Shield tunnel stratum detection muqiu precise positioning method and system

Abstract

The application provides a shield tunnel stratum detection muzi accurate positioning method and system, and relates to the technical field of shield tunnel stratum detection.The method comprises the following steps: obtaining a high-definition multispectral image of strata around a shield tunnel; determining a plurality of to-be-evaluated frost heaving deformation risk stratum blocks based on the high-definition multispectral image of the strata around the shield tunnel; determining a plurality of joint frost heaving test stratum block groups and the frost heaving test temperature of each joint frost heaving test stratum block group based on the stratum frost heaving deformation distribution data corresponding to each frost heaving test stratum block; and determining a frost heaving deformation high-significant stratum block of the strata around the shield tunnel based on the stratum frost heaving deformation distribution data corresponding to each frost heaving test stratum block and the stratum frost heaving deformation distribution data corresponding to each joint frost heaving test stratum block group.The method can accurately identify the frost heaving deformation risk of the strata around the shield tunnel and determine the frost heaving deformation high-significant stratum block.

Description

Technical Field

[0001] This invention relates to the field of shield tunnel geological detection technology, specifically to a method and system for precise positioning of muons in shield tunnel geological detection. Background Technology

[0002] When shield tunnels traverse cold regions or seasonally frozen soil areas, the surrounding strata are susceptible to varying degrees of frost heave deformation due to temperature changes, groundwater conditions, and differences in geological structure. Therefore, accurately identifying strata with high frost heave risk is crucial for tunnel construction and operational safety. Existing shield tunnel frost heave risk analysis typically combines geological survey data, on-site temperature and deformation monitoring results, and indoor freeze-thaw tests to assess the frost heave characteristics of different strata. However, the strata surrounding shield tunnels exhibit significant spatial heterogeneity, with substantial differences in water content, fissure development, and internal structure across different areas. Freeze-thaw test results from limited areas cannot fully represent the frost heave response of other strata. Furthermore, the degree of frost heave is highly sensitive to temperature conditions, and different test temperatures may result in significantly different deformation patterns. Results obtained using limited test conditions may fail to fully represent the frost heave characteristics of some high-risk strata. In addition, existing analyses often rely on separate risk assessments based on local test and monitoring results, which is easily influenced by the selection of test areas and conditions, leading to inaccurate identification of key risk areas with significant frost heave deformation.

[0003] Therefore, accurately identifying the risk of frost heave deformation in the strata surrounding shield tunnels and determining strata blocks with high frost heave deformation is an urgent problem to be solved. Summary of the Invention

[0004] The main technical problem addressed by this invention is how to accurately identify the risk of frost heave deformation in the strata surrounding a shield tunnel and determine the strata blocks with high frost heave deformation.

[0005] According to a first aspect, the present invention provides a method for precise positioning of muons in the strata of a shield tunnel, comprising: acquiring high-resolution multispectral images of the strata surrounding the shield tunnel; determining multiple strata blocks with frost heave deformation risk to be assessed based on the high-resolution multispectral images of the strata surrounding the shield tunnel; acquiring muon penetration detection data for each strata block with frost heave deformation risk to be assessed; determining multiple freeze-thaw test strata blocks and the freeze-thaw test temperature for each freeze-thaw test strata block based on the high-resolution multispectral images of each strata block with frost heave deformation risk to be assessed and the muon penetration detection data for each strata block with frost heave deformation risk to be assessed; and performing freeze-thaw cycle tests on each freeze-thaw test strata block based on the freeze-thaw test temperature of each freeze-thaw test strata block, and acquiring the strata corresponding to each freeze-thaw test strata block. Frost heave deformation distribution data; based on the frost heave deformation distribution data corresponding to each freeze-thaw test stratum block, determine multiple joint freeze-thaw test stratum block groups and the freeze-thaw test temperature of each joint freeze-thaw test stratum block group, wherein each joint freeze-thaw test stratum block group includes at least two freeze-thaw test stratum blocks; based on the freeze-thaw test temperature of each joint freeze-thaw test stratum block group, conduct freeze-thaw cycle tests on each joint freeze-thaw test stratum block group respectively, and obtain the frost heave deformation distribution data corresponding to each joint freeze-thaw test stratum block group; based on the frost heave deformation distribution data corresponding to each freeze-thaw test stratum block and the frost heave deformation distribution data corresponding to each joint freeze-thaw test stratum block group, determine the stratum blocks with high frost heave deformation around the shield tunnel.

[0006] In one possible implementation, determining the strata blocks with highly significant frost heave deformation around the shield tunnel based on the strata frost heave deformation distribution data corresponding to each freeze-thaw test strata block and the strata frost heave deformation distribution data corresponding to each combined freeze-thaw test strata block group includes: generating simulated frost heave deformation evolution of the strata around the shield tunnel under various extreme freeze-thaw test temperatures based on high-resolution multispectral images of the strata around the shield tunnel, the strata frost heave deformation distribution data corresponding to each freeze-thaw test strata block and the strata frost heave deformation distribution data corresponding to each combined freeze-thaw test strata block group. Data; Based on simulated frost heave deformation evolution data of the strata surrounding the shield tunnel under various extreme freeze-thaw test temperatures, information on multiple key strata frost heave deformation points under each extreme freeze-thaw test temperature is determined; a frost heave deformation evolution map is constructed, which includes multiple extreme freeze-thaw test nodes and multiple edges between the nodes, and the node characteristics of each extreme freeze-thaw test node are information on multiple key strata frost heave deformation points under each extreme freeze-thaw test temperature; the frost heave deformation evolution map is processed based on a significant block determination model to determine the strata blocks with high saliency in frost heave deformation around the shield tunnel.

[0007] In one possible implementation, determining multiple joint freeze-thaw test stratigraphic block groups and the freeze-thaw test temperature of each joint freeze-thaw test stratigraphic block group based on the formation frost heave deformation distribution data corresponding to each freeze-thaw test stratigraphic block, wherein each joint freeze-thaw test stratigraphic block group includes at least two freeze-thaw test stratigraphic blocks, comprising: clustering the formation frost heave deformation distribution data corresponding to each freeze-thaw test stratigraphic block to obtain multiple frost heave deformation response clusters; determining multiple frost heave deformation response transmission lines based on the multiple frost heave deformation response clusters and the multiple frost heave deformation response transmission lines; and determining the multiple joint freeze-thaw test stratigraphic block groups and the freeze-thaw test temperature of each joint freeze-thaw test stratigraphic block group based on the multiple frost heave deformation response clusters and the multiple frost heave deformation response transmission lines.

[0008] In one possible implementation, the step of clustering the frost heave deformation distribution data corresponding to each freeze-thaw test stratum block to obtain multiple frost heave deformation response clusters includes: using a clustering algorithm to cluster the frost heave deformation distribution data corresponding to each freeze-thaw test stratum block to obtain multiple frost heave deformation response clusters.

[0009] According to a second aspect, the present invention provides a precise muon positioning system for stratum detection in shield tunnels, comprising: an image acquisition module for acquiring high-resolution multispectral images of the strata surrounding the shield tunnel; a risk block determination module for determining multiple stratum blocks at risk of frost heave deformation to be assessed based on the high-resolution multispectral images of the strata surrounding the shield tunnel; a detection data acquisition module for acquiring muon penetration detection data for each stratum block at risk of frost heave deformation to be assessed; a single-unit test determination module for determining multiple freeze-thaw test stratum blocks and the freeze-thaw test temperature of each freeze-thaw test stratum block based on the high-resolution multispectral images of each stratum block at risk of frost heave deformation to be assessed and the muon penetration detection data of each stratum block at risk of frost heave deformation to be assessed; and a single-unit acquisition module for performing freeze-thaw cycle tests on each freeze-thaw test stratum block based on the freeze-thaw test temperature of each freeze-thaw test stratum block, and acquiring the temperature of each freeze-thaw test stratum block. The test stratum blocks include: a stratum frost heave deformation distribution data corresponding to each test stratum block; a joint test determination module, used to determine multiple joint freeze-thaw test stratum block groups and the freeze-thaw test temperature of each joint freeze-thaw test stratum block group based on the stratum frost heave deformation distribution data corresponding to each freeze-thaw test stratum block group, wherein each joint freeze-thaw test stratum block group includes at least two freeze-thaw test stratum blocks; a joint acquisition module, used to perform freeze-thaw cycle tests on each joint freeze-thaw test stratum block group based on the freeze-thaw test temperature of each joint freeze-thaw test stratum block group, and acquire the stratum frost heave deformation distribution data corresponding to each joint freeze-thaw test stratum block group; and a significant block determination module, used to determine the stratum blocks with highly significant frost heave deformation around the shield tunnel based on the stratum frost heave deformation distribution data corresponding to each freeze-thaw test stratum block and the stratum frost heave deformation distribution data corresponding to each joint freeze-thaw test stratum block group.

[0010] In one possible implementation, the salient block determination module is further configured to: generate simulated frost heave deformation evolution data of the strata surrounding the shield tunnel under multiple extreme freeze-thaw test temperatures based on high-resolution multispectral images of the strata surrounding the shield tunnel, frost heave deformation distribution data corresponding to each freeze-thaw test stratum block, and frost heave deformation distribution data corresponding to each joint freeze-thaw test stratum block group; determine multiple key stratum frost heave deformation point information under each extreme freeze-thaw test temperature based on the simulated frost heave deformation evolution data of the strata surrounding the shield tunnel under multiple extreme freeze-thaw test temperatures; construct a frost heave deformation evolution map, the frost heave deformation evolution map including multiple extreme freeze-thaw test nodes and multiple edges between multiple nodes, the node characteristics of each extreme freeze-thaw test node being multiple key stratum frost heave deformation point information under each extreme freeze-thaw test temperature; and process the frost heave deformation evolution map based on the salient block determination model to determine the strata blocks with highly significant frost heave deformation of the strata surrounding the shield tunnel.

[0011] In one possible implementation, the joint test determination module is further configured to: cluster the frost heave deformation distribution data corresponding to each freeze-thaw test stratigraphic block to obtain multiple frost heave deformation response clusters; determine multiple frost heave deformation response transmission lines based on the multiple frost heave deformation response clusters and the multiple frost heave deformation response transmission lines; and determine multiple joint freeze-thaw test stratigraphic block groups and the freeze-thaw test temperature of each joint freeze-thaw test stratigraphic block group based on the multiple frost heave deformation response clusters and the multiple frost heave deformation response transmission lines.

[0012] In one possible implementation, the step of clustering the frost heave deformation distribution data corresponding to each freeze-thaw test stratum block to obtain multiple frost heave deformation response clusters includes: using a clustering algorithm to cluster the frost heave deformation distribution data corresponding to each freeze-thaw test stratum block to obtain multiple frost heave deformation response clusters.

[0013] According to a third aspect, embodiments of the present invention provide an electronic device, including: a processor; a memory; and a computer program; wherein the computer program is stored in the memory and configured to be executed by the processor to implement the method described above, the method including: acquiring high-resolution multispectral images of strata surrounding a shield tunnel; determining multiple strata blocks to be assessed for frost heave deformation risk based on the high-resolution multispectral images of the strata surrounding the shield tunnel; acquiring muon penetration detection data for each strata block to be assessed for frost heave deformation risk; determining multiple freeze-thaw test strata blocks and a freeze-thaw test temperature for each freeze-thaw test strata block based on the high-resolution multispectral images of each strata block to be assessed for frost heave deformation risk and the muon penetration detection data of each strata block to be assessed for frost heave deformation risk; and performing freeze-thaw test tests on each freeze-thaw test strata block based on the freeze-thaw test temperature of each freeze-thaw test strata block. Freeze-thaw cycle tests are conducted on the blocks, and the distribution data of frost heave deformation corresponding to each freeze-thaw test stratum block is obtained. Based on the distribution data of frost heave deformation corresponding to each freeze-thaw test stratum block, multiple joint freeze-thaw test stratum block groups and the freeze-thaw test temperature of each joint freeze-thaw test stratum block group are determined. Each joint freeze-thaw test stratum block group includes at least two freeze-thaw test stratum blocks. Based on the freeze-thaw test temperature of each joint freeze-thaw test stratum block group, freeze-thaw cycle tests are conducted on each joint freeze-thaw test stratum block group, and the distribution data of frost heave deformation corresponding to each joint freeze-thaw test stratum block group is obtained. Based on the distribution data of frost heave deformation corresponding to each freeze-thaw test stratum block and the distribution data of frost heave deformation corresponding to each joint freeze-thaw test stratum block group, stratum blocks with high frost heave deformation around the shield tunnel are determined.

[0014] According to the fourth aspect, this embodiment provides a computer-readable storage medium storing a computer program thereon. When executed by a processor, the program implements the aforementioned method for precise positioning of muons in shield tunnel strata detection. The method includes: acquiring high-resolution multispectral images of the strata surrounding the shield tunnel; determining multiple strata blocks with frost heave deformation risk to be assessed based on the high-resolution multispectral images of the strata surrounding the shield tunnel; acquiring muon penetration detection data for each strata block with frost heave deformation risk to be assessed; determining multiple freeze-thaw test strata blocks and a freeze-thaw test temperature for each freeze-thaw test strata block based on the high-resolution multispectral images of each strata block with frost heave deformation risk to be assessed and the muon penetration detection data of each strata block with frost heave deformation risk to be assessed; and performing freeze-thaw cycles on each freeze-thaw test strata block based on the freeze-thaw test temperature of each freeze-thaw test strata block. The process involves: conducting a freeze-thaw cycle test and acquiring frost heave deformation distribution data for each freeze-thaw test stratum block; determining multiple joint freeze-thaw test stratum block groups and the freeze-thaw test temperature for each joint freeze-thaw test stratum block group, wherein each joint freeze-thaw test stratum block group includes at least two freeze-thaw test stratum blocks; conducting freeze-thaw cycle tests on each joint freeze-thaw test stratum block group based on the freeze-thaw test temperature, and acquiring frost heave deformation distribution data for each joint freeze-thaw test stratum block group; and identifying stratum blocks with high frost heave deformation around the shield tunnel based on the frost heave deformation distribution data of each freeze-thaw test stratum block and the frost heave deformation distribution data of each joint freeze-thaw test stratum block group.

[0015] The present invention has at least the following beneficial effects:

[0016] 1. This invention combines high-definition multispectral images of the strata surrounding the shield tunnel with muon penetration detection data to determine suitable freeze-thaw test strata blocks and corresponding test temperatures, so that the freeze-thaw test conditions can match the actual conditions of different strata areas, enhancing the pertinence of frost heave risk analysis.

[0017] 2. This invention combines the results of individual freeze-thaw tests with the spatial correlation of frost heave deformation between different stratigraphic blocks to determine the stratigraphic block group for joint freeze-thaw tests. This allows for further observation of the deformation performance of multiple related stratigraphic blocks when subjected to freeze-thaw action, making the analysis of stratigraphic frost heave response more comprehensive.

[0018] 3. This invention integrates the frost heave deformation results obtained under different test conditions to further screen the strata around the shield tunnel and identify strata blocks with high frost heave deformation, thereby achieving precise positioning of key frost heave risk areas and providing a basis for risk monitoring during shield tunnel construction. Attached Figure Description

[0019] Figure 1 A flowchart illustrating a method for precise positioning of muons in shield tunnel strata detection, provided as an embodiment of the present invention;

[0020] Figure 2 A typical cross-sectional schematic diagram of a geological block around a shield tunnel whose frost heave deformation risk needs to be assessed is provided for an embodiment of the present invention.

[0021] Figure 3 This is a schematic diagram showing the distribution of strata blocks around a shield tunnel that are subject to frost heave deformation risk assessment, provided by an embodiment of the present invention.

[0022] Figure 4 A frost heave tendency index distribution map of a stratigraphic block to be assessed for frost heave deformation risk is provided in an embodiment of the present invention.

[0023] Figure 5 A schematic diagram of the estimated phase change temperature range and frost heave temperature sensitivity index of a freeze-thaw test formation block provided in an embodiment of the present invention;

[0024] Figure 6 This is a schematic flowchart illustrating a process for determining multiple joint freeze-thaw test formation blocks and the freeze-thaw test temperature of each joint freeze-thaw test formation block group, as provided in an embodiment of the present invention.

[0025] Figure 7 A schematic diagram of a process for determining a stratum block with significant frost heave deformation around a shield tunnel, provided as an embodiment of the present invention;

[0026] Figure 8 This is a schematic diagram of a shield tunnel stratum detection muon precise positioning system provided in an embodiment of the present invention. Detailed Implementation

[0027] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of the invention. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to the present invention are not shown or described in the specification. This is to avoid obscuring the core parts of the invention with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0028] In this embodiment of the invention, the following are provided: Figure 1The method for precise positioning of muons in shield tunnel strata detection includes steps S1 to S8:

[0029] Step S1: Obtain high-resolution multispectral images of the strata surrounding the shield tunnel.

[0030] High-resolution multispectral images of the strata surrounding a shield tunnel are obtained by segmenting the internal structure of the strata using multispectral imaging equipment deployed in the strata detection holes around the shield tunnel, and then stitching the images together according to the acquisition locations.

[0031] High-resolution multispectral images of the strata surrounding a shield tunnel can reveal the color, texture, fracture boundaries, water-bearing area boundaries, and pixel reflectance intensity of the strata in different spectral bands.

[0032] In some embodiments, multiple stratum detection holes can be deployed around the shield tunnel, and multispectral imaging equipment can be sequentially sent into different depth positions to perform imaging, so that the local multispectral images obtained at different acquisition positions have corresponding spatial coordinates. Then, the images are stitched together according to the spatial coordinates to obtain a high-definition multispectral image covering the stratum detection range around the shield tunnel.

[0033] Step S2: Based on the high-resolution multispectral images of the strata surrounding the shield tunnel, identify multiple strata blocks with frost heave deformation risk to be assessed.

[0034] In some embodiments, determining multiple strata blocks at risk of frost heave deformation based on high-resolution multispectral images of the strata surrounding the shield tunnel includes: using a risk block determination model based on the high-resolution multispectral images of the strata surrounding the shield tunnel to determine multiple strata blocks at risk of frost heave deformation. The risk block determination model is a convolutional neural network. The input to the risk block determination model is the high-resolution multispectral image of the strata surrounding the shield tunnel, and the output of the risk block determination model is the multiple strata blocks at risk of frost heave deformation to be assessed.

[0035] A convolutional neural network (CNN) is a deep learning model that includes multiple layers of computational structure. A CNN consists of an input layer, convolutional layers, pooling layers, and fully connected layers stacked sequentially. CNNs can efficiently process image-like data and accurately identify specific spatial locations.

[0036] Multiple geological blocks with potential frost heave deformation risk were selected from the strata surrounding the shield tunnel using a risk block determination model. These blocks represent areas with potential frost heave deformation risk that require further assessment.

[0037] In some embodiments, Figure 2This invention provides a schematic diagram of a typical cross-section of a geological block surrounding a shield tunnel where the risk of frost heave deformation to be assessed. Figure 3 This is a schematic diagram showing the distribution of strata blocks around a shield tunnel that are subject to assessment for frost heave deformation risk, provided as an embodiment of the present invention.

[0038] Reference Figure 3 , Figure 3 This diagram illustrates the spatial distribution of geological blocks around a tunnel with assessed frost heave deformation risk, both longitudinally and circumferentially. The horizontal axis represents the tunnel's alignment, while the longitudinal axis represents the circumferential location of the surrounding geological strata. The diagram includes multiple geological blocks with assessed frost heave deformation risk, distributed at different longitudinal and circumferential locations within the tunnel, representing the spatial range of each block relative to the tunnel.

[0039] High-resolution multispectral images record the differences in reflectance of strata across different spectral bands. Areas within the strata with high water content, looser structure, and more developed fractures exhibit anomalous responses in specific bands of the multispectral images. These areas have a higher risk of frost heave deformation under temperature changes. Convolutional neural networks can use the spectral response and texture differences reflected in the multispectral images to filter out strata blocks around shield tunnels that pose a risk of frost heave deformation.

[0040] Convolutional neural networks (CNNs) can effectively extract local spatial features and multi-channel band information from images, and combine this local spatial information to process different image regions. By analyzing the image differences in different regions of high-resolution multispectral images of the strata surrounding a shield tunnel, CNNs can identify multiple strata blocks within the strata surrounding the shield tunnel that require further assessment of frost heave deformation, and designate these corresponding strata blocks as multiple strata blocks to be assessed for frost heave deformation risk.

[0041] Step S3: Obtain muon penetration detection data for each stratigraphic block to be assessed for frost heave deformation risk.

[0042] The muon penetration data for each stratigraphic block under assessment of frost heave deformation risk is obtained by penetrating the block with a muon detector and performing density inversion on the results. This data characterizes the internal density state of the corresponding stratigraphic block. The muon penetration data for each block includes the average stratigraphic density value, stratigraphic density values ​​at different spatial locations, the extent of density anomaly areas, and the average density value of the density anomaly areas.

[0043] In some embodiments, muon detection devices can be deployed inside the shield tunnel at locations corresponding to each stratum block with frost heave deformation risk to be assessed. The muon detection devices continuously record muons that reach the detection surface after passing through the corresponding stratum block with frost heave deformation risk to be assessed within a set detection time. The density inversion is performed on the collected muon detection results to obtain muon penetration detection data for each stratum block with frost heave deformation risk to be assessed.

[0044] Step S4: Based on the high-resolution multispectral image of each stratum block to be evaluated for frost heave deformation risk and the muon penetration detection data of each stratum block to be evaluated for frost heave deformation risk, determine multiple freeze-thaw test stratum blocks and the freeze-thaw test temperature of each freeze-thaw test stratum block.

[0045] The high-resolution multispectral image of each geological block with frost heave deformation risk to be assessed is the corresponding geological block image extracted from the high-resolution multispectral image of the strata surrounding the shield tunnel, based on the spatial location range of each geological block with frost heave deformation risk to be assessed.

[0046] In some embodiments, a freeze-thaw testing decision model can be used to determine multiple freeze-thaw test stratigraphic blocks and the freeze-thaw test temperature for each block. The freeze-thaw testing decision model is a Transformer model. The inputs to the freeze-thaw testing decision model are high-resolution multispectral images of each stratigraphic block to be assessed for frost heave deformation risk and muon penetration detection data for each block. The outputs of the freeze-thaw testing decision model are the multiple freeze-thaw test stratigraphic blocks and the freeze-thaw test temperature for each block.

[0047] The Transformer model is a deep learning model that uses self-attention computation as its core. It establishes correspondences between different positions in the input content through attention computation and processes the input content step-by-step using a multi-layered encoding structure. This allows input elements that are geographically distant to participate in the computation of the current output, ultimately generating the target output based on the correspondences between the input elements.

[0048] Multiple freeze-thaw test stratigraphic blocks are stratigraphic regions selected from multiple stratigraphic blocks whose frost heave deformation risk is to be assessed, for use in freeze-thaw cycle testing. Each freeze-thaw test stratigraphic block has a defined spatial extent and corresponds to all or part of the stratigraphic region within a stratigraphic block whose frost heave deformation risk is to be assessed.

[0049] The freeze-thaw test temperature for each freeze-thaw test formation block is the target temperature control value used to control the freeze-thaw cycle test conducted in the corresponding freeze-thaw test formation block. Each freeze-thaw test formation block corresponds to a separate freeze-thaw test temperature.

[0050] High-resolution multispectral images of stratigraphic blocks at risk of frost heave deformation can reflect the pixel response, regional texture, and spatial distribution differences of the corresponding stratigraphic blocks in different spectral bands. Meanwhile, muon penetration data of these blocks can reflect the average density, stratigraphic density at different spatial locations, and the distribution of density anomalies. Differences in surface appearance, internal density, and spatial distribution among different stratigraphic blocks will cause the high-resolution multispectral images and muon penetration data to exhibit different response results. The Transformer model, by combining the high-resolution multispectral images and muon penetration data of each stratigraphic block at risk of frost heave deformation, can distinguish the differences between different blocks and determine, based on this, multiple suitable freeze-thaw test stratigraphic blocks for freeze-thaw cycle testing and the freeze-thaw test temperature for each block.

[0051] The Transformer model can simultaneously establish a correspondence between high-resolution multispectral images of each stratigraphic block to be assessed for frost heave deformation risk and muon penetration data of that block using self-attention computation. It assigns different computational weights based on the importance of different inputs to the final judgment. Through multi-layer attention computation, the Transformer model can further learn the differences in multispectral response, internal density state, and the correspondence between different stratigraphic blocks, thereby distinguishing the frost heave deformation risk performance of different stratigraphic blocks. The learned correspondence is then mapped to the selection results of multiple freeze-thaw test stratigraphic blocks and the corresponding freeze-thaw test temperature for each block.

[0052] By determining the freeze-thaw test blocks and their freeze-thaw test temperatures, representative stratigraphic regions can be further selected from the blocks whose frost heave deformation risk is to be assessed. Freeze-thaw test conditions can then be configured for different test blocks to match their specific conditions. This avoids repeatedly testing all blocks with frost heave deformation risk at the same temperature, making subsequent freeze-thaw cycle tests more targeted and expanding the coverage of test results under different stratigraphic conditions and varying degrees of frost heave deformation risk.

[0053] In some embodiments, the determination of multiple freeze-thaw test blocks and the freeze-thaw test temperature of each freeze-thaw test block based on the high-resolution multispectral image of each stratigraphic block to be assessed for frost heave deformation risk and the muon penetration detection data of each freeze-thaw test block includes steps S41 to S43:

[0054] Step S41: Based on the high-resolution multispectral image of each stratigraphic block to be assessed for frost heave deformation risk and the muon penetration detection data of each stratigraphic block to be assessed for frost heave deformation risk, determine the spatial distribution sequence of shallow water-bearing fracture density, the volume ratio of deep low-density loose area, and the distribution map of frost heave tendency index for each stratigraphic block to be assessed for frost heave deformation risk.

[0055] In some embodiments, the Transformer model can be used to determine the spatial distribution sequence of shallow water-bearing fracture density, the volume percentage of deep low-density loose areas, and the distribution map of frost heave tendency index for each stratigraphic block to be assessed for frost heave deformation risk.

[0056] The spatial distribution sequence of shallow water-bearing fracture density for each stratigraphic block under assessment of frost heave deformation risk is a numerical sequence arranged according to the spatial location of the shallow area of ​​the corresponding stratigraphic block under assessment of frost heave deformation risk, used to represent the density of water-bearing fractures at different spatial locations. Different sequence positions in the spatial distribution sequence of shallow water-bearing fracture density correspond to different spatial locations within the shallow area of ​​the stratigraphic block under assessment of frost heave deformation risk, and each sequence position records the water-bearing fracture density value at that spatial location.

[0057] The percentage of deep low-density loose area volume in each stratigraphic block under assessment of frost heave deformation risk is a value used to represent the proportion of the total volume of low-density loose area in the deep part of the corresponding stratigraphic block under assessment of frost heave deformation risk to the total volume of the deep strata.

[0058] The frost heave tendency index distribution map for each stratigraphic block under assessment of frost heave deformation risk is used to represent the distribution of frost heave tendency at different locations within the corresponding stratigraphic block. Different locations on the frost heave tendency index distribution map correspond to different locations within the stratigraphic block under assessment of frost heave deformation risk, and each location has a corresponding frost heave tendency index value. A higher frost heave tendency index value indicates a higher tendency for significant frost heave deformation at the corresponding location. In some embodiments, Figure 4 This is a distribution map of the frost heave tendency index of a stratigraphic block to be assessed for frost heave deformation risk, provided as an embodiment of the present invention.

[0059] The Transformer model can jointly analyze high-resolution multispectral images and muon penetration data of each stratigraphic block to be assessed for frost heave deformation risk, and make judgments based on the distribution of water-bearing fractures and the formation density state at different spatial locations. The Transformer model can determine the spatial distribution sequence of shallow water-bearing fracture density for each stratigraphic block to be assessed for frost heave deformation risk based on the distribution differences of shallow water-bearing fractures. Simultaneously, it can determine the volume percentage of deep low-density loose areas based on the spatial extent of deep formation density anomalies, and judge the frost heave tendency at different spatial locations by combining the distribution of shallow water-bearing fractures and the deep low-density state, thereby generating a frost heave tendency index distribution map for each stratigraphic block to be assessed for frost heave deformation risk.

[0060] Step S42: Based on the spatial distribution sequence of shallow water-bearing fracture density, the volume ratio of deep low-density loose areas, and the distribution map of frost heave tendency index of each stratum block to be evaluated for frost heave deformation risk, determine multiple freeze-thaw test stratum blocks, the estimated phase change temperature range of each freeze-thaw test stratum block, and the frost heave temperature sensitivity index of each freeze-thaw test stratum block.

[0061] In some embodiments, the Transformer model can be used to determine multiple freeze-thaw test formation blocks, the estimated phase change temperature range for each freeze-thaw test formation block, and the frost heave temperature sensitivity index for each freeze-thaw test formation block.

[0062] The estimated phase change temperature range of the freeze-thaw test formation block is a temperature range used to represent the temperature range from the initial temperature during which the pore water inside the corresponding freeze-thaw test formation block undergoes a physical phase change from liquid to solid ice to the temperature range where the freezing phase change is basically completed.

[0063] The estimated phase transition temperature range for a freeze-thaw test formation block includes the upper limit of the freezing phase transition initiation temperature and the lower limit of the freezing phase transition essentially completed temperature. For example, the estimated phase transition temperature range for a certain freeze-thaw test formation block is -0.5°C to -2.5°C.

[0064] The frost heave temperature sensitivity index of a freeze-thaw test formation block is a numerical indicator used to quantify the sensitivity of frost heave deformation to temperature changes in the corresponding freeze-thaw test formation block. The larger the frost heave temperature sensitivity index of the freeze-thaw test formation block, the more easily the corresponding freeze-thaw test formation block will produce significant frost heave deformation changes when the temperature changes.

[0065] In some embodiments, Figure 5 This is a schematic diagram illustrating the estimated phase transition temperature range and frost heave temperature sensitivity index of a freeze-thaw test formation block provided in an embodiment of the present invention. (Refer to...) Figure 5 , Figure 5This diagram displays the estimated phase transition temperature ranges and frost heave temperature sensitivity indices for multiple freeze-thaw test stratigraphic blocks. The horizontal axis represents different freeze-thaw test stratigraphic blocks, the left vertical axis represents temperature in °C, and the right vertical axis represents the frost heave temperature sensitivity index. The vertical lines in the diagram represent the estimated phase transition temperature ranges for each freeze-thaw test stratigraphic block. The upper end of the range represents the upper limit of the freezing phase transition initiation temperature, and the lower end represents the lower limit of the freezing phase transition essentially completed temperature. The black dots represent the frost heave temperature sensitivity index for each freeze-thaw test stratigraphic block.

[0066] The Transformer model can compare and analyze the distribution of shallow water-bearing fractures, the degree of deep low-density loosening, and the frost heave tendency of different stratigraphic blocks under different assessments of frost heave deformation risk. It can also identify the range of stratigraphic blocks with representative frost heave responses, thereby determining multiple freeze-thaw test stratigraphic blocks. Based on the differences in water-bearing fractures, internal density state, and frost heave tendency among different freeze-thaw test stratigraphic blocks, the Transformer model can also determine the temperature range corresponding to a significant freeze-thaw phase transition in each test stratigraphic block and its sensitivity to temperature changes. This allows for the determination of the estimated phase transition temperature range and frost heave temperature sensitivity index for each freeze-thaw test stratigraphic block.

[0067] Step S43: Based on the multiple freeze-thaw test formation blocks, the estimated phase change temperature range of each freeze-thaw test formation block, and the frost heave temperature sensitivity index of each freeze-thaw test formation block, determine the freeze-thaw test temperature of each freeze-thaw test formation block.

[0068] In some embodiments, the Transformer model can be used to determine the freeze-thaw test temperature for each freeze-thaw test formation block.

[0069] The Transformer model can use the estimated phase transition temperature range of each freeze-thaw test formation block as the selectable range for the freeze-thaw test temperature, and combine it with the corresponding frost heave temperature sensitivity index to determine the degree of influence of different temperature values ​​on the frost heave deformation response. For freeze-thaw test formation blocks with different frost heave temperature sensitivities, the Transformer model can select temperature values ​​with strong test representativeness within the corresponding estimated phase transition temperature range, and finally determine the freeze-thaw test temperature for each freeze-thaw test formation block.

[0070] Step S5: Based on the freeze-thaw test temperature of each freeze-thaw test stratum block, perform freeze-thaw cycle tests on each freeze-thaw test stratum block and obtain the stratum frost heave deformation distribution data corresponding to each freeze-thaw test stratum block.

[0071] The frost heave deformation distribution data for each freeze-thaw test stratum block is obtained by collecting stratum deformation measurement data through displacement monitoring sensors deployed within the corresponding freeze-thaw test stratum block during freeze-thaw cycle tests conducted at the freeze-thaw test temperature for each stratum block. The frost heave deformation distribution data for each freeze-thaw test stratum block may include monitoring location coordinates, frost heave displacement amount, frost heave displacement direction, frost heave rate value, and frost heave influence depth.

[0072] In some embodiments, a temperature control device can be used to periodically apply temperature loading to the corresponding freeze-thaw test stratum block according to the freeze-thaw test temperature of each freeze-thaw test stratum block, and displacement monitoring sensors can be deployed at different horizontal positions and different burial depths of the corresponding freeze-thaw test stratum block to continuously collect displacement changes generated at different positions during the freeze-thaw cycle, thereby obtaining the distribution data of stratum frost heave deformation corresponding to each freeze-thaw test stratum block.

[0073] The temperature control device includes a circulating heat exchange pipe buried in the freeze-thaw test stratum block, a refrigeration and heating unit connected to the circulating heat exchange pipe, and a temperature controller. The refrigeration and heating unit supplies a circulating medium at a set temperature to the circulating heat exchange pipe, thereby cooling and heating the freeze-thaw test stratum block according to the set freeze-thaw test temperature.

[0074] Step S6: Based on the frost heave deformation distribution data corresponding to each freeze-thaw test stratum block, determine multiple joint freeze-thaw test stratum block groups and the freeze-thaw test temperature of each joint freeze-thaw test stratum block group, wherein each joint freeze-thaw test stratum block group includes at least two freeze-thaw test stratum blocks.

[0075] In some embodiments, Figure 6 This invention provides a schematic flowchart for determining the freeze-thaw test temperature of multiple joint freeze-thaw test stratigraphic blocks and each joint freeze-thaw test stratigraphic block group. The determination of the freeze-thaw test temperature of the multiple joint freeze-thaw test stratigraphic blocks and each joint freeze-thaw test stratigraphic block group includes steps S61-S63:

[0076] Step S61: Cluster the frost heave deformation distribution data corresponding to each freeze-thaw test stratum block to obtain multiple frost heave deformation response clusters.

[0077] In some embodiments, multiple frost heave deformation response clusters can be obtained by using a clustering algorithm based on the frost heave deformation distribution data corresponding to each freeze-thaw test stratum block. The clustering algorithm is the K-means clustering algorithm.

[0078] K-means clustering is a clustering algorithm that divides multiple samples into clusters based on the numerical distance between samples. K-means clustering continuously calculates the distance between each sample and the cluster center and redistributes the samples, gradually reducing the numerical differences between samples within the same cluster and gradually increasing the numerical differences between different clusters, thus completing the data grouping.

[0079] Multiple frost heave deformation response clusters are formed by dividing the frost heave deformation distribution data of each freeze-thaw test stratum block according to the similarity of the data using the K-means clustering algorithm. Each frost heave deformation response cluster retains the frost heave deformation distribution data of each freeze-thaw test stratum block within the cluster.

[0080] The number of frost heave deformation response clusters is the K value in the K-means clustering algorithm. The freeze-thaw test stratigraphic blocks within each frost heave deformation response cluster have similar stratigraphic frost heave deformation distribution data, while the stratigraphic frost heave deformation distribution data between different frost heave deformation response clusters show significant differences.

[0081] In some embodiments, the value of K can be determined by a preset relationship table between the value of K and the spatial distribution density of the freeze-thaw test stratigraphic block and the difference in the frost heave deformation response of the stratigraphic block. The higher the spatial distribution density of the freeze-thaw test stratigraphic block and the greater the difference in the frost heave deformation response of the stratigraphic block, the larger the value of K. The preset relationship table between the value of K and the spatial distribution density of the freeze-thaw test stratigraphic block and the difference in the frost heave deformation response of the stratigraphic block is artificially constructed in advance.

[0082] By obtaining multiple frost heave deformation response clusters, a large number of spatially dispersed freeze-thaw test stratigraphic blocks can be grouped together according to their frost heave deformation performance. This allows freeze-thaw test stratigraphic blocks with similar frost heave deformation distribution data to form the same response type, thereby more clearly presenting the differences in frost heave deformation of different stratigraphic blocks under freeze-thaw cycles and reducing the analysis complexity of a large number of discrete test results.

[0083] As an example, specifically, the frost heave deformation distribution data corresponding to each freeze-thaw test stratigraphic block is taken as a sample to be clustered. All samples to be clustered are processed at a uniform scale, and multiple initial cluster centers are selected according to a preset number of clusters. The distance between each sample to be clustered and each initial cluster center is calculated, and each sample to be clustered is assigned to the cluster corresponding to the nearest cluster center. Then, the cluster center of each cluster is recalculated based on all samples to be clustered within each cluster. The sample division and cluster center update are repeated until the positional change of each cluster center is less than a preset threshold or the clusters to which each sample to be clustered belongs remain stable, thereby obtaining multiple frost heave deformation response clusters.

[0084] Step S62: Determine multiple frost heave response transmission lines based on the multiple frost heave response clusters.

[0085] In some embodiments, a response transfer analysis model can be used to determine multiple frost heave deformation response transfer lines. The response transfer analysis model is a deep neural network. The input to the response transfer analysis model is the multiple frost heave deformation response clusters, and the output of the response transfer analysis model is the multiple frost heave deformation response transfer lines.

[0086] Multiple frost heave response transmission lines are path lines used to represent the spatial transmission paths that may form between different freeze-thaw test strata blocks due to frost heave deformation.

[0087] Each frost heave response transfer line connects two or more freeze-thaw test stratigraphic blocks that have spatial correlations in terms of frost heave effects. The connection order of the frost heave response transfer lines is used to indicate the spatial transfer direction and extension path that the frost heave effects may form between adjacent freeze-thaw test stratigraphic blocks.

[0088] Deep neural networks (DNNs) are multi-layered neural network models composed of an input layer, multiple hidden layers, and an output layer stacked sequentially. Through layer-by-layer combinations of linear transformations and nonlinear activation functions, DNNs can progressively abstract and represent input data. DNNs possess strong fitting capabilities in uncovering complex nonlinear mapping relationships. For example, a deep neural network can perform layer-by-layer analysis based on the spatial distribution of different freeze-thaw test strata blocks within multiple frost heave deformation response clusters, the differences in frost heave deformation responses, and the corresponding changes between different freeze-thaw test strata blocks. This analysis determines the transmission direction and spatial extension path of frost heave deformation effects between different freeze-thaw test strata blocks, ultimately identifying multiple frost heave deformation response transmission lines.

[0089] By determining multiple frost heave deformation response transmission lines, the transmission path of frost heave deformation influence between different freeze-thaw test strata blocks can be clearly represented, making the transmission direction and extension range of frost heave deformation influence in space more intuitive, and facilitating the analysis of the spatial propagation of frost heave deformation influence in the strata surrounding the shield tunnel.

[0090] Step S63: Based on the multiple frost heave deformation response clusters and the multiple frost heave deformation response transmission lines, determine multiple joint freeze-thaw test stratigraphic block groups and the freeze-thaw test temperature of each joint freeze-thaw test stratigraphic block group.

[0091] In some embodiments, a combined freeze-thaw test optimization model can be used to determine multiple combined freeze-thaw test stratigraphic blocks and the freeze-thaw test temperature of each combined freeze-thaw test stratigraphic block group. The combined freeze-thaw test optimization model is a deep neural network. The input to the combined freeze-thaw test optimization model is the multiple frost heave deformation response clusters and the multiple frost heave deformation response transfer lines, and the output of the combined freeze-thaw test optimization model is the multiple combined freeze-thaw test stratigraphic blocks and the freeze-thaw test temperature of each combined freeze-thaw test stratigraphic block group.

[0092] Multiple joint freeze-thaw test stratigraphic blocks are spatial test combinations consisting of two or more freeze-thaw test stratigraphic blocks that need to be subjected to temperature loading and frost heave deformation monitoring in the same test process.

[0093] Multiple freeze-thaw test stratigraphic blocks within the same joint freeze-thaw test stratigraphic block group are strongly related in terms of spatial location and the transmission of frost heave deformation effects. By testing these freeze-thaw test stratigraphic blocks as a whole, the changes in frost heave deformation caused by the mutual influence between different freeze-thaw test stratigraphic blocks can be observed.

[0094] The freeze-thaw test temperature for each joint freeze-thaw test stratigraphic block group is the target temperature value used when multiple freeze-thaw test stratigraphic blocks within the corresponding joint freeze-thaw test stratigraphic block group are subjected to freeze-thaw cycle tests together. Each joint freeze-thaw test stratigraphic block group corresponds to one freeze-thaw test temperature.

[0095] Frost heave deformation response clusters can reflect the typical frost heave deformation response types exhibited by different freeze-thaw test stratigraphic blocks under freeze-thaw action, while frost heave deformation response transfer lines can reflect the spatial transmission direction and extension path of frost heave deformation influence between different freeze-thaw test stratigraphic blocks. By combining multiple frost heave deformation response clusters and multiple frost heave deformation response transfer lines, the model can simultaneously grasp the frost heave deformation performance of different freeze-thaw test stratigraphic blocks, the corresponding freeze-thaw test temperatures, and the spatial transmission relationships formed between different freeze-thaw test stratigraphic blocks. This allows the model to determine which freeze-thaw test stratigraphic blocks are suitable for being assigned to the same joint freeze-thaw test stratigraphic block group, and to determine the freeze-thaw test temperature of each joint freeze-thaw test stratigraphic block group based on the frost heave deformation performance, corresponding freeze-thaw test temperatures, and spatial transmission relationships of each freeze-thaw test stratigraphic block within the group.

[0096] Deep neural networks can map the contents of multiple frost heave deformation response clusters and multiple frost heave deformation response transmission lines layer by layer through multiple hidden layers, and learn the complex nonlinear correspondence between different frost heave deformation response types, spatial transmission relationships, and joint test groups. Deep neural networks can analyze the similarity of frost heave deformation responses, spatial transmission directions, and spatial extension paths of different freeze-thaw test stratigraphic blocks, and further map the analysis results into grouping results and temperature matching results between freeze-thaw test stratigraphic blocks, thereby determining multiple joint freeze-thaw test stratigraphic block groups and the freeze-thaw test temperature of each joint freeze-thaw test stratigraphic block group.

[0097] Step S7: Based on the freeze-thaw test temperature of each joint freeze-thaw test stratigraphic block group, conduct freeze-thaw cycle tests on each joint freeze-thaw test stratigraphic block group respectively, and obtain the frost heave deformation distribution data of each joint freeze-thaw test stratigraphic block group.

[0098] The formation frost heave deformation distribution data for each joint freeze-thaw test stratigraphic block group is obtained through displacement monitoring sensors deployed within the corresponding joint freeze-thaw test stratigraphic block group during freeze-thaw cycle tests conducted at the freeze-thaw test temperatures of each joint freeze-thaw test stratigraphic block group. The formation frost heave deformation distribution data for each joint freeze-thaw test stratigraphic block group may include monitoring location coordinates, frost heave displacement, frost heave displacement direction, frost heave rate value, and frost heave influence depth.

[0099] Step S8: Based on the frost heave deformation distribution data of each freeze-thaw test stratum block and the frost heave deformation distribution data of each joint freeze-thaw test stratum block group, determine the stratum blocks with high frost heave deformation around the shield tunnel.

[0100] In some embodiments, Figure 7 This invention provides a schematic flowchart for determining a stratum block with significant frost heave deformation around a shield tunnel. The determination of this stratum block includes steps S81-S84:

[0101] Step S81: Based on the high-definition multispectral image of the strata surrounding the shield tunnel, the frost heave deformation distribution data of the strata corresponding to each freeze-thaw test strata block, and the frost heave deformation distribution data of the strata corresponding to each joint freeze-thaw test strata block group, simulated frost heave deformation evolution data of the strata surrounding the shield tunnel under various extreme freeze-thaw test temperatures are generated.

[0102] In some embodiments, a frost heave deformation evolution simulation model can be used to generate simulated frost heave deformation evolution data of the strata surrounding the shield tunnel under various extreme freeze-thaw test temperatures. The frost heave deformation evolution simulation model is a variational autoencoder. The inputs to the frost heave deformation evolution simulation model are high-resolution multispectral images of the strata surrounding the shield tunnel, frost heave deformation distribution data corresponding to each freeze-thaw test stratum block, and frost heave deformation distribution data corresponding to each joint freeze-thaw test stratum block group. The output of the frost heave deformation evolution simulation model is the simulated frost heave deformation evolution data of the strata surrounding the shield tunnel under various extreme freeze-thaw test temperatures.

[0103] A variational autoencoder (VAE) is a generative model that learns the probability distribution of input samples and generates new samples. A VAE uses an encoder to map the input content to a probability distribution in a continuous latent space, and then a decoder generates new output samples based on the sampling results in the latent space, ensuring that the generated results maintain a similar numerical distribution to the known input samples.

[0104] Simulated frost heave deformation evolution data of the strata surrounding a shield tunnel under various extreme freeze-thaw test temperatures are simulation data used to represent the predicted local frost heave deformation range and its spatial variation under different extreme freeze-thaw test temperatures. This data can include multiple predicted frost heave deformation regions at each extreme freeze-thaw test temperature, the frost heave displacement distribution, frost heave rate value, frost heave influence depth, frost heave deformation propagation direction, and the location of the region with the largest frost heave deformation at each extreme freeze-thaw test temperature.

[0105] Simulated frost heave deformation evolution data of the strata surrounding the shield tunnel under various extreme freeze-thaw test temperatures can cover different spatial locations around the shield tunnel and reflect the range, degree and spatial variation of frost heave deformation at different spatial locations under different extreme freeze-thaw test temperatures.

[0106] By generating simulated frost heave deformation evolution data of the strata surrounding the shield tunnel under various extreme freeze-thaw test temperatures, the range and spatial variation of local frost heave deformation that may occur in the strata surrounding the shield tunnel under different extreme freeze-thaw test temperatures can be expanded and simulated. This supplements the extreme temperature conditions that are difficult to be fully covered by actual freeze-thaw cycle tests alone, and thus more comprehensively reflects the frost heave deformation performance of the strata surrounding the shield tunnel under different extreme freeze-thaw test conditions.

[0107] The variational autoencoder (VAE) can learn the correspondence between the spatial representation of the strata, the actual frost heave deformation state, and different freeze-thaw test conditions based on high-resolution multispectral images of the strata surrounding the shield tunnel, frost heave deformation distribution data for each freeze-thaw test stratum block, and frost heave deformation distribution data for each joint freeze-thaw test stratum block group. It then probabilistically expresses and reconstructs these relationships using a latent space. Based on the learned frost heave deformation distribution patterns, the VAE can generate possible frost heave deformation results for the strata surrounding the shield tunnel under various extreme freeze-thaw test temperatures, thus obtaining simulated frost heave deformation evolution data of the strata surrounding the shield tunnel under multiple extreme freeze-thaw test temperatures.

[0108] Step S82: Based on the simulated frost heave deformation evolution data of the strata surrounding the shield tunnel under various extreme freeze-thaw test temperatures, determine the information of multiple key strata frost heave deformation points under each extreme freeze-thaw test temperature.

[0109] In some embodiments, a key point determination model can be used to determine information on multiple key strata frost heave deformation points at each extreme freeze-thaw test temperature. The key point determination model is a deep neural network. The input to the key point determination model is simulated frost heave deformation evolution data of the strata surrounding the shield tunnel under various extreme freeze-thaw test temperatures, and the output of the key point determination model is information on multiple key strata frost heave deformation points at each extreme freeze-thaw test temperature.

[0110] Information on multiple key strata frost heave deformation points under each extreme freeze-thaw test temperature is used to identify key strata locations exhibiting significant frost heave deformation under the corresponding extreme freeze-thaw test temperature. Information on each key strata frost heave deformation point may include its spatial coordinates, extreme freeze-thaw test temperature value, frost heave displacement value, frost heave rate value, frost heave influence depth, and frost heave deformation propagation direction.

[0111] Simulated frost heave deformation evolution data of the strata surrounding shield tunnels under various extreme freeze-thaw test temperatures can reflect the distribution range, frost heave displacement changes, frost heave rate, frost heave influence depth, and spatial expansion of multiple predicted frost heave deformation regions under different extreme freeze-thaw test temperatures. When the degree of frost heave deformation and its spatial variation differ at different spatial locations, they can exhibit varying degrees of importance. Deep neural networks can identify locations with prominent frost heave deformation and significant spatial variations from each predicted frost heave deformation region based on simulated frost heave deformation evolution data, thereby determining information on multiple key strata frost heave deformation points under each extreme freeze-thaw test temperature.

[0112] Deep neural networks can perform a unified analysis of frost heave displacement changes, frost heave rate differences, frost heave influence depth, and spatial expansion characteristics in predicted frost heave deformation regions under different extreme freeze-thaw test temperatures, based on a large number of simulated frost heave deformation evolution results. They can also identify locations that significantly characterize the frost heave deformation state of strata. Because different spatial locations differ in the degree of frost heave deformation, the magnitude of change, and the interaction with surrounding areas, some locations can more concentratedly reflect the anomalous changes in local strata under extreme freeze-thaw conditions. Deep neural networks can establish discriminant relationships between different frost heave deformation behaviors and key strata locations, comparing the characterization ability of each spatial location under different extreme freeze-thaw test temperatures. They can also identify locations with prominent frost heave deformation, obvious change trends, and representativeness of the surrounding strata deformation state. This allows for the selection of multiple key strata frost heave deformation points from a large number of simulated frost heave deformation results under each extreme freeze-thaw test temperature.

[0113] By determining the information of multiple key strata frost heave deformation points at each extreme freeze-thaw test temperature, key locations with significant frost heave deformation can be quickly screened out, and the identification efficiency of key locations of frost heave deformation can be improved.

[0114] Step S83: Construct a frost heave deformation evolution map. The frost heave deformation evolution map includes multiple extreme freeze-thaw test nodes and multiple edges between the nodes. The node characteristics of each extreme freeze-thaw test node are information on multiple key strata frost heave deformation points under each extreme freeze-thaw test temperature.

[0115] The frost heave deformation evolution map is a map data composed of multiple extreme freeze-thaw test nodes and multiple edges between the nodes. The frost heave deformation evolution map is used to organize the information of multiple key strata frost heave deformation points at different extreme freeze-thaw test temperatures and the connection relationships between different extreme freeze-thaw test temperatures in the same map.

[0116] Multiple extreme freeze-thaw test nodes are nodes in the frost heave deformation evolution map that correspond to different extreme freeze-thaw test temperatures. Each extreme freeze-thaw test node corresponds to one extreme freeze-thaw test temperature, and the node characteristics of each extreme freeze-thaw test node are information on multiple key strata frost heave deformation points at each extreme freeze-thaw test temperature.

[0117] Multiple edges between nodes represent connections between different extreme freeze-thaw test nodes in the frost heave deformation evolution map. The edge characteristic of each edge is the difference in extreme freeze-thaw test temperatures between the two extreme freeze-thaw test nodes it connects, representing the temperature difference between the two extreme freeze-thaw test nodes.

[0118] In some embodiments, multiple extreme freeze-thaw test nodes can be sorted according to the numerical values ​​of various extreme freeze-thaw test temperatures, and extreme freeze-thaw test nodes with adjacent temperature values ​​can be connected in sequence. The difference in extreme freeze-thaw test temperatures corresponding to two adjacent extreme freeze-thaw test nodes can be used as the feature of the corresponding edge to construct a freeze-thaw deformation evolution map.

[0119] Step S84: Based on the significant block determination model, the frost heave deformation evolution map is processed to determine the strata blocks with high significant frost heave deformation around the shield tunnel.

[0120] The salient block identification model is a graph neural network model. The input of the salient block identification model is the frost heave deformation evolution map, and the output of the salient block identification model is the frost heave deformation highly significant strata blocks around the shield tunnel.

[0121] The strata blocks with significant frost heave deformation around the shield tunnel were identified by the significant block determination model. These strata blocks exhibited prominent frost heave deformation and had a relatively high risk of frost heave deformation.

[0122] Stratigraphic blocks with significant frost heave deformation typically exhibit obvious frost heave deformation characteristics, and the corresponding stratigraphic locations are more prone to larger frost heave deformation changes compared to other surrounding stratigraphic blocks.

[0123] Graph Neural Networks (GNNs) are neural network models that can directly process graph-structured data. GNNs transmit and aggregate information about node features, edge features, and connections between nodes in a graph structure, enabling each node to update itself by combining its own features with information from neighboring nodes. This allows for the learning and analysis of the relationships between different nodes in the entire graph structure.

[0124] By constructing a frost heave deformation evolution map, information on frost heave deformation points in key formations dispersed under different extreme freeze-thaw test temperatures can be organized into a single map structure according to temperature change relationships. This transforms previously independent results from multiple extreme freeze-thaw tests into a coherent dataset with clear correlations. Multiple extreme freeze-thaw test nodes are connected according to their extreme freeze-thaw test temperatures, using the temperature difference between adjacent extreme freeze-thaw test temperatures. This allows the changes in frost heave deformation points in key formations under different temperature conditions to be correlated along the temperature change sequence. This facilitates comparison of frost heave deformation at the same formation location under different temperature conditions, determining whether significant frost heave deformation at a particular formation location is an occasional occurrence under a single temperature condition or exhibits a relatively stable and significant characteristic under different temperature conditions.

[0125] Graph neural networks (Graph Neural Networks) can propagate and aggregate information about node connections in a frost heave deformation evolution map. When analyzing a specific extreme freeze-thaw test node, a Graph Neural Network can simultaneously reference information on key frost heave deformation points in other connected extreme freeze-thaw test nodes, as well as temperature differences between nodes. This approach allows for correlation analysis of frost heave deformation at the same stratum location and its vicinity under different extreme freeze-thaw test temperatures, rather than relying solely on a single simulation result at a single extreme freeze-thaw test temperature. As the Graph Neural Network transmits and fuses information from multiple nodes, it can identify strata locations with significant frost heave deformation under different temperature conditions from the entire frost heave deformation evolution map, reducing the impact of local anomalies under a single temperature condition on judgment. This allows for more accurate identification of strata blocks with highly significant frost heave deformation around shield tunnels.

[0126] Based on the same inventive concept Figure 8 This invention provides a schematic diagram of a shield tunnel stratum detection muon precise positioning system, which includes:

[0127] Image acquisition module 91 is used to acquire high-resolution multispectral images of the strata surrounding the shield tunnel;

[0128] Risk block determination module 92 is used to determine multiple strata blocks with frost heave deformation risk to be evaluated based on high-definition multispectral images of the strata surrounding the shield tunnel.

[0129] The detection data acquisition module 93 is used to acquire muon penetration detection data for each stratigraphic block to be assessed for frost heave deformation risk;

[0130] The single-unit test determination module 94 is used to determine multiple freeze-thaw test strata blocks and the freeze-thaw test temperature of each freeze-thaw test strata block based on the high-definition multispectral image of each stratum block to be evaluated for frost heave deformation risk and the muon penetration detection data of each stratum block to be evaluated for frost heave deformation risk.

[0131] The single-unit acquisition module 95 is used to perform freeze-thaw cycle tests on each freeze-thaw test stratum block based on the freeze-thaw test temperature of each freeze-thaw test stratum block, and to acquire the stratum frost heave deformation distribution data corresponding to each freeze-thaw test stratum block.

[0132] The joint test determination module 96 is used to determine multiple joint freeze-thaw test stratigraphic block groups and the freeze-thaw test temperature of each joint freeze-thaw test stratigraphic block group based on the stratigraphic frost heave deformation distribution data corresponding to each freeze-thaw test stratigraphic block. Each joint freeze-thaw test stratigraphic block group includes at least two freeze-thaw test stratigraphic blocks.

[0133] The joint acquisition module 97 is used to perform freeze-thaw cycle tests on each joint freeze-thaw test stratigraphic block group based on the freeze-thaw test temperature of each joint freeze-thaw test stratigraphic block group, and to acquire the frost heave deformation distribution data of the stratigraphic blocks corresponding to each joint freeze-thaw test stratigraphic block group.

[0134] The significant block determination module 98 is used to determine the strata blocks with high significant frost heave deformation around the shield tunnel based on the strata frost heave deformation distribution data corresponding to each freeze-thaw test strata block and the strata frost heave deformation distribution data corresponding to each joint freeze-thaw test strata block group.

[0135] It should be noted that, in order to simplify the descriptions disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments of this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0136] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A method for precise positioning of muons in shield tunnel strata detection, characterized in that, include: Acquire high-resolution multispectral images of the strata surrounding the shield tunnel; Based on high-resolution multispectral images of the strata surrounding the shield tunnel, several strata blocks with frost heave deformation risk to be assessed were identified. Acquire muon penetration data for each stratigraphic block with frost heave deformation risk to be assessed; Based on high-resolution multispectral images of each stratigraphic block with frost heave deformation risk to be assessed and muon penetration detection data of each stratigraphic block with frost heave deformation risk to be assessed, multiple freeze-thaw test stratigraphic blocks and freeze-thaw test temperatures of each freeze-thaw test stratigraphic block are determined. Based on the freeze-thaw test temperature of each freeze-thaw test stratum block, freeze-thaw cycle tests are performed on each freeze-thaw test stratum block, and the distribution data of frost heave deformation of the stratum corresponding to each freeze-thaw test stratum block are obtained. Based on the distribution data of frost heave deformation corresponding to each freeze-thaw test stratum block, multiple joint freeze-thaw test stratum block groups and the freeze-thaw test temperature of each joint freeze-thaw test stratum block group are determined. Each joint freeze-thaw test stratum block group includes at least two freeze-thaw test stratum blocks. Based on the freeze-thaw test temperature of each joint freeze-thaw test stratigraphic block group, freeze-thaw cycle tests are performed on each joint freeze-thaw test stratigraphic block group, and the distribution data of frost heave deformation of the stratigraphic blocks corresponding to each joint freeze-thaw test stratigraphic block group are obtained. Based on the frost heave deformation distribution data corresponding to each freeze-thaw test stratum block and the frost heave deformation distribution data corresponding to each joint freeze-thaw test stratum block group, stratum blocks with high frost heave deformation around the shield tunnel are identified.

2. The method for precise positioning of muons in shield tunnel strata detection as described in claim 1, characterized in that, The determination of strata with highly significant frost heave deformation around the shield tunnel based on the strata frost heave deformation distribution data corresponding to each freeze-thaw test strata block and the strata frost heave deformation distribution data corresponding to each joint freeze-thaw test strata block group includes: Based on the high-definition multispectral images of the strata surrounding the shield tunnel, the frost heave deformation distribution data of the strata corresponding to each freeze-thaw test strata block, and the frost heave deformation distribution data of the strata corresponding to each joint freeze-thaw test strata block group, simulated frost heave deformation evolution data of the strata surrounding the shield tunnel under various extreme freeze-thaw test temperatures are generated. Based on the simulated frost heave deformation evolution data of the strata surrounding the shield tunnel under various extreme freeze-thaw test temperatures, information on multiple key strata frost heave deformation points under each extreme freeze-thaw test temperature was determined. A frost heave deformation evolution map is constructed, which includes multiple extreme freeze-thaw test nodes and multiple edges between the nodes. The node characteristics of each extreme freeze-thaw test node are information on multiple key strata frost heave deformation points under each extreme freeze-thaw test temperature. The frost heave deformation evolution map is processed based on the significant block determination model to identify the strata blocks with high significant frost heave deformation around the shield tunnel.

3. The method for precise positioning of muons in shield tunnel strata detection as described in claim 1, characterized in that, Based on the formation frost heave deformation distribution data corresponding to each freeze-thaw test formation block, multiple joint freeze-thaw test formation block groups and the freeze-thaw test temperature of each joint freeze-thaw test formation block group are determined. Each joint freeze-thaw test formation block group includes at least two freeze-thaw test formation blocks, including: Multiple frost heave deformation response clusters are obtained by clustering the frost heave deformation distribution data corresponding to each freeze-thaw test stratum block. Multiple frost heave deformation response transmission lines are determined based on the aforementioned multiple frost heave deformation response clusters; Based on the multiple frost heave deformation response clusters and the multiple frost heave deformation response transmission lines, multiple joint freeze-thaw test stratigraphic block groups and the freeze-thaw test temperature of each joint freeze-thaw test stratigraphic block group are determined.

4. The method for precise positioning of muons in shield tunnel strata detection as described in claim 3, characterized in that, The clustering of frost heave deformation distribution data corresponding to each freeze-thaw test stratum block yields multiple frost heave deformation response clusters, including: Based on the distribution data of frost heave deformation corresponding to each freeze-thaw test stratum block, a clustering algorithm is used to cluster multiple frost heave deformation response clusters.

5. A precise positioning system for muon detection in shield tunnel strata, characterized in that, include: The image acquisition module is used to acquire high-resolution multispectral images of the strata surrounding the shield tunnel; The risk block determination module is used to determine multiple strata blocks with frost heave deformation risk to be evaluated based on high-definition multispectral images of the strata surrounding the shield tunnel. The detection data acquisition module is used to acquire muon penetration detection data for each stratigraphic block to be assessed for frost heave deformation risk; The single-unit test determination module is used to determine multiple freeze-thaw test strata blocks and the freeze-thaw test temperature of each freeze-thaw test strata block based on the high-resolution multispectral image of each stratum block to be evaluated for frost heave deformation risk and the muon penetration detection data of each stratum block to be evaluated for frost heave deformation risk. The single-unit acquisition module is used to perform freeze-thaw cycle tests on each freeze-thaw test stratum block based on the freeze-thaw test temperature of each freeze-thaw test stratum block, and to acquire the stratum frost heave deformation distribution data corresponding to each freeze-thaw test stratum block. The joint test determination module is used to determine multiple joint freeze-thaw test stratigraphic block groups and the freeze-thaw test temperature of each joint freeze-thaw test stratigraphic block group based on the stratigraphic frost heave deformation distribution data corresponding to each freeze-thaw test stratigraphic block. Each joint freeze-thaw test stratigraphic block group includes at least two freeze-thaw test stratigraphic blocks. The joint acquisition module is used to perform freeze-thaw cycle tests on each joint freeze-thaw test stratigraphic block group based on the freeze-thaw test temperature of each joint freeze-thaw test stratigraphic block group, and to acquire the frost heave deformation distribution data of each joint freeze-thaw test stratigraphic block group. The significant block determination module is used to determine the strata blocks with high significant frost heave deformation around the shield tunnel based on the strata frost heave deformation distribution data corresponding to each freeze-thaw test strata block and the strata frost heave deformation distribution data corresponding to each joint freeze-thaw test strata block group.

6. The shield tunnel strata detection muon precise positioning system as described in claim 5, characterized in that, The significant block determination module is also used for: Based on the high-definition multispectral images of the strata surrounding the shield tunnel, the frost heave deformation distribution data of the strata corresponding to each freeze-thaw test strata block, and the frost heave deformation distribution data of the strata corresponding to each joint freeze-thaw test strata block group, simulated frost heave deformation evolution data of the strata surrounding the shield tunnel under various extreme freeze-thaw test temperatures are generated. Based on the simulated frost heave deformation evolution data of the strata surrounding the shield tunnel under various extreme freeze-thaw test temperatures, information on multiple key strata frost heave deformation points under each extreme freeze-thaw test temperature was determined. A frost heave deformation evolution map is constructed, which includes multiple extreme freeze-thaw test nodes and multiple edges between the nodes. The node characteristics of each extreme freeze-thaw test node are information on multiple key strata frost heave deformation points under each extreme freeze-thaw test temperature. The frost heave deformation evolution map is processed based on the significant block determination model to identify the strata blocks with high significant frost heave deformation around the shield tunnel.

7. The shield tunnel strata detection muon precise positioning system as described in claim 5, characterized in that, The joint test determination module is also used for: Multiple frost heave deformation response clusters are obtained by clustering the frost heave deformation distribution data corresponding to each freeze-thaw test stratum block. Multiple frost heave deformation response transmission lines are determined based on the aforementioned multiple frost heave deformation response clusters; Based on the multiple frost heave deformation response clusters and the multiple frost heave deformation response transmission lines, multiple joint freeze-thaw test stratigraphic block groups and the freeze-thaw test temperature of each joint freeze-thaw test stratigraphic block group are determined.

8. The shield tunnel strata detection muon precise positioning system as described in claim 7, characterized in that, The clustering of frost heave deformation distribution data corresponding to each freeze-thaw test stratum block yields multiple frost heave deformation response clusters, including: Based on the distribution data of frost heave deformation corresponding to each freeze-thaw test stratum block, a clustering algorithm is used to cluster multiple frost heave deformation response clusters.

9. An electronic device, characterized in that, include: processor; Memory; And a computer program; wherein the computer program is stored in the memory and configured to be executed by the processor to implement the shield tunnel strata detection muon precise positioning method as described in any one of claims 1 to 4.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the method for precise positioning of muons in shield tunnel strata detection as described in any one of claims 1 to 4.

Images