A method and system for automatic monitoring and early warning of tunnel surrounding rock stability

Abstract

The present invention relates to the field of geophysical exploration technology, and specifically discloses a method and system for automatic monitoring and early warning of tunnel surrounding rock stability. The method comprises: collecting surrounding rock microseismic signals, convergent displacement and seepage pressure data through a time-space synchronous array to generate a multi-source fusion data set; separating the dispersion effect based on a dynamic constitutive model, extracting the correlation characteristics between the microseismic event rate and the displacement rate; generating an instability energy density index; dynamically switching the impact or creep damage criterion according to the surrounding rock deformation behavior; inputting the index into the corresponding criterion to generate a damage status result; if the result exceeds the threshold, verifying the spatial focusing of the microseismic event cluster, and outputting an early warning instruction when the focusing degree exceeds the critical value. The present invention significantly improves the accuracy of tunnel surrounding rock instability early warning through multi-source fusion and dynamic criterion mechanism.

Description

Technical Field

[0001] The present invention relates to the field of geophysical exploration technology, and in particular to an automatic monitoring and early warning method and system for tunnel surrounding rock stability. 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, failing to incorporate the spatiotemporal correlations of multidimensional physical quantities such as seepage pressure and displacement. Interference from construction vibrations and equipment noise in tunnel environments distorts microseismic signals, making it difficult for a single indicator to accurately characterize the evolution of surrounding rock damage and prone to overlooking local instability risks.

[0004] Existing static damage criteria (such as fixed energy thresholds) are unable to adapt to the phased characteristics of surrounding rock, from gradual creep to sudden impact failure. Tunnel surrounding rock is subject to the combined influence of geological structure and construction disturbances, and its deformation mode changes dynamically. Rigid criteria are unable to respond in real time, resulting in poor early warning timeliness and a high false alarm rate. Summary of the Invention

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

[0006] To achieve the above-mentioned purpose, the present invention provides a method for automatic monitoring and early warning of tunnel surrounding rock stability, comprising:

[0007] S1. Collect microseismic signals, convergent displacement and seepage pressure data of geological bodies through time-space synchronous array to generate multi-source fusion data sets;

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

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

[0010] S4. Dynamically switching the damage criterion according to the deformation behavior of the geological body: using the impact damage criterion in the sudden deformation stage and the creep damage criterion in the gradual deformation stage;

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

[0012] S6. When the damage state judgment result exceeds the stage threshold, the spatial focusing verification of the microseismic event cluster is triggered. If the focusing degree of the microseismic event cluster in the preset space-time window is greater than the preset critical value, an early warning instruction is output, otherwise return to step S1 to re-collect data.

[0013] Optionally, the collecting of microseismic signals, convergence displacement and seepage pressure data of a geological body by a spatiotemporal synchronous array to generate a multi-source fusion data set includes:

[0014] Collect microseismic signals of geological bodies based on distributed sensor arrays;

[0015] measuring the convergence displacement of the geological body based on a displacement monitoring device;

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

[0017] aligning the timestamps of the microseismic signal, the converged displacement, and the seepage pressure data through a time synchronization protocol;

[0018] The aligned microseismic signals, the convergence displacement and the seepage pressure data are fused to generate a multi-source fusion data set.

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

[0020] The dispersion effect of the multi-source fusion data set is separated based on a preset dynamic constitutive model. The expression of the dynamic constitutive model is:

[0021]

[0022] in, is stress, is the elastic modulus, It's strain. is the viscosity coefficient that characterizes the stress-strain relationship of the geological body, It is a time marker;

[0023] Apply a decoupling algorithm to process the separated multi-source fusion data set, wherein the decoupling algorithm is:

[0024]

[0025] in, is the signal frequency component, is the frequency dependent delay, is the complex transfer function, is an imaginary unit;

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

[0027] Optionally, extracting correlation features between microseismic event rate and displacement rate includes:

[0028] The covariance between the microseismic event rate and the displacement rate is calculated for each sampling point;

[0029] The covariances are accumulated to form a correlation characteristic calculation formula, and a correlation coefficient in the correlation characteristic calculation formula is solved by the least square method, where the correlation coefficient represents the correlation characteristic 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 correlation feature includes:

[0031] 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:

[0032]

[0033] 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.

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

[0035] monitoring the deformation rate of the geological body in real time;

[0036] When the deformation rate suddenly exceeds a preset threshold, the impact damage criterion is triggered;

[0037] When the deformation rate suddenly changes and continues to be lower than a preset threshold, a creep damage criterion is triggered;

[0038] Damage criterion parameters are loaded based on the triggered impact damage criterion or the creep damage criterion.

[0039] Optionally, the trigger-based loading of damage criterion parameters for the impact damage criterion or the creep damage criterion includes:

[0040] Analyze the currently activated damage criterion type;

[0041] Calling a mechanical parameter library that matches the damage criterion type;

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

[0043] Optionally, generating a damage status determination result 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 damage criterion algorithm to generate a stability assessment value;

[0046] The stability assessment value is compared with a preset stability threshold, and a damage state determination result representing the critical failure probability of the rock mass is output.

[0047] Optionally, the verification of spatial focusing of the triggered microseismic event cluster includes:

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

[0049] Traversing the Euclidean distance between each event in the microseismic event cluster and the three-dimensional geometric centroid;

[0050] The average value of the Euclidean distance is used as the focusing degree, and the spatial focusing of the microseismic event cluster is verified based on the focusing degree and a preset critical value.

[0051] In order to solve the above problems, the present invention also provides an automatic monitoring and early warning system for tunnel surrounding rock stability, the system comprising:

[0052] The data acquisition module is used to collect microseismic signals, convergence displacement and seepage pressure data of geological bodies through a time-space synchronous array to generate a multi-source fusion data set;

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

[0054] an instability energy density index generating module, configured to generate an instability energy density index of the geological body based on the correlation feature;

[0055] A damage criterion adopting module is used to dynamically switch the damage criterion according to the deformation behavior of the geological body: the impact damage criterion is adopted in the sudden deformation stage, and the creep damage criterion is adopted in the gradual deformation stage;

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

[0057] The early warning instruction generation module is used to trigger the spatial focusing verification of the microseismic event cluster when the damage status judgment result exceeds the stage threshold. If the focusing degree of the microseismic event cluster in the preset space-time window is greater than the preset critical value, the early warning instruction is output; otherwise, the module returns to step S1 to re-collect data.

[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 spatiotemporal synchronized array. The dispersion effect is separated using a dynamic constitutive model to construct an interference-resistant multi-source fusion dataset. Sensor placement and decoupling algorithms are optimized for the narrow and long tunnel space to improve the accuracy of microseismic event location. The correlation between microseismic event rate and displacement rate is extracted based on covariance and the least squares method to quantify the coevolution of surrounding rock damage, overcoming the limitations of single parameters in complex tunnel environments.

[0060] 2. Based on the sudden change in the surrounding rock deformation rate, the impact / creep damage criterion is switched in real time, and a strength reduction factor is generated by linking the mechanical parameter library. This mechanism accurately distinguishes between sudden collapse and slow rheological risks induced by tunnel construction, avoiding misjudgments of static models. Furthermore, through the verification of the instability energy density index and the spatial focusing of microseismic event clusters (such as three-dimensional centroid distance analysis), the sensitivity of local tunnel instability warning is improved while ensuring a low false alarm rate, forming a closed-loop control logic. BRIEF DESCRIPTION OF THE DRAWINGS

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

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

[0063] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

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

[0065] The embodiment of the present application provides an automatic monitoring and early warning method for the stability of tunnel surrounding rocks. The execution subject of the automatic monitoring and early warning method for the stability of tunnel surrounding rocks includes but is not limited to at least one of the electronic devices such as a server and a terminal 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 rocks 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 it can be 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 networks (CDNs), and big data and artificial intelligence platforms.

[0066] Reference Figure 1 FIG. 1 is a flow chart of an automatic monitoring and early warning method for tunnel surrounding rock stability according to an embodiment of the present invention. In this embodiment, the automatic monitoring and early warning method for tunnel surrounding rock stability includes:

[0067] S1. Collect microseismic signals, convergent displacement and seepage pressure data of geological bodies through spatiotemporal synchronous arrays to generate multi-source fusion data sets.

[0068] In an embodiment of the present invention, the acquisition of microseismic signals, convergent displacement and seepage pressure data of a geological body by a spatiotemporal synchronous array to generate a multi-source fusion data set includes:

[0069] Collect microseismic signals of geological bodies based on distributed sensor arrays;

[0070] measuring the convergence displacement of the geological body based on a displacement monitoring device;

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

[0072] aligning the timestamps of the microseismic signal, the converged displacement, and the seepage pressure data through a time synchronization protocol;

[0073] The aligned microseismic signals, the convergence displacement and the seepage pressure data are fused to generate a multi-source fusion data set.

[0074] In detail, the spatiotemporal synchronization array refers to a three-dimensional monitoring network composed of a distributed sensor array, displacement monitoring equipment, and seepage pressure monitoring devices. The unified calibration of spatially distributed sensors and time references is achieved through a time synchronization protocol to ensure the spatiotemporal consistency of multi-source data.

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

[0076] In detail, displacement monitoring equipment refers to a device based on laser ranging or fiber optic sensing technology, which is 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 in the tunnel.

[0077] In detail, the seepage pressure monitoring device refers to a piezoresistive sensor installed in a borehole in the surrounding rock of a tunnel, which is used to monitor changes in groundwater pore pressure and reflect the permeability characteristics of the rock mass and the impact of water pressure on stability.

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

[0079] In detail, the multi-source fusion dataset refers to a structured data set containing multi-dimensional physical quantities formed by aligning microseismic signals, convergent displacement, and seepage pressure data according to a unified time axis, which is used for subsequent comprehensive analysis.

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

[0081] Furthermore, the convergence displacement measurement includes: arranging two groups of laser rangefinders (accuracy of ±0.01mm) in each monitoring section to monitor the tunnel vault subsidence and side wall convergence respectively, and the distance measurement baseline length is 1.2 times the tunnel diameter; the rangefinder automatically triggers the measurement, the sampling interval is 10 minutes, and the data is transmitted to the edge computing unit via the RS485 bus.

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

[0083] Furthermore, the piezometer was connected to a data acquisition instrument via a cable, with a sampling frequency of 1 time / hour, and the temperature compensation coefficient was automatically calibrated.

[0084] Furthermore, each monitoring device has a built-in GPS module to receive satellite clock signals (UTC time) and control the clock error within ±1ms 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 base.

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

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

[0087] In summary, this step, through the fusion of spatiotemporal synchronization arrays and multi-source data, addresses the existing problem of misjudgment of surrounding rock stability caused by single-parameter monitoring. Specifically, multi-source data collaboratively reflects the surrounding rock state, preventing missed judgments caused by noise interference from a single signal (such as a microseismic signal alone). Time synchronization accuracy ensures spatiotemporal consistency of the data, providing a reliable foundation for subsequent energy analysis and damage criterion switching. The fused data set encompasses multi-dimensional parameters including mechanics (microseismic), geometry (displacement), and hydraulics (seepage pressure), enhancing comprehensive monitoring.

[0088] In summary, the multi-source fusion dataset generated in S1 serves as the input foundation for subsequent steps: microseismic signals provide raw waveform data for microseismic event rate extraction in S2; converged displacement data is used to calculate displacement rates in S2; seepage pressure data assists in determining deformation behavior in S4, such as the potential for increased rock creep due to high seepage pressure; and unified timestamps ensure the accuracy of spatiotemporal window matching for microseismic event clusters in S6. This step achieves the transformation from physical signal acquisition to structured data, forming the data foundation for the entire monitoring and early warning process.

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

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

[0091] The dispersion effect of the multi-source fusion data set is separated based on a preset dynamic constitutive model. The expression of the dynamic constitutive model is:

[0092]

[0093] in, is stress, is the elastic modulus, It's strain. is the viscosity coefficient that characterizes the stress-strain relationship of the geological body, It is a time marker;

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

[0095]

[0096] in, is the signal frequency component, is the frequency dependent delay, is the complex transfer function, is an imaginary unit;

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

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

[0099]

[0100] in, is the number of sampling points, is the associated feature, is the average microseismic event rate of the geological body within the preset time and space window, is the average displacement rate of the geological body within the preset time and space window, is The microseismic event rate at the time, is The displacement rate at time, It is a time stamp.

[0101] In an embodiment of the present invention, extracting the correlation features between the microseismic event rate and the displacement rate includes:

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

[0103] The covariances are accumulated to form a correlation characteristic calculation formula, and a correlation coefficient in the correlation characteristic calculation formula is solved by the least square method, where the correlation coefficient represents the correlation characteristic of the coupling strength between the microseismic event rate and the displacement rate.

[0104] In detail, the dynamic constitutive model refers to a mathematical model used to describe the stress-strain relationship of a geological body, taking into account the coupling of elastic deformation and viscous deformation. Through this model, the dispersion effect in the microseismic signal can be separated to reflect the true mechanical state of the rock mass.

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

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

[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. It is used to correct the frequency-dependent delay caused by frequency dispersion so that the frequency components of the signal arrive synchronously.

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

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

[0110] Furthermore, the dynamic constitutive model parameters and their determination methods include: elastic modulus refers to the rock type determined according to the geological survey report (such as granite E = 60GPa, sandstone E = 30GPa), with a value range of 20-80GPa, and is calibrated through indoor triaxial compression tests; viscosity coefficient refers to the setting based on the rheological properties of the rock mass, and soft rock (such as mudstone) is set to 10 6 -10 8 Pa·s, for hard rock (such as granite), take 10 4 -10 6 Pa・s, fitted by creep test data; strain is calculated from the converged displacement data generated by S1, the formula is ,in is the displacement change (unit: mm), To monitor the baseline length (e.g. tunnel diameter 5m).

[0111] Furthermore, the model application and dispersion separation include: the stress σ (obtained from the inversion of microseismic signals), strain and strain rate Substitute into the dynamic constitutive model and separate the elastic component and viscosity component ; Perform frequency domain transform (FFT) on the separated microseismic signals to identify the phase offset characteristics of different frequency components.

[0112] Furthermore, the frequency-dependent time delay Δt(f) calibration includes: selecting two points with a known distance (e.g., 10 m) in the tunnel surrounding rock, exciting an artificial seismic source (e.g., an electric spark source), and collecting the arrival times of different frequency components; establishing and frequency The mapping relationship, for example: hour , hour , obtained by polynomial fitting ,in is the fitting coefficient.

[0113] Furthermore, the complex transfer function calculation and signal correction include: for each frequency component , calculate the complex transfer function ,in is an imaginary unit; by multiplying in the frequency domain Applied to dispersion signals, after phase correction, an inverse Fourier transform (IFFT) is performed to obtain a decoupled time domain signal with a waveform distortion rate controlled within 5%.

[0114] Furthermore, the calculation of microseismic event rate includes: setting a trigger threshold (3 times of the background noise, such as 200 μV) for the decoupled microseismic signal, using the STA / LTA algorithm (short time window / long time window ratio) to identify effective events; taking 1 hour as the time window, counting the number of events within the window, and obtaining The microseismic event rate at the moment (unit: times / h).

[0115] Furthermore, the displacement rate The calculation includes: based on the convergent displacement data of S1, taking the displacement difference of adjacent time points (1 day apart) ,calculate Day, the unit is mm / day, for example, The displacement at time is , The displacement at time is ,but sky.

[0116] Furthermore, the time-space window setting includes: selecting 24 hours of data, corresponding to n = 8640 sampling points (10ms / point) as the time window; taking the monitoring section as the center and extending 50m 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] In general, when the destabilizing energy density exceeds a critical value and the rate of increase suddenly changes, the impact damage criterion is triggered. If the energy density continues to be above 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 (S6 calculation) form a multi-dimensional verification. For example, when the destabilizing energy density index is greater than 0.8 J / m³·h and the event cluster focusing is less than 10m, the instability risk is doubly confirmed. The energy density calculation relies 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", ensuring the traceability of data at 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 adopt the creep damage criterion in the gradual deformation stage.

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

[0144] monitoring the deformation rate of the geological body in real time;

[0145] When the deformation rate suddenly exceeds a preset threshold, the impact damage criterion is triggered;

[0146] When the deformation rate suddenly changes and continues to be lower than a preset threshold, a creep damage criterion is triggered;

[0147] Damage criterion parameters are loaded based on the triggered impact damage criterion or the creep damage criterion.

[0148] In detail, the trigger-based loading of damage criterion parameters for the impact damage criterion or the creep damage criterion includes:

[0149] Analyze the currently activated damage criterion type;

[0150] Calling a mechanical parameter library that matches the damage criterion type;

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

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

[0153] In detail, dynamic switching of damage criteria refers to automatically switching the applicable damage assessment criteria according to the real-time deformation characteristics of the geological body, so that the criteria match the current failure mode of the rock mass and improve the accuracy of early warning.

[0154] In detail, the impact damage criterion refers to the criterion for the sudden deformation stage, which is applicable to the rapid rock fracture scenario, characterized by a sudden increase in microseismic event rate and energy density, and focuses on the instantaneous release of destructive energy.

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

[0156] Specifically, a sudden change in deformation rate refers to a significant change in the displacement rate of a geological body within a short period of time, exceeding the normal fluctuation range, which usually indicates a sudden instability of the internal structure of the rock mass.

[0157] In detail, the preset threshold refers to the critical value of deformation rate set in advance according to geological conditions and historical data, which is used to judge the type of deformation behavior and is divided into impact threshold and creep threshold.

[0158] In detail, the mechanical parameter library refers to a database that stores mechanical parameters corresponding to different damage criteria, including strength reduction coefficients, elastic modulus attenuation factors, etc., which are associated with parameters such as geological body lithology and burial depth.

[0159] In detail, the strength reduction factor refers to a dimensionless parameter used to quantify the attenuation of rock mass strength with the development of damage. The smaller the value, the greater the decrease in the bearing capacity of the rock mass and the higher the risk of instability.

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

[0161] Extract converged displacement data from the multi-source fusion data set generated by S1, and calculate the displacement rate using a sliding time window (e.g., 1 day). Perform a Kalman filter on the displacement rate sequence to eliminate monitoring noise interference. The filter parameters are set according to the rock type (the noise covariance Q for hard rock is 0.01, and the noise covariance Q for soft rock is 0.05). Calculate the time derivative of the deformation rate. , to characterize the degree of deformation acceleration, the five-point central difference method can be used: , the time step Δt = 1 hour to ensure real-time performance.

[0162] Furthermore, the preset threshold setting mechanism includes: impact damage threshold mechanism and creep damage threshold mechanism, wherein the impact damage threshold mechanism refers to 0.1mm / day² for hard rock and 0.5mm / day² for soft rock, which is determined based on indoor rock sample impact test. When this value is exceeded, the deformation rate is considered to have changed suddenly. The creep damage threshold is 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 judged to be gradual deformation.

[0163] Furthermore, the criterion trigger logic includes: impact damage criterion trigger and creep damage criterion trigger, wherein the impact damage criterion trigger means that when When the instability energy density index of S3 increases by more than 20% within 1 hour, the sudden damage warning logic is triggered; the creep damage criterion triggering means that when If the creep threshold is less than 48 hours and the microseismic event rate does not fluctuate significantly (the change is less than 10%), the progressive damage assessment process is activated.

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

[0165] Specifically, the mechanical parameter library uses key-value pairs, where 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 factor 0.6, elastic modulus decay rate 0.3 / day, energy release exponent 1.5}; Key: {mudstone + 200m + creep criterion}, Value: {strength reduction factor 0.8, elastic modulus decay rate 0.1 / day, energy release exponent 1.2}.

[0166] In detail, the instability energy density mapping includes:

[0167] A mapping function between the instability energy density index and the strength reduction coefficient is established using piecewise linear interpolation, including:

[0168] Impact criterion: When , the strength reduction factor is 0.8; when , with a coefficient of 0.5.

[0169] Creep criterion: When , the strength reduction factor is 0.9; when , with a coefficient of 0.6.

[0170] Real-time calculation of current The corresponding coefficients, such as , the interpolation result is a strength reduction factor of 0.65.

[0171] In general, this step solves the problem in the existing technology that fixed criteria cannot adapt to the differences in rock deformation stages by dynamically switching damage criteria: different criteria are used for sudden deformation and gradual deformation to avoid warning lags caused by the use of creep criteria in impact damage scenarios, or false alarms caused by misuse of impact criteria in gradual deformation; real-time mapping of strength reduction coefficient and energy density allows the criterion parameters to be dynamically adjusted with the degree of damage, thereby improving the timeliness and accuracy of surrounding rock stability assessment; combined with dual verification of deformation rate change rate and energy index, it reduces false triggering of criteria caused by fluctuations in a single parameter, such as eliminating sudden increases in deformation rate caused by external interference such as blasting construction.

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

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

[0174] In an embodiment of the present invention, generating the damage status determination result of the geological body includes:

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

[0176] Processing the instability energy density index through the damage criterion algorithm to generate a stability assessment value;

[0177] The stability assessment value is compared with a preset stability threshold, and a damage state determination result representing the critical failure probability of the rock mass is output.

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

[0179] In detail, the deformation stage refers to the different evolutionary states of the geological body under stress, including the sudden deformation stage (such as impact failure) and the gradual deformation stage (such as creep failure), which is dynamically determined by S4 based on the deformation rate.

[0180] In detail, damage criterion refers to the evaluation standard established based on rock mechanics theory, which is divided into impact damage criterion and creep damage criterion. It is dynamically switched according to the deformation stage and is used to convert energy indicators into stability evaluation values.

[0181] In detail, the stability assessment value refers to the quantitative value obtained after processing the instability energy density index through the damage criterion algorithm, which represents the current stability of the geological body. The smaller the value, the worse the stability.

[0182] In detail, the preset stability threshold refers to the critical value set in advance based on geological conditions and engineering experience, which is used to compare the stability assessment value and classify the rock damage status level (such as safe, warning, and dangerous).

[0183] In detail, the damage state determination result refers to the critical failure possibility of the rock mass output in the form of probability, which is generated by combining the comparison result of the stability assessment value with the threshold value to provide a quantitative basis for engineering decision-making.

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

[0185] The instability energy density index is retrieved from the real-time calculation results of S3. The data format is a floating-point value with a timestamp (unit: J / m³・h). The interface uses the REST API protocol and a data validity verification mechanism is set: if the instability energy density index is negative or exceeds the theoretical maximum value (such as 10J / m³・h), a data anomaly alarm is triggered and the average value of the previous hour is automatically retrieved as a replacement.

[0186] Furthermore, the deformation stage association includes: synchronously reading the currently activated damage criterion type (impact or creep) output by S4, and associating it through key-value pairs and criterion type to ensure that the energy index matches 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] In detail, the evaluation value calculation model of the impact damage criterion algorithm is: ,in, is the impact state stability assessment value, ranging from [0,1], the smaller the value, the closer to instability; is the impact criterion threshold, which is called from the mechanical parameter library of S4 (e.g. 1.0 J / m³·h for hard rock); is the lithology correction factor (granite ,sandstone ), automatically matched from the parameter library.

[0188] In detail, the evaluation value calculation model of the creep damage criterion algorithm is ,in, is the impact state stability evaluation value, 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 in 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.​​​​​​​​​​​​​​​​​​​​​​​​​​​In general, this step solves the problem in existing technologies that fixed criteria cannot adapt to differences in rock failure modes through the coupling analysis of dynamic criteria and energy indicators: different evaluation models are used for sudden deformation and gradual deformation to avoid early warning lags caused by the use of creep criteria in impact scenarios, for example, identifying energy surge characteristics in advance before sudden rock rupture; the three-level threshold system is combined with probability output to provide a quantitative basis for engineering decision-making, changing the traditional "black or white" early warning mode. For example, when the critical failure probability is 0.6, medium-strength reinforcement measures can be recommended; the real-time algorithm and data traceability meet engineering monitoring needs, and the calculation delay is controlled within 500ms to ensure dynamic response to changes in rock stability.

[0198] In general, the instability energy density index comes from the real-time calculation results of S3, and the damage criterion type is dynamically switched by S4 according to the deformation behavior to ensure that the assessment model matches the current state of the rock mass; the damage state judgment result serves as a prerequisite for the spatial focusing verification of the S6 microseismic event cluster. For example, when it is judged to be in a "dangerous" state, the spatiotemporal window of the focusing calculation is automatically narrowed (from 50m×24h to 20m×12h) to improve the warning sensitivity; the judgment result reversely verifies the parameter settings of S3-S4. For example, if three consecutive warnings do not trigger actual instability, the system automatically adjusts the threshold system (such as increasing the warning threshold from 0.3 to 0.4), forming an adaptive optimization mechanism.

[0199] S6. When the damage state judgment result exceeds the stage threshold, the spatial focusing verification of the microseismic event cluster is triggered. If the focusing degree of the microseismic event cluster in the preset space-time window is greater than the preset critical value, an early warning instruction is output, otherwise return to step S1 to re-collect data.

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

[0201] Identifying a microseismic event cluster within the preset space-time window and calculating the three-dimensional geometric centroid of the microseismic event cluster;

[0202] Traversing the Euclidean distance between each event in the microseismic event cluster and the three-dimensional geometric centroid;

[0203] The average value of the Euclidean distance is used as the focusing degree, and the spatial focusing of the microseismic event cluster is verified based on the focusing degree and a preset critical value.

[0204] Specifically, if the focus of the microseismic event cluster within a preset spatiotemporal window is greater than a preset critical value, outputting a warning instruction includes:

[0205] comparing the focus index with a preset critical value, and generating a warning signal when the focus 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, returning to step S1 to recollect data includes:

[0208] When the spatial focusing verification fails, resetting the acquisition parameters of the spatiotemporal synchronization array;

[0209] The acquisition of microseismic signals, converged displacement and seepage pressure data of the geological body by the spatiotemporal synchronous array is re-executed, and the multi-source fusion data set is updated to replace the historical data.

[0210] In detail, a microseismic event cluster refers to a group of microseismic events with similar spatial positions and short time intervals caused by the expansion or rupture of cracks inside a geological body within a preset time and space window, reflecting the spatial aggregation characteristics of rock damage.

[0211] In detail, spatial focusing verification refers to verifying the development state of the potential fracture surface of the rock mass by calculating the geometric distribution concentration of microseismic event clusters. It is a key spatial criterion for judging the instability risk of the surrounding rock mass.

[0212] In detail, the three-dimensional geometric centroid refers to the mass center of the microseismic event cluster in three-dimensional space, which is calculated by the weighted average of the coordinates of each event and is used to represent the spatial position center of the event cluster.

[0213] In detail, the Euclidean distance refers to the straight-line distance between two points in three-dimensional space. It is used to measure the spatial deviation of a microseismic event from the center of mass and is the basic parameter for calculating the focusing degree.

[0214] Specifically, the focality refers to the average value of the Euclidean distance from each event to the centroid in a microseismic event cluster. A smaller value indicates a more concentrated event distribution and a higher risk of rock mass failure.

[0215] In detail, the preset critical value refers to the focus threshold value set in advance based on tunnel engineering experience and geological conditions. It is used to determine whether the event cluster forms a potential rupture surface and is a key indicator for triggering early warning.

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

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

[0218] Furthermore, 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 event cluster development cycle; the spatial window is centered on the current monitoring section, extends 50 meters along the tunnel axis, and radially expands to the boundary of the surrounding rock plastic zone (calculated by the strength reduction coefficient of S4, such as a reduction coefficient of 0.6 corresponds to a plastic zone radius of 5 meters).

[0219] Furthermore, temporal association refers to setting an event interval threshold Tg = 10 minutes, and events with an interval less than Tg in the same spatial window are classified into the same cluster; spatial association 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] In detail, the 3D geometric centroid calculation includes: , , ,in, For the The three-dimensional coordinates of the microseismic event (obtained by waveform to time difference positioning after S2 decoupling, positioning accuracy rice); is the event weight, and the event magnitude ML (calculated by the amplitude of the microseismic signal, the formula is ,in is the amplitude, For the cycle, is the site correction factor).

[0221] In detail, a unified tunnel engineering coordinate system is 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 vertical upward. The coordinate reference is regularly calibrated by a total station (model Leica TS60).

[0222] In detail, for each event in the event cluster, calculate the centroid The Euclidean distance is calculated and the average distance is used as the focusing degree in 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 modified according to the following factors: burial depth correction and lithology correction, wherein the burial depth correction includes: for every 100 meters increase in burial depth, the critical value is reduced by 1 meter (through Calculation, where H is the burial depth / meter); lithology corrections include increasing the critical value for soft rock by 2 meters and decreasing it for hard rock by 2 meters (the preset critical value for soft rock is 10+2, and the preset critical value for hard rock is 10−2).

[0224] Furthermore, the focus degree continuous judgment logic refers to using a sliding time window (such as 3 consecutive calculation cycles, each cycle is 1 hour) to judge whether F continues to exceed :If all three cycles satisfy , it is determined to be effective focus, triggering an early warning; if any cycle , return to S1 to collect data again.

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

[0226] Furthermore, if the focusing verification fails, the following steps are performed: the sampling frequency of the spatiotemporal synchronous array is increased from 1000 Hz to 2000 Hz (hard rock scenario) or reduced to 500 Hz (soft rock scenario), and the GPS clock is recalibrated (error ≤ 1 ms); the historical 12-hour fused dataset is cleared, and data is collected according to the new parameters to ensure that subsequent analysis is based on the latest signals.

[0227] In general, this step solves the problem in the existing technology that a single energy indicator cannot locate the potential rupture surface through a spatial focus verification mechanism: combining the energy indicator with the spatial distribution characteristics to avoid false warnings caused by the dispersion of event clusters. For example, when the energy indicator is high but the event distribution is dispersed, it is judged to be a non-dangerous state; dynamically adjust the spatiotemporal window and critical value to adapt to the rupture characteristics under different geological conditions, improve the spatial positioning accuracy of the warning, and reduce the warning area to the vicinity of the actual rupture surface; continuously verify the logic to reduce interference from accidental factors, such as eliminating the misjudgment of event clusters caused by external earthquake sources such as blasting construction, and reduce the false alarm rate.

[0228] In general, the damage status determination result comes from S5, which serves as a prerequisite for triggering focus verification; the microseismic event coordinates and magnitude data are derived from the decoupling and feature extraction results of S2; the early warning instructions drive engineering intervention measures, and the verification results reversely optimize the S1-S5 parameters: if actual instability occurs after the early warning, the system automatically adds the current focus and energy indicators to the historical case library for threshold optimization; if the acquisition parameters are reset by returning to S1, the new data will update the energy calculation of S3 and the judgment switching logic of S4, forming an adaptive monitoring system.

[0229] like Figure 2 , which is a functional module diagram of an 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 tunnel surrounding rock stability described in the present invention can be installed in an electronic device. Depending on the functionality implemented, the system 100 may 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. A module, also referred to as a unit, is a series of computer program segments that can be executed by an electronic device processor and perform a fixed function. These modules are stored in the electronic device's memory.

[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 displacement and seepage pressure data of geological bodies through a time-space synchronous array to generate a multi-source fusion data set;

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

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

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

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

[0237] The warning instruction generation module 106 is used to trigger the spatial focusing verification of the microseismic event cluster when the damage status judgment result exceeds the stage threshold. If the focusing degree of the microseismic event cluster within the preset spatiotemporal window is greater than a preset critical value, a warning instruction is output; otherwise, the module returns to step S1 to re-collect data.

[0238] In the 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 merely illustrative. For example, the module division is merely a logical function division, and other division methods may be used in actual implementation.

[0239] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed across multiple network elements. Some or all of the modules may be selected to achieve the purpose of the solution of this embodiment according to actual needs.

[0240] In addition, the functional modules in various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or hardware plus software functional modules.

[0241] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0242] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to achieve optimal 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 limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for automatic monitoring and early warning of tunnel surrounding rock stability, characterized in that: The method comprises: S1. Collect microseismic signals, convergent displacement and seepage pressure data of geological bodies through time-space synchronous array to generate multi-source fusion data sets; S2. Extracting correlation features between microseismic event rate and displacement rate based on the multi-source fusion data set, including: calculating the covariance of the microseismic event rate and the displacement rate for each sampling point, accumulating the covariance to form a correlation feature calculation formula, solving the correlation coefficient in the correlation feature calculation formula by the least squares method, wherein the correlation coefficient represents the correlation feature of the coupling strength between the microseismic event rate and the displacement rate. Before extracting the correlation features between the microseismic event rate and the displacement rate based on the multi-source fusion data set, the method further includes: The dispersion effect of the multi-source fusion data set is separated based on a preset dynamic constitutive model. The expression of the dynamic constitutive model is: in, is stress, is the elastic modulus, It's strain. is the viscosity coefficient that characterizes the stress-strain relationship of the geological body, It is a time marker; Apply a decoupling algorithm to process the separated multi-source fusion data set, wherein the decoupling algorithm is: in, is the signal frequency component, is the frequency dependent delay, is the complex transfer function, is an imaginary unit; extracting microseismic event rates and displacement rates from the decoupled multi-source fusion dataset; S3. Generating an instability energy density index of the geological body based on the correlation feature, including: substituting the correlation feature into a preset energy calculation function to obtain the instability energy density index of the geological body, wherein the preset energy calculation function is: ,in, is the instability energy density index, is the associated feature, is the microseismic event rate, is the displacement rate, It is a time marker; S4. Dynamically switching the damage criterion according to the deformation behavior of the geological body: using the impact damage criterion in the sudden deformation stage and the creep damage criterion in the gradual deformation stage; S5, inputting the instability energy density index into a damage criterion corresponding to the deformation stage to generate a damage state determination result of the geological body; S6. When the damage state judgment result exceeds the stage threshold, the spatial focusing verification of the microseismic event cluster is triggered. If the focusing degree of the microseismic event cluster in the preset space-time window is greater than the preset critical value, an early warning instruction is output, otherwise return to step S1 to re-collect data. The spatial focusing verification of the microseismic event cluster is triggered, including: identifying the microseismic event cluster in the preset space-time window, and calculating the three-dimensional geometric centroid of the microseismic event cluster, traversing the Euclidean distance from each event in the microseismic event cluster to the three-dimensional geometric centroid, taking the average value of the Euclidean distance as the focusing degree, and performing spatial focusing verification on the microseismic event cluster based on the focusing degree and the preset critical value.

2. The automatic monitoring and early warning method for tunnel surrounding rock stability according to claim 1, characterized in that: The microseismic signals, convergent displacement and seepage pressure data of the geological body are collected by the time-space synchronous array to generate a multi-source fusion data set, including: Collect microseismic signals of geological bodies based on distributed sensor arrays; measuring the convergence displacement of the geological body based on a displacement monitoring device; Acquiring seepage pressure data of the geological body based on a seepage pressure monitoring device; aligning the timestamps of the microseismic signal, the converged displacement, and the seepage pressure data through a time synchronization protocol; The aligned microseismic signals, the convergence displacement and the seepage pressure data are fused to generate a multi-source fusion data set.

3. The automatic monitoring and early warning method for tunnel surrounding rock stability according to claim 1, characterized in that: The dynamically switching damage criterion according to the deformation behavior of the geological body includes: monitoring the deformation rate of the geological body in real time; When the deformation rate suddenly exceeds a preset threshold, the impact damage criterion is triggered; When the deformation rate suddenly changes and continues to be lower than a preset threshold, a creep damage criterion is triggered; Damage criterion parameters are loaded based on the triggered impact damage criterion or the creep damage criterion.

4. The automatic monitoring and early warning method for tunnel surrounding rock stability according to claim 3, characterized in that: The trigger-based loading of the impact damage criterion or the creep damage criterion with damage criterion parameters includes: Analyze the currently activated damage criterion type; Calling a mechanical parameter library that matches the damage criterion type; The instability energy density index is mapped to the mechanical parameter library to generate damage criterion parameters including a strength reduction coefficient.

5. The automatic monitoring and early warning method for tunnel surrounding rock stability according to claim 1, characterized in that: The generating of the damage status determination result of the geological body includes: Inputting the instability energy density index into the currently activated damage criterion; Processing the instability energy density index through the damage criterion algorithm to generate a stability assessment value; The stability assessment value is compared with a preset stability threshold, and a damage state determination result representing the critical failure probability of the rock mass is output.

6. An automatic monitoring and early warning system for tunnel surrounding rock stability, used to implement the automatic monitoring and early warning method for tunnel surrounding rock stability according to any one of claims 1 to 5, characterized in that: The system comprises: The data acquisition module is used to collect microseismic signals, convergence displacement and seepage pressure data of geological bodies through a time-space synchronous array to generate a multi-source fusion data set; A correlation feature extraction module is configured to extract correlation features between the microseismic event rate and the displacement rate based on the multi-source fusion data set, including: calculating the covariance of the microseismic event rate and the displacement rate for each sampling point, accumulating the covariance to form a correlation feature calculation formula, and solving the correlation coefficient in the correlation feature calculation formula by the least squares method, wherein the correlation coefficient represents the correlation feature of the coupling strength between the microseismic event rate and the displacement rate. Before extracting the correlation features between the microseismic event rate and the displacement rate based on the multi-source fusion data set, the module further includes: The dispersion effect of the multi-source fusion data set is separated based on a preset dynamic constitutive model. The expression of the dynamic constitutive model is: in, is stress, is the elastic modulus, It's strain. is the viscosity coefficient that characterizes the stress-strain relationship of the geological body, It is a time marker; Apply a decoupling algorithm to process the separated multi-source fusion data set, wherein the decoupling algorithm is: in, is the signal frequency component, is the frequency dependent delay, is the complex transfer function, is an imaginary unit; extracting microseismic event rates and displacement rates from the decoupled multi-source fusion dataset; The instability energy density index generation module is used to generate the instability energy density index of the geological body based on the correlation feature, including: substituting the correlation feature into a preset energy calculation function to obtain the instability energy density index of the geological body, wherein the preset energy calculation function is: ,in, is the instability energy density index, is the associated feature, is the microseismic event rate, is the displacement rate, It is a time marker; A damage criterion adopting module is used to dynamically switch the damage criterion according to the deformation behavior of the geological body: the impact damage criterion is adopted in the sudden deformation stage, and the creep damage criterion is adopted in the gradual deformation stage; A damage state determination module, configured to input the instability energy density index into a damage criterion corresponding to the deformation stage to generate a damage state determination result of the geological body; The early warning instruction generation module is used to trigger the spatial focusing verification of the microseismic event cluster when the damage status judgment result exceeds the stage threshold. If the focusing degree of the microseismic event cluster in the preset space-time window is greater than the preset critical value, the early warning instruction is output; otherwise, the module returns to step S1 to re-collect data.

Images