Optimization method of shield tunneling parameters based on digital strata and system reliability

By coupling generation of adversarial neural networks and physical information neural networks, digital twin strata is constructed, combined with system reliability design theory and global optimization genetic algorithm, the shield excavation parameters are optimized, and the problems of geotechnical parameter uncertainty and stratigraphic deformation control are solved, and the problems of geotechnical parameter uncertainty and strata deformation control are realized. The safety and efficiency of underpass construction of shield structures are achieved.

CN118673818BActive Publication Date: 2024-11-15CHINA RAILWAY WUJU GROUP ELECTRIC WORKS ENG CORP +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411146753.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-21
Publication Date
2024-11-15
Estimated Expiration
2044-08-21

AI Technical Summary

Technical Problem

The existing technology is difficult to accurately express and control the uncertainty of geotechnical parameters in shield excavation, which makes it difficult to ensure the safety of shield underpass construction, especially when the spatial variability of stratigraphic deformation is complex.

Method used

The digital twin strata is constructed using coupled generation and adversarial neural networks and physical information neural networks, combined with system reliability design theory and global optimization genetic algorithms, optimize shield excavation parameters, and realize accurate prediction and safety control of stratigraphic deformation by iteratively updating geotechnical parameters. .

Benefits of technology

The accuracy of shield tunnel boring mechanics analysis and surface settlement prediction accuracy are improved, the safety of shield construction is ensured, and construction energy consumption is reduced, and the robust design of shield underpass engineering is realized.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118673818B_ABST
    Figure CN118673818B_ABST
Patent Text Reader

Abstract

The present invention provides a shield tunneling parameter optimization method based on digital strata and system reliability. The method is based on geotechnical test data, stratum deformation measurement, shield tunneling parameters and other multi-source data. The method uses a physical information neural network coupled with a generative adversarial neural network to construct a digital twin stratum of the construction site, taking into account the uncertainty of geotechnical parameters, and then continuously updating the geotechnical parameters according to construction feedback, so as to achieve a refined characterization of the spatial variability of geotechnical parameters and an accurate prediction of stratum deformation. On this basis, based on the system reliability design theory and the global optimization genetic algorithm, a shield key tunneling mechanical parameter optimization method based on the system reliability index is established to achieve precise control of stratum deformation and fully ensure the safety of shield tunneling construction.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of safe and intelligent shield tunneling, and in particular to a shield tunneling parameter optimization method based on digital strata and system reliability. Background Art

[0002] In shield tunneling projects, the ground displacement caused by tunnel excavation activities extends to the surface. Improper control will cause large-scale settlement of the upper buildings, and in severe cases, cause cracks and collapse of houses. Therefore, the safe control of ground deformation in such projects is particularly important. To achieve this goal, it is crucial to set reasonable values ​​for key shield tunneling mechanical parameters. The key factor affecting shield tunneling mechanical analysis and construction safety control is the spatial variability of geotechnical physical and mechanical parameters. Geotechnical test data at sparse points alone is not sufficient to accurately estimate geotechnical parameters in the global range.

[0003] Most existing technologies are limited to the deterministic analysis level, which makes it difficult to fully quantify and express the inherent uncertainty of geotechnical parameters. The shield tunneling mechanical parameters set accordingly are not accurate, and therefore cannot fully guarantee the safety of shield tunneling construction.

[0004] Therefore, based on geotechnical test data, measured stratum deformation, shield tunneling parameters and other multi-source data, the present invention uses a physical information neural network coupled with a generative adversarial neural network to construct a digital twin stratum of the construction site, taking the uncertainty of geotechnical parameters into consideration, and then continuously updating the geotechnical parameters based on construction feedback such as monitoring deformation data, which not only makes up for the shortcomings of sparse geotechnical parameter data, but also effectively improves the accuracy of shield tunneling mechanics analysis and surface settlement prediction, thereby achieving a refined characterization of the spatial variability of geotechnical parameters and accurate prediction of stratum deformation. On this basis, based on the system reliability design theory and the global optimization genetic algorithm, an optimization method for key shield tunneling mechanics parameters based on system reliability indicators is established, providing a new way for precise settlement control, achieving safe and reliable shield construction control, and fully ensuring the safety of shield underpass construction. Summary of the invention

[0005] The purpose of the present invention is to provide a shield tunneling mechanical parameter optimization method based on digital twin strata, system reliability and allowable stratum deformation control, fully express the uncertainty of stratum and geotechnical parameters in shield tunneling construction control, realize refined decision-making of tunneling mechanical parameters, accurately control surface stratum deformation, ensure the safety of shield underpass tunneling in complex urban environments, and provide an effective solution for promoting the robust design of shield tunnel engineering.

[0006] In order to achieve the above object, the present invention provides a shield tunneling parameter optimization method based on digital strata and system reliability, which can effectively control stratum deformation during shield tunneling, comprising the following steps:

[0007] S1, using all the geotechnical test data of the project, summarize and organize the data of parameters such as compression modulus and shear strength, and establish and train a generative adversarial neural network for characterizing the random fields of these parameters based on the organized geotechnical test data ,in, Represents the spatial coordinate value of any point within the construction range. Generate adversarial neural networks The parameters determined by training in the generative adversarial neural network After training, it can be used to characterize the elastic modulus , Poisson's ratio The spatial distribution of random fields, generating adversarial neural networks Using the geotechnical test data at the sampling location, a preliminary estimate of the geotechnical parameter values ​​at any location within the construction area can be made;

[0008] S2, build coupled generative adversarial neural network Physical Information Neural Network The interaction problem between shield tunneling and strata is solved and analyzed. The strata deformation data obtained by real-time monitoring is used to generate adversarial neural networks. , Physical Information Neural Network Iteratively update the parameters in to obtain the elastic modulus (usually calculated from compression modulus), Poisson's ratio Compared with S1, the conditional random field information features are more accurate. is the shield tunneling parameter vector, which is set in the early stage with reference to similar engineering experience. is the coordinate of the excavation surface position, Physical Information Neural Network Subsequent construction will continuously receive new stratum deformation monitoring values ​​and use them to analyze the physical information neural network. The prediction results of formation deformation are verified and calibrated. The output result of the formation deformation prediction is consistent with the actual monitoring value, so it can be considered that the generated adversarial neural network Characterization of elastic modulus , Poisson's ratio The spatial distribution of the conditional random field conforms to the actual situation of the site.

[0009] S3, through physical information neural network The uncertainty of the spatial distribution of geotechnical parameters in the shield tunneling process is quantified, and the system reliability index to avoid surface settlement and uplift exceeding the allowable value is calculated. Then, the shield tunneling parameters are optimized based on the system reliability index to achieve safe control of shield construction. Furthermore, the iteratively updated physical information neural network in S2 The following conditions are met: the real-time stratum deformation prediction value at the corresponding monitoring location is equal to the true value, and the formula is:

[0010]

[0011] in, Represents the stratum deformation value, is the spatial coordinate of the monitoring point.

[0012] Furthermore, physical information neural network The loss function The error of the control equation describing the interaction between shield and formation , Monitoring data fitting error These two parts are composed of the formula:

[0013] ;

[0014] ;

[0015] ;

[0016] in, and denote the formation stress tensor and strain tensor respectively, represents the formation displacement tensor, and the subscript comma represents the partial derivative. is the Kronecker function, is the soil bulk density, and is the Lame constant, which characterizes the elastic modulus and Poisson's ratio Conditional Random Field The corresponding output is converted to get, , Represent the corresponding number of sampling points.

[0017] Furthermore, in S3, shield tunneling parameters are optimized for permissible stratum deformation control based on system reliability design theory and global optimization genetic algorithm.

[0018] Furthermore, S3 uses the trained and updated generative adversarial network Characterizing Geotechnical Parameters , The random field of shield tunneling parameters is , Shield machine thrust , Cutter speed The combined vector of the cutter head torque T is a variable, that is, using the Monte Carlo simulation algorithm, based on the trained generative adversarial network The random field of geotechnical parameters represented by , Random field development Random simulations, we get Group , Random field simulation results. At the same time, random initialization is performed and each parameter combination is input into the physical information neural network model after training and updating. Perform calculations, , All remain unchanged, and we get Surface deformation data The allowable stratum deformation range is ,in is the lower limit of the allowable value, corresponding to the maximum allowable settlement, is the upper limit of the allowable limit, corresponding to the maximum allowable uplift;

[0019] The ground deformation induced by shield tunneling includes settlement and uplift. Therefore, the ground deformation exceeding the limit can be defined as a system reliability problem, which includes two subsystems: when the ground uplift deformation exceeds the allowable upper limit, it corresponds to the failure of the first subsystem, which is defined as the event When the stratum settlement deformation exceeds the lower limit of the allowable limit, it corresponds to the failure of the second subsystem, which is defined as the event ;remember The number of groups of data that exceeds the upper limit is , the number exceeding the lower limit is , calculate the failure probability of the two subsystems as:

[0020] ;

[0021] ;

[0022] in, is the failure probability of the first subsystem, is the failure probability of the second subsystem; the Ditlevsen limit of the system reliability is calculated based on the Ditlevsen method. The range of this limit is usually very narrow. The average of the upper and lower limits is taken as the final reliability calculation result, so the probability of the system failure event F occurring is obtained as ; By the system failure probability , and the corresponding system reliability is obtained ; Assume that the system target reliability index of the shield tunneling stratum deformation does not exceed the limit is , quantitatively evaluate the risk probability of formation deformation exceeding the limit, so that the calculated system reliability is greater than the target reliability.

[0023] Further, maintain Geotechnical parameters , The random field simulation results remain unchanged, and the shield tunneling parameters are changed through genetic algorithm iterative optimization. The value of , under different values, the corresponding system reliability index of shield tunneling stratum deformation not exceeding the limit is calculated, and the shield tunneling parameter is System reliability at , the shield tunneling parameter value is System reliability at , through iterative optimization ; The optimization of shield tunneling parameters can be transformed into the following system reliability multi-objective optimization problem:

[0024] ;

[0025] in, For given shield tunneling parameters The system reliability index calculated by Monte Carlo simulation algorithm is as follows: is the energy consumption index of the shield machine, which is affected by the shield tunneling parameters The impact of and They are respectively the lower and upper limits of the reference value range of shield tunneling parameters. The calculation formula is as follows:

[0026] ;

[0027] In the formula, is the excavation time, which is determined during the excavation process. Since this problem involves complex multi-objective optimization, in most cases, it is impossible to obtain the global optimal value that maximizes or minimizes all the objectives to be optimized. In the actual optimization process, the shield excavation parameters undergo a certain number of iterations, and finally solve multiple Pareto optimal solutions (at this time, once an attempt is made to change the shield excavation parameters to improve reliability, it will inevitably lead to an increase in energy consumption indicators), and then select a set of shield excavation parameter values ​​from them to maximize the system reliability as much as possible on the basis of controlling energy consumption. Under the calculated reliability, the stratum deformation induced by shield tunneling can be effectively controlled, and at the same time, the energy consumption of shield construction is greatly reduced, with significant environmental protection effects.

[0028] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the shield tunneling parameter optimization method based on digital strata and system reliability as described above is implemented.

[0029] The present invention also provides an electronic device, comprising at least one processor and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the shield tunneling parameter optimization method based on digital strata and system reliability as described above.

[0030] The above scheme of the present invention has the following beneficial effects:

[0031] The shield tunneling parameter optimization method based on digital strata and system reliability provided by the present invention can fully quantify and express the uncertainty information of geotechnical parameters by generating adversarial neural networks and physical information neural networks, and accurately predict the stratum deformation induced by shield tunneling. On this basis, shield tunneling parameter optimization based on system reliability is carried out to further reduce costs and increase efficiency while ensuring the construction safety of shield tunneling projects.

[0032] Other beneficial effects of the present invention will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 is a flow chart of the steps of the present invention;

[0034] Figure 2 It is a schematic diagram of the construction principle of the physical information neural network digital twin stratum of the present invention;

[0035] Figure 3 It is a schematic diagram of the change of the reliability of the present invention with the update of shield tunneling parameters (taking one update optimization as an example);

[0036] Figure 4 The present invention is a schematic diagram of the process steps combined with a shield system. DETAILED DESCRIPTION

[0037] The following describes the embodiments of the present disclosure through specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present disclosure.

[0038] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein may be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present disclosure, it should be understood by those skilled in the art that an aspect described herein may be implemented independently of any other aspect, and two or more of these aspects may be combined in various ways. For example, any number of aspects described herein may be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein may be used to implement this device and / or practice this method.

[0039] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present disclosure, and the illustrations only show the components related to the present disclosure rather than the number, shape and size of the components in actual implementation. The type, quantity and proportion of each component in actual implementation may be changed at will, and the component layout type may also be more complicated. In addition, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that the aspects described can be practiced without these specific details.

[0040] like Figure 1 As shown, an embodiment of the present invention provides a shield tunneling parameter optimization method based on digital strata and system reliability, comprising the following steps:

[0041] S1, using all the geotechnical test parameters of the engineering project, the compression modulus and other geotechnical parameter data are summarized and sorted out, and based on the sorted geotechnical parameter data, a neural network for characterizing the random field of geotechnical parameters is established.

[0042] In this embodiment, a generative adversarial neural network is trained based on geotechnical parameter data. ,in, Represents the spatial coordinate value of any point within the construction range. Generate adversarial neural networks The parameters determined by training in the generative adversarial neural network After training, it can be used to characterize the spatial variability of elastic modulus and Poisson's ratio, that is, , can be used to represent the spatial distribution random field of elastic modulus and Poisson's ratio respectively. By using the geotechnical parameter data at the sampling locations, a preliminary estimate of the geotechnical parameters at any location within the construction area is achieved, which is equivalent to establishing the corresponding initial conditional random field, laying the foundation for subsequent analysis.

[0043] S2, build coupled generative adversarial neural network Physical Information Neural Network The interaction problem between shield tunneling and strata is solved and analyzed. The strata deformation data obtained by real-time monitoring is used to generate adversarial neural networks. , Physical Information Neural Network Iteratively update the parameters in to obtain the elastic modulus , Poisson's ratio Compared with S1, the conditional random field information features are more accurate. is the shield tunneling parameter vector, which is set in the early stage with reference to similar engineering experience. is the coordinate of the excavation surface position, Physical Information Neural Network Subsequent construction will continuously receive new stratum deformation monitoring values ​​and apply them to the physical information neural network. The prediction results of formation deformation are verified and calibrated. The output result of the formation deformation prediction is consistent with the actual monitoring value, so it can be considered that the generated adversarial neural network Characterization of elastic modulus , Poisson's ratio The spatial distribution of the conditional random field conforms to the actual situation of the site.

[0044] See also Figure 2 ,Different from traditional data driven neural networks, the physical information neural network ,can not only realize the fitting of the stratum deformation monitoring values ​​,at the corresponding position, but can also integrate the control equation of the ,mechanical mechanism of the interaction between shield tunneling and ,stratum into the training process.

[0045] Specifically in this embodiment, the final physical information neural network after iterative update The following conditions need to be met: the predicted value of the real-time stratum deformation at the corresponding monitoring location is equal to the actual value, which can be expressed in the following formula:

[0046] ;

[0047] in, Represents the stratum deformation value, is the spatial coordinate of the monitoring point.

[0048] In this embodiment, the iteratively updated physical information neural network The loss function By the control equation error (in this embodiment, the elastic constitutive relationship is used as an example) , data fitting error These two parts are composed and the formula is expressed as follows:

[0049] ;

[0050] ;

[0051] ;

[0052] in, and denote the stress tensor and strain tensor respectively, represents the displacement tensor, and the subscript comma represents the partial derivative. is the Kronecker function, is the weight of rock and soil, and is the Lame constant, which characterizes the elastic modulus and Poisson's ratio Conditional Random Field The corresponding output is converted to get, , Represent the corresponding number of sampling points.

[0053] Therefore, the loss function Error in the control equation Based on the physical information neural network representing the random field of spatial distribution of geotechnical parameters, the measured ground deformation data of the shield tunneling project were integrated. Iterative updates are performed to obtain physical information through the neural network Get accurate predictions of geotechnical parameters at the corresponding time.

[0054] S3, coupled with the generative adversarial neural network established in S2 Physical Information Neural Network The uncertainty in the spatial distribution of geotechnical parameters during shield tunneling is quantified, and the system reliability index to prevent surface subsidence and uplift from exceeding the allowable value is calculated. Then, shield tunneling parameter optimization based on the system reliability index is carried out to achieve safe control of shield construction.

[0055] In this embodiment, shield tunneling parameter optimization is based on system reliability design theory and global optimization genetic algorithm. Specifically, the shield tunneling parameter optimization process uses a trained and updated generative adversarial network. Characterizing Geotechnical Parameters , The random field of shield tunneling parameters is , Shield machine thrust , Cutter speed And cutter torque The combination vector is a variable, that is , using Monte Carlo simulation, based on a trained generative adversarial network The random field of geotechnical parameters represented by , Random field development Random simulations, we get Group , Random field simulation results. Random initialization , combine the parameters Input into the physical information neural network model after training and updating Perform calculation (at this time , remain unchanged), we get Surface deformation data The allowable stratum deformation range is ,in is the lower limit of the allowable value, corresponding to the maximum allowable settlement, is the upper limit of the tolerance, corresponding to the maximum allowable bulge.

[0056] The ground deformation induced by shield tunneling includes settlement and uplift. Therefore, the ground deformation exceeding the limit can be defined as a system reliability problem, which includes two subsystems. When the ground deformation exceeds the allowable upper limit, it corresponds to the failure of the first subsystem, which is defined as the event ; When the formation deformation exceeds the lower limit of the allowable limit, it corresponds to the failure of the second subsystem, which is defined as the event .remember The number of groups of data that exceeds the upper limit is , the number exceeding the lower limit is , the failure probability of the two subsystems is calculated as follows:

[0057] ;

[0058] ;

[0059] in, is the failure probability of the first subsystem, is the failure probability of the second subsystem. Based on the Ditlevsen method (a statistical method for estimating system reliability), the Ditlevsen limit of system reliability is calculated. The range of this limit is usually very narrow, and the average of the upper and lower limits is taken as the final reliability calculation result. Therefore, the probability of the occurrence of the system failure event F can be calculated as . The system failure probability , the corresponding system reliability can be converted . Set the system target reliability , quantitatively evaluate the risk probability of formation deformation exceeding the limit, and the ultimate goal is to make the calculated system reliability greater than the target reliability, so as to meet the safety control objectives.

[0060] Keep Geotechnical parameters , The random field simulation results remain unchanged, and the shield tunneling parameters are changed through genetic algorithm iterative optimization. The value of can be used to calculate the change process of reliability results under different values, such as Figure 3 As shown in the figure, take the first iteration as an example, and let the shield tunneling parameter be Reliability , the shield tunneling parameter value is Reliability , through iterative optimization , indicating that under the updated excavation parameter value conditions, the risk of over-limit of stratum deformation is reduced. In order to ensure construction safety and reduce energy consumption as much as possible, the optimization of shield excavation parameters can be transformed into the following multi-objective optimization problem:

[0061] ;

[0062] in, The reliability is obtained by multiple sampling simulations. Its value varies with shield tunneling parameters. Change and change, To set the target reliability, is the energy consumption index of the shield machine, which is affected by the shield tunneling parameters The impact of and They are respectively the lower and upper limits of the reference value range of shield tunneling parameters. The calculation formula is as follows:

[0063] ;

[0064] In the formula, It is the excavation time, which is determined during the excavation process.

[0065] Since this problem involves complex multi-objective optimization, in most cases, it is impossible to find the global optimal value that maximizes or minimizes all objectives. In the actual optimization process, the shield tunneling parameters undergo a certain number of iterations, and finally solve multiple Pareto optimal solutions, and then select a set of tunneling parameter values ​​from them to maximize the system reliability as much as possible on the basis of controlling energy consumption. Under the calculated reliability, the stratum deformation induced by shield tunneling can be effectively controlled, and at the same time, the energy consumption of shield construction is greatly reduced, with significant environmental protection effects. Figure 4 As shown, it comprehensively demonstrates the entire process of the method and the shield system involved.

[0066] Therefore, the method provided in this embodiment can fully quantify and express the uncertainty information of geotechnical parameters, and realize the accurate prediction of stratum deformation induced by shield tunneling. On this basis, the shield tunneling parameter optimization based on system reliability is carried out to achieve fine control of shield tunneling parameters, and further reduce costs and increase efficiency under the premise of ensuring the construction safety of shield tunneling projects.

[0067] Based on the same inventive concept, this embodiment also provides a shield tunneling parameter optimization system based on digital strata and system reliability, which collects shield tunneling parameter data recorded by the shield machine and on-site stratum deformation monitoring data. These data are uploaded in real time and continuously updated through the compiled physical information neural network program to complete the inversion of geotechnical parameters in the construction site and achieve accurate prediction of stratum deformation. On this basis, with the help of the compiled genetic algorithm program, shield tunneling parameter optimization based on system reliability is carried out to provide guidance for the setting of shield tunneling parameters in the cockpit.

[0068] Based on the same inventive concept, this embodiment also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the aforementioned shield tunneling parameter optimization method based on digital strata and system reliability is implemented.

[0069] The computer readable medium includes, but is not limited to, any type of disk (including floppy disk, hard disk, optical disk, CD-ROM, and magneto-optical disk), ROM, RAM, EPROM (Erasable Programmable Read-Only Memory), EEPROM, flash memory, magnetic card or optical card. In other words, the readable medium includes any medium that can store or transmit information in a readable form by a device (such as a computer).

[0070] Based on the same inventive concept, this embodiment also provides an electronic device, including at least one processor and a memory connected to the at least one processor in communication. The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the aforementioned shield tunneling parameter optimization method based on digital strata and system reliability.

[0071] The system, computer-readable storage medium, electronic device, etc. provided in this embodiment have the same inventive concept and the same beneficial effects as the aforementioned method, and will not be described in detail here.

[0072] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0073] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A shield tunneling parameter optimization method based on digital strata and system reliability, characterized in that: The steps include: S1, using all the geotechnical test parameters of the project, summarize and organize the geotechnical parameter data, and establish and train a generative adversarial neural network for characterizing the spatial variability of geotechnical parameters based on the geotechnical parameter data. ,in, Represents the spatial coordinate value of any point within the construction range. Generate adversarial neural networks The parameters determined by training in the generative adversarial network After training, it can be used to characterize the elastic modulus , Poisson's ratio The spatial distribution of random fields, generating adversarial neural networks Using the geotechnical parameter data at the sampling location, make a preliminary estimate of the geotechnical parameter values ​​at any location within the construction area; S2, build coupled generative adversarial neural network Physical Information Neural Network The interaction problem between shield tunneling and strata is solved and analyzed. The strata deformation data obtained by real-time monitoring is used to generate adversarial neural networks. , Physical Information Neural Network Iteratively update the parameters in to obtain the elastic modulus , Poisson's ratio Compared with the more accurate conditional random field information features in S1, is the shield tunneling parameter vector, is the coordinate of the excavation surface position, Physical Information Neural Network The parameters determined by training in the physical information neural network are used to continuously receive new stratum deformation monitoring values. Verify and calibrate the formation deformation prediction results; S3, through physical information neural network The uncertainty of the spatial distribution of geotechnical parameters in the shield tunneling process is quantified, and the system reliability index to prevent surface settlement and uplift from exceeding the allowable value is calculated. The shield tunneling parameter optimization based on the system reliability index is carried out to achieve safe control of shield construction.

2. The shield tunneling parameter optimization method based on digital strata and system reliability according to claim 1 is characterized in that: Iteratively update the final physical information neural network in S2 The following conditions are met: the real-time stratum deformation prediction value at the corresponding monitoring location is equal to the true value, and the formula is: ; in, Represents the stratum deformation value, is the spatial coordinate of the monitoring point.

3. The shield tunneling parameter optimization method based on digital strata and system reliability according to claim 2 is characterized in that: Iteratively updated physical information neural network The loss function The error in the control equation , data fitting error These two parts are composed of the formula: ; ; ; in, and denote the stress tensor and strain tensor respectively, represents the displacement tensor, and the subscript comma represents the partial derivative. is the Kronecker function, is the weight of rock and soil, and is the Lame constant, which characterizes the elastic modulus and Poisson's ratio Conditional Random Field The corresponding output is converted to get, , Represent the corresponding number of sampling points.

4. The shield tunneling parameter optimization method based on digital strata and system reliability according to claim 1 is characterized in that: In S3, shield tunneling parameter optimization for system reliability control is performed based on system reliability design theory and global optimization genetic algorithm.

5. The shield tunneling parameter optimization method based on digital strata and system reliability according to claim 4 is characterized in that: Using the trained and updated generative adversarial network in S3 Characterizing Geotechnical Parameters , The random field of shield tunneling parameters is , Shield machine thrust , Cutter speed And cutter torque The combination vector is a variable, that is , using Monte Carlo simulation, based on a trained generative adversarial network The random field of geotechnical parameters represented by , Random field development Random simulations, we get Group , Random field simulation results, with random initialization , combine the parameters Input into the physical information neural network model after training and updating Perform calculations, , All remain unchanged, and we get Surface deformation data The allowable stratum deformation range is ,in is the lower limit of the allowable value, corresponding to the maximum allowable settlement, is the upper limit of the allowable limit, corresponding to the maximum allowable uplift; The formation deformation exceeding the limit is defined as a system reliability problem, which includes two subsystems. When the formation deformation exceeds the allowable upper limit, it corresponds to the failure of the first subsystem, which is defined as the event ; When the formation deformation exceeds the lower limit of tolerance, it corresponds to the failure of the second subsystem, which is defined as the event ;remember The number of groups of data that exceeds the upper limit is , the number exceeding the lower limit is , calculate the failure probability of the two subsystems as: ; ; in, is the failure probability of the first subsystem, is the failure probability of the second subsystem; the Ditlevsen limit of the system reliability is calculated based on the Ditlevsen method, and the average of the upper and lower limits is taken as the final reliability calculation result. The probability of the system failure event F is calculated as ; The probability of system failure , and the corresponding system reliability is obtained ; Set system target reliability , quantitatively evaluate the risk probability of formation deformation exceeding the limit, so that the calculated system reliability is greater than the target reliability.

6. The shield tunneling parameter optimization method based on digital strata and system reliability according to claim 5 is characterized in that: Keep Geotechnical parameters , The random field simulation results remain unchanged, and the shield tunneling parameters are changed through genetic algorithm iterative optimization. The value of , under different values, the reliability result change process is calculated, and the shield tunneling parameter is set to Reliability , the shield tunneling parameter value is Reliability , through iterative optimization , and They are the probability curves of stratum deformation distribution under two sets of shield tunneling parameter values; the optimization of shield tunneling parameters can be transformed into the following multi-objective optimization problem: ; in, The reliability is obtained by multiple sampling simulations. Its value varies with the shield tunneling parameters. Change and change, To set the target reliability, is the energy consumption index of the shield machine, which is affected by the shield tunneling parameters The impact of and They are respectively the lower and upper limits of the reference value range of shield tunneling parameters.

7. The shield tunneling parameter optimization method based on digital strata and system reliability according to claim 6 is characterized in that: The calculation formula is as follows: ; in, It is the excavation time, which is determined during the excavation process.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, it implements the shield tunneling parameter optimization method based on digital strata and system reliability as described in any one of claims 1 to 7.

9. An electronic device, characterized in that: It includes at least one processor and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the shield tunneling parameter optimization method based on digital strata and system reliability as described in any one of claims 1 to 7.