Tunnel grouting diffusion form prediction system and method

Abstract

The invention discloses a tunnel grouting diffusion form prediction system and method. The system comprises a tunnel simulation device used for preparing a tunnel model with cracks to simulate a tunnel rock mass; the grouting simulation system is used for conducting grouting on the tunnel simulation device by adjusting grout diffusion influence factors, and the grout diffusion state under the influence of multiple factors is obtained. The grouting diffusion form quantification system is used for cutting the grouted tunnel simulation device to obtain a plurality of slices at depth positions, and performing grid division on the surface of the slice at each depth position to obtain a grout diffusion form quantification data set; and the grouting diffusion form prediction system is used for constructing a multi-modal space-time fusion depth prediction model and inputting the grout diffusion form quantitative data set into the multi-modal space-time fusion depth prediction model to predict the diffusion form of the future grout. According to the method, the prediction model is established, and the grout diffusion form quantitative data set is used as the space-time sequence data set, so that the space-time prediction of the grouting diffusion form is realized.

Description

Technical Field

[0001] The invention belongs to the technical field of tunnel grouting, and in particular relates to a morphology prediction system and method for tunnel grouting diffusion. Background Art

[0002] Tunnel engineering plays an important role in urban construction and transportation. Grouting technology, as one of the important construction methods in underground engineering, is widely used in tunnel engineering to improve the stability and sealing of underground structures. However, the study of the morphology of slurry diffusion in tunnel cracks and porous media and its influencing factors still faces many challenges. At present, the research on the slurry diffusion morphology in tunnel grouting engineering mainly focuses on two aspects: one is the quantitative analysis of the permeability and diffusion properties of the slurry during the grouting process, and the other is the evaluation and prediction of the stability and sealing of underground structures after grouting.

[0003] In terms of slurry permeability and diffusion performance, domestic and foreign scholars have explored the laws of slurry diffusion in different media such as rock and soil through laboratory simulation tests and numerical simulation methods. However, existing studies often ignore the impact of geological structure complexity on grouting effects and lack intuitive display and quantitative analysis of slurry diffusion morphology in fractures and porous media.

[0004] The evaluation and prediction of the stability and sealing of underground structures after grouting currently mainly rely on empirical formulas and numerical simulation methods. However, due to the complexity and uncertainty of underground projects, existing methods often have difficulty in accurately predicting the changing laws of slurry diffusion patterns at different times and spaces, thus affecting the effectiveness and safety of grouting projects.

[0005] In view of the above problems in the prior art, it is urgent to propose a morphology prediction system and method for tunnel grouting diffusion. Summary of the invention

[0006] In order to solve the above technical problems, the present invention proposes a tunnel grouting diffusion morphology prediction system and method to solve the above problems in the prior art.

[0007] To achieve the above-mentioned object, the present invention provides a tunnel grouting diffusion morphology prediction system, comprising: a grouting diffusion visualization system, a grouting diffusion morphology quantification system and a grouting diffusion morphology prediction system;

[0008] The grouting diffusion visualization system includes a grouting simulation system and a tunnel simulation device;

[0009] Wherein, the tunnel simulation device is used to prepare a tunnel model with cracks to simulate the tunnel rock mass;

[0010] The grouting simulation system is used to perform grouting on the tunnel simulation device by adjusting the factors affecting slurry diffusion, so as to obtain the slurry diffusion state under the influence of multiple factors;

[0011] The grouting diffusion morphology quantification system is used to cut the tunnel simulation device after grouting to obtain slices at several depth positions, and to grid the slice surface at each depth position, record the slurry diffusion morphology, and obtain a slurry diffusion morphology quantification data set;

[0012] The grouting diffusion morphology prediction system is used to construct a multimodal spatiotemporal fusion depth prediction model, and the slurry diffusion morphology quantization data set is input into the multimodal spatiotemporal fusion depth prediction model to predict the future slurry diffusion morphology.

[0013] Optionally, the preparation process of the tunnel simulation device includes: mixing river sand, paraffin and pebbles in a preset proportion and casting to obtain a tunnel model; cutting the tunnel model based on geological statistical crack distribution parameters to obtain simulated tunnel cracks; mixing coal powder, mica powder, perlite and foaming agent and applying them to the cut surface of the tunnel model to obtain a simulated tunnel crack surface, thereby completing the preparation of the tunnel simulation device.

[0014] Optionally, the tunnel model is horseshoe-shaped.

[0015] Optionally, the factors affecting slurry diffusion include the medium of the tunnel model itself, the crack opening, the crack length, the crack surface roughness, the medium porosity, and the slurry concentration and the grouting pressure.

[0016] Optionally, the grouting simulation system includes a grouting test operating table, a grouting bucket, slurry and pigment, which is used to grout the tunnel model based on the grouting bucket on the grouting test operating table, and add different pigments to different fracture slurries.

[0017] Optionally, the grouting diffusion morphology prediction system includes a model building module, a model training module and a morphology prediction module;

[0018] The model building module is used to build a multimodal spatiotemporal fusion deep prediction model based on convolutional neural networks, long short-term memory networks, graph neural networks, and spatiotemporal convolutional networks;

[0019] The model training module is used to train the multimodal spatiotemporal fusion depth prediction model in combination with fluid mechanics and permeability constraints in physical modeling;

[0020] The morphology prediction module is used to obtain the slurry diffusion morphology at each moment and each position in the future tunnel grouting process based on the trained multi-modal spatiotemporal fusion depth prediction model.

[0021] Optionally, the morphology prediction module includes a data input unit, a feature extraction unit, a feature fusion unit and a morphology prediction unit;

[0022] The data input unit is used to input the slurry diffusion influencing factors and the slurry diffusion morphology quantitative data set into the trained multi-modal spatiotemporal fusion depth prediction model;

[0023] The feature extraction unit is used to extract geometric features of the rock mass based on the convolutional neural network, extract dynamic features of slurry diffusion based on the long short-term memory network, and extract fracture correlation based on the graph neural network;

[0024] The feature fusion unit is used to fuse the features output by the feature extraction unit based on a fusion mechanism to obtain fused features;

[0025] The morphology prediction unit is used to predict the slurry diffusion morphology based on the fusion features and in combination with the spatiotemporal features extracted by the spatiotemporal convolutional network.

[0026] The present invention also provides a morphology prediction method for tunnel grouting diffusion, based on a morphology prediction system for tunnel grouting diffusion, comprising the following steps:

[0027] preparing a tunnel simulation device, and performing grouting on the tunnel simulation device based on a grouting simulation system;

[0028] Cut the tunnel simulation device after grouting to obtain slices at several depth positions;

[0029] Gridding the slice surface at each depth position, recording the slurry diffusion morphology, and obtaining a quantitative data set of the slurry diffusion morphology;

[0030] A multimodal spatiotemporal fusion deep prediction model is constructed, and the slurry diffusion morphology quantification data set is input into the multimodal spatiotemporal fusion deep prediction model to predict the future slurry diffusion morphology.

[0031] Optionally, the process of meshing the slice surface at each depth position, recording the slurry diffusion morphology, and obtaining a quantitative data set of the slurry diffusion morphology includes:

[0032] The slurry diffusion conditions in each grid are encoded with different numerical values. By recording the grid code at each slice, the slurry diffusion data at each depth position is obtained, and the slurry diffusion data set at each depth position constitutes a complete slurry diffusion morphology quantitative data set.

[0033] Optionally, a multimodal spatiotemporal fusion depth prediction model is constructed based on the slurry diffusion morphology quantification data set, and a process of predicting the future slurry diffusion morphology based on the multimodal spatiotemporal fusion depth prediction model includes:

[0034] Based on convolutional neural networks, long short-term memory networks, graph neural networks, and spatiotemporal convolutional networks, a multimodal spatiotemporal fusion deep prediction model is constructed;

[0035] Input the factors affecting slurry diffusion and the quantitative data set of slurry diffusion morphology into the trained multimodal spatiotemporal fusion depth prediction model, extract the geometric characteristics of the rock mass based on the convolutional neural network, extract the dynamic characteristics of slurry diffusion based on the long short-term memory network, and extract the fracture correlation based on the graph neural network;

[0036] Based on the fusion mechanism, the geometric characteristics of the rock mass, the dynamic characteristics of slurry diffusion and the fracture correlation are fused to obtain fusion features. Based on the fusion features and combined with the spatiotemporal features extracted by the spatiotemporal convolutional network, the slurry diffusion morphology is predicted.

[0037] Compared with the prior art, the present invention has the following advantages and technical effects:

[0038] The present invention proposes a tunnel fissure-porous medium grouting diffusion visualization device, which includes a tunnel simulation device and a grouting simulation system. Through the preparation of simulated rock mass and the adjustment of the grouting simulation system, the influence of multiple factors such as rock mass medium properties, fissure characteristics and slurry injection parameters on the grouting diffusion morphology can be considered at the same time, thereby simulating the actual engineering situation more realistically.

[0039] The present invention proposes a grouting diffusion morphology quantification system, which realizes accurate quantification of grout diffusion morphology by cutting the tunnel simulation device, taking photos and recording, and meshing. Compared with traditional methods, the system can more accurately record the diffusion state of grout on the tunnel face surface and generate a digital diffusion morphology data set, providing a more reliable basis for subsequent data processing and analysis.

[0040] The present invention proposes a grouting diffusion morphology prediction system, which realizes the spatiotemporal prediction of the grouting diffusion morphology by establishing a prediction model and using a slurry diffusion morphology quantization data set as a spatiotemporal sequence data set. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0042] Figure 1 A schematic diagram of a prediction process according to an embodiment of the present invention;

[0043] Figure 2 A schematic diagram of a horseshoe-shaped tunnel simulation device according to an embodiment of the present invention;

[0044] Figure 3 Schematic diagram of a grouting diffusion morphology quantification system according to an embodiment of the present invention;

[0045] Figure 4 It is a schematic diagram of grouting diffusion morphology prediction according to an embodiment of the present invention. DETAILED DESCRIPTION

[0046] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0047] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0048] like Figure 1 As shown, in this embodiment, a tunnel grouting diffusion morphology prediction system is provided, including: a grouting diffusion visualization system, a grouting diffusion morphology quantification system and a grouting diffusion morphology prediction system;

[0049] The grouting diffusion visualization system includes a grouting simulation system and a tunnel simulation device;

[0050] Wherein, the tunnel simulation device is used to prepare a tunnel model with cracks to simulate the tunnel rock mass;

[0051] The grouting simulation system is used to perform grouting on the tunnel simulation device by adjusting the factors affecting slurry diffusion, so as to obtain the slurry diffusion state under the influence of multiple factors;

[0052] The grouting diffusion morphology quantification system is used to cut the tunnel simulation device after grouting to obtain slices at several depth positions, and to grid the slice surface at each depth position, record the slurry diffusion morphology, and obtain a slurry diffusion morphology quantification data set;

[0053] The grouting diffusion morphology prediction system is used to construct a multimodal spatiotemporal fusion depth prediction model, and the slurry diffusion morphology quantization data set is input into the multimodal spatiotemporal fusion depth prediction model to predict the future slurry diffusion morphology.

[0054] As a specific embodiment, the tunnel simulation device is as follows: Figure 2The device uses river sand and paraffin mixed at a ratio of 100:10 to simulate the tunnel rock mass, and adds appropriate pebbles. After the mixed similar materials are heated evenly, they are poured into the test box in multiple molds. After the model is cooled, a horseshoe-shaped tunnel model is obtained. Referring to the geological statistical crack distribution parameters, the simulated rock mass is cut, and the tunnel cracks are simulated by cutting gaps. In order to more realistically simulate the surface morphology of the tunnel cracks, auxiliary materials such as coal powder, mica crushed, perlite and foaming agent are applied to the cut surface of the simulated rock mass to simulate the rough and irregular surface of the joints and cracks. The surface roughness and gap width of the crack surface can be simulated by adjusting the proportion of the above materials.

[0055] As a specific embodiment, the grouting simulation system includes a grouting test operating table, a grouting bucket, slurry and pigment. The diffusivity of the slurry is mainly affected by the following factors: the medium of the rock mass itself; the crack opening, the surface roughness of the crack, the length of the crack; the porosity of the porous medium; the slurry concentration and the grouting pressure. The permeability of the medium of the rock mass itself can be adjusted by adjusting the ratio of river sand to paraffin and the number of pebbles. The length and opening of the cracks are set based on the geostatistical method and with reference to multiple tunnel projects. The crack opening and crack length can be adjusted by adjusting the model cutting gap. The surface roughness of the cracks can be adjusted by changing the ratio of auxiliary materials such as coal powder, mica, perlite and foaming agent. The ratio of river sand to pebbles is increased to simulate the porous medium to adjust its porosity, and the slurry concentration and grouting pressure are continuously adjusted. By adding different pigments to different crack slurries, the diffusion of slurries in different cracks can be clearly distinguished. By changing the above parameters, the diffusion state of different slurries through tunnel cracks under the influence of multiple factors can be obtained.

[0056] As a specific embodiment, the grouting diffusion morphology quantification system cuts the horseshoe-shaped tunnel simulation device after grouting, simulates the process of tunnel face excavation exposure, takes photos to record the diffusion state of the slurry (including color) on the face surface, and performs grid division processing on the taken photos. Figure 3 , select the appropriate grid size according to the size of the crack to facilitate subsequent data processing and analysis. Each grid represents a small area, the grid with one color slurry diffusion is recorded as 1, the grid with another color slurry diffusion is recorded as 2, and so on, the grid without slurry diffusion is recorded as 0, record the vertical position of the slice and the slurry diffusion morphology, that is, digitally characterize the grid, obtain a dense data set of slurry diffusion morphology at multiple vertical positions, and establish a quantitative data set of slurry diffusion morphology.

[0057] As a specific embodiment, the grouting diffusion morphology prediction system includes a model construction module, a model training module and a morphology prediction module; the model construction module is used to construct a multimodal space-time fusion deep prediction model based on convolutional neural networks, long short-term memory networks, graph neural networks and space-time convolutional networks; the model training module is used to train the multimodal space-time fusion deep prediction model in combination with fluid mechanics and permeability constraints in physical modeling; the morphology prediction module is used to obtain the slurry diffusion morphology at each moment and each position in the future tunnel grouting process based on the trained multimodal space-time fusion deep prediction model.

[0058] Furthermore, the morphology prediction module includes a data input unit, a feature extraction unit, a feature fusion unit and a morphology prediction unit; the data input unit is used to input the factors affecting slurry diffusion and the quantitative data set of slurry diffusion morphology into the trained multimodal spatiotemporal fusion deep prediction model; the feature extraction unit is used to extract the geometric characteristics of the rock mass based on the convolutional neural network, extract the dynamic characteristics of slurry diffusion based on the long short-term memory network, and extract the fracture correlation based on the graph neural network; the feature fusion unit is used to fuse the features output by the feature extraction unit based on a fusion mechanism to obtain fused features; the morphology prediction unit is used to predict the slurry diffusion morphology based on the fused features and in combination with the spatiotemporal features extracted by the spatiotemporal convolutional network.

[0059] As an implementable method, the grouting diffusion morphology prediction system realizes the prediction of future grouting morphology by establishing a multi-modal spatiotemporal fusion depth prediction model. The prediction model compares the horseshoe tunneling depth to the "time" in the spatiotemporal sequence, such as Figure 4 As shown in the figure, the grouting diffusion morphology of the tunnel face at different depths is used as time series data. Based on the spatiotemporal series data, a prediction model for the grouting diffusion morphology is established. By adjusting the parameters and input data of the model, the model is repeatedly trained and optimized, and finally the model can accurately predict the slurry diffusion morphology under different construction conditions. The input data includes multiple parameters that affect the slurry diffusion, such as crack opening, surface roughness, crack length, porosity, slurry concentration, grouting pressure, etc. The final output of the prediction system is the future slurry diffusion morphology, which can provide a scientific basis for construction decisions in tunnel engineering.

[0060] Furthermore, the multimodal spatiotemporal fusion deep prediction model proposed in this embodiment integrates multimodal data and spatiotemporal information, and utilizes convolutional neural networks (CNN), long short-term memory networks (LSTM), graph neural networks (GNN) and other technologies in deep learning, combined with physical modeling methods, to achieve accurate prediction of tunnel grouting diffusion morphology.

[0061] Multimodal data input includes: Geometric attribute input: including physical attribute data such as the length, opening, surface roughness, and porosity of the crack. Environmental attribute input: including dynamic environmental parameters such as grouting pressure, slurry concentration, and temperature during construction. Spatiotemporal attribute input: including spatiotemporal information such as tunnel excavation depth and grouting time.

[0062] Then, convolutional neural networks (CNNs) are used to extract complex geometric features of cracks and rock surfaces, such as the distribution of surface roughness and gap width. Long short-term memory networks (LSTMs) are used to extract the dynamic characteristics of slurry diffusion during grouting, especially to capture long-term dependencies in time series. Graph neural networks (GNNs) are used to analyze the spatial correlation and mutual influence between cracks, and to optimize the path and morphology of slurry diffusion in the crack network.

[0063] The outputs from different feature extraction modules are integrated through a fusion mechanism. The attention mechanism can be used to dynamically adjust the weights of different features to enhance the model's perception of key features. Finally, the spatiotemporal attributes are input into the spatiotemporal convolutional network, and combined with the previously extracted features, the spatiotemporal prediction of the grouting diffusion morphology is achieved.

[0064] Furthermore, during the model training process, fluid mechanics and permeability constraints in physical modeling are added to ensure that the prediction results conform to physical laws. This module can be implemented by solving the Navier-Stokes equations or Darcy's law.

[0065] Furthermore, the final output of the model is the slurry diffusion pattern at each moment and each position in the future tunnel grouting process. Combining actual engineering data and prediction results, the model parameters are continuously optimized to make the prediction results more consistent with actual engineering conditions.

[0066] The present invention proposes a tunnel fissure-porous medium grouting diffusion visualization device, which includes a horseshoe-shaped tunnel simulation device and a grouting simulation system. Through the preparation of simulated rock mass and the adjustment of the grouting simulation system, the influence of multiple factors such as rock mass medium properties, fissure characteristics and slurry injection parameters on the grouting diffusion morphology can be considered at the same time, thereby more realistically simulating the actual engineering situation.

[0067] The present invention proposes a grouting diffusion morphology quantification system, which achieves accurate quantification of grout diffusion morphology by cutting, photographing and recording, and meshing the horseshoe-shaped tunnel simulation device. Compared with traditional methods, the system can more accurately record the diffusion state of grout on the tunnel face surface and generate a digital diffusion morphology data set, providing a more reliable basis for subsequent data processing and analysis.

[0068] The present invention proposes a grouting diffusion morphology prediction system, which realizes the prediction of grouting diffusion morphology by establishing a prediction model and using a spatiotemporal series data set. The system can predict the grouting diffusion morphology of the face at different depths based on different slurry influencing parameters, providing an effective prediction tool for the design and practice of underground engineering. By combining data in multiple modes such as geometry, dynamics, space, and time and space, the model's ability to predict slurry diffusion morphology in complex scenarios is improved; the spatiotemporal convolutional network in deep learning is combined with traditional physical constraints to ensure that the model not only has high prediction accuracy, but also follows physical laws; through an adaptive feedback mechanism, the model can continuously adjust and optimize itself to adapt to changes under different construction conditions.

[0069] On the other hand, based on the same inventive concept as the above embodiment, this embodiment further provides a method for predicting the morphology of tunnel grouting diffusion. The prediction method and the system for predicting the morphology of tunnel grouting diffusion provided by the above embodiment can be mutually referenced in effect. The prediction method includes the following steps:

[0070] (1) According to the actual project, a horseshoe-shaped tunnel simulation device was established, the ratio of river sand and paraffin was adjusted, and a horseshoe-shaped simulated tunnel was established. Coal powder, mica crushed, and perlite were used to simulate the roughness of the fracture surface;

[0071] (2) preparing grouting slurries of different viscosities and adding pigments, and then using a grouting simulation system to perform grouting on a horseshoe-shaped tunnel simulation device, and the grouting pressure can be adjusted;

[0072] (3) Slice the horseshoe-shaped tunnel simulation device and take photos of the slices;

[0073] (4) Quantify the slurry diffusion morphology using grid division and establish a quantitative data set of slurry diffusion morphology;

[0074] (5) The quantitative data set of slurry diffusion morphology is used as the spatiotemporal series data set of grouting diffusion morphology, and a neural network model is used to predict it;

[0075] (6) Use engineering excavation data to perform transfer learning on the spatiotemporal prediction neural network model of grouting diffusion morphology.

[0076] The process of meshing the slice surface at each depth position, recording the slurry diffusion morphology, and obtaining a quantitative data set of the slurry diffusion morphology includes:

[0077] (1) Mesh division: During the cutting and exposure process of the horseshoe-shaped tunnel simulation device after grouting, the tunnel face surface is divided into grids. The size of the grid is selected according to the size of the cracks to facilitate subsequent fine data processing.

[0078] (2) Color coding and numerical representation: The photographs taken are processed and the diffusion of the slurry in each grid is coded with different numerical values. For example, different colors of slurry are represented by 1, 2, etc., and the area without diffusion is recorded as 0.

[0079] (3) Slicing at depth: The slurry diffusion patterns at multiple depths are obtained by slicing the model. The slurry diffusion data at each depth is obtained by photographing and meshing.

[0080] (4) Establishment of quantitative data set: By recording the grid code at each slice, a complete quantitative data set of slurry diffusion morphology is finally established. This data set can be used for subsequent analysis and training of prediction models.

[0081] The feasible process of predicting the future slurry diffusion morphology based on the multimodal spatiotemporal fusion deep prediction model includes:

[0082] Based on convolutional neural networks, long short-term memory networks, graph neural networks, and spatiotemporal convolutional networks, a multimodal spatiotemporal fusion deep prediction model is constructed;

[0083] Input the factors affecting slurry diffusion and the quantitative data set of slurry diffusion morphology into the trained multimodal spatiotemporal fusion depth prediction model, extract the geometric characteristics of the rock mass based on the convolutional neural network, extract the dynamic characteristics of slurry diffusion based on the long short-term memory network, and extract the fracture correlation based on the graph neural network;

[0084] Based on the fusion mechanism, the geometric characteristics of the rock mass, the dynamic characteristics of slurry diffusion and the fracture correlation are fused to obtain fusion features. Based on the fusion features and combined with the spatiotemporal features extracted by the spatiotemporal convolutional network, the slurry diffusion morphology is predicted.

[0085] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A tunnel grouting diffusion morphology prediction system, characterized in that: include: Grouting diffusion visualization system, grouting diffusion morphology quantification system and grouting diffusion morphology prediction system; The grouting diffusion visualization system includes a grouting simulation system and a tunnel simulation device; Wherein, the tunnel simulation device is used to prepare a tunnel model with cracks to simulate the tunnel rock mass; The grouting simulation system is used to perform grouting on the tunnel simulation device by adjusting the factors affecting slurry diffusion, so as to obtain the slurry diffusion state under the influence of multiple factors; The grouting diffusion morphology quantification system is used to cut the tunnel simulation device after grouting to obtain slices at several depth positions, and to grid the slice surface at each depth position, record the slurry diffusion morphology, and obtain a slurry diffusion morphology quantification data set; The grouting diffusion morphology prediction system is used to construct a multimodal spatiotemporal fusion depth prediction model, and the slurry diffusion morphology quantization data set is input into the multimodal spatiotemporal fusion depth prediction model to predict the future slurry diffusion morphology.

2. The tunnel grouting diffusion morphology prediction system according to claim 1, characterized in that: The preparation process of the tunnel simulation device includes: mixing river sand, paraffin and pebbles according to a preset proportion and casting a mold to obtain a tunnel model; cutting the tunnel model based on geological statistical crack distribution parameters to obtain simulated tunnel cracks; mixing coal powder, mica powder, perlite and a foaming agent and applying them to the cut surface of the tunnel model to obtain a simulated tunnel crack surface, thereby completing the preparation of the tunnel simulation device.

3. The tunnel grouting diffusion morphology prediction system according to claim 2, characterized in that: The tunnel model is horseshoe-shaped.

4. The tunnel grouting diffusion morphology prediction system according to claim 1, characterized in that: The factors affecting slurry diffusion include the medium of the tunnel model itself, the crack opening, the crack length, the crack surface roughness, the medium porosity, the slurry concentration and the grouting pressure.

5. The tunnel grouting diffusion morphology prediction system according to claim 1, characterized in that: The grouting simulation system includes a grouting test operating table, a grouting bucket, slurry and pigments, and is used to grout the tunnel model on the grouting test operating table based on the grouting bucket, and add different pigments to different fracture slurries.

6. The tunnel grouting diffusion morphology prediction system according to claim 1, characterized in that: The grouting diffusion morphology prediction system includes a model building module, a model training module and a morphology prediction module; The model building module is used to build a multimodal spatiotemporal fusion deep prediction model based on convolutional neural networks, long short-term memory networks, graph neural networks, and spatiotemporal convolutional networks; The model training module is used to train the multimodal spatiotemporal fusion depth prediction model in combination with fluid mechanics and permeability constraints in physical modeling; The morphology prediction module is used to obtain the slurry diffusion morphology at each moment and each position in the future tunnel grouting process based on the trained multi-modal spatiotemporal fusion depth prediction model.

7. The tunnel grouting diffusion morphology prediction system according to claim 6, characterized in that: The morphology prediction module includes a data input unit, a feature extraction unit, a feature fusion unit and a morphology prediction unit; The data input unit is used to input the slurry diffusion influencing factors and the slurry diffusion morphology quantitative data set into the trained multi-modal spatiotemporal fusion depth prediction model; The feature extraction unit is used to extract geometric features of the rock mass based on the convolutional neural network, extract dynamic features of slurry diffusion based on the long short-term memory network, and extract fracture correlation based on the graph neural network; The feature fusion unit is used to fuse the features output by the feature extraction unit based on a fusion mechanism to obtain fused features; The morphology prediction unit is used to predict the slurry diffusion morphology based on the fusion features and in combination with the spatiotemporal features extracted by the spatiotemporal convolutional network.

8. A method for predicting the morphology of tunnel grouting diffusion, characterized in that: The morphology prediction system for tunnel grouting diffusion according to any one of claims 1 to 7 comprises the following steps: preparing a tunnel simulation device, and performing grouting on the tunnel simulation device based on a grouting simulation system; Cut the tunnel simulation device after grouting to obtain slices at several depth positions; Gridding the slice surface at each depth position, recording the slurry diffusion morphology, and obtaining a quantitative data set of the slurry diffusion morphology; A multimodal spatiotemporal fusion deep prediction model is constructed, and the slurry diffusion morphology quantification data set is input into the multimodal spatiotemporal fusion deep prediction model to predict the future slurry diffusion morphology.

9. The method for predicting the morphology of tunnel grouting diffusion according to claim 8, characterized in that: The process of meshing the slice surface at each depth position, recording the slurry diffusion morphology, and obtaining a quantitative data set of the slurry diffusion morphology includes: The slurry diffusion conditions in each grid are encoded with different numerical values. By recording the grid code at each slice, the slurry diffusion data at each depth position is obtained, and the slurry diffusion data set at each depth position constitutes a complete slurry diffusion morphology quantitative data set.

10. The method for predicting the morphology of tunnel grouting diffusion according to claim 8, characterized in that: A multimodal spatiotemporal fusion depth prediction model is constructed based on the slurry diffusion morphology quantification data set, and a process of predicting the future slurry diffusion morphology based on the multimodal spatiotemporal fusion depth prediction model includes: Based on convolutional neural networks, long short-term memory networks, graph neural networks, and spatiotemporal convolutional networks, a multimodal spatiotemporal fusion deep prediction model is constructed; Input the factors affecting slurry diffusion and the quantitative data set of slurry diffusion morphology into the trained multimodal spatiotemporal fusion depth prediction model, extract the geometric characteristics of the rock mass based on the convolutional neural network, extract the dynamic characteristics of slurry diffusion based on the long short-term memory network, and extract the fracture correlation based on the graph neural network; Based on the fusion mechanism, the geometric characteristics of the rock mass, the dynamic characteristics of slurry diffusion and the fracture correlation are fused to obtain fusion features. Based on the fusion features and combined with the spatiotemporal features extracted by the spatiotemporal convolutional network, the slurry diffusion morphology is predicted.

Images