Multi-objective optimization method and system for shield construction parameter

Abstract

A multi-objective optimization method and system for a shield construction parameter. The method comprises: acquiring an initial shield construction parameter; and optimizing the shield construction parameter by using a multi-objective optimization algorithm, during the optimization of the shield construction parameter, on the basis of the shield construction parameter, obtaining a tunneling speed by means of performing prediction on the basis of random forest, calculating a shield tunneling energy consumption ratio on the basis of the tunneling speed and the shield construction parameter, and taking the maximization of the tunneling speed and the minimization of the shield tunneling energy consumption ratio as two fitness functions of the multi-objective optimization algorithm. The multi-objective optimization method and system for a shield construction parameter take construction efficiency into consideration and reduce the construction cost while ensuring construction safety.

Description

A multi-objective optimization method and system for shield construction parameters

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This invention claims priority to Chinese patent application No. 202310826020.3 filed with the State Intellectual Property Office of China on July 6, 2023, entitled “A Multi-Objective Optimization Method and System for Shield Construction Parameters”, the entire contents of which are incorporated by reference into this invention and constitute a part of this invention for all purposes. Technical Field

[0003] The present invention belongs to the technical field of multi-objective optimization of shield parameters, and in particular relates to a multi-objective optimization method and system for shield construction parameters. Background Art

[0004] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0005] The shield method is the preferred construction method for urban underground tunnels. It is environmentally friendly, safe, and highly efficient, and has been widely used in urban tunnel construction in China. The shield method involves advancing a shield machine through the stratum along the designed axis, using the shield shell and segments to support the surrounding rock to prevent tunnel collapse. Compared to conventional tunnels of approximately 6 meters, the slurry pressure difference between the top and bottom of the cutterhead in large-diameter shield tunnels increases significantly, reducing its compatibility with the stratum water and soil pressure, thereby reducing the effective support for the excavation face. Furthermore, as the shield diameter increases, the probability of the shield traversing complex strata also increases significantly, placing higher demands on the shield's adaptability configuration and corresponding construction technology.

[0006] Safe and efficient shield tunneling is a pressing need for current underground space development. To maintain face stability during shield tunneling, sufficient support pressure must be provided within the slurry chamber. However, different combinations of tunneling parameters will produce different tunneling results. One issue worth noting is the shield machine's tunneling energy consumption, which determines the cost of shield construction. Furthermore, the shield machine's tunneling speed is a crucial parameter for safe and efficient shield tunneling, clearly guiding project schedule planning and cost control. Excessively fast tunneling speeds make it more difficult to control the shield's posture and result in greater surface undulations. However, excessively slow tunneling speeds can negatively impact construction efficiency.

[0007] Current research has found that the optimization objectives of shield construction parameters are relatively simple, and in actual engineering projects, it is impossible to consider construction efficiency and reduce construction costs while ensuring construction safety.

[0008] Summary of the Invention

[0009] In order to solve the technical problems existing in the above-mentioned background technology, the present invention provides a multi-objective optimization method and system for shield construction parameters, which combines a random forest algorithm with a multi-objective optimization algorithm, uses the tunneling speed predicted by the random forest as the first fitness function of the multi-objective optimization algorithm, and uses the shield tunneling energy consumption ratio as the second fitness function, to achieve multi-objective optimization of shield construction parameters, while ensuring construction safety, taking into account construction efficiency and reducing construction costs.

[0010] In order to achieve the above object, the present invention adopts the following technical solutions:

[0011] A first aspect of the present invention provides a multi-objective optimization method for shield construction parameters.

[0012] A multi-objective optimization method for shield construction parameters, comprising:

[0013] Obtain initial shield construction parameters;

[0014] A multi-objective optimization algorithm is used to optimize the shield construction parameters. In the process of optimizing the shield construction parameters, the tunneling speed is obtained based on the shield construction parameters through random forest prediction. The shield tunneling energy consumption ratio is calculated based on the tunneling speed and shield construction parameters. Maximizing the tunneling speed and minimizing the shield tunneling energy consumption ratio are used as the two fitness functions of the multi-objective optimization algorithm.

[0015] Furthermore, the Spearman correlation coefficient was used to identify several shield construction parameters with the highest correlation with the target tunneling speed. Based on the identified shield construction parameters, the tunneling speed was predicted using random forest prediction.

[0016] Furthermore, the identified shield construction parameters include: total thrust, cutterhead speed, cutterhead torque, cutterhead extrusion pressure, incision pressure, penetration, tunnel burial depth, cohesion and internal friction angle.

[0017] Furthermore, the shield tunneling energy consumption ratio is related to the tunneling speed and the total thrust, cutterhead speed, cutterhead torque and shield machine cutterhead radius among the shield construction parameters.

[0018] Furthermore, the random forest training steps include:

[0019] Obtain shield construction parameters and tunneling speed, and divide the tunnel into the empty push section, rising section, stable section, and ring construction section. Then select the shield construction parameters and tunneling speed of the stable section and ring construction section.

[0020] For the selected data, the isolation forest is used to detect outliers, and the detected outliers are replaced with missing values. The missing value data are then filled using the average method and normalized.

[0021] Use the normalized data to train the random forest.

[0022] Furthermore, the multi-objective optimization algorithm adopts a non-dominated genetic algorithm, and the non-dominated genetic algorithm adopts an arithmetic crossover operator during population crossover.

[0023] Furthermore, the non-dominated genetic algorithm introduces a cumulative sorting fitness strategy during Pareto sorting.

[0024] A second aspect of the present invention provides a multi-objective optimization system for shield construction parameters.

[0025] A multi-objective optimization system for shield construction parameters, comprising:

[0026] A data acquisition module is configured to: acquire initial shield construction parameters;

[0027] The optimization module is configured to: use a multi-objective optimization algorithm to optimize the shield construction parameters, and in the process of optimizing the shield construction parameters, obtain the excavation speed based on the shield construction parameters through random forest prediction, calculate the shield excavation energy consumption ratio based on the excavation speed and the shield construction parameters, and maximize the excavation speed and minimize the shield excavation energy consumption ratio as two fitness functions of the multi-objective optimization algorithm.

[0028] A third aspect of the present invention provides a computer-readable storage medium.

[0029] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the multi-objective optimization method for shield construction parameters as described in the first aspect above.

[0030] A fourth aspect of the present invention provides a computer device.

[0031] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps in the multi-objective optimization method for shield construction parameters as described in the first aspect above are implemented.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] The present invention takes into account a multi-objective optimization algorithm for shield tunneling parameters. Because in the shield construction process, not only construction efficiency but also construction cost must be considered, and there are many shield construction parameters, the RF algorithm is combined with NSGA-II, and the tunneling speed based on RF prediction is used as the first fitness function of NSGA-II, and the shield tunneling energy consumption ratio is used as the second fitness function. Multi-objective optimization of shield construction parameters is achieved, and the optimal parameter control range is given while meeting multiple objectives. While ensuring construction safety, construction efficiency is taken into account and construction cost is reduced. Compared with single-objective optimization, it is more in line with engineering practical significance.

[0034] This paper improves the NSGA-II algorithm by replacing the SBX crossover operator with an arithmetic crossover operator, enhancing its search performance. Furthermore, a cumulative sorting fitness strategy is introduced to improve the algorithm's global search capabilities and the uniformity of the Pareto front, ultimately yielding a multi-objective optimal solution set. Based on this optimal solution, the optimal range of tunneling parameters is determined, assisting shield tunneling operations and ensuring safe and efficient shield tunneling.

[0035] The present invention uses the Spearman correlation coefficient to select the parameters required for the model in the data, and improves the data quality and usability through data preprocessing. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0037] FIG1 is a flow chart of a multi-objective optimization method for shield construction parameters shown in Example 1 of the present invention;

[0038] FIG2 is a diagram showing the prediction effect of the RF model on the tunneling speed according to the first embodiment of the present invention;

[0039] FIG3 is a flow chart of an improved NSGA-II algorithm model according to the first embodiment of the present invention. DETAILED DESCRIPTION

[0040] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0041] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0042] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0043] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the methods and systems according to the various embodiments of the present disclosure. It should be noted that each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code can include one or more executable instructions for implementing the logical functions specified in the various embodiments. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, or they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the flowchart and / or block diagram, and the combination of the boxes in the flowchart and / or block diagram, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or can be implemented using a combination of dedicated hardware and computer instructions.

[0044] Example 1

[0045] This embodiment provides a multi-objective optimization method for shield construction parameters.

[0046] This embodiment provides a multi-objective optimization method for shield construction parameters, including: obtaining initial shield construction parameters; optimizing the shield construction parameters using a multi-objective optimization algorithm, and in the process of optimizing the shield construction parameters, obtaining a tunneling speed based on the shield construction parameters through random forest prediction, calculating a shield tunneling energy consumption ratio based on the tunneling speed and the shield construction parameters, and using maximizing the tunneling speed and minimizing the shield tunneling energy consumption ratio as two fitness functions of the multi-objective optimization algorithm.

[0047] Among them, the initial shield construction parameters can be randomly generated or set manually.

[0048] This embodiment provides a multi-objective optimization method for shield construction parameters, which selects parameters required by the model from the data and improves data quality and availability through data preprocessing; uses random forest (RF) for prediction, and improves prediction accuracy by adjusting its hyperparameters, which can be used as a fitness function for multi-objective optimization; uses RF-based predicted tunneling speed as the first fitness function of NSGA-II (Non-dominated Sorting Genetic Algorithms, NSGA); uses the shield tunneling energy consumption ratio as the second fitness function to obtain a multi-objective solution set; introduces an arithmetic crossover operator strategy and a cumulative sorting fitness strategy into the NSGA-II algorithm to achieve faster convergence and maintain population diversity; finally, based on the Pareto front, an optimal solution set under multiple objectives is obtained, and a reasonable range of parameter control is given, thereby assisting shield construction.

[0049] This embodiment provides a multi-objective optimization method for shield construction parameters, which specifically includes the following steps:

[0050] Step 1: Obtain shield construction parameters and excavation speed, and perform data segmentation: The sensor obtains four main parts of data for each ring, consisting of the empty push section, the rising section, the stable section, and the ring construction section. Most of the values ​​in the empty push section and the ring construction section are 0, and their analysis is meaningless. The data of the rising section and the stable section of 10 rings are selected for analysis, that is, the shield construction parameters and excavation speed of the stable section and the ring construction section are selected.

[0051] The shield tunneling process involves numerous parameters, including tunneling parameters, tunnel geometry, and geological parameters. Sensors capture data constantly, and analyzing data recorded during shield operation is meaningless; analyzing data recorded during downtime is useless.

[0052] Step 2: Outlier processing: For the selected data, use the isolation forest method to detect outliers, replace the outliers detected by the isolation forest with missing values, and then use the average method to fill the missing value data.

[0053] Isolation Forest is an unsupervised learning algorithm based on decision trees. The idea of ​​the algorithm is to identify samples distributed outside of high-density groups as outliers: a portion of the samples are extracted and placed at the root node of a single tree. The tree is split with a set cut value until the tree cannot be split or grows to a set height. The first batch of samples to be isolated are identified as outliers. This operation is repeated for each tree in the forest. The results of the entire forest split are then aggregated and the anomaly score of each sample is calculated according to the following formula:

[0054] Among them, h(x) is the depth of the sample in the tree, n is the number of samples, H is the harmonic constant, and samples with high scores are judged as outliers.

[0055] Due to sudden shutdowns or abnormal sensor monitoring data, some outliers will be obtained, which should be eliminated to avoid affecting subsequent model predictions.

[0056] Step 3: Data normalization calculation formula:

[0057] Normalize the data to eliminate the dimension between the data, so that it can be better put into the model for prediction. 70% of the processed data is used as the training set and 30% as the test set.

[0058] Step 4: Use the Spearman correlation coefficient to identify the correlation between the shield parameters in the data set and the target tunneling speed. The Spearman correlation coefficient is calculated as follows:

[0059] Where R i ,S i is the rank of observation i, are the actual values ​​of variables x and y, and n is the total number of parameters.

[0060] Step 5: Through correlation analysis, variables related to the tunneling speed target are obtained and used as the main parameters for rationally controlling shield construction. Total thrust, cutterhead speed, cutterhead torque, cutterhead extrusion pressure, incision pressure, penetration, tunnel burial depth, cohesion, internal friction angle, etc. are used as input parameters, and the tunneling speed output parameters are used to predict.

[0061] Step 6: Input the processed data from steps 1-3 into the RF regression model. When training the RF algorithm, it is necessary to adjust important model hyperparameters. By adjusting the number of decision trees, the maximum number of features, and the maximum depth, the regression fitting performance of the model can be improved.

[0062] Adjust the settings of RF hyperparameters, which directly affect the prediction accuracy of the model, including the number of decision trees n_estimators, the maximum number of features max_features, and the maximum depth max_depth; its prediction effect on the test set is shown in Figure 2. The determination coefficient R of the training set is obtained through its model quantitative indicators. 2 is 0.9057, the test set R 2 It is 0.8991, which proves that the model has high generalization ability.

[0063] In order to verify the prediction performance of the RF model, a quantitative indicator determination coefficient R is selected. 2 The performance index is calculated as follows:

[0064] Where yi is the actual value of the target value, is the predicted value of the target value, is the actual mean of the target values.

[0065] Step 7: Shield tunneling specific energy is the energy consumed by the shield machine per unit volume of tunneling, part of which is the energy consumed by the tool to excavate the soil W T , and the other part is the consumption of friction between the shield and the soil, and its formula is as follows:

[0066] Where F is the total thrust (kN), v is the tunneling speed (mm / min), n is the cutterhead speed (r / min), T is the cutterhead torque (MN·m), and R is the cutterhead radius of the shield machine.

[0067] From the perspective of RF-NSGA-II optimization, the soil and geometric parameters for a particular project are fixed. To achieve the goals of minimizing specific tunneling energy and maximizing tunneling speed during shield tunneling, the construction parameters should be optimized.

[0068] Step 8: After RF training, the relationship between shield tunneling parameters and tunneling speed is obtained, and the first objective function is established to determine the objective function maxF2 of tunneling speed; then, the second objective function is established to determine the objective function minF1 of shield tunneling specific energy.

[0069] Specifically, RF is used to accurately predict the tunneling speed, and the nonlinear relationship between the shield parameters and this target is used to construct the first fitness function of NSGA-II.

[0070] Specifically, the empirical formula for shield tunneling energy consumption ratio is used as the second fitness function. To ensure the reasonable implementation of the generated plan, it is necessary to set restrictions on each variable when generating the plan, and form variable constraints. This is to prevent large discrepancies between the optimized shield machine parameters and the actual shield machine parameters, which would lead to a large adjustment range and potential safety hazards.

[0071] Step 9: When optimizing construction parameters, the decision range of the population must be set and its variables constrained to ensure that the initial population has practical significance.

[0072] Step 10: Run the NSGA-II algorithm.

[0073] In multi-objective optimization research in other fields, various algorithms such as MODE and MOGA are used. The NSGA-II algorithm is relatively more common. It can generate optimal solutions to help decision makers make decisions. However, the crossover operator search performance it uses is weak, and the algorithm also has certain limitations in population diversity and convergence speed.

[0074] In this example, to ensure more uniform solutions obtained by the NSGA-II algorithm, an arithmetic crossover operator is used during population crossover, replacing the SBX crossover operator in NSGA-II. This preserves the superior traits of the parent individuals, improves the quality of the individuals after crossover, and enhances the algorithm's search performance. A cumulative ranking fitness strategy is introduced during Pareto sorting to preserve the fitness values ​​of outstanding individuals, thereby accelerating convergence and maintaining population diversity. Ultimately, the optimal Pareto front solution set under multiple objectives is obtained. The algorithm flow chart is shown in Figure 3. The population size, tournament size, crossover probability, and number of evolutionary generations are set according to the complexity of the objective to obtain the Pareto front.

[0075] The arithmetic crossover operator is a weighted average of the corresponding genes of the two parent individuals to obtain the corresponding genes of the offspring individual. Specifically, for two parent individuals P1 and P2, assuming that their genes are g(1) and g(2), respectively, the corresponding genes of the offspring individual are g(c) = λg(1) + (1-λ)g(2), where λ is a randomly generated weight factor, usually between [0,1]. Its advantage is that it retains the information of the parent individuals while introducing new mutation information, thereby expanding the search space of the genetic algorithm.

[0076] Step 11: Obtain the optimal solution using the NSGA-II algorithm. While meeting both objectives, the shield construction parameters are controlled within the optimization range, making shield tunneling safer and more efficient.

[0077] Example 2

[0078] This embodiment provides a multi-objective optimization system for shield construction parameters.

[0079] A multi-objective optimization system for shield construction parameters, comprising:

[0080] A data acquisition module is configured to: acquire initial shield construction parameters;

[0081] The optimization module is configured to: use a multi-objective optimization algorithm to optimize the shield construction parameters, and in the process of optimizing the shield construction parameters, obtain the excavation speed based on the shield construction parameters through random forest prediction, calculate the shield excavation energy consumption ratio based on the excavation speed and the shield construction parameters, and maximize the excavation speed and minimize the shield excavation energy consumption ratio as two fitness functions of the multi-objective optimization algorithm.

[0082] It should be noted that the examples and application scenarios implemented by the above modules are the same as those in the steps of Example 1, but are not limited to the contents disclosed in the above Example 1. It should be noted that the above modules, as part of the system, can be executed in a computer system such as a set of computer executable instructions.

[0083] Example 3

[0084] This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the steps of the multi-objective optimization method for shield construction parameters as described in the first embodiment above are implemented.

[0085] Example 4

[0086] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps in the multi-objective optimization method for shield construction parameters as described in the first embodiment above are implemented.

[0087] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A multi-objective optimization method for shield construction parameters, characterized in that: include: Obtain initial shield construction parameters; A multi-objective optimization algorithm is adopted to optimize the shield construction parameters. In the process of optimizing the shield construction parameters, the excavation speed is obtained based on the shield construction parameters through random forest prediction. The shield excavation energy consumption ratio is calculated based on the excavation speed and shield construction parameters. The maximization of the excavation speed and the minimization of the shield excavation energy consumption ratio are used as the two fitness functions of the multi-objective optimization algorithm.

2. The multi-objective optimization method for shield construction parameters according to claim 1 is characterized in that: The Spearman correlation coefficient was used to identify several shield construction parameters with the highest correlation with the target tunneling speed. Based on the identified shield construction parameters, the tunneling speed was predicted using random forest prediction.

3. The multi-objective optimization method for shield construction parameters according to claim 2 is characterized in that: The identified shield construction parameters include: total thrust, cutter head rotation speed, cutter head torque, cutter head extrusion pressure, incision pressure, penetration, tunnel burial depth, cohesion and internal friction angle.

4. The multi-objective optimization method for shield construction parameters according to claim 1 is characterized in that: The shield tunneling energy consumption ratio is related to the tunneling speed and the total thrust, cutter head rotation speed, cutter head torque and shield machine cutter head radius in the shield construction parameters.

5. The multi-objective optimization method for shield construction parameters according to claim 1 is characterized in that: The training steps of the random forest include: Obtain shield construction parameters and excavation speed, and divide them into empty push section, ascending section, stable section and ring construction section, and then select shield construction parameters and excavation speed of the stable section and ring construction section; For the selected data, the isolation forest is used to detect outliers, and the detected outliers are replaced with missing values. The missing value data are then filled using the average method and normalized. Use the normalized data to train the random forest.

6. The multi-objective optimization method for shield construction parameters according to claim 1 is characterized in that: The multi-objective optimization algorithm adopts a non-dominated genetic algorithm, and the non-dominated genetic algorithm adopts an arithmetic crossover operator when the population crosses.

7. The multi-objective optimization method for shield construction parameters according to claim 6 is characterized in that: The non-dominated genetic algorithm introduces a cumulative sorting fitness strategy during Pareto sorting.

8. A multi-objective optimization system for shield construction parameters, characterized in that: include: A data acquisition module is configured to: acquire initial shield construction parameters; The optimization module is configured as follows: adopting a multi-objective optimization algorithm to optimize the shield construction parameters, and in the process of optimizing the shield construction parameters, based on the shield construction parameters, obtaining the tunneling speed through random forest prediction, calculating the shield tunneling energy consumption ratio based on the tunneling speed and the shield construction parameters, and taking the maximization of the tunneling speed and the minimization of the shield tunneling energy consumption ratio as the two fitness functions of the multi-objective optimization algorithm.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in the multi-objective optimization method for shield construction parameters as described in any one of claims 1 to 7 are implemented.

10. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the multi-objective optimization method for shield construction parameters as described in any one of claims 1 to 7 are implemented.

Images