Diameter-variable shield disturbance mixing test method and system

Abstract

The invention belongs to the technical field of underground structure tests, and provides a variable-diameter shield disturbance mixing test method and system.The method comprises the steps that posterior probability distribution is obtained according to parameter samples obtained from soil parameters; acquiring boundary displacement data according to a Bayesian derivation result based on the numerical substructure, and performing numerical analysis on the boundary displacement data based on the data conversion system to obtain predictive force data; obtaining prediction response data according to the prediction force data based on the physical substructure; and updating the numerical substructure according to the predicted response data based on the data conversion system, and determining a target balance force according to the new feedback force based on the updated numerical substructure to drive the next stage until the interaction boundary between the numerical substructure and the physical substructure meets displacement coordination and force balance. According to the method, the posterior probability distribution of the rock-soil body parameters is obtained through Bayesian inference, and the adaptability and prediction precision of the variable-diameter shield tunneling process to the construction disturbance evolution process are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of underground structure test, in particular to a variable-diameter shield disturbance hybrid test method and system. BACKGROUND

[0002] The variable-diameter shield tunneling process will exert additional load on the adjacent underground structure, which may cause damage or even destruction of the structure. Hybrid testing not only studies the interaction between different substructures from a system scale, but also studies the structure damage and destruction rules from a structure scale. However, the fidelity of the numerical substructure in the hybrid test is usually difficult to guarantee, thereby making it difficult to ensure the accuracy of the boundary load input by the numerical substructure to the test substructure.

[0003] The related technology introduces digital twinning technology in hybrid testing, which updates the numerical model to make its simulation results consistent with part of the measured data in the construction site in real time, thereby ensuring the fidelity of the numerical substructure. However, due to the uncertainty of the construction site rock-soil body, the simulation results of the response of the above method to the multi-source measuring points in the construction site are not accurate, thereby making it difficult to evaluate the construction risk caused by uncertainty. SUMMARY

[0004] The present application provides a variable-diameter shield disturbance hybrid test method and system to solve the defect that the hybrid test method of the prior art is difficult to evaluate the uncertainty of the construction site rock-soil body, resulting in inaccurate simulation results of the response of the multi-source measuring points in the construction site. The method improves the adaptability and prediction accuracy of the variable-diameter shield tunneling process to the construction disturbance evolution process.

[0005] The present application provides a variable-diameter shield disturbance hybrid test method, applied to a hybrid test system, the hybrid test system comprising a numerical substructure, a data conversion system and a physical substructure, the numerical substructure being constructed based on a three-dimensional finite element model, the method comprising: At each stage of tunneling by the variable-diameter shield, the parameter sample is subjected to Bayesian derivation to obtain the posterior probability distribution of the soil parameter and the Bayesian derivation result; wherein the parameter sample is obtained based on the prior distribution of the soil parameter and the MCMC sampling method; Based on the numerical substructure, the boundary displacement data is obtained according to the Bayesian derivation result, the boundary displacement data is subjected to numerical analysis based on the data conversion system to obtain predicted force data; based on the physical substructure, the equivalent load is applied on the boundary of the physical model by driving the loading device according to the predicted force data to obtain predicted response data; wherein the predicted response data comprises force response data and displacement response data; updating the numerical substructure according to the predicted response data based on the data conversion system; determining a new feedback force according to the force response data and the predicted force data based on the data conversion system, and determining a target balance force according to the new feedback force based on the updated numerical substructure, to drive a next stage of variable-diameter shield tunneling until the interaction boundary between the numerical substructure and the physical substructure meets displacement coordination and force balance; wherein the prior probability distribution corresponding to the next stage is the posterior probability distribution of the current stage.

[0006] According to the variable-diameter shield disturbance hybrid test method provided by the application, the Bayesian inference on the parameter sample comprises: The Bayesian inference on the parameter sample is performed based on the surrogate model to obtain the posterior probability distribution of the soil body parameter and the Bayesian inference result; wherein the surrogate model is obtained by iteratively training a target neural network based on a target algorithm and a training sample pair of sample soil body parameters and sample response data; wherein the target algorithm comprises a multivariate adaptive regression spline algorithm or a polynomial chaos kriging algorithm.

[0007] According to the variable-diameter shield disturbance hybrid test method provided by the application, the posterior probability distribution is obtained by the following steps: The response prediction data is obtained by processing the parameter sample based on the surrogate model; A likelihood function is constructed based on the response prediction data and the soil body parameter monitored in the current stage; The posterior probability distribution is calculated based on the prior probability distribution and the likelihood function.

[0008] According to the variable-diameter shield disturbance hybrid test method provided by the application, the numerical analysis on the boundary displacement data based on the data conversion system to obtain the predicted force data comprises: The soil-structure numerical model is called based on the data conversion system according to the boundary displacement data; The soil-structure numerical model is called based on the data conversion system according to the boundary displacement data;

[0009] According to the variable-diameter shield disturbance hybrid test method provided by the application, the updating of the numerical substructure according to the predicted response data based on the data conversion system comprises: The high-fidelity constitutive parameters and stiffness of the numerical substructure are updated according to the predicted response data based on the data conversion system to obtain the updated numerical substructure.

[0010] The variable-diameter shield disturbance mixed test method provided by the application comprises the following steps: obtaining a soil parameter sample; and performing Bayesian inference on the soil parameter sample to obtain a posterior probability distribution of the soil parameter and a Bayesian inference result.

[0011] The application further provides a variable-diameter shield disturbance mixed test system, comprising: The Bayesian inference module is configured to perform Bayesian inference on the parameter sample at each stage of tunneling by using the variable-diameter shield to obtain the posterior probability distribution of the soil parameter and the Bayesian inference result; wherein the parameter sample is obtained based on the prior distribution of the soil parameter and the MCMC sampling method. The analysis and loading module is configured to obtain boundary displacement data based on the Bayesian inference result and the numerical substructure, perform numerical analysis on the boundary displacement data based on the data conversion system to obtain predicted force data, drive the loading device to apply equivalent loads on the boundary of the physical model based on the predicted force data and the physical substructure to obtain predicted response data, wherein the predicted response data comprises force response data and displacement response data, and the numerical substructure is constructed based on a three-dimensional finite element model. The updating module is configured to update the numerical substructure based on the data conversion system and the predicted response data, determine a new feedback force based on the data conversion system and the force response data and the predicted force data, and determine a target balancing force based on the updated numerical substructure and the new feedback force to drive the next stage of tunneling by the variable-diameter shield until the interaction boundary between the numerical substructure and the physical substructure satisfies displacement coordination and force balance, wherein the prior probability distribution corresponding to the next stage is the posterior probability distribution of the current stage.

[0012] The application further provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the variable-diameter shield disturbance mixed test method according to any one of the above-described methods when executing the computer program.

[0013] The application further provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program is executable on a processor to implement the variable-diameter shield disturbance mixed test method according to any one of the above-described methods.

[0014] The application further provides a computer program product comprising a computer program, wherein the computer program is executable on a processor to implement the variable-diameter shield disturbance mixed test method according to any one of the above-described methods.

[0015] The variable-diameter shield disturbance hybrid test method and system provided by the application obtain the posterior probability distribution and the Bayesian inference result of the soil body parameters through Bayesian inference on the parameter samples, obtain boundary displacement data based on the Bayesian inference result of the numerical substructure, perform numerical analysis on the boundary displacement data based on the data conversion system to obtain predicted force data, drive the loading device to apply equivalent load on the boundary of the physical model based on the physical substructure according to the predicted force data to obtain predicted response data, and update the numerical substructure based on the data conversion system according to the predicted response data; the data conversion system is used to determine new feedback force based on the force response data and the predicted force data, and determine a target balance force based on the updated numerical substructure according to the new feedback force to drive the next stage until the boundary meets the displacement coordination and force balance; the posterior probability distribution of the rock-soil body parameters is obtained through Bayesian inference, the uncertainty quantification and the multi-source response fitting are cooperatively improved, and the adaptability and the prediction accuracy of the variable-diameter shield tunneling process to the construction disturbance evolution process are improved. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative effort.

[0017] Figure 1 is a flowchart of the variable-diameter shield disturbance hybrid test method provided by the application.

[0018] Figure 2 is a structural schematic diagram of the variable-diameter shield disturbance hybrid test system provided by the application.

[0019] Figure 3 is a structural schematic diagram of the electronic device provided by the application. DETAILED DESCRIPTION

[0020] In order to make the objects, technical solutions and advantages of the application clearer, the technical solutions in the application will be described clearly and completely in the following with reference to the drawings in the application. Obviously, the described embodiments are some embodiments of the application, but not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative effort belong to the protection scope of the application.

[0021] The variable-diameter shield disturbance hybrid test method and system provided by the application will be described below. Figures 1-2

[0022] Figure 1 ​is a flowchart of a variable-diameter shield disturbance mixing test method provided by the present application, as shown in Figure 1 The method is applied to a mixing test system, the mixing test system comprising a numerical substructure, a data conversion system and a physical substructure, the numerical substructure being constructed based on a three-dimensional finite element model, and comprising the following steps: In step 110, at each stage of tunneling by the variable-diameter shield, Bayesian inference is performed on the parameter sample to obtain the posterior probability distribution of the soil parameters and the Bayesian inference result; wherein the parameter sample is obtained based on the prior distribution of the soil parameters and the MCMC sampling method.

[0023] In this step, at the initial stage of tunneling by the variable-diameter shield, the prior probability distribution of the soil parameters can be set by pre-experiments.

[0024] For example, through pre-experiments based on geological exploration, engineering experience and analogy analysis, the prior probability distribution of multiple key soil parameters is constructed to represent the preliminary understanding of the stratum characteristics.

[0025] Specifically, before the mixing test is carried out, based on the constructed prior distribution of the soil parameters, the soil parameters monitored in the real environment are sampled by the Markov Chain Monte Carlo (MCMC) sampling method to generate the corresponding parameter sample.

[0026] In this step, at the non-initial stage of tunneling by the variable-diameter shield, the prior probability distribution of the soil parameters can be the posterior probability distribution obtained by Bayesian inference on the soil parameters of the previous stage; that is, the posterior probability distribution of the previous stage is continuously rolled and iterated as the prior probability distribution of the next stage, new monitoring data is continuously integrated, and the stability of parameter identification and the reliability of prediction results are improved.

[0027] In this embodiment, the soil parameters include at least one of the soil elastic modulus, the cohesion and the internal friction angle.

[0028] In this embodiment, the association between the soil parameters and the predicted values can be simulated by constructing a mathematical model or a neural network model, the residual between the predicted values and the actual construction site monitoring data is determined through the association, the credibility of the current parameter sample is quantified, the corresponding posterior probability distribution is calculated in combination with the prior probability distribution and the credibility of the soil parameters, and the Bayesian inference result corresponding to the soil parameters is output as the input of the numerical sub-model in the current stage of the mixing test.

[0029] In step 120, boundary displacement data is obtained based on the numerical substructure according to the Bayesian inference result, numerical analysis is performed on the boundary displacement data based on the data conversion system to obtain predicted force data, the loading device is driven to apply equivalent load on the boundary of the physical model based on the predicted force data according to the physical substructure to obtain predicted response data; wherein the predicted response data includes force response data and displacement response data.

[0030] In this step, the numerical substructure can accurately describe the shield tunneling disturbance effect and the stratum-structure interaction based on the three-dimensional finite element model, and the physical substructure can accurately reproduce the local damage evolution mechanism such as shield segment cracking, joint opening and bolt yielding through a large-scale physical model; the boundary displacement coordination and force balance control are realized based on the OpenFresco-LabVIEW-servo loading platform to ensure the consistency and reliability of the system response.

[0031] In this embodiment, the variable-diameter shield disturbance hybrid test needs to be carried out multiple times of data iteration (once for each tunneling stage), and each iteration process is divided into a forward data flow and a reverse data flow; wherein the forward data flow includes the following processes: In this embodiment, the interactive boundary can be set between the numerical substructure and the physical substructure through a virtual interactive platform (such as OpenFresco); the Bayesian inference result output in step 110 is input into the numerical substructure, the numerical substructure can accurately describe the shield tunneling disturbance effect and the stratum-structure interaction based on the three-dimensional finite element model, and the OpenFresco platform is used to read out the interactive boundary predicted displacement (i.e. boundary displacement data).

[0032] In this embodiment, inputting the boundary displacement data into the data conversion system for numerical analysis includes: calling a soil-structure numerical model based on the data conversion system according to the boundary displacement data; performing numerical analysis based on the soil-structure numerical model with the boundary displacement data as the boundary displacement load to obtain predicted force data.

[0033] In this embodiment, the soil-structure model of the data conversion system accurately realizes the "displacement→force" nonlinear mapping, and provides a scientific decision basis for millimeter-level displacement control and kilonewton-level load prediction of the variable-diameter shield engineering.

[0034] In this embodiment, the boundary predicted force data is set as the boundary condition as the loading condition of the physical substructure, and the loading device is driven to apply equivalent load on the boundary of the pre-set physical model in the physical substructure, and the soil and structure mechanical response (corresponding to the interaction force between soil and structure) is measured to obtain the corresponding response data and displacement response data.

[0035] The reverse data flow of the variable-diameter shield disturbance hybrid test provided in this embodiment includes the following processes: Step 130, updating the numerical substructure according to the predicted response data based on the data conversion system; determining a new feedback force based on the force response data and the predicted force data based on the data conversion system, and determining a target balance force based on the updated numerical substructure according to the new feedback force, to drive the next stage of the variable-diameter shield tunneling until the interaction boundary between the numerical substructure and the physical substructure satisfies the displacement coordination and force balance; wherein the prior probability distribution corresponding to the next stage is the posterior probability distribution of the current stage.

[0036] In this step, the physical substructure sends the predicted response data to the data conversion system, and the data conversion system calls the parameter evaluation and inverse analysis module by comparing the force response data (actual value) with the predicted force data (predicted value), to identify the stress model or physical properties such as material of the soil and structure, and update the physical properties of the soil-structure numerical model (i.e. the numerical substructure), recalculate the equivalent feedback force and return it to the numerical substructure.

[0037] In this embodiment, the physical properties include at least one of the high-fidelity constitutive parameters and the stiffness properties.

[0038] For example, updating the numerical substructure according to the predicted response data based on the data conversion system includes: updating the high-fidelity constitutive parameters and the stiffness of the numerical substructure according to the predicted response data based on the data conversion system to obtain the updated numerical substructure; this embodiment realizes true constitutive updating and stiffness updating through real-time inversion, improving the accuracy of data simulation of the numerical substructure.

[0039] In this embodiment, the numerical subsystem introduces a new unbalanced force according to the updated feedback force, and the unbalanced force is the difference between the predicted force and the feedback force, to drive a new round of iteration (i.e. the next stage of the variable-diameter shield tunneling), and after multiple rounds of iteration, the unbalanced force approaches zero (or the unbalanced force is lower than a preset tolerance), i.e. the interaction boundary between the physical substructure and the numerical substructure satisfies the displacement coordination and force balance conditions at the same time, achieving iteration convergence.

[0040] It should be noted that the posterior probability distribution of the current stage is taken as the prior probability distribution of the next stage, and a group of parameter samples are generated by the MCMC sampling method and input into the high-efficiency surrogate model to quickly obtain the corresponding predicted value, calculate the residual between the predicted value and the actual construction site monitoring data, and then use the residual to construct a likelihood function, and finally use the Bayesian formula to combine the prior distribution of this stage and the likelihood function to obtain the posterior probability distribution and the Bayesian derivation result of the next stage; then input the target unbalanced force and the Bayesian derivation result of the next stage into the numerical substructure to obtain the corresponding boundary displacement data for the subsequent test process; in turn, until the interaction boundary between the numerical substructure and the physical substructure satisfies the displacement coordination and force balance, completing the variable-diameter shield tunneling test.

[0041] As can be seen from the above examples, after each round of Bayesian update is completed, a mixed test based on the latest parameters is carried out for the whole-stage tunneling process, and the mixed test outputs stratum deformation, structure response and damage evolution information; as the iterative update of the soil body parameters and the gradual development of the test, combined with the quantitative results of the posterior distribution uncertainty, the test prediction accuracy gradually improves, which can provide real-time risk early warning, parameter regulation and construction method optimization decision support for the construction process.

[0042] The variable-diameter shield disturbance mixed test method provided in the embodiment of the application obtains the posterior probability distribution of the soil body parameters and the Bayesian inference result by Bayesian inference on the parameter samples, obtains the boundary displacement data based on the Bayesian inference result according to the numerical substructure, performs numerical analysis on the boundary displacement data based on the data conversion system, and obtains the predicted force data; drives the loading device to apply equivalent load on the boundary of the physical model according to the predicted force data based on the physical substructure, obtains the predicted response data, and updates the numerical substructure according to the predicted response data based on the data conversion system; determines the new feedback force based on the force response data and the predicted force data based on the data conversion system, and determines the target balance force based on the updated numerical substructure according to the new feedback force to drive the next stage until the boundary satisfies the displacement coordination and force balance; the process obtains the posterior probability distribution of the rock and soil body parameters through Bayesian inference, realizes the collaborative improvement of uncertainty quantification and multi-source response fitting, and improves the adaptability and prediction accuracy of the variable-diameter shield tunneling process to the construction disturbance evolution process.

[0043] In some embodiments, the Bayesian inference on the parameter samples comprises: performing Bayesian inference on the parameter samples based on a surrogate model to obtain the posterior probability distribution of the soil body parameters and the Bayesian inference result; wherein the surrogate model is obtained based on iterative training of a target neural network with sample soil body parameters and sample response data as training sample pairs and in combination with a target algorithm; wherein the target algorithm comprises a multivariate adaptive regression spline algorithm or a polynomial chaos kriging algorithm.

[0044] In this embodiment, the efficient surrogate model is established by the following steps: (1) Latin hypercube sampling is performed on the soil body parameter set to generate sample soil body parameters.

[0045] In this embodiment, the parameter samples can be subjected to data cleaning or normalization preprocessing operations to reduce the influence of abnormal values and improve the quality of the parameter samples.

[0046] (2) Calculate the response data based on the preset finite element model according to the parameter samples to obtain the corresponding sample response data, and form training sample pairs with the sample soil body parameters.

[0047] (3) Using multiple adaptive regression splines (MARS), polynomial chaos kriging (PCK) algorithm to iteratively train the convolutional neural network or the Transformer model, and under the condition of meeting the iteration condition, a proxy model is obtained, which has the ability to realize the rapid prediction of the parameter-response relationship.

[0048] The variable-diameter shield disturbance mixed test method provided by the embodiment of the application can obtain the posterior probability distribution and the Bayesian inference result of the soil body parameters by performing Bayesian inference on the parameter sample through the proxy model, realizes efficient prediction of the soil body sample prediction and response, and improves the mixed test efficiency.

[0049] In some embodiments, the posterior probability distribution is obtained by the following steps: processing the parameter sample based on the proxy model to obtain response prediction data; constructing a likelihood function based on the response prediction data and the soil body parameters monitored in the current stage; and calculating the posterior probability distribution based on the prior probability distribution and the likelihood function.

[0050] In this embodiment, before the mixed test is carried out, a group of parameter samples are generated based on the constructed soil body parameter prior distribution through the Markov chain Monte Carlo (MCMC) sampling method, and are input into the efficient proxy model to quickly obtain corresponding response prediction data, calculate the residual between the predicted value and the actual construction site monitoring data, and construct a likelihood function by assuming that the residual is subject to a Gaussian distribution, which is used to quantify the credibility of the current parameter sample.

[0051] In this embodiment, the likelihood function L ( x ) is represented by the following formula: ; Wherein, i is the number of data observation points, is the construction site monitoring value, is the model prediction value, is the error variance; the prior distribution and the likelihood function are combined by using the Bayesian formula to obtain the posterior distribution of the parameters, and the distribution reflects the correction result of the current stage data on the parameters.

[0052] The variable-diameter shield disturbance hybrid test method provided by the embodiment of the present application processes parameter samples through a proxy model to obtain response prediction data; constructs a likelihood function based on the response prediction data and soil parameters monitored at a current stage; calculates a posterior probability distribution based on a prior probability distribution and the likelihood function; organically combines the hybrid test process with Bayesian inference to establish a hybrid test technical system for response identification and rolling optimization in the whole construction process, and significantly improves the scientificity and reliability of risk control and response prediction in the variable-diameter shield tunneling process.

[0053] The variable-diameter shield disturbance hybrid test system provided by the present application is described below, and the variable-diameter shield disturbance hybrid test system described below can be correspondingly referred to the variable-diameter shield disturbance hybrid test method described above.

[0054] Figure 2 is a structural schematic diagram of the variable-diameter shield disturbance hybrid test system provided by the present application, as Figure 2 indicated, the variable-diameter shield disturbance hybrid test system comprises a Bayesian inference module 210, an analysis and loading module 220, and an updating module 230.

[0055] The Bayesian inference module 210 is configured to perform Bayesian inference on parameter samples at each stage of tunneling by using a variable-diameter shield to obtain a posterior probability distribution of soil parameters and a Bayesian inference result; wherein the parameter samples are obtained based on a prior distribution of soil parameters and a MCMC sampling method. The analysis and loading module 220 is configured to obtain boundary displacement data based on a numerical substructure according to the Bayesian inference result, perform numerical analysis on the boundary displacement data based on a data conversion system to obtain predicted force data, drive a loading device to apply equivalent loads on the boundary of a physical model based on a physical substructure according to the predicted force data to obtain predicted response data; wherein the predicted response data comprises force response data and displacement response data; and the numerical substructure is constructed based on a three-dimensional finite element model. The updating module 230 is configured to update the numerical substructure based on the data conversion system according to the predicted response data, determine a new feedback force based on the data conversion system according to the force response data and the predicted force data, and determine a target balance force based on the updated numerical substructure according to the new feedback force to drive the next stage of tunneling by the variable-diameter shield until the interaction boundary between the numerical substructure and the physical substructure satisfies displacement coordination and force balance; wherein the prior probability distribution corresponding to the next stage is the posterior probability distribution of the current stage.

[0056] The variable-diameter shield disturbance mixed test system provided by the embodiment of the present application obtains the posterior probability distribution and the Bayesian inference result of the soil body parameter by Bayesian inference on the parameter sample, obtains boundary displacement data according to the Bayesian inference result based on a numerical substructure, performs numerical analysis on the boundary displacement data based on a data conversion system, and obtains predicted force data; drives a loading device to apply equivalent load on the boundary of the physical model according to the predicted force data based on a physical substructure, obtains predicted response data, and updates the numerical substructure according to the predicted response data based on the data conversion system; determines a new feedback force according to the force response data and the predicted force data based on the data conversion system, and determines a target balance force according to the new feedback force based on the updated numerical substructure to drive the next stage until the boundary meets the displacement coordination and the force balance; the process obtains the posterior probability distribution of the rock-soil body parameter through Bayesian inference, realizes the collaborative improvement of the uncertainty quantification and the multi-source response fitting, and improves the adaptability and the prediction accuracy of the variable-diameter shield tunneling process to the construction disturbance evolution process.

[0057] Figure 3 An example of a schematic diagram of a physical structure of an electronic device is shown in FIG. 1. Figure 3As shown, the electronic device can include a processor 310, a communications interface 320, a memory 330, and a communications bus 340, wherein the processor 310, the communications interface 320, and the memory 330 complete mutual communication through the communications bus 340. The processor 310 can invoke a logic instruction in the memory 330 to execute a variable-diameter shield disturbance mixing test method, which is applied to a mixing test system including a numerical substructure, a data conversion system, and a physical substructure. The numerical substructure is constructed based on a three-dimensional finite element model and includes: at each stage of tunneling using a variable-diameter shield, performing Bayesian inference on a parameter sample to obtain a posterior probability distribution of soil parameters and a Bayesian inference result; wherein the parameter sample is obtained based on a soil parameter prior distribution and an MCMC sampling method; obtaining boundary displacement data based on the numerical substructure according to the Bayesian inference result, performing numerical analysis on the boundary displacement data based on the data conversion system to obtain predicted force data; driving a loading device to apply equivalent loads on the boundary of the physical model based on the physical substructure according to the predicted force data to obtain predicted response data; wherein the predicted response data includes force response data and displacement response data; updating the numerical substructure based on the data conversion system according to the predicted response data; determining a new feedback force based on the data conversion system according to the force response data and the predicted force data, and determining a target balancing force based on the updated numerical substructure according to the new feedback force to drive the next stage of tunneling of the variable-diameter shield, until the interaction boundary between the numerical substructure and the physical substructure satisfies displacement coordination and force balance; wherein the prior probability distribution corresponding to the next stage is the posterior probability distribution of the current stage.

[0058] In addition, the logic instruction in the memory 330 described above can be implemented in the form of a software function unit and sold or used as an independent product, which can be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0059] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program being stored on a non-transitory computer readable storage medium, and the computer program being executable by a processor to enable a computer to perform the variable-diameter shield disturbance mixing test method provided by any of the above methods, the method being applied to a mixing test system comprising a numerical substructure, a data conversion system and a physical substructure, the numerical substructure being constructed based on a three-dimensional finite element model and comprising: performing Bayesian inference on a parameter sample at each stage of tunneling by the variable-diameter shield to obtain a posterior probability distribution of soil parameters and a Bayesian inference result; wherein the parameter sample is obtained based on a prior distribution of the soil parameters and an MCMC sampling method; obtaining boundary displacement data based on the numerical substructure according to the Bayesian inference result, performing numerical analysis on the boundary displacement data based on the data conversion system to obtain predicted force data; driving a loading device to apply equivalent loads on the boundary of the physical model based on the physical substructure according to the predicted force data to obtain predicted response data; wherein the predicted response data comprises force response data and displacement response data; updating the numerical substructure based on the data conversion system according to the predicted response data; determining a new feedback force based on the data conversion system according to the force response data and the predicted force data, and determining a target equilibrium force based on the updated numerical substructure according to the new feedback force to drive the next stage of tunneling by the variable-diameter shield, until the interaction boundary between the numerical substructure and the physical substructure satisfies displacement coordination and force balance; wherein the prior probability distribution corresponding to the next stage is the posterior probability distribution of the current stage.

[0060] In yet another aspect, the present application also provides a non-transitory computer readable storage medium having stored thereon a computer program, which, when executed by a processor, implements the variable-diameter shield disturbance mixing test method provided by the above method, and the method is applied to a mixing test system including a numerical substructure, a data conversion system and a physical substructure, the numerical substructure is constructed based on a three-dimensional finite element model, and the method comprises: at each stage of tunneling by using a variable-diameter shield, performing Bayesian inference on a parameter sample to obtain a posterior probability distribution of soil parameters and a Bayesian inference result; wherein the parameter sample is obtained based on a prior distribution of soil parameters and an MCMC sampling method; obtaining boundary displacement data based on the numerical substructure according to the Bayesian inference result, performing numerical analysis on the boundary displacement data based on the data conversion system to obtain predicted force data; driving a loading device to apply equivalent loads on the boundary of the physical model based on the physical substructure according to the predicted force data to obtain predicted response data; wherein the predicted response data includes force response data and displacement response data; updating the numerical substructure based on the data conversion system according to the predicted response data; determining a new feedback force based on the data conversion system according to the force response data and the predicted force data, and determining a target balance force based on the updated numerical substructure according to the new feedback force to drive the next stage of tunneling by the variable-diameter shield, until the interaction boundary between the numerical substructure and the physical substructure satisfies displacement coordination and force balance; wherein the prior probability distribution corresponding to the next stage is the posterior probability distribution of the current stage.

[0061] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0062] From the above description of the embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software plus necessary general hardware platforms, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0063] It should be pointed out finally that the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit the same; and although the present application has been described in detail with reference to the foregoing embodiments, it should be appreciated by those skilled in the art that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features thereof can be replaced equivalently; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A variable diameter shield disturbance mixing test method, characterized by, The method is applied to a hybrid test system comprising a numerical substructure, a data conversion system and a physical substructure, the numerical substructure being constructed based on a three-dimensional finite element model, and the method comprises: At each stage of tunneling by the variable-diameter shield, Bayesian inference is performed on the parameter sample to obtain a posterior probability distribution of the soil parameters and a Bayesian inference result; wherein the parameter sample is obtained based on a prior distribution of the soil parameters and an MCMC sampling method; Based on the numerical substructure, boundary displacement data is obtained according to the Bayesian inference result, numerical analysis is performed on the boundary displacement data based on the data conversion system to obtain predicted force data; based on the physical substructure, an equivalent load is applied on the boundary of the physical model by driving the loading device according to the predicted force data to obtain predicted response data; wherein the predicted response data includes force response data and displacement response data; Based on the data conversion system, the numerical substructure is updated according to the predicted response data; based on the data conversion system, a new feedback force is determined according to the force response data and the predicted force data, and a target balancing force is determined according to the new feedback force based on the updated numerical substructure to drive the next stage of tunneling by the variable-diameter shield, until the interaction boundary between the numerical substructure and the physical substructure satisfies displacement coordination and force balance; wherein the prior probability distribution corresponding to the next stage is the posterior probability distribution of the current stage.

2. The variable-diameter shield disturbance mixing test method according to claim 1, characterized by, The Bayesian inference on the parameter sample comprises: Based on the proxy model, Bayesian inference is performed on the parameter sample to obtain the posterior probability distribution of the soil parameters and the Bayesian inference result; wherein the proxy model is obtained by iteratively training a target neural network based on a training sample pair of sample soil parameters and sample response data, and combining a target algorithm; wherein the target algorithm includes a multivariate adaptive regression spline algorithm or a polynomial chaos kriging algorithm.

3. The variable-diameter shield disturbance mixing test method according to claim 2, characterized by, The posterior probability distribution is obtained by the following steps: Based on the proxy model, the parameter sample is processed to obtain response prediction data; Based on the response prediction data and the soil parameters monitored in the current stage, a likelihood function is constructed; Based on the prior probability distribution and the likelihood function, the posterior probability distribution is calculated.

4. The variable diameter shield disturbance hybrid test method of claim 1, wherein, The numerical analysis of the boundary displacement data based on the data conversion system to obtain the predicted force data comprises: Based on the data conversion system, a soil-structure numerical model is called according to the boundary displacement data; Based on the soil-structure numerical model, numerical analysis is performed with the boundary displacement data as boundary displacement load to obtain the predicted force data.

5. The variable-diameter shield disturbance mixing test method according to claim 1, wherein The updating of the numerical substructure based on the data conversion system according to the predicted response data comprises: Based on the data conversion system, the high-fidelity constitutive parameters and stiffness of the numerical substructure are updated according to the predicted response data to obtain the updated numerical substructure.

6. The variable-diameter shield disturbance mixing test method according to claim 1 or 2, characterized by, The soil parameters include at least one of the soil elastic modulus, cohesion and internal friction angle.

7. A variable diameter shield disturbance mixing test system, characterized in that, Comprise: The Bayesian derivation module is configured to perform Bayesian derivation on the parameter sample at each stage of tunneling by the variable-diameter shield to obtain a posterior probability distribution of the soil parameter and a Bayesian derivation result; the parameter sample is obtained based on a prior distribution of the soil parameter and a Markov Chain Monte Carlo (MCMC) sampling method; The analysis and loading module is configured to obtain boundary displacement data based on a numerical substructure according to the Bayesian derivation result, perform numerical analysis on the boundary displacement data based on a data conversion system to obtain predicted force data, drive a loading device to apply equivalent load on the boundary of the physical model according to the predicted force data based on a physical substructure to obtain predicted response data, and wherein the predicted response data includes force response data and displacement response data; the numerical substructure is constructed based on a three-dimensional finite element model; The updating module is configured to update the numerical substructure according to the predicted response data based on the data conversion system, determine a new feedback force according to the force response data and the predicted force data based on the data conversion system, and determine a target balance force according to the new feedback force based on the updated numerical substructure to drive the next stage of tunneling by the variable-diameter shield until the interaction boundary between the numerical substructure and the physical substructure satisfies displacement coordination and force balance; wherein the prior probability distribution corresponding to the next stage is the posterior probability distribution of the current stage.

8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and running on the processor, characterized in that, The processor executes the computer program to implement the variable-diameter shield disturbance hybrid test method according to any one of claims 1 to 6. 9.A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the variable-diameter shield disturbance hybrid test method according to any one of claims 1 to 6.

Images