Shield tunneling under bridge construction method and device based on BO-RF-NSGA-Ⅲ

Abstract

The application belongs to the technical field of shield construction, and particularly discloses a shield underpass construction pier displacement optimization method and equipment based on BO-RF-NSGA-III. The method comprises the following steps: constructing a shield construction parameter index system and obtaining real-time monitoring data; constructing and training a BO-RF regression model, and predicting the vertical displacement and horizontal displacement of the pier caused by the shield underpass construction according to the real-time monitoring data and the BO-RF regression model; constructing a BO-RF-NSGA-III multi-objective optimization model, taking the established BO-RF regression function as the NSGA-III fitness function, and optimizing the vertical displacement and horizontal displacement of the pier caused by the shield underpass construction; and determining the shield construction parameters that satisfy the optimal vertical displacement and horizontal displacement of the pier caused by the shield underpass construction according to the obtained Pareto optimal solution set and through the TOPSIS method. The application can reduce the vertical displacement and horizontal displacement of the pier caused by the shield underpass construction without relying on manual control.

Description

Technical Field

[0001] This invention belongs to the field of shield tunneling technology. Specifically, it relates to a method and equipment for optimizing the displacement of bridge piers during shield tunneling construction based on BO-RF-NSGA-Ⅲ. More specifically, it relates to a method and equipment for optimizing the displacement of bridge piers during shield tunneling construction based on Bayesian optimization (BO), random forest (RF), and nondominated sorting genetic algorithm-III (NSGA-III). Background Technology

[0002] With the increasing density of underground rail transit lines and underground spaces, tunnel construction near the surface or underground structures, such as subway shield tunneling, inevitably involves tunneling under existing buildings or bridges. The excavation of a shield tunnel causes the surrounding strata to contract and deform towards the tunnel, resulting in deformation of adjacent pile foundations. If the tunnel construction has an excessive impact on the pile foundations of nearby bridges, it may render the supported structures unsafe and unusable. Therefore, when developing a construction plan, it is necessary to estimate the impact of tunnel construction on the pile foundations of nearby bridges in advance and reduce this impact by adjusting the construction parameters of the shield machine, thereby mitigating potential risks during construction. Therefore, this paper develops a method and equipment based on BO-RF-NSGA-Ⅲ for optimizing bridge pier displacement during shield tunneling construction to address the pier displacement optimization problem caused by shield tunneling. Summary of the Invention

[0003] To address the aforementioned deficiencies or improvement needs of existing technologies, this invention provides a method and equipment for optimizing bridge pier displacement during shield tunneling construction based on BO-RF-NSGA-Ⅲ. This method combines the characteristics of shield tunneling itself and the specific features of pier displacement optimization during shield tunneling. Specifically, it establishes a shield tunneling construction parameter index system and acquires real-time monitoring data; constructs and trains a BO-RF regression model; and uses the BO-RF regression model to predict the vertical and horizontal displacements of the bridge piers caused by shield tunneling construction based on the real-time monitoring data. A BO-RF-NSGA-Ⅲ multi-objective optimization model is then constructed, using the established BO-RF regression function as the NSGA-Ⅲ fitness function to optimize the vertical and horizontal displacements of the bridge piers caused by shield tunneling construction. Based on the obtained Pareto optimal solution set, the TOPSIS method is used to determine the optimal shield tunneling parameters that satisfy the optimal vertical and horizontal displacements of the bridge piers caused by shield tunneling construction. This invention can reduce the vertical and horizontal displacements of bridge piers caused by shield tunneling construction without relying on manual operation.

[0004] To achieve the above objectives, according to one aspect of the present invention, a method for optimizing the displacement of bridge piers during shield tunneling construction based on BO-RF-NSGA-Ⅲ includes the following steps:

[0005] S1 constructs a parameter index system for shield tunneling construction and obtains real-time monitoring data;

[0006] S2 constructs and trains a BO-RF regression model. Based on real-time monitoring data, the BO-RF regression model is used to predict the vertical and horizontal displacements of the bridge piers caused by the shield tunneling construction.

[0007] S3 constructs a BO-RF-NSGA-Ⅲ multi-objective optimization model, using the established BO-RF regression function as the NSGA-Ⅲ fitness function to optimize the vertical and horizontal displacements of the bridge piers caused by the shield tunneling construction. Based on the obtained Pareto optimal solution set, the TOPSIS method is used to determine the shield construction parameters that satisfy the optimal vertical and horizontal displacements of the bridge piers caused by the shield tunneling construction.

[0008] As a further preferred option, in step S1, a shield tunneling construction parameter index system is constructed, and real-time monitoring data is obtained, which includes:

[0009] Obtain key working parameters for shield tunneling construction: soil pressure X1, total thrust X2, grouting pressure X3, advance speed X4, and grouting volume X5;

[0010] Characteristic parameters of the tunnel itself: distance between the bridge pier pile foundation and the outer edge of the tunnel structure X6, tunnel burial depth X7;

[0011] Geological parameters: Soft silty clay layer thickness X8, hard silty clay layer thickness X9;

[0012] Based on the above parameters, the numerical distribution of vertical and horizontal displacements of bridge piers caused by shield tunneling under different conditions was monitored.

[0013] As a further preferred embodiment, in step S2, the data sample set of the monitoring data is divided into a training set and a test set, and the parameters of the BO-RF regression model are optimized and the prediction accuracy is analyzed using the training set and the test set. The accuracy of the BO-RF regression model prediction is evaluated using root mean square error, goodness of fit, and mean absolute error. Preferably, the importance analysis of influencing factors based on the RF algorithm includes the following steps:

[0014] (1) Data preprocessing

[0015] Data normalization can scale input data within a certain range, thereby eliminating the impact of different metrics on prediction efficiency and accuracy. Considering that normalizing data to the [-1,1] interval is better than normalizing to the [0,1] interval in avoiding overloading the features of the input vector, this paper chooses to scale data samples for different targets to the [-1,1] interval.

[0016] (2) Determination of hyperparameters of RF prediction model

[0017] To achieve better prediction results for the RF regression model, this paper uses the Bayesian optimization algorithm to optimize the two parameters n_estimators and max_depth of the model, and combines it with the 5-fold cross-validation method to verify the accuracy of the model, thereby determining the parameter combination with the highest accuracy as the optimal parameters for the RF prediction model.

[0018] (3) Establish training model

[0019] The sample data is randomly divided into training sample set and test sample set. The obtained parameter optimization results are input into the model and the model is built using Python.

[0020] (4) Three commonly used metrics, root mean square error (RMSE), goodness of fit (R²), and mean absolute error (MAE), are introduced to evaluate the accuracy of the BO-RF prediction model. RMSE and MAE can better reflect the deviation between predicted and observed values, while R²... 2 These are used to measure the goodness of fit between predicted and observed values. The smaller the RMSE and MAE values, the better the model performance. 2 The closer the value is to 1, the higher the model's prediction accuracy.

[0021]

[0022]

[0023]

[0024] In the formula, n is the number of samples. For the predicted value, y i These are actual observed values. This represents the average of the actual observed values.

[0025] (5) Based on the RF model, using the vertical and horizontal displacements of the bridge piers as prediction indicators, a variable importance analysis was performed on the parameters in the shield tunneling parameter index system. Then, parameters with significant influence on vertical and horizontal displacements were selected as the main optimization parameters, and the five shield tunneling parameters with the highest importance ranking were optimized. More specifically, using the vertical and horizontal displacements of the bridge piers as prediction indicators, a variable importance analysis was performed on nine shield tunneling parameters. Then, parameters with significant influence on vertical and horizontal displacements were selected as the main optimization parameters, and the five shield tunneling parameters with the highest importance ranking were optimized to propose an executable construction plan.

[0026] As a further optimization, in step S2, the importance of each influencing factor is evaluated using the BO-RF algorithm's importance evaluation function to determine the soil pressure X1, total thrust X2, grouting pressure X3, advance speed X4, and grouting volume X5 that need adjustment. The established BO-RF regression function is used as the NSGA-Ⅲ fitness function to optimize the vertical and horizontal displacements of the bridge piers caused by the shield tunneling construction, including the following steps:

[0027] (1) The relationship between the shield tunneling parameters fitted by RF regression and the vertical and horizontal displacements of the bridge piers is introduced as the fitness function in the multi-objective genetic algorithm. The objective functions f1 and f2 of the vertical and horizontal displacements of the bridge piers are determined by the CatBoost prediction regression equation:

[0028]

[0029] Where f1 and f2 are the absolute values ​​of the output functions of each optimization objective, and X1, X2, X3, X4, and X5 are the earth pressure, total thrust, grouting pressure, grouting volume, and tunneling speed, respectively.

[0030] In the above steps, the relationship between the shield construction parameters and the vertical and horizontal displacements of the bridge piers, which is fitted by RF regression prediction, is introduced as the fitness function in the multi-objective genetic algorithm. This can clearly reflect the complex nonlinear relationship between the shield construction parameters and the vertical and horizontal displacements of the bridge piers.

[0031] (2) Determine the range of constraints

[0032] Constraints are set on the values ​​of decision parameters based on engineering requirements and existing project data. To ensure the practical significance of the optimized construction parameter design and to guarantee that the resulting construction parameter settings are more reasonable, feasible, and free from safety hazards, constraints are typically set on the values ​​of decision parameters based on engineering requirements and existing project data. These constraints are generally expressed as follows:

[0033]

[0034] in, and These represent the lower and upper limits of the values ​​that the operation parameter can take, respectively.

[0035] (3) NSGA-III multi-objective optimization: The NSGA-III algorithm is used to make multi-objective optimization decisions on shield tunneling construction parameters.

[0036] As a further optimization, the established BO-RF regression function is used as the NSGA-Ⅲ fitness function to optimize the vertical and horizontal displacements of the bridge piers caused by the shield tunneling construction. Specifically, based on the obtained Pareto optimal solution set, the TOPSIS method is used to determine the shield construction parameters that satisfy the optimal vertical and horizontal displacements of the bridge piers caused by the shield tunneling construction. This includes: the actual working conditions of the shield tunnel passing under the high-speed railway bridge; based on the parameter range obtained from the Pareto optimal solution set, the optimal shield construction parameters for the shield tunnel passing under the high-speed railway bridge are obtained to reduce the pier displacement during the shield tunneling construction. More specifically, this includes:

[0037] 1) First, randomly generate an initial population P of size N, perform fast non-dominated sorting on the initial population, and implement genetic algorithm selection, crossover, and mutation operations to generate a progeny population Q of size N;

[0038] 2) Merge the parent population P and the offspring population Q into a new population R with a size of 2N. Perform a fast non-dominated sort on population R and select individuals from different dominance levels to add to the next generation of offspring.

[0039] 3) Perform adaptive normalization on individuals, generate reference points based on subjective preferences or uniform settings, connect the origin and reference points to form a reference line, calculate the vertical distance between each individual and the reference line, and associate the individual with the smallest vertical distance with the reference point.

[0040] 4) Select the remaining individuals based on the minimum habitat number and combine them with the individuals selected by the previous fast non-dominated sorting to form a new generation of population. Repeat the above process until the convergent generation is reached, and finally obtain a Pareto optimal solution set.

[0041] 5) Obtaining the optimal solution using the TOPSIS method: The TOPSIS method is used to select the optimal solution from the Pareto solution set. This involves calculating the distances between each solution and the ideal solution and the negative ideal solution, and denoting the ideal solution as f. * =(f1) * f2 * The Pareto optimal solution set is denoted as The distance D between the Pareto solution set and the ideal solution is... i The calculation formula is as follows:

[0042]

[0043] In the formula, f1 * and These are the ideal values ​​corresponding to the objective functions f1 and f2, respectively. n and Let f1 and f2 be the values ​​corresponding to the nth Pareto optimal solution sets of objective functions f1 and f2, respectively; μ1 and μ2 be the average values ​​of the Pareto solution sets of objective functions f1 and f2, respectively; and σ1 and σ2 be the standard deviations of the Pareto solution sets of objective functions f1 and f2, respectively. Based on the minimum distance principle, the distance D from the ideal solution is selected. i The Pareto solution corresponding to the minimum value is taken as the optimal compromise solution.

[0044] As a further preferred embodiment, the constraints of the BO-RF-NSGA-Ⅲ multi-objective optimization model include:

[0045]

[0046] Where st represents the constraints, and X1, X2, X3, X4, and X5 represent the soil pressure, total thrust, grouting pressure, grouting volume, and tunneling speed, respectively.

[0047] According to another aspect of the present invention, a device for optimizing bridge pier displacement during shield tunneling construction based on BO-RF-NSGA-Ⅲ is also provided, comprising: a first main module for constructing a shield tunneling construction parameter index system and acquiring real-time monitoring data; a second main module for constructing and training a BO-RF regression model, and using the BO-RF regression model to predict the bridge pier displacement caused by shield tunneling construction based on real-time monitoring data; a third main module for constructing a BO-RF-NSGA-Ⅲ multi-objective optimization model, using the established BO-RF regression function as the NSGA-Ⅲ fitness function to optimize the vertical and horizontal displacements of the bridge piers caused by shield tunneling construction; and determining the optimal shield tunneling construction parameters for the vertical and horizontal displacements of the bridge piers caused by shield tunneling construction using the TOPSIS method based on the obtained Pareto optimal solution set.

[0048] According to another aspect of the invention, an electronic device includes:

[0049] At least one processor, at least one memory, and a communication interface; wherein,

[0050] The processor, memory, and communication interface communicate with each other;

[0051] The memory stores program instructions that can be executed by the processor, which calls the program instructions to execute the method described above.

[0052] According to another aspect of the present invention, a non-transitory computer-readable storage medium is also provided, the non-transitory computer-readable storage medium storing computer instructions that cause the computer to perform the above-described method.

[0053] In summary, compared with the prior art, the above-described technical solutions conceived by this invention mainly possess the following technical advantages:

[0054] 1. The method and equipment for optimizing the displacement of bridge piers during shield tunneling construction based on BO-RF-NSGA-Ⅲ provided in this invention establishes a high-precision prediction model for the vertical and horizontal displacements of the bridge piers using the RF algorithm. Based on the RF algorithm, the importance of variables for nine shield tunneling parameters is analyzed to ensure that the prediction results for the vertical and horizontal displacements of the bridge piers are more accurate and reliable.

[0055] 2. The method and equipment for optimizing bridge pier displacement during shield tunneling construction based on BO-RF-NSGA-Ⅲ provided in this invention replaces the traditional mathematical function with the BO-RF regression prediction function as the fitness function of the genetic algorithm for multi-objective optimization. This accurately maps the complex nonlinear relationship between the vertical and horizontal displacements of the bridge piers and the shield tunneling construction parameters, achieving more precise optimization.

[0056] 3. The method and equipment for optimizing the displacement of bridge piers during shield tunneling construction based on BO-RF-NSGA-Ⅲ provided in this invention uses NSGA-Ⅲ to establish a multi-objective optimization model. Compared with the traditional genetic algorithm, the NSGA-Ⅲ algorithm has an elite strategy, which ensures the superiority of individuals and the diversity of the population, and has high optimization accuracy and fast convergence speed.

[0057] 4. This invention utilizes the coefficient of determination (R²) 2 The performance of the prediction model was evaluated using three commonly used metrics: root mean square error (RMSE) and mean absolute error (MAE). The analysis verified the effectiveness and accuracy of the BO-RF model in predicting the vertical and horizontal displacements of bridge piers.

[0058] 5. The method and equipment for optimizing the displacement of bridge piers during shield tunneling construction based on BO-RF-NSGA-Ⅲ provided by this invention adopts the TOPSIS method to select the optimal decision scheme. It determines a set of optimal solutions that make the multi-objective function optimal from the Pareto front solution set obtained by the NSGA-Ⅲ algorithm. It comprehensively considers the state of the two objective functions, the acquisition method is simple, and the results are relatively objective. Attached Figure Description

[0059] Figure 1 A flowchart of a method for optimizing bridge pier displacement during shield tunneling construction based on BO-RF-NSGA-Ⅲ, provided in an embodiment of the present invention;

[0060] Figure 2 A schematic diagram of the structure of the shield tunneling pier displacement optimization device based on BO-RF-NSGA-Ⅲ provided in an embodiment of the present invention;

[0061] Figure 3 A schematic diagram of the physical structure of an electronic device provided in an embodiment of the present invention;

[0062] Figure 4 This is a search result diagram of the vertical displacement hyperparameters of bridge piers provided in an embodiment of the present invention;

[0063] Figure 5 This is a search result diagram of the horizontal displacement hyperparameter of the bridge pier provided in an embodiment of the present invention;

[0064] Figure 6 A schematic diagram of the prediction results of the training set for the vertical displacement of bridge piers provided in an embodiment of the present invention;

[0065] Figure 7 A schematic diagram of the prediction results of the test set for the vertical displacement of bridge piers provided in an embodiment of the present invention;

[0066] Figure 8 This is a schematic diagram of the prediction results of the training set for the horizontal displacement of bridge piers provided in an embodiment of the present invention;

[0067] Figure 9 This is a schematic diagram of the prediction results of the bridge pier horizontal displacement test set provided in an embodiment of the present invention;

[0068] Figure 10 A distribution diagram showing the importance of factors affecting the vertical displacement of bridge piers, provided in an embodiment of the present invention.

[0069] Figure 11 This is a distribution diagram showing the importance of factors affecting the horizontal displacement of bridge piers, provided in an embodiment of the present invention.

[0070] Figure 12 This is a schematic diagram of the ideal point effect of the Pareto front provided in an embodiment of the present invention. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0072] like Figure 1 As shown, this invention provides a method for optimizing the displacement of bridge piers during shield tunneling construction based on BO-RF-NSGA-Ⅲ, comprising the following steps:

[0073] S1. Construct a shield tunneling construction parameter index system and acquire real-time monitoring data. This step involves constructing the shield tunneling construction parameter index system and acquiring real-time monitoring data, including: acquiring key shield tunneling construction parameters: soil chamber pressure X1, total thrust X2, grouting pressure X3, advance speed X4, and grouting volume X5; tunnel-specific parameters: distance between the bridge pier pile foundation and the outer edge of the tunnel structure X6, and tunnel depth X7; geological parameters: thickness of soft silty clay layer X8 and thickness of hard silty clay layer X9; based on the above parameters, monitor the numerical distribution of vertical and horizontal displacements of the bridge piers caused by shield tunneling under different conditions.

[0074] S2 constructs and trains a BO-RF regression model, and uses the BO-RF regression model to predict the vertical and horizontal displacements of the bridge piers caused by shield tunneling construction based on real-time monitoring data.

[0075] In this step, the data sample set of the monitoring data is divided into a training set and a test set. The training set and the test set are used to optimize the parameters and analyze the prediction accuracy of the BO-RF regression model. The root mean square error, goodness of fit, and mean absolute error are used to evaluate the accuracy of the BO-RF regression model's predictions. The importance analysis of influencing factors is performed based on the RF algorithm, including the following steps:

[0076] (1) Data preprocessing

[0077] Data normalization can scale input data within a certain range, thereby eliminating the impact of different metrics on prediction efficiency and accuracy. Considering that normalizing data to the [-1,1] interval is better than normalizing to the [0,1] interval in avoiding overloading the features of the input vector, this paper chooses to scale data samples for different targets to the [-1,1] interval.

[0078] (2) Determination of hyperparameters of RF prediction model

[0079] To achieve better prediction results for the RF regression model, this paper uses the Bayesian optimization algorithm to optimize the two parameters n_estimators and max_depth of the model, and combines it with the 5-fold cross-validation method to verify the accuracy of the model, thereby determining the parameter combination with the highest accuracy as the optimal parameters for the RF prediction model.

[0080] (3) Establish a training model: Randomly divide the sample data into a training sample set and a test sample set, input the obtained parameter optimization results into the model, and establish the model based on Python.

[0081] (4) Three commonly used metrics, root mean square error (RMSE), goodness of fit (R²), and mean absolute error (MAE), are introduced to evaluate the accuracy of the BO-RF prediction model. RMSE and MAE can better reflect the deviation between predicted and observed values, while R²... 2 These are used to measure the goodness of fit between predicted and observed values. The smaller the RMSE and MAE values, the better the model performance. 2 The closer the value is to 1, the higher the model's prediction accuracy.

[0082]

[0083]

[0084]

[0085] In the formula, n is the number of samples. For the predicted value, y i These are actual observed values. This represents the average of the actual observed values.

[0086] (5) Based on the RF model, the vertical and horizontal displacements of the bridge piers were used as prediction indicators to conduct variable importance analysis on nine shield tunneling parameters. Then, the parameters with significant influence on vertical and horizontal displacements were selected as the main optimization parameters, and the five shield tunneling construction parameters with higher importance were optimized in order to propose an executable construction plan.

[0087] S3 constructs a BO-RF-NSGA-Ⅲ multi-objective optimization model, using the established BO-RF regression function as the NSGA-Ⅲ fitness function to optimize the vertical and horizontal displacements of the bridge piers caused by the shield tunneling construction. Based on the obtained Pareto optimal solution set, the TOPSIS method is used to determine the shield construction parameters that satisfy the optimal vertical and horizontal displacements of the bridge piers caused by the shield tunneling construction.

[0088] In this step, based on the importance evaluation function of the BO-RF algorithm for each influencing factor, the soil pressure X1, total thrust X2, grouting pressure X3, advance speed X4, and grouting volume X5 that need to be adjusted are determined. The established BO-RF regression function is used as the NSGA-Ⅲ fitness function to optimize the vertical and horizontal displacements of the bridge piers caused by the shield tunneling construction, including the following steps:

[0089] (1) Determine the objective function

[0090] The relationship between the shield tunneling parameters and the vertical and horizontal displacements of the bridge piers, as predicted and fitted by RF regression, is introduced as the fitness function in the multi-objective genetic algorithm. This clearly reflects the complex nonlinear relationship between the shield tunneling parameters and the vertical and horizontal displacements of the bridge piers. The objective functions f1 and f2 for the vertical and horizontal displacements of the bridge piers, determined by the CatBoost prediction regression equation, are as follows:

[0091]

[0092] Where f1 and f2 are the absolute values ​​of the output functions of each optimization objective, and X1, X2, X3, X4, and X5 are the earth pressure, total thrust, grouting pressure, grouting volume, and tunneling speed, respectively.

[0093] (2) Determine the range of constraints

[0094] To ensure the practical significance of optimized construction parameter design and to guarantee that the resulting parameter settings are more reasonable, feasible, and free from safety hazards, constraints need to be set on the values ​​of decision parameters based on project requirements and existing project data. These constraints are generally expressed as follows:

[0095]

[0096] (3) NSGA-III Multi-Objective Optimization

[0097] 1) First, randomly generate an initial population P of size N, perform fast non-dominated sorting on the initial population, and implement genetic algorithm selection, crossover, and mutation operations to generate a progeny population Q of size N;

[0098] 2) Merge the parent population P and the offspring population Q into a new population R with a size of 2N. Perform a fast non-dominated sort on population R and select individuals from different dominance levels to join the next generation of offspring.

[0099] 3) Perform adaptive normalization on individuals, generate reference points based on subjective preferences or uniform settings, connect the origin and reference points to form a reference line, calculate the vertical distance between each individual and the reference line, and associate the individual with the smallest vertical distance with the reference point.

[0100] 4) Select the remaining individuals based on the minimum habitat number, and combine them with the individuals selected by the previous fast non-dominated sorting to form a new generation population (population size N). Repeat the above process until the convergent generation is reached. Finally, a Pareto optimal solution set is obtained.

[0101] (4) Obtaining the optimal solution using the TOPSIS method

[0102] The TOPSIS method is used to select the optimal solution from the Pareto solution set. The core of this method is calculating the distances from each solution to the ideal solution and the negative ideal solution. The ideal solution is denoted as... The Pareto optimal solution set is denoted as The distance D between the Pareto solution set and the ideal solution is... i The calculation formula is as follows:

[0103]

[0104] In the formula, f1 * and These are the ideal values ​​corresponding to the objective functions f1 and f2, respectively. n and Let f1 and f2 be the values ​​corresponding to the nth Pareto optimal solution sets of objective functions f1 and f2, respectively; μ1 and μ2 be the average values ​​of the Pareto solution sets of objective functions f1 and f2, respectively; and σ1 and σ2 be the standard deviations of the Pareto solution sets of objective functions f1 and f2, respectively. Based on the minimum distance principle, the distance D from the ideal solution is selected. i The Pareto solution corresponding to the minimum value is taken as the optimal compromise solution.

[0105] Based on the above embodiments of the device, as an optional embodiment, the shield tunneling pier displacement optimization device based on BO-RF-NSGA-Ⅲ provided in this embodiment of the invention, such as... Figure 2 As shown, it also includes: a fourth submodule, used to implement the determination of the optimal shield construction parameters for the vertical and horizontal displacements of the bridge piers caused by the shield tunneling construction using the TOPSIS method based on the obtained Pareto optimal solution set. This includes: the actual working conditions of the shield tunnel passing under the high-speed railway bridge, and obtaining the optimal shield construction parameters when the shield tunnel passes under the high-speed railway bridge based on the parameter range obtained from the Pareto optimal solution set, thereby reducing the displacement of the bridge piers during the shield tunneling construction.

[0106] The method in this embodiment of the invention is implemented using an electronic device; therefore, it is necessary to describe the relevant electronic device. For this purpose, as... Figure 3 As shown, embodiments of the present invention provide an electronic device, such as... Figure 3 As shown, the electronic device includes at least one processor, a communications interface, at least one memory, and a communications bus, wherein the at least one processor, the communications interface, and the at least one memory communicate with each other via the communications bus. The at least one processor can invoke logical instructions stored in the at least one memory to execute all or part of the steps of the methods provided in the foregoing method embodiments.

[0107] Furthermore, when the logical instructions in at least one of the aforementioned memories can be implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a 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 described in the various method embodiments of the present invention. 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.

[0108] The device embodiments described above are merely illustrative. 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 modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0109] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0110] Example 1: This example provides a method for optimizing the displacement of bridge piers during shield tunneling construction based on BO-RF-NSGA-Ⅲ, mainly including the following steps:

[0111] (1) Data acquisition and preprocessing

[0112] Nine parameters were used as input variables: soil pressure, total thrust, grouting pressure, tunneling speed, grouting volume, distance between the pile foundation under the overpass pier and the outer edge of the tunnel structure, burial depth, thickness of soft silty clay layer, and thickness of hard silty clay layer. The vertical displacement and horizontal displacement of the pier were used as the two output variables. 156 sets of sample data were collected from the construction site, as shown in Table 1.

[0113] The samples were normalized, and 80% of the samples were randomly selected to form the training set for training the model. To test the generalization performance of the model, the remaining 20% ​​of the samples were used as the test set to verify the model's effectiveness.

[0114] Table 1 Sample Data

[0115]

[0116] (2) Determination of hyperparameters of RF prediction model

[0117] A Bayesian optimization algorithm was used to optimize the parameters `max_depth` and `n_estimators` of the model, and the model accuracy was verified using a 5-fold cross-validation method. The parameter combination with the highest accuracy was then determined as the optimal parameter set for the RF prediction model. Before optimizing the parameters, initial ranges for hyperparameter optimization were given: `max_depth` and `n_estimators` were set to [0, 200] and [0, 150], respectively. The hyperparameter search results for the vertical and horizontal displacements of the bridge piers are shown below. Figure 4 , Figure 5 As shown.

[0118] from Figure 4 As can be seen from the data, the RF prediction model for the vertical displacement of the bridge piers performs best when n_estimators = 90 and max_depth = 190.

[0119] from Figure 5 As can be seen from the data, the RF prediction model for the horizontal displacement of the bridge pier has the best performance when n_estimators = 120 and max_depth = 75.

[0120] (3) Analysis of prediction results

[0121] Based on the parameter optimization results, training and simulation were performed using the training set, and validation was performed using the test set. Prediction models for the vertical and horizontal displacements of the RF bridge piers were established, and the prediction results are as follows: Figure 6 , Figure 7 , Figure 8 , Figure 9 As shown.

[0122] Depend on Figure 6 and Figure 7It can be seen that the BO-RF model can effectively predict the changes in the vertical displacement of bridge piers. The root mean square error (RMSE) of the BO-RF model for predicting the vertical displacement of bridge piers on the training set is 0.225, the mean absolute error (MAE) is 0.117, and the goodness of fit R0 is [value missing]. 2 The R-value is 0.95, the root mean square error (RMSE) of the test set is 0.452, the mean absolute error (MAE) is 0.293, and the goodness of fit (Rfit) is 0. 2 The value of 0.823 indicates that the model has fully learned the relationship between the input and output variables of the sample set, and the fitting effect is good.

[0123] from Figure 8 and Figure 9 The results show that the BO-RF model can effectively predict changes in tunneling specific energy. The root mean square error (RMSE) of the training set for predicting pier horizontal displacement based on the BO-RF model is 0.452, the mean absolute error (MAE) is 0.293, and the goodness of fit R0 is [value missing]. 2 The R² value is 0.823, the root mean square error (RMSE) of the test set is 0.539, the mean absolute error (MAE) is 0.360, and the goodness of fit (R²) is 0.823. 2 The value of 0.826 indicates that the model has fully learned the relationship between the input and output variables of the sample set, and can obtain highly accurate prediction results.

[0124] (4) Importance analysis of influencing factors based on RF

[0125] Based on the RF model, using the vertical and horizontal displacements of the bridge piers as prediction indicators, a variable importance analysis was conducted on nine shield tunneling parameters. The ranking of the importance of the influencing factors is as follows: Figure 10 and Figure 11 As shown.

[0126] The factors influencing vertical displacement of the tunnel boring machine (TBM) parameters, ranked as follows: distance between the pile foundation under the pier and the outer edge of the tunnel structure > soil pressure > tunneling speed > grouting pressure > total thrust > grouting volume > burial depth > thickness of hard silty clay layer > thickness of soft silty clay layer. The factors influencing horizontal displacement of the TBM parameters, ranked as follows: distance between the pile foundation under the pier and the outer edge of the tunnel structure > soil pressure > tunneling speed > grouting pressure > total thrust > grouting volume > burial depth > thickness of hard silty clay layer > thickness of soft silty clay layer.

[0127] The distance between the pile foundation under the bridge pier and the outer edge of the tunnel structure is the most important factor affecting the vertical and horizontal displacement of the bridge pier due to the shield tunneling parameters. The effects of shield tunneling parameters such as grouting volume, soil chamber pressure, grouting pressure, tunneling speed, and total thrust on bridge pier displacement are similar. Construction parameters and geometric parameters have a greater impact on surface settlement than geological parameters.

[0128] (5) Determine the objective function and the range of variable constraints.

[0129] Based on the BO-RF prediction model, a non-dominated sorting genetic algorithm (NSGA-III) with an elitist strategy is applied to perform multi-objective optimization of the vertical and horizontal displacements of the bridge piers. The optimization objective is to simultaneously minimize both the vertical and horizontal displacements of the bridge piers. The objective function of the corresponding optimization model is obtained from the RF prediction regression functions of the vertical and horizontal displacements of the bridge piers, f1 and f2 respectively:

[0130]

[0131] Where |f1| and |f2| are the absolute values ​​of the output functions of each optimization objective, and X1, X2, X3, X4, and X5 are the earth chamber pressure, total thrust, grouting pressure, grouting volume, and tunneling speed, respectively.

[0132] Specifically, the fitting relationship between the vertical and horizontal displacements of the bridge piers predicted by the BO-RF intelligent algorithm and the shield tunneling construction parameters is used. The minimum vertical and horizontal displacements of the bridge piers are taken as the objective function of multi-objective optimization. The minimum displacement of the bridge piers during shield tunneling is shown in equation (15).

[0133] When performing parameter optimization calculations, initial data, i.e., population, is required first. The decision range of the population needs to be set to ensure that the initial population has practical significance. In order to avoid the optimized shield machine parameter results from differing too much from the actual engineering parameters of the shield machine, resulting in a large parameter adjustment range and potential safety hazards, the range of measured data is used as the main reference to set the range of initial parameters. The specific constraint is expressed as equation (16).

[0134]

[0135] Where st represents the constraint condition.

[0136] (6) Multi-objective optimization based on NSGA-III

[0137] Based on the analysis of the importance and influence of construction parameters, five shield tunneling operation parameters with the greatest impact on the vertical and horizontal displacement of the bridge piers were selected for adjustment: earth chamber pressure, total thrust, grouting pressure, grouting volume, and tunneling speed. This resulted in five scenarios. In scenario 1, only earth chamber pressure was adjusted; in scenario 2, earth chamber pressure and total thrust were adjusted; in scenario 3, earth chamber pressure, total thrust, and grouting pressure were adjusted; in scenario 4, earth chamber pressure, total thrust, grouting pressure, and grouting volume were adjusted; and in scenario 5, earth chamber pressure, total thrust, grouting pressure, grouting volume, and tunneling speed were adjusted.

[0138] With the vertical and horizontal displacements of the bridge piers as the two optimization objectives, the population type of the BO-RF-NSGA-Ⅲ algorithm was set to bivector, the number of objectives in the genetic algorithm was set to three, the population size to 100, and the maximum number of generations to six. After determining the above algorithm parameter settings, multi-objective optimization was performed according to the steps of the NSGA-Ⅲ algorithm, obtaining the optimal Pareto front, which contained 58 pairs of optimal solutions. When the shield tunneling parameters were basically consistent with the construction parameters of the optimal solution, the vertical and horizontal displacements of the bridge piers were also basically consistent, which verifies the reference value of the multi-objective optimization results. Therefore, to reduce the vertical and horizontal displacements of the bridge piers, the shield tunneling parameters of the optimal solution can be used as reference values ​​to guide the shield tunneling construction.

[0139] Based on the obtained Pareto optimal solution set, the TOPSIS method was used to determine the optimal shield tunneling parameters that satisfy the vertical and horizontal displacements of the bridge piers caused by the shield tunneling construction. Scenario 5 showed the best pier displacement optimization effect, with soil chamber pressure of 241 bar, total thrust of 17600 kN, grouting pressure of 418 bar, and grouting volume of 4.55 m³. 3 With a tunneling speed of 29.8 m / d, the vertical and horizontal displacements of the piers are 0.35 mm and 0.35 mm, respectively. The percentage improvements in the vertical and horizontal displacements of the piers are 49.92% and 58.48%, respectively. The vertical and horizontal displacements of the piers have been significantly optimized. It can be considered that the values ​​of the above parameters are relatively ideal and can provide a reference for setting the values ​​of various parameters in actual engineering projects.

[0140] Table 2. Pier Displacement Results for 5 Scenarios

[0141]

[0142] Note: The values ​​in parentheses represent the percentage improvement of the optimized value compared to the average of the original data.

[0143] Based on the statistical analysis of shield tunneling parameters obtained through the multi-objective optimization described above, and considering the actual working conditions of a rail transit shield tunnel passing under a high-speed railway bridge, as well as feedback from construction monitoring information, and taking into account economic, technical, and social benefits, the optimal construction parameters for shield tunneling under the bridge are obtained using the parameter range obtained from Pareto front optimization as a reference. (See...) Figure 12 This is to reduce the displacement of bridge piers caused by shield tunneling.

[0144] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Based on this understanding, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, or sometimes in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0145] In this patent, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0146] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention 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; and these 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 the present invention.

[0147] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for optimizing the displacement of bridge piers during shield tunneling construction based on BO-RF-NSGA-Ⅲ, characterized in that, Includes the following steps: S1 constructs a parameter index system for shield tunneling construction and obtains real-time monitoring data; S2 constructs and trains a BO-RF regression model. Based on real-time monitoring data, the BO-RF regression model is used to predict the vertical and horizontal displacements of the bridge piers caused by the shield tunneling construction. S3 constructs a BO-RF-NSGA-Ⅲ multi-objective optimization model, uses the established BO-RF regression function as the NSGA-Ⅲ fitness function, and optimizes the vertical and horizontal displacements of the bridge piers caused by the shield tunneling construction. Based on the obtained Pareto optimal solution set, the TOPSIS method is used to determine the shield construction parameters that satisfy the optimal vertical and horizontal displacements of the bridge piers caused by the shield tunneling construction. In step S2, the importance of each influencing factor is evaluated using the BO-RF algorithm's importance assessment function. This determines the soil pressure X1, total thrust X2, grouting pressure X3, advance speed X4, and grouting volume X5 that need adjustment. The established BO-RF regression function is used as the NSGA-Ⅲ fitness function to optimize the vertical and horizontal displacements of the bridge piers caused by the shield tunneling construction. This includes the following steps: (1) The relationship between the shield construction parameters fitted by RF regression and the vertical and horizontal displacements of the bridge piers is introduced as the fitness function in the multi-objective genetic algorithm. The objective functions of the vertical and horizontal displacements of the bridge piers are determined by the CatBoost prediction regression equation. and They are respectively: in, , The absolute value of the output function for each optimization objective. These are earth chamber pressure, total thrust, grouting pressure, grouting volume, and tunneling speed, respectively. (2) Determine the range of constraints Constraints are set on the values ​​of decision parameters based on engineering requirements and existing project data; (3) NSGA-III multi-objective optimization: The NSGA-III algorithm is used to make multi-objective optimization decisions on shield tunneling construction parameters.

2. The method for optimizing the displacement of bridge piers during shield tunneling construction based on BO-RF-NSGA-Ⅲ as described in claim 1, characterized in that, In step S1, a shield tunneling construction parameter index system is constructed, and real-time monitoring data is obtained, including: Obtain key working parameters for shield tunneling construction: soil pressure X1, total thrust X2, grouting pressure X3, advance speed X4, and grouting volume X5; Characteristic parameters of the tunnel itself: distance between the bridge pier pile foundation and the outer edge of the tunnel structure X6, tunnel burial depth X7; Geological parameters: Soft silty clay layer thickness X8, hard silty clay layer thickness X9; Based on the above parameters, the numerical distribution of vertical and horizontal displacements of bridge piers caused by shield tunneling under different conditions was monitored.

3. The method for optimizing the displacement of bridge piers during shield tunneling construction based on BO-RF-NSGA-Ⅲ according to claim 2, characterized in that, In step S2, the data sample set of the monitoring data is divided into a training set and a test set. The training set and test set are then used to optimize the parameters and analyze the prediction accuracy of the BO-RF regression model. The root mean square error, goodness of fit, and mean absolute error are used to evaluate the accuracy of the BO-RF regression model's predictions. Preferably, the importance analysis of influencing factors is performed based on the RF algorithm, including the following steps: (1) Data preprocessing The monitoring data is normalized to [−1, 1]. (2) Determination of hyperparameters of RF prediction model The Bayesian optimization algorithm was used to optimize the two parameters n_estimators and max_depth of the model, and the model accuracy was verified by combining it with the 5-fold cross-validation method. (3) Establish a training model The sample data is randomly divided into training sample set and test sample set. The obtained parameter optimization results are input into the model and the model is built using Python. (4) Three indicators, root mean square error, goodness of fit and mean absolute error, are introduced to evaluate the accuracy of the BO-RF prediction model; (5) Based on the RF model, the vertical and horizontal displacements of the piers are used as prediction indicators to conduct variable importance analysis on the parameters in the shield tunneling parameter index system. Then, the parameters with significant influence on vertical and horizontal displacements are selected as the main optimization parameters, and the five shield tunneling parameters with higher importance ranking are optimized.

4. The method for optimizing the displacement of bridge piers during shield tunneling construction based on BO-RF-NSGA-Ⅲ according to claim 3, characterized in that, Based on the established BO-RF regression function as the NSGA-Ⅲ fitness function, the vertical and horizontal displacements of the bridge piers caused by shield tunneling construction are optimized, including: 1) First, randomly generate an initial population P of size N, perform fast non-dominated sorting on the initial population, and implement genetic algorithm selection, crossover, and mutation operations to generate a progeny population Q of size N; 2) Merge the parent population P and the offspring population Q into a new population R with a size of 2N. Perform a fast non-dominated sort on population R and select individuals from different dominance levels to add to the next generation of offspring. 3) Perform adaptive normalization on individuals, generate reference points based on subjective preferences or uniform settings, connect the origin and reference points to form a reference line, calculate the vertical distance between each individual and the reference line, and associate the individual with the smallest vertical distance with the reference point; 4) Select the remaining individuals based on the minimum habitat number and combine them with the individuals selected by the previous fast non-dominated sorting to form a new generation of population. Repeat the above process until the convergent generation is reached, and finally obtain a Pareto optimal solution set. (4) Obtaining the optimal solution using the TOPSIS method: The TOPSIS method is used to select the optimal solution from the Pareto solution set, that is, to calculate the distance between each solution and the ideal solution and the negative ideal solution, and to denote the ideal solution as... The Pareto optimal solution set is denoted as Then the distance between the Pareto solution set and the ideal solution is... The calculation formula is as follows: In the formula, and These are the objective functions. and The corresponding ideal value, and These are the objective functions. and No. The values ​​corresponding to each Pareto optimal solution set. and Objective function and The average value of the Pareto solution set. and Objective function and The standard deviation of the Pareto solution set; Based on the principle of minimum distance, select the distance from the ideal solution. The Pareto solution corresponding to the minimum value is taken as the optimal compromise solution.

5. The method for optimizing the displacement of bridge piers during shield tunneling construction based on BO-RF-NSGA-Ⅲ according to claim 1, characterized in that, The constraints of the BO-RF-NSGA-Ⅲ multi-objective optimization model include: Where st represents the constraint condition. These are earth chamber pressure, total thrust, grouting pressure, grouting volume, and tunneling speed, respectively.

6. A pier displacement optimization system for shield tunneling construction based on BO-RF-NSGA-Ⅲ, characterized in that, include: The first main module is used to construct the shield tunneling construction parameter index system and acquire real-time monitoring data. The second main module is used to construct and train the BO-RF regression model, and predict the pier displacement caused by shield tunneling construction based on real-time monitoring data. The third main module is used to construct the BO-RF-NSGA-Ⅲ multi-objective optimization model, and use the established BO-RF regression function as the NSGA-Ⅲ fitness function to optimize the vertical and horizontal displacements of the piers caused by shield tunneling construction. Based on the obtained Pareto optimal solution set, the TOPSIS method is used to determine the shield tunneling construction parameters that satisfy the optimal vertical and horizontal displacements of the piers caused by shield tunneling construction. The importance of each influencing factor was evaluated using the BO-RF algorithm, determining the soil pressure X1, total thrust X2, grouting pressure X3, advance speed X4, and grouting volume X5 that needed adjustment. The established BO-RF regression function was used as the NSGA-Ⅲ fitness function to optimize the vertical and horizontal displacements of the bridge piers caused by the shield tunneling construction, including the following steps: (1) The relationship between the shield construction parameters fitted by RF regression and the vertical and horizontal displacements of the bridge piers is introduced as the fitness function in the multi-objective genetic algorithm. The objective functions of the vertical and horizontal displacements of the bridge piers are determined by the CatBoost prediction regression equation. and They are respectively: in, , The absolute value of the output function for each optimization objective. These are earth chamber pressure, total thrust, grouting pressure, grouting volume, and tunneling speed, respectively. (2) Determine the range of constraints Constraints are set on the values ​​of decision parameters based on engineering requirements and existing project data; (3) NSGA-III multi-objective optimization: The NSGA-III algorithm is used to make multi-objective optimization decisions on shield tunneling construction parameters.

7. An electronic device, characterized in that, include: At least one processor, at least one memory, and a communication interface; wherein, The processor, memory, and communication interface communicate with each other; The memory stores program instructions that can be executed by the processor, which invokes the program instructions to perform the method described in any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores computer instructions that cause the computer to perform the method described in any one of claims 1 to 5.

Images