Micro-seismic discrimination method and system for vibration and excavation unloading fracture of deep-buried tunnel TBM (Tunnel Boring Machine)

Abstract

The invention provides a deep-buried tunnel TBM vibration and excavation unloading fracture micro-seismic discrimination method and system, and relates to the technical field of tunnel construction, and the method comprises the steps: obtaining micro-seismic event data in a target area; spatial classification is carried out according to the micro-seismic event data, and the influence range of the excavation unloading effect is obtained; performing region division according to the influence range to obtain a rock fracture signal after noise reduction; performing multi-layer decomposition on the rock fracture signal by using wavelet transform to obtain a characteristic parameter set; performing energy grading according to the characteristic parameter set to obtain a grading result; constructing a discrimination standard according to the grading result to obtain the discrimination standard; and performing discrimination based on the discrimination standard to obtain a discrimination result. According to the method, the dynamic model is constructed based on the multi-energy-level discrimination standard, main control factor classification is performed on the real-time micro-seismic event, real-time dynamic discrimination of TBM vibration and excavation unloading fracture in the construction process is achieved, and the early warning timeliness of time-delay rock burst is remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of tunnel construction technology, and more specifically, to a microseismic discrimination method and system for TBM vibration and excavation unloading fracture in deep-buried tunnels. Background Technology

[0002] Rockburst is a dynamic instability phenomenon in deep-buried tunnel engineering, caused by the combined effects of excavation unloading and dynamic disturbance. It manifests as a sudden and violent release of elastic strain energy accumulated within high-energy-storage rock, leading to rock bursting and accompanying dynamic damage. In deep-buried TBM tunnel construction in plate-active regions, the surrounding rock is simultaneously subjected to two types of mechanical forces: firstly, the stress field redistribution caused by excavation unloading; and secondly, the vibrational wave disturbance generated by the TBM's cutting tool. Under these influences, the rock mass in the potential rockburst zone undergoes progressive damage evolution, with micro-fractures continuously initiating, expanding, and fracturing within the rock mass, ultimately leading to rockburst. While existing microseismic monitoring technology can capture fracture signals throughout the entire cycle, it cannot effectively distinguish the controlling factors of the fracture—that is, it is difficult to determine whether the fracture event originates from excavation unloading or is induced by TBM vibration.

[0003] Existing patented technologies, such as the TBM construction simulation system proposed in CN118190479A, can only restore the redistribution of static stress field and do not consider the disturbance effect of vibration waves, resulting in deviations between the test results and the actual dynamic response of the project. Although CN119294144B introduces neural networks to predict rockburst, its model is based on the quasi-static unloading assumption, ignores the influence of TBM vibration stress waves on the evolution of micro-fractures, and lacks real-time correlation analysis of micro-seismic parameters, making it difficult to accurately identify the causes of fracture.

[0004] Based on the shortcomings of the existing technologies, there is an urgent need for a microseismic discrimination method and system for the vibration and excavation unloading rupture of deep-buried tunnel TBMs. Summary of the Invention

[0005] The purpose of this invention is to provide a microseismic discrimination method and system for vibration and excavation unloading fracture of deep-buried tunnel TBMs, so as to improve the above-mentioned problems. To achieve the above objective, the technical solution adopted by this invention is as follows: Firstly, this application provides a microseismic discrimination method for vibration and excavation unloading fracture of a deep-buried tunnel TBM, including: Acquire microseismic event data within the target area; Spatial classification is performed based on the microseismic event data to obtain the influence range of the excavation unloading effect. The affected area is divided into regions, and rock fracture waveforms, microseismic parameters and stress parameters of each region are collected simultaneously. The collected rock fracture signals are filtered and the noise-reduced rock fracture signals are extracted. The rock fracture signal was decomposed into multiple layers using wavelet transform. The frequency band energy ratio parameter, detail coefficient and approximation coefficient generated after each wavelet decomposition were collected, and the entropy value after each decomposition was calculated to obtain the set of characteristic parameters. Energy classification is performed based on the set of characteristic parameters to obtain classification results, which include low-energy events, medium-energy events, and high-energy events. Based on the classification results, a discrimination criterion is constructed. The discrimination criterion is obtained by extracting microseismic parameters, stress parameters, approximate energy ratio, detailed energy ratio, approximate sample entropy, and detailed sample entropy in different energy level intervals. Based on the aforementioned discrimination criteria, a dynamic discrimination model is established to classify the main controlling factors of microseismic events within the excavation unloading range, and the discrimination results are obtained.

[0006] Secondly, this application also provides a microseismic discrimination system for vibration and excavation unloading fracture of deep-buried tunnel TBMs, including: The acquisition module is used to acquire microseismic event data within the target area; The classification module is used to perform spatial classification based on the microseismic event data to obtain the influence range of the excavation unloading effect; The segmentation module is used to divide the region according to the influence range, synchronously collect rock fracture waveforms, microseismic parameters and stress parameters in each region, and filter the collected rock fracture signals to extract the noise-reduced rock fracture signals. The decomposition module uses wavelet transform to perform multi-level decomposition of the rock fracture signal, collects the frequency band energy ratio parameters, detail coefficients and approximation coefficients generated after each level of wavelet decomposition, and calculates the entropy value after each level of decomposition to obtain a set of feature parameters. The classification module is used to perform energy classification based on the set of feature parameters to obtain classification results, which include low-energy events, medium-energy events and high-energy events. The construction module is used to construct the discrimination criteria based on the classification results. The discrimination criteria are obtained by extracting microseismic parameters, stress parameters, approximate energy ratio, detailed energy ratio, approximate sample entropy and detailed sample entropy in different energy level intervals. The discrimination module makes a judgment based on the discrimination criteria. It classifies the main controlling factors of microseismic events within the excavation unloading range by establishing a dynamic discrimination model and obtains the discrimination result.

[0007] The beneficial effects of this invention are as follows: This invention constructs a dynamic model based on a multi-level discrimination standard, classifies the main controlling factors of real-time microseismic events, and realizes real-time dynamic discrimination of TBM vibration and excavation unloading rupture during construction, significantly improving the early warning timeliness of time-delay rockburst. Attached Figure Description

[0008] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a schematic diagram of the microseismic discrimination method for vibration and excavation unloading rupture of a deep-buried tunnel TBM as described in an embodiment of the present invention; Figure 2 This is a schematic diagram of a microseismic discrimination system for vibration and excavation unloading fracture of a deep-buried tunnel TBM, as described in an embodiment of the present invention. Figure 3 The distribution characteristics of microseismic events in a deeply buried TBM tunnel in this embodiment of the invention; Figure 4 This is a schematic diagram illustrating the importance of microseismic event classification features from 0J to 10J in an embodiment of the present invention; Figure 5 This is a schematic diagram illustrating the importance of microseismic event classification features for 10J-100J events in an embodiment of the present invention. Figure 6 This is a distribution diagram of the magnitude of the dominant frequency of excavation unloading-induced rupture corresponding to microseismic events with energy levels of 100J-1000J in an embodiment of the present invention; Figure 7 This is a distribution diagram of the dominant frequency of TBM vibration-induced rupture corresponding to microseismic events with energy levels of 100J-1000J in an embodiment of the present invention; The diagram is labeled as follows: 901, Acquisition Module; 902, Classification Module; 903, Division Module; 904, Decomposition Module; 905, Hierarchical Module; 906, Construction Module; 907, Discrimination Module. Detailed Implementation

[0010] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0011] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0012] Example 1: This embodiment provides a microseismic discrimination method for vibration and excavation unloading rupture of a deep-buried tunnel TBM.

[0013] See Figure 1 The figure shows that the method includes steps S100 to S600.

[0014] Step S100: Obtain microseismic event data within the target area; Understandably, during tunnel construction, a microseismic monitoring system is first deployed in areas at risk of rockburst (such as behind the tunnel face and in areas with high ground stress). The system uses a sensor array to collect micro-fracture signals inside the surrounding rock in real time, covering key areas including the excavation unloading impact zone and the TBM vibration disturbance zone.

[0015] Step S200: Perform spatial classification based on microseismic event data to obtain the influence range of excavation unloading effect; It should be noted that this step, by accurately delineating the dominant area of ​​the excavation unloading effect, provides a spatial classification basis for subsequent parameter statistics and supports the identification of the main controlling factors.

[0016] Further, step S200 includes steps S210 to S230.

[0017] Step S210: Analyze the event location distribution based on the spatial coordinates in the microseismic event data. By calculating the relative distance between the microseismic events and the tunnel face, the event location distribution dataset is obtained. Step S220: Perform spatial density gradient statistics based on the event location distribution dataset. By analyzing the spatial density variation trend of microseismic events along the tunnel axis, obtain the key density gradient abrupt change points that characterize the stress adjustment boundary. Step S230: Delineate the influence range based on the key density gradient abrupt change point, and determine the stress adjustment range dominated by the excavation unloading effect through the preset tunnel diameter multiple relationship, thus obtaining the influence range.

[0018] Specifically, based on the three-dimensional spatial coordinate data of microseismic events, the axial relative distance between each event and the tunnel face is first calculated (with the tunnel face as the origin and the excavation direction as the positive direction), generating a dataset of event location distribution. Then, the spatial density distribution of microseismic events along the tunnel axis is statistically analyzed, and the trend of density gradient changes is analyzed to identify the abrupt change point of density drop, i.e., the boundary of the excavation unloading effect's influence range is located three times the tunnel diameter behind the tunnel face. Figure 3 Finally, based on the tunnel diameter multiple relationship, the stress adjustment range dominated by excavation unloading (X∈[-3D,0]) is defined, and events outside the range (X<-3D) are attributed to TBM vibration disturbance. Here, X represents the stress adjustment range, and D represents the tunnel diameter. Generally, 6:00-12:00 is the TBM maintenance period within 24 hours. During this period, the TBM is stopped and does not tunnel. Therefore, the micro-vibration events generated during this period are all micro-vibration events caused by excavation unloading, realizing the spatiotemporal separation of the two types of mechanical action domains.

[0019] Step S300: Divide the affected area into regions, simultaneously collect rock fracture waveforms, microseismic parameters and stress parameters in each region, and filter the collected rock fracture signals to extract the noise-reduced rock fracture signals. It should be noted that during TBM construction, the mechanical vibrations generated during TBM excavation cause the on-site microseismic sensors to continuously collect excavation vibration signals, which may interfere with the rock fracture signals, distorting their time-frequency characteristics. Therefore, it is necessary to filter and reduce noise from the collected rock fracture signals to extract the true rock fracture characteristics.

[0020] Step S400: Use wavelet transform to perform multi-level decomposition on the rock fracture signal, collect the frequency band energy ratio parameters, detail coefficients and approximation coefficients generated after each wavelet decomposition, and calculate the entropy value after each decomposition to obtain the set of characteristic parameters.

[0021] Further, step S400 includes steps S410 to S430.

[0022] Step S410: Based on the denoised rock fracture signal, perform wavelet decomposition layer selection and multi-layer decomposition processing to obtain the approximation coefficients and detail coefficients after each layer decomposition. Step S420: Calculate the frequency band energy ratio and extract time-domain features based on the approximation coefficients and detail coefficients to obtain the energy ratio parameters and time-domain feature set; Step S430: Calculate the sample entropy based on the energy ratio parameter and the time-domain feature set to obtain the feature parameter set.

[0023] In this invention, wavelet transform uses "db5" as the wavelet basis and has 5 decomposition levels. Practice has shown that this wavelet basis and decomposition level effectively improve the signal-to-noise ratio and optimize feature extraction, making it the best choice for identifying microseismic source types. By processing rock fracture signals using wavelet transform, the following parameters can be obtained: approximate energy percentage, detailed energy percentage (covering layers 1 to 5), approximate sample entropy, and detailed sample entropy (also covering layers 1 to 5). Furthermore, by determining the relative relationship between the location, time, and boundary of microseismic events (TBM shutdown and non-excavation are excavation unloading events, X<-3D is a TBM disturbance event), events are classified into excavation unloading effect-dominated areas and TBM vibration disturbance-dominated areas, forming a preliminary regional division dataset. Subsequently, preset formulas are applied to events inside and outside the range to calculate microseismic release energy, apparent volume, and waveform dominant frequency, generating a microseismic parameter set containing multidimensional parameters. In practice, the microseismic parameters include the number of instantaneous microseismic events, the instantaneous microseismic release energy, and the instantaneous microseismic apparent volume; the stress parameters include apparent stress; and the spectral physical information features include approximate energy proportion, detailed energy proportion (layer 1 to layer 5), approximate sample entropy, and detailed sample entropy (layer 1 to layer 5).

[0024] The energy released by microseismic events can be calculated using the following formula: ; in, It is the energy released by microseismic events; Density of the rock; for Wave, Wave speed; The distance from the sensor to the rupture site; The waveform velocity pulse is a time function; Duration; Indicates time; Represents the differential symbol.

[0025] The apparent volume of microseismic activity is calculated using the following formula: ; in, The apparent volume of microseismic events; The shear modulus of the rock; For radiated microseismic energy; This is a change in the potential of a microseismic body.

[0026] Apparent stress is calculated using the following formula: ; in, For apparent stress; The shear modulus of the rock; For radiated microseismic energy; It is a moment magnitude earthquake.

[0027] Understandably, this step uses wavelet transform to analyze regional statistical parameters, revealing the difference in response between TBM vibration and excavation unloading fracture, thus providing data support for the subsequent construction of discrimination criteria.

[0028] Step S500: Perform energy classification based on the set of feature parameters to obtain classification results, which include low-energy events, medium-energy events, and high-energy events. Further, step S500 includes steps S510 to S530.

[0029] Step S510: Energy parameter extraction processing is performed based on the feature parameter set. By analyzing the released energy values ​​of microseismic events in the region, an energy parameter dataset is obtained. Step S520: Perform dynamic threshold division processing based on the energy parameter dataset. By statistically analyzing the energy accumulation distribution characteristics of historical microseismic events, obtain the dynamic energy thresholds for low, medium, and high energy levels. Step S530: Perform event energy level classification processing based on dynamic energy threshold. By matching the energy released by microseismic events with the threshold range, a classification result containing energy classification labels is generated.

[0030] Specifically, the classification is based on the following: After wavelet transform, statistical analysis of the parameters revealed differences in the proportion of detailed energy (layer 2), detailed sample entropy (layer 4), and detailed sample entropy (layer 3) between low and medium energy levels. Therefore, microseismic events are classified into three intervals: 0J-10J (low energy level), 10J-100J (medium energy level), and 100J-1000J (high energy level). Furthermore, during TBM excavation, elastic and inelastic deformations occur within the rock mass. The elastic potential energy accumulated in the rock mass is gradually or suddenly released outwards as vibration waves along the surrounding rock medium during inelastic deformation, resulting in microseismic events. During the transition from elastic to inelastic deformation, the rock mass may crack or undergo frictional sliding, releasing the energy stored within it as elastic waves—this energy is the microseismic release energy.

[0031] The formation of rockbursts exhibits a progressive failure characteristic, and its evolution mechanism can be divided into several stages: the initiation of microcracks parallel to the free surface, followed by crack propagation and penetration, ultimately forming a macroscopic fracture surface. This failure evolution process is accompanied by energy release, and the intensity of energy release is positively correlated with the scale and propagation rate of crack development. Specifically, the larger the crack propagation range and the faster the fracture speed, the stronger the microseismic energy radiation generated. Based on a high-precision microseismic monitoring system, elastic wave signals generated by crack development within the rock mass can be captured in real time, and the energy during crack propagation can be calculated. For example, authorized patent CN103742156B determines the rockburst level based on the microseismic energy released from microfractures, thereby reducing the risk of columnar rockbursts. Generally, higher microseismic energy means that cracks within the rock mass continuously expand and enlarge, eventually penetrating. Different microseismic energies correspond to different fracture scales, and different fracture scales correspond to different microseismic parameters, stress parameters, time domain, frequency domain, and time-frequency characteristics. Therefore, microseismic energy can be used as an indicator to classify fractures within the rock mass.

[0032] Step S600: Construct discrimination criteria based on the classification results. The discrimination criteria are obtained by extracting microseismic parameters, stress parameters, approximate energy ratio, detailed energy ratio, approximate sample entropy, and detailed sample entropy in different energy level ranges. Further, step S600 includes steps S610 to S630.

[0033] Step S610: Based on the microseismic parameters and detailed energy proportions of low-energy events in the classification results, perform frequency domain-volume joint analysis to generate low-energy discrimination intervals; Step S620: Perform stress-entropy value fusion processing based on the stress parameters, detail sample entropy, and detail energy ratio of the intermediate-level events in the classification results to generate the intermediate-level stress discrimination interval; Step S630: Based on the distribution characteristics of the spectral parameters of high-energy events in the classification results, perform frequency domain analysis, extract the difference in the spectral response range between vibration disturbance and excavation unloading rupture, and generate high-energy spectral discrimination intervals.

[0034] It should be noted that relying solely on a single or partial parameter from the microseismic parameters, stress parameters, dominant frequency parameters, and spectral parameters is insufficient to effectively distinguish between microseismic events caused by TBM construction vibration and those caused by excavation unloading. Preferably, this invention employs a multi-parameter joint judgment method, combining decision trees and random forest models for microseismic event classification. The decision tree automatically learns and generates judgment rules based on feature thresholds by analyzing the relationship between features and labels in the training data. Figures 4 and 5 illustrate the importance of microseismic parameters, stress parameters, dominant frequency parameters, and spectral parameters for microseismic events at different energy levels. Based on the differences in the importance of each parameter for microseismic events at different energy levels, a threshold is set to determine whether the parameter exceeds the threshold for splitting, thus progressively classifying the microseismic events into two categories: TBM disturbance and excavation unloading. After five cross-validations, the accuracy of the decision tree classification is confirmed (>90%). This method enables the decision tree to effectively identify the differences in characteristic parameters between the two types of microseismic events, achieving accurate classification.

[0035] Step S700: Based on the discrimination criteria, make a discrimination. By establishing a dynamic discrimination model, classify the main controlling factors of microseismic events within the excavation unloading range and obtain the discrimination results.

[0036] It should be noted that this step is based on the energy classification results of real-time microseismic events, matching the corresponding energy level discrimination rules. If the event parameters meet the preset TBM vibration disturbance characteristic range or boundary, the main controlling factor is determined to be TBM vibration disturbance; otherwise, it is attributed to the excavation unloading effect. This model achieves real-time adaptive classification by dynamically updating the discrimination threshold, and finally outputs the discrimination results, providing a direct basis for time-delay rockburst risk early warning and TBM tunneling parameter optimization, solving the technical problem in existing technologies that cannot quantify the contribution of two types of mechanical effects to rockburst.

[0037] Further, step S700 includes steps S710 to S730.

[0038] Step S710: Based on the discrimination criteria, a dynamic model is constructed by integrating hierarchical threshold rules of different energy level ranges with multi-parameter combination logic to establish a dynamic discrimination model containing multi-parameter discrimination logic. Step S720: Extract parameters from the microseismic events within the excavation unloading range acquired in real time. Generate a real-time microseismic parameter set by calculating apparent volume, apparent stress, approximate energy ratio, detailed energy ratio, approximate sample entropy, and detailed sample entropy. Step S730: Classify the main control factors according to the dynamic discrimination model and the real-time microseismic parameter set, and output the main control factor discrimination results of rock fracture events by matching the discrimination rules corresponding to the energy level intervals.

[0039] Specifically, using features extracted from wavelet decomposition and microseismic parameters and stress parameters, including apparent volume, apparent stress, approximate energy percentage, detail energy percentage (layers 1 to 5), approximate sample entropy, and detail sample entropy (layers 1 to 5), decision tree and random forest models were trained respectively. These features can effectively capture the time-frequency characteristics of rock fracture signals. Through training, the model can learn different characteristic patterns of TBM vibration and excavation unloading fracture responses. Cross-validation was used to optimize the parameters of the trained model, ensuring good generalization ability on different datasets and maintaining high recognition accuracy.

[0040] This step compares and analyzes the microseismic parameters, stress parameters, and spectral physical information characteristics after wavelet transform within and outside the excavation unloading range, and comprehensively identifies the main controlling factors of rock fracture based on the multi-parameter characteristics.

[0041] The following example, a deep TBM tunnel project in Southwest China, provides a detailed verification of the practical application effectiveness of this invention: First, in this deep TBM tunnel project, based on the spatial distribution characteristics of microseismic events (such as... Figure 3 As shown in the figure, the influence range of the excavation unloading effect was determined to be three times the tunnel diameter behind the tunnel face (X∈[-3D,0]). Based on the on-site TBM construction schedule, 6:00-12:00 was designated as the TBM maintenance time, during which the TBM would be shut down and not operating. Subsequently, microseismic events outside the excavation unloading influence range (X<-3D) and microseismic events generated during TBM shutdown (caused by excavation unloading) were collected. Rock fracture signals and microseismic parameters (waveform dominant frequency) were extracted. Micro-vibration release energy , depending on volume ) and stress parameters (apparent stress) The approximate energy ratio, detail energy ratio (layers 1 to 5), approximate sample entropy and detail sample entropy (layers 1 to 5), microseismic parameters and stress parameters were calculated after wavelet transform.

[0042] Table 1 presents the statistical results of microseismic events occurring outside the excavation unloading effect range (X<-3D) in this project, including approximate energy percentage, detailed energy percentage (layers 1 to 5), approximate sample entropy, detailed sample entropy (layers 1 to 5), and waveform dominant frequency (…). Micro-vibration energy release ( ), apparent volume ( ), apparent stress ( They are divided into three levels according to energy: 0J-10J (low energy level), 10J-100J (medium energy level), and 100J-1000J (high energy level).

[0043] Table 1. Statistical results of microseismic events outside the influence range of excavation unloading effect. ;

[0044] Table 2 presents the statistical results of microseismic events within the influence range of the excavation unloading effect (X∈[-3D,0]) of this project, including approximate energy proportion, detailed energy proportion (layers 1 to 5), approximate sample entropy, detailed sample entropy (layers 1 to 5), and waveform dominant frequency (…). Micro-vibration energy release ( ), apparent volume ( ), apparent stress ( They are divided into three levels according to energy: 0J-10J (low energy level), 10J-100J (medium energy level), and 100J-1000J (high energy level).

[0045] Table 2. Statistical results of microseismic events within the influence range of excavation unloading effect. ;

[0046] Based on the microseismic event data in Table 1 (outside the influence range of excavation unloading effect) and Table 2 (within the influence range of excavation unloading effect), the determination parameters for each energy level (low energy level, medium energy level, and high energy level) were calculated, and the results are as follows. Figures 4-7 As shown.

[0047] In the subsequent construction of this TBM tunnel, based on the established discrimination criteria, the main controlling factors (TBM vibration disturbance or excavation unloading effect) of microseismic events within the excavation unloading range were determined. Specifically, the following combination of discrimination parameters was used for microseismic events in different energy ranges: When the microseismic energy is 0-10 J, if the energy percentage of detail (second layer) is less than 10.986, then the energy percentage of detail (first layer) is further determined: if it is less than 0.0293, it is determined to be dominated by excavation unloading; if it is greater than or equal to 0.0293, it is determined to be dominated by TBM disturbance. When the energy percentage of detail (second layer) is greater than or equal to 10.986, based on the apparent volume ( The following classifications are made: if the apparent volume is less than 506.97 m³, it is determined to be dominated by TBM disturbance; if the apparent volume is greater than or equal to 506.97 m³ and less than 3755.2 m³, it is determined to be dominated by excavation unloading; if the apparent volume is greater than or equal to 3755.2 m³, it is determined to be dominated by TBM disturbance.

[0048] When the microseismic energy is 10J-100J, first determine if the apparent stress (Pa) is less than 80617.2. If it is, check if the entropy of the detail sample (layer 1) is less than 0.530. If it is, further check if the entropy of the detail sample (layer 2) is less than 0.054. If it is, it is determined to be a TBM disturbance. Otherwise, determine whether it is excavation unloading or a TBM disturbance based on whether the energy ratio of the detail sample (layer 3) is less than 58.932. If the apparent stress is greater than or equal to 80617.2, check if the energy ratio of the detail sample (layer 3) is less than 23.306. If it is, further check if the apparent stress (Pa) is less than 4741.72 to determine the type. If the apparent stress is greater than or equal to 4741.72, it is a TBM disturbance. If the apparent stress is less than 4741.72, it is an excavation unloading. Otherwise, if the energy ratio of the detail sample is greater than or equal to 23.306, it is directly determined to be a TBM disturbance.

[0049] When the micro-vibration energy is 100J-1000J: with the dominant frequency of the waveform ( ) is the criterion, if If the frequency is within the range of [10Hz, 200Hz], it is determined to be a TBM vibration disturbance-dominated fracture (characteristic of low-frequency vibration waves), while the dominant frequency of a fracture dominated by excavation unloading effect is usually concentrated in the high-frequency range. (∈[200Hz, 600Hz]).

[0050] Outside the range of excavation unloading effect (X<-3D), although the stress field and energy distribution of the surrounding rock have tended to a secondary equilibrium state, the continuous action of TBM cutter vibration disturbance leads to energy accumulation in the rock mass in the area >3D behind the tunnel face. >1000 J) and the accumulation of microseismic events. During this process, the TBM vibration stress wave causes the expansion of microcracks in the rock mass, triggering cumulative damage to the surrounding rock. The above-mentioned damage accumulation effect significantly weakens the bearing capacity of the rock mass, ultimately leading to the rapid release of strain energy and triggering time-delayed rockburst.

[0051] Example 2: like Figure 2 As shown, this embodiment provides a microseismic discrimination system for vibration and excavation unloading fracture of a deep-buried tunnel TBM. The system includes: The acquisition module 901 is used to acquire microseismic event data within the target area; Classification module 902 is used to perform spatial classification based on microseismic event data to obtain the influence range of excavation unloading effect; The segmentation module 903 is used to divide the area according to the influence range, synchronously collect rock fracture waveforms, microseismic parameters and stress parameters in each area, and filter the collected rock fracture signals to extract the noise-reduced rock fracture signals. Decomposition module 904 uses wavelet transform to perform multi-level decomposition of rock fracture signal, collects the frequency band energy ratio parameter, detail coefficient and approximation coefficient generated after each level of wavelet decomposition, and calculates the entropy value after each level of decomposition to obtain the set of characteristic parameters. The classification module 905 is used to perform energy classification based on the set of characteristic parameters to obtain classification results, which include low-energy events, medium-energy events and high-energy events. Module 906 is constructed to build discrimination criteria based on the classification results. The discrimination criteria are obtained by extracting microseismic parameters, stress parameters, approximate energy ratio, detailed energy ratio, approximate sample entropy, and detailed sample entropy in different energy level ranges. The discrimination module 907 makes discrimination based on discrimination criteria. It classifies the main controlling factors of microseismic events within the excavation unloading range by establishing a dynamic discrimination model and obtains the discrimination results.

[0052] In one specific embodiment of the present invention, the classification module 902 includes: The first classification unit is used to analyze the distribution of event locations based on the spatial coordinates in the microseismic event data. By calculating the relative distance between the microseismic event and the tunnel face, the event location distribution dataset is obtained. The second classification unit is used to determine the spatial density variation trend of microseismic events along the tunnel axis, and to obtain the key density gradient abrupt point that characterizes the stress adjustment boundary. The third classification unit is used to delineate the influence range based on the key density gradient abrupt change point. It determines the stress adjustment range dominated by the excavation unloading effect through a preset tunnel diameter multiple relationship, thus obtaining the influence range.

[0053] In one specific embodiment of the present invention, the decomposition module 904 includes: The first decomposition unit is used to select the wavelet decomposition layer and perform multi-layer decomposition processing based on the denoised rock fracture signal to obtain the approximation coefficients and detail coefficients after each layer decomposition. The second decomposition unit is used to calculate the frequency band energy ratio and extract time-domain features based on the approximation coefficients and detail coefficients to obtain the energy ratio parameters and time-domain feature set. The third decomposition unit is used to calculate the sample entropy based on the energy ratio parameter and the time-domain feature set to obtain the feature parameter set.

[0054] In one specific embodiment of the present invention, the hierarchical module 905 includes: The first hierarchical unit performs energy parameter extraction processing based on the feature parameter set. By analyzing the released energy values ​​of microseismic events in the region, an energy parameter dataset is obtained. The second hierarchical unit is used to perform dynamic threshold division processing based on the energy parameter dataset. By statistically analyzing the energy accumulation distribution characteristics of historical microseismic events, dynamic energy thresholds for low, medium, and high energy levels are obtained. The third classification unit performs event energy level classification based on dynamic energy thresholds. By matching the energy released by microseismic events with the threshold range, it generates classification results containing energy classification labels.

[0055] In one specific embodiment of the present invention, the construction module 906 includes: The first building unit is used to perform frequency-volume joint analysis based on the microseismic parameters and detailed energy proportions of low-energy events in the classification results, and generate low-energy discrimination intervals. The second building unit is used to perform stress-entropy value fusion processing based on the stress parameters, detail sample entropy, and detail energy ratio of intermediate-level events in the classification results, and to generate intermediate-level stress discrimination intervals. The third building block performs frequency domain analysis based on the spectral parameter distribution characteristics of high-energy events in the classification results, extracts the difference in spectral response range between TBM vibration disturbance and excavation unloading rupture, and generates high-energy spectral discrimination intervals.

[0056] In one specific embodiment of the present invention, the discrimination module 907 includes: The first discrimination unit performs dynamic model construction based on the discrimination criteria. By integrating hierarchical threshold rules and multi-parameter combination logic for different energy level ranges, a dynamic discrimination model containing multi-parameter discrimination logic is established. The second discrimination unit is used to extract parameters from microseismic events within the excavation unloading range acquired in real time. By calculating apparent volume, apparent stress, approximate energy ratio, detailed energy ratio, approximate sample entropy and detailed sample entropy, a real-time microseismic parameter set is generated. The third discrimination unit is used to classify the main control factors based on the dynamic discrimination model and the real-time microseismic parameter set. By matching the discrimination rules corresponding to the energy level intervals, it outputs the discrimination results of the main control factors of rock fracture events.

[0057] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A microseismic discrimination method for vibration and excavation unloading fracture of a deep-buried tunnel TBM, characterized in that, include: Acquire microseismic event data within the target area; Spatial classification is performed based on the microseismic event data to obtain the influence range of the excavation unloading effect. The affected area is divided into regions, and rock fracture waveforms, microseismic parameters and stress parameters of each region are collected simultaneously. The collected rock fracture signals are filtered and the noise-reduced rock fracture signals are extracted. The rock fracture signal was decomposed into multiple layers using wavelet transform. The frequency band energy ratio parameter, detail coefficient and approximation coefficient generated after each wavelet decomposition were collected, and the entropy value after each decomposition was calculated to obtain the set of characteristic parameters. Energy classification is performed based on the set of characteristic parameters to obtain classification results, which include low-energy events, medium-energy events, and high-energy events. Based on the classification results, a discrimination criterion is constructed. The discrimination criterion is obtained by extracting microseismic parameters, stress parameters, approximate energy ratio, detailed energy ratio, approximate sample entropy, and detailed sample entropy in different energy level intervals. Based on the aforementioned discrimination criteria, a dynamic discrimination model is established to classify the main controlling factors of microseismic events within the excavation unloading range, and the discrimination results are obtained.

2. The microseismic discrimination method for vibration and excavation unloading fracture of deep-buried tunnel TBMs according to claim 1, characterized in that, Spatial classification is performed based on the microseismic event data, including: Based on the spatial coordinates in the microseismic event data, the event location distribution is analyzed. By calculating the relative distance between the microseismic events and the tunnel face, the event location distribution dataset is obtained. Based on the event location distribution dataset, spatial density gradient statistics are performed. By analyzing the spatial density variation trend of microseismic events along the tunnel axis, key density gradient abrupt change points characterizing the stress adjustment boundary are obtained. The influence range is defined based on the key density gradient abrupt change point, and the stress adjustment range dominated by the excavation unloading effect is determined by the preset tunnel diameter multiple relationship, thus obtaining the influence range.

3. The microseismic discrimination method for vibration and excavation unloading fracture of deep-buried tunnel TBMs according to claim 1, characterized in that, The rock fracture signal is decomposed into multiple layers using wavelet transform, including: Based on the denoised rock fracture signal, wavelet decomposition layer selection and multi-layer decomposition processing are performed to obtain the approximation coefficients and detail coefficients of each decomposition layer. Based on the approximation coefficients and detail coefficients, the frequency band energy ratio is calculated and the time domain feature is extracted to obtain the energy ratio parameters and the time domain feature set. The sample entropy is calculated based on the energy ratio parameter and the time-domain feature set to obtain the feature parameter set.

4. The microseismic discrimination method for vibration and excavation unloading fracture of deep-buried tunnel TBMs according to claim 1, characterized in that, The set of feature parameters is used to perform energy classification, resulting in classification results, including: Based on the regional division results, energy parameters are extracted and processed. By analyzing the released energy values ​​of microseismic events within the region, an energy parameter dataset is obtained. Dynamic thresholding is performed based on the energy parameter dataset. By statistically analyzing the cumulative energy distribution characteristics of historical microseismic events, dynamic energy thresholds for low, medium, and high energy levels are obtained. Based on the dynamic energy threshold, event energy level classification is performed. By matching the energy released by microseismic events with the threshold range, a classification result containing energy classification labels is generated.

5. The microseismic discrimination method for vibration and excavation unloading fracture of deep-buried tunnel TBMs according to claim 1, characterized in that... Based on the classification results, the discrimination criteria are constructed, including: Based on the microseismic parameters and detailed energy proportions of low-energy events in the classification results, frequency domain-volume joint analysis is performed to generate low-energy discrimination intervals. Based on the stress parameters, detailed sample entropy, and detailed energy ratio of the intermediate-level events in the classification results, stress-entropy value fusion processing is performed to generate the intermediate-level stress discrimination interval. Frequency domain analysis is performed based on the spectral parameter distribution characteristics of high-energy events in the classification results to extract the difference in spectral response range between vibration disturbance and excavation unloading rupture, and to generate high-energy spectral discrimination intervals.

6. A microseismic discrimination system for vibration and excavation unloading fracture of a deep-buried tunnel TBM, characterized in that, include: The acquisition module is used to acquire microseismic event data within the target area; The classification module is used to perform spatial classification based on the microseismic event data to obtain the influence range of the excavation unloading effect; The segmentation module is used to divide the region according to the influence range, synchronously collect rock fracture waveforms, microseismic parameters and stress parameters in each region, and filter the collected rock fracture signals to extract the noise-reduced rock fracture signals. The decomposition module uses wavelet transform to perform multi-level decomposition of the rock fracture signal, collects the frequency band energy ratio parameters, detail coefficients and approximation coefficients generated after each level of wavelet decomposition, and calculates the entropy value after each level of decomposition to obtain a set of feature parameters. The classification module is used to perform energy classification based on the set of feature parameters to obtain classification results, which include low-energy events, medium-energy events and high-energy events. The construction module is used to construct the discrimination criteria based on the classification results. The discrimination criteria are obtained by extracting microseismic parameters, stress parameters, approximate energy ratio, detailed energy ratio, approximate sample entropy and detailed sample entropy in different energy level intervals. The discrimination module makes a judgment based on the discrimination criteria. It classifies the main controlling factors of microseismic events within the excavation unloading range by establishing a dynamic discrimination model and obtains the discrimination result.

7. The microseismic discrimination system for vibration and excavation unloading fracture of a deep-buried tunnel TBM according to claim 6, characterized in that, The classification module includes: The first classification unit is used to analyze the distribution of event locations based on the spatial coordinates in the microseismic event data. By calculating the relative distance between the microseismic event and the tunnel face, the event location distribution dataset is obtained. The second classification unit is used to determine the spatial density variation trend of microseismic events along the tunnel axis, and to obtain the key density gradient abrupt point that characterizes the stress adjustment boundary. The third classification unit is used to delineate the influence range based on the key density gradient abrupt change point. It determines the stress adjustment range dominated by the excavation unloading effect through a preset tunnel diameter multiple relationship, thus obtaining the influence range.

8. The microseismic discrimination system for vibration and excavation unloading fracture of a deep-buried tunnel TBM according to claim 6, characterized in that, The decomposition module includes: The first decomposition unit is used to select the wavelet decomposition layer and perform multi-layer decomposition processing based on the denoised rock fracture signal to obtain the approximation coefficients and detail coefficients after each layer decomposition. The second decomposition unit is used to calculate the frequency band energy ratio and extract time-domain features based on the approximation coefficients and detail coefficients to obtain the energy ratio parameters and time-domain feature set. The third decomposition unit is used to calculate the sample entropy based on the energy ratio parameter and the time-domain feature set to obtain the feature parameter set.

9. The microseismic discrimination system for vibration and excavation unloading fracture of a deep-buried tunnel TBM according to claim 6, characterized in that, The hierarchical module includes: The first hierarchical unit performs energy parameter extraction processing based on the region division results, and obtains the energy parameter dataset by analyzing the released energy values ​​of microseismic events within the region; The second classification unit is used to perform dynamic threshold division processing based on the energy parameter dataset. By statistically analyzing the energy accumulation distribution characteristics of historical microseismic events, dynamic energy thresholds for low, medium, and high energy levels are obtained. The third grading unit performs event energy level classification processing based on the dynamic energy threshold. By matching the energy released by microseismic events with the threshold range, it generates a grading result containing energy grading labels.

10. A microseismic discrimination system for vibration and excavation unloading fracture of a deep-buried tunnel TBM according to claim 6, characterized in that, The classification results include low-energy events, medium-energy events, and high-energy events, and the construction module includes: The first building unit is used to perform frequency-volume joint analysis based on the microseismic parameters and detailed energy proportions of low-energy events in the classification results, and generate low-energy discrimination intervals. The second building unit is used to perform stress-entropy value fusion processing based on the stress parameters, detail sample entropy, and detail energy ratio of intermediate-level events in the classification results, and to generate intermediate-level stress discrimination intervals. The third building block performs frequency domain analysis based on the spectral parameter distribution characteristics of high-energy events in the classification results, extracts the difference in spectral response range between vibration disturbance and excavation unloading rupture, and generates high-energy spectral discrimination intervals.

Images