterizes the current stability degree of the geological body, and the smaller the value, the worse the stability.

[0182] Specifically, the preset stability threshold refers to the critical value preset according to geological conditions and engineering experience, which is used to compare the stability evaluation value and divide the rock mass damage state level (such as safe, warning, dangerous).

[0183] Specifically, the damage state determination result refers to the probability of critical failure of the rock mass output in the form of probability, which is generated by combining the comparison result of the stability evaluation value and the threshold, and provides a quantitative basis for engineering decision-making.

[0184] Furthermore, the input of the instability energy density index includes:

[0185] Retrieve the instability energy density index from the real-time calculation result of S3. The data format is a floating-point numerical value with a time stamp (unit: J / m³・h). The interface adopts the RESTAPI protocol, and a data validity verification mechanism is set: if the instability energy density index appears negative or exceeds the theoretical maximum value (such as 10 J / m³・h), trigger a data anomaly alarm and automatically retrieve the average value of the previous 1 hour for replacement.

[0186] Furthermore, the deformation stage association includes: synchronously read the currently activated damage criterion type (impact or creep) output by S4, and associate through key-value pairs and the criterion type to ensure the matching of the energy index and the applicable criterion. For example: when the impact criterion is activated, the input data is marked as ; when the creep criterion is activated, it is marked as .

[0187] Specifically, the evaluation value calculation model of the impact damage criterion algorithm is , where is the impact state stability evaluation value, and the value range is [0,1]. The smaller the value, the closer it is to instability; is the impact criterion threshold, which is called from the mechanical parameter library of S4 (such as taking 1.0 J / m³·h for hard rock); is the lithology correction coefficient (granite , sandstone ), which is automatically matched from the parameter library.

[0188] Specifically, the evaluation value calculation model of the creep damage criterion algorithm is , where is the stability evaluation value of the impact state, with a value range of [0, 1]. The smaller the value, the closer it is to instability; is the impact criterion threshold, which is called from the mechanical parameter library of S4 (for example, for hard rock, take ); is the lithology correction coefficient (for granite , for sandstone ), which is automatically matched from the parameter library.

[0189] Furthermore, GPU parallel computing is used to accelerate the calculation of the evaluation value. The floating-point operation precision is optimized for the linear model of the impact criterion and the exponential model of the creep criterion respectively (single-precision floating-point error ≤ 1e-6), and the intermediate variables of the algorithm are saved in real time to form a traceable calculation log.

[0190] Furthermore, the stability threshold system uses a three-level threshold to divide the damage state:

[0191] Safety threshold T1: For example, 0.6. When the evaluation value S ≥ T1, it is determined to be in a safe state;

[0192] Warning threshold T2: For example, 0.3. When T2 < S < T1, a warning is triggered;

[0193] Danger threshold T3: For example, 0.1. When S ≤ T2, it is determined to be in a dangerous state.

[0194] Furthermore, the threshold is dynamically adjusted according to the burial depth. For example, for every 100m increase in the burial depth, each level of the threshold is reduced by 10% (through calculation, where H is the burial depth).

[0195] Furthermore, the calculation of the critical failure probability is based on the relative position of the evaluation value and the threshold, and an S-shaped function is used to map the probability: , is the critical failure probability, with a value range of [0, 1]; is the median value of the current threshold interval (such as the median value of the warning interval ); is the slope parameter (for hard rock = 10, for soft rock = 5), which controls the probability change gradient.

[0196] Furthermore, the structured output includes: timestamp (accurate to seconds); damage state level (safe / warning / dangerous); critical failure probability (such as "0.75, indicating a 75% probability of entering the critical failure state"); associated criterion type and parameters.

[0197] Generally speaking, through the coupled analysis of dynamic criteria and energy indicators, this step solves the problem that the fixed criteria in the prior art cannot adapt to the differences in rock mass failure modes: different evaluation models are adopted for sudden deformation and progressive deformation to avoid early warning lag caused by the use of creep criteria in impact scenarios. For example, the characteristics of sudden energy increase are identified in advance before the sudden rupture of the rock mass; the three-level threshold system combined with probability output provides a quantitative basis for engineering decision-making and changes the traditional "black or white" early warning mode. For example, when the critical failure probability is 0.6, it is recommended to take medium-strength reinforcement measures; the real-time performance of the algorithm and data traceability meet the requirements of engineering monitoring, and the calculation delay is controlled within 500 ms to ensure dynamic response to the changes in rock mass stability.

[0198] Generally speaking, the index of instability energy density comes from the real-time calculation results of S3, and the type of damage criterion is dynamically switched by S4 according to the deformation behavior to ensure that the evaluation model matches the current state of the rock mass; the determination result of the damage state is used as a prerequisite for the spatial focusing verification of the S6 microseismic event cluster. For example, when it is determined to be in a "dangerous" state, the spatio-temporal window for focusing calculation is automatically reduced (from 50 m × 24 h to 20 m × 12 h) to improve the early warning sensitivity; the determination result is used to reverse-check the parameter settings of S3 - S4. For example, if no actual instability is triggered after 3 consecutive early warnings, the system automatically adjusts the threshold system (such as increasing the early warning threshold from 0.3 to 0.4) to form an adaptive optimization mechanism.

[0199] S6. When the determination result of the damage state exceeds the stage threshold, trigger the spatial focusing verification of the microseismic event cluster. If the focusing degree of the microseismic event cluster is greater than the preset critical value within the preset spatio-temporal window, output an early warning instruction; otherwise, return to step S1 to collect data again.

[0200] In the embodiment of the present invention, the triggering of the spatial focusing verification of the microseismic event cluster includes:

[0201] Identify the microseismic event cluster within the preset spatio-temporal window and calculate the three-dimensional geometric centroid of the microseismic event cluster;

[0202] Traverse the Euclidean distance from each event in the microseismic event cluster to the three-dimensional geometric centroid;

[0203] Take the average value of the Euclidean distances as the focusing degree, and perform spatial focusing verification on the microseismic event cluster based on the focusing degree and the preset critical value.

[0204] Specifically, the step of outputting an early warning instruction if the focusing degree of the microseismic event cluster is greater than the preset critical value within the preset spatio-temporal window includes:

[0205] Compare the magnitude relationship between the focusing degree and the preset critical value, and generate an early warning signal when the focusing degree index continuously exceeds the preset critical value;

[0206] Encode the warning signal into an executable instruction format and drive the monitoring terminal to perform an alarm action.

[0207] In the embodiment of the present invention, the step of returning to step S1 to re-collect data includes:

[0208] When the spatial focusing verification fails, reset the acquisition parameters of the spatio-temporal synchronization array;

[0209] Re-execute the acquisition of microseismic signals, convergence displacements, and seepage pressure data of the geological body through the spatio-temporal synchronization array, and update the multi-source fusion data set to replace the historical data.

[0210] Specifically, a microseismic event cluster refers to a set of microseismic events with similar spatial positions and short time intervals generated by the expansion or rupture of internal fractures in a geological body within a preset spatio-temporal window, reflecting the spatial aggregation characteristics of rock mass damage.

[0211] Specifically, the spatial focusing verification refers to verifying the development state of the potential fracture surface of the rock mass by calculating the geometric distribution concentration degree of the microseismic event cluster, which is the key spatial criterion for judging the risk of surrounding rock instability.

[0212] Specifically, the three-dimensional geometric centroid refers to the mass center of the microseismic event cluster in three-dimensional space, which is obtained by weighted averaging of the coordinates of each event and is used to characterize the spatial position center of the event cluster.

[0213] Specifically, the Euclidean distance refers to the straight-line distance between two points in three-dimensional space, which is used to measure the spatial deviation degree between the microseismic event and the centroid and is the basic parameter for calculating the focusing degree.

[0214] Specifically, the focusing degree refers to the average value of the Euclidean distances from each event in the microseismic event cluster to the centroid. The smaller the value, the more concentrated the event distribution and the higher the risk of rock mass rupture.

[0215] Specifically, the preset critical value refers to the focusing degree threshold preset according to tunnel engineering experience and geological conditions, which is used to judge whether the event cluster forms a potential fracture surface and is the key index for triggering an alarm.

[0216] Furthermore, the damage state threshold comparison includes: obtaining the critical failure probability P in the damage state determination result from S5, and when P exceeds the stage threshold (such as 0.7), triggering the focusing verification process.

[0217] Furthermore, the threshold dynamic adjustment includes: reducing the stage threshold by 10%-20% in soft rock engineering and increasing it by 10%-15% in hard rock engineering, which is achieved through the lithology correction coefficient k3 (k3 = 0.8 for soft rock and k3 = 1.2 for hard rock).

[0218] Further, the time window is automatically adjusted according to the damage state, with the impact damage scenario set to 12 hours and the creep scenario set to 72 hours to ensure coverage of the development cycle of the event cluster; the spatial window is centered on the current monitoring section and extends 50 meters along the tunnel axis and radially to the boundary of the surrounding rock plastic zone (calculated by the strength reduction factor of S4, e.g., a reduction factor of 0.6 corresponds to a plastic zone radius of 5 meters).

[0219] Further, time correlation means setting the event interval threshold Tg = 10 minutes, and events with an interval less than Tg within the same spatial window are grouped into the same cluster; spatial correlation uses the DBSCAN density clustering algorithm, with the centroid as the core point, the neighborhood radius set to 10 meters, and the density threshold set to 3 events / cubic meter.

[0220] Specifically, the calculation of the three-dimensional geometric centroid includes: , , , where is the three-dimensional coordinate of the th microseismic event (obtained by locating the waveform arrival time difference after decoupling by S2, with a positioning accuracy of meters); is the event weight, taking the event magnitude ML (calculated from the microseismic signal amplitude, with the formula , where is the amplitude, is the period, is the site correction factor).

[0221] Specifically, the tunnel engineering coordinate system is uniformly adopted: the origin is the starting point of the tunnel entrance, the x-axis is along the tunnel axis, the y-axis is horizontal and perpendicular to the tunnel, and the z-axis is vertically upward. The coordinate benchmark is regularly calibrated by a total station (model Leica TS60).

[0222] Specifically, for each event in the event cluster, calculate the Euclidean distance to the centroid , and calculate the average distance as the focusing degree, with the unit of meters. The historical curve is updated and stored in real time for trend analysis.

[0223] Specifically, the preset critical value dynamic adjustment includes: the basic critical value is set to 10 meters and is corrected according to the following factors: burial depth correction and lithology correction. Among them, the burial depth correction includes: for every 100-meter increase in burial depth, the critical value decreases by 1 meter (calculated by , where H is the burial depth in meters); the lithology correction includes: the critical value for soft rock increases by 2 meters and decreases by 2 meters for hard rock (the preset critical value for soft rock is 10 + 2, and the preset critical value for hard rock is 10 - 2).

[0224] Further, the focusing degree continuous judgment logic means using a sliding time window (such as 3 consecutive calculation periods, each period being 1 hour) to judge whether F continuously exceeds : If all three cycles meet , it is determined as effective focusing and an early warning is triggered; if any cycle , return to S1 to collect data again.

[0225] Furthermore, the alarm action drive means sending it to the monitoring terminal through the Modbus protocol to trigger the following linkages: audible and visual alarm (sound pressure level ≥ 85 dB, red light flashing); SMS notification (sent to the mobile phone numbers of 3 preset responsible persons); the tunnel access control system automatically locks the entrance of the dangerous area.

[0226] Furthermore, when the focusing verification fails, the following steps are executed: the sampling frequency of the space-time synchronization array is increased from 1000 Hz to 2000 Hz (hard rock scenario) or decreased to 500 Hz (soft rock scenario), and the GPS clock is recalibrated (error ≤ 1 ms); the historical 12-hour fusion data set is cleared, and data is collected according to the new parameters to ensure that subsequent analysis is based on the latest signal.

[0227] Generally speaking, this step solves the problem in the prior art that a single energy index cannot locate potential fracture surfaces through a spatial focusing verification mechanism: by combining the energy index with the spatial distribution characteristics, false early warnings caused by the dispersion of event clusters are avoided. For example, when the energy index is high but the event distribution is dispersed, it is determined as a non-dangerous state; the space-time window and the critical value are dynamically adjusted to adapt to the fracture characteristics under different geological conditions, improving the accuracy of early warning spatial positioning and narrowing the early warning area to near the actual fracture surface; the continuous verification logic reduces the interference of accidental factors, such as excluding the misjudgment of event clusters caused by external seismic sources such as blasting construction, and reducing the false alarm rate.

[0228] Generally speaking, the damage state determination result comes from S5, which is the premise for triggering the focusing verification; the coordinates and magnitude data of microseismic events are derived from the decoupling and feature extraction results of S2; the early warning instruction drives engineering intervention measures, and at the same time, the verification result reversely optimizes the parameters of S1 - S5: if actual instability occurs after the early warning, the system automatically adds the current focusing degree and energy index to the historical case library for optimizing the threshold; if it returns to S1 to reset the acquisition parameters, the new data will update the energy calculation of S3 and the criterion switching logic of S4 to form an adaptive monitoring system.

[0229] As Figure 2 shown, it is the functional module diagram of the automatic monitoring and early warning system for tunnel surrounding rock stability provided by an embodiment of the present invention.

[0230] The automatic monitoring and early warning system 100 for the stability of tunnel surrounding rock according to the present invention can be installed in an electronic device. According to the functions achieved, the automatic monitoring and early warning system 100 for the stability of tunnel surrounding rock can include a data acquisition module 101, a correlation feature extraction module 102, an instability energy density index generation module 103, a damage criterion adoption module 104, a damage state determination module 105, and an early warning instruction generation module 106. The modules in the present invention can also be referred to as units, which refer to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, and are stored in the memory of the electronic device.

[0231] In this embodiment, the functions of each module / unit are as follows:

[0232] The data acquisition module 101 is used to collect microseismic signals, convergence displacements, and seepage pressure data of the geological body through a spatio-temporal synchronization array, and generate a multi-source fusion data set;

[0233] The correlation feature extraction module 102 is used to extract the correlation features between the microseismic event rate and the displacement rate based on the multi-source fusion data set;

[0234] The instability energy density index generation module 103 is used to generate the instability energy density index of the geological body based on the correlation features;

[0235] The damage criterion adoption module 104 is used to dynamically switch the damage criterion according to the deformation behavior of the geological body: adopt the impact damage criterion in the sudden deformation stage and the creep damage criterion in the progressive deformation stage;

[0236] The damage state determination module 105 is used to input the instability energy density index into the damage criterion corresponding to the deformation stage to generate the damage state determination result of the geological body;

[0237] The early warning instruction generation module 106 is used to trigger the spatial focusing verification of the microseismic event cluster when the damage state determination result exceeds the stage threshold. If the focusing degree of the microseismic event cluster within the preset spatio-temporal window is greater than the preset critical value, an early warning instruction is output; otherwise, return to step S1 to collect data again.

[0238] In several embodiments provided by the present invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are only illustrative. For example, the division of the modules is only a logical function division, and there can be other division methods in actual implementation.

[0239] The module described as a separation component may or may not be physically separated. The component shown as a module may or may not be a physical unit, that is, it may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0240] In addition, in each embodiment of the present invention, each functional module can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware, or in the form of hardware plus software functional modules.

[0241] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms.

[0242] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence is a theory, method, technology and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0243] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not restrictive. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An automatic monitoring and early warning method for the stability of tunnel surrounding rock, characterized in that, The method includes: S1. Collect microseismic signals, convergence displacements, and seepage pressure data of a geological body through a spatio-temporal synchronization array to generate a multi-source fusion dataset; S2. Extract the correlation features between the microseismic event rate and the displacement rate based on the multi-source fusion dataset; S3. Generate the instability energy density index of the geological body based on the correlation features; S4. Dynamically switch the damage criterion according to the deformation behavior of the geological body: adopt the impact damage criterion in the sudden deformation stage and the creep damage criterion in the progressive deformation stage; S5. Input the instability energy density index into the damage criterion corresponding to the deformation stage to generate the determination result of the damage state of the geological body; S6. When the determination result of the damage state exceeds the stage threshold, trigger the spatial focusing verification of the microseismic event cluster. If the focusing degree of the microseismic event cluster within a preset spatio-temporal window is greater than the preset critical value, output a warning instruction; otherwise, return to step S1 to collect data again.

2. The automatic monitoring and early warning method for tunnel surrounding rock stability according to claim 1, characterized in that The collection of microseismic signals, convergence displacements, and seepage pressure data of a geological body through a spatio-temporal synchronization array to generate a multi-source fusion dataset includes: Collect microseismic signals of the geological body based on a distributed sensor array; Measure the convergence displacement of the geological body based on a displacement monitoring device; Obtain the seepage pressure data of the geological body based on a seepage pressure monitoring device; Align the timestamps of the microseismic signals, the convergence displacements, and the seepage pressure data through a time synchronization protocol; Fuse the aligned microseismic signals, the convergence displacements, and the seepage pressure data to generate a multi-source fusion dataset.

3. The automatic monitoring and early warning method for the stability of tunnel surrounding rock according to claim 1, characterized in that, Before extracting the correlation features between the microseismic event rate and the displacement rate based on the multi-source fusion dataset, it further includes: Separate the dispersion effect of the multi-source fusion dataset based on a preset dynamic constitutive model, and the expression of the dynamic constitutive model is: Among them, is stress, is elastic modulus, is strain, is the viscosity coefficient characterizing the stress-strain relationship of the geological body, is the time identifier; Process the separated multi-source fusion dataset using a decoupling algorithm, where the decoupling algorithm is: wherein, is the signal frequency component, is the frequency-dependent time delay, is the complex transfer function, is the imaginary unit; Extract the microseismic event rate and the displacement rate from the decoupled multi-source fusion dataset.

4. The automatic monitoring and early warning method for tunnel surrounding rock stability according to claim 1, characterized in that, The extraction of the correlation features between the microseismic event rate and the displacement rate includes: Calculate the covariance between the microseismic event rate and the displacement rate for each sampling point; Accumulate the covariance to form a correlation feature calculation formula, and solve the correlation coefficient in the correlation feature calculation formula by the least squares method. The correlation coefficient represents the correlation feature of the coupling strength between the microseismic event rate and the displacement rate.

5. The automatic monitoring and early warning method for the stability of tunnel surrounding rock according to claim 1, characterized in that The generation of the instability energy density index of the geological body based on the correlation features includes: Substitute the correlation features into a preset energy calculation function to obtain the instability energy density index of the geological body, where the preset energy calculation function is: Among them, is the instability energy density index, is the correlation feature, is the microseismic event rate, is the displacement rate, is the time identifier.

6. The automatic monitoring and early warning method for tunnel surrounding rock stability according to claim 1, characterized in that, The dynamic switching of the damage criterion according to the deformation behavior of the geological body includes: Real-time monitor the deformation rate of the geological body; When the deformation rate mutation exceeds a preset threshold, trigger the impact damage criterion; When the deformation rate mutation continuously is lower than the preset threshold, trigger the creep damage criterion; Load the damage criterion parameters based on the triggered impact damage criterion or creep damage criterion.

7. The automatic monitoring and early warning method for tunnel surrounding rock stability according to claim 6, characterized in that, The loading of the damage criterion parameters based on the triggered impact damage criterion or creep damage criterion includes: Analyze the type of the currently activated damage criterion; Call the mechanical parameter library matching the type of damage criterion; Map the instability energy density index to the mechanical parameter library to generate damage criterion parameters including the strength reduction coefficient.

8. The automatic monitoring and early warning method for tunnel surrounding rock stability according to claim 1, characterized in that, The generation of the damage state determination result of the geological body includes: Input the instability energy density index into the currently activated damage criterion; Process the instability energy density index through the algorithm of the damage criterion to generate a stability evaluation value; Compare the stability evaluation value with a preset stability threshold and output the damage state determination result representing the critical failure probability of the rock mass.

9. The automatic monitoring and early warning method for tunnel surrounding rock stability according to claim 1, characterized in that, The triggering of the spatial focusing verification of the microseismic event cluster includes: Identify the microseismic event cluster within a preset spatio-temporal window and calculate the three-dimensional geometric centroid of the microseismic event cluster; Traverse the Euclidean distance from each event in the microseismic event cluster to the three-dimensional geometric centroid; Take the average value of the Euclidean distances as the focusing degree, and perform spatial focusing verification on the microseismic event cluster based on the focusing degree and a preset critical value.

10. An automatic monitoring and early warning system for the stability of tunnel surrounding rock, characterized in that, The system includes: A data acquisition module for collecting microseismic signals, convergence displacements, and seepage pressure data of the geological body through a spatio-temporal synchronization array to generate a multi-source fusion data set; An associated feature extraction module for extracting the associated features of the microseismic event rate and displacement rate based on the multi-source fusion data set; An instability energy density index generation module for generating the instability energy density index of the geological body based on the associated features; A damage criterion adoption module for dynamically switching the damage criterion according to the deformation behavior of the geological body: adopting an impact damage criterion in the sudden deformation stage and a creep damage criterion in the progressive deformation stage; A damage state determination module for inputting the instability energy density index into the damage criterion corresponding to the deformation stage to generate the damage state determination result of the geological body; An early warning instruction generation module for triggering the spatial focusing verification of the microseismic event cluster when the damage state determination result exceeds the stage threshold. If the focusing degree of the microseismic event cluster is greater than the preset critical value within the preset spatio-temporal window, output an early warning instruction, otherwise return to step S1 to collect data again.

Images