Knowledge-Assisted Prediction Method, System, Terminal and Storage Medium for Mud Cake Blockage in Tunnel Construction

Abstract

The present invention discloses a method, system, terminal and storage medium for predicting mud cake blockage in tunnel construction based on knowledge assistance. The method includes: generating knowledge feature data and corresponding mud cake blockage risk category labels through a shield mud cake risk map; extracting a target feature distribution from actual engineering data, and using a kernel density estimation algorithm to simulate the target feature distribution to generate supplementary feature data; processing the knowledge feature data, mud cake blockage risk category labels and supplementary feature data to obtain a synthetic data set, constructing a deep classification neural network model, and using the data set to pre-train the deep classification neural network model to obtain a preliminary classification model; collecting actual data at the shield tunneling site, fine-tuning the preliminary classification model using transfer learning technology, and then inputting the actual data into the model for classification prediction to output the mud cake blockage situation at the shield tunneling site. The present invention provides a real-time and accurate blockage risk warning for the construction site.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a method, system, terminal and computer-readable storage medium for predicting mud cake blockage in tunnel construction based on knowledge assistance. Background Art

[0002] Shield tunneling is a common tunnel construction technology, and its construction efficiency and safety have an important impact on the project progress. However, the occurrence of mud cake blockage during tunneling may lead to construction stagnation and damage to equipment. Therefore, predicting the risk of mud cake blockage is one of the key technologies.

[0003] There are mainly two knowledge application paradigms for predicting the risk of mud cake blockage in the engineering field: on the one hand, knowledge paradigms such as analytical formulas and risk charts formed based on expert experience summaries. Although these methods contain rich engineering experience, since the experience information often exists in the form of unstructured charts or texts and is difficult to accurately quantify, the prediction results often have inaccurate problems; on the other hand, data-driven machine learning models construct prediction models through a large amount of on-site actual data. Although they have a certain self-learning ability, due to the high cost of obtaining high-quality data, limited sample size, and complex and changeable engineering environments, the generalization and robustness of the models are insufficient, and it is difficult to play a stable role in actual engineering.

[0004] That is to say, most of the current application methods are only limited to a single paradigm and lack a method that can organically integrate expert experience and data-driven, so as to be unable to make full use of the complementary advantages of the two.

[0005] Therefore, the existing technology still needs to be improved and developed. Summary of the Invention

[0006] The main purpose of the present invention is to provide a method, system, terminal and computer-readable storage medium for predicting mud cake blockage in tunnel construction based on knowledge assistance, aiming to solve the problem that the prediction of the risk of mud cake blockage in the prior art is limited to a single paradigm, resulting in the inability to accurately predict the mud cake blockage situation at the shield tunneling site.

[0007] To achieve the above object, the present invention provides a method for predicting mud cake blockage in tunnel construction based on knowledge assistance, and the method for predicting mud cake blockage in tunnel construction based on knowledge assistance includes the following steps:

[0008] Convert the boundary conditions in the shield mud cake risk map into mathematical expressions through mathematical modeling, generate knowledge feature data based on the mathematical expressions, and label the corresponding mud cake blockage risk category labels for the knowledge feature data according to the shield mud cake risk map;

[0009] Extract the target feature distribution from the actual engineering data, and use the kernel density estimation algorithm to simulate the target feature distribution to generate supplementary feature data;

[0010] Perform data standardization processing and splicing processing on the knowledge feature data, the mud cake blockage risk category labels, and the supplementary feature data to obtain a synthetic data set, and preprocess the synthetic data set to obtain a target data set;

[0011] Construct a deep classification neural network model, and use the target data set to pre-train the deep classification neural network model to obtain a preliminary classification model;

[0012] Collect the actual data at the shield tunneling site, preprocess the actual data to obtain an actual data set, and based on the actual data set, use transfer learning technology to fine-tune the preliminary classification model to obtain a target classification model, and input the actual data into the target classification model for classification prediction, and output the mud cake blockage situation at the shield tunneling site.

[0013] Optionally, in the knowledge-assisted tunnel construction mud cake blockage prediction method, the shield mud cake risk map includes: plastic limit moisture content, liquid limit moisture content, and viscosity index;

[0014] The viscosity index is jointly represented by the plastic limit moisture content and the liquid limit moisture content.

[0015] Optionally, in the knowledge-assisted tunnel construction mud cake blockage prediction method, generating knowledge feature data based on the mathematical expression and labeling the corresponding mud cake blockage risk category labels for the knowledge feature data according to the shield mud cake risk map specifically includes:

[0016] Within the boundary range of the mathematical expression, generate multiple knowledge feature data describing key indicators by random sampling;

[0017] According to the shield mud cake risk map, label the corresponding mud cake blockage risk category labels for multiple pieces of knowledge feature data.

[0018] Optionally, in the knowledge-assisted tunnel construction mud cake blockage prediction method, using the kernel density estimation algorithm to simulate the target feature distribution to generate supplementary feature data specifically includes:

[0019] Select a Gaussian kernel function as the kernel function of the kernel density estimation algorithm, use the grid search method to determine the optimal bandwidth parameter of the kernel density estimation algorithm, and construct a target kernel density estimation algorithm according to the kernel function and the optimal bandwidth parameter;

[0020] Simulate the target feature distribution using the target kernel density estimation algorithm to generate supplementary feature data, where the supplementary feature data includes the shield cutterhead rotation speed, cutterhead torque, average tunneling speed, penetration, total thrust, and porosity.

[0021] Optionally, in the knowledge-assisted tunnel construction mud cake blockage prediction method, where the knowledge feature data, the mud cake blockage risk category label, and the supplementary feature data are subjected to data standardization processing and splicing processing to obtain a synthetic data set, specifically including:

[0022] Perform data standardization on the knowledge feature data, the mud cake blockage risk category label, the shield cutterhead rotation speed, the cutterhead torque, the average tunneling speed, the penetration, the total thrust, and the porosity to obtain standard knowledge feature data, a standard mud cake blockage risk category label, a standard shield cutterhead rotation speed, a standard cutterhead torque, a standard average tunneling speed, a standard penetration, a standard total thrust, and a standard porosity;

[0023] Splice the standard knowledge feature data, the standard mud cake blockage risk category label, the standard shield cutterhead rotation speed, the standard cutterhead torque, the standard average tunneling speed, the standard penetration, the standard total thrust, and the standard porosity to obtain a synthetic data set;

[0024] Among them, the input features of the synthetic data set are the standard knowledge feature data, the standard shield cutterhead rotation speed, the standard cutterhead torque, the standard average tunneling speed, the standard penetration, the standard total thrust, and the standard porosity, and the output feature of the synthetic data set is the standard mud cake blockage risk category label.

[0025] Optionally, in the knowledge-assisted tunnel construction mud cake blockage prediction method, where the actual data is preprocessed to obtain an actual data set, and based on the actual data set, the preliminary classification model is fine-tuned using transfer learning technology to obtain a target classification model, specifically including:

[0026] Perform normalization processing, standardization processing, and outlier removal processing on the actual data to obtain preprocessed actual data, and construct an actual data set according to the preprocessed actual data;

[0027] Based on the actual data set, the preliminary classification model is fine-tuned using transfer learning technology, keeping the overall network structure of the preliminary classification model unchanged, freezing all the parameters of the preliminary classification model, and unfreezing the weights of the last two layers of the preliminary classification model to obtain a target classification model.

[0028] Optionally, in the knowledge-assisted tunnel construction mud cake blockage prediction method, the step of inputting the actual data into the target classification model for classification prediction and outputting the mud cake blockage situation at the shield tunneling site specifically includes:

[0029] Input the actual data into the target classification model for classification prediction. When the classification result output by the target classification model is non-blockage, it is determined that there is no risk of mud cake blockage at the shield tunneling site;

[0030] When the classification result output by the target classification model is blockage, it is determined that there is a risk of mud cake blockage at the shield tunneling site, and the classification result is reported as the basis for on-site early warning.

[0031] In addition, to achieve the above object, the present invention also provides a knowledge-assisted tunnel construction mud cake blockage prediction system. The knowledge-assisted tunnel construction mud cake blockage prediction system includes:

[0032] A knowledge feature data generation module, which is used to convert the boundary conditions in the shield mud cake risk map into mathematical expressions through mathematical modeling, generate knowledge feature data based on the mathematical expressions, and label the corresponding mud cake blockage risk category labels for the knowledge feature data according to the shield mud cake risk map;

[0033] A supplementary feature data generation module, which is used to extract the target feature distribution from the actual engineering data, simulate the target feature distribution using the kernel density estimation algorithm, and generate supplementary feature data;

[0034] A dataset construction module, which is used to perform data standardization processing and splicing processing on the knowledge feature data, the mud cake blockage risk category labels, and the supplementary feature data to obtain a synthetic dataset, and perform preprocessing on the synthetic dataset to obtain a target dataset;

[0035] A model pre-training module, which is used to construct a deep classification neural network model, and use the target dataset to pre-train the deep classification neural network model to obtain a preliminary classification model;

[0036] A blockage prediction module, which is used to collect the actual data at the shield tunneling site, preprocess the actual data to obtain an actual dataset, use transfer learning technology to fine-tune the preliminary classification model to obtain a target classification model, input the actual data into the target classification model for classification prediction, and output the mud cake blockage situation at the shield tunneling site.

[0037] In addition, to achieve the above object, the present invention further provides a terminal, wherein the terminal includes: a memory, a processor, and a knowledge-assisted tunnel construction mud cake blockage prediction program stored on the memory and executable on the processor. When the knowledge-assisted tunnel construction mud cake blockage prediction program is executed by the processor, the steps of the knowledge-assisted tunnel construction mud cake blockage prediction method described above are implemented.

[0038] In addition, to achieve the above object, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a knowledge-assisted tunnel construction mud cake blockage prediction program. When the knowledge-assisted tunnel construction mud cake blockage prediction program is executed by a processor, the steps of the knowledge-assisted tunnel construction mud cake blockage prediction method described above are implemented.

[0039] In the present invention, knowledge feature data and corresponding mud cake blockage risk category labels are generated through a shield mud cake risk map; a target feature distribution is extracted from actual engineering data, and the kernel density estimation algorithm is used to simulate the target feature distribution to generate supplementary feature data; the knowledge feature data, mud cake blockage risk category labels, and supplementary feature data are processed to obtain a synthetic data set, a deep classification neural network model is constructed, and the data set is used to pre-train the deep classification neural network model to obtain a preliminary classification model; actual data at the shield tunneling site is collected, and after fine-tuning the preliminary classification model using transfer learning technology, the actual data is input into the model for classification prediction, and the mud cake blockage situation at the shield tunneling site is output. The present invention provides real-time and accurate blockage risk warnings for shield construction sites. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 is a flowchart of a preferred embodiment of the knowledge-assisted tunnel construction mud cake blockage prediction method of the present invention;

[0041] Figure 2 is a principle architecture diagram of the knowledge-assisted tunnel construction mud cake blockage prediction method of the present invention;

[0042] Figure 3 is a shield mud cake risk determination diagram of implicit knowledge in the knowledge-assisted tunnel construction mud cake blockage prediction method of the present invention;

[0043] Figure 4 is a data scatter plot matrix based on actual engineering data in the knowledge-assisted tunnel construction mud cake blockage prediction method of the present invention;

[0044] Figure 5 is a constructed data scatter plot matrix based on KDE in the knowledge-assisted tunnel construction mud cake blockage prediction method of the present invention;

[0045] Figure 6 It is a structural diagram of a preferred embodiment of the knowledge-assisted tunnel construction mud cake blockage prediction system of the present invention;

[0046] Figure 7 It is a structural diagram of a preferred embodiment of the terminal of the present invention. Specific embodiments

[0047] The present application provides a knowledge-assisted tunnel construction mud cake blockage prediction method, system, terminal and storage medium. To make the purpose, technical solution and effect of the present application clearer and more definite, the following further elaborates the present application with reference to the accompanying drawings and by way of examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0048] Those skilled in the art of the present technology can understand that, unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as the general understanding of those of ordinary skill in the field to which the present application belongs. It should also be understood that terms such as those defined in a general dictionary should be understood to have a meaning consistent with the meaning in the context of the prior art, and will not be interpreted in an idealized or overly formal sense unless specifically defined as here.

[0049] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, such descriptions of "first", "second", etc. are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, features defined with "first", "second" may explicitly or implicitly include at least one such feature. In addition, the technical solutions between various embodiments can be combined with each other, but it must be based on the fact that those of ordinary skill in the art can implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the protection scope required by the present invention.

[0050] The knowledge-assisted tunnel construction mud cake blockage prediction method described in the preferred embodiment of the present invention, as Figure 1 and Figure 2 shown, the knowledge-assisted tunnel construction mud cake blockage prediction method includes the following steps:

[0051] Step S10: Convert the boundary conditions in the shield mud cake risk map into mathematical expressions through mathematical modeling, generate knowledge feature data based on the mathematical expressions, and label the corresponding mud cake blockage risk category labels for the knowledge feature data according to the shield mud cake risk map.

[0052] Specifically, for the shield mud cake risk map (such as Figure 3Perform mathematical modeling as shown in the figure, and convert each boundary condition in the shield mud-caking risk map into a mathematical expression. In this embodiment, the shield mud-caking risk map includes: plastic limit water content (WP-Wn), liquid limit water content (WL-Wn), plasticity index (Ip), and viscosity index (Ic). Among them, the plastic limit water content is the abscissa x, and the value range of the plastic limit water content is [-160%, 50%]; the liquid limit water content is the ordinate y, and the value range of the liquid limit water content is [-20%, 200%]; the viscosity index is jointly represented by the plastic limit water content and the liquid limit water content.

[0053] Furthermore, the ordinate (WL-Wn) of all sample points in the shield mud-caking risk map must be greater than the abscissa (WP-Wn). At the same time, risk classification is carried out according to the viscosity index, and the viscosity index is converted into a relationship that can be expressed by the plastic limit water content and the liquid limit water content. For example: Ic = 0, that is, it can be represented by y = 0% in the shield mud-caking risk map; Ic = 0.5, that is, it can be represented by y = -x in the shield mud-caking risk map; Ic = 0.75, that is, it can be represented by y = -3x in the shield mud-caking risk map; Ic = 1, that is, it can be represented by x = 0% in the shield mud-caking risk map.

[0054] Furthermore, generate knowledge feature data based on the mathematical expression, and label corresponding mud-caking risk category labels for the knowledge feature data according to the shield mud-caking risk map, specifically including:

[0055] Within the boundary range of the mathematical expression, generate multiple knowledge feature data describing key indicators by means of random sampling;

[0056] According to the shield mud-caking risk map, label corresponding mud-caking risk category labels for multiple pieces of the knowledge feature data.

[0057] In this embodiment, use random sampling to generate knowledge feature data describing key indicators such as plastic limit water content and liquid limit water content within the above boundary range, and label each sample with the corresponding mud-caking risk category according to the risk map.

[0058] Table 1: Assignment of sample point labels

[0059]

[0060] Step S20: Extract the target feature distribution from the actual engineering data, and use the kernel density estimation algorithm to simulate the target feature distribution to generate supplementary feature data.

[0061] Specifically, a Gaussian kernel function is selected as the kernel function of the kernel density estimation algorithm, and a grid search method is used to determine the optimal bandwidth parameter of the kernel density estimation algorithm. The target kernel density estimation algorithm is constructed according to the kernel function and the optimal bandwidth parameter.

[0062] It can be understood that in order to supplement the shield tunneling related parameters not covered in the shield mud cake risk map (such as Figure 4 shown), the target feature distribution is extracted from the actual engineering data, and the kernel density estimation (KernelDensity Estimation, KDE) method is used to generate supplementary feature data similar to the real data distribution (such as Figure 5 shown), where, Figure 4 and Figure 5 a, b, c, d, e in represent the cutter head rotation speed, cutter head torque, average tunneling speed, penetration, total thrust and porosity of the shield respectively.

[0063] Furthermore, the target kernel density estimation algorithm is used to simulate the target feature distribution to generate supplementary feature data, where the supplementary feature data includes the cutter head rotation speed, cutter head torque, average tunneling speed, penetration, total thrust and porosity of the shield.

[0064] In this embodiment, the extracted engineering parameters include the cutter head rotation speed, cutter head torque, average tunneling speed, penetration, total thrust and porosity of the shield. The specific parameter settings of the kernel density estimation algorithm are as follows: the Gaussian kernel function is selected as the kernel function, the bandwidth parameter is determined by the grid search method, the bandwidth parameter range is set to [0.1, 10], and the logarithmic scale uniform distribution is adopted. A total of 20 candidate values are selected, and the best bandwidth is evaluated to be 0.6951. Since only the liquid limit water content and the plastic limit water content are covered in the shield mud cake risk, and other characteristic data are wanted to be generated, then find other actual engineering data, and use the kernel density estimation method to imitate the characteristic data of these actual projects, and then other characteristics are generated. Then the final formed data set can be expressed as: [liquid limit water content, plastic limit water content, cutter head rotation speed, cutter head torque,... porosity].

[0065] Step S30: Perform data standardization processing and splicing processing on the knowledge feature data, the mud cake blockage risk category label, and the supplementary feature data to obtain a synthetic data set, and perform preprocessing on the synthetic data set to obtain a target data set.

[0066] The process of performing data standardization processing and splicing processing on the knowledge feature data, the mud cake blockage risk category label, and the supplementary feature data to obtain a synthetic data set specifically includes:

[0067] Normalize the knowledge feature data, the mud cake blockage risk category label, the shield cutter head rotation speed, the cutter head torque, the average tunneling speed, the penetration degree, the total thrust, and the porosity to obtain the standard knowledge feature data, the standard mud cake blockage risk category label, the standard shield cutter head rotation speed, the standard cutter head torque, the standard average tunneling speed, the standard penetration degree, the standard total thrust, and the standard porosity;

[0068] Concatenate the standard knowledge feature data, the standard mud cake blockage risk category label, the standard shield cutter head rotation speed, the standard cutter head torque, the standard average tunneling speed, the standard penetration degree, the standard total thrust, and the standard porosity to obtain a synthetic dataset;

[0069] Among them, the input features of the synthetic dataset are the standard knowledge feature data, the standard shield cutter head rotation speed, the standard cutter head torque, the standard average tunneling speed, the standard penetration degree, the standard total thrust, and the standard porosity, and the output feature of the synthetic dataset is the standard mud cake blockage risk category label.

[0070] It can be understood that in this application, the knowledge feature data and the generated supplementary feature data are unified in dimension and then concatenated to form a synthetic dataset containing all features and risk labels, organically integrating expert experience and data-driven, so as to make full use of the complementary advantages of both.

[0071] Furthermore, preprocess the synthetic dataset, including normalization, standardization, and outlier removal, to obtain a target dataset, which is used to pre-train a deep classification neural network model to enable it to have the preliminary classification ability of mud cake blockage risk.

[0072] Step S40: Construct a deep classification neural network model, and use the target dataset to pre-train the deep classification neural network model to obtain a preliminary classification model.

[0073] It can be understood that using the constructed target dataset, pre-train a deep classification neural network model (8×64×128×64×1). And divide the target dataset into a training set, a test set, and a validation set according to a preset ratio, use the training set to train the deep classification neural network model, use the test set to evaluate the deep classification neural network model for each round of training to obtain a trained model, and use the validation set to evaluate the trained model to obtain a preliminary classification model.

[0074] Further, during the pre-training process of the model, training parameters are set, and the optimizer is set to Adam (the main function of the Adam optimizer is to update the neural network parameters according to the gradient information to minimize the loss function), the loss function is set to cross-entropy loss, the learning rate is set to 0.001, and the number of iterations is set to 100. After the pre-training is completed, a preliminary classification model is obtained. The preliminary classification model can initially realize the classification prediction of the mud cake blockage risk.

[0075] Step S50: Collect the actual data at the shield tunneling site, preprocess the actual data to obtain an actual data set, and based on the actual data set, use transfer learning technology to fine-tune the preliminary classification model to obtain a target classification model. Input the actual data into the target classification model for classification prediction, and output the mud cake blockage situation at the shield tunneling site.

[0076] Specifically, preprocessing the actual data includes normalization processing, standardization processing, and outlier removal processing to obtain preprocessed actual data, and constructing an actual data set according to the preprocessed actual data.

[0077] Normalization is a way to simplify calculations, that is, a dimensional expression is transformed into a dimensionless expression to become a scalar. Standardization (Normalization) usually transforms the data into a distribution with the same standard deviation and mean. Standardization can make the data easier to compare and analyze. Outlier removal refers to a method of processing or excluding extreme data that deviates from the normal range in data analysis.

[0078] Further, based on the actual data set, use transfer learning technology to fine-tune the preliminary classification model, keep the overall network structure of the preliminary classification model unchanged, freeze all the parameters of the preliminary classification model, and unfreeze the weights of the last two layers of the preliminary classification model to obtain a target classification model.

[0079] It can be understood that transfer learning is a machine learning method, which means that a pre-trained model is reused in another task. In this embodiment, based on the actual data set, transfer learning technology is used to fine-tune the pre-trained preliminary classification model. And during this process, keep the overall network structure of the model unchanged, freeze some parameters of the model, and only adjust the weights of specific layers, so that the model can quickly adapt to the real working conditions and realize the classification result that directly reflects the on-site blockage risk. For example, first freeze all the parameters of the pre-trained model, and only unfreeze the weights of the last two layers (i.e., 128×64×1) partially, so that the model can quickly adapt to the actual on-site working conditions, thereby further improving the classification performance.

[0080] Furthermore, input the actual data into the target classification model for classification prediction. When the classification result output by the target classification model is non-blockage, it is determined that there is no risk of mud cake blockage at the shield tunneling site; when the classification result output by the target classification model is blockage, it is determined that there is a risk of mud cake blockage at the shield tunneling site, and the classification result is reported as the basis for on-site early warning.

[0081] It can be seen that this application converts the expert experience in the shield risk map into mathematical model to generate knowledge feature data, and uses kernel density estimation to generate supplementary feature data, thus constructing a synthetic data set with domain knowledge. Then, this data set is used to pre-train a deep classification neural network, and the pre-trained model is fine-tuned through transfer learning, realizing the effective combination of expert experience knowledge and data-driven methods. This method not only reduces the data acquisition cost, but also greatly improves the generalization ability and prediction accuracy of the model under actual engineering conditions, providing real-time and accurate blockage risk early warning for the shield construction site, and having broad engineering application prospects.

[0082] The beneficial effects of the present invention are as follows:

[0083] (1). Using the risk map to generate synthetic data for pre-training realizes the explicit injection of domain knowledge and reduces the data acquisition and annotation costs;

[0084] (2). Adopting the deep transfer learning technology enables the pre-trained model to quickly adapt to the actual working conditions on site, directly output the classification result of the blockage risk, and simplifies the model structure;

[0085] (3). This method innovatively realizes the effective connection of the expert experience knowledge paradigm and the data-driven machine learning paradigm. It not only overcomes the problems of insufficient quantification and accuracy of traditional expert experience methods, but also avoids the deficiencies of poor generalization and robustness of pure data-driven methods in the case of insufficient samples, thus significantly improving the prediction accuracy and engineering adaptability;

[0086] (4). It provides real-time and accurate blockage risk early warning for the shield construction site, and has broad engineering application prospects.

[0087] Further, as Figure 6 shown, based on the above knowledge-assisted tunnel construction mud cake blockage prediction method, the present invention also correspondingly provides a knowledge-assisted tunnel construction mud cake blockage prediction system, wherein, the knowledge-assisted tunnel construction mud cake blockage prediction system includes:

[0088] The knowledge feature data generation module 51 is used to convert the boundary conditions in the shield mud-caking risk map into mathematical expressions through mathematical modeling, generate knowledge feature data based on the mathematical expressions, and label the corresponding mud-caking risk category labels for the knowledge feature data according to the shield mud-caking risk map;

[0089] The supplementary feature data generation module 52 is used to extract the target feature distribution from the actual engineering data, simulate the target feature distribution by using the kernel density estimation algorithm, and generate supplementary feature data;

[0090] The data set construction module 53 is used to perform data standardization processing and splicing processing on the knowledge feature data, the mud-caking risk category labels, and the supplementary feature data to obtain a synthetic data set, and perform preprocessing on the synthetic data set to obtain a target data set;

[0091] The model pre-training module 54 is used to construct a deep classification neural network model, and use the target data set to pre-train the deep classification neural network model to obtain a preliminary classification model;

[0092] The blockage prediction module 55 is used to collect the actual data at the shield tunneling site, preprocess the actual data to obtain an actual data set, use the transfer learning technology to fine-tune the preliminary classification model to obtain a target classification model, input the actual data into the target classification model for classification prediction, and output the mud-caking situation at the shield tunneling site.

[0093] Further, as Figure 7 shown, based on the above knowledge-assisted tunnel construction mud-caking prediction method and system, the present invention also correspondingly provides a terminal, and the terminal includes a processor 10, a memory 20, and a display 30. Figure 7 Only some components of the terminal are shown, but it should be understood that it is not required to implement all the shown components, and more or fewer components can be alternatively implemented.

[0094] In some embodiments, the memory 20 may be an internal storage unit of the terminal, such as the hard disk or memory of the terminal. In some other embodiments, the memory 20 may also be an external storage device of the terminal, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped on the terminal. Further, the memory 20 may also include both the internal storage unit of the terminal and the external storage device. The memory 20 is used to store application software installed on the terminal and various types of data, such as the program code of the installed terminal, etc. The memory 20 may also be used to temporarily store data that has been output or will be output. In one embodiment, a knowledge-assisted tunnel construction mud cake blockage prediction program 40 is stored on the memory 20, and the knowledge-assisted tunnel construction mud cake blockage prediction program 40 can be executed by the processor 10, so as to implement the knowledge-assisted tunnel construction mud cake blockage prediction method in the present application.

[0095] In some embodiments, the processor 10 may be a central processing unit (CPU), a microprocessor or other data processing chips, and is used to run the program code stored in the memory 20 or process data, such as executing the knowledge-assisted tunnel construction mud cake blockage prediction method, etc.

[0096] In some embodiments, the display 30 may be an LED display, a liquid crystal display, a touch liquid crystal display, and an OLED (Organic Light-Emitting Diode) toucher, etc. The display 30 is used to display information on the terminal and to display a visual user interface. The components of the terminal communicate with each other through a system bus.

[0097] In one embodiment, when the processor 10 executes the knowledge-assisted tunnel construction mud cake blockage prediction program 40 in the memory 20, the following steps are implemented:

[0098] Convert the boundary conditions in the shield mud cake risk map into mathematical expressions through mathematical modeling, generate knowledge feature data based on the mathematical expressions, and label the corresponding mud cake blockage risk category labels for the knowledge feature data according to the shield mud cake risk map;

[0099] Extract the target feature distribution from the actual engineering data, and use the kernel density estimation algorithm to simulate the target feature distribution to generate supplementary feature data;

[0100] Perform data standardization processing and splicing processing on the knowledge feature data, the mud cake blockage risk category labels, and the supplementary feature data to obtain a synthetic data set, and perform preprocessing on the synthetic data set to obtain a target data set;

[0101] Construct a deep classification neural network model, and use the target data set to pre-train the deep classification neural network model to obtain a preliminary classification model;

[0102] Collect the actual data at the shield tunneling site, perform preprocessing on the actual data to obtain an actual data set, and based on the actual data set, use transfer learning technology to fine-tune the preliminary classification model to obtain a target classification model. Input the actual data into the target classification model for classification prediction, and output the mud cake blockage situation at the shield tunneling site.

[0103] Among them, the shield mud cake risk map includes: plastic limit moisture content, liquid limit moisture content, and viscosity index;

[0104] The viscosity index is jointly represented by the plastic limit moisture content and the liquid limit moisture content.

[0105] Among them, generating knowledge feature data based on the mathematical expression and labeling the corresponding mud cake blockage risk category labels for the knowledge feature data according to the shield mud cake risk map specifically includes:

[0106] Within the boundary range of the mathematical expression, generate multiple knowledge feature data describing key indicators by means of random sampling;

[0107] According to the shield mud cake risk map, label the corresponding mud cake blockage risk category labels for multiple pieces of the knowledge feature data.

[0108] Among them, using the kernel density estimation algorithm to simulate the target feature distribution and generate supplementary feature data specifically includes:

[0109] Select the Gaussian kernel function as the kernel function of the kernel density estimation algorithm, use the grid search method to determine the optimal bandwidth parameter of the kernel density estimation algorithm, and construct a target kernel density estimation algorithm according to the kernel function and the optimal bandwidth parameter;

[0110] Use the target kernel density estimation algorithm to simulate the target feature distribution and generate supplementary feature data. Among them, the supplementary feature data includes shield cutterhead rotation speed, cutterhead torque, average tunneling speed, penetration, total thrust, and porosity.

[0111] Among them, the data standardization and splicing of the knowledge feature data, the mud cake blockage risk category label, and the supplementary feature data to obtain a synthetic dataset specifically include:

[0112] Perform data standardization on the knowledge feature data, the mud cake blockage risk category label, the shield cutter head rotation speed, the cutter head torque, the average tunneling speed, the penetration rate, the total thrust, and the porosity to obtain standard knowledge feature data, a standard mud cake blockage risk category label, a standard shield cutter head rotation speed, a standard cutter head torque, a standard average tunneling speed, a standard penetration rate, a standard total thrust, and a standard porosity;

[0113] Splice the standard knowledge feature data, the standard mud cake blockage risk category label, the standard shield cutter head rotation speed, the standard cutter head torque, the standard average tunneling speed, the standard penetration rate, the standard total thrust, and the standard porosity to obtain a synthetic dataset;

[0114] Among them, the input features of the synthetic dataset are the standard knowledge feature data, the standard shield cutter head rotation speed, the standard cutter head torque, the standard average tunneling speed, the standard penetration rate, the standard total thrust, and the standard porosity, and the output feature of the synthetic dataset is the standard mud cake blockage risk category label.

[0115] Among them, the preprocessing of the actual data to obtain an actual dataset, and based on the actual dataset, using transfer learning technology to fine-tune the preliminary classification model to obtain a target classification model specifically includes:

[0116] Perform normalization, standardization, and outlier removal on the actual data to obtain preprocessed actual data, and construct an actual dataset according to the preprocessed actual data;

[0117] Based on the actual dataset, use transfer learning technology to fine-tune the preliminary classification model, keep the overall network structure of the preliminary classification model unchanged, freeze all the parameters of the preliminary classification model, and unfreeze the weights of the last two layers of the preliminary classification model to obtain a target classification model.

[0118] Among them, the inputting of the actual data into the target classification model for classification prediction and outputting the mud cake blockage situation at the shield tunneling site specifically includes:

[0119] Input the actual data into the target classification model for classification prediction. When the classification result output by the target classification model is non-blockage, it is determined that there is no mud cake blockage risk at the shield tunneling site;

[0120] When the classification result output by the target classification model is "blockage", it is determined that there is a risk of mud cake blockage at the shield tunneling site, and the classification result is reported as the basis for on-site early warning.

[0121] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a knowledge-assisted tunnel construction mud cake blockage prediction program. When the knowledge-assisted tunnel construction mud cake blockage prediction program is executed by a processor, the steps of the knowledge-assisted tunnel construction mud cake blockage prediction method as described above are implemented.

[0122] In summary, the present invention proposes a knowledge-assisted tunnel construction mud cake blockage prediction method, system, terminal and storage medium. The method includes: generating knowledge feature data and corresponding mud cake blockage risk category labels through a shield mud cake risk map; extracting a target feature distribution from actual engineering data, and using a kernel density estimation algorithm to simulate the target feature distribution to generate supplementary feature data; processing the knowledge feature data, mud cake blockage risk category labels and supplementary feature data to obtain a synthetic data set, constructing a deep classification neural network model, and using the data set to pre-train the deep classification neural network model to obtain a preliminary classification model; collecting actual data at the shield tunneling site, using transfer learning technology to fine-tune the preliminary classification model, and then inputting the actual data into the model for classification prediction to output the mud cake blockage situation at the shield tunneling site. The present invention pre-trains the deep network by extracting professional knowledge from the shield risk map to generate synthetic data, and then uses the actual on-site data for transfer fine-tuning to enable the model to directly output the blockage risk classification result. This method effectively reduces the data collection cost, makes up for the deficiency of traditional transfer learning lacking domain knowledge injection, and significantly improves the recognition accuracy and robustness of the model for low-probability blockage events.

[0123] It should be noted that in this article, the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or terminal including a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or terminal. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article or terminal including the element.

[0124] Of course, those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided by the present invention can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0125] It should be understood that the application of the present invention is not limited to the above examples. For those of ordinary skill in the art, improvements or transformations can be made according to the above description. All such improvements and transformations should fall within the protection scope of the appended claims of the present invention.

Claims

1. A knowledge-assisted prediction method for mud cake blockage in tunnel construction, characterized in that The described knowledge-assisted tunnel construction mud cake blockage prediction method includes: Converting the boundary conditions in the shield mud cake risk map into mathematical expressions through mathematical modeling, generating knowledge feature data based on the mathematical expressions, and labeling corresponding mud cake blockage risk category labels for the knowledge feature data according to the shield mud cake risk map; Extracting the target feature distribution from the actual engineering data, and using the kernel density estimation algorithm to simulate the target feature distribution to generate supplementary feature data; Performing data standardization processing and splicing processing on the knowledge feature data, the mud cake blockage risk category labels, and the supplementary feature data to obtain a synthetic data set, and preprocessing the synthetic data set to obtain a target data set; Constructing a deep classification neural network model, and using the target data set to pre-train the deep classification neural network model to obtain a preliminary classification model; Collecting the actual data at the shield tunneling site, preprocessing the actual data to obtain an actual data set, and based on the actual data set, using transfer learning technology to fine-tune the preliminary classification model to obtain a target classification model, and inputting the actual data into the target classification model for classification prediction to output the mud cake blockage situation at the shield tunneling site.

2. The method for predicting mud cake blockage in tunnel construction based on knowledge assistance according to claim 1, wherein The shield mud cake risk map includes: plastic limit water content, liquid limit water content, and viscosity index; The viscosity index is jointly represented by the plastic limit water content and the liquid limit water content.

3. The knowledge-assisted tunnel construction mud cake blockage prediction method according to claim 1, characterized in that The generating knowledge feature data based on the mathematical expressions and labeling corresponding mud cake blockage risk category labels for the knowledge feature data according to the shield mud cake risk map specifically includes: Generating multiple knowledge feature data describing key indicators by means of random sampling within the boundary range of the mathematical expressions; Labeling corresponding mud cake blockage risk category labels for the multiple knowledge feature data according to the shield mud cake risk map.

4. The method for predicting mud cake blockage in tunnel construction based on knowledge assistance according to claim 1, wherein The using the kernel density estimation algorithm to simulate the target feature distribution to generate supplementary feature data specifically includes: Selecting a Gaussian kernel function as the kernel function of the kernel density estimation algorithm, using a grid search method to determine the optimal bandwidth parameter of the kernel density estimation algorithm, and constructing a target kernel density estimation algorithm according to the kernel function and the optimal bandwidth parameter; Using the target kernel density estimation algorithm to simulate the target feature distribution to generate supplementary feature data, where the supplementary feature data includes shield cutter head rotation speed, cutter head torque, average tunneling speed, penetration, total thrust, and porosity.

5. The knowledge-assisted tunnel construction mud cake blockage prediction method according to claim 4, wherein The performing data standardization processing and splicing processing on the knowledge feature data, the mud cake blockage risk category labels, and the supplementary feature data to obtain a synthetic data set specifically includes: Perform data standardization on the knowledge feature data, the mud cake blockage risk category label, the shield cutterhead rotation speed, the cutterhead torque, the average tunneling speed, the penetration rate, the total thrust, and the porosity to obtain standardized knowledge feature data, a standardized mud cake blockage risk category label, a standardized shield cutterhead rotation speed, a standardized cutterhead torque, a standardized average tunneling speed, a standardized penetration rate, a standardized total thrust, and a standardized porosity; Concatenate the standardized knowledge feature data, the standardized mud cake blockage risk category label, the standardized shield cutterhead rotation speed, the standardized cutterhead torque, the standardized average tunneling speed, the standardized penetration rate, the standardized total thrust, and the standardized porosity to obtain a synthetic dataset; Among them, the input features of the synthetic dataset are the standardized knowledge feature data, the standardized shield cutterhead rotation speed, the standardized cutterhead torque, the standardized average tunneling speed, the standardized penetration rate, the standardized total thrust, and the standardized porosity, and the output feature of the synthetic dataset is the standardized mud cake blockage risk category label.

6. The knowledge-assisted tunnel construction mud cake blockage prediction method according to claim 5, wherein The preprocessing of the actual data to obtain an actual dataset, and based on the actual dataset, using transfer learning technology to fine-tune the preliminary classification model to obtain a target classification model specifically includes: Perform normalization processing, standardization processing, and outlier removal processing on the actual data to obtain preprocessed actual data, and construct an actual dataset according to the preprocessed actual data; Based on the actual dataset, use transfer learning technology to fine-tune the preliminary classification model, keep the overall network structure of the preliminary classification model unchanged, freeze all the parameters of the preliminary classification model, and unfreeze the weights of the last two layers of the preliminary classification model to obtain a target classification model.

7. The method for predicting mud cake blockage in tunnel construction based on knowledge assistance according to claim 1, wherein The inputting the actual data into the target classification model for classification prediction and outputting the mud cake blockage situation at the shield tunneling site specifically includes: Input the actual data into the target classification model for classification prediction. When the classification result output by the target classification model is non-blockage, it is determined that there is no mud cake blockage risk at the shield tunneling site; When the classification result output by the target classification model is blockage, it is determined that there is a mud cake blockage risk at the shield tunneling site, and the classification result is reported as the basis for on-site early warning.

8. A knowledge-assisted prediction system for mud cake blockage in tunnel construction, characterized in that, The knowledge-assisted tunnel construction mud cake blockage prediction system based on includes: A knowledge feature data generation module for converting the boundary conditions in the shield mud cake risk map into mathematical expressions through mathematical modeling, generating knowledge feature data based on the mathematical expressions, and labeling the corresponding mud cake blockage risk category label for the knowledge feature data according to the shield mud cake risk map; A supplementary feature data generation module for extracting the target feature distribution from the actual engineering data, simulating the target feature distribution using the kernel density estimation algorithm, and generating supplementary feature data; A dataset construction module, configured to perform data standardization processing and splicing processing on the knowledge feature data, the mud cake plugging risk category labels, and the supplementary feature data to obtain a synthetic dataset, and perform preprocessing on the synthetic dataset to obtain a target dataset; A model pre-training module, configured to construct a deep classification neural network model, and use the target dataset to pre-train the deep classification neural network model to obtain a preliminary classification model; A plugging prediction module, configured to collect actual data at the shield tunneling site, perform preprocessing on the actual data to obtain an actual dataset, use transfer learning technology to fine-tune the preliminary classification model to obtain a target classification model, input the actual data into the target classification model for classification prediction, and output the mud cake plugging situation at the shield tunneling site.

9. A terminal, characterized in that, The terminal includes: a memory, a processor, and a knowledge-assisted tunnel construction mud cake plugging prediction program stored on the memory and executable on the processor. When the knowledge-assisted tunnel construction mud cake plugging prediction program is executed by the processor, the steps of the knowledge-assisted tunnel construction mud cake plugging prediction method according to any one of claims 1-7 are implemented.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a knowledge-assisted tunnel construction mud cake plugging prediction program. When the knowledge-assisted tunnel construction mud cake plugging prediction program is executed by a processor, the steps of the knowledge-assisted tunnel construction mud cake plugging prediction method according to any one of claims 1-7 are implemented.

Images