Tunnel water inrush early warning method and device based on microseismic and resistivity fusion

Abstract

The application discloses a tunnel water inrush early warning method and device based on microseism and resistivity fusion. The method comprises the following steps: synchronously arranging a microseism and resistivity monitoring array, collecting microseismic waveform and apparent resistivity data; processing to obtain microseismic space-time distribution, energy release and rupture type, and resistivity space-time evolution image; establishing a space-time coupling model fusing two types of parameters, extracting rock-water correlation characteristics, calculating a comprehensive risk index and grading early warning. The device comprises a comprehensive monitoring array, a data acquisition and transmission unit and a data processing center, and the data processing center is provided with a microseismic processing module, a resistivity inversion module, a fusion modeling module and an early warning module. The application realizes full-chain three-dimensional monitoring and intelligent early warning of water inrush disasters, and improves accuracy and advance nature.

Description

Technical Field

[0001] This invention relates to the field of tunnel and underground engineering safety monitoring technology, and in particular to a method and corresponding device for early warning of water inrush and sudden water disasters during tunnel construction by combining microseismic monitoring technology in geophysical exploration with resistivity method technology. Background Technology

[0002] When tunnels traverse water-rich and complex strata, water inrush hazards pose a significant safety risk. The mechanism typically involves two key processes: first, the rock mass fractures and expands under the influence of construction disturbance and water pressure (mechanical process); second, groundwater accumulates and migrates along the fractured channels until it breaches the free face (hydrological process). Existing early warning technologies mostly focus on monitoring a single process.

[0003] 1. Microseismic monitoring technology: By capturing the elastic waves generated by rock mass fractures, it can accurately locate fracture events and reflect the process of rock mass damage accumulation. However, it has poor direct sensitivity to water, making it difficult to distinguish between tectonic fractures and hydraulically driven fractures, and it cannot directly detect the existence and scale of water-bearing bodies.

[0004] 2. Resistivity method: By measuring changes in formation resistivity, it is exceptionally sensitive to water-bearing bodies (low-resistivity bodies) and can characterize their spatial distribution. However, it has a weak ability to directly reflect rock mass fracturing activity, making it difficult to determine whether low-resistivity bodies are in an active migration state, and it is easily affected by interference from construction metal components, etc.

[0005] In summary, the main drawback of existing technologies lies in their limited information dimensions. Neither microseismic nor resistivity methods can independently and completely capture the entire chain of precursory information from "rock mass fracturing initiation" to "waterway formation" and then to "water outburst." This results in early warning systems either relying too heavily on mechanical criteria (potentially leading to false alarms) or too heavily on hydrological criteria (potentially leading to missed alarms or delayed warnings), resulting in insufficient accuracy and predictability.

[0006] Therefore, there is an urgent need for an innovative method and device that can integrate mechanical and hydrological information, reveal the chain reaction process of disasters, and achieve earlier and more accurate early warning. Summary of the Invention

[0007] In view of the above-mentioned problems in the existing technology, the technical problem to be solved by the present invention is: how to overcome the shortcomings of existing single monitoring technology in that it cannot fully capture the entire chain of early warning information of water inrush disasters from "rock fracture" to "water outburst", resulting in insufficient accuracy and foresight in early warning.

[0008] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention achieves spatial collaborative monitoring by simultaneously deploying microseismic and resistivity monitoring arrays in S1; simultaneously acquiring monitoring data in S2 to ensure a unified time series reference; analyzing microseismic data to obtain the spatiotemporal distribution, energy release, and fracture type of rock mass fracturing; conducting three-dimensional temporal inversion of resistivity to characterize the dynamic evolution law of the aquifer; constructing a spatiotemporal coupling model to quantify the correlation characteristics between rock mass fracturing and water transport; and calculating a comprehensive risk index and implementing graded early warning, thereby achieving accurate and advanced early warning of tunnel water inrush disasters.

[0009] This invention provides a method and device for monitoring and early warning of tunnel water inrush disasters based on the fusion of microseismic and resistivity methods. This method achieves three-dimensional perception and intelligent early warning of the co-evolution process of rock mass damage and water migration through synchronous monitoring, information fusion, and coupled modeling.

[0010] A method for monitoring and early warning of tunnel water inrush disasters based on the fusion of microseismic and resistivity methods includes the following steps:

[0011] S1: Simultaneously deploy microseismic monitoring arrays and resistivity monitoring arrays in the area behind and in front of the tunnel face during construction.

[0012] S2: Synchronously acquire and transmit the microseismic event waveform data obtained by the microseismic monitoring array, and the apparent resistivity data obtained by the resistivity monitoring array.

[0013] S3: Process the waveform data of the microseismic events, including event identification, P-wave / S-wave arrival time picking, source location, source parameter calculation and source mechanism inversion, to obtain the spatiotemporal distribution, energy release characteristics and rupture type of the microseismic events, and then extract the microseismic activity parameters based on the spatiotemporal distribution, energy release characteristics and rupture type of the microseismic events.

[0014] S4: Perform time-series three-dimensional inversion on the apparent resistivity data, including generating a resistivity spatiotemporal evolution image, dynamically depicting the spatial distribution, morphology and migration characteristics of water bodies within the monitoring area, and extracting resistivity anomaly parameters based on the resistivity spatiotemporal evolution image.

[0015] S5: Establish a spatiotemporal coupling model between microseismic activity parameters and resistivity anomaly parameters. This spatiotemporal coupling model takes microseismic activity parameters, resistivity anomaly parameters and a unified three-dimensional space-time reference as inputs. Through spatial superposition, temporal cross-correlation and intensity correlation analysis, it extracts the spatiotemporal correlation characteristics of "rock-water" and finally outputs the spatiotemporal correlation characteristics, coupling factor C and comprehensive risk index RISK.

[0016] The results of S3 and S4 are superimposed and correlated in a unified three-dimensional spatiotemporal grid to analyze the spatiotemporal response law of rock mass fracture and aquifer evolution, and to construct a coupled model characterizing the intensity of "rock-water interaction".

[0017] The microseismic activity parameters include at least one of the following: microseismic event rate, cumulative released energy, event spatial concentration, b-value, and the proportion of tensile rupture reflected by the source mechanism; the resistivity anomaly parameters include at least one of the following: volume of the low-resistivity anomaly zone, resistivity decrease rate, distance between the low-resistivity anomaly zone and the working face, and spatial overlap between the low-resistivity anomaly zone and the microseismic event concentration zone.

[0018] S6: Based on the output of the spatiotemporal coupling model, calculate the comprehensive risk index of the inrush water disaster and conduct graded early warning based on the preset risk threshold.

[0019] Four risk thresholds are preset, and the real-time RISK level is compared with the thresholds to determine the warning level:

[0020] Blue indicates normal, RISK≤0.25;

[0021] Yellow indicates caution; 0.25 < RISK ≤ 0.50.

[0022] Orange indicates a warning; 0.50 < RISK ≤ 0.75.

[0023] Red indicates an alarm, RISK > 0.75, and triggers the corresponding on-site response plan. For example, when the model shows a highly overlapping area of ​​dense microseismic activity and a newly formed low-resistivity area in front of the tunnel face, and both areas show increased activity simultaneously, the comprehensive risk index will rise rapidly, triggering a high-level warning.

[0024] As an improvement, the steps in S3 for obtaining the spatiotemporal distribution, energy release characteristics, and rupture type of microseismic events are as follows:

[0025] S3-1: Spatiotemporal distribution of microseismic events:

[0026] S3-1-1: Event Recognition. For continuously acquired microseismic waveform data, the short-term mean ratio (STA / LTA) method is used to automatically identify valid microseismic events. When the ratio of the short-term mean to the long-term mean exceeds a preset threshold, it is determined to be a microseismic event, and a segment of the event waveform is extracted for subsequent processing.

[0027] S3-1-2: P-wave / S-wave arrival time acquisition. Automatically calibrates the arrival times of the P-wave (longitudinal wave) and S-wave (transverse wave) on each event waveform segment. STA / LTA combined with the AIC criterion or polarization analysis methods can be used to provide accurate arrival time data for positioning purposes.

[0028] S3-1-3: Source location. This involves using location algorithms based on uniform or layered velocity models (such as the Geiger method) to locate microseismic events. The arrival times of the P-wave and S-wave are picked up and combined with the medium velocity model (uniform or layered model). The Geiger iterative location method is used to calculate the three-dimensional spatial coordinates (x, y, z) and the time of origin t of the event. The three-dimensional spatial coordinates (x, y, z) of the source and the time of origin t are combined to obtain the location (x, y, z, t) of each microseismic event. Combining all the microseismic event locations yields the spatiotemporal distribution of the microseismic events.

[0029] S3-2: Energy release characteristics of microseismic events:

[0030] S3-2-1: Source parameter calculation: This includes calculating the energy E and moment magnitude Mw of the microseismic event, and analyzing the waveform amplitude and spectrum of the event waveform segment corresponding to the microseismic event to obtain the energy of the microseismic event.

[0031] Moment magnitude calculation:

[0032] Mw=(2 / 3)lgM0-6.0

[0033] M0 = μ·A·D;

[0034] Where μ represents the shear modulus of the surrounding rock, A represents the fracture surface area, and D represents the average dislocation amount on the fracture surface;

[0035] The energy of microseismic events is calculated using two methods:

[0036] Method 1: Based on moment magnitude lgE=1.5Mw+4.8;

[0037] Method 2: Based on the integral of the corrected displacement waveform E=(8 / 5)πρv s R 2

[0038] Where ρ represents the density of the surrounding rock, kg / m³; v s The S-wave velocity is represented in m / s; R represents the distance from the source to the sensor in m; u_corr(t) represents the particle displacement time history after instrument, path, and site correction in m; t represents the effective duration of the waveform in s.

[0039] Energy calculation from waveform amplitude: After noise and propagation correction of the micro-vibration waveform, the relative energy is obtained by integral of the square of the amplitude, and then the absolute energy is calculated by combining the medium parameters.

[0040] Energy can be calculated from the spectrum: the displacement spectrum is obtained by performing a Fourier transform on the corrected waveform, the source spectrum is inverted and the zero-frequency amplitude is extracted, the seismic moment is calculated and then the energy is converted; or the radiated energy can be obtained by integrating the spectrum and combining it with Parseval's theorem.

[0041] S3-2-2: Energy release characteristics of microseismic events are obtained through microseismic event energy and moment magnitude. The moment magnitude Mw of a microseismic event has a logarithmic linear quantitative relationship with the radiated energy E, satisfying lgE=1.5Mw+4.8; the energy release characteristics are comprehensively characterized by the moment magnitude, the magnitude of the radiated energy, the cumulative release rate, spatial concentration, and temporal distribution law. The larger the moment magnitude, the more exponentially the radiated energy of the corresponding event increases, and the more intense the energy release.

[0042] S3-3: Fracturing Types of Microseismic Events: Moment tensor inversion is performed using P-wave initial polarity and amplitude ratio information. The moment tensor inversion steps are: waveform preprocessing → Green's function calculation → moment tensor least squares solution → tensor decomposition and accuracy verification → focal mechanism parameter output; the inversion results include moment tensor components, the proportion of dual-couple DC / isotropic ISO / compensated linear CLVD components, and the focal mechanism solution; rupture types are classified according to component proportions: DC ≥ 70% is shear rupture, ISO ≥ 60% is tensile rupture, and DC and ISO between 30% and 70% are complex ruptures.

[0043] Among them, the ratio of tensile fracturing is a key indicator for judging the degree of hydraulic fracturing activity.

[0044] As an improvement, the process of generating a resistivity spatiotemporal evolution image in S4 to dynamically depict the spatial distribution, morphology, and migration characteristics of water bodies within the monitoring area is as follows:

[0045] S41: Using the resistivity monitoring array detection range set up in step S1 as the boundary, a unified inversion grid model is constructed by combining the three-dimensional spatial coordinates of the tunnel.

[0046] S42: For the microseismic event waveform data acquired in step S2, noise removal, terrain correction, and electrode arrangement error correction are performed sequentially.

[0047] S43: The least squares smoothing constraint inversion algorithm is used to carry out three-dimensional resistivity inversion time-by-time to obtain the true three-dimensional resistivity distribution data of the formation at each time.

[0048] S44: Spatiotemporal registration and fusion of multi-time series three-dimensional resistivity data are performed to generate a resistivity spatiotemporal evolution image, which dynamically depicts the spatial distribution, morphology and migration characteristics of low resistivity anomaly regions.

[0049] As an improvement, step S5, the process of establishing the spatiotemporal coupling model, includes:

[0050] S51: Extract microseismic activity parameters based on the spatiotemporal distribution, energy release characteristics, and rupture type of microseismic events obtained in step S3; extract resistivity anomaly parameters based on the spatiotemporal evolution image of resistivity obtained in step S4; and compare and analyze the spatiotemporal evolution patterns of dense microseismic event areas and low resistivity anomaly areas under a unified three-dimensional space-time coordinate system.

[0051] S52: Based on the aforementioned microseismic activity parameters and resistivity anomaly parameters, the spatiotemporal correlation characteristics of the two are extracted through quantitative calculation and time series analysis, including:

[0052] Spatial overlap: defined as the three-dimensional spatial overlap rate between the microseismic event cluster area and the low resistivity anomaly area, extracted by calculating the ratio of the intersection volume of the two types of areas to the volume of the microseismic event cluster area;

[0053] Time series response relationship: By performing lag cross-correlation analysis on microseismic activity time series data and resistivity anomaly time series data, the time lag of microseismic activity peak and resistivity anomaly abrupt change point is extracted.

[0054] Correlation of activity intensity: The Pearson correlation coefficient was used to calculate the linear correlation between the microseismic activity intensity parameters and the resistivity anomaly intensity parameters. The correlation coefficient ranged from [-1, 1], and the larger the absolute value, the stronger the correlation.

[0055] S53: Based on the aforementioned spatiotemporal correlation characteristics, the calculation function for the comprehensive risk index RISK, used to quantify the intensity of precursors to sudden water inrush, is as follows:

[0056] RISK = f(α·MSI,β·ERI,γ·C)

[0057] Wherein, MSI is the microseismic risk factor calculated based on microseismic activity parameters, ERI is the resistivity risk factor calculated based on resistivity anomaly parameters, C is the coupling factor characterizing the spatial overlap between the microseismic event area and the resistivity anomaly area, and α, β, and γ are the weight coefficients of each factor, satisfying α+β+γ=1.

[0058] As an improvement, in step S1, the resistivity monitoring array is a high-density electrical resistivity array or a transpore resistivity CT array; the sensor deployment range of the microseismic monitoring array covers the detection area of ​​the resistivity monitoring array.

[0059] An early warning device for implementing the above method, the tunnel inrush water disaster monitoring and early warning device comprising:

[0060] The microseismic monitoring array deployed in the surrounding rock of the tunnel includes a microseismic sensor array and a resistivity electrode array;

[0061] The data synchronization acquisition unit is connected to the integrated monitoring array and is used to synchronously trigger and acquire waveform data and apparent resistivity data of microseismic events;

[0062] The data transmission unit is used to transmit the data collected by the data synchronization acquisition unit to the data processing center;

[0063] The data processing center includes:

[0064] The microseismic data processing module is used to process the waveform data of microseismic events to obtain the spatiotemporal distribution, energy release characteristics, and rupture type of the microseismic events.

[0065] The resistivity data inversion module is used to perform time-series three-dimensional inversion on the apparent resistivity data to generate a resistivity spatiotemporal evolution image.

[0066] The multi-source information fusion and coupling modeling module is used to establish a spatiotemporal coupling model;

[0067] The integrated early warning module is used to perform risk calculation and graded early warning based on the output of the spatiotemporal coupling model. The early warning levels of the integrated early warning module include at least: normal (blue), attention (yellow), warning (orange), and alarm (red), with different levels corresponding to different on-site response plans.

[0068] As an improvement, the microseismic sensor array includes a three-component accelerometer or velocity meter arranged in the tunnel wall and borehole; the resistivity electrode array includes multiple sets of electrodes arranged in the tunnel wall, the ground surface and in the advance borehole.

[0069] As an improvement, the data synchronization acquisition unit includes a GPS or BeiDou satellite synchronization clock module to ensure that the acquisition time of microseismic event waveform data and apparent resistivity data is synchronized.

[0070] As an improvement, the early warning device also includes a visual early warning platform for three-dimensional dynamic display:

[0071] Microseismic event distribution cloud map generated based on the spatiotemporal distribution results of microseismic events; resistivity inversion slice map generated based on the resistivity spatiotemporal evolution image; spatiotemporal evolution map of comprehensive risk index generated based on the spatiotemporal coupling model output and the comprehensive risk index of step S6; and early warning information output by the graded early warning results.

[0072] Compared with the prior art, the present invention has at least the following beneficial effects:

[0073] 1. Achieved “full-chain” monitoring of the disaster process: Deeply integrated microseismic information reflecting “force” with resistivity information reflecting “water”, and for the first time achieved continuous and three-dimensional tracking of the entire process of water inrush disaster from its incubation (rock mass damage) to its occurrence (waterway connection) at the technical level, breaking through the limitations of a single monitoring dimension.

[0074] 2. Significantly improves the accuracy and predictability of early warnings: By using a coupled model, interference from single methods (such as non-water-burst-related tectonic microseisms and inactive static aquifers) is eliminated, focusing on the core precursor of water inrush: "rock-water interaction." This allows for earlier identification of "synergistic anomalies" caused by hydraulic fracturing and the formation of dominant water-conducting channels, significantly advancing the warning window before water inrush occurs.

[0075] 3. Provides three-dimensional spatial early warning capability: It can not only warn "when" the risk is high, but also accurately point out "where" (which direction, how far from the working face) the risk is the highest, as well as the "scale" of the risk source, providing a precise decision-making basis for formulating accurate treatment plans (such as target grouting).

[0076] 4. High degree of intelligence and automation: From data collection and processing to fusion analysis and early warning output, the entire process is automated and intelligent, which greatly reduces the reliance on human expert experience, improves the objectivity, stability and efficiency of early warning, and adapts to the long-term continuous monitoring needs of complex tunnel construction environments. Attached Figure Description

[0077] Figure 1 This is a simplified flowchart of the method of the present invention.

[0078] Figure 2 This is a block diagram of the functional modules of the data processing center of this invention.

[0079] Figure 3 This is a schematic diagram illustrating the spatiotemporal coupling analysis of microseismic events and resistivity anomalies in an exemplary scenario for applying the present invention. (a) Time T1, (b) Time T2.

[0080] Figure 4 This is an example curve showing the evolution of the comprehensive risk index over time, along with a schematic diagram of the corresponding warning levels.

[0081] Figure 5 This is a schematic diagram of the monitoring system layout of the present invention. Detailed Implementation

[0082] The present invention will now be described in further detail.

[0083] Example: Figure 1 and Figure 5As shown, the method of this invention is implemented in tunnel 1 under construction. A spatial array consisting of multiple three-component microseismic sensors 3 is deployed in the excavated section of the tunnel wall behind the tunnel face and in the advanced geological boreholes. Simultaneously, a high-density resistivity electrode array 4 is deployed in the tunnel wall, the ground surface, and the advanced boreholes. The spatial arrangement of the two types of arrays has been optimized to ensure an effective joint monitoring area is formed within a range of at least 50-100 meters in front of the tunnel face.

[0084] All three-component microseismic sensors 3 and density resistivity electrode array 4 are connected to the data transmission unit (5) located inside the tunnel. The data synchronization acquisition unit has a built-in high-precision BeiDou / GPS synchronization clock to ensure that each microseismic waveform data and each set of resistivity measurement data has a unified and accurate timestamp. The acquired data is transmitted in real time to the data processing center 6 via industrial Ethernet or wireless relay.

[0085] like Figure 2 As shown, the data processing center 6 is the "brain" of this invention, consisting of four core software modules:

[0086] Microseismic data processing module 61: This module processes microseismic event waveform data to obtain the spatiotemporal distribution, energy release characteristics, and rupture type of the microseismic events. It automatically detects microseismic events using the STA / LTA algorithm, locates the events using location algorithms based on uniform or layered velocity models (such as the Geiger method), calculates parameters such as energy and frequency band of the events, and analyzes the focal mechanism using the moment tensor inversion method.

[0087] Resistivity data inversion module 62: used to perform time-series three-dimensional inversion on the apparent resistivity data to generate a resistivity spatiotemporal evolution image; using a three-dimensional resistivity inversion algorithm based on finite element or finite difference (such as least squares smoothing constraint inversion) to invert hundreds of thousands of sets of apparent resistivity data collected daily into a three-dimensional distribution model of real resistivity, generating resistivity volume data.

[0088] Multi-source information fusion and coupled modeling module 63: This module reads the microseismic event spatiotemporal intensity and focal mechanism catalog output by module 61, and the time-series resistivity volume data output by module 62. It maps the two types of data into the same three-dimensional spatial grid.

[0089] The spatiotemporal coupling model constructed in this module takes as input: microseismic activity parameters such as microseismic event rate, cumulative released energy, event spatial clustering, b-value, and tensile rupture ratio; resistivity anomaly parameters such as low-resistivity anomaly zone volume, resistivity decrease rate, distance between low-resistivity anomaly zone and working face, and spatial overlap; and a unified three-dimensional spatiotemporal reference.

[0090] The model outputs include: the spatial correlation index between dense microseismic areas and low-resistivity anomaly areas, the cross-correlation results of time-series responses, and the coupling factor C, providing core coupling parameters for the calculation of the comprehensive risk index.

[0091] The module's built-in algorithm performs the following key operations:

[0092] 1. Spatial correlation analysis: Calculate the three-dimensional spatial correlation index between the spatial density field of microseismic events and the spatial distribution of resistivity anomalies (areas below the background value by a certain proportion) within each time window.

[0093] The three-dimensional spatial correlation index is calculated using a coupled calculation method of the three-dimensional mesh Pearson correlation coefficient and volume overlap, as shown in the following formula:

[0094] SCI=α·Υ spatial +(1−α)·O V

[0095] In the formula: SCI is the three-dimensional spatial correlation index; α is the weighting coefficient, with a value of 0.4 to 0.6;

[0096] Υ spatial Pearson correlation coefficient between microseismic energy density and resistivity anomaly coefficient in a 3D grid:

[0097] Υ spatial =

[0098] M represents the total number of three-dimensional mesh elements; Let J be the energy density of the microseismic event in the j-th grid. This represents the average microseismic energy density. Let J be the resistivity anomaly coefficient of the j-th grid. This is the average resistivity anomaly coefficient;

[0099] O V The volume overlap between the microseismic anomaly region and the low resistivity anomaly region:

[0100] O V

[0101] V(·) represents the volume of the three-dimensional region; This is a microseismic anomaly zone; This is the region of low resistivity and abnormal resistance.

[0102] 2. Time-series response analysis: In the identified key anomaly areas (such as the direction of potential water inrush channels in front of the tunnel face), extract the time series of microseismic event rates (especially the proportion of tensile events) and the time series of average resistivity values ​​in the area, and perform cross-correlation analysis to determine whether there is a hysteresis response mode of "active microseismic events → decreased resistivity".

[0103] 3. Coupling Factor Calculation: Based on the above analysis, calculate the coupling factor C as described in claim 4. For example, C can be defined as "the reciprocal of the distance between the centroid of the microseismic event cloud and the geometric center of the low-resistivity anomaly zone," with a larger value for C as the distance is closer.

[0104] 4. Time-series response analysis: In the identified key anomaly areas (such as the direction of potential water inrush channels in front of the tunnel face), extract the time series of microseismic event rates (especially the proportion of tensile events) and the time series of average resistivity values ​​within these areas, and perform cross-correlation analysis: First, extract and preprocess two sets of time-series data: microseismic cumulative energy and resistivity anomaly coefficient; set a time lag τ, calculate the cross-correlation coefficients under different τ values, and obtain the optimal lag time and the maximum correlation coefficient. Determine whether there exists a lag response pattern of "microseismic activity → resistivity decrease" (optimal lag time τ > 0 (microseismic advance), maximum correlation coefficient is negative (microseismic activity accompanied by resistivity decrease), and absolute value of correlation coefficient ≥ 0.4).

[0105] Coupling factor calculation: Based on the above analysis, the coupling factor C is calculated. For example, C can be defined as "the reciprocal of the distance between the centroid of the microseismic event cloud and the geometric center of the low-resistivity anomaly zone". The closer the distance, the larger the value of C.

[0106] Coupling factor C combines the three-dimensional spatial correlation index (SCI) and the time-series maximum cross-correlation coefficient (R). max The strength of coupling is characterized by the following calculation steps:

[0107] ① Normalize the three-dimensional spatial correlation index (SCI) to the 0-1 range;

[0108] ② Extract the effective correlation coefficient |R from the time series analysis max |;

[0109] ③According to C=βSCI+(1-β)∣R max | Calculate the coupling factor, where β is the spatial weighting coefficient, ranging from 0.5 to 0.7;

[0110] ④C takes values ​​from 0 to 1. The larger the value, the higher the degree of coupling between microseismic activity and resistivity anomaly.

[0111] Integrated Early Warning Module 64: This module pre-defines the calculation formulas and weighting coefficients (α, β, γ) for risk factors (MSI, ERI). MSI can be calculated based on the microseismic event rate, cumulative energy, and the degree of b-value decrease; ERI can be calculated based on the volume growth rate of low-resistivity anomaly zones, the rate of resistivity decrease, and their distance from the tunnel face. The module calculates the comprehensive risk index RISK in real time and, based on preset thresholds (such as...),... Figure 4(As shown) Issue warnings: Normal (I, Blue), Caution (II, Yellow, Strengthen Observation), Warning (III, Orange, Prepare for Treatment), Alert (IV, Red, Immediately Stop Work and Evacuate People).

[0112] A specific early warning scenario (such as) Figure 3 As shown in the figure): Taking the Youyang Karst Tunnel of the Chongqing-Xiangtan Expressway as the actual monitoring object, eight microseismic sensors were deployed in the tunnel, arranged in three dimensions on the arch of the surrounding rock 0-50m in front of the tunnel face, the left and right sidewalls and the invert arch; 48 resistivity monitoring electrodes were deployed, symmetrically arranged longitudinally along the left and right sidewalls of the tunnel, with a longitudinal spacing of 5m and a lateral spacing of 2m, and the detection range covered 80m in front of the tunnel face.

[0113] At time T1, monitoring showed that there was a relatively stable low-resistivity zone (9, old aquifer) in front of the tunnel face, with sporadic historical microseismic events around it.

[0114] As the tunnel face progressed, at time T2, the coupled modeling module processed the monitoring data: The STA / LTA algorithm was used to identify microseismic events, the Geiger method to locate the seismic source, and moment tensor inversion to determine the rupture type. A new microseismic event cluster 12 was identified between the stable low-resistivity zone 9 and the tunnel face, with tensile rupture being the dominant type. After correcting the resistivity data, a three-dimensional time-series inversion was performed to obtain a low-resistivity channel 12 extending from the stable low-resistivity zone to the newly formed microseismic zone. Through three-dimensional spatial correlation and time-series cross-correlation calculations, the three-dimensional spatial correlation index SCI = 0.76 and the maximum cross-correlation coefficient |R0 were obtained. max With a weighting factor of β=0.67 and a weighting coefficient of β=0.6, the coupling factor C=0.72 is calculated. Simultaneously, resistivity inversion shows that a new low-resistivity channel is extending from the stable low-resistivity region 9 towards the newly formed microseismic region, with the two spatially overlapping significantly.

[0115] At this point, the MSI (due to newly formed dense tensile microseismic activity) and ERI (due to newly formed low-resistivity channels) increased, and the C value (due to high spatial overlap) reached its maximum. The comprehensive risk index RISK in early warning module 64 rose sharply. This tunnel adopts a four-level real-time graded early warning system: Level I no risk (C < 0.3), Level II low risk (0.3 < C < 0.5), Level III medium risk (0.5 < C < 0.7), and Level IV high risk (C ≥ 0.7). This scenario triggered a Level IV high-risk alarm, quickly jumping from "attention" (Level II) to "early warning" (Level III) and even "alarm" (Level IV), gaining valuable emergency response time for the site.

[0116] All analysis results, including 3D microseismic event cloud maps, resistivity slices, spatiotemporal evolution cloud maps of risk indices, and early warning signals, are displayed in real time on the visualization and early warning platform 7 for decision-makers to use.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A tunnel inrush water early warning method based on the fusion of microseismic and resistivity data, characterized in that, Includes the following steps: S1: Simultaneously deploy microseismic monitoring arrays and resistivity monitoring arrays in the area behind and in front of the tunnel face during construction. S2: Synchronously acquire and transmit the microseismic event waveform data obtained by the microseismic monitoring array, and the apparent resistivity data obtained by the resistivity monitoring array; S3: Process the waveform data of the microseismic events, including event identification, P-wave / S-wave arrival time picking, source location, source parameter calculation and source mechanism inversion, to obtain the spatiotemporal distribution, energy release characteristics and rupture type of the microseismic events, and then extract the microseismic activity parameters based on the spatiotemporal distribution, energy release characteristics and rupture type of the microseismic events. S4: Perform time-series three-dimensional inversion on the apparent resistivity data, including generating a resistivity spatiotemporal evolution image, dynamically depicting the spatial distribution, morphology and migration characteristics of water bodies within the monitoring area, and extracting resistivity anomaly parameters based on the resistivity spatiotemporal evolution image; S5: Establish a spatiotemporal coupling model between microseismic activity parameters and resistivity anomaly parameters. This spatiotemporal coupling model takes microseismic activity parameters, resistivity anomaly parameters and a unified three-dimensional space-time reference as inputs. Through spatial superposition, temporal cross-correlation and intensity correlation analysis, it extracts the spatiotemporal correlation characteristics of "rock-water" and finally outputs the spatiotemporal correlation characteristics, coupling factor C and comprehensive risk index RISK. S6: Based on the output of the spatiotemporal coupling model, calculate the comprehensive risk index of the inrush water disaster, and conduct graded early warning based on the preset risk threshold; Four risk thresholds are preset, and the real-time RISK level is compared with the thresholds to determine the warning level: Blue indicates normal, RISK≤0.25; Yellow indicates caution; 0.25 < RISK ≤ 0.

50. Orange indicates a warning; 0.50 < RISK ≤ 0.

75. Red indicates an alarm, RISK > 0.75, and triggers the corresponding on-site response plan.

2. The tunnel inrush water early warning method based on the fusion of microseismic and resistivity data according to claim 1, characterized in that, The steps in S3 to obtain the spatiotemporal distribution, energy release characteristics, and rupture type of microseismic events are as follows: S3-1: Spatiotemporal distribution of microseismic events S3-1-1: Event Recognition. For continuously acquired microseismic waveform data, the short-term mean ratio (STA / LTA) method is used to automatically identify valid microseismic events. When the ratio of the short-term mean to the long-term mean exceeds a preset threshold, it is determined to be a microseismic event, and a segment of the event waveform is extracted for subsequent processing. S3-1-2: P-wave / S-wave arrival time acquisition. Automatically calibrates the arrival times of the P-wave and S-wave on each event waveform segment. STA / LTA combined with the AIC criterion or polarization analysis method can be used to provide accurate arrival time data for positioning purposes. S3-1-3: Source location, using a location algorithm based on a uniform or layered velocity model to locate microseismic events. Utilizing the acquired P-wave and S-wave arrival time information, combined with a medium velocity model (uniform or layered model); the Geiger iterative location method is used to calculate the three-dimensional spatial coordinates (x, y, z) and the time of origin t of the event; the three-dimensional spatial coordinates (x, y, z) of the source and the time of origin t are combined to obtain the location (x, y, z, t) of each microseismic event, and the spatiotemporal distribution of all microseismic event locations is obtained by combining the locations of all microseismic events. S3-2: Energy release characteristics of microseismic events S3-2-1: Source parameter calculation: including calculating the energy E and moment magnitude Mw of the microseismic event, and analyzing the waveform amplitude and spectrum of the event waveform segment corresponding to the microseismic event to obtain the energy of the microseismic event; in: Moment magnitude calculation: Mw=(2 / 3)lgM0-6.0 M0 = μ·A·D; Where μ represents the shear modulus of the surrounding rock, A represents the fracture surface area, and D represents the average dislocation amount on the fracture surface; The energy of microseismic events is calculated using two methods: Method 1: Based on moment magnitude lgE=1.5Mw+4.8; Method 2: Based on the integral of the corrected displacement waveform E=(8 / 5)πρv s R 2 Where ρ represents the density of the surrounding rock, kg / m³; v s The S-wave velocity is represented in m / s; R represents the distance from the source to the sensor in m; u_corr(t) represents the particle displacement time history after instrument, path, and site correction in m; t represents the effective duration of the waveform in s. S3-2-2: The energy release characteristics of microseismic events are obtained through the energy and moment magnitude of microseismic events; the moment magnitude Mw of a microseismic event has a logarithmic linear quantitative relationship with the radiated energy E, satisfying lgE=1.5Mw+4.8; the energy release characteristics are comprehensively characterized by the moment magnitude, the magnitude of the radiated energy, the cumulative release rate, the spatial clustering, and the temporal distribution law. The larger the moment magnitude, the more exponentially the radiated energy of the corresponding event increases, and the more intense the energy release. S3-3: Rupture Types of Microseismic Events Moment tensor inversion is performed using the initial polarity and amplitude ratio information of the P-wave. The moment tensor inversion steps are as follows: waveform preprocessing → Green's function calculation → moment tensor least squares solution → tensor decomposition and accuracy verification → focal mechanism parameter output; the inversion results include moment tensor components, the proportion of dual-couple DC / isotropic ISO / compensated linear CLVD components, and focal mechanism solution; the rupture type is classified according to the component proportion: DC ≥ 70% is shear rupture, ISO ≥ 60% is tensile rupture, and DC and ISO between 30% and 70% is composite rupture.

3. The tunnel inrush water early warning method based on microseismic and resistivity fusion according to claim 1, characterized in that, The process of generating a resistivity spatiotemporal evolution image in S4, dynamically depicting the spatial distribution, morphology, and migration characteristics of water bodies within the monitoring area, is as follows: S41: Using the resistivity monitoring array detection range set up in step S1 as the boundary, a unified inversion grid model is constructed by combining the three-dimensional spatial coordinates of the tunnel. S42: For the microseismic event waveform data acquired in step S2, noise removal, terrain correction, and electrode arrangement error correction are performed sequentially. S43: The least squares smoothing constraint inversion algorithm is used to carry out three-dimensional resistivity inversion time-by-time to obtain the true three-dimensional resistivity distribution data of the formation at each time. S44: Spatiotemporal registration and fusion of multi-time series three-dimensional resistivity data are performed to generate a resistivity spatiotemporal evolution image, which dynamically depicts the spatial distribution, morphology and migration characteristics of low resistivity anomaly regions.

4. The tunnel inrush water early warning method based on microseismic and resistivity fusion according to claim 1, characterized in that, Step S5, the process of establishing the spatiotemporal coupling model includes: S51: Extract microseismic activity parameters based on the spatiotemporal distribution, energy release characteristics, and rupture type of microseismic events obtained in step S3; extract resistivity anomaly parameters based on the spatiotemporal evolution image of resistivity obtained in step S4; and compare and analyze the spatiotemporal evolution patterns of dense microseismic event areas and low resistivity anomaly areas under a unified three-dimensional space-time coordinate system. S52: Based on the aforementioned microseismic activity parameters and resistivity anomaly parameters, the spatiotemporal correlation characteristics of the two are extracted through quantitative calculation and time series analysis, including: Spatial overlap: defined as the three-dimensional spatial overlap rate between the microseismic event cluster area and the low resistivity anomaly area, extracted by calculating the ratio of the intersection volume of the two types of areas to the volume of the microseismic event cluster area; Time series response relationship: By performing lag cross-correlation analysis on microseismic activity time series data and resistivity anomaly time series data, the time lag of microseismic activity peak and resistivity anomaly abrupt change point is extracted. Correlation of activity intensity: The Pearson correlation coefficient was used to calculate the linear correlation between the microseismic activity intensity parameters and the resistivity anomaly intensity parameters. The correlation coefficient ranged from [-1, 1], and the larger the absolute value, the stronger the correlation. S53: Based on the aforementioned spatiotemporal correlation characteristics, the calculation function for the comprehensive risk index RISK, used to quantify the intensity of precursors to sudden water inrush, is as follows: RISK = f(α·MSI,β·ERI,γ·C) Wherein, MSI is the microseismic risk factor calculated based on microseismic activity parameters, ERI is the resistivity risk factor calculated based on resistivity anomaly parameters, C is the coupling factor characterizing the spatial overlap between the microseismic event area and the resistivity anomaly area, and α, β, and γ are the weight coefficients of each factor, satisfying α+β+γ=1.

5. The tunnel inrush water early warning method based on microseismic and resistivity fusion according to claim 1, characterized in that, In step S1, the resistivity monitoring array is a high-density electrical resistivity array or a transpore resistivity CT array; the sensor deployment range of the microseismic monitoring array covers the detection area of ​​the resistivity monitoring array.

6. An early warning device for implementing the tunnel inrush water early warning method based on the fusion of microseismic data and resistivity as described in any one of claims 1-5, characterized in that, The tunnel flood inrush disaster monitoring and early warning device includes: The microseismic monitoring array deployed in the surrounding rock of the tunnel includes a microseismic sensor array and a resistivity electrode array; The data synchronization acquisition unit is connected to the integrated monitoring array and is used to synchronously trigger and acquire waveform data and apparent resistivity data of microseismic events; The data transmission unit is used to transmit the data collected by the data synchronization acquisition unit to the data processing center; The data processing center includes: The microseismic data processing module is used to process the waveform data of microseismic events to obtain the spatiotemporal distribution, energy release characteristics, and rupture type of the microseismic events. The resistivity data inversion module is used to perform time-series three-dimensional inversion on the apparent resistivity data to generate a resistivity spatiotemporal evolution image. The multi-source information fusion and coupling modeling module is used to establish a spatiotemporal coupling model; The integrated early warning module is used to perform risk calculation and graded early warning based on the output of the spatiotemporal coupling model.

7. The apparatus according to claim 6, characterized in that, The microseismic sensor array includes a three-component accelerometer or velocity meter arranged in the tunnel wall and borehole; the resistivity electrode array includes multiple sets of electrodes arranged in the tunnel wall, on the ground surface and in the advance borehole.

8. The apparatus according to claim 7, characterized in that, The data synchronization acquisition unit includes a GPS or BeiDou satellite synchronization clock module to ensure that the acquisition time of microseismic event waveform data and apparent resistivity data is synchronized.

Images