CN115563831B

CN115563831B - Tunnel stratum mechanical parameter acquisition method and device, electronic equipment and storage medium

Info

Publication number: CN115563831B
Application number: CN202211289730.9A
Authority: CN
Inventors: 闫鹏洋; 王长欣; 田淑明; 赵洪斌
Original assignee: Beijing Yunlu Technology Co Ltd
Current assignee: Beijing Yunlu Technology Co Ltd
Priority date: 2022-10-20
Filing date: 2022-10-20
Publication date: 2023-07-21
Anticipated expiration: 2042-10-20
Also published as: CN115563831A

Abstract

The application provides a tunnel stratum mechanical parameter acquisition method and device, electronic equipment and a storage medium. The method comprises the following steps: constructing an associated finite element CAE stratum mechanical inversion parameter-earth surface subsidence data set according to the stratum mechanical inversion parameter and the finite element CAE simulation model, wherein the stratum mechanical inversion parameter is a mechanical parameter corresponding to a plurality of stratum in a preset range; constructing a PSO-BP neural network through parameters in the earth surface subsidence data set, and performing model training; and inputting the original surface subsidence data into the trained PSO-BP neural network to obtain mechanical parameters of the tunnel stratum. According to the method, the PSO-BP neural network is constructed and trained through parameters in the finite element CAE stratum mechanics inversion parameter-earth surface subsidence data set, the PSO-BP neural network model parameters are dynamically adjusted, the training efficiency and the convergence speed are improved, further more accurate tunnel stratum mechanics parameters are obtained through the PSO-BP neural network, and the accuracy of the tunnel stratum mechanics parameters is improved.

Description

Tunnel stratum mechanical parameter acquisition method and device, electronic equipment and storage medium

Technical Field

The present disclosure relates to the field of tunnel engineering, and in particular, to a method and an apparatus for obtaining mechanical parameters of a tunnel stratum, an electronic device, and a storage medium.

Background

When the subway shield tunnel is constructed and excavated, the problems of complex mechanical properties, discontinuity, irregular defects and the like of the rock-soil body occur, so that the disturbance of the excavation construction to the rock-soil body is large, and the safety of an engineering body, surrounding buildings and structures can be greatly influenced. The finite element CAE simulation calculation method is used for simulating the working conditions of the underground engineering to verify the feasibility, safety and stability of the engineering, and is an important measure for improving the safety of the underground engineering. Because of the complexity and uncertainty of the rock and soil mass, inaccurate mechanical parameter supply is a bottleneck for restricting the development of the rock and soil mass, conventional geological survey and empirical data can only provide preliminary references, so that finite element CAE simulation of tunnel engineering is in a random dilemma of parameter selection for a long time, the due function of the finite element CAE simulation cannot be fully exerted, and the acquired mechanical parameter of the tunnel stratum is low in accuracy.

Disclosure of Invention

In view of the foregoing, an objective of an embodiment of the present application is to provide a method, an apparatus, an electronic device, and a storage medium for obtaining mechanical parameters of a tunnel stratum. The accuracy of the acquired tunnel stratum mechanical parameters can be improved.

In a first aspect, an embodiment of the present application provides a method for obtaining a mechanical parameter of a tunnel stratum, including: constructing an associated finite element CAE stratum mechanical inversion parameter-earth surface subsidence data set according to the stratum mechanical inversion parameter and the finite element CAE simulation model, wherein the stratum mechanical inversion parameter is a mechanical parameter corresponding to a plurality of stratum in a preset range; constructing a PSO-BP neural network through parameters in the earth surface subsidence data set, and performing model training; inputting the original earth surface subsidence data into the PSO-BP neural network after training to obtain mechanical parameters of the tunnel stratum; the original earth surface subsidence data are subsidence data acquired at preset earth surface subsidence measuring points.

In the implementation process, the PSO algorithm is fused in the BP neural network, and the problems of local extremum, low convergence speed and the like of the BP neural network algorithm can be solved based on the advantages of high convergence speed, strong global searching capability and high robustness of the PSO algorithm. In addition, the PSO-BP neural network is constructed and trained through the parameters in the finite element CAE stratum mechanics inversion parameter-earth surface subsidence data set, the PSO-BP neural network model parameters are dynamically adjusted, the training efficiency and the convergence speed are improved, further more accurate tunnel stratum mechanics parameters are obtained through the PSO-BP neural network, and the accuracy of the tunnel stratum mechanics parameters is improved.

In one embodiment, the constructing the associated finite element CAE formation mechanical inversion parameter-surface subsidence dataset according to the formation mechanical inversion parameter and the finite element CAE simulation model comprises: determining an orthogonal test table according to the stratum mechanics inversion parameters and the corresponding stratum mechanics inversion parameter grading level conditions; grading the stratum mechanics inversion parameters according to the orthogonal test table to obtain a plurality of groups of orthogonal test combination parameters; inputting the multiple groups of orthogonal test combination parameters into the finite element CAE simulation model to construct an associated finite element CAE stratum mechanical inversion parameter-surface subsidence data set.

In the implementation process, in the inversion process of the stratum mechanical parameters of the shield tunnel, the stratum mechanical inversion parameters are classified by adopting an orthogonal test method, and the orthogonal test combination parameters are obtained to serve as input parameters of a finite element CAE simulation model so as to construct an associated finite element CAE stratum mechanical inversion parameter-surface subsidence data set. The different mechanical inversion parameters of different strata can be classified according to the corresponding classification levels, so that unified processing of the mechanical inversion parameters of different strata is realized, and the applicable scene for constructing the related finite element CAE stratum mechanical inversion parameter-earth surface subsidence data set is increased.

In one embodiment, the plurality of sets of orthogonal test combination parameters include graded mechanical parameters, and the method further includes, before constructing a PSO-BP neural network from the parameters in the surface subsidence dataset and performing model training: normalizing the classified stratum mechanics inversion parameters; the PSO-BP neural network is constructed through parameters in the earth surface subsidence data set, model training is carried out, and the method comprises the following steps: and constructing a PSO-BP neural network according to the classified mechanical parameters after normalization treatment, and performing model training.

In the implementation process, the PSO-BP neural network is constructed and trained based on the mechanical parameters after the normalization processing by carrying out normalization processing on the classified stratum mechanical inversion parameters, so that the PSO-BP neural network training efficiency and convergence speed can be improved.

In one embodiment, the constructing the PSO-BP neural network according to the normalized classified mechanical parameters includes: determining the super parameters of the PSO-BP neural network according to the input parameters of the PSO-BP neural network and the output parameters of the PSO-BP neural network; constructing the PSO-BP neural network according to the super parameters and a preset network structure; the output parameters of the PSO-BP neural network are determined according to the classified mechanical parameters after normalization processing, the input parameters of the PSO-BP neural network are determined according to ground surface subsidence simulation values, and the ground surface subsidence simulation values are obtained through calculation of the multiple groups of orthogonal test combination parameters and the finite element CAE simulation model.

In the implementation process, because the finite element CAE stratum mechanics inversion parameters-the surface subsidence data set contains parameters such as elastic modulus, poisson ratio, cohesive force, internal friction angle, grouting layer elastic modulus, surface subsidence simulation value and the like, different parameters have different dimension orders of magnitude, the model can promote the weight of the large-magnitude data, and normally, the PSO-BP model can not accurately identify the implicit relation in the data. According to the embodiment, through determining each super parameter in the PSO-BP neural network according to the classified parameters, the association relation between each parameter in the finite element CAE stratum mechanical inversion parameter-earth surface subsidence data set and the super parameter in the PSO-BP neural network is established, and then the implicit relation in the data can be accurately identified. In addition, the mechanical inversion parameters of finite element CAE stratum mechanics-the mechanical parameters in the earth surface subsidence data set are normalized, and can be mapped to a fixed interval, so that the training efficiency of the PSO-BP neural network model is improved.

In one embodiment, the method further comprises: inputting the tunnel stratum mechanical parameters into the finite element CAE simulation model to verify the accuracy of the tunnel stratum mechanical parameters so as to update the tunnel stratum mechanical parameters according to a verification result.

In the implementation process, after the mechanical parameters of the tunnel stratum are determined, the parameters are verified through the finite element CAE simulation model, the defect of insufficient optimization of the PSO-BP neural network is found in time, the mechanical parameters of the tunnel stratum are updated through updating the PSO-BP neural network, the accuracy of the mechanical parameters of the tunnel stratum is guaranteed, and the accuracy and reliability of a parameter inversion result are improved.

In one embodiment, the inputting the tunnel formation mechanical parameters into the finite element CAE simulation model verifies accuracy of the tunnel formation mechanical parameters to update the tunnel formation mechanical parameters according to the verification result, including: determining a judging function according to the ground surface subsidence simulation value and the original ground surface subsidence data; verifying the accuracy of the tunnel stratum mechanical parameters according to the evaluation function so as to update the tunnel stratum mechanical parameters according to a verification result; the ground surface subsidence simulation value is obtained through calculation of the orthogonal test combination parameters and the finite element CAE simulation model.

In the implementation process, the average value of the relative errors of the ground surface subsidence simulation value and the original ground surface subsidence data is determined according to the formula of the judging function, so that whether the tunnel stratum mechanical parameter meets the accuracy requirement is judged, the tunnel stratum mechanical parameter is updated according to the accuracy judgment result of the tunnel stratum mechanical parameter, the average value of the relative errors of the ground surface subsidence simulation value and the original ground surface subsidence data can reflect the deviation degree of the simulation result and the actual result obtained by inversion analysis of the output mechanical parameter, and the tunnel stratum mechanical parameter exceeding the error range is updated, so that the accuracy of the tunnel stratum mechanical parameter is improved.

In one embodiment, the verifying the accuracy of the tunnel formation mechanical parameter according to the evaluation function to update the tunnel formation mechanical parameter according to the verification result includes: judging whether the evaluation function is not larger than a first preset value or not; and if the evaluation function is larger than the first preset value, updating the tunnel stratum mechanical parameter according to the magnitude relation between the iteration times and the second preset value.

In the implementation process, whether the error values of the simulation result and the actual result obtained by the mechanical parameter inversion analysis are in the error range or not is determined according to the magnitude relation between the judging function and the first preset value, if the error values of the simulation result and the actual result obtained by the mechanical parameter inversion analysis exceed the error range, the mechanical parameters of the tunnel stratum are updated, the errors of the mechanical parameters of the tunnel stratum are ensured to be in the error range, and the accuracy of the mechanical parameters of the tunnel stratum is improved.

In one embodiment, the updating the tunnel formation mechanical parameter according to the magnitude relation between the iteration number and the second preset value includes: judging whether the iteration times are not larger than a second preset value or not; if the iteration times are not greater than the second preset value, updating the iteration times, reconstructing a PSO-BP neural network and performing model training; inputting the original surface subsidence data into the PSO-BP neural network after training to obtain tunnel stratum mechanical parameters until the evaluation function is not more than the first preset value; if the iteration times are larger than the second preset value, updating the hierarchical level number of the formation mechanical inversion parameter, and reconstructing an associated finite element CAE formation mechanical inversion parameter-surface subsidence data set according to the updated formation mechanical inversion parameter and the finite element CAE simulation model; constructing a PSO-BP neural network through parameters in the earth surface subsidence data set, and performing model training; inputting the original surface subsidence data into the PSO-BP neural network after training to obtain the mechanical parameters of the tunnel stratum until the evaluation function is not more than the first preset value.

In the implementation process, when the judging function is larger than the first preset value, the magnitude relation between the iteration times and the second preset value is further judged, whether the iteration times meet the optimization conditions is further determined, when the iteration times meet the optimization conditions, the formation mechanical inversion parameter grading level number is updated, the associated finite element CAE formation mechanical inversion parameter-earth surface subsidence data set is reconstructed, and then the PSO-BP neural network is reconstructed and model training is performed. And when the iteration times do not meet the optimization conditions, reconstructing the PSO-BP neural network directly according to the updated current iteration times and the maximum iteration times, and performing model training. And inputting the original surface subsidence data into the retrained PSO-BP neural network to obtain the mechanical parameters of the tunnel stratum and verifying the accuracy of the inversion parameters again until the judging function is not larger than a first preset value. Through the judgment and circulation, the PSO-BP neural network is adjusted and optimized, so that the error of the mechanical parameters of the tunnel stratum is gradually reduced, the performance of the PSO-BP neural network is improved, and the accuracy of the mechanical parameters of the tunnel stratum is improved.

In one embodiment, the formation mechanical inversion parameters include: modulus of elasticity of the slip layer.

In the implementation process, the elastic modulus of the grouting layer is increased in the stratum mechanical inversion parameters, the influence of the grouting layer on the surface subsidence is considered, the accuracy of inversion of the stratum mechanical parameters is further improved, and the accuracy of the tunnel stratum mechanical parameters is further improved.

In a second aspect, an embodiment of the present application further provides a device for obtaining a mechanical parameter of a tunnel stratum, including: the first construction module is used for constructing an associated finite element CAE stratum mechanical inversion parameter-earth surface subsidence data set according to the stratum mechanical inversion parameter and the finite element CAE simulation model, wherein the stratum mechanical inversion parameter is a mechanical parameter corresponding to a plurality of strata in a preset range; the second construction module is used for constructing a PSO-BP neural network through parameters in the earth surface subsidence data set and performing model training; the calculation module is used for inputting the original earth surface subsidence data into the PSO-BP neural network after training to obtain mechanical parameters of the tunnel stratum, wherein the original earth surface subsidence data are subsidence data obtained at preset earth surface subsidence measuring points.

In a third aspect, embodiments of the present application further provide an electronic device, including: a processor, a memory storing machine-readable instructions executable by the processor, which when executed by the processor, perform the steps of the method of the first aspect, or any of the possible implementations of the first aspect.

In a fourth aspect, the embodiments of the present application further provide a computer readable storage medium, on which a computer program is stored, which when executed by a processor performs the steps of the tunnel soil body parameter obtaining method of the first aspect, or any one of the possible implementation manners of the first aspect.

In order to make the above objects, features and advantages of the present application more comprehensible, embodiments accompanied with figures are described in detail below.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered limiting the scope, and that other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flowchart of a method for obtaining mechanical parameters of a tunnel formation according to an embodiment of the present application;

fig. 2 is a spatial position relationship diagram of a tunnel model according to an embodiment of the present application;

FIG. 3 is a schematic diagram of a tunnel model boundary condition setup according to an embodiment of the present disclosure;

FIG. 4 is a flow chart for verifying mechanical parameters of a tunnel formation according to an embodiment of the present application;

fig. 5 is a schematic functional block diagram of a device for obtaining mechanical parameters of a tunnel stratum according to an embodiment of the present application;

fig. 6 is a block schematic diagram of an electronic device according to an embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.

It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only to distinguish the description, and are not to be construed as indicating or implying relative importance.

Stratum factors and construction factors influencing surface subsidence in the subway shield tunnel excavation process are more, such as physical mechanical parameters including stratum elasticity modulus, stratum poisson ratio, stratum cohesion, stratum internal friction angle and the like. The existing subway shield tunnel displacement inverse analysis method generally combines on-site earth surface subsidence monitoring data with a finite element CAE simulation technology. Calculating the existing finite element CAE simulation model under different input parameters by establishing a finite element CAE simulation calculation model, establishing a deep learning database, introducing an intelligent optimization algorithm, combining on-site earth surface subsidence monitoring data, inverting stratum mechanical parameters, applying the mechanical parameters obtained through inversion calculation to calculation of the finite element CAE simulation model, calculating to obtain a finite element CAE simulation earth surface subsidence value, and comparing and verifying the finite element CAE simulation earth surface subsidence value with on-site corresponding earth surface subsidence monitoring data, thereby selecting optimal stratum mechanical parameters.

However, the existing technical application scene of inversion of stratum mechanical parameters of the front shield tunnel is mostly used for a single stratum, and the actual stratum of the shield tunnel construction site is usually a plurality of stratum, so that the existing technical application scene of stratum parameters cannot meet the actual production requirements. In the actual construction process, a grouting layer formed by mixing soil and mortar is formed around the tunnel when the shield tunnel is synchronously grouting, and the grouting layer is used for filling gaps between shield segments and surrounding soil and reducing earth surface subsidence. The influence of the current shield tunnel stratum mechanical parameter inversion technology on the earth surface subsidence caused by the grouting layer elastic modulus is not considered, so that the inversion can not obtain more accurate stratum mechanical parameters.

In view of the above, the inventor of the application provides a tunnel stratum mechanical parameter acquisition method by long-term research aiming at the situation that the application scene of the existing shield tunnel stratum mechanical parameter inversion technology is limited and accurate stratum mechanical parameters are difficult to obtain, and establishes a multi-stratum shield tunnel finite element CAE simulation model by taking mechanical parameters such as elastic modulus, poisson ratio, cohesive force, internal friction angle, grouting layer elastic modulus and the like as inversion parameters, and adopts a 'finite element CAE stratum mechanical inversion parameter-earth surface subsidence data set', and a PSO-BP neural network is fused, so that more accurate mechanical parameters can be acquired, and more accurate and reliable rock-soil body mechanical parameters are provided for later construction simulation and field construction optimization design.

Referring to fig. 1, a flowchart of a method for obtaining mechanical parameters of a tunnel formation according to an embodiment of the present application is shown. The specific flow shown in fig. 2 will be described in detail.

Step 201, constructing a related finite element CAE stratum mechanical inversion parameter-surface subsidence data set according to the stratum mechanical inversion parameter and the finite element CAE simulation model.

The stratum mechanical inversion parameters are mechanical inversion parameters corresponding to a plurality of stratum within a preset range. The preset range may be a range in which tunnel construction may be affected, a tunnel construction range, or a position range determined according to a set calculation method. The multiple strata can be one stratum, two strata, three strata … n strata and the like, the number of the strata can be determined according to the actual stratum soil conditions of tunnel construction, and the application is not particularly limited.

The inversion parameters of the formation mechanics here include: elastic modulus of soil body, poisson ratio, cohesive force, internal friction angle, elastic modulus of grouting layer, etc. The inversion parameters of the stratum mechanics can be obtained through acquisition equipment, can be obtained according to investigation of staff, and can be obtained according to images of soil bodies. For example, the cohesion and internal friction angle can be measured by a direct fast shear method, a consolidation fast shear method or a triaxial test method, the elastic modulus can be obtained by a static triaxial apparatus, and the poisson ratio can be measured by a triaxial compression method. The acquisition mode of the inversion parameters of the stratum mechanics can be adjusted according to actual requirements, and the method is not particularly limited.

The finite element CAE simulation model can be established for the first time when the mechanical parameters of the tunnel stratum are acquired, and can also be established when the mechanical parameters of the tunnel stratum are acquired each time. The finite element CAE simulation model building method comprises the following steps:

determining a transverse section inverted by stratum mechanical parameters according to the acquired tunnel data, and determining the central position of the shield tunnel and the thickness distribution condition of each soil layer on the transverse section at the periphery of the tunnel; determining the diameter size of the shield tunnel, the thickness of the grouting layer and the outer diameter of the grouting ring according to the actual tunnel construction data (as shown in fig. 2, the diameter size of the shield tunnel shown in fig. 2 is D, the thickness eta of the pipe piece, the thickness delta of the grouting layer and the outer diameter D of the grouting ring); determining the distance between the two side boundaries of the finite element CAE simulation model and the central axis of the tunnel and the distance between the bottom of the model and the bottom of the grouting layer of the tunnel according to the influence range of shield and grouting construction on the stratum (the distance between the two side boundaries of the finite element CAE simulation model and the central axis of the tunnel is 3.5D, and the distance between the bottom of the model and the bottom of the grouting layer of the tunnel is 3D shown in FIG. 2); determining the shape and size conditions of the model according to the distance between the boundaries of the two sides of the finite element CAE simulation model and the central axis of the tunnel and the distance between the bottom of the model and the bottom of the grouting layer of the tunnel, and establishing the CAE simulation model; mesh dissection is carried out on the CAE simulation model; checking the quality of the grid, and modifying the grid at the position with poor quality; determining a material model of each stratum according to a set material selection principle, and defining mechanical inversion parameters of each stratum material model; boundary conditions are applied to the CAE simulation model, displacement constraints in the abscissa direction are set on the boundaries of the left side and the right side of the CAE simulation model, and displacement constraints in the abscissa direction and the ordinate direction are set on the boundary of the bottom of the CAE simulation model, so that the finite element CAE simulation model is created (as shown in FIG. 3, the X direction is the abscissa direction, and the Y direction is the ordinate direction in FIG. 3).

It can be understood that the finite element CAE stratum mechanical inversion parameter-surface subsidence data set comprises a plurality of groups of mechanical inversion parameters and surface subsidence simulation values, and each group comprises the mechanical inversion parameters of each stratum and the surface subsidence simulation values calculated by the finite element CAE simulation under the condition of the parameters. For example, the mechanical inversion parameters of the stratum 1, the mechanical inversion parameters of the stratum 2, the mechanical inversion parameters of the stratum 3 and the mechanical inversion parameters of the stratum …, and the surface subsidence simulation values obtained by the finite element CAE simulation calculation under the condition of the inversion parameters of the n strata 4 n.

And 202, constructing a PSO-BP neural network through parameters in the surface subsidence data set, and performing model training.

The PSO-BP neural network is a neural network obtained by fusing a PSO algorithm on a BP neural network model and optimizing the BP neural network model.

Alternatively, the transfer function of the BP neural network model may use a tan sig function, a purelin function, a log sig function, or the like. The training function of the BP neural network model may use a tranlm function, a tranbfg function, a tranrp function, or the like. The transfer function and the training function of the BP neural network model can be selected according to actual conditions, and the application is not particularly limited.

It will be appreciated that when model training the PSO-BP neural network, a portion of the data set in the surface subsidence data set may be used as a training set, model training the PSO-BP neural network, and model testing the PSO-BP neural network using another portion of the data set in the surface subsidence data set as a test set. For example, 75% of the total set of parameters in the surface subsidence dataset may be selected as the training set and 25% as the test set. 70% of the total set of parameters in the surface subsidence dataset may also be selected as the training set, with 30% being the test set. 80% of the total set of parameters in the surface subsidence dataset may also be selected as the training set, 20% as the test set, and so on. The number of the test groups and the training groups can be adjusted according to the practical conditions of the training effect, and the method is not particularly limited.

And 203, inputting the original surface subsidence data into a trained PSO-BP neural network to obtain mechanical parameters of the tunnel stratum.

The original earth surface subsidence data are subsidence data acquired at preset earth surface subsidence measuring points. The preset earth surface settlement measuring point can take the central axis of the tunnel as a central axis, earth surface axis measuring points are distributed above the central axis of the tunnel, and measuring points are distributed at the positions of the set horizontal intervals on two sides of the earth surface axis measuring points. The set horizontal intervals can be set at equal intervals or unequal intervals, for example, the set horizontal intervals can be 5m, 5m and 5m, 3m, 4m and 5m, 1m, 4m and 6m, and the set horizontal intervals can be adjusted according to practical situations, and the set horizontal intervals are not particularly limited.

In one possible implementation, step 201 includes: determining an orthogonal test table according to the inversion parameters of the stratum mechanics and the corresponding grading level conditions of the stratum parameters; grading the stratum inversion parameters according to the orthogonal test table to obtain a plurality of groups of orthogonal test combination parameters; inputting a plurality of groups of orthogonal test combination parameters into a finite element CAE simulation model to construct an associated finite element CAE stratum mechanical inversion parameter-surface subsidence data set.

Wherein, the multiple groups of orthogonal test combination parameters comprise classified mechanical inversion parameters.

The stratum mechanics inversion parameter grading level condition can be obtained from external equipment, can be directly input, and can be obtained from a memory. The stratigraphic mechanical inversion parameter classification level condition may be a preset classification rule.

It can be appreciated that when determining the orthogonal test table, the formation mechanical inversion parameters can be classified according to the number of layers of the formation and a preset formula, and the classification level and the maximum value and the minimum value of each parameter of each formation mechanical inversion parameter, so as to determine the type of the corresponding orthogonal test table.

After the orthogonal test table is determined, the grading level of each parameter in the inversion parameters of the stratum mechanics can be further determined according to the orthogonal test table, and then a plurality of groups of orthogonal test combination parameters are obtained.

For example, if the predetermined formula of the orthogonal test table is determined to be 4n+1. When the tunnel construction soil layer is a single layer, n=1 can be determined, and the orthogonal test table can be selected from the 5-factor orthogonal test table (L) ₁₆ (4 ⁵ ) Modulus of elasticity E _i Poisson ratio v _i Cohesive force c _i Angle of internal frictionModulus of elasticity E of slip casting layer _g Parameter grading level number m of (2) _E 、m _V 、m _c 、/>m _g 4, the number of orthogonal test combination parameter combinations N is 16, and the orthogonal test combination sequences are shown in table 1:

Table 1:

when the tunnel construction soil layer is 3 layers, n=3 can be determined, and the orthogonal test table can be selected from 13-factor orthogonal test tables (L ₂₇ (3 ¹³ ) Modulus of elasticity E _i Poisson ratio v _i Cohesive force c _i Angle of internal frictionModulus of elasticity E of slip casting layer _g Parameter grading level number m of (2) _E 、m _V 、m _c 、/>m _g 3, and the number of orthogonal test combination parameter combinations N is 27, and the orthogonal test combination sequences are shown in table 2:

table 2:

In one possible implementation, before step 202, the method further includes: and carrying out normalization treatment on the classified mechanical inversion parameters.

The normalization process here is formulated as:

wherein x is _norm Normalized value of mechanical inversion parameter, x is classified mechanical inversion parameter, x _max Maximum value of classified mechanical inversion parameter, x _min The minimum value of the mechanical inversion parameters after grading is obtained.

Step 202, including: and constructing a PSO-BP neural network according to the parameters after normalization processing, and performing model training.

It can be understood that after the original surface subsidence data is obtained, the original surface subsidence data is input into a trained PSO-BP neural network, dimensionless results after normalization of the mechanical parameters of multiple strata are obtained through inversion, and all normalized dimensionless results are converted into corresponding mechanical parameter values through a normalization value conversion and restoration formula.

The formula for converting and restoring the normalized value of the mechanical parameter into the corresponding mechanical parameter value can be as follows:

x ^* ＝x _norm ·(x _max -x _min )+x _min ；

wherein x is _norm Normalized value, x, for mechanical parameter obtained by inversion _max Maximum value of classified mechanical inversion parameter, x _min Is the minimum value of the classified mechanical inversion parameter, x ^* Is used for transforming the mechanical parameter value after reduction.

In some embodiments, the transformed and restored mechanical parameters are input into a finite element CAE simulation model to obtain a ground subsidence simulation value.

In the implementation process, the PSO-BP neural network is constructed and trained based on the mechanical inversion parameters after the normalization processing by carrying out normalization processing on the mechanical inversion parameters after the classification processing, so that the PSO-BP neural network training efficiency and convergence speed can be improved.

In one possible implementation manner, constructing a PSO-BP neural network according to the classified mechanical parameters after normalization processing, and performing model training, including: determining the super parameters of the PSO-BP neural network according to the input parameters of the PSO-BP neural network and the output parameters of the PSO-BP neural network; and constructing the PSO-BP neural network according to the super parameters and a preset network structure.

The output parameters of the PSO-BP neural network are determined according to the classified mechanical parameters after normalization processing, the input parameters of the PSO-BP neural network are determined according to ground surface subsidence simulation values, and the ground surface subsidence simulation values are obtained through calculation of a plurality of groups of orthogonal test combination parameters and a finite element CAE simulation model.

The super-parameters here may include the number of input layer nodes, the number of output layer nodes, the number of neurons of the hidden layer, the coding length of the PSO algorithm, etc. of the PSO-BP neural network.

In some embodiments, determining the super-parameters of the PSO-BP neural network from the input parameters of the PSO-BP neural network and the output parameters of the PSO-BP neural network comprises: determining the number of nodes of an input layer of a PSO-BP neural network according to the input parameters of the PSO-BP neural network, and determining the number of nodes of an output layer of the PSO-BP neural network according to the output parameters of the PSO-BP neural network; and determining the neuron number of the hidden layer and the coding length of a PSO algorithm according to the input layer node number and the output layer node number.

The number of nodes of the output layer of the PSO-BP neural network can be determined according to the number of the classified mechanical parameters after normalization processing. For example, if the stratum is 1 layer, the number of output layer nodes of the PSO-BP neural network is 5 if the output normalized mechanical parameters after the classification are 5. And if the stratum is 3 layers, 13 classified mechanical parameters after the output normalization processing are output, and the number of nodes of the output layer of the PSO-BP neural network is 13. The number of nodes of the input layer of the PSO-BP neural network can be determined according to the number of preset earth surface subsidence measuring points of the finite element CAE simulation model. For example, if the finite element CAE simulation model presets 7 earth surface subsidence measuring points, and the earth surface subsidence simulation values at the corresponding 7 measuring points are obtained through calculation, the number of nodes of the input layer of the PSO-BP neural network is 7.

In some embodiments, the neuron number of the hidden layer is determined by the following formula:

wherein t is the number of output layer nodes, r is the number of input layer nodes, s is the number of neurons of the hidden layer, and a is a constant of 1-10.

The code length of the PSO algorithm is determined by the following formula:

L＝r×s+s+s×t+t；

In one possible implementation, the method further includes: inputting the mechanical parameters of the tunnel stratum into a finite element CAE simulation model to verify the accuracy of the mechanical parameters of the tunnel stratum so as to update the mechanical parameters of the tunnel stratum according to the verification result.

It can be understood that the mechanical parameters of the tunnel stratum acquired through the trained PSO-BP neural network may deviate due to the influence of parameters or operations, in order to ensure the accuracy of the mechanical parameters of the tunnel stratum, after determining the mechanical parameters of the tunnel stratum, the mechanical parameters of the tunnel stratum may be further input into a finite element CAE simulation model for verification, and if the mechanical parameters of the tunnel stratum are verified to be within the accurate range, the process is ended; if the tunnel stratum mechanical parameters are verified to be not in the accurate range, the tunnel stratum mechanical parameters are further updated after the PSO-BP neural network is updated.

In one possible implementation, inputting the tunnel formation mechanical inversion parameter into a finite element CAE simulation model to verify the accuracy of the tunnel formation mechanical inversion parameter to update the tunnel formation mechanical inversion parameter according to the verification result, including: determining a judging function according to the ground surface subsidence simulation value and the original ground surface subsidence data; and verifying the accuracy of the tunnel stratum mechanical inversion parameter according to the judging function so as to update the tunnel stratum mechanical inversion parameter according to the verification result.

The ground surface subsidence simulation value is obtained through calculation of a plurality of groups of orthogonal test combination parameters and a finite element CAE simulation model.

The judging function is an average value of relative errors between the ground surface subsidence simulation value of the finite element CAE simulation model and the original ground surface subsidence data, and is used for reducing fluctuation influence of the relative errors caused by different measuring point numbers.

The formula for determining the judging function according to the ground surface subsidence simulation value and the original ground surface subsidence data is as follows:

wherein, the liquid crystal display device comprises a liquid crystal display device,s is the simulation value of the earth surface subsidence _k For raw surface subsidence data, k=1, 2,3,4,5,6,7, …, n is the preset surface subsidence point number.

It can be understood that by inputting the ground subsidence simulation value and the original ground subsidence data into the above formula, the value of the judging function can be determined, the accuracy of the mechanical parameters of the tunnel stratum can be determined by judging the range of the value of the judging function, and if the accuracy of the mechanical parameters of the tunnel stratum does not meet the requirement, the inversion parameters of the stratum mechanics can be further updated by updating the PSO-BP neural network.

In one possible implementation, as shown in fig. 4, verifying accuracy of the tunnel formation mechanical parameters according to the evaluation function to update the tunnel formation mechanical parameters according to the verification result includes: judging whether the judging function is not larger than a first preset value or not; if the judging function is larger than the first preset value, the mechanical parameters of the tunnel stratum are updated according to the magnitude relation between the iteration times and the second preset value.

The first preset value may be an industry standard error demarcation value, an error critical value determined empirically, or a fixed value calculated according to a specific algorithm. The first preset value can be stored in a memory, can be input each time tunnel formation mechanical parameter calculation is performed, and can be obtained from a third party device or a server. The first preset value may be 0.01, 0.05, 0.07, etc., and the determination, the obtaining manner and the numerical value of the first preset value may be adjusted according to practical situations, which is not specifically limited in this application.

It can be understood that the first preset value is an error critical value, and when it is determined that the evaluation function is not greater than the first preset value, it is indicated that the error value of the simulation result and the actual result obtained by the mechanical parameter inversion analysis is within the specified error value range, and then the mechanical parameter of the tunnel stratum obtained at this time is an accurate mechanical parameter of the tunnel stratum. At this time, the tunnel stratum mechanical parameter determination flow is ended.

When the judging function is determined to be larger than the first preset value, the fact that the error value of the simulation result and the actual result obtained by mechanical parameter inversion analysis exceeds the specified error value range is determined, at the moment, the PSO-BP neural network is required to be updated, and then the mechanical parameters of the tunnel stratum are updated, so that the error value of the mechanical parameters of the tunnel stratum is in the specified error range, and the accuracy of the mechanical parameters of the tunnel stratum is ensured.

In the implementation process, whether the error values of the simulation result and the actual result obtained by the mechanical parameter inversion analysis are within a specified error range is determined according to the magnitude relation between the judging function and the first preset value, if the error values of the simulation result and the actual result obtained by the mechanical parameter inversion analysis exceed the specified error range, the mechanical parameters of the tunnel stratum are updated, the errors of the mechanical parameters of the tunnel stratum are ensured to be within the specified error range, and the accuracy of the mechanical parameters of the tunnel stratum is improved.

In one possible implementation, updating the tunnel formation mechanical parameter according to the magnitude relation between the iteration number and the second preset value includes: judging whether the iteration times are not more than a second preset value; if the iteration times are not greater than the second preset value, updating the iteration times, reconstructing a PSO-BP neural network and performing model training; inputting the original surface subsidence data into a trained PSO-BP neural network to obtain mechanical parameters of the tunnel stratum until the judging function is not more than a first preset value; if the iteration times are greater than a second preset value, updating the stratum mechanics inversion parameter grading level number, and reconstructing an associated finite element CAE stratum mechanics inversion parameter-earth surface subsidence data set according to the updated stratum mechanics inversion parameter and the finite element CAE simulation model; constructing a PSO-BP neural network through parameters in the earth surface subsidence data set, and performing model training; inputting the original surface subsidence data into the trained PSO-BP neural network to obtain the mechanical parameters of the tunnel stratum until the judging function is not more than a first preset value.

The iteration number here is the iteration number of the PSO-BP neural network. When the PSO-BP neural network is constructed for the first time, a maximum iteration number can be determined, wherein the maximum iteration number can be 20, 30, 50 and the like, the maximum iteration number can be determined according to actual conditions, and the application is not particularly limited.

It can be understood that the second preset value is an optimization condition range, and when it is determined that the iteration number is not greater than the second preset value, and the optimization condition is satisfied, the current iteration number and the maximum iteration number are updated at this time, and the PSO-BP neural network is reconstructed and model training is performed. And inputting the original earth surface subsidence data into the retrained PSO-BP neural network to obtain the mechanical parameters of the tunnel stratum and verifying the accuracy of the mechanical parameters of the tunnel stratum again until the judging function is not more than a first preset value.

And when the iteration times are determined to be larger than a second preset value, and the optimization condition cannot be met, updating the hierarchical level number of the formation mechanical inversion parameter, and reconstructing an associated finite element CAE formation mechanical inversion parameter-surface subsidence data set according to the updated formation mechanical inversion parameter and the finite element CAE simulation model. And then constructing a PSO-BP neural network through the parameters in the surface subsidence data set again and performing model training. And inputting the original earth surface subsidence data into the retrained PSO-BP neural network to obtain the mechanical parameters of the tunnel stratum and verifying the accuracy of the mechanical parameters of the tunnel stratum again until the judging function is not more than a first preset value.

The stratum mechanics inversion parameter grading level number m _E 、m _V 、m _c 、m _g Can be updated to (m) _E +1、m _v +1、m _c +1、/>m _g +1)、(m _E +2、m _v +2、m _c +2、/>m _g +2)、(m _E +3、m _v +3、m _c +3、/>m _g +3) … (the number of mechanical inversion parameter grading levels is shown as being updated sequentially to (m in FIG. 4 _E +1、m _v +1、m _c +1、/>m _g +1)). The number of iterations may be updated to s=s+1, s=s+2, s=s+3 … (the number of iterations is shown updated to s=s+1 in fig. 4), and the maximum number of iterations may be updated to T _max ＝T _max +30、T _max ＝T _max +50、T _max ＝T _max +60 … (the maximum number of iterations is shown updated to T in FIG. 4) _max ＝T _max +50)。

In the implementation process, when the judging function is larger than the first preset value, the magnitude relation between the iteration times and the second preset value is further judged, whether the iteration times meet the optimization conditions is further determined, when the iteration times meet the optimization conditions, the formation mechanical inversion parameter grading level number is updated, the associated finite element CAE formation mechanical inversion parameter-earth surface subsidence data set is reconstructed, and then the PSO-BP neural network is reconstructed and model training is performed. And when the iteration times do not meet the optimization conditions, reconstructing the PSO-BP neural network directly according to the updated current iteration times and the maximum iteration times, and performing model training. And inputting the original earth surface subsidence data into the retrained PSO-BP neural network to obtain the mechanical parameters of the tunnel stratum and verifying the accuracy of the mechanical parameters of the tunnel stratum again until the judging function is not more than a first preset value. Through the judgment and circulation, the PSO-BP neural network is adjusted and optimized, so that the error of the mechanical parameters of the tunnel stratum is gradually reduced, the performance of the PSO-BP neural network is improved, and the accuracy of the mechanical parameters of the tunnel stratum is improved.

In one possible implementation, the formation mechanical inversion parameters include: modulus of elasticity of the slip layer.

The grouting layer is a soil and mortar mixed stratum formed around the tunnel during synchronous grouting of the shield tunnel. The grouting layer is used for filling gaps between shield segments and surrounding soil bodies and reducing earth surface subsidence.

In the implementation process, the elastic modulus of the supplementary grouting layer is added in the stratum mechanical inversion parameters, the influence of the grouting layer on stratum settlement is considered, the accuracy of stratum mechanical parameter inversion is further improved, and the accuracy of tunnel stratum mechanical parameters is further improved.

Based on the same application conception, the embodiment of the application also provides a tunnel stratum mechanical parameter obtaining device corresponding to the tunnel stratum mechanical parameter obtaining method, and because the principle of solving the problem of the device in the embodiment of the application is similar to that of the embodiment of the tunnel stratum mechanical parameter obtaining method, the implementation of the device in the embodiment of the application can be referred to the description in the embodiment of the method, and the repetition is omitted.

Fig. 5 is a schematic functional block diagram of a tunnel formation mechanical parameter obtaining device according to an embodiment of the present application. The modules in the tunnel formation mechanical parameter obtaining device in this embodiment are configured to execute the steps in the method embodiment. The tunnel stratum mechanical parameter acquisition device comprises a first construction module 301, a second construction module 302 and a calculation module 303; wherein, the liquid crystal display device comprises a liquid crystal display device,

The first construction module 301 is configured to construct an associated finite element CAE formation mechanical inversion parameter-surface subsidence data set according to a formation mechanical inversion parameter and a finite element CAE simulation model, where the formation mechanical inversion parameter is a mechanical parameter corresponding to a plurality of formations within a preset range.

The second construction module 302 is configured to construct a PSO-BP neural network through parameters in the surface subsidence data set, and perform model training.

The calculation module 303 is configured to input the raw surface subsidence data into the trained PSO-BP neural network to obtain mechanical parameters of the tunnel formation.

In a possible implementation manner, the first construction module 301 is further configured to: determining an orthogonal test table according to the stratum mechanics inversion parameters and the corresponding stratum mechanics inversion parameter grading level conditions; grading the stratum mechanics inversion parameters according to the orthogonal test table to obtain a plurality of groups of orthogonal test combination parameters; inputting the multiple groups of orthogonal test combination parameters into the finite element CAE simulation model to construct an associated finite element CAE stratum mechanical inversion parameter-surface subsidence data set.

In a possible implementation manner, the tunnel formation mechanical parameter obtaining device further comprises a normalizing device, which is used for normalizing the classified formation mechanical inversion parameters.

In a possible implementation manner, the second construction module 302 is further configured to: and constructing a PSO-BP neural network according to the classified mechanical parameters after normalization treatment, and performing model training.

In a possible implementation manner, the second construction module 302 is specifically configured to: determining the super parameters of the PSO-BP neural network according to the input parameters of the PSO-BP neural network and the output parameters of the PSO-BP neural network; constructing the PSO-BP neural network according to the super parameters and a preset network structure; the output parameters of the PSO-BP neural network are determined according to the classified mechanical parameters after normalization processing, the input parameters of the PSO-BP neural network are determined according to ground surface subsidence simulation values, and the ground surface subsidence simulation values are obtained through calculation of the multiple groups of orthogonal test combination parameters and the finite element CAE simulation model.

In a possible implementation manner, the tunnel stratum mechanical parameter obtaining device further comprises a verifying device, which is used for: inputting the tunnel stratum mechanical parameters into the finite element CAE simulation model to verify the accuracy of the tunnel stratum mechanical parameters so as to update the stratum mechanical inversion parameters according to a verification result.

In a possible embodiment, the verification device is further configured to: determining a judging function according to the ground surface subsidence simulation value and the original ground surface subsidence data; verifying the accuracy of the tunnel stratum mechanical parameters according to the evaluation function so as to update the tunnel stratum mechanical parameters according to a verification result; the ground surface subsidence simulation value is obtained through calculation of the plurality of groups of orthogonal test combination parameters and the finite element CAE simulation model.

In a possible embodiment, the verification device is specifically configured to: judging whether the evaluation function is not larger than a first preset value or not; and if the evaluation function is larger than the first preset value, updating the tunnel stratum mechanical parameter according to the magnitude relation between the iteration times and the second preset value.

In a possible embodiment, the verification device is specifically configured to: judging whether the iteration times are not larger than a second preset value or not; if the iteration times are not greater than the second preset value, updating the iteration times, reconstructing a PSO-BP neural network and performing model training; inputting the original surface subsidence data into the PSO-BP neural network after training to obtain tunnel stratum mechanical parameters until the evaluation function is not more than the first preset value; if the iteration times are larger than the second preset value, updating the hierarchical level number of the formation mechanical inversion parameter, and reconstructing an associated finite element CAE formation mechanical inversion parameter-surface subsidence data set according to the updated formation mechanical inversion parameter and the finite element CAE simulation model; constructing a PSO-BP neural network through parameters in the earth surface subsidence data set, and performing model training; inputting the original surface subsidence data into the PSO-BP neural network after training to obtain the mechanical parameters of the tunnel stratum until the evaluation function is not more than the first preset value.

For the sake of understanding the present embodiment, the following describes in detail an electronic device that executes the method for obtaining the mechanical parameters of the tunnel formation disclosed in the embodiments of the present application.

As shown in fig. 6, a block schematic diagram of the electronic device is shown. The electronic device 100 may include a memory 111, a processor 113, and an input-output unit 115. It will be appreciated by those of ordinary skill in the art that the configuration shown in fig. 6 is merely illustrative and is not limiting of the configuration of the electronic device 100. For example, the electronic device 100 may also include more or fewer components than shown in fig. 6, or have a different configuration than shown in fig. 6.

The above-mentioned memory 111, processor 113 and input/output unit 115 are electrically connected to each other directly or indirectly to realize data transmission or interaction. For example, the components may be electrically connected to each other via one or more communication buses or signal lines. The processor 113 is used to execute executable modules stored in the memory.

The Memory 111 may be, but is not limited to, a random access Memory (Random Access Memory, RAM), a Read Only Memory (ROM), a programmable Read Only Memory (Programmable Read-Only Memory, PROM), an erasable Read Only Memory (Erasable Programmable Read-Only Memory, EPROM), an electrically erasable Read Only Memory (Electric Erasable Programmable Read-Only Memory, EEPROM), etc. The memory 111 is configured to store a program, and the processor 113 executes the program after receiving an execution instruction, and a method executed by the electronic device 100 defined by the process disclosed in any embodiment of the present application may be applied to the processor 113 or implemented by the processor 113.

The processor 113 may be an integrated circuit chip having signal processing capabilities. The processor 113 may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU for short), a network processor (Network Processor, NP for short), etc.; but also digital signal processors (digital signal processor, DSP for short), application specific integrated circuits (Application Specific Integrated Circuit, ASIC for short), field Programmable Gate Arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The disclosed methods, steps, and logic blocks in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The input-output unit 115 described above is used to provide input data to a user. The input/output unit 115 may be, but is not limited to, a mouse, a keyboard, and the like.

The electronic device 100 in this embodiment may be used to perform each step in each method provided in the embodiments of the present application.

In addition, the embodiment of the application further provides a computer readable storage medium, and a computer program is stored on the computer readable storage medium, and when the computer program is executed by a processor, the steps of the tunnel soil body parameter obtaining method in the embodiment of the method are executed.

The computer program product of the method for obtaining the mechanical parameters of the tunnel formation provided in the embodiments of the present application includes a computer readable storage medium storing program codes, where the instructions included in the program codes may be used to execute the steps of the method for obtaining the mechanical parameters of the tunnel formation described in the embodiments of the method, and the detailed description of the embodiments of the method may be omitted herein.

In the several embodiments provided in this application, it should be understood that the disclosed apparatus and method may be implemented in other manners as well. The apparatus embodiments described above are merely illustrative, for example, flow diagrams and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

In addition, the functional modules in the embodiments of the present application may be integrated together to form a single part, or each module may exist alone, or two or more modules may be integrated to form a single part.

The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes. It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.

The foregoing description is only of the preferred embodiments of the present application and is not intended to limit the same, but rather, various modifications and variations may be made by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application. It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures.

The foregoing is merely specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily think about changes or substitutions within the technical scope of the present application, and the changes and substitutions are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. The method for acquiring the mechanical parameters of the tunnel stratum is characterized by comprising the following steps of:

constructing an associated finite element CAE stratum mechanical inversion parameter-earth surface subsidence data set according to the stratum mechanical inversion parameter and the finite element CAE simulation model, wherein the stratum mechanical inversion parameter is a mechanical parameter corresponding to a plurality of stratum in a preset range;

Constructing a PSO-BP neural network through parameters in the earth surface subsidence data set, and performing model training;

inputting the original earth surface subsidence data into the PSO-BP neural network after training to obtain mechanical parameters of the tunnel stratum;

the original earth surface subsidence data are subsidence data acquired at preset earth surface subsidence measuring points;

the constructing the related finite element CAE stratum mechanical inversion parameter-surface subsidence data set according to the stratum mechanical inversion parameter and the finite element CAE simulation model comprises the following steps:

determining an orthogonal test table according to the stratum mechanics inversion parameters and the corresponding stratum mechanics inversion parameter grading level conditions;

grading the stratum mechanics inversion parameters according to the orthogonal test table to obtain a plurality of groups of orthogonal test combination parameters;

inputting the multiple groups of orthogonal test combination parameters into the finite element CAE simulation model to construct an associated finite element CAE stratum mechanical inversion parameter-surface subsidence data set;

the multiple sets of orthogonal test combination parameters comprise classified mechanical parameters, and before the PSO-BP neural network is constructed through the parameters in the earth surface subsidence data set and model training is carried out, the method further comprises:

Normalizing the classified stratum mechanics inversion parameters;

the PSO-BP neural network is constructed through parameters in the earth surface subsidence data set, model training is carried out, and the method comprises the following steps:

constructing a PSO-BP neural network according to the classified mechanical parameters after normalization treatment, and performing model training;

the PSO-BP neural network is constructed according to the classified mechanical parameters after normalization processing, and the PSO-BP neural network comprises:

determining the super parameters of the PSO-BP neural network according to the input parameters of the PSO-BP neural network and the output parameters of the PSO-BP neural network;

constructing the PSO-BP neural network according to the super parameters and a preset network structure;

the output parameters of the PSO-BP neural network are determined according to the classified mechanical parameters after normalization processing, the input parameters of the PSO-BP neural network are determined according to ground surface subsidence simulation values, and the ground surface subsidence simulation values are obtained through calculation of the multiple groups of orthogonal test combination parameters and the finite element CAE simulation model.

2. The method according to claim 1, wherein the method further comprises:

inputting the tunnel stratum mechanical parameters into the finite element CAE simulation model to verify the accuracy of the tunnel stratum mechanical parameters so as to update the stratum mechanical inversion parameters according to a verification result.

3. The method of claim 2, wherein the inputting the tunnel formation mechanical parameters into the finite element CAE simulation model verifies accuracy of the tunnel formation mechanical parameters to update the tunnel formation mechanical parameters based on the verification result comprises:

determining a judging function according to the ground surface subsidence simulation value and the original ground surface subsidence data;

verifying the accuracy of the tunnel stratum mechanical parameters according to the evaluation function so as to update the tunnel stratum mechanical parameters according to a verification result;

the ground surface subsidence simulation value is obtained through calculation of the plurality of groups of orthogonal test combination parameters and the finite element CAE simulation model.

4. A method according to claim 3, wherein verifying the accuracy of the tunnel formation mechanical parameters according to the evaluation function to update the tunnel formation mechanical parameters according to the verification result comprises:

judging whether the evaluation function is not larger than a first preset value or not;

and if the evaluation function is larger than the first preset value, updating the tunnel stratum mechanical parameter according to the magnitude relation between the iteration times and the second preset value.

5. The method of claim 4, wherein updating the tunnel formation mechanical parameter according to the magnitude relationship of the number of iterations and the second preset value comprises:

Judging whether the iteration times are not larger than a second preset value or not;

if the iteration times are not greater than the second preset value, updating the iteration times, reconstructing a PSO-BP neural network and performing model training; inputting the original surface subsidence data into the PSO-BP neural network after training to obtain tunnel stratum mechanical parameters until the evaluation function is not more than the first preset value;

if the iteration times are larger than the second preset value, updating the hierarchical level number of the formation mechanical inversion parameter, and reconstructing an associated finite element CAE formation mechanical inversion parameter-surface subsidence data set according to the updated formation mechanical inversion parameter and the finite element CAE simulation model; constructing a PSO-BP neural network through parameters in the earth surface subsidence data set, and performing model training; inputting the original surface subsidence data into the PSO-BP neural network after training to obtain the mechanical parameters of the tunnel stratum until the evaluation function is not more than the first preset value.

6. The method of claim 1, wherein the formation mechanical inversion parameters comprise: modulus of elasticity of the slip layer.

7. A tunnel formation mechanical parameter acquisition device, comprising:

The first construction module is used for constructing an associated finite element CAE stratum mechanical inversion parameter-earth surface subsidence data set according to the stratum mechanical inversion parameter and the finite element CAE simulation model, wherein the stratum mechanical inversion parameter is a mechanical parameter corresponding to a plurality of strata in a preset range;

the second construction module is used for constructing a PSO-BP neural network through parameters in the earth surface subsidence data set and performing model training;

the calculation module is used for inputting the original earth surface subsidence data into the PSO-BP neural network after training to obtain mechanical parameters of the tunnel stratum, wherein the original earth surface subsidence data are subsidence data obtained at preset earth surface subsidence measuring points;

the first construction module is further configured to: determining an orthogonal test table according to the stratum mechanics inversion parameters and the corresponding stratum mechanics inversion parameter grading level conditions; grading the stratum mechanics inversion parameters according to the orthogonal test table to obtain a plurality of groups of orthogonal test combination parameters; inputting the multiple groups of orthogonal test combination parameters into the finite element CAE simulation model to construct an associated finite element CAE stratum mechanical inversion parameter-surface subsidence data set;

the normalization device is used for normalizing the classified stratum mechanics inversion parameters;

The second building module is further configured to: constructing a PSO-BP neural network according to the classified mechanical parameters after normalization treatment, and performing model training;

the second construction module is specifically configured to: determining the super parameters of the PSO-BP neural network according to the input parameters of the PSO-BP neural network and the output parameters of the PSO-BP neural network; constructing the PSO-BP neural network according to the super parameters and a preset network structure; the output parameters of the PSO-BP neural network are determined according to the classified mechanical parameters after normalization processing, the input parameters of the PSO-BP neural network are determined according to ground surface subsidence simulation values, and the ground surface subsidence simulation values are obtained through calculation of the multiple groups of orthogonal test combination parameters and the finite element CAE simulation model.

8. An electronic device, comprising: a processor, a memory storing machine-readable instructions executable by the processor, which when executed by the processor perform the steps of the method of any of claims 1 to 6 when the electronic device is run.

9. A computer-readable storage medium, characterized in that it has stored thereon a computer program which, when executed by a processor, performs the steps of the method according to any of claims 1 to 6.