Shield tunnel segment floating prediction method, device and equipment
By establishing a pipe-filled floating prediction model based on CPO-LSSVM and Adaboost integrated learner, the problem of insufficient prediction accuracy of the pipe-filled floating prediction in the prior art is solved, and higher prediction accuracy and stability are achieved.
Patent Information
- Application Number
- CN202411947483.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-27
- Publication Date
- 2025-04-22
AI Technical Summary
The existing shield tunnel pipe sheet floating prediction methods are insufficient in accuracy and reliability. The traditional method relies on theoretical formulas, rules of thumb or numerical models. It is difficult to obtain parameters and time-consuming to calculate, resulting in deviations in prediction results.
A prediction regression model based on CPO-LSSVM is adopted, combined with Adaboost integrated learner, optimizes hyperparameters through 50% cross-validation, and builds an Adaboost-CPO-LSSVM model to predict the floating amount of pipe slices.
The accuracy and stability of the floating prediction of shield tunnel pipe sheets are improved. The model has good prediction effect on the test set and is suitable for shield tunnel projects of different types and sizes.
Smart Images
Figure CN119862548A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of shield tunnel construction, and in particular to a shield tunnel segment floating prediction method, device and equipment. Background Art
[0002] With the continuous development of urban underground transportation systems and infrastructure construction, shield tunnel construction technology has been widely used in related tunnel construction. The shield method has significant advantages under conditions of complex geology and difficult ground construction, but during the construction process, the floating phenomenon of shield tunnel segments has always been a key factor affecting construction safety and progress. As a typical construction deformation phenomenon, the floating of shield tunnel segments may cause damage to the segments, loose joints, and even affect the safety and stability of the tunnel. A reasonable and effective segment floating prediction method can provide guidance for the construction site and is of great significance to the construction of shield tunnels.
[0003] The occurrence of shield tunnel segment floating is closely related to multiple factors, such as shield tunneling parameters, ground conditions, tunnel geometry conditions, etc. The complex relationship between these factors makes it difficult for traditional prediction methods to fully and accurately reflect the actual situation. Existing segment floating prediction methods mostly rely on theoretical formulas, empirical rules or numerical models. Theoretical formulas and related empirical rules often require more parameters for prediction, and the relevant parameters are difficult or costly to obtain, resulting in insufficient accuracy and reliability of the prediction results; the calculation of numerical models requires simplified processing of on-site conditions, and modeling and subsequent calculations are time-consuming and labor-intensive, and the prediction results are also deviated from the actual situation.
[0004] In recent years, with the rapid development of machine learning technology, data-driven prediction methods have gradually become an important means to solve such problems. As a powerful supervised learning model, the least squares support vector machine (LSSVM) has been widely used in regression and classification tasks and has achieved remarkable results. However, the use of LSSVM alone is often difficult to meet the prediction accuracy of segment floating in shield tunnel construction.
[0005] The background description provided herein is for the purpose of generally presenting the context of the present disclosure. Unless otherwise indicated herein, the materials described in this section are not prior art to the claims of this application and are not admitted to be prior art by inclusion in this section. Summary of the invention
[0006] In view of the above technical problems in the related art, the present invention proposes a shield tunnel segment floating prediction method, device and equipment.
[0007] In a first aspect, the present invention provides a method for predicting the floating of a shield tunnel segment, comprising the following steps:
[0008] S1. Constructing a segment floating prediction model database based on the collected data, wherein the collected data includes segment floating amount, engineering geological parameters, excavation parameters and tunnel geometric parameters;
[0009] S2. Establish a prediction regression model based on CPO-LSSVM, determine the hyperparameter optimization range in LSSVM and the parameters of the CPO optimization algorithm, and use 5-fold cross validation to optimize the hyperparameters;
[0010] S3, using the Adaboost ensemble learner to integrate multiple weak learners into the predictive regression model to obtain the Adaboost-CPO-LSSVM model;
[0011] S4, using 80% of the collected data to train the Adaboost-CPO-LSSVM model, and using the remaining 20% of the collected data to test the Adaboost-CPO-LSSVM model to obtain model evaluation indicators; the model evaluation indicators include determination coefficient, root mean square and mean absolute error;
[0012] S5. Input the newly collected parameters into the Adaboost-CPO-LSSVM model to obtain the predicted value of the segment floating amount.
[0013] Specifically, step S2 specifically includes:
[0014] S21, determine the LSSVM model structure: initialize the LSSVM model, determine the kernel function and initial hyperparameters; the kernel function is a radial basis function RBF; the hyperparameters include a regularization parameter γ and a width parameter σ of the Gaussian kernel function 2 ;
[0015] S22, setting the hyperparameter optimization range of LSSVM and CPO optimization algorithm parameters; the two hyperparameters of LSSVM, the regularization parameter γ and the width parameter σ of the Gaussian kernel function 2 ; The CPO optimization algorithm parameters include the initial population number and the maximum number of iterations;
[0016] S23, using CPO optimization algorithm to search for the optimal hyperparameters within the hyperparameter optimization range of LSSVM using 5-fold cross validation;
[0017] Specifically, the search range of the regularization parameter γ is [0.001, 1000], and the width parameter σ of the Gaussian kernel function is 2 The search range is [0.001, 100]; the initial population size is 20, and the maximum number of iterations is 100.
[0018] Specifically, step S3 specifically includes:
[0019] S31, based on the prediction regression model of CPO-LSSVM, generate N weak learners with different data partitions;
[0020] S32, initialize the weight of each weak learner to 1 / N;
[0021] S33, for each weak learner, use the weighted data set to perform iterative training, and calculate its prediction error on the training set;
[0022] S34, dynamically adjust the weight of each weak learner according to the prediction error; the learner with a smaller prediction error obtains a higher weight so as to play a greater role in the final prediction;
[0023] S35, repeating steps S33 to S34 until the number of iterations of the weak learner reaches a preset maximum number of iterations;
[0024] S36. Use weighted voting or weighted averaging to combine the prediction results of multiple weak learners to obtain a final prediction value.
[0025] Specifically, step S4 is as follows:
[0026] 80% of the collected data were randomly selected from the segment flotation prediction model database as a training set, and the segment engineering geological parameters, excavation parameters and tunnel geometric parameters in the training set were used as input, and the segment flotation height was used as output to train the Adaboost-CPO-LSSVM model. After the model training was completed, the remaining 20% of the collected data were used as a test set to test the Adaboost-CPO-LSSVM model. Finally, the model evaluation indicators, including the determination coefficient, root mean square error and mean absolute error, were calculated to evaluate the prediction performance of the model.
[0027] Specifically, step S5 specifically includes the following steps:
[0028] S51, obtaining new collection parameters, including segment floating volume, engineering geological parameters, excavation parameters and tunnel geometric parameters;
[0029] S52. Input the new collected parameters into the trained Adaboost-CPO-LSSVM model to obtain the predicted value of the floating amount of the pipe segment.
[0030] Specifically, step S5 also includes:
[0031] S53, comparing the predicted value of the buoyancy of the pipe segment with the buoyancy of the pipe segment actually measured;
[0032] S54. Analyze the difference between the predicted value and the true value to evaluate the prediction accuracy and reliability of the model.
[0033] Specifically, step S1 is as follows:
[0034] The collected data are cleaned, outliers and missing values are removed, and normalized to ensure that the data are on the same scale to ensure numerical stability and convergence during model training. The processed collected data are then stored in the database.
[0035] Specifically, the normalization process is to use Z-score normalization to process the collected data.
[0036] Specifically, the engineering geological parameters include cohesion, internal friction angle, permeability coefficient, compression modulus and soil weight; the excavation parameters include propulsion speed, cutter head torque, total propulsion force, propulsion pressure, average soil bin pressure, average grouting pressure, grouting pressure difference, shield machine pitch angle and grouting volume; the tunnel geometric parameters include tunnel burial depth, shield diameter and segment width.
[0037] In a second aspect, the present invention provides a shield tunnel segment floating prediction device, based on a shield tunnel segment floating prediction method described in any one of the first aspects, comprising the following units:
[0038] A database construction unit, used to construct a segment floating prediction model database based on the collected data, wherein the collected data includes segment floating amount, engineering geological parameters, excavation parameters and tunnel geometric parameters;
[0039] The prediction regression model unit is used to establish a prediction regression model based on CPO-LSSVM, determine the hyperparameter optimization range in LSSVM and the parameters of the CPO optimization algorithm, and use 5-fold cross validation to perform hyperparameter optimization;
[0040] The prediction regression model unit is optimized to integrate multiple weak learners of the prediction regression model using the Adaboost ensemble learner to obtain the Adaboost-CPO-LSSVM model;
[0041] A training unit, used to train the Adaboost-CPO-LSSVM model using 80% of the collected data, and to test the Adaboost-CPO-LSSVM model using the remaining 20% of the collected data to obtain model evaluation indicators; the model evaluation indicators include determination coefficient, root mean square and mean absolute error;
[0042] The prediction unit is used to input the newly collected parameters into the Adaboost-CPO-LSSVM model to obtain the predicted value of the floating amount of the pipe segment.
[0043] In a third aspect, the present invention also provides an electronic device comprising a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the program includes instructions for executing the steps in the method described in any one of the first aspects.
[0044] The present invention provides a method, device and equipment for predicting the floating of a shield tunnel segment. The present invention constructs a segment floating prediction database based on on-site segment floating monitoring data, excavation parameters, geological parameters, etc., and successfully applies it to the prediction method. From the final results, the segment floating prediction method provided by the present invention has good accuracy and stability in the engineering implementation case. The good effect on the test set indicates that the prediction effect of the model on new data also maintains a high level, which further indicates that the model can accurately predict the floating of the segment to a certain extent, and the method is applicable to different shield tunnel projects, and is not restricted by the type, mode and size of the shield, as well as the stratum in which the shield tunnel is located, which is of great significance for shield tunnel construction. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] 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.
[0046] Figure 1 A schematic diagram of a shield tunnel segment floating prediction method provided by an embodiment of the present invention;
[0047] Figure 2 A schematic diagram of the principle of the Adaboost integrated algorithm provided by an embodiment of the present invention;
[0048] Figure 3 A schematic diagram of the CPO-LSSVM-Adaboost model process provided by an embodiment of the present invention;
[0049] Figure 4 A schematic diagram of a shield tunnel segment flotation prediction device provided by an embodiment of the present invention;
[0050] Figure 5 A schematic diagram of a shield tunnel segment flotation prediction device provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0051] The present invention can be explained in detail by the following examples. The purpose of providing the present invention is to protect all technical improvements within the scope of the present invention. In the description of the present invention, it should be understood that the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Thus, the features defined as "first" and "second" can explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.
[0052] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.
[0053] Embodiment 1
[0054] refer to Figure 1 This embodiment provides a shield tunnel segment floating prediction method, comprising the following steps:
[0055] S1. Constructing a segment floating prediction model database based on the collected data, wherein the collected data includes segment floating amount, engineering geological parameters, excavation parameters and tunnel geometric parameters;
[0056] The engineering geological parameters include cohesion, internal friction angle, permeability coefficient, compression modulus and soil weight; the excavation parameters include propulsion speed, cutter head torque, total propulsion force, propulsion pressure, soil bin pressure average, grouting pressure average, grouting pressure difference, shield machine pitch angle and grouting volume; the tunnel geometric parameters include tunnel burial depth, shield diameter and segment width;
[0057] The excavation parameters and the floating volume of the pipe segments are continuously collected during the construction process, and the database is continuously expanded;
[0058] The collected data are cleaned, outliers and missing values are removed, and normalized to ensure that the data are on the same scale to ensure numerical stability and convergence during model training. The processed collected data are then stored in the database.
[0059] The normalization method is selected according to the distribution characteristics of the data, such as minimum-maximum normalization or Z-score normalization; preferably, in this example, Z-score normalization is used to process the collected data.
[0060] Z-score normalization is a method for converting data into a standard normal distribution, which is a prior art and will not be described in detail here.
[0061] This embodiment does not limit the type and storage method of the database.
[0062] S2. Establish a prediction regression model based on CPO-LSSVM, determine the hyperparameter optimization range in LSSVM and the parameters of the CPO optimization algorithm, and use 5-fold cross validation to optimize the hyperparameters;
[0063] Specifically, step S2 specifically includes the following steps:
[0064] S21, determine the LSSVM model structure: initialize the LSSVM model, determine the kernel function and initial hyperparameters; the kernel function is a radial basis function RBF; the hyperparameters include a regularization parameter γ and a width parameter σ of the Gaussian kernel function 2 ;
[0065] LSSVM (Least Squares Support Vector Machine) is an improved support vector machine algorithm with efficient solution method and good generalization performance. It is widely used in pattern recognition and regression analysis. LSSVM realizes nonlinear classification by mapping data into high-dimensional feature space. Compared with traditional SVM, LSSVM uses the least squares loss function, which makes it more stable and less affected by outliers.
[0066] S22, setting the hyperparameter optimization range of LSSVM and CPO optimization algorithm parameters; the two hyperparameters of LSSVM, the regularization parameter γ and the width parameter σ of the Gaussian kernel function 2 ; The CPO optimization algorithm parameters include the initial population number and the maximum number of iterations;
[0067] The CPO optimization algorithm (Crested Porcupine Optimizer) is a new type of meta-heuristic optimization algorithm. As an intelligent optimization algorithm, the CPO algorithm has a wide range of application potential. It can be used to solve various complex optimization problems, such as benchmark function optimization, engineering optimization problems, etc. Since the CPO algorithm simulates the survival strategy of crested porcupines in complex environments, it is particularly suitable for optimization problems that require simulation of biological behaviors or natural phenomena. In addition, the CPO algorithm can also be combined with other optimization algorithms to form a hybrid optimization algorithm to further improve the optimization performance.
[0068] In this example, the search range of the regularization parameter γ is [0.001, 1000];
[0069] The width parameter σ of the Gaussian kernel function 2 The search range is [0.001, 100]; the initial population size is 20, and the maximum number of iterations is 100;
[0070] S23, using CPO optimization algorithm to search for the optimal hyperparameters within the hyperparameter optimization range of LSSVM using 5-fold cross validation;
[0071] A 5-fold cross-validation strategy is used to avoid overfitting of the model due to the selected hyperparameters:
[0072] The data set is divided into 5 subsets, which are used as test sets in turn, and the rest are used as training sets. The average prediction error (such as MSE, MAE) is used as the optimization target, and the CPO optimization algorithm is used to search for the optimal γ and σ within the set hyperparameter optimization range. 2 , record the optimal solution of each iteration until the maximum number of iterations is reached or the convergence condition is met.
[0073] By optimizing the hyperparameters, the hyperparameter optimization results of LSSVM after CPO optimization are shown in Table 1:
[0074] Table 1
[0075] Hyperparameter Name Optimization range Optimal results Regularization parameter γ [0.001,1000] 87.429501 <![CDATA[Width parameter σ of the Gaussian kernel function 2 > [0.001,100] 10.066997
[0076] S3, using the Adaboost ensemble learner to integrate multiple weak learners into the predictive regression model to obtain the Adaboost-CPO-LSSVM model;
[0077] The AdaBoost (Adaptive Boosting) ensemble learner is an iterative algorithm. Its core idea is to train different weak classifiers (weak learners) for the same training set, and then combine these weak classifiers to form a stronger final classifier (strong classifier).
[0078] refer to Figure 2 , step S3 specifically includes the following steps:
[0079] S31. Construct AdaBoost weak learner set: Based on the CPO-LSSVM prediction regression model, generate N weak learners with different data partitions to ensure that each learner has certain differences and independence;
[0080] The diversity of the learner can also be increased by adjusting the kernel parameters and regularization parameters of LSSVM.
[0081] The number N of weak learners ranges from [1, 100], and the number N of weak learners is determined according to the minimum error in the training model corresponding to the number of weak learners.
[0082] S32, initialize the weight of each weak learner to 1 / N;
[0083] S33, for each weak learner, use the weighted data set to perform iterative training, and calculate its prediction error on the training set;
[0084] S34, dynamically adjust the weight of each weak learner according to the prediction error; the learner with a smaller prediction error obtains a higher weight so as to play a greater role in the final prediction;
[0085] S35. Repeat steps S33 to S34 until the number of iterations of the weak learner reaches a preset maximum number of iterations.
[0086] The value of the preset maximum number of iterations is selected according to actual conditions. In this embodiment, the preset maximum number of iterations is 1000;
[0087] S36. Use weighted voting or weighted averaging to combine the prediction results of multiple weak learners to obtain a final prediction value.
[0088] After integrating the prediction results of multiple weak learners, the prediction accuracy of CPO-LSSVM is further improved. The optimized model is called Adaboost-CPO-LSSVM model.
[0089] S4, using 80% of the collected data to train the Adaboost-CPO-LSSVM model, and using the remaining 20% of the collected data to test the Adaboost-CPO-LSSVM model to obtain model evaluation indicators; the model evaluation indicators include determination coefficient, root mean square and mean absolute error;
[0090] refer to Figure 3 80% of the collected data were randomly selected from the segment floating prediction model database as the training set, and the segment engineering geological parameters, excavation parameters and tunnel geometric parameters in the training set were used as input, and the segment floating height was used as output to train the Adaboost-CPO-LSSVM model. After the model training was completed, the remaining 20% of the collected data were used as the test set to test the Adaboost-CPO-LSSVM model. Finally, the model evaluation indicators were calculated, including the determination coefficient (R 2 ), root mean square error (RMSE), and mean absolute error (MAE) were used to evaluate the prediction performance of the model.
[0091] The model evaluation index is specifically the determination coefficient (R 2 ), root mean square error (RMSE), and mean absolute error (MAE) are as follows:
[0092]
[0093] Where: n is the number of data samples, y iis the actual value of the pipe segment floating up, is the predicted value of segment floating, It is the average value of the actual floating value of the pipe segment.
[0094] S5, input the newly collected parameters into the Adaboost-CPO-LSSVM model to obtain the predicted value of the floating amount of the pipe segment;
[0095] We will continue to acquire new data, expand the database, and obtain new segment floatation predictions, which will then be compared with the actual values.
[0096] Step S5 specifically includes the following steps:
[0097] S51, obtaining new collection parameters, including segment floating volume, engineering geological parameters, excavation parameters and tunnel geometric parameters;
[0098] S52, inputting the new collected parameters into the trained Adaboost-CPO-LSSVM model to obtain the predicted value of the floating amount of the pipe segment;
[0099] Step S5 also includes:
[0100] S53, comparing the predicted value of the buoyancy of the pipe segment with the buoyancy of the pipe segment actually measured;
[0101] S54. Analyze the difference between the predicted value and the true value to evaluate the prediction accuracy and reliability of the model; the true value is the actual measured floating amount of the pipe segment.
[0102] After predicting the new collected data in this case, the evaluation index results are shown in Table 2.
[0103] Table 2
[0104] Evaluation indicators Indicator results <![CDATA[Coefficient of determination (R 2 )]]> 0.93 Root mean square error (RMSE) / mm 5.45 Mean absolute error (MAE) / mm 3.08
[0105] From Table 2, we can see that R 2 =0.93 shows that the model has successfully captured the main law of segment floating, and the prediction effect of segment floating in the test set is good with high prediction accuracy; for RMSE, its value is 5.45mm, which means that the root mean square error of the model is relatively small, indicating that the prediction result is relatively accurate. Considering the various complex factors in shield tunnel construction, this error range is reasonable and acceptable. Therefore, the CPO-LSSVM-Adaboost model can achieve a lower prediction error to a certain extent; for MAE, the results show that the average prediction error of the model is 3.08mm, which means that the model is accurate as a whole and the prediction error is relatively uniform among different samples.
[0106] This implementation case constructed a segment floating prediction database based on on-site segment floating monitoring data, excavation parameters, geological parameters, etc., and successfully applied it to the prediction method. From the final results, the segment floating prediction method provided by the present invention has good accuracy and stability in this engineering implementation case. The good effect on the test set means that the prediction effect of the model on new data also maintains a high level, which further shows that the model can accurately predict the floating of the segment to a certain extent, and the method is applicable to different shield tunnel projects, and is not restricted by the type, mode and size of the shield, as well as the stratum in which the shield tunnel is located, which is of great significance for shield tunnel construction.
[0107] Embodiment 2
[0108] refer to Figure 4 This embodiment provides a shield tunnel segment floating prediction device, including the following units:
[0109] A database construction unit, used to construct a segment floating prediction model database based on the collected data, wherein the collected data includes segment floating amount, engineering geological parameters, excavation parameters and tunnel geometric parameters;
[0110] The prediction regression model unit is used to establish a prediction regression model based on CPO-LSSVM, determine the hyperparameter optimization range in LSSVM and the parameters of the CPO optimization algorithm, and use 5-fold cross validation to perform hyperparameter optimization;
[0111] The prediction regression model unit is optimized to integrate multiple weak learners of the prediction regression model using the Adaboost ensemble learner to obtain the Adaboost-CPO-LSSVM model;
[0112] A training unit, used to train the Adaboost-CPO-LSSVM model using 80% of the collected data, and to test the Adaboost-CPO-LSSVM model using the remaining 20% of the collected data to obtain model evaluation indicators; the model evaluation indicators include determination coefficient, root mean square and mean absolute error;
[0113] The prediction unit is used to input the newly collected parameters into the Adaboost-CPO-LSSVM model to obtain the predicted value of the floating amount of the pipe segment.
[0114] Embodiment 3
[0115] refer to Figure 5 , Figure 5: is a structural schematic diagram of a shield tunnel segment floating prediction device of this embodiment. The shield tunnel segment floating prediction device 20 of this embodiment includes a processor 21, a memory 22, and a computer program stored in the memory 22 and executable on the processor 21. When the processor 21 executes the computer program, the steps in the above method embodiment are implemented. Alternatively, when the processor 21 executes the computer program, the functions of each module / unit in the above device embodiments are implemented.
[0116] Exemplarily, the computer program can be divided into one or more modules / units, and the one or more modules / units are stored in the memory 22 and executed by the processor 21 to complete the present invention. The one or more modules / units can be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the shield tunnel segment floating prediction device 20. For example, the computer program can be divided into the modules in Example 2. For the specific functions of each module, please refer to the working process of the device described in the above embodiment, which will not be repeated here.
[0117] The shield tunnel segment floating prediction device 20 may include, but is not limited to, a processor 21 and a memory 22. Those skilled in the art will appreciate that the schematic diagram is merely an example of the shield tunnel segment floating prediction device 20 and does not constitute a limitation on the shield tunnel segment floating prediction device 20. The shield tunnel segment floating prediction device 20 may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the shield tunnel segment floating prediction device 20 may also include input and output devices, network access devices, buses, and the like.
[0118] The processor 21 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor 21 is the control center of the shield tunnel segment floating prediction device 20, and uses various interfaces and lines to connect various parts of the entire shield tunnel segment floating prediction device 20.
[0119] The memory 22 can be used to store the computer program and / or module. The processor 21 realizes various functions of the shield tunnel segment floating prediction device 20 by running or executing the computer program and / or module stored in the memory 22 and calling the data stored in the memory 22. The memory 22 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.). In addition, the memory 22 can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.
[0120] Wherein, if the module / unit integrated in the shield tunnel segment floating prediction device 20 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor 21. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media does not include electrical carrier signals and telecommunication signals.
[0121] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.
[0122] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 Processes Multiple processes and / or boxes Figure 1 A device that has the functions specified in one or more boxes.
[0123] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0124] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0125] The parts of the present invention that are not described in detail are prior art. It is obvious to those skilled in the art that the present invention is not limited to the details of the above-mentioned exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention; therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive, and it is intended that all changes that fall within the meaning and scope of equivalent elements are included in the present invention.
Claims
1. A shield tunnel segment floating prediction method, characterized in that: The following steps are involved: S1. Constructing a segment floating prediction model database based on the collected data, wherein the collected data includes segment floating amount, engineering geological parameters, excavation parameters and tunnel geometric parameters; S2. Establish a prediction regression model based on CPO-LSSVM, determine the hyperparameter optimization range in LSSVM and the parameters of the CPO optimization algorithm, and use 5-fold cross validation to optimize the hyperparameters; S3, using the Adaboost ensemble learner to integrate multiple weak learners into the predictive regression model to obtain the Adaboost-CPO-LSSVM model; S4, using 80% of the collected data to train the Adaboost-CPO-LSSVM model, and using the remaining 20% of the collected data to test the Adaboost-CPO-LSSVM model to obtain model evaluation indicators; the model evaluation indicators include determination coefficient, root mean square and mean absolute error; S5. Input the newly collected parameters into the Adaboost-CPO-LSSVM model to obtain the predicted value of the segment floating amount.
2. The shield tunnel segment floating prediction method according to claim 1 is characterized in that: Step S2 specifically includes: S21, determine the LSSVM model structure: initialize the LSSVM model, determine the kernel function and initial hyperparameters; the kernel function is a radial basis function RBF; the hyperparameters include a regularization parameter γ and a width parameter σ of the Gaussian kernel function 2 ; S22, setting the hyperparameter optimization range of LSSVM and CPO optimization algorithm parameters; the two hyperparameters of LSSVM, the regularization parameter γ and the width parameter σ of the Gaussian kernel function 2 ; The CPO optimization algorithm parameters include the initial population number and the maximum number of iterations; S23. The CPO optimization algorithm is used to search for the optimal hyperparameters within the hyperparameter optimization range of LSSVM using 5-fold cross validation.
3. The shield tunnel segment floating prediction method according to claim 2 is characterized in that: The search range of the regularization parameter γ is [0.001, 1000], and the width parameter σ of the Gaussian kernel function is 2 The search range is [0.001, 100]; the initial population size is 20, and the maximum number of iterations is 100.
4. The shield tunnel segment floating prediction method according to claim 1 is characterized in that: Step S3 specifically includes: S31, based on the prediction regression model of CPO-LSSVM, generate N weak learners with different data partitions; S32, initialize the weight of each weak learner to 1 / N; S33, for each weak learner, use the weighted data set to perform iterative training, and calculate its prediction error on the training set; S34, dynamically adjust the weight of each weak learner according to the prediction error; the learner with a smaller prediction error obtains a higher weight so as to play a greater role in the final prediction; S35, repeating steps S33 to S34 until the number of iterations of the weak learner reaches a preset maximum number of iterations; S36. Use weighted voting or weighted averaging to combine the prediction results of multiple weak learners to obtain a final prediction value.
5. The shield tunnel segment floating prediction method according to claim 1 is characterized in that: Step S4 is specifically as follows: 80% of the collected data were randomly selected from the segment flotation prediction model database as a training set, and the segment engineering geological parameters, excavation parameters and tunnel geometric parameters in the training set were used as input, and the segment flotation height was used as output to train the Adaboost-CPO-LSSVM model. After the model training was completed, the remaining 20% of the collected data were used as a test set to test the Adaboost-CPO-LSSVM model. Finally, the model evaluation indicators, including the determination coefficient, root mean square error and mean absolute error, were calculated to evaluate the prediction performance of the model.
6. The shield tunnel segment floating prediction method according to claim 1 is characterized in that: Step S1 is specifically as follows: The collected data are cleaned, outliers and missing values are removed, and normalized to ensure that the data are on the same scale to ensure numerical stability and convergence during model training. The processed collected data are then stored in the database.
7. The shield tunnel segment floating prediction method according to claim 6 is characterized in that: The normalization process is to use Z-score normalization to process the collected data.
8. The shield tunnel segment floating prediction method according to claim 1 is characterized in that: The engineering geological parameters include cohesion, internal friction angle, permeability coefficient, compression modulus and soil weight; the excavation parameters include propulsion speed, cutter head torque, total propulsion force, propulsion pressure, average soil bin pressure, average grouting pressure, grouting pressure difference, shield machine pitch angle and grouting volume; the tunnel geometric parameters include tunnel burial depth, shield diameter and segment width.
9. A shield tunnel segment floating prediction device, characterized in that: The following units are included: A database construction unit, used to construct a segment floating prediction model database based on the collected data, wherein the collected data includes segment floating amount, engineering geological parameters, excavation parameters and tunnel geometric parameters; The prediction regression model unit is used to establish a prediction regression model based on CPO-LSSVM, determine the hyperparameter optimization range in LSSVM and the parameters of the CPO optimization algorithm, and use 5-fold cross validation to perform hyperparameter optimization; The prediction regression model unit is optimized to integrate multiple weak learners of the prediction regression model using the Adaboost ensemble learner to obtain the Adaboost-CPO-LSSVM model; A training unit, used to train the Adaboost-CPO-LSSVM model using 80% of the collected data, and to test the Adaboost-CPO-LSSVM model using the remaining 20% of the collected data to obtain model evaluation indicators; the model evaluation indicators include determination coefficient, root mean square and mean absolute error; The prediction unit is used to input the newly collected parameters into the Adaboost-CPO-LSSVM model to obtain the predicted value of the floating amount of the pipe segment.
10. An electronic device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, characterized in that: When the processor executes the computer program, the shield tunnel segment floating prediction method described in any one of claims 1 to 8 is implemented.