A method and system for evaluating the stability of underwater tunnel shield excavation face

Abstract

The present invention relates to the technical field of tunnel shield construction, and discloses a method and system for evaluating the stability of an excavation face of an underwater tunnel shield excavation, the method comprising the following steps: collecting monitoring data of the excavation face and correcting the monitoring data; simulating the stability of the excavation face under different working conditions, identifying key influencing factors and potential risk points; identifying the geological structure characteristics of the excavation face, and evaluating the stability of the excavation face in combination with the acoustic wave velocity in the acoustic wave data; evaluating the stability of the excavation face in combination with the real-time monitoring data of the evaluation index; calculating the comprehensive risk score of the excavation face based on the results of the simulation, acoustic wave and index evaluation, and determining the risk level of the excavation face according to the comprehensive risk score. The present invention can realize real-time monitoring and response to changes in the excavation face, adapt to complex geological conditions, and comprehensively consider multiple factors to provide a more accurate and real-time stability evaluation, which is helpful to improve the accuracy and real-time performance of the excavation face stability evaluation during underwater tunnel shield excavation.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel shield construction, and in particular to a method and system for evaluating the stability of an underwater tunnel shield excavation face. Background Art

[0002] During underwater tunneling, the stability of the excavation face is crucial. The instability of the excavation face may lead to the destruction of the tunnel structure, construction accidents and environmental impacts. Traditional stability assessment methods rely on empirical formulas and on-site monitoring, which often cannot reflect the dynamic changes of the excavation face in real time and have limited adaptability to complex geological conditions.

[0003] During underwater tunneling, the stability assessment of the excavation face is a complex and challenging task. The stability of the excavation face is directly related to the safety and success of the entire tunnel project. If the excavation face is unstable, it may cause damage to the tunnel structure, such as tunnel collapse, cracks, deformation, etc., seriously affecting the service life and safety of the tunnel. In addition, the instability of the excavation face may also lead to construction accidents, such as casualties, equipment damage, etc., causing huge economic losses and social impacts to the project. At the same time, the instability of the excavation face may also have a negative impact on the surrounding environment, such as groundwater pollution, ground subsidence, etc.

[0004] Traditional stability assessment methods mainly rely on empirical formulas and on-site monitoring. These methods are often based on historical data and expert experience, and evaluate the stability of the excavation face by establishing empirical formulas. However, these methods often fail to reflect the dynamic changes of the excavation face in real time because they cannot monitor and respond to changes in the excavation face in real time. In addition, these methods have limited adaptability to complex geological conditions. Complex geological conditions, such as soft soil, rock, groundwater, etc., will have a significant impact on the stability of the excavation face, and these traditional methods often fail to fully consider these complex factors.

[0005] Therefore, a new method is urgently needed to evaluate the stability of the excavation face during underwater tunnel shield excavation. Therefore, the present invention proposes a method and system for evaluating the stability of the excavation face during underwater tunnel shield excavation. Summary of the invention

[0006] In view of the problems in the related art, the present invention proposes a method and system for evaluating the stability of an underwater tunnel shield excavation face to overcome the above-mentioned technical problems existing in the existing related art.

[0007] To this end, the specific technical solution adopted by the present invention is as follows:

[0008] According to one aspect of the present invention, a method for evaluating the stability of an underwater tunnel shield excavation face is provided, and the method for evaluating the stability of an underwater tunnel shield excavation face comprises the following steps:

[0009] S1. Collect monitoring data of the underwater tunnel shield excavation face, and identify and correct abnormal data in the monitoring data based on the excavation data correction algorithm;

[0010] S2. Construct a numerical model based on the excavation surface and set the model parameters; simulate the stability of the excavation surface under different working conditions based on the numerical model, and identify key influencing factors and potential risk points based on the simulation results;

[0011] S3, based on the acoustic wave data in the corrected monitoring data, identifying the geological structure characteristics of the excavation surface, and evaluating the stability of the excavation surface in combination with the acoustic wave velocity in the acoustic wave data;

[0012] S4. Determine the evaluation index affecting the stability of the excavation surface based on the identified key influencing factors and potential risk points, and evaluate the stability of the excavation surface in combination with the real-time monitoring data of the evaluation index;

[0013] S5. Based on the results of simulation, acoustic wave and index evaluation, the comprehensive risk score of the excavation surface is calculated in combination with the preset weights, and the risk level of the excavation surface is determined based on the comprehensive risk score.

[0014] Furthermore, the collecting of monitoring data of the underwater tunnel shield excavation face and identifying and correcting abnormal data in the monitoring data based on the excavation data correction algorithm includes the following steps:

[0015] S11, collecting monitoring data of the underwater tunnel shield excavation face, and sequentially performing data cleaning, normalization and data labeling processing on the monitoring data to obtain a data set, wherein the monitoring data includes geological and soil parameters, groundwater parameters, structural response data, environmental parameters and acoustic wave monitoring data;

[0016] S12. Using the Bayesian optimization algorithm to solve the optimal combination parameters of the bidirectional long short-term memory model based on the convolutional neural network, and training the bidirectional long short-term memory model based on the convolutional neural network in combination with the data set;

[0017] S13, obtaining the monitoring data of the excavation surface at the previous moment, and using the trained bidirectional long short-term memory model based on the convolutional neural network to output the predicted monitoring data at the current moment corresponding to the monitoring data at the previous moment;

[0018] S14. Calculate the difference between the real-time monitoring data at the current moment and the predicted monitoring data, and determine whether the difference exceeds the preset difference threshold. If not, determine that the real-time monitoring data is normal data. If so, determine that the real-time monitoring data is abnormal data, and use the predicted monitoring data at the current moment as the corrected real-time monitoring data.

[0019] Furthermore, the method of solving the optimal combination parameters of the bidirectional long short-term memory model based on the convolutional neural network by using the Bayesian optimization algorithm and training the bidirectional long short-term memory model based on the convolutional neural network in combination with the data set includes the following steps:

[0020] S121, setting the hyperparameters to be optimized and defining a search space for each hyperparameter; initializing the Bayesian optimizer and setting initial hyperparameter values;

[0021] S122, constructing a bidirectional long short-term memory model based on a convolutional neural network using the initial hyperparameter group obtained from the Bayesian optimizer, and training and verifying the bidirectional long short-term memory model based on the convolutional neural network using the data set; based on the verification result, the Bayesian optimizer updates the probability distribution of the hyperparameters;

[0022] S123. Repeat S122 until the maximum number of iterations, the time limit, or the performance indicator reaches a preset threshold; according to the final result of the Bayesian optimizer, select the best hyperparameter combination, and use the best hyperparameter combination to reconstruct and train the bidirectional long short-term memory model based on the convolutional neural network.

[0023] Furthermore, the construction of a numerical model based on the excavation surface, setting model parameters; simulating the stability of the excavation surface under different working conditions based on the numerical model, and identifying key influencing factors and potential risk points according to the simulation results includes the following steps:

[0024] S21. constructing a three-dimensional model according to the geometric shape and size of the actual excavation surface, and meshing the three-dimensional model to generate a finite element mesh or a discrete element mesh;

[0025] S22. Based on the geological survey report, the soil in the three-dimensional model is assigned corresponding physical and mechanical parameters, and the supporting structure is assigned corresponding material properties;

[0026] S23. Setting the boundary conditions and initial stress field of the three-dimensional model according to the actual geological conditions and construction parameters, and simulating the excavation surface stability under normal working conditions and extreme working conditions;

[0027] S24. Use finite element analysis to simulate the excavation process, record relevant indicators of excavation surface stability, and analyze simulation results to identify key influencing factors and potential risk points of excavation surface stability;

[0028] S25. Conduct sensitivity analysis on key influencing factors to evaluate their impact on the stability of the excavation face, and evaluate the risk level of the excavation face based on the simulation results and sensitivity analysis results.

[0029] Furthermore, the identifying of geological structural features of the excavation surface based on the acoustic wave data in the corrected monitoring data and evaluating the stability of the excavation surface in combination with the acoustic wave velocity in the acoustic wave data includes the following steps:

[0030] S31, obtaining the sound wave data in the corrected monitoring data, calculating the sound wave velocity according to the sound wave propagation time, and analyzing the change trend of the sound wave velocity;

[0031] S32. Analyze the structural characteristics of geological bodies based on the reflection and refraction characteristics of sound waves and identify potential geological defects that affect the stability of the excavation surface;

[0032] S33, inverting the physical parameters of the geological body through the acoustic wave velocity, and constructing an acoustic wave risk index based on the acoustic wave data and related geological parameters, wherein the acoustic wave risk index includes the acoustic wave velocity change rate and the geological defect density;

[0033] S34. Use the analytic hierarchy process to assign weights to the acoustic wave risk indicators, calculate the risk scores based on the acoustic wave risk indicators and the corresponding weights, and evaluate the risk level of the excavation surface based on the risk scores.

[0034] Furthermore, determining the evaluation index affecting the stability of the excavation surface according to the identified key influencing factors and potential risk points, and evaluating the stability of the excavation surface in combination with the real-time monitoring data of the evaluation index includes the following steps:

[0035] S41. Determine the evaluation indicators affecting the stability of the excavation surface based on key influencing factors and potential risk points;

[0036] S42. Use the analytic hierarchy process to assign weights to the evaluation indicators, and construct a risk assessment model based on the evaluation indicators and weights;

[0037] S43, obtaining real-time monitoring data related to the evaluation indicators, and calculating the score of each evaluation indicator according to the risk assessment model;

[0038] S44, multiplying the score of each evaluation indicator by its corresponding weight to obtain a weighted score, and adding the weighted scores of all evaluation indicators to obtain a total risk score of the excavation surface;

[0039] S45. Evaluate the risk level of the excavation surface based on the total risk score of the excavation surface.

[0040] Furthermore, the calculation of the comprehensive risk score of the excavation face based on the results of simulation, acoustic wave and index evaluation in combination with preset weights, and the determination of the risk level of the excavation face according to the comprehensive risk score includes the following steps:

[0041] S51. Define key criteria for assessing risk levels of excavation faces, including simulation-based risk levels, acoustic-based risk levels, and indicator-based risk levels;

[0042] S52, respectively obtaining the risk level assessment results based on simulation, the risk level assessment results based on sound waves, and the risk level assessment results based on indicators, and performing standardization processing on the assessment results;

[0043] S53, using an entropy weight method to respectively determine the weights of the risk level based on simulation, the risk level based on sound waves, and the risk level based on indicators;

[0044] S54, constructing a decision matrix based on the acquired data, wherein the rows in the decision matrix represent different key criteria, and the columns represent different risk level assessment results;

[0045] S55. Calculate the comprehensive risk level of each risk level assessment result using the weighted average method, and determine the final risk level of the excavation surface based on the ranking results of the comprehensive risk levels and the risk threshold.

[0046] Furthermore, the method of using the entropy weight method to respectively determine the weights of the risk level based on simulation, the risk level based on sound waves, and the risk level based on indicators includes the following steps:

[0047] S531, calculating the information entropy of each key criterion, and determining the difference coefficient according to the difference between the information entropy of each key criterion and the maximum information entropy;

[0048] S532. Calculate the weight of each key criterion according to the difference coefficient, and normalize the calculated weight.

[0049] Furthermore, the calculation formula of the information entropy is:

[0050]

[0051] The weight calculation formula of key criteria is:

[0052]

[0053] The calculation formula of the normalized weight is:

[0054]

[0055] The calculation formula for the comprehensive risk level is:

[0056]

[0057] Where H(X) represents the information entropy of random variable X; n represents the number of key criteria; x i represents the value of the i-th key criterion; p(x i ) represents the probability of the i-th key criterion; ω irepresents the weight of the i-th key criterion; ω j represents the weight of the jth key criterion; ω i ′ represents the normalized weight of key criterion i; d i represents the coefficient of difference of the i-th key criterion; d j represents the coefficient of difference of the jth key criterion; R represents the comprehensive risk level; r i represents the score of the i-th key criterion.

[0058] According to another aspect of the present invention, there is provided an underwater tunnel shield excavation face stability evaluation system, comprising a data acquisition correction module, a simulation-based risk level evaluation module, an acoustic-wave-based risk level evaluation module, an indicator-based risk level evaluation module, and a comprehensive risk level evaluation module;

[0059] The data acquisition and correction module is used to collect monitoring data of the underwater tunnel shield excavation face, and identify and correct abnormal data in the monitoring data based on the excavation data correction algorithm;

[0060] The simulation-based risk level assessment module is used to construct a numerical model based on the excavation surface and set model parameters; simulate the stability of the excavation surface under different working conditions based on the numerical model, and identify key influencing factors and potential risk points based on the simulation results;

[0061] The acoustic wave-based risk level assessment module is used to identify the geological structure characteristics of the excavation surface based on the acoustic wave data in the corrected monitoring data, and to assess the stability of the excavation surface in combination with the acoustic wave velocity in the acoustic wave data;

[0062] The indicator-based risk level assessment module is used to determine the assessment indicators that affect the stability of the excavation surface according to the identified key influencing factors and potential risk points, and to assess the stability of the excavation surface in combination with the real-time monitoring data of the assessment indicators;

[0063] The comprehensive risk level assessment module is used to calculate the comprehensive risk score of the excavation surface based on the results of simulation, acoustic wave and index assessment combined with preset weights, and determine the risk level of the excavation surface according to the comprehensive risk score.

[0064] The beneficial effects of the present invention are:

[0065] 1) The present invention can not only use the correction algorithm based on the excavation data to identify and correct abnormal data in the monitoring data, but also construct a numerical model to simulate the stability of the excavation face under different working conditions, and use the acoustic wave data to identify the geological structure characteristics, and evaluate the stability of the excavation face in combination with the acoustic wave velocity. At the same time, the evaluation index can be determined according to the key influencing factors and potential risk points identified, and the stability of the excavation face can be evaluated in combination with the real-time monitoring data of the evaluation index, so that the comprehensive risk score of the excavation face can be calculated based on the results of simulation, acoustic wave and index evaluation, combined with preset weights, and the final risk level of the excavation face can be determined according to the comprehensive risk score. Compared with the traditional evaluation method, the present invention can realize real-time monitoring and response to the changes in the excavation face, adapt to complex geological conditions, and comprehensively consider multiple factors to provide more accurate and real-time stability evaluation, which is helpful to improve the accuracy and real-time performance of the excavation face stability evaluation during the shield excavation of underwater tunnels, thereby effectively ensuring the construction safety of the project and improving the construction quality and efficiency.

[0066] 2) The present invention can not only use the Bayesian optimization algorithm to optimize the bidirectional long short-term memory model based on the convolutional neural network, thereby effectively improving the performance of the model in the abnormality recognition task, but also use the optimized model to automatically identify and correct abnormal data, significantly improving the accuracy and reliability of the monitoring data, thereby ensuring that the monitoring data can timely reflect the dynamic changes of the excavation face, and providing a strong technical guarantee for the stability assessment of the excavation face during the shield excavation of underwater tunnels. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0068] Figure 1 The present invention is a flowchart of a method for evaluating the stability of an underwater tunnel shield excavation face according to an embodiment of the present invention. DETAILED DESCRIPTION

[0069] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in the field should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0070] According to an embodiment of the present invention, a method and system for evaluating the stability of an underwater tunnel shield excavation face are provided.

[0071] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to one embodiment of the present invention, a method for evaluating the stability of an underwater tunnel shield excavation face is provided, and the method for evaluating the stability of an underwater tunnel shield excavation face comprises the following steps:

[0072] S1. Collect monitoring data of the underwater tunnel shield excavation face, and identify and correct abnormal data in the monitoring data based on the excavation data correction algorithm;

[0073] The method of collecting monitoring data of the underwater tunnel shield excavation face and identifying and correcting abnormal data in the monitoring data based on the excavation data correction algorithm includes the following steps:

[0074] S11, collecting monitoring data of the underwater tunnel shield excavation face, and sequentially performing data cleaning, normalization and data labeling on the monitoring data (labeling the monitoring data based on professional knowledge to distinguish normal data from abnormal data) to obtain a data set;

[0075] Specifically, the monitoring data include:

[0076] 1) Geological and soil parameters: physical and mechanical properties of soil, such as density, cohesion, internal friction angle, etc.; stratification of strata, including the thickness and properties of different strata;

[0077] 2) Groundwater parameters: changes in groundwater level, groundwater flow speed and direction;

[0078] 3) Structural response data: displacement, stress, strain and other data during the advancement of the shield machine; stress, strain and deformation data of the support structure (such as segments, anchors, etc.);

[0079] 4) Environmental parameters: groundwater pressure near the excavation surface, soil pressure near the excavation surface;

[0080] 5) Acoustic wave monitoring data: acoustic wave propagation time, waveform, amplitude, etc.; acoustic wave velocity and its changes, used to assess geological structure and stability;

[0081] These monitoring data are collected in real time by sensors and monitoring equipment installed on the shield machine, excavation face, support structure and surrounding environment. By analyzing these data, the stability of the excavation face can be evaluated and corresponding measures can be taken to ensure construction safety.

[0082] S12. Using the Bayesian optimization algorithm to solve the optimal combination parameters of the bidirectional long short-term memory model based on the convolutional neural network, and training the bidirectional long short-term memory model based on the convolutional neural network in combination with the data set;

[0083] Specifically, the method of solving the optimal combination parameters of the bidirectional long short-term memory model based on the convolutional neural network by using the Bayesian optimization algorithm and training the bidirectional long short-term memory model based on the convolutional neural network in combination with the data set includes the following steps:

[0084] S121, set the hyperparameters to be optimized, such as learning rate, batch size, number of hidden layer units, number of filters, convolution kernel size, etc., and define a search space for each hyperparameter, which should cover all possible parameter values; initialize the Bayesian optimizer and set the initial hyperparameter values;

[0085] S122. Construct a bidirectional long short-term memory model based on a convolutional neural network using the initial hyperparameter group obtained from the Bayesian optimizer (the convolutional neural network layer is used to extract features, and the bidirectional long short-term memory network layer is used to process time series data), and use the data set to train and verify the bidirectional long short-term memory model based on the convolutional neural network; based on the verification results, the Bayesian optimizer updates the probability distribution of the hyperparameters;

[0086] Specifically, the training and verification of the bidirectional long short-term memory model based on the convolutional neural network using the data set includes:

[0087] Train the model on the training set in the dataset; monitor the loss function value during training to ensure that the model improves on the training data; evaluate the performance of the model on the validation set in the dataset, recording key indicators such as cross entropy loss, accuracy, etc.

[0088] S123. Repeat S122 until the maximum number of iterations, the time limit, or the performance indicator reaches a preset threshold; according to the final result of the Bayesian optimizer, select the best hyperparameter combination, and use the best hyperparameter combination to reconstruct and train the bidirectional long short-term memory model based on the convolutional neural network.

[0089] S13, obtaining the monitoring data of the excavation surface at the previous moment, and using the trained bidirectional long short-term memory model based on the convolutional neural network to output the predicted monitoring data at the current moment corresponding to the monitoring data at the previous moment;

[0090] S14, calculate the difference between the real-time monitoring data and the predicted monitoring data at the current moment, and determine whether the difference exceeds the preset difference threshold. If not, the real-time monitoring data is determined to be normal data. If so, the real-time monitoring data is determined to be abnormal data, and the predicted monitoring data at the current moment is used as the corrected real-time monitoring data. Among them, the preset difference threshold is determined by historical data analysis and statistical methods, and enough historical monitoring data is collected, including real-time monitoring data and corresponding predicted monitoring data. The difference between the real-time monitoring data and the predicted monitoring data at each time point in the historical data is calculated, and the difference data set is statistically analyzed to calculate the mean (Mean, μ) and standard deviation (Standard Deviation, σ) of the difference. According to the mean and standard deviation, the difference threshold range is set. In general, the threshold range can be set to the mean plus or minus several times the standard deviation, such as set to μ±kσ, where k is a constant, usually 2 or 3, to ensure that most normal differences are covered.

[0091] S2. Construct a numerical model based on the excavation surface and set the model parameters; simulate the stability of the excavation surface under different working conditions based on the numerical model, and identify key influencing factors and potential risk points based on the simulation results;

[0092] The construction of a numerical model based on the excavation surface, setting model parameters, simulating the stability of the excavation surface under different working conditions based on the numerical model, and identifying key influencing factors and potential risk points according to the simulation results include the following steps:

[0093] S21. Construct a three-dimensional model according to the geometric shape and size of the actual excavation surface, including the geological model, shield machine model, support structure model, etc., and mesh the three-dimensional model to generate a finite element mesh or a discrete element mesh to ensure that the mesh quality meets the calculation requirements;

[0094] S22. Based on the geological survey report, the soil in the three-dimensional model is assigned corresponding physical and mechanical parameters, such as density, cohesion, internal friction angle, etc., and the supporting structure (such as segments, anchors, etc.) is assigned corresponding material properties;

[0095] In order to assign corresponding physical and mechanical parameters to the soil and supporting structure in the 3D model, the following steps are included:

[0096] 1) Geological survey report analysis

[0097] Understand geological conditions: Carefully analyze the geological survey report to understand the distribution of soil layers, rock properties, groundwater conditions, etc.

[0098] Identify soil types: Based on the geotechnical test data in the report, identify the types of different soil layers, such as sand, silt, clay, etc.

[0099] 2) Assignment of soil parameters

[0100] Density: Assign density parameters to the soil based on the type of soil layer and test data.

[0101] Cohesion: Based on the results of geotechnical tests, the soil is assigned a cohesion parameter, which reflects the shear strength of the soil.

[0102] Internal friction angle: Based on the results of geotechnical tests, the soil is assigned an internal friction angle parameter, which also reflects the shear strength of the soil.

[0103] Elastic modulus: The elastic modulus is assigned to the soil, which reflects the stiffness of the soil.

[0104] Poisson's ratio: The soil is assigned Poisson's ratio, which describes the relationship between the lateral strain and the longitudinal strain of the soil when subjected to stress.

[0105] 3) Assignment of support structure parameters

[0106] Segment density: The density of the segment is usually determined by the density of the material (such as concrete).

[0107] Segment elastic modulus: The segment elastic modulus reflects its ability to resist deformation.

[0108] Segment Poisson's ratio: The Poisson's ratio of the segment is usually consistent with the Poisson's ratio of the material.

[0109] Segment tensile strength and compressive strength: Assign corresponding strength parameters based on the segment design and material properties.

[0110] Anchor rod elastic modulus: The elastic modulus of the anchor rod is usually determined by the material of the rod body (such as steel).

[0111] Anchor tensile strength: The tensile strength of an anchor reflects its ability to withstand tensile loads.

[0112] Anchor bond strength: The bond strength between the anchor and the surrounding soil reflects the interaction between the anchor and the soil.

[0113] 4) Parameter verification: Ensure that the assigned parameter values ​​are consistent with relevant engineering specifications and standards. If necessary, consult geological and structural engineering experts to verify and adjust the parameters.

[0114] 5) Parameter input: Input the above determined parameters into the three-dimensional numerical model to assign corresponding material properties to the soil and support structure.

[0115] Through the above steps, it can be ensured that the parameter settings of the soil and support structure in the numerical model are reasonable, thereby improving the accuracy and reliability of the model analysis.

[0116] S23. Set the boundary conditions (such as fixed boundary, displacement boundary or stress boundary) and initial stress field (such as self-weight stress field) of the three-dimensional model according to the actual geological conditions and construction parameters, and simulate the excavation face stability under normal working conditions and extreme working conditions, such as high water pressure, earthquake, etc.;

[0117] In order to set the boundary conditions and initial stress field in the 3D model and simulate the excavation face stability under different working conditions, the following steps are included:

[0118] 1) Determine boundary conditions

[0119] Fixed Boundaries: Apply fixed supports to the sides and bottom of the model to prevent these boundaries from moving during the simulation. This is often used to simulate soils far from the excavation area, which are less affected during construction.

[0120] Displacement boundaries: Apply known displacement values ​​to certain boundaries of the model to simulate the influence of supporting structures or adjacent areas in actual engineering. For example, when simulating tunnel excavation, the soil on both sides of the tunnel may be subject to certain displacement restrictions.

[0121] Stress boundary: Stress values ​​are applied on certain boundaries of the model to simulate load conditions in actual engineering, such as ground load, building load, etc.

[0122] 2) Set the initial stress field

[0123] Self-weight stress field: The self-weight stress of the soil is calculated based on the density and gravity acceleration of the soil as the initial stress field of the simulation. The self-weight stress field is usually calculated based on the gravity of the soil and the thickness of the soil layer.

[0124] 3) Simulate normal working conditions

[0125] Excavation simulation: Remove soil gradually in the model to simulate the actual excavation process.

[0126] Support application: Apply appropriate support structures during the excavation process, such as segments, anchors, shotcrete, etc.

[0127] 4) Simulate extreme working conditions

[0128] High Water Pressure: Apply additional water pressure to the model to simulate the stability of the excavation face under high water pressure conditions. This can be achieved by applying hydrostatic pressure on the model boundaries.

[0129] Earthquake: To simulate the stability of the excavation face under earthquake action, dynamic loads are usually applied. Earthquake dynamic analysis (such as time history analysis) can be used to simulate the propagation of seismic waves and the dynamic response of the soil.

[0130] 5) Simulation steps

[0131] Model building: Build a three-dimensional model in numerical simulation software and divide the mesh.

[0132] Input parameters: Assign physical and mechanical parameters to soil and support structure.

[0133] Set boundary conditions and initial stress field: Set boundary conditions and initial stress field according to actual geological conditions and construction parameters.

[0134] Run the simulation:

[0135] Normal working conditions: simulate the excavation and support process.

[0136] Extreme working conditions: simulate extreme conditions such as high water pressure and earthquake.

[0137] Result analysis: Analyze the simulation results and evaluate the stability of the excavation face.

[0138] 6) Result evaluation

[0139] Stability analysis: Based on the simulation results, the stability of the excavation face under different working conditions is evaluated.

[0140] Identification of key factors: Identify key factors that affect the stability of the excavation face, such as soil properties, support structure performance, external loads, etc.

[0141] Through the above steps, the stability of the excavation surface under different working conditions can be effectively simulated and evaluated, providing an important reference for actual construction.

[0142] S24. Use finite element analysis to simulate the excavation process, record relevant indicators of excavation surface stability, such as displacement, stress, plastic zone, etc., and analyze the simulation results to identify key influencing factors of excavation surface stability (such as soil parameters, groundwater conditions, support structure performance, etc.) and potential risk points (such as areas where excessive displacement or stress concentration may occur);

[0143] S25. Conduct sensitivity analysis on key influencing factors to evaluate their impact on the stability of the excavation face, and evaluate the risk level of the excavation face based on the simulation results and sensitivity analysis results.

[0144] To conduct sensitivity analysis on key influencing factors and evaluate their influence on the stability of the excavation face, as well as to evaluate the risk level of the excavation face based on the simulation results and sensitivity analysis results, the following steps can be followed:

[0145] 1) Conduct sensitivity analysis

[0146] Single factor sensitivity analysis: Change the values ​​of key influencing factors one by one and observe their influence on the stability of the excavation face.

[0147] Multi-factor sensitivity analysis: Change the values ​​of multiple key influencing factors simultaneously and observe their combined effects on the stability of the excavation face.

[0148] 2) Sensitivity analysis method

[0149] Local sensitivity analysis: By changing the value of a single factor, analyze its direct impact on the results.

[0150] Global sensitivity analysis: Analyze the overall impact of all factors on the results by changing the values ​​of multiple factors.

[0151] 3) Evaluation of sensitivity analysis results

[0152] Impact assessment: Based on the results of sensitivity analysis, evaluate the impact of each key factor on the stability of the excavation face.

[0153] Sensitivity ranking: Rank key factors according to their impact on stability.

[0154] 4) Risk level assessment

[0155] Determination of risk indicators: Based on the results of sensitivity analysis, determine the main risk indicators that affect the stability of the excavation face.

[0156] Risk level classification: According to the size of the risk index, the excavation surface is divided into different risk levels.

[0157] Risk assessment: Combine the simulation results and sensitivity analysis results to conduct a comprehensive assessment of the risk level of the excavation surface.

[0158] Through sensitivity analysis and risk level assessment, we can have a deeper understanding of the key influencing factors of excavation face stability and provide a scientific basis for risk management during the construction process.

[0159] In addition to the above methods of identifying key influencing factors and potential risk points, they can also be identified in the following ways:

[0160] Define basic parameters and influencing factors: soil type and properties: such as soil type (clay, sand, etc.), effective cohesion c′ and effective internal friction angle φ′; excavation depth H, which affects potential landslides and lateral pressure, is determined according to engineering requirements; water level hw, hw determines the pore water pressure and soil saturation, which is obtained by averaging the water level height from historical data; excavation slope β, β affects the self-weight stability of the soil, which is determined according to engineering safety standards. Correlation formula construction:

[0161] Safety factor formula:

[0162]

[0163] Where FS is the safety factor, σ v is the vertical stress, σ v =ρ*g*H, ρ is the soil density, which is determined according to the test results of the soil density testing equipment, g is the acceleration of gravity, u is the pore water pressure, u=ρw*g*hw, ρw is the density of water, hw is the water level.

[0164] Excavation influence coefficient formula:

[0165]

[0166] In the formula, H max The maximum excavation depth allowed for the corresponding soil type stored in the database is determined based on the construction requirements.

[0167] Comparison and analysis:

[0168] Comparing the safety factors and excavation influence factors calculated by the above formulas for different situations with the ideal safety factors and ideal excavation influence factors stored in the database can identify which parameters are most likely to cause stability problems. The specific steps include: calculating the difference between the actual value of each parameter and its ideal value. For example, if the actual calculated value of FS is 1.3 instead of the ideal 1.5, this indicates a potential stability problem. Systematically changing each parameter, such as increasing or decreasing the excavation depth H, adjusting the excavation slope β, or simulating water level changes hw, to observe how these changes affect FS can help determine which factors have the most significant impact on stability. The ideal safety factor and ideal excavation influence factor are determined by industry standards.

[0169] S3, based on the acoustic wave data in the corrected monitoring data, identifying the geological structure characteristics of the excavation surface, and evaluating the stability of the excavation surface in combination with the acoustic wave velocity in the acoustic wave data;

[0170] The method of identifying the geological structure characteristics of the excavation surface based on the acoustic wave data in the corrected monitoring data and evaluating the stability of the excavation surface in combination with the acoustic wave velocity in the acoustic wave data includes the following steps:

[0171] S31, obtain the sound wave data in the revised monitoring data, calculate the sound wave velocity according to the sound wave propagation time, and analyze the change trend of the sound wave velocity (statistical analysis is performed on the calculated sound wave velocity, including the average value, standard deviation, minimum value and maximum value, etc.; analyze the change trend of the sound wave velocity, including the change over time, the change with depth, etc.), because the sound wave velocity is related to the density and elastic modulus of the geological body, and can reflect the physical state of the geological body;

[0172] S32. Analyze the structural characteristics of geological bodies such as rocks, soils, cavities, faults, cracks, etc. based on the reflection and refraction characteristics of sound waves. Structural analysis helps identify potential geological defects that affect the stability of the excavation surface;

[0173] The structural characteristics of geological bodies can be analyzed based on the reflection and refraction characteristics of sound waves by following the steps below:

[0174] 1) Analysis of sound wave propagation characteristics:

[0175] Analyze the propagation characteristics of sound waves in geological bodies, including reflection and refraction.

[0176] Reflected waves can provide information about the interface of geological bodies, and refracted waves can provide information about the internal structure of geological bodies.

[0177] 2) Structural feature recognition:

[0178] Based on the reflection and refraction characteristics of sound waves, the structural characteristics of geological bodies such as rocks, soil, cavities, faults, cracks, etc. are identified.

[0179] The structural characteristics are analyzed and interpreted by using parameters such as the speed, amplitude, and frequency of sound waves combined with known information about the geological body.

[0180] 3) Geological defect assessment:

[0181] The identified geological defects may affect the stability of the excavation face, such as cavities that may lead to collapse, faults and cracks that may cause groundwater leakage, etc.

[0182] Identified geological defects are evaluated to determine the extent of their impact on the stability of the excavation face.

[0183] 4) Result verification:

[0184] Compare the acoustic wave analysis results with the actual geological conditions to verify the accuracy of the analysis results.

[0185] Adjust the analysis methods and parameters based on the validation results to improve the accuracy of the analysis results.

[0186] Through the above steps, the acoustic wave data can be effectively used to analyze the structural characteristics of the geological body, identify potential geological defects that affect the stability of the excavation surface, and provide important reference for engineering design and construction.

[0187] S33. Invert the physical parameters of the geological body, such as density and elastic modulus, through data such as acoustic wave velocity (parameter inversion provides a quantitative basis for risk assessment), and construct acoustic wave risk indicators based on acoustic wave data and related geological parameters (wind acoustic wave risk indicators can reflect the stability of the excavation surface), where the acoustic wave risk indicators include the acoustic wave velocity change rate and geological defect density, etc.;

[0188] The physical parameters of the geological body can be inverted through data such as acoustic wave velocity, and the acoustic wave risk index can be constructed by following the steps below:

[0189] 1) Parameter inversion

[0190] Establish an inversion model: select an appropriate inversion algorithm, such as the least squares method, genetic algorithm, etc., to link the acoustic velocity data with the physical parameters of the geological body (such as density, elastic modulus).

[0191] Inversion calculation: Use the acoustic wave velocity data to calculate the physical parameters of the geological body through the inversion model.

[0192] Model verification: Verify the accuracy of inversion parameters through field drilling sampling or laboratory test results.

[0193] 2) Constructing acoustic risk indicators

[0194] Sound wave velocity change rate: Calculate the rate of change of sound wave velocity with time or depth to reflect the dynamic changes of geological bodies.

[0195] Geological defect density: Identify geological defects and calculate their density based on sound wave reflection and refraction data.

[0196] 3) Definition of acoustic risk indicators

[0197] Acoustic wave risk index: Combining factors such as the rate of change of acoustic wave velocity and the density of geological defects, one or more indicators are constructed to reflect the stability of the excavation face.

[0198] 4) Risk Assessment

[0199] Risk level classification: According to the size of the acoustic risk index, the stability of the excavation surface is divided into different risk levels.

[0200] Risk management and decision-making: Based on the risk assessment results, formulate corresponding risk management measures and construction decisions.

[0201] 5) Continuous monitoring and updating

[0202] Monitoring data update: Regularly collect new acoustic data and update the physical parameters of geological bodies and acoustic risk indicators.

[0203] Dynamic risk assessment: Dynamically assess the stability status and risk level of the excavation face based on new data.

[0204] Through such a process, a quantitative assessment of the stability of the excavation face can be achieved, providing a scientific basis for engineering risk management.

[0205] S34. Use the analytic hierarchy process to assign weights to the acoustic wave risk indicators, calculate the risk scores based on the acoustic wave risk indicators and the corresponding weights, and evaluate the risk level of the excavation surface based on the risk scores.

[0206] Using the Analytic Hierarchy Process (AHP) to assign weights to acoustic risk indicators and calculate risk scores based on these indicators and weights can be done in the following steps:

[0207] 1) Determine the risk indicator system

[0208] Acoustic risk indicators: Determine key acoustic risk indicators used to assess excavation face risk, such as the rate of change of acoustic velocity, geological defect density, etc.

[0209] Indicator classification: Risk indicators are divided into different levels, such as primary indicators, secondary indicators, etc., to form a hierarchical structure.

[0210] 2) Construct judgment matrix

[0211] Expert consultation: Invite experts in related fields to conduct pairwise comparisons of risk indicators and construct a judgment matrix.

[0212] Paired comparison: Experts conduct pairwise comparisons based on the importance of indicators and give a relative importance score for each indicator.

[0213] 3) Calculate weights using the analytic hierarchy process

[0214] Weight calculation: Use the analytic hierarchy process to calculate the weight of each risk indicator.

[0215] By calculating the maximum eigenvalue and the corresponding eigenvector of the judgment matrix, the weight of each indicator is obtained.

[0216] Consistency check: Conduct consistency check to ensure the consistency of expert evaluation.

[0217] 4) Risk score calculation

[0218] Indicator scoring: Based on the actual monitoring data, score each acoustic risk indicator.

[0219] The indicator score is obtained based on the scoring criteria of the acoustic risk indicators established in advance. The scoring criteria should be set according to the specific monitoring data range and actual conditions, and can be scored quantitatively or qualitatively. Quantitative scoring is usually based on the actual measurement value of the indicator, while qualitative scoring is based on the experience and judgment of experts.

[0220] For example, the scoring criteria for the acoustic risk indicators established in advance are as follows:

[0221] Rate of change of sound wave velocity:

[0222] Change rate ≤ 5%: score 10; 5% < change rate ≤ 10%: score 8; 10% < change rate ≤ 15%: score 6; 15% < change rate ≤ 20%: score 4; change rate > 20%: score 2.

[0223] Geological defect density: defect density ≤ 1 / m 2 : Rating 10; 1<defect density ≤ 3 pieces / m 2 : Rating 8; 3<Defect density ≤ 5 pieces / m 2 : Score 6; 5< Defect density ≤ 7 pieces / m 2 : Rating 4; Defect density>7 / m 2 :Rating 2.

[0224] Collect data related to each acoustic risk indicator from actual on-site monitoring. This data can come from acoustic detection instruments, geological survey reports, etc. Score each acoustic risk indicator based on the actual monitoring data and the established scoring criteria. For example, if the actual monitored acoustic velocity change rate is 12%, then according to the scoring criteria, the indicator scores 6.

[0225] Weighted score: Multiply the indicator score by its corresponding weight to get the weighted score.

[0226] 5) Risk level assessment

[0227] Risk level classification: According to the size of the risk score, the risk level of the excavation surface is divided into different levels, such as low risk, medium risk, high risk, etc.

[0228] Risk Assessment Report: Prepare a risk assessment report including risk levels, major risk factors and recommended risk management measures.

[0229] 6) Risk Management

[0230] Develop measures: Develop appropriate risk management measures based on the risk level.

[0231] Monitoring and adjustment: Implement risk management measures and continuously monitor changes in risk indicators, and adjust risk levels and measures when necessary.

[0232] Through the above steps, the risk level of the excavation surface can be systematically evaluated, providing a scientific basis for engineering safety management. The analytic hierarchy process has high practicality and effectiveness in dealing with complex decision-making problems.

[0233] In addition to the above evaluation methods, the stability of the excavation face can also be evaluated in the following ways:

[0234] Sonic wave tests are carried out before and after excavation. Data are collected at different locations and depths through special equipment (such as sonic probes), and the theoretical sound wave velocity is calculated based on the known geological material parameters stored in the database. The known geological material parameters include the elastic modulus, material density and Poisson's ratio of the geotechnical material. These parameters are measured in advance by professional equipment and stored in the database.

[0235] The theoretical sound wave velocity is calculated as follows:

[0236]

[0237] In the formula, v t is the theoretical sound wave velocity, E is the elastic modulus of the geomaterial, ρt is the material density, and A is the Poisson's ratio.

[0238] Design a formula to evaluate the stability of the excavation face, taking into account the relationship between the change in sound wave velocity and the strength of rock and soil. The calculation formula is as follows:

[0239]

[0240] In the formula, v m is the measured sound wave velocity, which is detected by an acoustic wave detector. S is the stability index, and C is the correction coefficient stored in the database. It is obtained by simulating the propagation of sound waves in different geological structures using numerical simulation technology. By changing the parameters in the model (such as rock elastic modulus, density, Poisson's ratio, etc.) and observing how these changes affect the sound wave velocity, the correction coefficient can be estimated. If S is close to or greater than 1, it indicates that the actual conditions of the excavation face are at least consistent with the theoretical expectations and are considered to be stable. If S is significantly less than 1, it indicates that the measured sound wave velocity is lower than the theoretical value, and there may be unstable geological structures such as cracks or voids.

[0241] S4. Determine the evaluation index affecting the stability of the excavation surface based on the identified key influencing factors and potential risk points, and evaluate the stability of the excavation surface in combination with the real-time monitoring data of the evaluation index;

[0242] The step of determining the evaluation index affecting the stability of the excavation surface according to the identified key influencing factors and potential risk points, and evaluating the stability of the excavation surface in combination with the real-time monitoring data of the evaluation index comprises the following steps:

[0243] S41. Determine the evaluation indicators affecting the stability of the excavation surface based on key influencing factors and potential risk points;

[0244] S42. Assign weights to the evaluation indicators using the analytic hierarchy process, and construct a risk assessment model based on the evaluation indicators and the weights. The risk assessment model may be a mathematical model, such as a linear weighted model, or a more complex model, such as a fuzzy comprehensive evaluation model.

[0245] S43, obtaining real-time monitoring data related to the evaluation indicators, and calculating the score of each evaluation indicator according to the risk assessment model;

[0246] S44, multiplying the score of each evaluation indicator by its corresponding weight to obtain a weighted score, and adding the weighted scores of all evaluation indicators to obtain a total risk score of the excavation surface;

[0247] S45. Evaluate the risk level of the excavation surface according to the total risk score of the excavation surface, such as low risk, medium risk, high risk, etc. The evaluation process is to compare the total risk score with the risk level corresponding to each total risk score stored in the database, and obtain the risk level corresponding to the total risk score stored in the database that has successfully compared. For example, if the total risk score is 1, and the total risk score stored in the database is 1, which corresponds to low risk, then the risk level of the excavation surface determined is low risk.

[0248] S5. Based on the results of simulation, acoustic wave and index evaluation, the comprehensive risk score of the excavation surface is calculated in combination with the preset weights, and the risk level of the excavation surface is determined based on the comprehensive risk score.

[0249] The method of calculating the comprehensive risk score of the excavation surface based on the results of simulation, acoustic wave and index evaluation in combination with preset weights, and determining the risk level of the excavation surface according to the comprehensive risk score includes the following steps:

[0250] S51. Define key criteria for assessing risk levels of excavation faces, including simulation-based risk levels, acoustic-based risk levels, and indicator-based risk levels;

[0251] S52, respectively obtaining the risk level assessment results based on simulation, the risk level assessment results based on sound waves, and the risk level assessment results based on indicators, and performing standardization on the assessment results, for example, converting the risk level into a numerical score, such as low risk = 1, medium risk = 2, and high risk = 3;

[0252] S53, using an entropy weight method to respectively determine the weights of the risk level based on simulation, the risk level based on sound waves, and the risk level based on indicators;

[0253] Specifically, the method of using the entropy weight method to respectively determine the weights of the risk level based on simulation, the risk level based on sound waves, and the risk level based on indicators includes the following steps:

[0254] S531, calculating the information entropy of each key criterion, and determining the difference coefficient according to the difference between the information entropy of each key criterion and the maximum information entropy;

[0255] S532. Calculate the weight of each key criterion according to the difference coefficient, and normalize the calculated weight.

[0256] The calculation formula of information entropy is:

[0257]

[0258] The weight calculation formula of key criteria is:

[0259]

[0260] The calculation formula of the normalized weight is:

[0261]

[0262] Where H(X) represents the information entropy of random variable X; n represents the number of key criteria; x i represents the value of the i-th key criterion; p(x i ) represents the probability of the i-th key criterion; ω i represents the weight of the i-th key criterion; ω j represents the weight of the jth key criterion; ω i ′ represents the normalized weight of key criterion i; d i represents the coefficient of difference of the i-th key criterion; d j represents the coefficient of variation of the jth key criterion;

[0263] S54, constructing a decision matrix based on the acquired data, wherein the rows in the decision matrix represent different key criteria, and the columns represent different risk level assessment results;

[0264] S55. Calculate the comprehensive risk level of each risk level assessment result using the weighted average method, and determine the final risk level of the excavation surface based on the ranking results of the comprehensive risk levels and the risk threshold.

[0265] The calculation formula for the comprehensive risk level is:

[0266]

[0267] In the formula, R represents the comprehensive risk level; r i represents the score of the i-th key criterion.

[0268] In order to better understand the final risk level assessment process, the following is a specific example:

[0269] Assuming there is an excavation face, its risk level is assessed through three methods: simulation, acoustic wave monitoring and indicator evaluation:

[0270] The risk level assessment results based on the simulation are: medium risk (assumed score = 2);

[0271] The risk level assessment results based on sound waves are: low risk (hypothetical score = 1);

[0272] The risk level assessment results based on the indicators are: high risk (assumed score = 3);

[0273] Then, the entropy weight method is used to determine the weight of each evaluation method:

[0274] The risk level weight based on the simulation is 0.4.

[0275] The risk level based on sound waves has a weight of 0.3.

[0276] The indicator-based risk level weight is 0.3.

[0277] Finally, calculate the comprehensive risk score: R = (0.4 × 2) + (0.3 × 1) + (0.3 × 3) = 2;

[0278] Based on the comprehensive risk score, the risk level of the excavation face can be determined, that is, the comprehensive risk score is 2. According to the preset risk level classification, this may mean that the excavation face is at a medium risk level.

[0279] According to another embodiment of the present invention, there is provided an underwater tunnel shield excavation face stability evaluation system, the underwater tunnel shield excavation face stability evaluation system comprising a data acquisition correction module, a simulation-based risk level evaluation module, a sound wave-based risk level evaluation module, an indicator-based risk level evaluation module and a comprehensive risk level evaluation module;

[0280] The data acquisition and correction module is used to collect monitoring data of the underwater tunnel shield excavation face, and identify and correct abnormal data in the monitoring data based on the excavation data correction algorithm;

[0281] The simulation-based risk level assessment module is used to construct a numerical model based on the excavation surface and set model parameters; simulate the stability of the excavation surface under different working conditions based on the numerical model, and identify key influencing factors and potential risk points based on the simulation results;

[0282] The acoustic wave-based risk level assessment module is used to identify the geological structure characteristics of the excavation surface based on the acoustic wave data in the corrected monitoring data, and to assess the stability of the excavation surface in combination with the acoustic wave velocity in the acoustic wave data;

[0283] The indicator-based risk level assessment module is used to determine the assessment indicators that affect the stability of the excavation surface according to the identified key influencing factors and potential risk points, and to assess the stability of the excavation surface in combination with the real-time monitoring data of the assessment indicators;

[0284] The comprehensive risk level assessment module is used to calculate the comprehensive risk score of the excavation surface based on the results of simulation, acoustic wave and index assessment combined with preset weights, and determine the risk level of the excavation surface according to the comprehensive risk score.

[0285] In summary, with the help of the above-mentioned technical scheme of the present invention, the present invention can not only use the correction algorithm based on the excavation data to identify and correct the abnormal data in the monitoring data, but also construct a numerical model to simulate the stability of the excavation face under different working conditions, and use the acoustic wave data to identify the geological structure characteristics, and evaluate the stability of the excavation face in combination with the acoustic wave velocity. At the same time, the evaluation index can be determined according to the key influencing factors and potential risk points identified, and the stability of the excavation face can be evaluated in combination with the real-time monitoring data of the evaluation index, so that the comprehensive risk score of the excavation face can be calculated based on the results of simulation, acoustic wave and index evaluation, combined with preset weights, and the final risk level of the excavation face can be determined according to the comprehensive risk score. Compared with the traditional evaluation method, the present invention can realize real-time monitoring and response to the changes in the excavation face, adapt to complex geological conditions, and comprehensively consider multiple factors to provide more accurate and real-time stability evaluation, which is helpful to improve the accuracy and real-time performance of the excavation face stability evaluation during the shield excavation of underwater tunnels, thereby effectively ensuring the construction safety of the project and improving the construction quality and efficiency. In addition, the present invention can not only use the Bayesian optimization algorithm to optimize the bidirectional long short-term memory model based on the convolutional neural network, thereby effectively improving the performance of the model in the abnormality recognition task, but also use the optimized model to automatically identify and correct abnormal data, significantly improving the accuracy and reliability of the monitoring data, thereby ensuring that the monitoring data can timely reflect the dynamic changes of the excavation face, and providing a strong technical guarantee for the stability assessment of the excavation face during the shield excavation of underwater tunnels.

[0286] The technical features of the above-mentioned embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification. A person of ordinary skill in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program. The program can be stored in a computer-readable storage medium. When the program is executed, it includes the steps described in the above method, and the storage medium, such as ROM / RAM, a disk, an optical disk, etc.

[0287] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for evaluating the stability of an underwater tunnel shield excavation face, characterized in that: The underwater tunnel shield excavation face stability evaluation method comprises the following steps: S1. Collect monitoring data of the underwater tunnel shield excavation face, and identify and correct abnormal data in the monitoring data based on the excavation data correction algorithm; S2. Construct a numerical model based on the excavation surface and set the model parameters; simulate the stability of the excavation surface under different working conditions based on the numerical model, and identify key influencing factors and potential risk points based on the simulation results; S3, based on the acoustic wave data in the corrected monitoring data, identifying the geological structure characteristics of the excavation surface, and evaluating the stability of the excavation surface in combination with the acoustic wave velocity in the acoustic wave data; S4. Determine the evaluation index affecting the stability of the excavation surface based on the identified key influencing factors and potential risk points, and evaluate the stability of the excavation surface in combination with the real-time monitoring data of the evaluation index; S5. Based on the results of simulation, acoustic wave and index evaluation, the comprehensive risk score of the excavation surface is calculated in combination with the preset weights, and the risk level of the excavation surface is determined according to the comprehensive risk score; The method of calculating the comprehensive risk score of the excavation face based on the results of simulation, acoustic wave and index evaluation in combination with preset weights, and determining the risk level of the excavation face according to the comprehensive risk score includes the following steps: S51. Define key criteria for assessing risk levels of excavation faces, including simulation-based risk levels, acoustic-based risk levels, and indicator-based risk levels; S52, respectively obtaining the risk level assessment results based on simulation, the risk level assessment results based on sound waves, and the risk level assessment results based on indicators, and performing standardization processing on the assessment results; S53, using an entropy weight method to respectively determine the weights of the risk level based on simulation, the risk level based on sound waves, and the risk level based on indicators; S54, constructing a decision matrix based on the acquired data, wherein the rows in the decision matrix represent different key criteria, and the columns represent different risk level assessment results; S55. Calculate the comprehensive risk level of each risk level assessment result using the weighted average method, and determine the final risk level of the excavation surface based on the ranking results of the comprehensive risk level and the risk threshold.

2. The method for evaluating the stability of an underwater tunnel shield excavation face according to claim 1, characterized in that: The method of collecting monitoring data of the underwater tunnel shield excavation face and identifying and correcting abnormal data in the monitoring data based on the excavation data correction algorithm includes the following steps: S11, collecting monitoring data of the underwater tunnel shield excavation face, and sequentially performing data cleaning, normalization and data labeling processing on the monitoring data to obtain a data set, wherein the monitoring data includes geological and soil parameters, groundwater parameters, structural response data, environmental parameters and acoustic wave monitoring data; S12. Using the Bayesian optimization algorithm to solve the optimal combination parameters of the bidirectional long short-term memory model based on the convolutional neural network, and training the bidirectional long short-term memory model based on the convolutional neural network in combination with the data set; S13, obtaining the monitoring data of the excavation surface at the previous moment, and using the trained bidirectional long short-term memory model based on the convolutional neural network to output the predicted monitoring data at the current moment corresponding to the monitoring data at the previous moment; S14. Calculate the difference between the real-time monitoring data at the current moment and the predicted monitoring data, and determine whether the difference exceeds the preset difference threshold. If not, determine that the real-time monitoring data is normal data. If so, determine that the real-time monitoring data is abnormal data, and use the predicted monitoring data at the current moment as the corrected real-time monitoring data.

3. The method for evaluating the stability of an underwater tunnel shield excavation face according to claim 2, characterized in that: The method of solving the optimal combination parameters of the bidirectional long short-term memory model based on the convolutional neural network by using the Bayesian optimization algorithm and training the bidirectional long short-term memory model based on the convolutional neural network in combination with the data set comprises the following steps: S121, setting the hyperparameters to be optimized and defining a search space for each hyperparameter; initializing the Bayesian optimizer and setting initial hyperparameter values; S122, constructing a bidirectional long short-term memory model based on a convolutional neural network using the initial hyperparameter group obtained from the Bayesian optimizer, and training and verifying the bidirectional long short-term memory model based on the convolutional neural network using the data set; based on the verification result, the Bayesian optimizer updates the probability distribution of the hyperparameters; S123. Repeat S122 until the maximum number of iterations, the time limit, or the performance indicator reaches a preset threshold; according to the final result of the Bayesian optimizer, select the best hyperparameter combination, and use the best hyperparameter combination to reconstruct and train the bidirectional long short-term memory model based on the convolutional neural network.

4. The method for evaluating the stability of an underwater tunnel shield excavation face according to claim 1, characterized in that: The method of constructing a numerical model based on the excavation surface, setting model parameters, simulating the stability of the excavation surface under different working conditions based on the numerical model, and identifying key influencing factors and potential risk points according to the simulation results includes the following steps: S21. constructing a three-dimensional model according to the geometric shape and size of the actual excavation surface, and meshing the three-dimensional model to generate a finite element mesh or a discrete element mesh; S22. Based on the geological survey report, the soil in the three-dimensional model is assigned corresponding physical and mechanical parameters, and the supporting structure is assigned corresponding material properties; S23. Setting the boundary conditions and initial stress field of the three-dimensional model according to the actual geological conditions and construction parameters, and simulating the excavation surface stability under normal working conditions and extreme working conditions; S24. Use finite element analysis to simulate the excavation process, record relevant indicators of excavation surface stability, and analyze simulation results to identify key influencing factors and potential risk points of excavation surface stability; S25. Conduct sensitivity analysis on key influencing factors to evaluate their impact on the stability of the excavation face, and evaluate the risk level of the excavation face based on the simulation results and sensitivity analysis results.

5. The method for evaluating the stability of an underwater tunnel shield excavation face according to claim 1, characterized in that: The method of identifying the geological structure characteristics of the excavation surface based on the acoustic wave data in the corrected monitoring data and evaluating the stability of the excavation surface in combination with the acoustic wave velocity in the acoustic wave data includes the following steps: S31, obtaining the sound wave data in the corrected monitoring data, calculating the sound wave velocity according to the sound wave propagation time, and analyzing the change trend of the sound wave velocity; S32. Analyze the structural characteristics of geological bodies based on the reflection and refraction characteristics of sound waves and identify potential geological defects that affect the stability of the excavation surface; S33, inverting the physical parameters of the geological body through the acoustic wave velocity, and constructing an acoustic wave risk index based on the acoustic wave data and related geological parameters, wherein the acoustic wave risk index includes the acoustic wave velocity change rate and the geological defect density; S34. Use the analytic hierarchy process to assign weights to the acoustic wave risk indicators, calculate the risk scores based on the acoustic wave risk indicators and the corresponding weights, and evaluate the risk level of the excavation surface based on the risk scores.

6. The method for evaluating the stability of an underwater tunnel shield excavation face according to claim 1, characterized in that: Determining the evaluation index affecting the stability of the excavation surface according to the identified key influencing factors and potential risk points, and evaluating the stability of the excavation surface in combination with the real-time monitoring data of the evaluation index includes the following steps: S41. Determine the evaluation indicators affecting the stability of the excavation surface based on key influencing factors and potential risk points; S42. Use the analytic hierarchy process to assign weights to the evaluation indicators, and construct a risk assessment model based on the evaluation indicators and weights; S43, obtaining real-time monitoring data related to the evaluation indicators, and calculating the score of each evaluation indicator according to the risk assessment model; S44, multiplying the score of each evaluation indicator by its corresponding weight to obtain a weighted score, and adding the weighted scores of all evaluation indicators to obtain a total risk score of the excavation surface; S45. Evaluate the risk level of the excavation surface based on the total risk score of the excavation surface.

7. The method for evaluating the stability of an underwater tunnel shield excavation face according to claim 1, characterized in that: The method of using the entropy weight method to respectively determine the weights of the risk level based on simulation, the risk level based on sound waves, and the risk level based on indicators includes the following steps: S531, calculating the information entropy of each key criterion, and determining the difference coefficient according to the difference between the information entropy of each key criterion and the maximum information entropy; S532. Calculate the weight of each key criterion according to the difference coefficient, and normalize the calculated weight.

8. The method for evaluating the stability of an underwater tunnel shield excavation face according to claim 7, characterized in that: The calculation formula of the information entropy is: The weight calculation formula of key criteria is: The calculation formula of the normalized weight is: The calculation formula for the comprehensive risk level is: Where H(X) represents the information entropy of random variable X; n represents the number of key criteria; x i represents the value of the i-th key criterion; p(x i ) represents the probability of the i-th key criterion; ω i represents the weight of the i-th key criterion; ω j represents the weight of the jth key criterion; ω i ′ represents the normalized weight of key criterion i; d i represents the coefficient of variation of the i-th key criterion; d j represents the coefficient of variation of the jth key criterion; R represents the comprehensive risk level; r i represents the score of the i-th key criterion.

9. A system for evaluating the stability of an underwater tunnel shield excavation face, used to implement the steps of the method for evaluating the stability of an underwater tunnel shield excavation face as described in any one of claims 1 to 8, characterized in that: The underwater tunnel shield excavation face stability evaluation system includes a data acquisition correction module, a simulation-based risk level evaluation module, an acoustic-wave-based risk level evaluation module, an indicator-based risk level evaluation module, and a comprehensive risk level evaluation module; The data acquisition and correction module is used to collect monitoring data of the underwater tunnel shield excavation face, and identify and correct abnormal data in the monitoring data based on the excavation data correction algorithm; The simulation-based risk level assessment module is used to construct a numerical model based on the excavation surface and set model parameters; simulate the stability of the excavation surface under different working conditions based on the numerical model, and identify key influencing factors and potential risk points based on the simulation results; The acoustic wave-based risk level assessment module is used to identify the geological structure characteristics of the excavation surface based on the acoustic wave data in the corrected monitoring data, and to assess the stability of the excavation surface in combination with the acoustic wave velocity in the acoustic wave data; The indicator-based risk level assessment module is used to determine the assessment indicators that affect the stability of the excavation surface according to the identified key influencing factors and potential risk points, and to assess the stability of the excavation surface in combination with the real-time monitoring data of the assessment indicators; The comprehensive risk level assessment module is used to calculate the comprehensive risk score of the excavation surface based on the results of simulation, acoustic wave and index assessment combined with preset weights, and determine the risk level of the excavation surface according to the comprehensive risk score.

Images