A method and related equipment for predicting the floor heave level of gently dipping layered tunnels

A tunnel floor heave level prediction model was established using FDEM numerical software and XGBoost machine learning algorithm. By utilizing the depth of the disturbance zone at the bottom of the tunnel and the relative displacement index, the accuracy and efficiency problems of floor heave level prediction in traditional methods were solved, achieving high-precision and rapid automated assessment, which is suitable for construction safety management under different geological conditions.

CN123116407APending Publication Date: 2026-11-17INST OF ROCK & SOIL MECHANICS CHINESE ACAD OF SCI +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611344192.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-01
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Traditional methods are difficult to accurately and efficiently predict the floor heave grade of gently dipping layered tunnels, especially under complex geological conditions where their generalization ability is poor and they lack an intelligent classification mechanism, leading to construction safety hazards and difficulties in management.

Method used

Using a data-driven approach, combining FDEM numerical software and XGBoost machine learning algorithm, a tunnel floor heave level prediction model is established through tunnel excavation model simulation and machine learning training. The depth and relative displacement of the disturbance zone at the bottom of the tunnel are used as evaluation indicators to construct a standardized labeling system and achieve automated prediction.

Benefits of technology

It significantly improves the accuracy and speed of bottom heave level prediction, reduces human experience errors, supports real-time assessment and rapid early warning, reduces construction safety hazards and engineering treatment costs, and is adaptable to different geological types and construction methods.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123116407A_ABST
    Figure CN123116407A_ABST
Patent Text Reader

Abstract

This invention discloses a method and related equipment for predicting the floor heave level of gently dipping layered tunnels, relating to the field of tunnel engineering. The method includes: acquiring basic information of known gently dipping layered tunnels and determining known values ​​of target influencing factors; using target simulation software to establish a tunnel excavation model for simulation based on the known values ​​of the target influencing factors to obtain numerical simulation results, determining the depth of the disturbance zone at the bottom of the tunnel and extracting the maximum longitudinal displacement value at the bottom of the tunnel as the floor heave amount; combining known target factors and corresponding tunnel floor heave levels to establish a database, and using a target machine learning algorithm to establish a tunnel floor heave prediction model; acquiring the actual numerical information corresponding to the target influencing factors of the target tunnel and inputting it into the tunnel floor heave prediction model to obtain the floor heave level prediction results of the target tunnel. These results can serve as a preliminary reference for support design optimization and risk management, identifying floor heave risks in advance and formulating reasonable support schemes, effectively reducing construction safety hazards and engineering management costs.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This specification relates to the field of tunnel engineering, and more specifically, this application relates to a method and related equipment for predicting the floor heave level of gently dipping layered tunnels. Background Technology

[0002] Tunnel engineering projects often face the risk of floor heave when traversing gently dipping layered geological formations (i.e., strata with small dip angles and layered rock distribution). Floor heave refers to the upward bulging of the tunnel floor due to stress redistribution, surrounding rock instability, or water pressure, and is characterized by its suddenness, destructiveness, and difficulty in remediation. Once floor heave occurs, it not only affects the construction progress but may also trigger tunnel collapse accidents, endangering the safety of personnel and equipment.

[0003] Traditional methods for determining the level of tunnel floor heave often rely on manual experience or simplified engineering analogies, which limits their accuracy and adaptability. Some studies predict floor heave risk by establishing geomechanical models or conducting numerical simulations, but the selection of parameters is highly subjective, and there is a lack of systematic modeling of influencing factors; the prediction process is computationally complex and inefficient, making it difficult to meet the needs of real-time assessment; the prediction results have errors compared to actual engineering performance, especially with poor generalization ability under complex geological conditions; and there is a lack of intelligent classification mechanisms, with floor heave level assessment standards being scattered and dependent on human judgment.

[0004] Especially in gently sloping layered tunnels, the geological structure is highly continuous, and the factors affecting the occurrence of floor heave are complex and intertwined. For example, the coupling effect of factors such as rock strength, bedding angle, water pressure state, and burial depth variation is particularly significant, making it difficult for traditional methods to accurately model and efficiently predict.

[0005] Therefore, there is an urgent need for a method that integrates data-driven and physical modeling to make high-precision, fast, and automated predictions of the floor heave level, so as to achieve proactive control over tunnel construction safety. Summary of the Invention

[0006] The summary section introduces a series of simplified concepts, which will be further explained in detail in the detailed description section. This summary section is not intended to limit the key and essential technical features of the claimed technical solution, nor is it intended to determine the scope of protection of the claimed technical solution.

[0007] Firstly, this application proposes a method for predicting the floor heave level of gently dipping layered tunnels, including: Obtain basic information about known gently dipping layered tunnels and determine known values ​​of target influencing factors; A tunnel excavation model is established using target software. The known values ​​of the aforementioned known target influencing factors are used to perform simulation using target simulation software to obtain numerical simulation results. Based on the numerical simulation results, the depth of the disturbance zone at the bottom of the tunnel is determined, and the maximum longitudinal displacement value at the bottom of the tunnel is extracted as the bottom heave. Obtain bass drum grading information; A database is established based on the above known values, the above floor heave amount, the above depth of the disturbance zone at the bottom of the tunnel, and the above floor heave classification information. A target machine learning algorithm is used to learn from the dataset in the above database to establish a tunnel floor heave prediction model; Obtain the actual numerical information corresponding to the above-mentioned target influencing factors of the target tunnel; The above actual numerical information is input into the above tunnel floor heave prediction model to obtain the tunnel floor heave level of the target tunnel.

[0008] In one feasible implementation, obtaining the bass drum level classification information includes: relative displacement at the bottom of the tunnel and the depth of the disturbance zone at the bottom of the tunnel The base drum is classified into different grades based on the indicators.

[0009] In one feasible implementation, The above-mentioned relative displacement at the bottom of the tunnel Calculated based on the following formula:

[0010] The equivalent radius of the above tunnel Calculated based on the following formula:

[0011] The depth of the disturbance zone at the bottom of the aforementioned tunnel Calculated based on the following formula:

[0012] in, The cross-sectional area of ​​the tunnel. , These are the x and y coordinates of a fully damaged joint element, respectively. , These represent the x and y coordinates of a fully damaged rock element; , , where are the horizontal and vertical coordinates of the tunnel center, and D is the aforementioned under-bulging amount.

[0013] In one feasible implementation, the aforementioned target machine learning algorithm includes XGBoost.

[0014] In one feasible implementation, the target software includes FDEM numerical software.

[0015] In one feasible implementation, the aforementioned influencing factors include rock stratum thickness, tunnel depth, horizontal lateral pressure coefficient, and uniaxial compressive strength of rock.

[0016] In one feasible implementation, during the learning process of the aforementioned tunnel floor heave prediction model, a random grid search is used to determine the hyperparameter values.

[0017] Secondly, the present invention also proposes a device for predicting the floor heave level of gently dipping layered tunnels, comprising: The first acquisition unit is used to acquire basic information about known gently dipping layered tunnels and determine known values ​​of target influencing factors. The second acquisition unit is used to establish a tunnel excavation model using the target software, and to perform simulation using the known values ​​of the aforementioned known target influencing factors to obtain numerical simulation results. Based on the aforementioned numerical simulation results, the depth of the disturbance zone at the bottom of the tunnel is determined, and the maximum longitudinal displacement value at the bottom of the tunnel is extracted as the bottom heave. The third acquisition unit is used to acquire information on the classification of the bass drum level. The first establishment unit is used to establish a database based on the above-mentioned known value information, the above-mentioned bottom heave amount, the above-mentioned depth of the disturbance zone at the bottom of the tunnel, and the above-mentioned bottom heave level classification information; The second establishment unit is used to learn from the dataset in the above database using a target machine learning algorithm to establish a tunnel floor drum level prediction model. The fourth acquisition unit is used to acquire the actual numerical information corresponding to the aforementioned target influencing factors of the target tunnel; The fifth acquisition unit is used to input the above-mentioned actual numerical information into the above-mentioned tunnel floor heave prediction model in order to obtain the tunnel floor heave level of the target tunnel.

[0018] Thirdly, the present invention also proposes an electronic device comprising: a memory and a processor, characterized in that the processor is configured to execute a computer program stored in the memory to implement the steps of the method for predicting the floor heave level of gently sloping layered tunnels as described in any of the first aspects.

[0019] Fourthly, the present invention also proposes a computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of the method for predicting the floor heave level of gently dipping layered tunnels as described in any one of the first aspects.

[0020] In summary, this embodiment proposes a method for predicting the floor heave level of gently dipping layered tunnels. This invention utilizes machine learning technology to construct a nonlinear mapping model, automatically learning the complex relationship between influencing factors and floor heave level, significantly improving prediction accuracy and avoiding the errors and subjectivity caused by "human experience judgment" in traditional methods. During the prediction model establishment stage, numerical simulation software is used to generate floor heave data. By simulating tunnel deformation behavior under different geological conditions, the training data is ensured to have physical rationality and engineering representativeness, effectively solving the problem of insufficient training samples. After model training is completed, new tunnel parameters can be input into the model and output within seconds, saving a significant amount of simulation time compared to traditional numerical simulation processes, meeting the needs of rapid on-site assessment and early warning. The depth H of the disturbance zone at the bottom of the tunnel and the relative displacement R of the tunnel bottom are used to determine the relationship between the depth H and the relative displacement R of the tunnel bottom. d As a dual-indicator evaluation criterion, a standardized labeling system is formed in conjunction with the floor heave grading standard to improve the objectivity and consistency of floor heave grading. The training model can be continuously updated with new engineering data, supporting generalized predictions under different geological types and construction methods, and has broad application value. The prediction results can serve as a preliminary reference for support design optimization and risk management, identifying floor heave risks in advance, formulating reasonable support schemes, and effectively reducing construction safety hazards and engineering treatment costs.

[0021] Other advantages, objectives and features of this application will be apparent in part from the description which follows, and in part from what those skilled in the art will understand through study and practice of this application. Attached Figure Description

[0022] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit this specification. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A schematic flowchart of a method for predicting the floor heave level of a gently dipping layered tunnel provided in this application embodiment; Figure 2 A structural schematic diagram of a gently sloping layered tunnel floor heave level prediction device provided in this application embodiment; Figure 3 This is a schematic diagram of an electronic device structure provided in an embodiment of this application. Detailed Implementation

[0023] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus. The technical solutions of the embodiments of this application will now be clearly and completely described in conjunction with the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them.

[0024] Please see Figure 1 This is a flowchart illustrating a method for predicting the floor heave level of a gently dipping layered tunnel according to an embodiment of this application. Specifically, it may include: Firstly, this application proposes a method for predicting the floor heave level of gently dipping layered tunnels, including: S110. Obtain basic information about the known gently dipping layered tunnel and determine the known values ​​of the target influencing factors; S120. Use the target software to establish a tunnel excavation model, use the known values ​​of the above-mentioned known target influencing factors to perform simulation using the target simulation software to obtain numerical simulation results, determine the depth of the disturbance zone at the bottom of the tunnel based on the above-mentioned numerical simulation results, and extract the maximum longitudinal displacement value at the bottom of the tunnel as the bottom heave amount. S130, Obtain bass drum level classification information; S140. Based on the above known values, the above floor heave amount, the above tunnel bottom disturbance zone depth, and the above floor heave classification information, establish a database; S150. Use the target machine learning algorithm to learn from the dataset in the above database and establish a tunnel floor heave prediction model. S160. Obtain the actual numerical information corresponding to the above-mentioned target influencing factors of the target tunnel; S170. Input the above actual numerical information into the above tunnel floor heave prediction model to obtain the tunnel floor heave level of the target tunnel.

[0025] For example, firstly, basic information on gently dipping layered tunnels is extracted and summarized from historically known projects. This information covers several key geological and engineering factors that influence the bottom heave phenomenon, such as: rock layer thickness, tunnel depth, stratum dip angle, rock strength, and horizontal lateral pressure coefficient. This data constitutes the feature vector of influencing factors, which serves as the input basis for subsequent modeling.

[0026] Next, a tunnel excavation model was established using numerical simulation software (such as FLAC3D, ABAQUS, FDEM, etc.). The above-mentioned characteristic parameters were used as input conditions to perform numerical simulation of the entire tunnel excavation process. The depth of the disturbance zone at the bottom of the tunnel was extracted and the maximum vertical displacement at the bottom of the tunnel was used as the bottom heave amount to measure the severity of the bottom heave phenomenon.

[0027] Based on the simulation results of the tunnel bottom disturbance zone depth and floor heave amount, and combined with existing floor heave classification standards, each sample is labeled with its corresponding floor heave level. Floor heave levels are generally divided into: no floor heave, slight floor heave (Level I), moderate floor heave (Level II), and severe floor heave (Level III). The classification criteria comprehensively consider the tunnel bottom disturbance zone depth H and the relative displacement R of the tunnel bottom. d Two indicators.

[0028] Subsequently, the aforementioned input parameters (target influencing factors) and bass drum level labels were organized into a standardized structure to construct a database, and the training sample set and test set of the machine learning model were divided in an 8:2 ratio.

[0029] High-performance machine learning algorithms such as XGBoost (Extreme Gradient Boosting) are introduced to train the model on a pre-constructed training set, and the test set is used to verify the model's accuracy, thereby establishing a tunnel floor drum level prediction model. This model can automatically capture the complex nonlinear relationship between input parameters and floor drum level, significantly improving prediction accuracy and generalization ability.

[0030] After the model training is completed, the prediction phase begins. For a specific target tunnel to be evaluated, its actual geological and structural parameters are collected as input features for the sample to be predicted.

[0031] Finally, these actual characteristics are input into the trained prediction model, which will automatically output the predicted result of the floor heave level of the tunnel, thereby achieving a fast and accurate floor heave classification assessment and assisting the construction unit in formulating targeted support measures or setting up early warning mechanisms in advance.

[0032] In summary, this invention utilizes machine learning technology to construct a nonlinear mapping model, automatically learning the complex relationship between influencing factors and floor heave levels, significantly improving prediction accuracy and avoiding the errors and subjectivity caused by "human experience judgment" in traditional methods. During the prediction model establishment phase, numerical simulation software is used to generate floor heave data. By simulating tunnel deformation behavior under different geological conditions, the physical rationality and engineering representativeness of the training data are ensured, effectively solving the problem of insufficient training samples. After model training is completed, new tunnel parameters can be input into the model and output within seconds, saving a significant amount of simulation time compared to traditional numerical simulation processes, meeting the needs for rapid on-site assessment and early warning. The depth H of the disturbance zone at the tunnel bottom and the relative displacement R of the tunnel bottom are compared. d As a dual-indicator evaluation criterion, a standardized labeling system is formed in conjunction with the floor heave grading standard to improve the objectivity and consistency of floor heave grading. The training model can be continuously updated with new engineering data, supporting generalized predictions under different geological types and construction methods, and has broad application value. The prediction results can serve as a preliminary reference for support design optimization and risk management, identifying floor heave risks in advance, formulating reasonable support schemes, and effectively reducing construction safety hazards and engineering treatment costs.

[0033] In one feasible implementation, obtaining the bass drum level classification information includes: relative displacement at the bottom of the tunnel and the depth of the disturbance zone at the bottom of the tunnel The base drum is classified into different grades based on the indicators.

[0034] For example, the classification standard for tunnel floor heave can be determined according to Table 1.

[0035]

[0036] Table 1 Classification Standards for Tunnel Floor Heave This embodiment introduces the depth H of the disturbance zone at the bottom of the tunnel and the relative displacement R at the bottom of the tunnel. d Two structural indicators enable a quantitative description and graded assessment of floor heave, overcoming the problems of strong subjectivity and inconsistency in traditional grading methods. This standard combines the dimensions of displacement ratio and disturbance range, adapting to the development characteristics of floor heave under different geological types (such as weak surrounding rock, gently dipping layers, and water-rich strata), thus improving its applicability and engineering suitability. Based on formulaic expression, this grading standard can be directly embedded into tunnel deformation monitoring systems or intelligent simulation platforms, combining sensor or numerical simulation results to output the floor heave level in real time, meeting the needs of intelligent construction management.

[0037] In one feasible implementation, the aforementioned relative displacement of the tunnel bottom Calculated based on the following formula:

[0038] The equivalent radius of the above tunnel Calculated based on the following formula:

[0039] The depth of the disturbance zone at the bottom of the aforementioned tunnel Calculated based on the following formula:

[0040] in, The cross-sectional area of ​​the tunnel. , These are the x and y coordinates of a fully damaged joint element, respectively. , These represent the x and y coordinates of a fully damaged rock element; , , where are the horizontal and vertical coordinates of the tunnel center, and D is the aforementioned under-bulging amount.

[0041] For example, this embodiment constructs mathematical expressions to accurately geometrically model and relatively evaluate the shape of the tunnel floor heave, providing a solid quantitative basis for subsequent floor heave level identification, support design adjustment and risk warning. It is a typical demonstration of integrating traditional rock and soil deformation monitoring engineering with modern mathematical modeling methods, and has high practicality and promotion value.

[0042] In one feasible implementation, the aforementioned target machine learning algorithm includes XGBoost.

[0043] In one feasible implementation, the target software includes FDEM numerical software.

[0044] In one feasible implementation, during the learning process of the aforementioned tunnel floor heave prediction model, a random grid search is used to determine the hyperparameter values.

[0045] For example, the embodiment constructs an intelligent modeling process for predicting the level of tunnel floor heave by integrating FDEM numerical simulation software, XGBoost machine learning algorithm and random grid search parameter tuning method, aiming to achieve high-precision automated identification of the level of tunnel floor heave under gently dipping layered geological conditions.

[0046] Specifically, the tunnel excavation process was first simulated using FDEM (Finite-Discrete Element Method) numerical software. FDEM combines the precise calculation capabilities of the finite element method for continuous medium deformation with the descriptive capabilities of the discrete element method for crack propagation and structural separation, making it particularly suitable for simulating nonlinear large deformation behaviors such as floor heave. In the actual modeling process, multiple excavation models under different working conditions were constructed in FDEM based on different geological parameter configurations (such as rock layer thickness, tunnel depth, bedding angle, rock strength, lateral pressure coefficient, etc.), and corresponding boundary and load conditions were applied for simulation. The maximum longitudinal displacement D at the bottom and the depth H of the disturbed zone were extracted from the simulation results, and the relative displacement R at the bottom of the tunnel was further calculated. d And according to the criteria for judging the bass drum grade (such as R) d Each set of working condition data is labeled with a level (e.g., no bottoming, slight, moderate, severe) using H threshold.

[0047] Subsequently, the large amount of sample data obtained from the above simulation was used to construct a structured training dataset. Each sample contains multiple input features (rock parameters, burial depth, water pressure, etc.) and a floor heave level label. Based on this, the XGBoost (eXtreme Gradient Boosting) algorithm was used to construct a tunnel floor heave level classification model. As an ensemble boosting algorithm, XGBoost can progressively optimize the residuals by training weak classifiers (decision trees) through multiple rounds of iteration, capturing the nonlinear and complex relationship between input features and the prediction target.

[0048] To further improve model performance and avoid the inefficiencies and large biases caused by manual parameter tuning, a random grid search mechanism is introduced during model training. By setting candidate value ranges for hyperparameters (such as learning rate, tree depth, subsample ratio, etc.), combinations of hyperparameters are randomly sampled, and the performance of each combination is evaluated on the validation set. Finally, the parameter configuration that optimizes accuracy or F1-score is selected for model training, thereby achieving automatic parameter tuning and accuracy optimization.

[0049] After the model training is complete, the actual geological parameters of the tunnel to be evaluated can be input into the trained model, and the system will automatically output the predicted floor heave level for that tunnel. This data-driven prediction method can significantly improve the speed and accuracy of prediction and is applicable to multiple stages such as pre-construction assessment, design optimization, and construction monitoring and early warning.

[0050] In summary, this embodiment generates high-quality physical samples through FDEM simulation, establishes an accurate classification model using XGBoost, and achieves efficient parameter tuning through random grid search. This not only effectively improves the accuracy and engineering adaptability of tunnel floor heave level prediction, but also has good real-time performance and automation capabilities, and has broad engineering promotion value and application prospects.

[0051] In one feasible implementation, the aforementioned influencing factors include rock stratum thickness, tunnel depth, horizontal lateral pressure coefficient, and uniaxial compressive strength of rock.

[0052] For example, the thickness of rock strata affects the stability of the surrounding rock after tunnel excavation. For gently dipping layered strata, if the rock strata are thin, they are prone to bending, fracturing, or interlayer slippage. After tunnel excavation, the bottom rock strata bulge upwards due to stress release, forming a bottom bulge.

[0053] Tunnel burial depth directly affects the magnitude of the original ground stress. The greater the burial depth, the higher the vertical stress, and the greater the risk of floor heave. Under the same surrounding rock conditions, deeply buried tunnels are more prone to floor heave or floor slab uplift during the excavation stage due to uneven stress release. In the modeling process, burial depth is used as a continuous variable input, which helps the model capture the evolution law of "stress-dominated" floor heave.

[0054] The Lateral Pressure Coefficient (K) is the ratio of lateral stress to vertical stress (K = σ). h / σ v The K value reflects the strength of horizontal stress constraint on the surrounding rock. When the K value is large (>1), the horizontal compressive stress increases, which may exacerbate the development of bottom heave through the squeezing effect. In the model, the K value can be estimated by combining field monitoring or geological inversion results, serving as an important mechanical input parameter.

[0055] Uniaxial compressive strength (UCS) of rock reflects its ability to resist compressive failure and is an important indicator for assessing the deformation resistance of the base plate. Soft rocks with low UCS are more prone to dilatation or shear failure under stress concentration or water pressure, manifesting as floor bulging. UCS can be obtained as an input feature from uniaxial compression tests or empirical formulas, or typical values ​​can be extracted from databases.

[0056] In this embodiment, one or more of the aforementioned target influencing factors will be used as input feature vectors, combined with the floor heave displacement, disturbance depth, and grade labels extracted from FDEM simulation or field data, and input into the XGBoost model for learning. The model will automatically capture the complex nonlinear relationship between these influencing factors and the floor heave grade, and after training, it can predict the grade of new tunnel samples.

[0057] It should be noted that, if indoor testing is feasible, the mechanical parameters (tensile strength, Poisson's ratio, cohesion, and internal friction angle) of the tunnel surrounding rock should be determined based on uniaxial compression tests, Brazilian splitting tests, and shear tests. If the existing conditions do not allow for testing, the mechanical parameters (tensile strength, Poisson's ratio, cohesion, and internal friction angle) of the tunnel surrounding rock can be determined according to the relevant standards "Code for Design of Railway Tunnels" TB10003-2016 and "Standard for Classification of Engineering Rock Mass" GB / T 50218-2014.

[0058] In this embodiment, rock stratum thickness, tunnel burial depth, horizontal lateral pressure coefficient, and uniaxial compressive strength of rock are explicitly identified as key influencing factors. This not only demonstrates the model's ability to accurately depict engineering geological mechanisms but also greatly enhances its practicality, flexibility, and prediction accuracy, making it of significant engineering value in scenarios such as tunnel design evaluation, safety early warning, and intelligent construction.

[0059] Secondly, this invention also proposes a device for predicting the floor heave level of gently dipping layered tunnels, such as... Figure 2 As shown, it includes: The first acquisition unit 21 is used to acquire basic information of known gently dipping layered tunnels and determine known value information of target influencing factors; The second acquisition unit 22 is used to establish a tunnel excavation model using the target software, and to perform simulation using the known value information of the known target influencing factors to obtain numerical simulation results. The numerical simulation results determine the depth of the disturbance zone at the bottom of the tunnel, and extract the maximum longitudinal displacement value at the bottom of the tunnel as the bottom heave. The third acquisition unit 23 is used to acquire bass drum level classification information; The first establishment unit 24 is used to establish a database based on the above-mentioned known value information, the above-mentioned bottom heave amount, the above-mentioned depth of the disturbance zone at the bottom of the tunnel and the above-mentioned bottom heave level classification information; The second establishment unit 25 is used to learn from the dataset in the above database using a target machine learning algorithm to establish a tunnel floor drum prediction model. The fourth acquisition unit 26 is used to acquire the actual numerical information corresponding to the above-mentioned target influencing factors of the target tunnel; The fifth acquisition unit 27 is used to input the above-mentioned actual numerical information into the above-mentioned tunnel floor heave prediction model in order to obtain the tunnel floor heave level of the target tunnel.

[0060] The aforementioned device for predicting the floor heave level of gently sloping layered tunnels can also perform the following operations: In one feasible implementation, obtaining the bass drum level classification information includes: relative displacement at the bottom of the tunnel and the depth of the disturbance zone at the bottom of the tunnel The base drum is classified into different grades based on the indicators.

[0061] In one feasible implementation, the aforementioned relative displacement of the tunnel bottom Calculated based on the following formula:

[0062] The equivalent radius of the above tunnel Calculated based on the following formula:

[0063] The depth of the disturbance zone at the bottom of the aforementioned tunnel Calculated based on the following formula:

[0064] in, The cross-sectional area of ​​the tunnel. , These are the x and y coordinates of a fully damaged joint element, respectively. , These represent the x and y coordinates of a fully damaged rock element; , , where are the horizontal and vertical coordinates of the tunnel center, and D is the aforementioned under-bulging amount.

[0065] In one feasible implementation, the aforementioned target machine learning algorithm includes XGBoost.

[0066] In one feasible implementation, the target software includes FDEM numerical software.

[0067] In one feasible implementation, the aforementioned influencing factors include rock stratum thickness, tunnel depth, horizontal lateral pressure coefficient, and uniaxial compressive strength of rock.

[0068] In one feasible implementation, during the learning process of the aforementioned tunnel floor heave prediction model, a random grid search is used to determine the hyperparameter values.

[0069] Thirdly, the present invention also proposes an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor. When the processor 320 executes the computer program 311, it implements the steps of the method for predicting the floor heave level of gently sloping layered tunnels as described in any of the first aspects.

[0070] Fourthly, the present invention also proposes a computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of the method for predicting the floor heave level of gently dipping layered tunnels as described in any one of the first aspects.

[0071] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0072] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0073] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0074] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0075] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0076] This application also provides a computer program product, which includes computer software instructions. When the computer software instructions are executed on a processing device, the processing device performs the voice-based identity recognition process in the corresponding embodiment. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0077] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0078] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between apparatuses or units, and may be electrical, mechanical, or other forms.

[0079] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0080] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0081] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0082] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for predicting the floor heave level of gently dipping layered tunnels, characterized in that, include: Obtain basic information about known gently dipping layered tunnels and determine known values ​​of target influencing factors; A tunnel excavation model is established using target software. The known values ​​of the known target influencing factors are used to perform simulation using target simulation software to obtain numerical simulation results. The depth of the disturbance zone at the bottom of the tunnel is determined based on the numerical simulation results, and the maximum longitudinal displacement value at the bottom of the tunnel is extracted as the bottom heave. Obtain bass drum grading information; A database is established based on the known values, the heave volume, the depth of the disturbance zone at the bottom of the tunnel, and the heave level classification information. A target machine learning algorithm is used to learn from the dataset in the database to establish a tunnel floor heave prediction model; Obtain the actual numerical information corresponding to the target influencing factors of the target tunnel; The actual numerical information is input into the tunnel floor heave prediction model to obtain the floor heave level of the target tunnel.

2. The method for predicting the floor heave level of gently dipping layered tunnels according to claim 1, characterized in that, The acquisition of bass drum level classification information includes: relative displacement at the bottom of the tunnel and the depth of the disturbance zone at the bottom of the tunnel The base drum is classified into different grades based on the indicators.

3. The method for predicting the floor heave level of gently dipping layered tunnels according to claim 2, characterized in that, The relative displacement at the bottom of the tunnel Calculated based on the following formula: equivalent radius of tunnel Calculated based on the following formula: The depth of the disturbance zone at the bottom of the tunnel Calculated based on the following formula: in, The cross-sectional area of ​​the tunnel. , These are the x and y coordinates of a fully damaged joint element, respectively. , These represent the x and y coordinates of a fully damaged rock element; , , where are the horizontal and vertical coordinates of the tunnel center, and D is the amount of floor bulge.

4. The method for predicting the floor heave level of gently dipping layered tunnels according to claim 1, characterized in that, The target machine learning algorithm includes XGBoost.

5. The method for predicting the floor heave level of gently dipping layered tunnels according to claim 1, characterized in that, The target software includes FDEM numerical software.

6. The method for predicting the floor heave level of gently dipping layered tunnels according to claim 1, characterized in that, The target influencing factors include rock stratum thickness, tunnel burial depth, horizontal lateral pressure coefficient, and uniaxial compressive strength of rock.

7. The method for predicting the floor heave level of gently dipping layered tunnels according to claim 1, characterized in that, During the learning process of the tunnel floor heave prediction model, a random grid search is used to determine the hyperparameter values.

8. A device for predicting the level of floor heave in gently sloping layered tunnels, characterized in that, include: The first acquisition unit is used to acquire basic information about known gently dipping layered tunnels and determine known values ​​of target influencing factors. The second acquisition unit is used to establish a tunnel excavation model using target software, and to perform simulation using target simulation software with the known values ​​of the known target influencing factors to obtain numerical simulation results. Based on the numerical simulation results, the depth of the disturbance zone at the bottom of the tunnel is determined, and the maximum longitudinal displacement value at the bottom of the tunnel is extracted as the bottom heave. The third acquisition unit is used to acquire information on the classification of the bass drum level. The first establishment unit is used to establish a database based on the known value information, the floor heave amount, the depth of the disturbance zone at the bottom of the tunnel, and the floor heave level classification information; The second establishment unit is used to learn from the dataset in the database using a target machine learning algorithm to establish a tunnel floor drum prediction model; The fourth acquisition unit is used to acquire the actual numerical information corresponding to the target influencing factors of the target tunnel; The fifth acquisition unit is used to input the actual numerical information into the tunnel floor heave prediction model to obtain the tunnel floor heave level of the target tunnel.

9. An electronic device, comprising: The memory and processor are characterized in that the processor, when executing a computer program stored in the memory, implements the steps of the method for predicting the floor heave level of gently sloping layered tunnels as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for predicting the floor heave level of gently dipping layered tunnels as described in any one of claims 1-7.