Double-line shield tunnel induced ground surface settlement curve intelligent prediction method and product

Abstract

The invention provides a double-line shield tunnel induced ground surface settlement curve intelligent prediction method and product, and relates to the technical field of shield tunnel engineering and ground surface settlement prediction. In the embodiment of the invention, the VL-GP (volume loss and grouting pressure combination) mixed double-line shield tunnel two-dimensional simplified numerical model considering the construction process is established, and the calculation cost is greatly reduced and the simulation precision and efficiency are improved by adopting the two-dimensional simulation method while the influence of the shield tunnel construction process is considered; ground surface settlement curve data are obtained through simulation based on the two-dimensional simplified numerical model, control parameters of a ground surface settlement curve are obtained through empirical formula fitting, and an efficient database is established. An AutoML method is adopted to establish an intelligent prediction model of ground surface settlement curve control parameters, the workload of manual modeling and feature extraction is greatly reduced, and the modeling efficiency and prediction precision are improved.

Description

TECHNICAL FIELD

[0001] The embodiment of the present application relates to the technical field of shield tunnel engineering and ground surface settlement prediction, and particularly relates to a double-line shield tunnel induced ground surface settlement curve intelligent prediction method and product. BACKGROUND

[0002] With the continuous acceleration of urbanization, the development and utilization of underground space is increasingly widespread. Shield tunnel construction has become the preferred method in urban tunnel construction due to its high degree of mechanization, fast tunneling speed, good adaptability to geological conditions, small environmental impact, and high safety. In particular, double-line shield tunnel construction is more common in urban areas with complex geological conditions and dense ground buildings. Although shield tunneling technology has been greatly developed, the shield tunneling process inevitably causes disturbance to the surrounding strata, leading to ground settlement problems. If the settlement is too large or the settlement difference is obvious, it is easy to cause engineering accidents such as cracking of ground buildings, damage to roads, and damage to pipelines, which seriously threatens urban safety operations. Therefore, accurately predicting the settlement curve caused by double-line shield tunnel construction is of great significance for construction risk control, design optimization, and protection of the surrounding environment. Compared with single-line shield tunnels, double-line tunnel construction involves two tunneling processes, and the soil disturbance and stress release are more complex, with mutual influence and superposition effects, making the settlement curve have more significant asymmetric and stage characteristics. If the interaction between double-line tunneling is ignored and the single-line tunnel settlement prediction method is directly applied to double-line situations, it is easy to cause prediction errors, which further affects construction safety and cost control.

[0003] Therefore, there is an urgent need to develop an efficient, accurate, and adaptive settlement curve prediction method for double-line shield tunnel characteristics. SUMMARY

[0004] The embodiment of the present application provides a double-line shield tunnel induced ground surface settlement curve intelligent prediction method and product, which realizes intelligent prediction of double-line shield tunnel induced ground surface settlement curve by combining FLAC3D numerical simulation, automatic machine learning (AutoML) and Python programming, and aims to solve the problems of insufficient prediction accuracy, low efficiency and lack of intelligence in the prior art.

[0005] The first aspect of the embodiment of the present application provides a double-line shield tunnel induced ground surface settlement curve intelligent prediction method, which comprises: A VL-GP hybrid double-line shield tunnel two-dimensional simplified numerical model considering the construction process is established, and based on stratum mechanical parameters, tunnel geometric parameters and shield construction parameters, numerical simulation is performed through the VL-GP hybrid double-line shield tunnel two-dimensional simplified numerical model to generate ground surface settlement curve data; The ground surface settlement curve data are fitted by using a double-line tunnel ground surface settlement curve empirical formula to extract control parameters of the ground surface settlement curve, and a database containing stratum mechanical parameters, tunnel geometric parameters, shield construction parameters and ground surface settlement curve control parameters is established; An automatic machine learning method is used for training based on the database, in which the input is stratum mechanical parameters, tunnel geometric parameters and shield construction parameters, and the output is settlement curve control parameters, and an intelligent prediction model of ground surface settlement curve control parameters is established; The stratum mechanical parameters, tunnel geometric parameters and shield construction parameters of the target working condition are input into the intelligent prediction model to predict the target ground surface settlement curve control parameters, and the target double-line tunnel ground surface settlement curve is obtained according to the double-line tunnel ground surface settlement curve empirical formula and the target ground surface settlement curve control parameters.

[0006] Optionally, a VL-GP mixed double-line shield tunnel two-dimensional simplified numerical model considering the construction process is established, including: The shield tunnel excavation process is divided into three construction stages: shield machine passing stage, shield tail closing and grouting stage and grouting hardening stage, and the three construction stages are sequentially realized by using two-dimensional simulation technology to construct the VL-GP mixed double-line shield tunnel two-dimensional simplified numerical model considering the construction process.

[0007] Optionally, the shield tunnel excavation process is divided into three construction stages: shield machine passing stage, shield tail closing and grouting stage and grouting hardening stage, and the three construction stages are sequentially realized by using two-dimensional simulation technology, including: A complete numerical model is established according to the working condition requirements, a plane strain boundary condition is set, and an initial ground stress field is generated under K0 condition; In the shield machine passing stage, the soil in the excavation part is deactivated, displacement control technology is used to make the tunnel boundary converge inward non-uniformly, and the convergence speed decreases linearly with depth to reflect the geometric relationship between the shield machine shell slightly smaller than the excavation diameter and the excavation boundary under the action of gravity, and the calculation is stopped when the volume loss rate reaches a predetermined value; In the shield tail closing and grouting stage, the liner element simulated lining is activated, grouting pressure is applied on the tunnel excavation boundary, the grouting pressure increases linearly with depth, the physical gap between the lining and the excavation boundary is filled with fresh grouting, and the fresh grouting is simulated by using elastic element and is given physical and mechanical parameters of fresh grouting; In the grouting hardening stage, the grouting pressure is removed, and hardened grouting is used to replace fresh grouting, and the grouting body is given physical and mechanical parameters of hardened grouting.

[0008] Optionally, the stratum mechanical parameters include soil layer specific weight γ, compression modulus E s, cohesion c, internal friction angle φ; tunnel geometric parameters include: tunnel diameter D, tunnel depth ratio C / D, double line tunnel center distance ratio B / D and double line tunnel deflection angle θ; shield construction parameters include the excavation volume loss rate and grouting pressure of the two tunnels respectively.

[0009] Optionally, a database containing stratum mechanical parameters, tunnel geometric parameters, shield construction parameters and ground settlement curve control parameters is established, including: Parameterizing control is performed on the stratum mechanical parameters, tunnel geometric parameters and shield construction parameters, The VL-GP mixed double line shield tunnel two-dimensional simplified numerical model is adjusted based on different stratum mechanical parameters, tunnel geometric parameters and shield construction parameters, so as to obtain ground settlement curve data corresponding to each of the different stratum mechanical parameters, tunnel geometric parameters and shield construction parameters; The ground settlement curve data is fitted by using a double line tunnel ground settlement curve empirical formula, and control parameters of the ground settlement curve corresponding to each of the different stratum mechanical parameters, tunnel geometric parameters and shield construction parameters are extracted, so as to establish a database containing stratum mechanical parameters, tunnel geometric parameters, shield construction parameters and ground settlement curve control parameters.

[0010] Optionally, the ground settlement curve data is fitted by using a double line tunnel ground settlement curve empirical formula, including: The numerical simulation generated ground settlement curve is fitted by using a double line tunnel ground settlement curve empirical formula based on the superposition principle, and the empirical formula is as follows:

[0011] wherein, x is the horizontal distance from the middle line of the double line tunnel; S max，1 and S max，2 are the maximum ground settlements caused by the preceding and following tunnels respectively; i 1 and i 2 are the distances from the inflection points of the preceding and following tunnel settlement troughs to the tunnel center line respectively, d is half of the horizontal center distance of the double line tunnel; through fitting, the control parameters of the ground settlement curve S max，1 , S max，2 , i 1 and i 2.

[0012] The second aspect of the embodiment of the application provides a double line shield tunnel induced ground settlement curve intelligent prediction device, the device comprising: The numerical model establishing module is configured to establish a two-dimensional simplified numerical model of a VL-GP mixed double-line shield tunnel considering a construction process, generate surface subsidence curve data through numerical simulation of the two-dimensional simplified numerical model of the VL-GP mixed double-line shield tunnel based on stratum mechanical parameters, tunnel geometric parameters and shield construction parameters, and the like. The fitting module is configured to fit the surface subsidence curve data by using an empirical formula of a double-line tunnel surface subsidence curve, extract control parameters of the surface subsidence curve, and establish a database containing stratum mechanical parameters, tunnel geometric parameters, shield construction parameters and surface subsidence curve control parameters. The training module is configured to train based on the database by using an automatic machine learning method, in which the input is stratum mechanical parameters, tunnel geometric parameters and shield construction parameters, the output is subsidence curve control parameters, and an intelligent prediction model of the surface subsidence curve control parameters is established. The prediction module is configured to input stratum mechanical parameters, tunnel geometric parameters and shield construction parameters of a target working condition into the intelligent prediction model, predict target surface subsidence curve control parameters, and obtain a target double-line tunnel surface subsidence curve based on the empirical formula of the double-line tunnel surface subsidence curve and the target surface subsidence curve control parameters.

[0013] The third aspect of the embodiment of the present application provides an electronic device, which includes a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the intelligent prediction method of the double-line shield tunnel induced surface subsidence curve when executed.

[0014] The fourth aspect of the embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the program implements the intelligent prediction method of the double-line shield tunnel induced surface subsidence curve when executed by a processor.

[0015] The fifth aspect of the embodiment of the present application provides a computer program product, which includes a computer program / instruction, and the computer program / instruction implements the steps of the intelligent prediction method of the double-line shield tunnel induced surface subsidence curve when executed by a processor.

[0016] In the embodiment of the present application, a VL-GP (volume loss and grouting pressure combined) mixed double-line shield tunnel two-dimensional simplified numerical model considering the construction process is established. While considering the influence of the shield tunnel construction process, the two-dimensional simulation method greatly reduces the calculation cost, improves the simulation accuracy and efficiency; based on the two-dimensional simplified numerical model, the ground surface settlement curve data is simulated, the control parameters of the ground surface settlement curve are obtained through the empirical formula fitting, and an efficient database is established. The AutoML method is used to establish an intelligent prediction model of the ground surface settlement curve control parameters, which greatly reduces the workload of manual modeling and feature extraction, improves the modeling efficiency and prediction accuracy.

[0017] Therefore, in the embodiment of the present application, by combining FLAC3D numerical simulation and AutoML method, the double-line shield tunnel ground surface settlement curve can be more accurately predicted; and the AutoML method greatly reduces the workload of manual modeling and feature extraction, improves the prediction efficiency; and in the embodiment of the present application, the intelligent prediction model realizes prediction based on stratum mechanical parameters, tunnel geometric parameters and shield construction parameters, which can adapt to different geological conditions without manual intervention; the technical scheme provided in the embodiment of the present application can be applied to double-line shield tunnel construction under complex geological conditions, and has wide engineering application value.

[0018] In summary, the embodiment of the present application combines FLAC3D numerical simulation, automatic machine learning (AutoML) and Python programming to realize intelligent prediction of double-line shield tunnel induced ground surface settlement curve, which not only improves construction safety and economy, but also has important engineering significance and social value for enriching shield tunnel construction settlement control theory and promoting sustainable development of underground space. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the description of the embodiments of the present application. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0020] Figure 1 is a step flow chart of the double-line shield tunnel induced ground surface settlement curve intelligent prediction method provided by the embodiment of the present application; Figure 2 is a two-dimensional simplified simulation process schematic diagram of the double-line shield tunnel induced ground surface settlement curve intelligent prediction method provided by the embodiment of the present application; Figure 3It is an example fitting schematic diagram of a double-line shield tunnel ground settlement curve empirical formula based on superposition technology in the double-line shield tunnel induced ground settlement curve intelligent prediction method provided by the embodiment of the application. Figure 4 It is an Auto-Sklearn architecture diagram in the double-line shield tunnel induced ground settlement curve intelligent prediction method provided by the embodiment of the application. Figure 5 It is a double-line shield tunnel ground settlement curve prediction flowchart in the double-line shield tunnel induced ground settlement curve intelligent prediction method provided by the embodiment of the application. Figure 6 It is a hardware structure diagram of any device with data processing capability where the double-line shield tunnel induced ground settlement curve intelligent prediction device provided by the embodiment of the application is located. DETAILED DESCRIPTION

[0021] In order to make the above-mentioned purposes, features and advantages of the application more obvious and easy to understand, the application will be further described in detail below with reference to the drawings and specific embodiments.

[0022] At present, the commonly used settlement prediction methods include empirical formula method, numerical simulation method and machine learning method. The empirical formula method is simple and intuitive, and is widely used in the preliminary design stage of double-line shield tunnels. However, the simple empirical formula method has poor adaptability to complex geological conditions and complex working conditions; the numerical simulation method can comprehensively analyze various interactions occurring in the construction process of shield tunnels, such as soil-shield interaction, soil-lining interaction and construction load action, while considering the nonlinear mechanical behavior of soil, but the three-dimensional numerical simulation modeling process is complex and the calculation cost is high, and the model parameters are highly dependent, which makes it difficult to provide timely and effective guidance for shield tunnel construction; the machine learning method shows good potential in dealing with complex nonlinear relationships, but it has high requirements for data quantity and quality, but the data used for machine learning training in most studies comes from literature or field tests, and the data quantity is small and the quality is difficult to guarantee, in addition, the traditional machine learning method needs manual design of features and models, and the modeling process is complex and inefficient.

[0023] Therefore, the embodiment of the application proposes to combine the advantages of numerical simulation, empirical formula and machine learning method, develop a robust and reliable data set generation method and machine learning modeling technology to realize accurate prediction of double-line shield tunnel induced ground settlement curve.

[0024] Specifically, as shown in Figure 1 It shows the step flowchart of the double-line shield tunnel induced ground settlement curve intelligent prediction method provided by the embodiment of the application, which comprises the following steps: S101, a VL-GP mixed double-line shield tunnel two-dimensional simplified numerical model considering the construction process is established, and based on the stratum mechanics parameters, tunnel geometric parameters and shield construction parameters, numerical simulation is carried out through the VL-GP mixed double-line shield tunnel two-dimensional simplified numerical model to generate the ground surface settlement curve data.

[0025] In the embodiment of the application, in order to obtain a database with sufficient data to complete the training of the intelligent prediction model, a large number of numerical simulation processes need to be performed to obtain sufficient data pairs (stratum mechanics parameters, tunnel geometric parameters, shield construction parameters and ground surface settlement curve control parameters).

[0026] However, the current three-dimensional numerical simulation modeling process is relatively complex, time-consuming and computationally intensive, and the calculation cost of multiple simulations using three-dimensional data simulation is huge. Based on this, the embodiment of the application proposes to establish a VL-GP (volume loss and grouting pressure joint) mixed double-line shield tunnel two-dimensional simplified numerical model considering the construction process.

[0027] Specifically, the VL-GP mixed double-line shield tunnel two-dimensional simplified numerical model does not aim to truly reflect the three-dimensional space-time effect of the double-line shield tunnel, but rather to exhibit the three-dimensional construction effect from a two-dimensional perspective. It is equivalent to cutting a cross section of the double-line shield tunnel and sequentially simulating the tunnel excavation, cross section convergence, grouting and segment support on this cross section. In the two-dimensional simulation process, a cross section is positioned and then the construction steps are simulated in sequence to complete the numerical simulation.

[0028] Therefore, the embodiment of the application can exhibit the three-dimensional construction process through two-dimensional simplified numerical simulation, and ultimately complete the real simulation of the entire tunnel construction process to obtain the ground surface settlement data.

[0029] Specifically, in the embodiment of the application, the VL-GP mixed double-line shield tunnel two-dimensional simplified numerical model considering the construction process is established, including: The shield tunnel excavation process is divided into three construction stages: shield machine passing stage, shield tail closure and grouting stage and grouting hardening stage, and the two-dimensional simulation technology is used to sequentially realize the three construction stages to construct the VL-GP mixed double-line shield tunnel two-dimensional simplified numerical model considering the construction process.

[0030] In the embodiment of the application, the VL-GP mixed double-line shield tunnel two-dimensional simplified numerical model considering the construction process considers the entire construction process, and the accuracy is improved. Therefore, the VL-GP mixed double-line shield tunnel two-dimensional simplified numerical model considering the construction process can improve the accuracy on the one hand, and can improve the efficiency compared to three-dimensional simulation on the other hand, achieving both near three-dimensional simulation accuracy and providing calculation efficiency.

[0031] In this embodiment of the invention, the steps for establishing a simplified two-dimensional numerical model of a VL-GP hybrid double-track shield tunnel that takes into account the construction process specifically include: A complete numerical model is established according to the working conditions, plane strain boundary conditions are set, and the initial geostress field is generated under K0 conditions.

[0032] During the tunnel boring machine (TBM) passage phase, the soil in the excavated section is deactivated, and displacement control technology is used to make the tunnel boundary converge inward non-uniformly. The convergence speed decreases linearly with depth to reflect the geometric relationship between the TBM shell, which is slightly smaller than the excavation diameter, and the excavation boundary under gravity. The calculation stops when the volume loss rate reaches a predetermined value.

[0033] During the tail shield closure and grouting stage, the lining simulated by Liner elements is activated, and grouting pressure is applied to the tunnel excavation boundary. The grouting pressure increases linearly with depth. The physical gap between the lining and the excavation boundary is filled with fresh grout. Elastic elements are used to simulate and assign physical and mechanical parameters to the fresh grout. During the grouting hardening stage, the grouting pressure is removed, and hardened grouting is used instead of fresh grouting to impart the physical and mechanical parameters of hardened grouting to the grouting body.

[0034] Specifically, such as Figure 2 The diagram illustrates a simplified two-dimensional simulation process for a VL-GP hybrid double-track shield tunnel considering the construction process in an embodiment of the present invention. Stage 0 refers to determining the excavation boundary, tunnel boundary, and physical gaps according to the working conditions, establishing a complete numerical model, setting plane strain boundary conditions, and generating the initial geostress field under K0 conditions. Stage 1 refers to simulating volume loss during the shield machine's passage. Stage 2 refers to simulating the filling of fresh grout into the lining, grouting pressure, and physical gaps between the lining and excavation boundary during the shield tail closure and grouting stages. Stage 3 refers to simulating the replacement of fresh grout with hardened grout during the grout hardening stage, assigning the hardened grout's physical and mechanical parameters to the grout body. After the final simulation calculation is completed, the simulation result file "result.sav" is saved and loaded. At the ground surface (y=0), in the horizontal x-direction, the vertical (y-direction) displacement of the nodes is extracted and saved at 0.5m intervals (i.e., Δx=0.5m), thereby obtaining the surface settlement curve data under this working condition.

[0035] In this embodiment of the invention, the geological mechanical parameters include the soil unit weight γ and the compression modulus E. s The parameters include: cohesion c, internal friction angle φ; tunnel geometric parameters include: tunnel diameter D, tunnel burial depth ratio C / D, center-to-center distance ratio of the two tunnels B / D, and deflection angle of the two tunnels θ; shield tunneling construction parameters include the excavation volume loss rate and grouting pressure of the two tunnels respectively.

[0036] S102, the ground surface settlement curve data is fitted by using a double-line tunnel ground surface settlement curve empirical formula, control parameters of the ground surface settlement curve are extracted, and a database containing stratum mechanics parameters, tunnel geometric parameters, shield construction parameters and ground surface settlement curve control parameters is established.

[0037] In the embodiment of the application, the control parameters of the ground surface settlement curve refer to fixed parameters of the double-line tunnel ground surface settlement curve empirical formula.

[0038] In the embodiment of the application, all parameters in the stratum mechanics parameters, the tunnel geometric parameters and the shield construction parameters can be controlled in a parameterized manner, so that the model parameters can be flexibly adjusted according to different working conditions, and large-scale batch simulation can be conveniently realized.

[0039] Specifically, the database containing the stratum mechanics parameters, the tunnel geometric parameters, the shield construction parameters and the ground surface settlement curve control parameters is established, including: controlling the stratum mechanics parameters, the tunnel geometric parameters and the shield construction parameters in a parameterized manner; adjusting the VL-GP mixed double-line shield tunnel two-dimensional simplified numerical model based on different stratum mechanics parameters, tunnel geometric parameters and shield construction parameters to obtain ground surface settlement curve data corresponding to each of the different stratum mechanics parameters, tunnel geometric parameters and shield construction parameters; fitting the ground surface settlement curve data by using a double-line tunnel ground surface settlement curve empirical formula, and extracting control parameters of the ground surface settlement curve corresponding to each of the different stratum mechanics parameters, tunnel geometric parameters and shield construction parameters to establish the database containing the stratum mechanics parameters, the tunnel geometric parameters, the shield construction parameters and the ground surface settlement curve control parameters.

[0040] In the embodiment of the application, fitting the ground surface settlement curve data by using a double-line tunnel ground surface settlement curve empirical formula includes: fitting the ground surface settlement curve generated by numerical simulation by using a double-line tunnel ground surface settlement curve empirical formula based on the superposition principle, and the empirical formula is as follows:

[0041] wherein, x is a horizontal distance from the middle line of the double-line tunnel; S max，1 and S max，2 are maximum ground surface settlements caused by the preceding and following tunnels respectively; i 1 and i 2 are distances from the inflection points of the preceding and following tunnel settlement troughs to the center line of the tunnel respectively, d is half of the horizontal center distance of the double-line tunnel; by fitting, the control parameters of the ground surface settlement curve are obtainedS max，1 、 S max，2 、 i 1 and i 2.

[0042] Specifically, as shown in Figure 3 , which shows an example fitting diagram of the ground settlement curve empirical formula of the double-line shield tunnel based on the superposition technology in the embodiment of the application. Among them, the horizontal distance from the middle line of the double-line tunnel is represented by the abscissa, specifically, the middle line of the double-line tunnel is taken as the 0 point, the horizontal distance from the middle line of the double-line tunnel to the left area is negative, and the horizontal distance from the middle line of the double-line tunnel to the right area is positive.

[0043] S103, training based on the database by using an automatic machine learning method, in the training process, the input is the stratum mechanics parameter, the tunnel geometric parameter and the shield construction parameter, the output is the settlement curve control parameter, and the intelligent prediction model of the ground settlement curve control parameter is established.

[0044] In the embodiment of the application, the AutoML method adopted is realized based on the Auto-Sklearn library of Python, Auto-Sklearn is one of the most commonly used automatic machine learning methods at present, and specifically, as shown in Figure 4 , Figure 4The automatic machine learning library Auto-Sklearn architecture diagram of the embodiment of the present application is shown, the core architecture of Auto-Sklearn includes two levels, internally adopts a classic machine learning (ML) framework, that is, the process of data preprocessing-feature engineering-regression analysis is completed, wherein 5 data preprocessing methods (such as imputation, rescaling, balancing, etc.), 18 feature processing methods (such as densifier, polynomial, PCA, etc.) and 13 regression algorithms (such as Adaboost, Gaussian process, Random forest, etc.) are encapsulated; externally adopts an automatic machine learning (AutoML) framework, uses a meta-learner, a Bayesian optimizer and an integrator to perform algorithm selection and hyperparameter optimization, the meta-learner compares the similarity between a new data set and an existing data set through learning of historical tasks, infers a more effective model and hyperparameter configuration on the new data set, the Bayesian optimizer automatically finds the best model and hyperparameter combination in performance by continuously adjusting model weights and hyperparameters, and the integrator integrates the models evaluated in the optimization process to combine multiple models with excellent performance. Auto-Sklearn reduces the burden of users and improves the performance and generalization ability of models by automating the entire machine learning process, including feature engineering, model selection and hyperparameter tuning.

[0045] The input for the AutoML intelligent prediction model training is the formation mechanics parameter, the tunnel geometric parameter and the shield construction parameter, the output is the ground settlement curve control parameter, 80% is randomly selected as the training set, the remaining 20% is selected as the test set, the 10-fold cross-validation method is adopted, that is, the model performance is evaluated multiple times under different training and validation set divisions, thereby effectively utilizing the data and reducing the dependence on a single training set division. By setting the automatic machine learning model training time (such as 9000s), the model automatically completes the training and obtains the optimal AutoML intelligent prediction model, and the mean square error MSE and the correlation coefficient R 2 The performance of the AutoML intelligent prediction model is quantitatively evaluated, if the performance meets the requirements, the model is used; if not, the training time is increased to obtain a more optimal and required prediction model.

[0046] S104, the formation mechanics parameter, the tunnel geometric parameter and the shield construction parameter of the target working condition are input into the intelligent prediction model, the target ground settlement curve control parameter is predicted, and the target double-line tunnel ground settlement curve is obtained according to the double-line tunnel ground settlement curve empirical formula and the target ground settlement curve control parameter.

[0047] Specifically, as Figure 5As shown in the figure, it shows the double-line shield tunnel ground settlement curve prediction flowchart of the embodiment of the application, in steps S101-S103, the database containing the stratum mechanics parameters, tunnel geometric parameters, shield construction parameters and ground settlement curve control parameters is constructed, and the database is trained by using the automatic machine learning method to obtain the intelligent prediction model of the ground settlement curve control parameters. In step S104, the stratum mechanics parameters, tunnel geometric parameters and tunnel construction parameters of the target double-line tunnel target working condition to be predicted can be input into the intelligent prediction model of the ground settlement curve control parameters to obtain the ground settlement curve control parameter prediction result. Based on the prediction result, the empirical formula can be superimposed to obtain the ground settlement curve prediction result. Specifically, the double-line tunnel ground settlement curve can be back calculated according to the ground settlement curve control parameters and the empirical formula.

[0048] Based on the same inventive concept, the embodiment of the application also provides a double-line shield tunnel induced ground settlement curve intelligent prediction device, the device comprises: A numerical model establishing module is configured to establish a VL-GP mixed double-line shield tunnel two-dimensional simplified numerical model considering the construction process, and generate ground settlement curve data by numerical simulation based on stratum mechanics parameters, tunnel geometric parameters and shield construction parameters through the VL-GP mixed double-line shield tunnel two-dimensional simplified numerical model. A fitting module is configured to fit the ground settlement curve data by using a double-line tunnel ground settlement curve empirical formula, extract the control parameters of the ground settlement curve, and establish a database containing stratum mechanics parameters, tunnel geometric parameters, shield construction parameters and ground settlement curve control parameters. A training module is configured to train based on the database by using an automatic machine learning method, and the input is stratum mechanics parameters, tunnel geometric parameters and shield construction parameters, and the output is settlement curve control parameters, so as to establish an intelligent prediction model of the ground settlement curve control parameters. A prediction module is configured to input the stratum mechanics parameters, tunnel geometric parameters and shield construction parameters of the target working condition into the intelligent prediction model to predict the target ground settlement curve control parameters, and obtain the target double-line tunnel ground settlement curve according to the double-line tunnel ground settlement curve empirical formula and the target ground settlement curve control parameters.

[0049] Optionally, the numerical model establishing module is configured to: The shield tunnel excavation process is divided into three construction stages: shield machine passing stage, shield tail closing and grouting stage and grouting hardening stage, and the two-dimensional simulation technology is used to sequentially realize the three construction stages to construct the VL-GP mixed double-line shield tunnel two-dimensional simplified numerical model considering the construction process.

[0050] Optionally, the numerical model establishing module is configured to: According to the working condition requirements, a complete numerical model is established, a plane strain boundary condition is set, and an initial ground stress field is generated under the K0 condition; In the shield passing stage, the soil in the excavation part is deactivated, a displacement control technique is used to make the tunnel boundary converge inward non-uniformly, the convergence speed linearly decreases with the depth, to reflect the geometric relationship between the shield shell slightly smaller than the excavation diameter and the excavation boundary under the action of gravity, and the calculation is stopped when the volume loss rate reaches a predetermined value; In the shield tail closing and grouting stage, the lining simulated by the Liner unit is activated, the grouting pressure is applied on the tunnel excavation boundary, the grouting pressure linearly increases with the depth, the physical gap between the lining and the excavation boundary is filled with fresh grouting, the elastic unit is used for simulation, and the physical and mechanical parameters of the fresh grouting are given; In the grouting hardening stage, the grouting pressure is removed, the hardened grouting is used to replace the fresh grouting, and the physical and mechanical parameters of the hardened grouting are given to the grouting body.

[0051] Optionally, the stratum mechanical parameters include the soil layer specific weight γ, the compression modulus E s , the cohesion c, and the internal friction angle φ; the tunnel geometric parameters include the tunnel diameter D, the tunnel burial depth ratio C / D, the double-line tunnel center distance ratio B / D, and the double-line tunnel deflection angle θ; and the shield construction parameters include the excavation volume loss rates of the two tunnels respectively and the grouting pressure.

[0052] Optionally, the fitting module is configured to: Parameterize control is performed on the stratum mechanical parameters, the tunnel geometric parameters, and the shield construction parameters; The VL-GP mixed double-line shield tunnel two-dimensional simplified numerical model is adjusted based on different stratum mechanical parameters, tunnel geometric parameters, and shield construction parameters, to obtain respective ground surface settlement curve data corresponding to the stratum mechanical parameters, the tunnel geometric parameters, and the shield construction parameters; The ground surface settlement curve data are fitted by using a double-line tunnel ground surface settlement curve empirical formula, control parameters of the ground surface settlement curves corresponding to different stratum mechanical parameters, tunnel geometric parameters, and shield construction parameters are extracted, to establish a database containing the stratum mechanical parameters, the tunnel geometric parameters, the shield construction parameters, and the ground surface settlement curve control parameters.

[0053] Optionally, the fitting module is further configured to: The ground surface settlement curve generated by the numerical simulation is fitted by using a double-line tunnel ground surface settlement curve empirical formula based on the superposition principle, and the empirical formula is as follows:

[0054] wherein, xis the horizontal distance from the center line of the double-line tunnel to the middle line of the double-line tunnel; S max，1 and S max，2 are the maximum ground surface settlements caused by the preceding and following tunnels, respectively; i 1 and i 2 are the distances from the inflection point of the settlement trough to the center line of the preceding and following tunnels, respectively, d is half of the center distance of the double-line tunnel; the control parameters of the ground surface settlement curve are obtained by fitting S max，1 , S max，2 , i 1 and i 2.

[0055] The embodiment of the double-line shield tunnel induced ground surface settlement curve intelligent prediction device provided by the application can be applied to any device with data processing capability, which can be a device or apparatus such as a computer. The device embodiment can be realized by software, or by hardware or a combination of software and hardware. Taking software realization as an example, as a logical device, it is formed by reading the corresponding computer program instructions in the non-volatile memory into the memory and running by the processor of the device with data processing capability. From the hardware level, as shown in Figure 6 , it is a hardware structure diagram of the device with data processing capability of the double-line shield tunnel induced ground surface settlement curve intelligent prediction device provided by the application. In addition to the processor, memory, network interface, and non-volatile memory shown in Figure 6 , the device with data processing capability in the embodiment usually includes other hardware according to the actual functions of the device with data processing capability, which will not be described here.

[0056] Based on the same inventive concept, the embodiment of the application further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps in the double-line shield tunnel induced ground surface settlement curve intelligent prediction method according to any of the above embodiments when executed.

[0057] Based on the same inventive concept, the embodiment of the application further provides a computer readable storage medium, which stores a computer program, and the program is executed by the processor to implement the steps in the double-line shield tunnel induced ground surface settlement curve intelligent prediction method according to any of the above embodiments.

[0058] Based on the same inventive concept, the embodiment of the present application provides a computer program product comprising computer programs / instructions which, when executed by a processor, implement the steps of the intelligent prediction method for double-line shield tunnel induced ground settlement curve.

[0059] Each of the embodiments in the specification is described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The same or similar parts between the embodiments can be referred to each other.

[0060] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, device, or computer program product. Therefore, the embodiments of the present application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of the present application can be in the form of a computer program product implemented on one or more computer usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0061] The embodiments of the present application are described with reference to flowcharts and / or block diagrams according to the method, terminal device (apparatus), and computer program product of the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable terminal device to produce a machine, so that the instructions executed by the processor of the computer or other programmable terminal device produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one flow or multiple flows and / or blocks Figure 1 The device that implements the functions specified in one block or multiple blocks.

[0062] These computer program instructions can also be stored in a computer readable storage medium that can guide the computer or other programmable terminal device to work in a specific manner, so that the instructions stored in the computer readable storage medium produce a product including instruction devices that implement the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one flow or multiple flows and / or blocks Figure 1 The device that implements the functions specified in one block or multiple blocks.

[0063] These computer program instructions can also be loaded into a computer or other programmable terminal device, so that a series of operation steps are performed on the computer or other programmable terminal device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable terminal device provide a process for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one flow or multiple flows and / or blocks Figure 1steps of the functions specified in the one or more blocks.

[0064] While the preferred embodiments of the application have been described above, it should be understood that many modifications and variations to these embodiments will be apparent to those skilled in the art once they learn of the basic inventive concepts. Therefore, the attached claims are intended to cover all such modifications and variations.

[0065] Finally, it is to be understood that the phraseology or terminology employed herein, such as "first" and "second", etc., are for descriptive purposes only and should not be construed to be indicative of a necessary order of occurrence, the relationships or sequences of elements or steps, etc., unless expressly so limited. Moreover, the use of the term "including" or "comprising" or any other variant thereof is intended to cover the non-exclusive inclusion of the elements or steps set forth, such that additional elements or steps are not precluded. The use of the term "comprising" does not exclude the presence of other elements or steps than those listed in the claims, and the use of the term "comprising" does not exclude the presence of other elements or steps than those listed in the claims. The term "comprising" is used herein to mean that the claims include the recited elements or steps, but do not exclude additional elements or steps.

[0066] The above describes in detail a double-line shield tunnel induced ground settlement curve intelligent prediction method provided by the present application, and the principles and implementation manners of the present application are described by using specific examples; the above description of the embodiments is only used to help understand the method of the present application and its core idea; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manners and application ranges will be changed; in conclusion, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A method for intelligent prediction of ground settlement curve induced by double-line shield tunneling, characterized in that, The method comprises: A two-dimensional simplified numerical model of a VL-GP mixed double-line shield tunnel considering the construction process is established, and based on stratum mechanical parameters, tunnel geometric parameters and shield construction parameters, numerical simulation is performed through the two-dimensional simplified numerical model of the VL-GP mixed double-line shield tunnel to generate ground surface settlement curve data; The ground surface settlement curve data is fitted by using an empirical formula of a double-line tunnel ground surface settlement curve, control parameters of the ground surface settlement curve are extracted, and a database containing stratum mechanical parameters, tunnel geometric parameters, shield construction parameters and ground surface settlement curve control parameters is established; An automatic machine learning method is used for training based on the database, in the training process, the input is stratum mechanical parameters, tunnel geometric parameters and shield construction parameters, the output is settlement curve control parameters, and an intelligent prediction model of the ground surface settlement curve control parameters is established; Stratum mechanical parameters, tunnel geometric parameters and shield construction parameters of a target working condition are input into the intelligent prediction model, target ground surface settlement curve control parameters are predicted, and a target double-line tunnel ground surface settlement curve is obtained according to the empirical formula of the double-line tunnel ground surface settlement curve and the target ground surface settlement curve control parameters. 2.The method of claim 1, wherein, A two-dimensional simplified numerical model of a VL-GP mixed double-line shield tunnel considering the construction process is established, including: The shield tunnel excavation process is divided into three construction stages: shield machine passing stage, shield tail closing and grouting stage and grouting hardening stage, and two-dimensional simulation technology is used to sequentially realize the three construction stages to construct the two-dimensional simplified numerical model of the VL-GP mixed double-line shield tunnel considering the construction process. 3.The method of claim 2, wherein, The shield tunnel excavation process is divided into three construction stages: shield machine passing stage, shield tail closing and grouting stage and grouting hardening stage, and two-dimensional simulation technology is used to sequentially realize the three construction stages, including: A complete numerical model is established according to the working condition requirements, a plane strain boundary condition is set, and an initial ground stress field is generated under the K0 condition; In the shield machine passing stage, the soil in the excavation part is deactivated, displacement control technology is used to make the tunnel boundary converge inward non-uniformly, the convergence speed decreases linearly with the depth to reflect the geometric relationship between the slightly smaller shield machine shell than the excavation diameter and the excavation boundary under the action of gravity, and the calculation is stopped when the volume loss rate reaches the predetermined value; In the shield tail closing and grouting stage, the liner element simulated lining is activated, grouting pressure is applied on the tunnel excavation boundary, the grouting pressure increases linearly with the depth, the physical gap between the lining and the excavation boundary is filled with fresh grouting, the fresh grouting is simulated by using an elastic element, and the physical and mechanical parameters of the fresh grouting are assigned; In the grouting hardening stage, the grouting pressure is removed, the fresh grouting is replaced by hardened grouting, and the physical and mechanical parameters of the hardened grouting are assigned to the grouting body. 4.The method of claim 1, wherein, The mechanical parameters of the stratum include the specific weight γ, the compression modulus E, the cohesion c and the internal friction angle φ s The geometric parameters of the tunnel include the diameter D, the buried depth ratio C / D, the center distance ratio B / D of the double-line tunnel and the deflection angle θ of the double-line tunnel. The shield construction parameters include the excavation volume loss rates of the two tunnels and the grouting pressures. 5.The method of claim 1, wherein, The database containing stratum mechanical parameters, tunnel geometric parameters, shield construction parameters and ground surface settlement curve control parameters is established, including: The stratum mechanical parameters, tunnel geometric parameters and shield construction parameters are parameterized controlled; Adjust the two-dimensional simplified numerical model of the VL-GP mixed double-line shield tunnel based on different stratum mechanical parameters, tunnel geometric parameters and shield construction parameters to obtain stratum surface settlement curve data corresponding to each of the stratum mechanical parameters, tunnel geometric parameters and shield construction parameters; Fitting the stratum surface settlement curve data using a double-line tunnel stratum surface settlement curve empirical formula to extract control parameters of the stratum surface settlement curve corresponding to each of the stratum mechanical parameters, tunnel geometric parameters and shield construction parameters, thereby establishing a database containing stratum mechanical parameters, tunnel geometric parameters, shield construction parameters and stratum surface settlement curve control parameters. 6.The method of claim 1, wherein, Fitting the stratum surface settlement curve data using a double-line tunnel stratum surface settlement curve empirical formula, including: Fitting the stratum surface settlement curve data using a double-line tunnel stratum surface settlement curve empirical formula based on the superposition principle, and the empirical formula is as follows: wherein, x is the horizontal distance from the center line of the twin-tunnel; S max，1 and S max，2 are the maximum ground surface settlements caused by the preceding and following tunnels, respectively; i 1 and i 2 are the distances from the inflection point of the settlement trough to the center line of the preceding and following tunnels, respectively, d is half of the horizontal center-to-center distance of the twin-tunnel; the control parameters of the ground surface settlement curve are obtained by fitting S max，1 , S max，2 , i 1 and i 2.

7. A device for intelligently predicting ground surface settlement curve induced by double-line shield tunneling, characterized in that, The device includes: A numerical model establishment module for establishing a two-dimensional simplified numerical model of a VL-GP mixed double-line shield tunnel considering the construction process, generating stratum surface settlement curve data through numerical simulation based on stratum mechanical parameters, tunnel geometric parameters and shield construction parameters; A fitting module for fitting the stratum surface settlement curve data using a double-line tunnel stratum surface settlement curve empirical formula, extracting control parameters of the stratum surface settlement curve and establishing a database containing stratum mechanical parameters, tunnel geometric parameters, shield construction parameters and stratum surface settlement curve control parameters; A training module for training based on the database using an automatic machine learning method, wherein the input is stratum mechanical parameters, tunnel geometric parameters and shield construction parameters, and the output is settlement curve control parameters, and an intelligent prediction model of stratum surface settlement curve control parameters is established; A prediction module for inputting target stratum mechanical parameters, tunnel geometric parameters and shield construction parameters into the intelligent prediction model to predict target stratum surface settlement curve control parameters, and obtaining a target double-line tunnel stratum surface settlement curve based on a double-line tunnel stratum surface settlement curve empirical formula and the target stratum surface settlement curve control parameters.

8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to realize the double-line shield tunnel induced stratum surface settlement curve intelligent prediction method of any one of claims 1-6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the double-line shield tunnel induced stratum surface settlement curve intelligent prediction method of any one of claims 1-6.

10. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instructions are executed by the processor to realize the steps in the double-line shield tunnel induced stratum surface settlement curve intelligent prediction method of any one of claims 1-6.

Images