Shield tunneling equipment posture prediction method, device, equipment, medium and product

Abstract

The present invention relates to shield tunneling technology, and provides a method, device, equipment, medium and product for predicting the posture of shield tunneling equipment. The method includes: constructing a global geological feature map through geological drilling data in combination with the Kriging algorithm; extracting a local geological feature map around the shield according to the current position information of the shield tunneling equipment and the global geological feature map; obtaining the operating parameters of the shield tunneling equipment, inputting the operating parameters and the local geological feature map around the shield into a pre-trained posture prediction model, and predicting the posture evolution of the shield tunneling equipment in the subsequent tunneling section; wherein the posture prediction model includes a geological information convolution module, a temporal feature extraction fusion module and a prediction module. Through the technical solution provided by the present application, the spatial characteristics of geological information and the temporal characteristics of operating parameters are integrated to effectively improve the prediction accuracy of the tunneling posture of the shield through complex strata.

Description

Technical Field

[0001] The present invention relates to the technical field of shield tunneling, and in particular to a posture prediction method, device, equipment, medium and product of shield tunneling equipment. Background Art

[0002] Three-dimensional geological models can directly display the geological structure, lithology, spatial distribution and other properties of the strata, with good visualization effects, and usually assist in the development and decision-making of underground space. Although three-dimensional geological modeling technology is relatively mature, the fusion method of geological information in high-dimensional space has not been deeply studied, and the geological information fusion method dedicated to deep learning has not been reported. On the other hand, shield construction has become the main means of underground space development in my country in recent years. At present, there are still large-scale urban subways or highway tunnels under construction or planning. Intelligent prediction and control of shield tunneling posture is one of the core modules of shield intelligent driving system. At present, a series of shield tunneling posture prediction models have been developed based on deep learning methods. However, most of the existing models only take shield tunneling parameters as input, which limits the prediction accuracy of the existing models when the shield passes through complex strata. Summary of the invention

[0003] The present invention provides a method, device, equipment, medium and product for predicting the posture of a shield tunneling equipment, which are used to solve the defect of low accuracy in predicting the posture of a shield tunneling equipment passing through complex strata in the prior art.

[0004] The present invention provides a method for predicting the posture of a shield tunneling equipment, comprising:

[0005] The geological characteristic map of the whole area is constructed by combining geological drilling data with Kriging algorithm;

[0006] Extracting a local geological characteristic map around the shield according to the current position information of the shield tunneling equipment and the global geological characteristic map;

[0007] The operating parameters of the shield tunneling equipment are obtained, and the operating parameters and the local geological feature map around the shield are input into a pre-trained posture prediction model to predict the posture evolution of the shield tunneling equipment in the subsequent tunneling section; wherein the posture prediction model includes a geological information convolution module, a time series feature extraction and fusion module and a prediction module.

[0008] According to a method for predicting the posture of a shield tunneling equipment provided by the present invention, the operating parameters and the local geological feature map around the shield are input into a pre-trained posture prediction model to predict the posture evolution of the shield tunneling equipment in a subsequent tunneling section, including:

[0009] Inputting the local geological feature map around the shield into the geological information convolution module for processing to obtain the geological information spatial features in the tunneling axis direction of the shield tunneling equipment;

[0010] Inputting the geological information spatial features and the operation parameters into the temporal feature extraction and fusion module to perform multimodal feature fusion to obtain fusion features;

[0011] The fusion features are input into the prediction module for processing to predict the posture evolution of the shield tunneling equipment in the subsequent tunneling section.

[0012] According to a posture prediction method for shield tunneling equipment provided by the present invention, the geological information convolution module includes a first convolution layer, a pooling layer and a second convolution layer connected in sequence, the temporal feature extraction and fusion module is a long short-term memory network, and the prediction module includes a fully connected layer and a Dropout layer.

[0013] According to a method for predicting the posture of a shield tunneling equipment provided by the present invention, the local geological feature map around the shield is input into the geological information convolution module for processing to obtain the geological information spatial features in the tunneling axis direction of the shield tunneling equipment, including:

[0014] Through the geological information convolution module, the initial stress, compression modulus, and compressive strength are extracted based on the local geological characteristic map around the shield, and the initial stress, compression modulus, and compressive strength are fused to obtain the geological information spatial characteristics.

[0015] According to a method for predicting the posture of a shield tunneling equipment provided by the present invention, a global geological feature map is constructed by combining geological drilling data with a Kriging algorithm, including:

[0016] Acquiring the geological drilling data of the shield tunneling equipment during the tunneling process;

[0017] Based on the geological drilling data, geological modeling is performed by using a Kriging algorithm;

[0018] The geological category of each point in the geological modeling is divided by three-dimensional spatial coordinates to establish the geological feature map.

[0019] According to a method for predicting the posture of a shield tunneling equipment provided by the present invention, the posture evolution includes a pitch angle, a roll angle, a front shield vertical deviation, a front shield horizontal deviation, a shield tail vertical deviation and a shield tail horizontal deviation.

[0020] The present invention also provides a posture prediction device for shield tunneling equipment, comprising the following modules:

[0021] A construction module is configured to construct a global geological feature map by using geological drilling data combined with a Kriging algorithm;

[0022] An extraction module is configured to extract a local geological characteristic map around the shield according to the current position information of the shield tunneling equipment and the global geological characteristic map;

[0023] The prediction module is configured to obtain the operating parameters of the shield tunneling equipment, input the operating parameters and the local geological feature map around the shield into a pre-trained posture prediction model, and predict the posture evolution of the shield tunneling equipment in the subsequent tunneling section; wherein the posture prediction model includes a geological information convolution module, a time series feature extraction and fusion module and a prediction module.

[0024] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a posture prediction method for shield tunneling equipment as described in any one of the above-mentioned methods is implemented.

[0025] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the posture prediction method of the shield tunneling equipment as described in any one of the above-mentioned methods is implemented.

[0026] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the posture prediction method of the shield tunneling equipment as described in any one of the above-mentioned methods is implemented.

[0027] The method, device, equipment, medium and product for predicting the posture of shield tunneling equipment provided by the present invention construct a global geological feature map in a high-dimensional matrix format based on geological drilling data. The corresponding local geological feature map around the shield is extracted according to the current position information of the shield tunneling equipment, and then the operating parameters of the shield tunneling equipment and the local geological feature map around the shield are input into the posture prediction model for processing to predict the posture evolution of the shield tunneling equipment in the subsequent tunneling section. This application integrates the spatial characteristics of geological information and the temporal characteristics of operating parameters to improve the prediction accuracy of the tunneling posture of the shield through complex strata. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0029] Figure 1 It is a schematic flow chart of the posture prediction method of shield tunneling equipment provided by the present invention.

[0030] Figure 2 It is a processing schematic diagram of the posture prediction model provided by the present invention.

[0031] Figure 3 Schematic diagram of a geological information convolution module provided in an embodiment of the present invention.

[0032] Figure 4 It is a schematic diagram of constructing a geological feature map provided by an embodiment of the present invention.

[0033] Figure 5 It is a structural schematic diagram of the posture prediction device of the shield tunneling equipment provided by the present invention.

[0034] Figure 6 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] Combine the following Figure 1-Figure 4 The present invention describes a method for predicting the posture of a shield tunneling equipment.

[0037] Figure 1 FIG. 1 is a flow chart of a method for predicting the posture of a shield tunneling equipment according to an exemplary embodiment. Figure 1 As shown, in an exemplary embodiment, the posture prediction method of a shield tunneling equipment includes steps 110 to 130, which are described in detail as follows.

[0038] Step 110, constructing a global geological feature map through geological drilling data combined with the Kriging algorithm.

[0039] In the embodiment of the present application, a global geological feature map in a high-dimensional matrix format is constructed based on geological drilling data, and the global geological feature map has geological categories and geological features corresponding to global position information. When posture prediction is required, the local geological feature map around the shield is obtained through the current position information index of the shield tunneling equipment.

[0040] Step 120, extracting a local geological characteristic map around the shield according to the current position information of the shield tunneling equipment and the global geological characteristic map.

[0041] In the embodiment of the present application, based on the acquired current position information and the global geological characteristic map, the shield-surrounding local geological characteristic map corresponding to the current position information is determined in the geological characteristic map.

[0042] Step 130, obtaining the operating parameters of the shield tunneling equipment, inputting the operating parameters and the local geological feature map around the shield into a pre-trained posture prediction model, and predicting the posture evolution of the shield tunneling equipment in the subsequent tunneling section; wherein the posture prediction model includes a geological information convolution module, a time series feature extraction and fusion module, and a prediction module.

[0043] In an embodiment of the present application, the operating parameters of the shield tunneling equipment are obtained, and the obtained operating parameters can be obtained based on the update frequency of the operating parameters, that is, the operating parameters updated for the most recent n times are obtained. The operating parameters and the local geological feature map around the shield are input into the pre-trained posture prediction model for processing to predict the posture evolution of the shield tunneling equipment in the subsequent tunneling section. When predicting, the posture prediction model predicts the posture evolution in the future period based on the operating parameters in the past period. For example, if the update frequency of the operating parameters is 1 minute, the geological information features spanned within n min can be integrated to predict the posture evolution in the future period, such as within m min. For example, if n=8 and m=4, the posture evolution in the next 4 minutes is predicted by the operating parameters of the past 8 minutes.

[0044] In the embodiment of the present application, a global geological feature map in a high-dimensional matrix format is constructed based on geological drilling data. The corresponding local geological feature map around the shield is determined according to the current position information of the shield tunneling equipment, and then the operating parameters of the shield tunneling equipment and the local geological feature map around the shield are input into the posture prediction model for processing. The posture evolution of the shield tunneling equipment in the subsequent tunneling section integrates the spatial characteristics of geological information and the temporal characteristics of operating parameters, realizing the accurate prediction of the tunneling posture of the shield through complex strata.

[0045] In an exemplary embodiment of the present application, the operation parameters and the target geological feature map are input into a pre-trained posture prediction model to predict the posture evolution of the shield tunneling equipment in a subsequent tunneling section, including the following steps, which are described in detail as follows.

[0046] The local geological feature map around the shield is input into the geological information convolution module for processing to obtain the geological information spatial features in the tunneling axis direction of the shield tunneling equipment.

[0047] The geological information spatial features and the operating parameters are input into the temporal feature extraction and fusion module to perform multimodal feature fusion to obtain fusion features.

[0048] The fusion features are input into the prediction module for processing to predict the posture evolution of the shield tunneling equipment in the subsequent tunneling section.

[0049] In the embodiment of the present application, the posture prediction model proposed by the present invention is as follows: Figure 2 , 3 As shown in the figure, it consists of three modules: geological information convolution module, temporal feature extraction and fusion module and prediction module. The geological information convolution module is equipped with multiple three-dimensional convolution kernels, which can simultaneously extract the geological information spatial features in the direction of the excavation axis. The initial hidden state c0 of the temporal feature extraction and fusion module is given by the geological information spatial features output by the geological information convolution module. This is different from the commonly used randomly initialized hidden state. The temporal feature extraction and fusion module allows the geological information spatial features to flow through the entire network module, thereby realizing the multimodal feature fusion of geological information spatial features and operating parameters. The prediction module uses the hidden state h0 of the temporal feature extraction and fusion module as the input. n As input, output is the posture evolution over a period of time in the future.

[0050] The hyperparameters involved in the posture prediction model also include the network hyperparameters of each module, such as the input data dimension of each module and the size, number and activation function of each layer, as well as the hyperparameters that control model training and optimization, such as learning rate and number of rounds.

[0051] In an exemplary embodiment of the present application, the geological information convolution module includes a first convolution layer, a pooling layer, and a second convolution layer connected in sequence, the temporal feature extraction fusion module is a long short-term memory network, and the prediction module includes a fully connected layer and a Dropout layer.

[0052] In the present application embodiment, Figure 3 As shown, the geological information convolution module includes the first convolution layer (Conv3D), the pooling layer (Pool3D) and the second convolution layer (Conv3D) connected in sequence, and the temporal feature extraction fusion module is a long short-term memory network (LSTM). The prediction module includes a fully connected layer and a Dropout layer, and the Dropout layer is connected after the fully connected layer to reduce the risk of overfitting of the model.

[0053] In an exemplary embodiment of the present application, the local geological feature map around the shield is input into the geological information convolution module for processing to obtain the geological information spatial characteristics in the direction of the tunneling axis of the shield tunneling equipment, including the following steps, which are described in detail as follows.

[0054] Through the geological information convolution module, the initial stress, compression modulus, and compressive strength are extracted based on the local geological characteristic map around the shield, and the initial stress, compression modulus, and compressive strength are fused to obtain the geological information spatial characteristics.

[0055] In the present application embodiment, Figure 3 and Figure 4 As shown in the figure, multiple layers of 3D convolution kernels are stacked to construct a geological information convolution module. The local geological feature map of the shield is used as input, and the initial stress (S) is extracted based on this architecture. i )、Compression modulus(E s ), compressive strength (q c ) and other geological information, and output the spatial characteristics of geological information represented by high-dimensional vectors that integrate the surrounding geological information.

[0056] In an exemplary embodiment of the present application, the construction of a global geological feature map through geological drilling data combined with a Kriging algorithm includes the following steps, which are described in detail as follows.

[0057] The geological drilling data of the shield tunneling equipment during the tunneling process is obtained.

[0058] Based on the geological drilling data, geological modeling is performed using the Kriging algorithm.

[0059] The geological category of each point in the geological modeling is divided by three-dimensional spatial coordinates to establish the global geological characteristic map.

[0060] In the embodiments of the present application, the three-dimensional geological modeling technology is relatively mature, but the fusion method for high-dimensional spatial geological information has not been deeply studied, and the geological information fusion method dedicated to deep learning has not been reported. This has brought limitations on the interpretability and accuracy of the model for some problems that are easily affected by stratum characteristics (such as shield tunneling posture, thrust, torque prediction, etc.). The present invention proposes a spatial geological information convolution method to achieve the extraction and fusion of high-dimensional spatial geological information, providing a technical basis for the subsequent development of a more accurate shield tunneling posture prediction model. The technical route and algorithm architecture of spatial geological information convolution are as follows: Figure 3 , Figure 4 Specifically, the present invention is based on the geological drilling data obtained by geological survey, uses the Kriging algorithm to perform geological modeling, uses three-dimensional spatial coordinates to divide the geological categories of each point in the geological modeling, and establishes a geological feature map in a high-dimensional matrix format. Figure 4 As shown, Figure 4 Part a in the figure is a schematic diagram of geological modeling using the Kriging algorithm. Figure 4 Part b in the figure is a schematic diagram of the 3D convolution along the excavation axis. Figure 4 Part c in the figure is a schematic diagram of high-dimensional geological information convolution.

[0061] In an exemplary embodiment of the present application, the attitude evolution includes a pitch angle (Pitch), a roll angle (Roll), a front shield vertical deviation (VDsH), a front shield horizontal deviation (HDsH), a shield tail vertical deviation (VDsT) and a shield tail horizontal deviation (HDsT).

[0062] In the embodiment of the present invention, the following Tables 1 to 6 show the prediction performance of the attitude prediction model (GeoFilter) and LSTM proposed in the present invention for the six attitude evolution components, and are presented in three evaluation quantities: mean absolute error (MAE), determination coefficient (R²), and root mean square error (RMSE). This embodiment predicts the attitude evolution of the next 4 minutes through the operating parameters of the past 8 minutes. According to the results, the attitude prediction model has significantly higher prediction accuracy for the pitch angle (Pitch) and the horizontal deviation of the shield tail (HDsT) than LSTM; at the same time, for the six attitude evolution components, the long-term prediction average accuracy of the attitude prediction model GeoFilter is higher than that of LSTM. This shows that the proposed attitude prediction method for shield tunneling equipment effectively improves the accuracy of shield attitude prediction.

[0063] Table 1

[0064]

[0065] Table 2

[0066]

[0067] Table 3

[0068]

[0069] Table 4

[0070]

[0071] Table 5

[0072]

[0073] Table 6

[0074]

[0075] The following is a description of the posture prediction device for shield tunneling equipment provided by the present invention. The posture prediction device for shield tunneling equipment described below and the posture prediction method for shield tunneling equipment described above can be referred to each other. It should be noted that the device provided in the following embodiment and the method provided in the above embodiment belong to the same concept, and the specific way in which each module and unit performs the operation has been described in detail in the method embodiment, which will not be repeated here.

[0076] In an exemplary embodiment of the present invention, see Figure 5 , Figure 5 A device for predicting the posture of a shield tunneling equipment according to an exemplary embodiment is provided, comprising:

[0077] A construction module 510 is configured to construct a global geological feature map by using geological drilling data in combination with a Kriging algorithm;

[0078] An extraction module 520 is configured to extract a local geological characteristic map around the shield according to the current position information of the shield tunneling equipment and the global geological characteristic map;

[0079] The prediction module 530 is configured to obtain the operating parameters of the shield tunneling equipment, input the operating parameters and the local geological feature map around the shield into a pre-trained posture prediction model, and predict the posture evolution of the shield tunneling equipment in the subsequent tunneling section; wherein the posture prediction model includes a geological information convolution module, a time series feature extraction and fusion module and a prediction module.

[0080] In an exemplary embodiment of the present invention, the prediction module 530 includes:

[0081] A first input submodule is configured to input the shield surrounding local geological feature map into the geological information convolution module for processing, so as to obtain the geological information spatial features in the tunneling axis direction of the shield tunneling equipment;

[0082] A second input submodule is configured to input the geological information spatial features and the operation parameters into the temporal feature extraction and fusion module to perform multimodal feature fusion to obtain fusion features;

[0083] The third input submodule is configured to input the fusion feature into the prediction module for processing, and predict the posture evolution of the shield tunneling equipment in the subsequent tunneling section.

[0084] In an exemplary embodiment of the present invention, the geological information convolution module includes a first convolution layer, a pooling layer and a second convolution layer connected in sequence, the temporal feature extraction fusion module is a long short-term memory network, and the prediction module includes a fully connected layer and a Dropout layer.

[0085] In an exemplary embodiment of the present invention, the first input submodule includes:

[0086] The extraction unit is configured to extract initial stress, compression modulus, and compressive strength based on the local geological characteristic map of the shield through the geological information convolution module, and fuse the initial stress, compression modulus, and compressive strength to obtain the geological information spatial feature.

[0087] In an exemplary embodiment of the present invention, the building block 510 includes:

[0088] An acquisition submodule configured to acquire geological drilling data of the shield tunneling equipment during the tunneling process;

[0089] A geological modeling submodule, configured to perform geological modeling based on the geological drilling data by using a Kriging algorithm;

[0090] A submodule is established, configured to classify the geological categories of each point in the geological modeling by three-dimensional spatial coordinates, and establish the global geological characteristic map.

[0091] In an exemplary embodiment of the present invention, the attitude evolution includes a pitch angle, a roll angle, a front shield vertical deviation, a front shield horizontal deviation, a shield tail vertical deviation and a shield tail horizontal deviation.

[0092] Figure 6 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 6 As shown, the electronic device may include: a processor 610, a communication interface 620, a memory 630 and a communication bus 640, wherein the processor 610, the communication interface 620 and the memory 630 communicate with each other through the communication bus 640. The processor 610 may call the logic instructions in the memory 630 to execute the posture prediction method of the shield tunneling equipment, the method comprising: constructing a global geological feature map through geological drilling data in combination with the Kriging algorithm; extracting a shield-surrounding local geological feature map according to the current position information of the shield tunneling equipment and the global geological feature map; obtaining the operating parameters of the shield tunneling equipment, inputting the operating parameters and the shield-surrounding local geological feature map into a pre-trained posture prediction model, and predicting the posture evolution of the shield tunneling equipment in the subsequent tunneling section; wherein the posture prediction model comprises a geological information convolution module, a temporal feature extraction and fusion module and a prediction module.

[0093] In addition, the logic instructions in the above-mentioned memory 630 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0094] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the posture prediction method of the shield tunneling equipment provided by the above-mentioned methods, and the method includes: constructing a global geological characteristic map through geological drilling data in combination with the Kriging algorithm; extracting a local geological characteristic map around the shield according to the current position information of the shield tunneling equipment and the global geological characteristic map; obtaining the operating parameters of the shield tunneling equipment, and inputting the operating parameters and the local geological characteristic map around the shield into a pre-trained posture prediction model to predict the posture evolution of the shield tunneling equipment in subsequent tunneling sections; wherein the posture prediction model includes a geological information convolution module, a time series feature extraction and fusion module, and a prediction module.

[0095] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the posture prediction method of shield tunneling equipment provided by the above-mentioned methods, the method comprising: constructing a global geological characteristic map through geological drilling data in combination with the Kriging algorithm; extracting a local geological characteristic map around the shield according to the current position information of the shield tunneling equipment and the global geological characteristic map; obtaining the operating parameters of the shield tunneling equipment, inputting the operating parameters and the local geological characteristic map around the shield into a pre-trained posture prediction model, and predicting the posture evolution of the shield tunneling equipment in subsequent tunneling sections; wherein the posture prediction model comprises a geological information convolution module, a time series feature extraction and fusion module, and a prediction module.

[0096] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0097] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0098] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting the posture of a shield tunneling equipment, characterized in that: include: The geological characteristic map of the whole area is constructed by combining geological drilling data with Kriging algorithm; Extracting a local geological characteristic map around the shield according to the current position information of the shield tunneling equipment and the global geological characteristic map; Acquire the operating parameters of the shield tunneling equipment, input the operating parameters and the local geological feature map around the shield into a pre-trained posture prediction model, and predict the posture evolution of the shield tunneling equipment in the subsequent tunneling section; wherein the posture prediction model includes a geological information convolution module, a temporal feature extraction and fusion module, and a prediction module; The geological drilling data is combined with the Kriging algorithm to construct a global geological feature map, including: Acquiring the geological drilling data of the shield tunneling equipment during the tunneling process; Based on the geological drilling data, geological modeling is performed by using a Kriging algorithm; Divide the geological categories of each point in the geological modeling by three-dimensional spatial coordinates to establish the global geological characteristic map; Inputting the operating parameters and the local geological feature map around the shield into the pre-trained posture prediction model to predict the posture evolution of the shield tunneling equipment in the subsequent tunneling section, including: Inputting the local geological feature map around the shield into the geological information convolution module for processing to obtain the geological information spatial features in the tunneling axis direction of the shield tunneling equipment; Inputting the geological information spatial features and the operation parameters into the temporal feature extraction and fusion module to perform multimodal feature fusion to obtain fusion features; The fusion features are input into the prediction module for processing to predict the posture evolution of the shield tunneling equipment in the subsequent tunneling section.

2. The method for predicting the posture of shield tunneling equipment according to claim 1, characterized in that: The geological information convolution module includes a first convolution layer, a pooling layer and a second convolution layer connected in sequence, the temporal feature extraction and fusion module is a long short-term memory network, and the prediction module includes a fully connected layer and a Dropout layer.

3. The method for predicting the posture of shield tunneling equipment according to claim 1, characterized in that: The shield surrounding local geological feature map is input into the geological information convolution module for processing to obtain the geological information spatial features in the tunneling axis direction of the shield tunneling equipment, including: Through the geological information convolution module, the initial stress, compression modulus, and compressive strength are extracted based on the local geological characteristic map around the shield, and the initial stress, compression modulus, and compressive strength are fused to obtain the geological information spatial characteristics.

4. The method for predicting the posture of a shield tunneling equipment according to any one of claims 1 to 3, characterized in that: The attitude evolution includes a pitch angle, a roll angle, a front shield vertical deviation, a front shield horizontal deviation, a shield tail vertical deviation and a shield tail horizontal deviation.

5. A posture prediction device for shield tunneling equipment, characterized in that: include: A construction module is configured to construct a global geological feature map by using geological drilling data combined with a Kriging algorithm; An extraction module is configured to extract a local geological characteristic map around the shield according to the current position information of the shield tunneling equipment and the global geological characteristic map; A prediction module is configured to obtain the operating parameters of the shield tunneling equipment, input the operating parameters and the local geological feature map around the shield into a pre-trained posture prediction model, and predict the posture evolution of the shield tunneling equipment in a subsequent tunneling section; wherein the posture prediction model includes a geological information convolution module, a temporal feature extraction and fusion module, and a prediction module; Building blocks, including: An acquisition submodule configured to acquire geological drilling data of the shield tunneling equipment during the tunneling process; A geological modeling submodule, configured to perform geological modeling based on the geological drilling data by using a Kriging algorithm; Establishing a submodule configured to classify the geological categories of each point in the geological modeling by three-dimensional spatial coordinates to establish the global geological characteristic map; Prediction module, including: A first input submodule is configured to input the shield surrounding local geological feature map into the geological information convolution module for processing, so as to obtain the geological information spatial features in the tunneling axis direction of the shield tunneling equipment; A second input submodule is configured to input the geological information spatial features and the operation parameters into the temporal feature extraction and fusion module to perform multimodal feature fusion to obtain fusion features; The third input submodule is configured to input the fusion feature into the prediction module for processing, and predict the posture evolution of the shield tunneling equipment in the subsequent tunneling section.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the posture prediction method of the shield tunneling equipment as described in any one of claims 1 to 4 is implemented.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the posture prediction method of the shield tunneling equipment as described in any one of claims 1 to 4 is implemented.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the posture prediction method of the shield tunneling equipment as described in any one of claims 1 to 4 is implemented.

Images