Stratum disturbance analysis method and system under shield construction coupling effect

Abstract

The invention provides a stratum disturbance analysis method and system under a shield construction coupling effect. Cutting vibration spectrum data are collected in real time through a cutter vibration sensor, and a feature fingerprint database containing a vibration energy distribution mode and a critical grouting interval is constructed in combination with soil parameters. And aligning the vibration spectrum with the soil bin pressure in a space-time manner, and generating a disturbance field distribution diagram for displaying an energy gradient distribution curve and a stress diffusion path topology. And matching the energy distribution curve and correcting formation interface propagation parameters through the feature matching network optimized by transfer learning, and outputting a formation type identification result and disturbance dynamic parameters. And based on the mapping relation between the parameters and the critical grouting interval, the ground surface displacement data are linked to dynamically regulate and control the grouting pressure, graded early warning and parameter regulation instructions are generated, and a stratum disturbance monitoring-regulation and control closed loop is formed. According to the technical scheme, dynamic optimization of the grouting pressure is achieved, and the stratum deformation risk caused by shield construction is remarkably reduced.

Description

Technical Field

[0001] This application relates to the technical field of intelligent perception and Internet of Things, and particularly to a method and system for analyzing formation disturbance under the coupling action of shield tunneling construction. Background Art

[0002] In the development of urban underground space, when shield tunneling passes through a composite formation of interbedded clay and gravel, the dynamic contact behavior between the cutter head and the soil will cause the spread of formation disturbance. It is necessary to capture the coupling characteristics of cutter head vibration and soil response in real time at the construction site, establish an association model between formation type identification and disturbance propagation, so as to dynamically adjust the synchronous grouting pressure and propulsion speed parameters during the construction process. However, due to the sudden change of formation parameters, the sensor data lags behind the actual disturbance state. The vibration wave generated by cutter head cutting and the soil stress wave form a non-linear superposition effect. Existing monitoring methods are difficult to quantify the spatial topological characteristics of the disturbance propagation path, and the energy dissipation mechanism at the interface between sand and clay lacks dynamic modeling ability, resulting in a time difference between the warning signal and the actual formation damage.

[0003] Currently, the existing technical solution adopts a prediction method combining cutter head vibration spectrum analysis and BP neural network. By arranging an acceleration sensor array on the surface of the cutter head to collect vibration spectrum data at different rotational speeds, and using wavelet packet transform to extract the time-frequency domain energy distribution characteristics, a training data set containing formation type labels is constructed. Based on the BP neural network, a mapping relationship between the peak frequency band of vibration energy and soil parameters is established. After outputting the formation identification result, the grouting pressure adjustment instruction is triggered according to the preset vibration intensity threshold. Summary of the Invention

[0004] This application provides a method and system for analyzing formation disturbance under the coupling action of shield tunneling construction to solve the problem of insufficient adaptability in the prior art.

[0005] In the first aspect, this application provides a method for analyzing formation disturbance under the coupling action of shield tunneling construction, including:

[0006] Collect vibration spectrum data generated when the cutter head cuts the soil through vibration sensors arranged on the surface of the shield machine cutter head;

[0007] Based on the obtained plastic index of the clay formation and the particle size distribution parameters of the gravel formation, construct a formation feature map library, and combine the soil shear modulus, groundwater permeability coefficient and synchronous grouting pressure threshold to generate vibration feature fingerprints, where the vibration feature fingerprints include the vibration energy distribution patterns under different formation combinations and the corresponding critical grouting pressure intervals;

[0008] Align the vibration spectrum data in different regions of the cutter head with the monitored values of the soil chamber pressure in space and time to generate a disturbance field distribution map, which includes the energy gradient distribution curve of the vibration field and the topological stress diffusion path of the soil mass;

[0009] Input the disturbance field distribution map and the vibration characteristic fingerprint into a feature matching network optimized by transfer learning, so as to perform spectral shape matching between the energy gradient distribution curve and the vibration energy distribution pattern through the feature matching network, and at the same time correct the propagation path parameters in the formation interface transition zone according to the stress diffusion path topology, so as to output the recognition result of the current formation type and extract the disturbance propagation dynamic parameters;

[0010] According to the mapping relationship between the disturbance propagation dynamic parameters and the critical grouting pressure interval in the vibration characteristic fingerprint, combined with the surface displacement monitoring data, dynamically adjust the synchronous grouting pressure threshold, generate a graded warning signal and a grouting parameter adjustment instruction, and complete the closed-loop control of formation disturbance.

[0011] Optionally, based on the cutter head rotation angle encoding signal and the spatial position information of the vibration monitoring points, perform a space-time alignment operation on the vibration spectrum data in each region of the cutter head surface and the soil chamber pressure monitoring value to generate a space-time synchronous vibration pressure joint dataset;

[0012] Extract the vibration frequency spectrum segment synchronized with the cutter head rotation phase and the sudden change interval of the soil chamber pressure from the vibration pressure joint dataset, and establish a space-time correlation mapping relationship between the vibration energy and the pressure change;

[0013] According to the space-time correlation mapping relationship, calculate the cumulative increment of the vibration energy in each region within the cutter head tunneling section during consecutive rotation cycles, and generate an energy gradient distribution curve of the vibration representing the formation disturbance intensity;

[0014] Based on the spatial superposition relationship between the amplitude of the soil chamber pressure fluctuation and the energy gradient distribution curve of the vibration, identify the soil layer stress concentration area and calibrate the stress diffusion direction, and generate a stress diffusion topological network penetrating the soil layer;

[0015] Fuse the energy gradient distribution curve of the vibration and the stress diffusion topological network to form a disturbance field distribution map containing space-time correlated disturbance characteristics.

[0016] Optionally, based on the mapping relationship between the disturbance propagation dynamic parameters and the critical grouting pressure interval in the vibration characteristic fingerprint, determine the range of the critical grouting pressure threshold in the formation interface transition zone;

[0017] Extract the synchronous grouting reference pressure value corresponding to the current formation combination from the formation characteristic atlas library, and combine it with the instantaneous change amount of the surface settlement rate in the surface displacement monitoring data to generate a grouting pressure dynamic adjustment gradient table including the pressure fluctuation tolerance;

[0018] Collect the dynamic difference between the outlet pressure of the grouting pipe and the face support pressure in real time, and synchronously monitor the displacement conduction rate of the formation interface in the stress diffusion topology network. When the dynamic difference exceeds the tolerance range of the grouting pressure dynamic adjustment gradient table or the displacement conduction rate breaks through the preset diffusion rate threshold, trigger multi-level warning signals according to the degree of exceeding the standard;

[0019] Based on the multi-level warning signals, activate the linkage pressurization operation of the target grouting hole position and its adjacent hole positions, synchronously collect the surface displacement rebound amount and the stress diffusion direction offset parameter after pressurization, and calculate the grouting pressure compensation coefficient;

[0020] According to the grouting pressure compensation coefficient, adjust the critical grouting pressure threshold range, generate a grouting parameter adjustment instruction and bind it to the corresponding formation combination in the formation feature map library, and form a closed-loop control link based on stress diffusion topology and surface displacement feedback.

[0021] Optionally, based on the obtained plastic index threshold of the clay formation and the particle size distribution parameters of the gravel formation, divide the formation units in combination with the spatial coupling relationship between the plastic index and the particle size distribution. Each formation unit includes a shear modulus change interval and a permeability coefficient stratification interval;

[0022] Dynamically match the real-time torque fluctuation data of the screw conveyor with the shear modulus change interval of the formation unit, establish a real-time correlation rule between the mechanical load response and the soil shear modulus, and generate a compensation and correction coefficient for the grouting pressure based on the dynamic difference between the outlet pressure of the grouting pipe and the face support pressure;

[0023] Integrate the compensation and correction coefficient, the permeability coefficient stratification interval and the real-time correlation rule according to the formation unit, and construct a formation feature map library including shear modulus correlation factors, grouting pressure compensation rules and permeability influence parameters;

[0024] Based on the plastic index weight, particle size distribution weight and grouting pressure gradient distribution of each unit in the formation feature map library, fuse and generate vibration energy distribution patterns under different formation combinations and vibration characteristic fingerprints corresponding to the critical grouting pressure intervals.

[0025] Optionally, input the vibration energy gradient distribution curve in the disturbance field distribution map and the vibration energy distribution patterns of different formation combinations in the vibration characteristic fingerprints into a feature matching network optimized by transfer learning, and perform spectral shape matching on the frequency band energy ratio and waveform similarity of the energy distribution curve through the feature matching network to generate a formation type matching degree index;

[0026] Extract the propagation path node parameters of the formation interface transition zone in the stress diffusion path topology, and correct the propagation path node parameters of the transition zone based on the stress conduction direction offset angle and connection strength between nodes;

[0027] Associate the corrected propagation path node parameters with the formation type matching degree index, extract the critical grouting pressure interval corresponding to the current formation combination from the vibration characteristic fingerprint, and generate a dynamic parameter set for the disturbance propagation in the formation interface transition zone;

[0028] According to the path offset and energy attenuation rate in the dynamic parameter set, output the recognition result of the current formation type and extract the disturbance propagation dynamic parameters.

[0029] Optionally, calculate the pressure - energy superposition coefficient representing the combined action intensity of the pressure amplitude and energy gradient in each region according to the spatial distribution consistency of the soil chamber pressure fluctuation amplitude and the vibration energy gradient distribution curve within the cutter head tunneling section;

[0030] Screen the regions where the pressure - energy superposition coefficient exceeds the preset threshold, mark them as initial stress concentration regions, and extract the energy gradient change slope between adjacent initial stress concentration regions;

[0031] Based on the directivity of the energy gradient change slope, calibrate the stress diffusion direction between the initial stress concentration regions, and merge adjacent regions with the same direction into stress conduction links;

[0032] Generate a stress diffusion topology network penetrating the soil layer according to the attenuation rate of the pressure - energy superposition coefficient and the link extension length in the stress conduction link.

[0033] Optionally, according to the level of the multi - level warning signal, activate the linkage pressurization device of the target grouting hole and its adjacent hole positions according to the preset priority, so that the grouting pressures of the target hole and the adjacent holes are synchronously increased to the preset pressurization value;

[0034] Synchronously collect the surface displacement rebound amount and stress diffusion direction offset angle data in the pressurized area. The surface displacement rebound amount is the difference between the surface uplift height after pressurization and the settlement amount before pressurization, and the stress diffusion direction offset angle is the angular deviation between the actual conduction direction and the preset diffusion path;

[0035] Based on the ratio of the absolute value of the surface displacement rebound amount to the preset rebound expected value, calculate the displacement rebound compensation weight, and at the same time, based on the difference ratio between the stress diffusion direction offset angle and the allowable offset angle threshold, calculate the direction offset correction factor;

[0036] Weight the displacement springback compensation weight and the direction offset correction factor and superimpose them to generate a grouting pressure compensation coefficient for quantifying the correction effect of the coupled boost operation on the suppression of formation disturbance.

[0037] In a second aspect, the present application provides a formation disturbance analysis system under the coupled action of shield construction, including:

[0038] An acquisition module that acquires vibration spectrum data generated when the cutter head cuts the soil through vibration sensors arranged on the surface of the shield machine cutter head;

[0039] A construction module that constructs a formation feature map library based on the obtained soil plasticity index of the clay formation and the particle size distribution parameters of the gravel formation, combines the soil shear modulus, the groundwater permeability coefficient, and the synchronous grouting pressure threshold, and generates a vibration feature fingerprint, where the vibration feature fingerprint includes the vibration energy distribution pattern and the corresponding critical grouting pressure interval under different formation combinations;

[0040] A generation module that aligns the vibration spectrum data in different regions of the cutter head and the monitored values of the soil chamber pressure in space and time to generate a disturbance field distribution map, where the disturbance field distribution map includes the energy gradient distribution curve of the vibration field and the topological structure of the stress diffusion path of the soil body;

[0041] A correction module that inputs the disturbance field distribution map and the vibration feature fingerprint into a feature matching network optimized by transfer learning, so as to perform spectral shape matching between the energy gradient distribution curve and the vibration energy distribution pattern through the feature matching network, and at the same time correct the propagation path parameters of the formation interface transition zone according to the topological structure of the stress diffusion path, so as to output the recognition result of the current formation type and extract the disturbance propagation dynamic parameters;

[0042] An adjustment module that dynamically adjusts the synchronous grouting pressure threshold according to the mapping relationship between the disturbance propagation dynamic parameters and the critical grouting pressure interval in the vibration feature fingerprint, combines the surface displacement monitoring data to generate a hierarchical warning signal and a grouting parameter adjustment instruction, and completes the closed-loop control of formation disturbance.

[0043] In a third aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a method for analyzing formation disturbance under the coupled action of shield construction as described in the first aspect above.

[0044] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program, and when the computer program is executed by a computer, it implements a method for analyzing formation disturbance under the coupled action of shield construction as described in the first aspect.

[0045] In the embodiments of the present application, vibration spectrum data generated when the cutter head cuts the soil body is collected through vibration sensors arranged on the surface of the cutter head of the shield machine; based on the soil plasticity index of the clay stratum and the particle size distribution parameters of the gravel stratum obtained, a stratum characteristic map library is constructed, and combined with the soil shear modulus, the groundwater permeability coefficient, and the synchronous grouting pressure threshold, a vibration characteristic fingerprint is generated. The vibration characteristic fingerprint includes the vibration energy distribution pattern under different stratum combinations and the corresponding critical grouting pressure range; the vibration spectrum data in different regions of the cutter head is aligned in space and time with the monitored value of the soil chamber pressure to generate a disturbance field distribution map. The disturbance field distribution map includes the energy gradient distribution curve of the vibration field and the topological structure of the stress diffusion path of the soil body; the disturbance field distribution map and the vibration characteristic fingerprint are input into a feature matching network optimized by transfer learning, so as to perform spectral shape matching between the energy gradient distribution curve and the vibration energy distribution pattern through the feature matching network, and at the same time correct the propagation path parameters in the stratum interface transition area according to the stress diffusion path topology, so as to output the recognition result of the current stratum type and extract the disturbance propagation dynamic parameters; according to the mapping relationship between the disturbance propagation dynamic parameters and the critical grouting pressure range in the vibration characteristic fingerprint, combined with the surface displacement monitoring data, the synchronous grouting pressure threshold is dynamically adjusted to generate a hierarchical warning signal and a grouting parameter adjustment instruction, and the closed-loop control of stratum disturbance is completed.

[0046] The present application has the following beneficial effects:

[0047] By collecting the vibration spectrum data of the cutter head in real time to improve the perception accuracy of the cutting state, generating a vibration characteristic fingerprint by combining multi-source geological parameters to achieve accurate mapping of stratum combinations and construction parameters, using space-time alignment to generate a disturbance field distribution map to quantify the correlation characteristics of energy gradient and stress diffusion path, and leveraging the feature matching network optimized by transfer learning to enhance the dynamic analysis ability of the energy dissipation mechanism at the complex stratum interface. Finally, by dynamically adjusting the grouting threshold in combination with the surface displacement data to establish a hierarchical warning and closed-loop control mechanism, it effectively solves the problems of insufficient modeling of disturbance propagation in composite strata and lagging warning in traditional methods.

[0048] Furthermore, by fusing the space-time alignment data of the cutter head vibration spectrum and the soil chamber pressure, the vibration energy gradient distribution and stress diffusion path of stratum disturbance during tunneling are accurately quantified, realizing the dynamic calibration of the soil layer stress concentration area and the real-time tracking of the disturbance propagation direction; based on the space-time correlated disturbance field distribution map, the risk of local stratum mutation within the cutter head rotation period can be dynamically identified, and combined with the stress diffusion topological network to predict the disturbance propagation trend, thereby optimizing construction parameters such as cutter head rotation speed and grouting pressure, and actively suppressing problems such as stratum collapse and settlement caused by vibration energy accumulation or stress concentration, significantly improving the safety of shield construction and the active prevention and control ability of stratum disturbance.

[0049] These aspects or other aspects of the present application will be more clearly understood in the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0051] Figure 1 Shows a flowchart of a method for analyzing formation disturbance under the coupling action of shield construction provided by the present application;

[0052] Figure 2 Shows a scenario diagram of a method for analyzing formation disturbance under the coupling action of shield construction provided by the present application;

[0053] Figure 3 Shows a schematic structural diagram of a system for analyzing formation disturbance under the coupling action of shield construction provided by the present application;

[0054] Figure 4 Shows a schematic structural diagram of a computing device provided by the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0055] In order to enable those skilled in the art to better understand the solutions of the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application.

[0056] In some processes described in the specification, claims and the above-mentioned drawings of the present application, there are multiple operations that appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order in which they appear herein or may be executed in parallel. The operation numbers such as 101, 102, etc. are only used to distinguish different operations, and the numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions such as "first" and "second" in this article are used to distinguish different messages, devices, modules, etc., and do not represent a sequence, nor do they limit that "first" and "second" are of different types.

[0057] Researchers found that in shield construction, the sudden change of formation parameters causes the sensor data to lag behind the actual disturbance state. The vibration wave generated by the cutter head cutting and the soil stress wave form a non-linear superposition effect. Existing monitoring means are difficult to quantify the spatial topological characteristics of the disturbance propagation path, and the energy dissipation mechanism at the interface between sand and clay lacks dynamic modeling ability, resulting in a time difference between the warning signal and the actual formation damage. Therefore, there is an urgent need for a method to dynamically analyze formation disturbance by fusing multi-source data.

[0058] In view of the above problems, the present invention proposes a method for analyzing formation disturbance under the coupling action of shield construction, the core of which lies in constructing a dynamic matching mechanism between the formation feature map library and the vibration feature fingerprint. Specifically, the vibration spectrum data of cutterhead cutting is collected by vibration sensors, and the vibration feature fingerprint including the energy distribution pattern and the critical pressure interval is generated by combining the soil plasticity index, the particle size distribution parameters and the grouting pressure threshold; the vibration spectrum and the soil chamber pressure are fused by the spatio-temporal alignment technology to generate a disturbance field distribution map, revealing the energy gradient and the stress diffusion path; then the feature matching network optimized by transfer learning is used to perform spectral shape matching between the energy distribution pattern and the formation feature map, and dynamically correct the formation interface propagation path parameters. This method solves the problem of sensor data lag, improves the formation recognition accuracy by quantifying the topological features of disturbance propagation, dynamically models the energy dissipation mechanism at the interface, and shortens the time difference between the warning signal and the actual formation damage.

[0059] Next, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts shall fall within the protection scope of the present application.

[0060] Figure 1 The flowchart of a method for analyzing formation disturbance under the coupling action of shield construction provided by the embodiments of the present application is as Figure 1 shown, and the method includes:

[0061] 101. Collect the vibration spectrum data generated when the cutterhead of the shield machine cuts the soil through the vibration sensors arranged on the surface of the cutterhead of the shield machine;

[0062] In the above step, the cutterhead of the shield machine refers to the rotary metal disc structure at the front end of the shield machine for cutting soil, and the surface thereof is distributed with cutters and soil discharge ports. The vibration sensor refers to a piezoelectric or acceleration measuring device installed at a specific position on the surface of the cutterhead for detecting mechanical vibration signals, which can convert mechanical vibration into electrical signals. The vibration spectrum data refers to the frequency-amplitude distribution data obtained by converting the time-domain vibration signal into a frequency-domain signal through Fourier transform, and includes the vibration energy characteristics of different frequency components.

[0063] In the embodiments of the present application, first, a triaxial vibration sensor array is evenly arranged on the working surface of the shield machine cutter head and connected to a data acquisition module at the center of the cutter head through a shielded cable to form a monitoring network. When the cutter head rotates to cut the soil, each sensor real-time collects three-dimensional vibration acceleration signals. The data acquisition module synchronously collects the vibration waveform data of each channel at a sampling frequency of 10 kHz and associates and records the cutter head rotation speed and propulsion pressure parameters. Subsequently, the digital signal processor performs band-pass filtering on the original signal in the range of 20 - 2000 Hz, uses the fast Fourier transform algorithm to convert the time-domain signal into a frequency-domain spectrogram, and extracts the 1 / 3 octave band spectrum features to form a vibration spectrum matrix. Finally, the processed spectrum data is spatially and temporally aligned with the geological exploration parameters and transmitted to the ground database server through the industrial Ethernet, and the geological unit number and mileage label are added according to the tunneling ring to form a standardized vibration feature data set.

[0064] In practical applications, for example, triaxial acceleration sensor arrays are respectively arranged in the central area of the shield machine cutter head and at the hob installation base, and the vibration signals during the rotation and cutting of the cutter head are real-time collected at a sampling frequency of 4000 Hz. During on-site construction, when the shield machine advances at a speed of 15 r / min to the section with interbedded silty clay and gravel, the monitoring system obtains the axial, radial, and tangential vibration data of the cutter head through a 6-channel synchronous acquisition module. After eliminating low-frequency mechanical noise and high-frequency electromagnetic interference through a 150 - 2500 Hz digital band-pass filter, the time-domain waveform data is segmented and compressed for storage in 60-second durations. By performing a fast Fourier transform on 320 groups of vibration signals collected during the propulsion of the 8th ring, key frequency components of 380 ± 20 Hz and 1350 ± 50 Hz are detected in the spectrogram. Combining with the geological radar detection data, it is confirmed that this characteristic spectral line corresponds to the composite vibration mode generated by the interaction between the cutter head and the pebble layer, successfully realizing the effective feature extraction of the cutter-rock / soil coupling vibration at the interface of complex strata and providing a high-precision raw data basis for subsequent disturbance level classification.

[0065] In the overall solution of step 101 above, by real-time collecting the vibration spectrum data of the shield cutter head and combining multi-dimensional feature fusion analysis methods, the dynamic response characteristics under the interaction between the cutter head and the soil are accurately extracted, the spatial distribution law and energy transfer mode of the cutting resistance in complex strata are effectively identified, a non-linear mapping relationship based on the spectral feature vector and the formation disturbance intensity is established, a multi-level early warning threshold system is formed, the real-time monitoring of the formation disturbance state and the risk classification prediction are realized. At the same time, the shield tunneling parameter configuration is optimized through a dynamic feedback mechanism, significantly improving the accuracy and adaptability of shield tunneling disturbance control under complex geological conditions, providing intelligent decision-making support for the safe advancement of the project, and ensuring the construction efficiency while reducing the risk of excessive formation deformation.

[0066] 102. Based on the obtained plasticity index of the clay stratum and the particle size distribution parameters of the gravel stratum, a stratum characteristic map library is constructed. Combining the soil shear modulus, groundwater permeability coefficient, and synchronous grouting pressure threshold, a vibration characteristic fingerprint is generated. The vibration characteristic fingerprint includes the vibration energy distribution pattern and the corresponding critical grouting pressure interval under different stratum combinations;

[0067] Optionally, step 102 may specifically include the following steps:

[0068] 1021. Based on the obtained plasticity index threshold of the clay stratum and the particle size distribution parameters of the gravel stratum, combined with the spatial coupling relationship between the plasticity index and the particle size distribution, the stratum units are divided. Each stratum unit includes a shear modulus change interval and a permeability coefficient stratification interval;

[0069] 1022. Dynamically match the real-time torque fluctuation data of the screw conveyor with the shear modulus change interval of the stratum unit, establish a real-time correlation rule between the mechanical load response and the soil shear modulus, and generate a compensation correction coefficient for the grouting pressure based on the dynamic difference between the grouting pipe outlet pressure and the face support pressure;

[0070] 1023. Integrate the compensation correction coefficient, the permeability coefficient stratification interval, and the real-time correlation rule according to the stratum unit to construct a stratum characteristic map library including the shear modulus correlation factor, the grouting pressure compensation rule, and the permeability influence parameter;

[0071] 1024. Based on the plasticity index weight, particle size distribution weight, and grouting pressure gradient distribution of each unit in the stratum characteristic map library, fuse and generate a vibration characteristic fingerprint including the vibration energy distribution pattern and the corresponding critical grouting pressure interval under different stratum combinations.

[0072] In the above steps, the plasticity index threshold of the clay stratum refers to the determination standard when the difference between the liquid limit and the plastic limit water content reaches the characteristic value of the cohesive soil; the particle size distribution parameters of the gravel stratum include the coefficient of uniformity and the coefficient of curvature of the particle size distribution curve; the stratum unit refers to a three-dimensional geological body unit divided based on the spatial variation of geological parameters; the shear modulus change interval reflects the range of the soil shear strength varying with the water content; the permeability coefficient stratification interval characterizes the difference in the hydraulic conductivity characteristics of different geological horizons; the grouting pressure compensation coefficient refers to the correction amount of the grouting pressure dynamically adjusted according to the face support pressure; the vibration characteristic fingerprint is a set of vibration modes with stratum identification characteristics generated by fusing multi-source geological parameters.

[0073] In the embodiment of the present application, first, through step 1021, based on the clay plasticity index and gravel particle size distribution data obtained from engineering surveys, a three-dimensional geological parameter field model is constructed using the geostatistical Kriging interpolation algorithm. The construction area is divided into cubic formation units of 2m×2m×2m according to the plasticity index mutation threshold and the particle size distribution critical value. The shear modulus variation range with water content is determined for each formation unit through direct shear tests, and a vertical stratification interval table of permeability coefficients is established in combination with pumping test data to form a basic geological unit containing mechanical parameters and hydraulic characteristics.

[0074] Secondly, through step 1022, real-time torque fluctuation data is collected by the current sensor of the screw conveyor drive motor. The dynamic time warping algorithm is used to perform pattern matching between the torque fluctuation characteristics and the shear modulus variation range of the formation unit, and an exponential correlation equation between the torque amplitude and the soil shear strength is established. The real-time data of the grouting pipe outlet pressure sensor and the face pressure sensor of the earth pressure balance shield machine are collected synchronously. Based on the fuzzy PID control algorithm, the dynamic pressure difference between the two is calculated, and a grouting pressure compensation correction coefficient that changes with the formation characteristics is generated to realize the dynamic adjustment of grouting parameters.

[0075] Then, through step 1023, the permeability coefficient stratification interval of each formation unit is quantified into five-level permeability influence parameters, and the shear modulus correlation equation is transformed into a correlation factor in the form of a weight coefficient. Parameter coding is carried out in combination with the grouting pressure compensation rule. The knowledge graph technology is used to construct a three-dimensional geological model. Each graph node stores the spatial coordinates, mechanical parameter set, and grouting strategy coding of the corresponding formation unit, and adjacent units are connected by relationship edges to form a complete formation feature graph library.

[0076] Finally, through step 1024, the principal influence factors of the plasticity index and particle size distribution are extracted using the principal component analysis method, and a three-dimensional parameter space is constructed in combination with the grouting pressure gradient distribution. The Mahalanobis distance algorithm is used to calculate the distribution density of each parameter combination in the space, and the vibration main frequency band distribution mode with a probability of more than 85% is screened out to determine the critical grouting pressure range threshold required to maintain formation stability. A vibration characteristic fingerprint library containing the vibration energy characteristics of typical formation combinations and grouting control parameters is generated through multi-source parameter fusion, providing a discrimination benchmark for subsequent formation disturbance analysis.

[0077] In practical applications, for example, in the shield tunneling of a certain composite stratum, based on the measured data of the plastic index of 18 - 26 in the silty clay stratum and the proportion of 65% of particles with a particle size of 0.5 to 20 mm in the gravel stratum, the tunneling section is divided into stratum units of 1.2 m × 1.2 m × 0.5 m using three-dimensional space grid modeling technology. The dynamic range of the clay shear modulus is set to 25 - 65 MPa, and the gravel shear modulus is 80 - 150 MPa. The measured parameters of the permeability coefficient of the clay stratum of 1×10^-6 cm / s and the gravel stratum of 5×10^-3 cm / s are associated. By collecting the torque fluctuation data of the screw conveyor in real time from 0 to 850 kN·m, the dynamic time warping algorithm is used to match it with the shear modulus of the stratum unit. When the torque standard deviation exceeds 45 kN·m, the associated rule is triggered for update. Synchronously, combined with the real-time difference between the outlet pressure of the grouting pipe of 2.5 - 3.2 MPa and the face support pressure of 1.7 - 2.4 MPa, a grouting compensation coefficient of 0.85 - 1.15 is generated when the pressure difference breaks through the threshold of 0.8 MPa, and the closed-loop regulation of the grouting pump speed is realized through a PID controller. The finally constructed characteristic atlas library of 325 stratum units integrates a 5D shear modulus correlation vector, a 3-level grouting pressure gradient rule, and double-layer permeability correction parameters. Using a fusion algorithm with a plastic index weight of 0.6 and a particle size distribution weight of 0.4, a vibration fingerprint containing the critical grouting pressure of 1.8 - 2.3 MPa for the clay-cobble interlayer and the characteristic of the main frequency band of 200 - 800 Hz in the gravel-dominated stratum is generated, realizing the real-time identification of the mutation characteristics of the stratum interface and the dynamic optimization of grouting parameters.

[0078] In the overall solution of step 102 above, a multi-parameter coupling analysis framework is constructed by fusing the plastic index of the clay stratum and the particle size distribution parameters of the gravel stratum. The stratum units are divided based on the spatial correlation law of the plastic state and particle size distribution. Combining the dynamic range of the shear modulus, the hierarchical threshold of the permeability coefficient, and the grouting pressure difference compensation mechanism, a stratum characteristic atlas library with mechanical load response characteristics is formed. Through the real-time mapping relationship between the torque fluctuation data of the screw conveyor and the change range of the shear modulus, a dynamic feedback mechanism between the shear resistance characteristics of the soil and the mechanical behavior of the shield is established. Based on the synergistic effect of the grouting pressure compensation rule and the penetration influence parameters, a vibration characteristic fingerprint containing the vibration energy distribution pattern and the critical grouting pressure range is generated, realizing the precise matching of the stratum combination type and the shield dynamic response, effectively predicting the disturbance propagation trend at the interface of different strata, providing a quantitative basis for the dynamic optimization of grouting parameters and the prevention of stratum instability, and significantly improving the foresight and adaptability of shield tunneling disturbance control under composite stratum conditions.

[0079] 103. Align the vibration spectrum data of different regions of the cutter head and the monitored values of the soil chamber pressure in space and time to generate a disturbance field distribution map, which includes the energy gradient distribution curve of the vibration field and the topological structure of the stress diffusion path of the soil body.

[0080] Optionally, step 103 may specifically include the following steps:

[0081] 1031. Based on the cutterhead rotation angle encoding signal and the spatial position information of the vibration monitoring points, perform a spatio-temporal alignment operation on the vibration spectrum data and the soil chamber pressure monitoring values of each area on the cutterhead surface to generate a spatio-temporally synchronized vibration-pressure joint dataset;

[0082] 1032. Extract the vibration spectrum segment synchronized with the cutterhead rotation phase and the soil chamber pressure mutation interval from the vibration-pressure joint dataset, and establish a spatio-temporal correlation mapping relationship between the vibration energy and the pressure change;

[0083] 1033. According to the spatio-temporal correlation mapping relationship, calculate the cumulative increment of the vibration energy of each area in the cutterhead tunneling section within consecutive rotation periods, and generate a vibration energy gradient distribution curve characterizing the formation disturbance intensity;

[0084] 1034. Based on the spatial superposition relationship between the soil chamber pressure fluctuation amplitude and the vibration energy gradient distribution curve, identify the soil layer stress concentration area and calibrate the stress diffusion direction to generate a stress diffusion topological network penetrating the soil layer;

[0085] Among them, step 1034 may specifically include the following process: According to the spatial distribution consistency of the soil chamber pressure fluctuation amplitude and the vibration energy gradient distribution curve in the cutterhead tunneling section, calculate the pressure-energy superposition coefficient characterizing the combined action intensity of the pressure amplitude and the energy gradient of each area; Screen the areas where the pressure-energy superposition coefficient exceeds the preset threshold, mark them as the initial stress concentration areas, and extract the energy gradient change slope between adjacent initial stress concentration areas; Based on the directivity of the energy gradient change slope, calibrate the stress diffusion direction between the initial stress concentration areas, and merge adjacent areas with the same direction into stress conduction links; According to the attenuation rate of the pressure-energy superposition coefficient and the link extension length in the stress conduction link, generate a stress diffusion topological network penetrating the soil layer.

[0086] 1035. Integrate the vibration energy gradient distribution curve and the stress diffusion topological network to form a disturbance field distribution map containing spatio-temporal correlation disturbance characteristics.

[0087] In the above steps, the cutter head rotation angle encoding signal refers to the angle pulse signal output by the photoelectric encoder installed on the cutter head drive shaft, which is used to accurately calibrate the rotation position of the cutter head; the spatial position information of the vibration monitoring points refers to the three-dimensional coordinate parameters of the vibration sensors arranged on the surface of the cutter head in the cutter head coordinate system; the spatio-temporal synchronous vibration pressure joint data set refers to the multi-dimensional data set of the vibration frequency spectrum data and the soil chamber pressure value with aligned timestamps; the vibration energy gradient distribution curve refers to the spatial distribution curve generated based on the cumulative increment of vibration energy; the stress diffusion topological network refers to the directed graph network structure reflecting the stress transmission path of the soil layer; the pressure energy superposition coefficient is the action intensity index calculated by synthesizing the pressure fluctuation amplitude and the vibration energy gradient; the stress conduction link refers to the adjacent area connection path with continuous stress transmission characteristics.

[0088] In the embodiment of the present application, first, the data spatio-temporal alignment operation is completed through step 1031. Based on the rotation angle signal output by the photoelectric encoder on the cutter head drive shaft, combined with the three-dimensional spatial coordinate parameters of the vibration sensors on the surface of the cutter head, the coordinate transformation algorithm is used to convert the vibration frequency spectrum data of each monitoring point to the fixed geographic coordinate system. The monitoring values of the four pressure zones in the soil chamber are synchronously collected, and the vibration data and the pressure data are unified to the millisecond-level time reference through the timestamp alignment algorithm, forming a vibration pressure joint data set including spatial position marks and time synchronization.

[0089] Secondly, the spatio-temporal correlation mapping is established through step 1032. The vibration frequency spectrum segments in each 15-degree phase interval of the cutter head are extracted from the joint data set, and the soil chamber pressure change data within the corresponding time window is intercepted. The dynamic time warping algorithm is used to analyze the correlation between the main vibration frequency band energy fluctuation and the pressure change curve, establish the exponential correlation equation between the vibration energy peak value and the pressure change rate in each phase interval, and generate the vibration pressure mapping matrix with spatio-temporal correspondence.

[0090] Then, step 1033 is executed to generate the vibration energy gradient distribution curve. The vibration energy in each phase interval is integrated within three consecutive rotation periods of the cutter head, and the energy cumulative increment of each 1°×1° grid unit in the tunneling section is calculated. The Kriging spatial interpolation method is used to convert the discrete point energy values into a continuous distribution surface, generating a gradient distribution curve reflecting the formation disturbance intensity, and the peak area of the curve corresponds to the position of strong soil disturbance.

[0091] Subsequently, a stress diffusion topological network is constructed through step 1034. The product of the amplitude of the soil bin pressure fluctuation and the vibration energy gradient in each grid cell is calculated as the pressure energy superposition coefficient. The cells with a coefficient exceeding the threshold of 2.5 are marked as the initial stress concentration areas. The energy gradient change rate vectors between adjacent concentrated areas are extracted, and the stress diffusion direction is determined based on the consistency of the vector direction angles. The adjacent areas with a direction deviation less than 10 degrees are connected to form a stress conduction link. By analyzing the coefficient attenuation gradient and the extension length in the link, a tree-like topological network including a main path and branch paths is constructed.

[0092] Finally, through step 1035, a disturbance field distribution map is generated by fusing the contour line data of the vibration energy gradient curve and the directed edge data of the stress topological network in three-dimensional space. The color rendering technology is used to implement a three-color grading display of the disturbance intensity, and the stress conduction direction is marked with arrow symbols. Finally, a visualization model of the disturbance field with spatio-temporal correlation characteristics is generated. The red-highlighted areas in the model represent strong disturbance zones, and the arrow links show the main paths of stress diffusion.

[0093] In practical applications, for example, during the shield tunneling in a certain composite stratum, based on the real-time signal with a resolution of 0.5° of the cutterhead rotation angle encoder and the three-dimensional space coordinates of 12 vibration monitoring points, the spatio-temporal interpolation algorithm is used to align the vibration spectrum data on the cutterhead surface and the monitoring values of the soil bin pressure sensor in the range of 0.18 - 0.35 MPa, generating a spatio-temporal synchronous data set with 240 groups per minute. By extracting the vibration spectrum segments corresponding to every 15° rotation phase of the cutterhead and the intervals where the sudden change in the soil bin pressure exceeds 0.08 MPa, a correlation matrix of the vibration energy and the pressure change rate in 6 cutting zones of the cutterhead is established. According to the calculation of the cumulative increment of the vibration energy in each area within 3 consecutive rotation cycles, a disturbance intensity distribution curve with an axial gradient of 0.8 - 3.2 kJ / m² and a radial gradient of 0.5 - 2.4 kJ / m² is generated. Combining with the spatial distribution of the amplitude of the soil bin pressure fluctuation in the range of 0.12 - 0.28 MPa, it is calculated that the pressure energy superposition coefficient in the area at the 3 o'clock direction of the cutterhead reaches 0.76, which is marked as the initial stress concentration area. Through the directional analysis of the energy gradient slope of 0.35 - 0.68 / m in the adjacent areas, a stress conduction link with an extension length of 1.8 meters is constructed. Finally, by fusing 6 stress conduction links and the vibration gradient curve, a three-dimensional disturbance field distribution map including 12 disturbance hot zones and 5 main diffusion paths is generated, realizing the dynamic visualization of the disturbance characteristics of the stratum within a range of 2.4 meters in front of the cutterhead.

[0094] In the overall solution of step 103 above, through the dual calibration of the cutter head rotation angle encoding and the spatial position of the vibration monitoring points, the spatio-temporal synchronous alignment of the vibration spectrum data and the soil chamber pressure monitoring values is realized, and a joint dataset with phase correlation characteristics is constructed; based on the spatio-temporal mapping relationship between the vibration energy and the pressure mutation, the cumulative increment of the vibration energy in each area of the cutter head within a continuous rotation period is extracted to generate a gradient distribution curve characterizing the formation disturbance intensity, and combined with the spatial superposition effect of the soil chamber pressure fluctuation amplitude, through the calculation of the pressure energy superposition coefficient and the analysis of the energy gradient slope, the stress concentration area is accurately identified and the diffusion direction is calibrated to form a multi-level stress conduction link network penetrating the soil layer; finally, through the dynamic fusion of the disturbance field distribution map, the three-dimensional disturbance field evolution law of the vibration energy gradient distribution, the stress diffusion path and the spatio-temporal correlation characteristics inside the formation is completely presented, providing a visual analysis basis for predicting the formation disturbance propagation range and dynamic regulation during the shield tunneling process, and effectively supporting the real-time optimization of construction parameters and the accurate prevention and control of disturbance risks.

[0095] 104. Input the disturbance field distribution map and the vibration feature fingerprint into a feature matching network optimized by transfer learning, so as to perform spectral shape matching between the energy gradient distribution curve and the vibration energy distribution pattern through the feature matching network, and at the same time correct the propagation path parameters of the formation interface transition area according to the stress diffusion path topology, so as to output the recognition result of the current formation type and extract the disturbance propagation dynamic parameters;

[0096] Optionally, step 104 may specifically include the following steps:

[0097] 1041. Input the vibration energy gradient distribution curve in the disturbance field distribution map and the vibration energy distribution patterns of different formation combinations in the vibration feature fingerprint into a feature matching network optimized by transfer learning, and perform spectral shape matching on the frequency band energy ratio and waveform similarity of the energy distribution curve through the feature matching network to generate a formation type matching degree index;

[0098] 1042. Extract the propagation path node parameters of the formation interface transition area in the stress diffusion path topology, and correct the propagation path node parameters of the transition area based on the stress conduction direction offset angle and connection strength between nodes;

[0099] 1043. Associate the corrected propagation path node parameters with the formation type matching degree index, extract the critical grouting pressure interval corresponding to the current formation combination from the vibration feature fingerprint, and generate a dynamic parameter set for the disturbance propagation in the formation interface transition area;

[0100] 1044. According to the path offset and energy attenuation rate in the dynamic parameter set, output the recognition result of the current formation type and extract the disturbance propagation dynamic parameters.

[0101] In the above steps, the feature matching network optimized by transfer learning refers to a deep learning model with cross-domain matching ability formed by fine-tuning a pre-trained convolutional neural network architecture through a formation feature dataset; spectral shape matching refers to a pattern recognition process that compares the frequency band energy ratio and waveform similarity of vibration energy distribution curves; the formation interface transition zone refers to a geological transition zone with a gradual change characteristic at the junction of different formation combinations; the propagation path node parameters include the stress conduction direction angle, connection strength coefficient, and energy attenuation rate; the dynamic parameter set refers to a time-varying parameter group that reflects the path offset and energy attenuation characteristics during the disturbance propagation process; the path offset is the angle difference between the stress conduction direction and the formation interface strike; the energy attenuation rate is the attenuation gradient of vibration energy per unit distance along the propagation path.

[0102] In the embodiment of the present application, first, spectral shape matching analysis is performed through step 1041. The vibration energy gradient distribution curve in the disturbance field distribution map and the typical formation combination energy pattern pre-stored in the vibration feature fingerprint library are input into the feature matching network. This network adopts a transfer learning strategy, loads the ResNet-50 model pre-trained on ImageNet, and realizes cross-domain adaptation by freezing the underlying convolutional layer and fine-tuning the top fully connected layer. The dynamic time warping algorithm is used to calculate the energy ratio difference degree between the input curve and the feature fingerprint in the 1 / 3 octave frequency band, and the formation type matching degree index in the range of 0-1 is generated by combining the waveform correlation coefficient.

[0103] Secondly, the propagation path parameters are corrected through step 1042. The propagation path node data of the formation interface transition zone is extracted from the stress diffusion topology network. The graph attention network is used to analyze the direction offset angle and connection strength coefficient between nodes, and the deviation value of the stress conduction direction angle is calculated and corrected through the direction cosine. Pruning processing is performed on the weak conduction links with a connection strength lower than 0.7, and the effective path nodes with a main conduction direction deviation less than 15 degrees are retained, and the propagation direction angle parameters and connection strength coefficient of the transition zone are updated.

[0104] Then, step 1043 is executed to generate a dynamic parameter set, and the corrected path node direction angle parameters and the formation type matching degree index are associated with multi-modal data. Based on the sorting result of the matching degree index, the critical grouting pressure intervals corresponding to the top three candidate formation combinations are retrieved from the vibration feature fingerprint library. According to the node distribution density of the current stress diffusion path, the energy attenuation gradient along the main conduction direction is calculated, and a dynamic parameter matrix including the path offset, attenuation rate, and critical pressure value is generated.

[0105] Finally, the recognition result is output through step 1044. The weighted voting algorithm is adopted to comprehensively consider the formation type matching degree index and the path parameter consistency index, and the formation combination with a comprehensive score exceeding 0.85 is selected as the final recognition result. The median offset and average attenuation rate of the main conduction path are extracted from the dynamic parameter matrix to generate a dynamic parameter report including the disturbance propagation speed, energy dissipation coefficient, and recommended grouting pressure range, providing a decision-making basis for the optimization of shield construction parameters.

[0106] In practical applications, for example, during the construction of a certain subway shield section, the vibration energy distribution curve containing an axial gradient of 0.8 - 3.2 kJ / m² and the vibration characteristic fingerprint in the main frequency band of 200 - 800 Hz are input into the feature matching network optimized based on the ResNet-18 architecture. Through the time-series feature extraction layer with a three-layer convolution kernel of 5×1, the cosine similarity between the energy proportion of each sub-band in the frequency range of 50 - 2000 Hz of the energy distribution curve and the standard vibration mode is calculated to generate a formation type matching degree index of 0.72 - 0.89. At the same time, the propagation path parameters of 12 transition zone nodes in the stress diffusion topological network are extracted. For the weak sections where the offset angle of the stress conduction direction between nodes exceeds 15 degrees and the connection strength is lower than 0.6, the cubic spline interpolation algorithm is used to correct the path curvature radius to the range of 0.8 - 1.5 m. The corrected node coordinate parameters and the formation matching degree index are subjected to multi-dimensional space mapping, and combined with the critical grouting pressure standard of 1.8 - 2.3 MPa for the clay-cobble interlayer in the vibration characteristic fingerprint, a dynamic parameter set including a path offset of 0.3 - 1.2 m and an energy attenuation rate of 0.15 - 0.35 per second is generated. Finally, by fusing and analyzing the characteristics of the strongly correlated regions with a matching degree index above 0.85 and the extension direction of the stress conduction link, the transition interface from silty clay to cobble formation is successfully identified, and the dynamic control parameters of the grouting pressure compensation coefficient of 1.05 - 1.12 and the inclination angle of the main disturbance propagation path of 28 - 35 degrees are extracted.

[0107] In the overall solution of step 104 above, a feature matching network optimized by transfer learning is used to achieve in-depth coupling analysis of the disturbance field distribution map and the vibration feature fingerprint. The spectral shape matching mechanism of the frequency band energy distribution and the waveform morphology is used to accurately calculate the formation type matching degree. Combining the dynamic correction of the node conduction direction offset angle and the connection strength in the stress diffusion path topology, a spatial correlation model of the disturbance propagation parameters in the formation interface transition zone is established; through the multi-dimensional dynamic parameter fusion of the energy attenuation rate and the path offset amount, the formation combination type recognition result and the disturbance propagation characteristics of the interface transition zone are synchronously output. At the same time, based on the matching critical grouting pressure interval, the shield grouting control strategy is adaptively optimized, significantly improving the prediction accuracy of the disturbance propagation path and the parameter adaptability at the interface of complex formations, providing real-time data support for the dynamic adjustment of the grouting pressure and the disturbance suppression when the shield crosses the overlapping area of multi-textured formations, and effectively ensuring the active prevention and control ability of the construction process against the risk of formation interface mutation.

[0108] 105. According to the mapping relationship between the disturbance propagation dynamic parameters and the critical grouting pressure interval in the vibration feature fingerprint, and combining the surface displacement monitoring data, dynamically adjust the synchronous grouting pressure threshold, generate a hierarchical warning signal and a grouting parameter adjustment instruction, and complete the closed-loop control of formation disturbance.

[0109] Optionally, step 105 may specifically include the following steps:

[0110] 1051. Based on the mapping relationship between the disturbance propagation dynamic parameters and the critical grouting pressure interval in the vibration feature fingerprint, determine the range of the critical grouting pressure threshold in the formation interface transition zone;

[0111] 1052. Extract the synchronous grouting reference pressure value corresponding to the current formation combination from the formation feature map library, and combine the instantaneous change amount of the surface settlement rate in the surface displacement monitoring data to generate a grouting pressure dynamic adjustment gradient table including the pressure fluctuation tolerance;

[0112] 1053. Real-time collect the dynamic difference between the grouting pipe outlet pressure and the face support pressure, and synchronously monitor the displacement conduction rate of the formation interface in the stress diffusion topology network. When the dynamic difference exceeds the tolerance range of the grouting pressure dynamic adjustment gradient table or the displacement conduction rate breaks through the preset diffusion rate threshold, trigger a multi-level warning signal according to the degree of exceeding the standard;

[0113] 1054. Based on the multi-level warning signal, activate the linkage pressurization operation of the target grouting hole position and its adjacent hole positions, synchronously collect the surface displacement rebound amount and the stress diffusion direction offset parameter after pressurization, and calculate the grouting pressure compensation coefficient;

[0114] Among them, step 1054 may specifically include the following process: According to the level of the multi-level warning signal, activate the linkage pressurization device of the target grouting hole and its adjacent hole positions according to the preset priority, so that the grouting pressures of the target hole and the adjacent holes are synchronously increased to the preset pressurization value; synchronously collect the data of the surface displacement rebound amount and the stress diffusion direction deviation angle in the pressurization area, where the surface displacement rebound amount is the difference between the surface uplift height after pressurization and the settlement amount before pressurization, and the stress diffusion direction deviation angle is the angle deviation between the actual conduction direction and the preset diffusion path; based on the ratio of the absolute value of the surface displacement rebound amount to the preset rebound expected value, calculate the displacement rebound compensation weight, and at the same time, based on the difference ratio between the stress diffusion direction deviation angle and the allowable deviation angle threshold, calculate the direction deviation correction factor; perform weighted superposition on the displacement rebound compensation weight and the direction deviation correction factor to generate a grouting pressure compensation coefficient for quantifying the correction effect of the linkage pressurization operation on the suppression of formation disturbance.

[0115] 1055. According to the grouting pressure compensation coefficient, adjust the critical grouting pressure threshold range, generate a grouting parameter adjustment instruction and bind it to the corresponding formation combination in the formation feature map library to form a closed-loop control link based on stress diffusion topology and surface displacement feedback.

[0116] In the above steps, the disturbance propagation dynamic parameter refers to a real-time data set including the path offset and the energy attenuation rate. The vibration characteristic fingerprint library refers to a database storing the correlation between the vibration energy modes of different formation combinations and the critical pressure. The formation interface transition zone refers to the gradually changing geological area at the junction of different formation combinations. The pressure buffer zone refers to the safety redundancy interval outside the critical pressure threshold range. The formation feature map library refers to a structured database storing the mechanical parameters, grouting rules and permeability coefficients of formation units. The synchronous grouting reference pressure value refers to the initial set pressure value required for formation stability. The instantaneous change amount of the surface settlement rate refers to the change value of the surface settlement height per minute. The pressure fluctuation tolerance refers to the maximum amplitude allowed for the grouting pressure to deviate from the reference value. The grouting pipe outlet pressure refers to the real-time pressure value at the end of the grouting system. The face support pressure refers to the pressure value in the soil bin in front of the shield machine to maintain the stability of the soil body. The dynamic difference refers to the real-time deviation between the grouting and the support pressure. The displacement conduction rate refers to the transmission speed of the formation displacement in the stress diffusion topology network (unit: mm / s). The multi-level warning signal refers to a three-level alarm divided according to the degree of parameter exceeding the standard. The linkage pressurization operation refers to the coordinated adjustment of the pressures of the target grouting hole and the adjacent holes. The surface displacement rebound amount refers to the change amount of the surface height after pressurization. The stress diffusion direction deviation angle refers to the angle deviation between the actual conduction direction and the preset path. The grouting pressure compensation coefficient refers to the pressure correction factor calculated by comprehensively considering the rebound amount and the direction deviation. The closed-loop control link refers to a periodic control process composed of monitoring, adjustment and feedback. The grouting parameter adjustment instruction refers to an operation command including the target pressure value, the response time and the control logic.

[0117] In the embodiment of the present application, first, the critical grouting pressure range is determined through step 1051. Based on the path offset of 20 mm and the energy attenuation rate of 1.5 dB / m in the disturbance propagation dynamic parameters, combined with the reference pressure value of 0.8 MPa for the matching formation combination in the vibration characteristic fingerprint library, the upper and lower limits of the critical pressure in the formation interface transition zone are calculated using the cubic spline interpolation algorithm. For example, for every 1 mm increase in the path offset, the lower limit of the pressure decreases by 0.005 MPa, and for every 0.1 dB / m increase in the energy attenuation rate, the upper limit of the pressure increases by 0.003 MPa. Secondly, the reference pressure value of 0.8 MPa is expanded with positive and negative tolerances of 0.1 MPa to generate a safe operation range of 0.6 MPa to 0.945 MPa, and redundant control to ensure formation stability is achieved by increasing the pressure buffer zone. For example, the actual control range for a certain sandy soil layer is set to 0.62 - 0.93 MPa.

[0118] Secondly, a dynamic adjustment gradient table is generated through step 1052. The synchronous grouting reference pressure value of 0.8 MPa for the current formation combination is extracted from the formation characteristic map library, combined with the surface settlement rate monitoring data of 4 mm / h, and the adjustment amplitude is calculated based on the proportional relationship of 0.1 MPa / mm / h between the settlement rate increment and the pressure adjustment amount. For example, a settlement rate of 4 mm / h corresponds to a pressure adjustment amount of 0.4 MPa, and a grading strategy is generated in combination with the safety tolerance range: when the settlement rate is 2 - 5 mm / h, a pressure fluctuation of ±0.15 MPa is allowed, and the response time ≤ 10 s; when the settlement rate is 5 - 10 mm / h, the tolerance is expanded to ±0.25 MPa, and the response time ≤ 5 s. Finally, the gradient table is written into the formation characteristic map library. For example, in a certain silty sand formation, a settlement rate of 6 mm / h triggers a pressure adjustment to 0.8 ± 0.25 MPa.

[0119] Next, real-time monitoring and multi-level early warning are implemented through step 1053. The outlet pressure of the grouting pipe of 0.95 MPa and the face support pressure of 0.7 MPa are collected in real time through a pressure sensor, and the dynamic difference of 0.25 MPa is calculated. The displacement conduction rate of 6 mm / s at the key nodes in the stress diffusion topology network is synchronously monitored. When the dynamic difference exceeds the ±0.15 MPa tolerance allowed by the gradient table or the displacement rate breaks through the 5 mm / s safety threshold, a hierarchical early warning mechanism is triggered. For example, a dynamic difference of 0.25 MPa triggers an orange early warning, the system automatically starts the standby grouting pump and records the operation log. If the difference further rises to 0.36 MPa, a red early warning is triggered and the shield tunneling is stopped urgently.

[0120] Subsequently, linkage boost control is executed through step 1054. When the orange warning is triggered, the pressure of the target grouting hole No. 5 is increased to 1.2 times the reference value, i.e., 0.96 MPa. The distance weighting factor of 1.033 is calculated for the adjacent 3 hole positions with a spacing of 1.5 meters, and the pressure is adjusted to 0.826 MPa. After boosting, the surface displacement rebound amount of 3 mm (preset expectation of 5 mm) and the direction deviation angle of 12 degrees (allowable threshold of 30 degrees) are collected, and the rebound compensation weight 3 / 5 = 0.6 and the direction correction factor 1 - 2 / 30 ≈ 0.93 are calculated. The compensation coefficient 0.75 is generated by superimposing the weights of 0.6 and 0.4. For example, in a certain clay formation, the compensation coefficient of 0.8 triggers the pressure threshold to be extended to 0.55 - 1.0 MPa.

[0121] Finally, a closed-loop control link is formed through step 1055. The critical pressure threshold range is adjusted to 0.6 - 0.945 MPa according to the compensation coefficient of 0.75, and the updated parameters are bound to the formation characteristic atlas library. The grouting parameter adjustment instruction is sent through the industrial bus, with a target pressure of 0.96 MPa and a response time of 5 seconds, and the actuator is controlled to complete the pressure adjustment within 5 seconds. A monitoring cycle with a period of 30 seconds is started. For example, in a certain composite formation, the rebound amount data is collected every 30 seconds. If the rebound amount is less than 2 mm, boosting is triggered again, forming a continuously optimized closed-loop control process.

[0122] In practical applications, for example, in a certain cross-river shield tunnel project, based on the critical grouting pressure threshold of 1.8 - 2.3 MPa for the clay-cobble interlayer, combined with the settlement rate data of 0.5 - 1.2 mm / h collected by the surface displacement monitoring system, a three-level dynamic adjustment gradient table with a pressure fluctuation tolerance of ±0.15 MPa is generated using the gradient descent algorithm. The pressure difference data is obtained in real-time through the pressure transmitters installed at the outlet of the grouting pipe and the tunnel face. When the shield advances to the formation interface, it is monitored that the pressure difference suddenly increases from 0.6 MPa to 1.05 MPa. At the same time, the stress diffusion topology network shows that the displacement conduction rate in the interface area reaches 3.8 mm / min, exceeding the preset threshold of 2.5 mm / min, triggering a secondary orange warning. The system immediately activates the linkage pressurization device for the No. 3 main grouting hole and its adjacent No. 2 and No. 4 hole positions, and stepwise increases the grouting pressure from 1.9 MPa to 2.25 MPa within 12 seconds. Through the inclination sensor, it is monitored that the deviation angle of the stress diffusion direction is corrected from 28 degrees to 15 degrees, and the surface rebound amount reaches 0.7 mm. Based on the ratio of the displacement rebound amount of 0.7 mm to the preset expected value of 1.0 mm, the displacement compensation weight of 0.7 is calculated. Combining with the difference ratio of the direction deviation angle of 0.43, the direction correction factor of 0.85 is calculated. Through weighted superposition, the grouting pressure compensation coefficient of 0.79 is generated. Finally, the corrected critical pressure threshold is adjusted to 2.05 - 2.18 MPa and updated to the clay-cobble interlayer unit in the formation characteristic atlas library, forming a multi-parameter closed-loop control link of grouting pressure - surface displacement - stress diffusion with a period of 0.5 seconds, achieving the engineering effect of increasing the shrinkage rate of the formation disturbance range and reducing the loss of grouting materials.

[0123] In the overall scheme of step 105 above, by establishing a real-time mapping mechanism between the dynamic parameters of disturbance propagation and the critical grouting pressure range, combined with the multi-source feedback of surface displacement monitoring data and the dynamic difference of grouting pressure, an adaptive closed-loop control system including linkage pressurization compensation and stress diffusion correction is constructed; based on the critical grouting pressure threshold range in the formation interface transition zone, a dynamic adjustment gradient table is generated by synchronizing the reference pressure value of grouting and the instantaneous change of the surface settlement rate, and multiple warning signals are triggered using the dual criteria of exceeding the limit of the grouting pressure difference and breaking through the threshold of the displacement conduction rate; relying on the linkage pressurization operation of the target grouting hole and its adjacent hole positions, the surface displacement rebound amount and the stress diffusion direction deviation parameters are collected in real-time, and the grouting pressure compensation coefficient is generated through the weighted superposition of the rebound compensation weight and the direction deviation correction factor, dynamically correcting the critical pressure threshold range and optimizing the adjustment instructions; by real-time binding the compensation coefficient with the formation characteristic atlas library, a closed-loop control link of stress diffusion path topology and surface displacement feedback is formed, realizing the dynamic adaptation of grouting pressure and formation response, significantly improving the disturbance warning accuracy of multi-formation interfaces and the regulation efficiency of grouting parameters, while reducing the risk of formation mutations caused by shield tunneling and ensuring the adaptive regulation ability of the construction process.

[0124] The following is a complete embodiment for steps 101 to 105:

[0125] As Figure 2 shown, in a shield tunneling project of a certain urban subway, 12 groups of triaxial acceleration sensor arrays arranged on the cutter head surface are used to collect vibration spectrum data during the cutter head cutting process in real time at a sampling frequency of 4000 Hz. When the shield machine advances at a rotation speed of 18 r / min, the monitoring system synchronously obtains the vibration main frequency band energy distribution in the 3 o'clock, 6 o'clock, and 9 o'clock zones of the cutter head. Among them, the energy ratio of the 180 - 450 Hz frequency band in the clay layer reaches 62%, and there is a sudden increase in high-frequency energy in the 850 - 1300 Hz frequency band in the gravel layer section, with the maximum vibration acceleration peak reaching 12.3 m / s², providing high-precision original data for subsequent formation characteristic analysis.

[0126] In the shield tunneling section of the composite formation, based on the measured data of the plastic index of 22.3 in the clay layer and the proportion of 65% of particles with a particle size of 0.5 to 20 mm in the gravel layer, a characteristic map library containing 325 formation units is constructed. Through 3D geological modeling technology, the dynamic range of the clay shear modulus is set to 30 - 60 MPa, and the shear modulus of the gravel layer is 90 - 140 MPa, and the stratified data of the permeability coefficient is associated (clay layer 8×10^-7 cm / s, gravel layer 3×10^-3 cm / s). Combining with the historical data of the synchronous grouting system, the vibration characteristic fingerprints of typical formation combinations are generated. Among them, the critical grouting pressure range corresponding to the clay-gravel interlayer is 1.7 - 2.2 MPa, and the main frequency band of the vibration energy is concentrated in the 200 - 700 Hz range, establishing a standardized comparison benchmark for formation identification.

[0127] In the shield construction of the cross-river tunnel, using the 0.2° resolution signal of the cutter head rotation angle encoder, the vibration spectrum data of 6 zones on the cutter head surface and the monitored values of the soil chamber pressure of 0.2 - 0.4 MPa are aligned in space and time. By constructing a spatio-temporal correlation matrix, a disturbance field distribution map containing an axial vibration energy gradient of 0.6 - 2.8 kJ / m² and a radial gradient of 0.4 - 1.9 kJ / m² is generated. Combining with the stress diffusion topological network analysis, the main stress conduction path with an extension length of 2.1 m in the 2 o'clock direction of the cutter head is identified, and its energy attenuation rate of 0.18 / s has a significant spatial correlation with the pressure fluctuation amplitude of 0.15 - 0.28 MPa.

[0128] In the shield tunneling section of the geological interface, the disturbance field distribution map containing the high-frequency energy mutation of 650 - 1300 Hz is input into the ResNet-34 feature matching network optimized by transfer learning. The waveform features of the vibration energy gradient curve are extracted through three layers of convolution kernels, and spectral matching is performed with the standard patterns in the formation feature map library to obtain a formation matching degree index of 0.83. At the same time, the node parameters of three transition zone paths in the stress diffusion topology network are corrected, and the path curvature radius is optimized from 1.2 m to 0.9 m. Finally, the identification result of the gravel-dominated formation is output, and the dynamic parameters of the inclination angle of the main disturbance propagation path of 32 degrees and the energy attenuation rate of 0.25 / s are extracted.

[0129] During the construction in the sensitive building section, based on the critical grouting pressure standard of 2.0 - 2.4 MPa for the gravel formation and combined with the settlement data of 0.3 - 0.9 mm / h collected by the surface displacement monitoring system, a three-level pressure adjustment gradient table is dynamically generated. When the difference between the outlet pressure of the grouting pipe and the face support pressure is detected to break through the threshold of 0.7 MPa, a yellow warning is triggered and the linkage pressurization of the No. 5 grouting hole group is activated, and the grouting pressure is increased from 1.9 MPa to 2.3 MPa within 8 seconds. Through the real-time feedback of the surface rebound of 0.5 mm and the correction amount of the stress diffusion direction offset angle of 12 degrees, a grouting compensation coefficient of 0.87 is calculated, and a closed-loop control link of grouting pressure - formation deformation with a period of 10 seconds is formed to effectively control the surface settlement within the warning threshold range.

[0130] Figure 3 The structure schematic diagram of a formation disturbance analysis system under the coupling action of shield construction is provided for the embodiment of this application, as Figure 2 shown, the system includes:

[0131] An acquisition module 31, which acquires the vibration spectrum data generated when the cutter head cuts the soil through the vibration sensors arranged on the surface of the shield machine cutter head;

[0132] A construction module 32, which constructs a formation feature map library based on the soil plasticity index of the clay formation and the particle size distribution parameters of the gravel formation obtained, and generates vibration feature fingerprints in combination with the soil shear modulus, the groundwater permeability coefficient, and the synchronous grouting pressure threshold. The vibration feature fingerprints include the vibration energy distribution patterns under different formation combinations and the corresponding critical grouting pressure intervals;

[0133] A generation module 33, which aligns the vibration spectrum data of different regions of the cutter head and the monitored values of the soil chamber pressure in space and time to generate a disturbance field distribution map, and the disturbance field distribution map includes the energy gradient distribution curve of the vibration field and the stress diffusion path topology of the soil body;

[0134] Correction module 34 inputs the disturbance field distribution map and the vibration characteristic fingerprint into the feature matching network optimized by transfer learning, so as to perform spectral shape matching between the energy gradient distribution curve and the vibration energy distribution pattern through the feature matching network, and at the same time correct the propagation path parameters of the formation interface transition zone according to the stress diffusion path topology, so as to output the recognition result of the current formation type and extract the disturbance propagation dynamic parameters;

[0135] Adjustment module 35 dynamically adjusts the synchronous grouting pressure threshold according to the mapping relationship between the disturbance propagation dynamic parameters and the critical grouting pressure interval in the vibration characteristic fingerprint, combines the surface displacement monitoring data to generate a hierarchical early warning signal and a grouting parameter adjustment instruction, and completes the closed-loop control of formation disturbance.

[0136] Figure 3 The described formation disturbance analysis system under the coupling action of shield tunneling construction can execute Figure 1 The described formation disturbance analysis method under the coupling action of shield tunneling construction in the illustrated embodiment, its implementation principle and technical effects will not be elaborated. For the formation disturbance analysis system under the coupling action of shield tunneling construction in the above embodiment, the specific ways for each module and unit to execute operations have been described in detail in the embodiment related to the method, and will not be elaborated here.

[0137] In a possible design, Figure 3 The formation disturbance analysis system in the illustrated embodiment under the coupling action of shield tunneling construction can be implemented as a computing device, such as Figure 4 shown, this computing device can include a storage component 41 and a processing component 42;

[0138] The storage component 41 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 42.

[0139] The processing component 42 is used for the above Figure 1 The formation disturbance analysis method in the illustrated embodiment under the coupling action of shield tunneling construction.

[0140] Among them, the processing component 42 can include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component can also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components for executing the above method.

[0141] The storage component 41 is configured to store various types of data to support the operation of the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0142] Of course, the computing device may also necessarily include other components, such as input / output interfaces, display components, communication components, etc.

[0143] The input / output interface provides an interface between the processing component and the peripheral interface module, and the above-mentioned peripheral interface module can be an output device, an input device, etc.

[0144] The communication component is configured to facilitate communication between the computing device and other devices in a wired or wireless manner, etc.

[0145] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. At this time, the computing device can refer to a cloud server, and the above-mentioned processing component, storage component, etc. can be basic server resources leased or purchased from a cloud computing platform.

[0146] The embodiment of the present application also provides a computer storage medium storing a computer program, and when the computer program is executed by a computer, it can implement the above Figure 1 shown embodiment of a method for analyzing formation disturbance under the coupling action of shield construction.

[0147] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0148] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0149] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0150] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for analyzing formation disturbance under the coupling action of shield tunneling construction, characterized in that Including: Collect vibration spectrum data generated when the cutter head cuts the soil through vibration sensors arranged on the surface of the shield machine cutter head; Based on the soil plasticity index of the clay stratum and the particle size distribution parameters of the gravel stratum obtained, construct a stratum characteristic atlas library, and combine the soil shear modulus, groundwater permeability coefficient and synchronous grouting pressure threshold to generate vibration characteristic fingerprints, where the vibration characteristic fingerprints include the vibration energy distribution patterns and corresponding critical grouting pressure intervals under different stratum combinations; Perform spatio-temporal alignment on the vibration spectrum data of different regions of the cutter head and the monitored values of the soil chamber pressure to generate a disturbance field distribution map, where the disturbance field distribution map includes the energy gradient distribution curve of the vibration field and the stress diffusion path topology of the soil body; Input the disturbance field distribution map and the vibration characteristic fingerprints into a feature matching network optimized by transfer learning, so as to perform spectral shape matching between the energy gradient distribution curve and the vibration energy distribution pattern through the feature matching network, and at the same time correct the propagation path parameters of the stratum interface transition area according to the stress diffusion path topology, so as to output the recognition result of the current stratum type and extract the disturbance propagation dynamic parameters; According to the mapping relationship between the disturbance propagation dynamic parameters and the critical grouting pressure interval in the vibration characteristic fingerprints, dynamically adjust the synchronous grouting pressure threshold in combination with the surface displacement monitoring data, generate a hierarchical warning signal and a grouting parameter adjustment instruction, and complete the closed-loop control of stratum disturbance.

2. The method according to claim 1, characterized in that, Perform spatio-temporal alignment on the vibration spectrum data of different regions of the cutter head and the monitored values of the soil chamber pressure to generate a disturbance field distribution map, where the disturbance field distribution map includes the energy gradient distribution curve of the vibration field and the stress diffusion path topology of the soil body, including: Based on the cutter head rotation angle encoding signal and the spatial position information of the vibration monitoring points, perform spatio-temporal alignment operations on the vibration spectrum data of each region on the surface of the cutter head and the soil chamber pressure monitoring values to generate a spatio-temporally synchronized vibration pressure joint dataset; Extract the vibration spectrum segment synchronized with the cutter head rotation phase and the sudden change interval of the soil chamber pressure from the vibration pressure joint dataset, and establish a spatio-temporal correlation mapping relationship between the vibration energy and the pressure change; According to the spatio-temporal correlation mapping relationship, calculate the cumulative increment of the vibration energy of each region in the cutter head tunneling section within consecutive rotation cycles, and generate an energy gradient distribution curve of the vibration representing the stratum disturbance intensity; Based on the spatial superposition relationship between the soil chamber pressure fluctuation amplitude and the vibration energy gradient distribution curve, identify the soil layer stress concentration area and calibrate the stress diffusion direction, and generate a stress diffusion topological network penetrating the soil layer; Fuse the vibration energy gradient distribution curve and the stress diffusion topological network to form a disturbance field distribution map including spatio-temporal correlation disturbance characteristics.

3. The method according to claim 1, wherein According to the mapping relationship between the disturbance propagation dynamic parameters and the critical grouting pressure interval in the vibration characteristic fingerprints, dynamically adjust the synchronous grouting pressure threshold in combination with the surface displacement monitoring data, generate a hierarchical warning signal and a grouting parameter adjustment instruction, and complete the closed-loop control of stratum disturbance, including: Based on the mapping relationship between the disturbance propagation dynamic parameters and the critical grouting pressure interval in the vibration characteristic fingerprints, determine the range of the critical grouting pressure threshold for the stratum interface transition area; Extract the synchronous grouting reference pressure value corresponding to the current formation combination from the formation characteristic map library, and generate a dynamic adjustment gradient table of grouting pressure including pressure fluctuation tolerance in combination with the instantaneous change amount of the ground settlement rate in the ground displacement monitoring data; Real-time collect the dynamic difference between the pressure at the grouting pipe outlet and the face support pressure, and synchronously monitor the displacement conduction rate of the formation interface in the stress diffusion topology network. When the dynamic difference exceeds the tolerance range of the grouting pressure dynamic adjustment gradient table or the displacement conduction rate breaks through the preset diffusion rate threshold, trigger multi-level warning signals according to the exceeding degree; Based on the multi-level warning signals, activate the linkage pressurization operation of the target grouting hole position and its adjacent hole positions, synchronously collect the ground displacement rebound amount and the stress diffusion direction offset parameter after pressurization, and calculate the grouting pressure compensation coefficient; According to the grouting pressure compensation coefficient, adjust the critical grouting pressure threshold range, generate a grouting parameter adjustment instruction and bind it to the corresponding formation combination of the formation characteristic map library, and form a closed-loop control link based on stress diffusion topology and ground displacement feedback.

4. The method according to claim 1, characterized in that Based on the obtained plastic index of the clay formation and the particle size distribution parameters of the gravel formation, construct a formation characteristic map library, and generate a vibration characteristic fingerprint in combination with the soil shear modulus, the groundwater permeability coefficient and the synchronous grouting pressure threshold. The vibration characteristic fingerprint includes the vibration energy distribution pattern and the corresponding critical grouting pressure interval under different formation combinations, including: Based on the obtained plastic index threshold of the clay formation and the particle size distribution parameters of the gravel formation, divide the formation units in combination with the spatial coupling relationship between the plastic index and the particle size distribution. Each formation unit includes a shear modulus change interval and a permeability coefficient stratification interval; Dynamically match the real-time torque fluctuation data of the screw conveyor with the shear modulus change interval of the formation unit, establish a real-time correlation rule between the mechanical load response and the soil shear modulus, and generate a compensation correction coefficient of the grouting pressure based on the dynamic difference between the pressure at the grouting pipe outlet and the face support pressure; Integrate the compensation correction coefficient, the permeability coefficient stratification interval and the real-time correlation rule according to the formation unit, and construct a formation characteristic map library including the shear modulus correlation factor, the grouting pressure compensation rule and the permeability influence parameter; Based on the plastic index weight, the particle size distribution weight and the grouting pressure gradient distribution of each unit in the formation characteristic map library, fuse and generate a vibration characteristic fingerprint including the vibration energy distribution pattern and the corresponding critical grouting pressure interval under different formation combinations.

5. The method according to claim 1, wherein Input the disturbance field distribution map and the vibration characteristic fingerprint into a feature matching network optimized by transfer learning, so as to perform spectral shape matching between the energy gradient distribution curve and the vibration energy distribution pattern through the feature matching network, and at the same time correct the propagation path parameters of the formation interface transition zone according to the stress diffusion path topology, so as to output the recognition result of the current formation type and extract the disturbance propagation dynamic parameters, including: Input the vibration energy gradient distribution curve in the disturbance field distribution map and the vibration energy distribution patterns of different formation combinations in the vibration characteristic fingerprint into the feature matching network optimized by transfer learning. Through the feature matching network, perform spectral shape matching on the frequency band energy ratio and waveform similarity of the energy distribution curve to generate a formation type matching degree index; Extract the propagation path node parameters in the formation interface transition zone of the stress diffusion path topology, and based on the stress conduction direction offset angle and connection strength between nodes, correct the propagation path node parameters in the transition zone; Associate the corrected propagation path node parameters with the formation type matching degree index, extract the critical grouting pressure interval corresponding to the current formation combination from the vibration characteristic fingerprint, and generate a dynamic parameter set for disturbance propagation in the formation interface transition zone; According to the path offset and energy attenuation rate in the dynamic parameter set, output the recognition result of the current formation type and extract the disturbance propagation dynamic parameters.

6. The method according to claim 2, characterized in that, Based on the spatial superposition relationship between the soil chamber pressure fluctuation amplitude and the vibration energy gradient distribution curve, identify the soil layer stress concentration area and calibrate the stress diffusion direction to generate a stress diffusion topology network penetrating the soil layer, including: According to the spatial distribution consistency between the soil chamber pressure fluctuation amplitude and the vibration energy gradient distribution curve in the cutter head tunneling section, calculate the pressure energy superposition coefficient representing the combined action intensity of the pressure amplitude and energy gradient in each area; Screen the areas where the pressure energy superposition coefficient exceeds the preset threshold, mark them as initial stress concentration areas, and extract the energy gradient change slope between adjacent initial stress concentration areas; Based on the directivity of the energy gradient change slope, calibrate the stress diffusion direction between the initial stress concentration areas, and merge adjacent areas with the same direction into stress conduction links; According to the attenuation rate of the pressure energy superposition coefficient and the link extension length in the stress conduction link, generate a stress diffusion topology network penetrating the soil layer.

7. The method according to claim 3, wherein Based on the multi-level warning signals, activate the linkage pressurization operation of the target grouting hole position and its adjacent hole positions, synchronously collect the surface displacement rebound amount and stress diffusion direction offset parameters after pressurization, and calculate the grouting pressure compensation coefficient, including: According to the level of the multi-level warning signals, activate the linkage pressurization device of the target grouting hole and its adjacent hole positions according to the preset priority, and synchronously increase the grouting pressure of the target hole and the adjacent hole to the preset pressurization value; Synchronously collect the surface displacement rebound amount and stress diffusion direction offset angle data in the pressurized area. The surface displacement rebound amount is the difference between the surface uplift height after pressurization and the settlement amount before pressurization, and the stress diffusion direction offset angle is the angle deviation between the actual conduction direction and the preset diffusion path; Based on the ratio of the absolute value of the surface displacement rebound amount to the preset rebound expected value, calculate the displacement rebound compensation weight, and at the same time, based on the difference ratio between the stress diffusion direction offset angle and the allowable offset angle threshold, calculate the direction offset correction factor; Perform weighted superposition on the displacement rebound compensation weight and the direction offset correction factor to generate a grouting pressure compensation coefficient for quantifying the correction effect of the linkage pressurization operation on the formation disturbance suppression.

8. A formation disturbance analysis system under the coupling action of shield tunneling construction, characterized in that Including: The acquisition module collects the vibration spectrum data generated when the cutter head cuts the soil through the vibration sensors arranged on the surface of the shield machine cutter head; The construction module constructs a formation feature map library based on the soil plasticity index of the clay formation and the particle size distribution parameters of the gravel formation obtained, combines the soil shear modulus, the groundwater permeability coefficient and the synchronous grouting pressure threshold value to generate a vibration feature fingerprint, and the vibration feature fingerprint includes the vibration energy distribution pattern and the corresponding critical grouting pressure interval under different formation combinations; The generation module performs spatio-temporal alignment on the vibration spectrum data of different regions of the cutter head and the monitored values of the soil chamber pressure to generate a disturbance field distribution map, and the disturbance field distribution map includes the energy gradient distribution curve of the vibration field and the topological structure of the stress diffusion path of the soil body; The correction module inputs the disturbance field distribution map and the vibration feature fingerprint into a feature matching network optimized by transfer learning, so as to perform spectral shape matching between the energy gradient distribution curve and the vibration energy distribution pattern through the feature matching network, and at the same time correct the propagation path parameters of the formation interface transition zone according to the stress diffusion path topology, so as to output the recognition result of the current formation type and extract the disturbance propagation dynamic parameters; The adjustment module dynamically adjusts the synchronous grouting pressure threshold value according to the mapping relationship between the disturbance propagation dynamic parameters and the critical grouting pressure interval in the vibration feature fingerprint, combines the surface displacement monitoring data to generate a hierarchical warning signal and a grouting parameter adjustment instruction, and completes the closed-loop control of the formation disturbance.

9. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a method for analyzing formation disturbance under the coupling action of shield construction as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, A computer program is stored, and when the computer program is executed by the computer, a method for analyzing formation disturbance under the coupling action of shield construction as described in any one of claims 1 to 7 is implemented.

Images