Intelligent prediction method and system for settlement of TBM passing through pebble stratum

Abstract

The invention provides an intelligent prediction method and system for settlement of a TBM penetrating through a pebble stratum, and belongs to the technical field of stratum settlement prediction. The method comprises the following steps: acquiring ground surface monitoring point position information, excavation surface position information and burial depth, inputting the information into a first deep neural network model, and acquiring and outputting tunnel vault displacement, haunch convergence and ground surface settlement; the second deep neural network model takes a position coordinate formed by earth surface monitoring point position information, excavation surface position information and burial depth and earth surface settlement as input, and outputs earth surface elastic modulus and Poisson ratio; obtaining stress-strain components of ground surface settlement, and constructing a control equation and boundary conditions based on the stress-strain components, the elastic modulus and the Poisson's ratio; obtaining prior geological information by using the residual error of the control equation; and by taking the control equation, the boundary condition and the prior geological information as constraint conditions, obtaining a settlement prediction value when the TBM passes through the pebble stratum based on the tunnel vault displacement, and realizing accurate prediction of the settlement of the pebble stratum.

Description

Technical Field

[0001] The present invention belongs to the technical field of stratum settlement prediction, and in particular relates to an intelligent prediction method and system for settlement of a TBM passing through a pebble stratum. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] With the accelerating pace of urbanization, the development and utilization of underground space has become an effective way to alleviate the pressures of urban transportation and infrastructure construction. Tunnel boring machines (TBMs), a key piece of equipment in modern tunnel construction, are widely used in various tunnel projects due to their high efficiency and minimal surface disturbance. However, when traversing complex geological conditions, especially gravelly strata, TBMs are prone to causing ground deformation and surface subsidence, posing potential risks to surrounding buildings and infrastructure. Therefore, accurately predicting TBM-induced subsidence to ensure construction safety has become a key technical issue that urgently needs to be addressed in the field of tunnel engineering.

[0004] Traditional settlement prediction methods are primarily based on empirical formulas, analytical elastic-plastic equations, and numerical simulations. However, the physical knowledge models underlying these methods struggle to integrate information with measured surface settlement, resulting in low efficiency in actual engineering applications. Machine learning is an effective method for predicting TBM-induced surface settlement, but it relies on high-quality data mining and mapping relationships, neglecting the physics of tunnel-induced ground displacement. Most projects have relatively limited physical knowledge, and obtaining sufficient monitoring data to train machine learning models during construction is difficult. This is especially true in gravelly strata, where the heterogeneity of the strata, uncertainty in geological parameters, and complex stress distribution complicate settlement prediction. Summary of the Invention

[0005] To overcome the shortcomings of the prior art, the present invention provides an intelligent prediction method and system for TBM settlement during traversal of pebble formations. By leveraging the respective strengths of physical knowledge and data-driven approaches, a deep learning framework integrating physical constraints and prior geological information is proposed for intelligently predicting surface settlement during TBM traversal of pebble formations. This system provides more intelligent prediction and decision-making support for TBM construction in complex geological conditions and has broad application prospects.

[0006] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions: The first aspect of the present invention provides an intelligent prediction method for settlement of a TBM passing through a pebble stratum; An intelligent prediction method for settlement of a TBM traveling through pebble strata, comprising: Obtain the location of surface monitoring points, excavation surface location, and burial depth; Input the surface monitoring point location, excavation surface location and burial depth into the trained settlement prediction knowledge data dual-driven model, and output the surface settlement value when the TBM passes through the pebble stratum; The trained settlement prediction knowledge data dual-driven model includes a first deep neural network model and a second deep neural network model connected in series; the first deep neural network model takes the position of surface monitoring points, the position of the excavation surface and the burial depth as input, and outputs the tunnel vault displacement, arch waist convergence and surface settlement; the second deep neural network model takes the position of surface monitoring points, the position of the excavation surface and the burial depth as input, and outputs the surface elastic modulus and Poisson's ratio; Based on the basic principles of elasticity, the stress and strain components of surface settlement are solved, and the governing equations and boundary conditions are constructed based on the stress and strain components, elastic modulus, and Poisson's ratio. A priori geological information constraints are imposed based on the residual terms of the governing equations. A dual-driven model loss function is constructed, which includes data-driven terms, residual terms of the control equations, and boundary conditions. With the goal of minimizing the loss function, an intelligent algorithm is used to optimize the hyperparameter combination to obtain the final settlement prediction value.

[0007] As a further technical solution, the tunnel crown displacement includes: uniform radial displacement caused by stratum loss during tunnel excavation and elliptical deformation of the tunnel profile caused by anisotropic initial stress.

[0008] As a further technical solution, the control equation includes: The governing equations satisfied by the surface stress, strain, and displacement components are as follows:

[0009]

[0010]

[0011] Where, x The location information of the surface monitoring points, z is the burial depth; Find the partial derivative of the stress tensor with respect to the coordinates; is the Kronecker function; is the volume strain tensor; is the strain tensor; for x - z In-plane elastic modulus; 、 Find the partial derivative of the displacement tensor with respect to the coordinates; The governing equations for tunnel crown displacement and arch waist convergence are as follows:

[0012] Where, y is the excavation face position; is the vault displacement; To converge for the arch waist; r is the tunnel excavation radius; The governing equation between tunnel crown displacement and ground settlement is as follows:

[0013] Where, For surface subsidence; is the uniform radial displacement caused by ground loss during tunnel excavation; The elliptical deformation of the tunnel profile is caused by the anisotropic initial stress; u is Poisson's ratio; h is the distance from the tunnel axis to the ground.

[0014] As a further technical solution, the surface settlement is:

[0015] Where, i is the angle between the line connecting the boundary point around the hole and the center of the circle and the horizontal direction, 、 For different excavation face positions The displacement of the vault at .

[0016] As a further technical solution, the boundary conditions are boundary conditions satisfied by the surface stress and displacement components, and the boundary conditions are:

[0017]

[0018] Where, for x Normal stress in the direction; for z Normal stress in the direction; On the surface x Direction displacement.

[0019] As a further technical solution, the process of applying a priori geological information constraints based on the residual term of the control equation is as follows: The original value of the second deep neural network model output Apply exponential function constraints to ensure E Always positive:

[0020] The original value of the second deep neural network model output Apply the scaled Sigmoid function constraint:

[0021] After the constraint E and u Substitute it into the control equation to calculate the residual term of the control equation.

[0022] As a further technical solution, the trained settlement prediction knowledge data dual-driven model also includes a loss function, which is:

[0023] Where, is the total number of sample points containing position coordinates and actual displacement values; is the actual value of surface settlement; is the predicted value of surface subsidence; is the total number of sample points in the study area that do not require actual displacement values; is the total number of sample points on the boundary that do not require actual displacement values; ~ is the control equation; l is the weight coefficient of the physical knowledge driving item.

[0024] A second aspect of the present invention provides an intelligent prediction system for settlement of a TBM passing through a pebble stratum.

[0025] An intelligent prediction system for TBM settlement during traversal of pebble strata, comprising: The settlement information acquisition module is configured to: obtain the location of the surface monitoring point, the location of the excavation surface and the burial depth; The data processing module is configured to: input the surface monitoring point position, excavation surface position and burial depth into a first deep neural network model to obtain tunnel vault displacement, arch waist convergence and surface settlement; input the surface monitoring point position, excavation surface position and burial depth and the surface settlement into a second deep neural network model to output the surface elastic modulus and Poisson's ratio; The control equation acquisition module is configured to: solve the stress and strain components of surface settlement based on the basic principles of elastic mechanics, construct the control equations and boundary conditions based on the stress and strain components, elastic modulus and Poisson's ratio; and impose a priori geological information constraints based on the residual terms of the control equations; The settlement prediction value acquisition module is configured to: construct a dual-drive model loss function including data-driven terms, residual terms of the control equation and boundary conditions, with the goal of minimizing the loss function, and use intelligent algorithms to optimize the hyperparameter combination to obtain the final settlement prediction value.

[0026] A third aspect of the present invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps of the intelligent prediction method for settlement of a TBM passing through a pebble formation as described in the first aspect of the present invention.

[0027] The fourth aspect of the present invention provides an electronic device, comprising a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, the steps of the intelligent prediction method for TBM settlement when passing through pebble formations as described in the first aspect of the present invention are implemented.

[0028] One or more of the above technical solutions have the following beneficial effects: This method couples the physical and mechanical mechanisms of TBM-induced surface settlement to a deep learning framework. Using governing equations, boundary conditions, and prior geological information as constraints, it effectively overcomes the problem of neglecting physical laws in data-driven approaches, reduces reliance on large amounts of measured data, and improves prediction accuracy and generalization capabilities. This method provides more intelligent prediction and decision-making support for TBM construction in complex geological conditions and has broad application prospects.

[0029] Advantages of additional aspects of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0031] Figure 1 This is a flow chart of the method of the first embodiment.

[0032] Figure 2 This is the architecture diagram of the settlement prediction knowledge and data dual-driven model in the first embodiment.

[0033] Figure 3 This is a system structure diagram of the second embodiment. DETAILED DESCRIPTION

[0034] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0035] It should be noted that the terms used herein are for describing particular embodiments only and are not intended to limit the exemplary embodiments according to the present invention.

[0036] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.

[0037] Example 1 This embodiment discloses an intelligent prediction method for settlement of a TBM passing through a pebble stratum; like Figure 1 and Figure 2 As shown in FIG, an intelligent prediction method for settlement of a TBM traveling through a pebble stratum includes: Step S1, obtaining the position information of the surface monitoring point, the position information of the excavation surface and the burial depth; In this embodiment, the buried depth of the TBM tunnel is less than 5 to 7 times the tunnel excavation radius. The location information of the surface monitoring point is obtained based on the construction information. x , excavation surface location information y and burial depth z。

[0038] Step S2: the surface monitoring point location information x , excavation surface location information y and burial depth z The data is input into the trained settlement prediction knowledge dual-driven model, and the surface settlement value when the TBM passes through the pebble layer is output; The trained settlement prediction knowledge data dual-driven model includes two parallel first deep neural network models and second deep neural network models; the first deep neural network model is based on the position of the surface monitoring point. x , excavation surface location y and burial depth z As input, output is tunnel crown displacement, arch waist convergence and surface settlement The tunnel crown displacement consists of two parts: the uniform radial displacement caused by the loss of strata during tunnel excavation and the and anisotropic initial stress induced elliptical deformation of tunnel profile . The tunnel circumference converges radially u ( i ) can be expressed by the displacement of the dome:

[0039] Where, i is the angle between the line connecting the boundary point around the hole and the center of the circle and the horizontal direction, 、 For different excavation face positions The displacement of the vault at .

[0040] From the classical VB analytical solution of the ground displacement induced by a semi-infinite space tunnel, it can be seen that once the tunnel crown displacement is determined, the ground settlement can be determined by the following formula:

[0041] Where, i is the angle between the line connecting the boundary point around the hole and the center of the circle and the horizontal direction, 、 For different excavation face positions The displacement of the vault at .

[0042] The second deep neural network model is based on the location of the surface monitoring points x , excavation surface location y and burial depth z and the surface subsidence As input, it outputs the surface elastic modulus and Poisson's ratio.

[0043] Step S3: Based on the basic principles of elastic mechanics, the stress and strain components of surface settlement are solved, and the governing equations and boundary conditions are constructed based on the stress and strain components, elastic modulus, and Poisson's ratio. A priori geological information constraints are imposed based on the residual terms of the governing equations. A dual-driven model loss function is constructed, which includes data-driven terms, governing equation residual terms, and boundary conditions. With the goal of minimizing the loss function, an intelligent algorithm is used to optimize the combination of hyperparameters (such as learning rate, number of hidden layers, and weight coefficients of knowledge-driven terms). The surface settlement value at which the loss function is minimized is the final settlement prediction value. Specifically: Step S31, based on the basic principles of elastic mechanics, solving the stress and strain components of the surface settlement, and constructing the control equations and boundary conditions based on the stress and strain components, elastic modulus and Poisson's ratio; The surface settlement vector can be further solved for its strain component according to the basic principles of elastic mechanics : 、 、 , For rock and soil x The normal strain in the direction, For rock and soil z The normal strain in the direction, For rock and soil x - z In-plane shear strain; On the surface x Directional displacement; On the surface z Direction displacement.

[0044] The stress and strain components satisfy the governing equations ~ and boundary conditions; governing equations ~ The elastic modulus involved E and Poisson's ratio u Output by the second deep neural network model and used as the control equation ~ 、 The control equation is ~ As shown in the following formula:

[0045]

[0046]

[0047] Where, Find the partial derivative of the stress tensor with respect to the coordinates; is the Kronecker function; is the volume strain tensor; is the strain tensor; for x - z In-plane elastic modulus; 、 Find the partial derivative of the displacement tensor with respect to the coordinates; The boundary conditions are as follows:

[0048]

[0049] Where, for x Normal stress in the direction; for z Normal stress in the direction; On the surface x Directional displacement; The displacement of the arch crown and the convergence of the arch waist satisfy the governing equation , as shown below:

[0050] Where, is the vault displacement; To converge for the arch waist; r is the tunnel excavation radius.

[0051] The vault displacement and ground settlement satisfy the governing equation , as shown below:

[0052] Where, For surface subsidence; The elliptical deformation of the tunnel profile is caused by the anisotropic initial stress; u is Poisson's ratio; h is the distance from the tunnel axis to the ground.

[0053] Step S32, applying a priori geological information constraints based on the residual term of the control equation, wherein in the output layer of the second deep neural network model, the elastic modulus E and Poisson's ratio u A priori geological information constraints are imposed by: (1) The original value of the output of the second deep neural network model Apply exponential function constraints to ensure E Always positive:

[0054] (2) The original value of the second deep neural network model output Apply the scaled Sigmoid function constraint:

[0055] After the constraint E and u Substitute into the governing equation 、 Calculate the residual term of the control equation.

[0056] Step S33: construct a dual-drive model loss function including a data-driven term, a residual term of the control equation, and boundary conditions, wherein the loss function is as follows:

[0057] Where, is the total number of sample points containing position coordinates and actual displacement values; is the actual value of surface settlement; is the predicted value of surface subsidence; is the total number of sample points in the study area that do not require actual displacement values; is the total number of sample points on the boundary that do not require actual displacement values; ~ is the control equation; l is the weight coefficient of the physical knowledge driving item.

[0058] In this embodiment, the loss function of the dual-drive model includes but is not limited to the form of adding the data-driven term, the residual term of the control equation, and the boundary condition: , can also be constructed by multiplication: ,in 、 and They represent the bias coefficients of the data-driven term, the residual term of the control equation, and the boundary condition, respectively. They are small positive numbers that ensure that the loss function does not become numerically unstable due to over-reliance on a certain term. Similarly, they can also be regarded as hyperparameters of the dual-drive model, and the optimal values ​​are solved through intelligent algorithms to reduce the loss value of the model.

[0059] Step S33, calculate the loss function value, with the goal of minimizing the loss function, and use an intelligent algorithm to optimize the combination of hyperparameters (such as learning rate, number of hidden layers, and weight coefficient of knowledge-driven items, etc.). The surface settlement value when the loss function is minimized is the final settlement prediction value.

[0060] Specifically, for samples whose actual displacement values ​​within the study area and on the boundary are unknown, only the residual terms of the control equation and the residual terms of the boundary conditions are calculated respectively. For samples whose actual displacement values ​​within the study area and on the boundary are known, in addition to calculating the above residual terms, it is also necessary to calculate the mean square error (MSE) between the predicted displacement value and the actual displacement value. The above mean square error, the residual terms of the control equation and the residual terms of the boundary conditions are substituted into the loss function of the dual-drive model to calculate the loss value of the first iteration, where the weight coefficient of the initial physical knowledge-driven term is set to 0.1.

[0061] Through multiple iterative training, the total loss value is gradually minimized, and the surface settlement when the total loss value is the minimum is the final output value.

[0062] Example 2 This embodiment discloses an intelligent prediction system for settlement of a TBM passing through a pebble stratum; like Figure 3 As shown in FIG, an intelligent prediction system for settlement of a TBM passing through a pebble stratum includes: An intelligent prediction system for TBM settlement during traversal of pebble strata, comprising: The settlement information acquisition module is configured to: obtain the location of the surface monitoring point, the location of the excavation surface and the burial depth; The data processing module is configured to: input the surface monitoring point position, excavation surface position and burial depth into a first deep neural network model to obtain tunnel vault displacement, arch waist convergence and surface settlement; input the surface monitoring point position, excavation surface position and burial depth and the surface settlement into a second deep neural network model to output the surface elastic modulus and Poisson's ratio; The control equation acquisition module is configured to: solve the stress and strain components of surface settlement based on the basic principles of elastic mechanics, construct the control equations and boundary conditions based on the stress and strain components, elastic modulus and Poisson's ratio; and impose a priori geological information constraints based on the residual terms of the control equations; The settlement prediction value acquisition module is configured to: construct a dual-drive model loss function including data-driven terms, residual terms of the control equation and boundary conditions, with the goal of minimizing the loss function, and use intelligent algorithms to optimize the hyperparameter combination to obtain the final settlement prediction value.

[0063] Example 3 The purpose of this embodiment is to provide a computer-readable storage medium.

[0064] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the intelligent prediction method for settlement of a TBM passing through a pebble formation as described in Example 1.

[0065] Example 4 The purpose of this embodiment is to provide an electronic device.

[0066] An electronic device includes a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, the steps of the intelligent prediction method for settlement of a TBM passing through a pebble formation as described in Example 1 are implemented.

[0067] The steps involved in the apparatuses of Examples 2, 3, and 4 above correspond to those of Method Example 1. For detailed implementations, please refer to the relevant description of Example 1. The term "computer-readable storage medium" should be understood to mean a single medium or multiple media containing one or more instruction sets; it should also be understood to include any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and causing the processor to perform any method of the present invention.

[0068] Those skilled in the art will appreciate that the modules or steps of the present invention described above can be implemented using a general-purpose computer device. Alternatively, they can be implemented using program code executable by a computing device, which can then be stored in a storage device and executed by the computing device. Alternatively, they can be fabricated into separate integrated circuit modules, or multiple modules or steps can be fabricated into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0069] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without any creative work are still within the scope of protection of the present invention.

Claims

1. An intelligent prediction method for TBM settlement when it passes through pebble strata, characterized in that: include: Obtain the location of surface monitoring points, excavation surface location, and burial depth; Input the surface monitoring point location, excavation surface location and burial depth into the trained settlement prediction knowledge data dual-driven model, and output the surface settlement value when the TBM passes through the pebble stratum; The trained settlement prediction knowledge data dual-driven model includes a first deep neural network model and a second deep neural network model connected in series; the first deep neural network model takes the position of surface monitoring points, the position of the excavation surface and the burial depth as input, and outputs the tunnel vault displacement, arch waist convergence and surface settlement; the second deep neural network model takes the position of surface monitoring points, the position of the excavation surface and the burial depth as input, and outputs the surface elastic modulus and Poisson's ratio; Based on the basic principles of elasticity, the stress and strain components of surface settlement are solved, and the governing equations and boundary conditions are constructed based on the stress and strain components, elastic modulus, and Poisson's ratio. A priori geological information constraints are imposed based on the residual terms of the governing equations. A dual-driven model loss function is constructed, which includes data-driven terms, residual terms of the control equations, and boundary conditions. With the goal of minimizing the loss function, an intelligent algorithm is used to optimize the hyperparameter combination to obtain the final settlement prediction value.

2. The intelligent prediction method for TBM settlement when passing through pebble strata according to claim 1, characterized in that: The tunnel crown displacement includes: uniform radial displacement caused by stratum loss during tunnel excavation and tunnel profile elliptical deformation caused by anisotropic initial stress.

3. The intelligent prediction method for TBM settlement when passing through pebble strata according to claim 1, characterized in that: The control equations include: The governing equations satisfied by the surface stress, strain, and displacement components are as follows: Where, x The location information of the surface monitoring points, z is the burial depth; Find the partial derivative of the stress tensor with respect to the coordinates; is the Kronecker function; is the volume strain tensor; is the strain tensor; for x - z In-plane elastic modulus; 、 Find the partial derivative of the displacement tensor with respect to the coordinates; The governing equations for tunnel crown displacement and arch waist convergence are as follows: Where, y is the excavation face position; is the vault displacement; To converge for the arch waist; r is the tunnel excavation radius; The governing equation between tunnel crown displacement and ground settlement is as follows: Where, For surface subsidence; is the uniform radial displacement caused by ground loss during tunnel excavation; The elliptical deformation of the tunnel profile is caused by the anisotropic initial stress; υ is Poisson's ratio; h is the distance from the tunnel axis to the ground.

4. The intelligent prediction method for TBM settlement when passing through pebble strata according to claim 3, characterized in that: The surface subsidence is: Where, θ is the angle between the line connecting the boundary point around the hole and the center of the circle and the horizontal direction, 、 For different excavation face positions The displacement of the vault at .

5. The intelligent prediction method for TBM settlement when passing through pebble strata according to claim 1, characterized in that: The boundary conditions are boundary conditions satisfied by the surface stress and displacement components. Specifically, the boundary conditions are: Where, for x Normal stress in the direction; for z Normal stress in the direction; On the surface x Direction displacement.

6. The intelligent prediction method for TBM settlement when passing through pebble strata according to claim 1, characterized in that: The process of applying prior geological information constraints based on the residual term of the governing equation is: The original value of the second deep neural network model output Apply exponential function constraints to ensure E Always positive: The original value of the second deep neural network model output Apply the scaled Sigmoid function constraint: After the constraint E and υ Substitute into the governing equation 、 Calculate the residual term of the control equation.

7. The intelligent prediction method for TBM settlement when passing through pebble strata according to claim 1, characterized in that: The trained settlement prediction knowledge data dual-driven model also includes a loss function, which is: Where, is the total number of sample points containing position coordinates and actual displacement values; is the actual value of surface settlement; is the predicted value of surface subsidence; is the total number of sample points in the study area that do not require actual displacement values; is the total number of sample points on the boundary that do not require actual displacement values; ~ is the control equation; λ is the weight coefficient of the physical knowledge driving item.

8. An intelligent prediction system for TBM settlement in gravel formations, characterized by: include: The settlement information acquisition module is configured to: obtain the location of the surface monitoring point, the location of the excavation surface and the burial depth; The data processing module is configured to: input the surface monitoring point position, excavation surface position and burial depth into a first deep neural network model to obtain tunnel vault displacement, arch waist convergence and surface settlement; input the surface monitoring point position, excavation surface position and burial depth and the surface settlement into a second deep neural network model to output the surface elastic modulus and Poisson's ratio; The control equation acquisition module is configured to: solve the stress and strain components of surface settlement based on the basic principles of elastic mechanics, construct the control equations and boundary conditions based on the stress and strain components, elastic modulus and Poisson's ratio; and impose a priori geological information constraints based on the residual terms of the control equations; The settlement prediction value acquisition module is configured to: construct a dual-drive model loss function including data-driven terms, residual terms of the control equation and boundary conditions, with the goal of minimizing the loss function, and use intelligent algorithms to optimize the hyperparameter combination to obtain the final settlement prediction value.

9. A computer-readable storage medium having a program stored thereon, characterized in that: When the program is executed by a processor, the steps of the intelligent prediction method for settlement of a TBM passing through a pebble formation as described in any one of claims 1 to 7 are implemented.

10. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the intelligent prediction method for settlement of a TBM passing through a pebble formation as described in any one of claims 1 to 7 are implemented.

Images