A tunnel construction surrounding rock collapse advanced prediction analysis method and system

Abstract

The application discloses a tunnel construction surrounding rock collapse advanced prediction analysis method and system, the method comprises the following steps: setting various sensors in the tunnel, combining to form a perturbation sensing grid, and extracting a disturbance waveform response vector group of each observation point; according to the disturbance waveform response vector group, the space-time coherence index between the observation points is calculated, the two observation points with the space-time coherence index exceeding the threshold value are clustered together, and a plurality of disturbance coherence maps are formed; the graph neural clustering is performed on the plurality of disturbance coherence maps, the plurality of disturbance coherence maps are clustered into a plurality of blocks as potential instability blocks; the risk intensity of each position in each potential instability block is calculated, the risk intensity of each position is rendered by color spectrum and then superimposed into a tunnel BIM model, and then imported into a mixed reality platform for real-time display.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of tunnel construction surrounding rock collapse advanced prediction analysis, more specifically, relates to a tunnel construction surrounding rock collapse advanced prediction analysis method and system. BACKGROUND

[0002] Surrounding rock collapse is one of the most common and dangerous disasters in tunnel construction, mainly occurring under conditions such as insufficient self-stability of surrounding rock, lagging or improper support, complex geological structure, etc. Construction disturbance can destroy the original stress balance of the surrounding rock, leading to crack propagation, shear slip or instability collapse of the rock mass, and forming phenomena such as collapse, block falling, and arch crown subsidence. Surrounding rock collapse often has the characteristics of strong suddenness, inconspicuous precursors, and rapid expansion, which can easily cause damage to equipment on the work face, injuries to construction personnel, and interruption of construction progress. Soft interlayer, fault fracture zone, high ground stress area, and groundwater development section are high-risk areas for collapse. The causes include engineering operation problems such as overbreak or underbreak, blasting disturbance, inadequate support, and poor drainage, and may also be caused by sudden geological changes or continuous rainfall. Before the initial support of the tunnel forms a closed structure, the surrounding rock is in a highly unstable state and is prone to local or overall instability.

[0003] Therefore, there is an urgent need for a technical solution to accurately identify surrounding rock collapse disasters during tunnel construction. SUMMARY

[0004] To solve the above technical problems, the present application provides a tunnel construction surrounding rock collapse advanced prediction analysis method, comprising:

[0005] A plurality of sensors are arranged in the tunnel to form a perturbation perception grid, and a disturbance waveform response vector group of each observation point is extracted;

[0006] According to the disturbance waveform response vector group, the spatio-temporal coherence index between the observation points is calculated, the two observation points with a spatio-temporal coherence index exceeding a threshold value are clustered together, and a plurality of disturbance coherence maps are formed;

[0007] The plurality of disturbance coherence maps are graphically clustered, and the plurality of disturbance coherence maps are clustered into a plurality of clusters as potential instability clusters;

[0008] The risk intensity of each position in each potential instability cluster is calculated, the risk intensity of each position is rendered by color spectrum and superimposed into a tunnel BIM model, and imported into a mixed reality platform for real-time display.

[0009] Further, the plurality of sensors include an acoustic emission sensor and a microseismometer;

[0010] The acoustic energy release of each observation point is obtained in real time by the acoustic emission sensor;

[0011] The first frequency offset of each observation point is obtained in real time by the acoustic emission sensor, and the second frequency offset of each observation point is obtained in real time by the microseismometer. The first frequency offset and the second frequency offset of each observation point are normalized and then weighted and fused into the final frequency offset.

[0012] The acoustic emission sensor is used to acquire the signal entropy increment of each observation point in real time.

[0013] The acoustic energy release, final frequency shift, and signal entropy increment at each observation point are used as the disturbance waveform response vector set for each observation point.

[0014] Furthermore, before calculating the spatiotemporal coherence index between observation points, the process includes performing a time sliding window operation on the disturbance waveform response vector group to form a disturbance waveform response vector group with multiple time windows.

[0015] Furthermore, calculating the spatiotemporal coherence index between observation points includes:

[0016]

[0017] Among them, STCI ij Let t be the spatiotemporal coherence index between the i-th and j-th observation points, K be the number of time windows, and t be the time window index. k Let t be the time in the k-th time window. <D i (t k ), D j (t k D is the set of disturbance waveform response vectors for the i-th observation point at time t in the k-th time window. i (t k The disturbance waveform response vector set D of the j-th observation point at time t in the k-th time window. j (t k The dot product of ).

[0018] Furthermore, forming multiple perturbation coherence graphs involves connecting pairs of observation points whose spatiotemporal coherence indices exceed a threshold, with nodes representing observation points and edge weights representing spatiotemporal coherence indices, ultimately forming multiple perturbation coherence graphs.

[0019] Furthermore, graph neural clustering of the multiple perturbation coherence maps includes: performing graph neural clustering on the multiple perturbation coherence maps using a graph neural network to identify regions that simultaneously face the risk of structural instability in space as clusters.

[0020] Furthermore, calculating the risk intensity at each location within each potentially unstable cluster includes:

[0021]

[0022] Where R is the risk intensity at the current location, N is the number of clumps, and α q Let K be the perturbation intensity factor of the q-th cluster, K be the Gaussian kernel function, and r be the coordinates of the current position. q Let Γ be the centroid coordinates of the q-th cluster. q Let Γ be the cluster density factor of the q-th cluster, where Γ q α is the number of observation points within the cluster multiplied by the average spatiotemporal coherence index within the cluster. q This is the normalized value of the maximum acoustic energy release within the cluster.

[0023] Furthermore, the risk intensity at each location is rendered using color chromatograms and then superimposed onto the tunnel BIM model. This includes using transparent materials to map the risk intensity to a risk intensity layer, with higher risk intensity resulting in a darker color.

[0024] This invention also proposes a system for predicting and analyzing the collapse of surrounding rock during tunnel construction, comprising:

[0025] A sensing network module is set up to install multiple sensors inside the tunnel, which are combined to form a micro-perturbation sensing grid and extract the disturbance waveform response vector group of each observation point.

[0026] The coherence graph generation module is used to calculate the spatiotemporal coherence index between observation points based on the disturbance waveform response vector group, cluster the observation points with spatiotemporal coherence indices exceeding the threshold together, and form multiple disturbance coherence graphs.

[0027] A clumping generation module is used to perform graph neural clustering on the multiple perturbation coherence maps, and cluster the multiple perturbation coherence maps into multiple clumping blocks as potential unstable clumping blocks;

[0028] The identification module is used to calculate the risk intensity at each location in each potential unstable cluster. The risk intensity at each location is then rendered using colorimetry and superimposed onto the tunnel BIM model, and imported into the mixed reality platform for real-time display.

[0029] Furthermore, various sensors are used, including acoustic emission sensors and microseismometers;

[0030] The acoustic energy release at each observation point is acquired in real time using the acoustic emission sensor.

[0031] The first frequency offset of each observation point is obtained in real time by the acoustic emission sensor, and the second frequency offset of each observation point is obtained in real time by the microseismometer. The first frequency offset and the second frequency offset of each observation point are normalized and then weighted and fused into the final frequency offset.

[0032] The acoustic emission sensor is used to acquire the signal entropy increment of each observation point in real time.

[0033] The acoustic energy release, final frequency shift, and signal entropy increment at each observation point are used as the disturbance waveform response vector set for each observation point.

[0034] In summary, the technical solutions conceived by this invention have the following beneficial effects compared with the prior art:

[0035] This invention is the first to combine the features of acoustic emission sensors and microseismometers into a time vector to characterize the precursors of unstable states in surrounding rock, which is different from the traditional single-point threshold triggering method.

[0036] The perturbation behavior is constructed into a behavior graph, and a graph neural network is used for clumping identification, rather than simple statistical density analysis.

[0037] For the first time, disaster identification, dynamic simulation, and risk distribution results are integrated into an MR scenario to form an interactive pre-drill / training / scheduling support system. Attached Figure Description

[0038] Figure 1 This is a flowchart of the method in Embodiment 1 of the present invention;

[0039] Figure 2 This is a system structure diagram of Embodiment 2 of the present invention. Detailed Implementation

[0040] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0041] The method provided by this invention can be implemented in a terminal environment that may include one or more of the following components: a processor, a storage medium, and a display screen. The storage medium stores at least one instruction, which is loaded and executed by the processor to implement the method described in the following embodiments.

[0042] A processor may include one or more processing cores. The processor uses various interfaces and lines to connect various parts of the terminal, and performs various functions and processes data by running or executing instructions, programs, code sets or instruction sets stored in the storage medium, and by calling data stored in the storage medium.

[0043] Storage media can include random access memory (RAM) or read-only memory (ROM). Storage media can be used to store instructions, programs, code, code sets, or instructions.

[0044] The display screen is used to show the user interface of each application.

[0045] In addition, those skilled in the art will understand that the above-described terminal structure does not constitute a limitation on the terminal. The terminal may include more or fewer components, or combine certain components, or have different component arrangements. For example, the terminal may also include radio frequency circuits, input units, sensors, audio circuits, power supplies, and other components, which will not be described in detail here.

[0046] Example 1

[0047] like Figure 1 This embodiment proposes a method for predicting and analyzing the collapse of surrounding rock during tunnel construction, including:

[0048] Step 101: Set up multiple sensors inside the tunnel, combine them to form a micro-perturbation sensing grid, and extract the disturbance waveform response vector group for each observation point;

[0049] Specifically, the various sensors include: acoustic emission sensors (acoustic energy release is a typical precursor to high-frequency fracture activity, and acoustic emission sensors (AE sensors) are sensitive to frequencies above 100kHz) and microseismometers;

[0050] The acoustic energy release at each observation point is acquired in real time using the acoustic emission sensor.

[0051] Both high-frequency disturbances (AE sensor) and low-frequency disturbances (microseismometer) can cause changes in the spectral structure; therefore, their spectra should be calculated separately, normalized, and then fused. This characteristic reflects the evolutionary trend of the structure from a steady state to an active state. Therefore, the first frequency offset of each observation point is obtained in real time through the acoustic emission sensor, and the second frequency offset of each observation point is obtained in real time through the microseismometer. The first and second frequency offsets of each observation point are normalized and then weighted and fused into the final frequency offset.

[0052] The acoustic emission sensor is used to acquire the signal entropy increment of each observation point in real time.

[0053] The acoustic energy release, final frequency shift, and signal entropy increment at each observation point are used as the disturbance waveform response vector set for each observation point.

[0054] Preferably, acoustic energy release is a typical precursor to high-frequency fissure activity, and the AE sensor is sensitive to frequencies above 100kHz.

[0055] Step 102: Calculate the spatiotemporal coherence index between observation points based on the disturbance waveform response vector group, cluster the observation points with spatiotemporal coherence indices exceeding the threshold together, and form multiple disturbance coherence maps. For example, high coherence may indicate the existence of a common control source (such as fault triggering or rock bridge instability).

[0056] Preferably, through normalization, the spatiotemporal coherence index is always between [0,1].

[0057] Specifically, before calculating the spatiotemporal coherence index between observation points, the process includes: performing a time sliding window operation on the disturbance waveform response vector group to form a disturbance waveform response vector group with multiple time windows.

[0058] Specifically, calculating the spatiotemporal coherence index between observation points includes:

[0059]

[0060] Among them, STCI ij Let t be the spatiotemporal coherence index between the i-th and j-th observation points, K be the number of time windows, and t be the time window index. k Let t be the time in the k-th time window. <D i (t k ), D j (t k D is the set of disturbance waveform response vectors for the i-th observation point at time t in the k-th time window. i (t k The disturbance waveform response vector set D of the j-th observation point at time t in the k-th time window. j (t k The dot product of ).

[0061] Specifically, forming multiple perturbation coherence graphs involves connecting pairs of observation points whose spatiotemporal coherence indices exceed a threshold, with nodes representing observation points and edge weights representing spatiotemporal coherence indices, ultimately forming multiple perturbation coherence graphs.

[0062] Step 103: Perform graph neural clustering on the multiple perturbation coherence maps to cluster the multiple perturbation coherence maps into multiple clusters as potential unstable clusters;

[0063] Specifically, graph neural clustering of the multiple perturbation coherence maps includes: using a graph neural network (GNN) to perform graph neural clustering on the multiple perturbation coherence maps to identify regions that simultaneously face the risk of structural instability in space as clusters.

[0064] Specifically, calculating the risk intensity at each location within each potentially unstable cluster includes:

[0065]

[0066] Where R is the risk intensity at the current location, N is the number of clumps, and α q Let K be the perturbation intensity factor of the q-th cluster, K be the Gaussian kernel function, and r be the coordinates of the current position. q Let Γ be the centroid coordinates of the q-th cluster. q Let Γ be the cluster density factor of the q-th cluster, where Γq α is the number of observation points within the cluster multiplied by the average spatiotemporal coherence index within the cluster. q This is the normalized value of the maximum acoustic energy release within the cluster.

[0067] Step 104: Calculate the risk intensity at each location within each potential unstable cluster. After color rendering of the risk intensity at each location, overlay it onto the tunnel BIM model and import it into a mixed reality platform (MR platform such as Unity and MRTK) for real-time display. This step serves on-site decision-making by visually displaying the risk location and level to users through color hotspots, 3D rendering, and other intuitive methods. With the help of mixed reality (MR) technology, the technological achievements form an interactive and perceptible 3D risk scene on the construction site, assisting users in making predictions and responses.

[0068] Specifically, the risk intensity of each location is rendered using color chromatographic rendering and then superimposed onto the tunnel BIM model. This includes using transparent materials to map the risk intensity to a risk intensity layer, with the color becoming darker as the risk intensity increases. Preferably, high-risk areas (R>0.8) are highlighted with a red flashing indicator.

[0069] Example 2

[0070] like Figure 2 As shown in the figure, this embodiment proposes a system for predicting and analyzing the collapse of surrounding rock during tunnel construction, including:

[0071] A sensing network module is set up to install multiple sensors inside the tunnel, which are combined to form a micro-perturbation sensing grid and extract the disturbance waveform response vector group of each observation point.

[0072] Specifically, the various sensors include: acoustic emission sensors and microseismometers;

[0073] The acoustic energy release at each observation point is acquired in real time using the acoustic emission sensor.

[0074] The first frequency offset of each observation point is obtained in real time by the acoustic emission sensor, and the second frequency offset of each observation point is obtained in real time by the microseismometer. The first frequency offset and the second frequency offset of each observation point are normalized and then weighted and fused into the final frequency offset.

[0075] The acoustic emission sensor is used to acquire the signal entropy increment of each observation point in real time.

[0076] The acoustic energy release, final frequency shift, and signal entropy increment at each observation point are used as the disturbance waveform response vector set for each observation point.

[0077] The coherence graph generation module is used to calculate the spatiotemporal coherence index between observation points based on the disturbance waveform response vector group, cluster the observation points with spatiotemporal coherence indices exceeding the threshold together, and form multiple disturbance coherence graphs.

[0078] Specifically, before calculating the spatiotemporal coherence index between observation points, the process includes: performing a time sliding window operation on the disturbance waveform response vector group to form a disturbance waveform response vector group with multiple time windows.

[0079] Specifically, calculating the spatiotemporal coherence index between observation points includes:

[0080]

[0081] Among them, STCI ij Let t be the spatiotemporal coherence index between the i-th and j-th observation points, K be the number of time windows, and t be the time window index. k Let t be the time in the k-th time window. <D i (t k ),D j (t k D is the set of disturbance waveform response vectors for the i-th observation point at time t in the k-th time window. i (t k The disturbance waveform response vector set D of the j-th observation point at time t in the k-th time window. j (t k The dot product of ).

[0082] Specifically, forming multiple perturbation coherence graphs involves connecting pairs of observation points whose spatiotemporal coherence indices exceed a threshold, with nodes representing observation points and edge weights representing spatiotemporal coherence indices, ultimately forming multiple perturbation coherence graphs.

[0083] A clumping generation module is used to perform graph neural clustering on the multiple perturbation coherence maps, and cluster the multiple perturbation coherence maps into multiple clumping blocks as potential unstable clumping blocks;

[0084] Specifically, graph neural clustering of the multiple perturbation coherence maps includes: using a graph neural network to cluster the multiple perturbation coherence maps, identifying regions that simultaneously face the risk of structural instability in space as clusters.

[0085] Specifically, calculating the risk intensity at each location within each potentially unstable cluster includes:

[0086]

[0087] Where R is the risk intensity at the current location, N is the number of clumps, and α qLet K be the perturbation intensity factor of the q-th cluster, K be the Gaussian kernel function, and r be the coordinates of the current position. q Let Γ be the centroid coordinates of the q-th cluster. q Let Γ be the cluster density factor of the q-th cluster, where Γ q α is the number of observation points within the cluster multiplied by the average spatiotemporal coherence index within the cluster. q This is the normalized value of the maximum acoustic energy release within the cluster.

[0088] The identification module is used to calculate the risk intensity at each location in each potential unstable cluster. The risk intensity at each location is then rendered using colorimetry and superimposed onto the tunnel BIM model, and imported into the mixed reality platform for real-time display.

[0089] Specifically, the risk intensity of each location is rendered using color chromatograms and then superimposed onto the tunnel BIM model. This includes using transparent materials to map the risk intensity to a risk intensity layer, with higher risk intensity resulting in a darker color.

[0090] Example 3

[0091] This invention also proposes a storage medium storing multiple instructions, which are used to implement the aforementioned method for predicting and analyzing the collapse of surrounding rock during tunnel construction.

[0092] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0093] Optionally, in this embodiment, the storage medium is configured to store program code for performing the following method steps: Step 101, setting up multiple sensors in the tunnel, combining them to form a perturbation sensing grid, and extracting the perturbation waveform response vector group for each observation point;

[0094] Specifically, the various sensors include: acoustic emission sensors and microseismometers;

[0095] The acoustic energy release at each observation point is acquired in real time using the acoustic emission sensor.

[0096] The first frequency offset of each observation point is obtained in real time by the acoustic emission sensor, and the second frequency offset of each observation point is obtained in real time by the microseismometer. The first frequency offset and the second frequency offset of each observation point are normalized and then weighted and fused into the final frequency offset.

[0097] The acoustic emission sensor is used to acquire the signal entropy increment of each observation point in real time.

[0098] The acoustic energy release, final frequency shift, and signal entropy increment at each observation point are used as the disturbance waveform response vector set for each observation point.

[0099] Step 102: Calculate the spatiotemporal coherence index between observation points based on the disturbance waveform response vector group, cluster the observation points with spatiotemporal coherence indices exceeding the threshold together, and form multiple disturbance coherence maps.

[0100] Specifically, before calculating the spatiotemporal coherence index between observation points, the process includes: performing a time sliding window operation on the disturbance waveform response vector group to form a disturbance waveform response vector group with multiple time windows.

[0101] Specifically, calculating the spatiotemporal coherence index between observation points includes:

[0102]

[0103] Among them, STCI ij Let t be the spatiotemporal coherence index between the i-th and j-th observation points, K be the number of time windows, and t be the time window index. k Let t be the time in the k-th time window. <D i (t k ), D j (t k D is the set of disturbance waveform response vectors for the i-th observation point at time t in the k-th time window. i (t k The disturbance waveform response vector set D of the j-th observation point at time t in the k-th time window. j (t k The dot product of ).

[0104] Specifically, forming multiple perturbation coherence graphs involves connecting pairs of observation points whose spatiotemporal coherence indices exceed a threshold, with nodes representing observation points and edge weights representing spatiotemporal coherence indices, ultimately forming multiple perturbation coherence graphs.

[0105] Step 103: Perform graph neural clustering on the multiple perturbation coherence maps to cluster the multiple perturbation coherence maps into multiple clusters as potential unstable clusters;

[0106] Specifically, graph neural clustering of the multiple perturbation coherence maps includes: using a graph neural network to cluster the multiple perturbation coherence maps, identifying regions that simultaneously face the risk of structural instability in space as clusters.

[0107] Specifically, calculating the risk intensity at each location within each potentially unstable cluster includes:

[0108]

[0109] Where R is the risk intensity at the current location, N is the number of clumps, and α q Let K be the perturbation intensity factor of the q-th cluster, K be the Gaussian kernel function, and r be the coordinates of the current position. q Let Γ be the centroid coordinates of the q-th cluster. q Let Γ be the cluster density factor of the q-th cluster, where Γ q α is the number of observation points within the cluster multiplied by the average spatiotemporal coherence index within the cluster. q This is the normalized value of the maximum acoustic energy release within the cluster.

[0110] Step 104: Calculate the risk intensity at each location in each potential unstable cluster, perform chromatographic rendering of the risk intensity at each location, overlay it onto the tunnel BIM model, and import it into the mixed reality platform for real-time display.

[0111] Specifically, the risk intensity of each location is rendered using color chromatograms and then superimposed onto the tunnel BIM model. This includes using transparent materials to map the risk intensity to a risk intensity layer, with higher risk intensity resulting in a darker color.

[0112] Example 4

[0113] This invention also proposes an electronic device, including a processor and a storage medium connected to the processor. The storage medium stores multiple instructions, which can be loaded and executed by the processor to enable the processor to perform the aforementioned method for predicting and analyzing rock collapse during tunnel construction.

[0114] Specifically, the electronic device in this embodiment can be a computer terminal, which may include one or more processors and a storage medium.

[0115] The storage medium can be used to store software programs and modules, such as the method for predicting and analyzing rock collapse during tunnel construction in this embodiment of the invention. The corresponding program instructions / modules allow the processor to execute various functional applications and data processing by running the software programs and modules stored in the storage medium, thus realizing the aforementioned method for predicting and analyzing rock collapse during tunnel construction. The storage medium may include high-speed random access storage media, and may also include non-volatile storage media, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state storage media. In some instances, the storage medium may further include storage media remotely configured relative to the processor, which can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0116] The processor can call the information and application stored in the storage medium through the transmission system to perform the following steps: Step 101, set up a variety of sensors in the tunnel, combine them to form a micro-perturbation sensing grid, and extract the disturbance waveform response vector group of each observation point;

[0117] Specifically, the various sensors include: acoustic emission sensors and microseismometers;

[0118] The acoustic energy release at each observation point is acquired in real time using the acoustic emission sensor.

[0119] The first frequency offset of each observation point is obtained in real time by the acoustic emission sensor, and the second frequency offset of each observation point is obtained in real time by the microseismometer. The first frequency offset and the second frequency offset of each observation point are normalized and then weighted and fused into the final frequency offset.

[0120] The acoustic emission sensor is used to acquire the signal entropy increment of each observation point in real time.

[0121] The acoustic energy release, final frequency shift, and signal entropy increment at each observation point are used as the disturbance waveform response vector set for each observation point.

[0122] Step 102: Calculate the spatiotemporal coherence index between observation points based on the disturbance waveform response vector group, cluster the observation points with spatiotemporal coherence indices exceeding the threshold together, and form multiple disturbance coherence maps.

[0123] Specifically, before calculating the spatiotemporal coherence index between observation points, the process includes: performing a time sliding window operation on the disturbance waveform response vector group to form a disturbance waveform response vector group with multiple time windows.

[0124] Specifically, calculating the spatiotemporal coherence index between observation points includes:

[0125]

[0126] Among them, STCI ij Let t be the spatiotemporal coherence index between the i-th and j-th observation points, K be the number of time windows, and t be the time window index. k Let t be the time in the k-th time window. <D i (t k ),D j (t k D is the set of disturbance waveform response vectors for the i-th observation point at time t in the k-th time window. i (t k The disturbance waveform response vector set D of the j-th observation point at time t in the k-th time window. j (t k The dot product of ).

[0127] Specifically, forming multiple perturbation coherence graphs involves connecting pairs of observation points whose spatiotemporal coherence indices exceed a threshold, with nodes representing observation points and edge weights representing spatiotemporal coherence indices, ultimately forming multiple perturbation coherence graphs.

[0128] Step 103: Perform graph neural clustering on the multiple perturbation coherence maps to cluster the multiple perturbation coherence maps into multiple clusters as potential unstable clusters;

[0129] Specifically, graph neural clustering of the multiple perturbation coherence maps includes: using a graph neural network to cluster the multiple perturbation coherence maps, identifying regions that simultaneously face the risk of structural instability in space as clusters.

[0130] Specifically, calculating the risk intensity at each location within each potentially unstable cluster includes:

[0131]

[0132] Where R is the risk intensity at the current location, N is the number of clumps, and α q Let K be the perturbation intensity factor of the q-th cluster, K be the Gaussian kernel function, and r be the coordinates of the current position. q Let Γ be the centroid coordinates of the q-th cluster. q Let Γ be the cluster density factor of the q-th cluster, where Γ q α is the number of observation points within the cluster multiplied by the average spatiotemporal coherence index within the cluster. q This is the normalized value of the maximum acoustic energy release within the cluster.

[0133] Step 104: Calculate the risk intensity at each location in each potential unstable cluster, perform chromatographic rendering of the risk intensity at each location, overlay it onto the tunnel BIM model, and import it into the mixed reality platform for real-time display.

[0134] Specifically, the risk intensity of each location is rendered using color chromatograms and then superimposed onto the tunnel BIM model. This includes using transparent materials to map the risk intensity to a risk intensity layer, with higher risk intensity resulting in a darker color.

[0135] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0136] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0137] In the several embodiments provided by this invention, it should be understood that the disclosed technical content can be implemented in other ways. The system embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection between units or modules, and may be electrical or other forms.

[0138] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0139] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0140] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, optical disks, and other media capable of storing program code.

[0141] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for analyzing and predicting the collapse of surrounding rock in advance during tunnel construction, characterized in that, The method comprises the following steps: a plurality of sensors are arranged in the tunnel to form a micro-disturbance sensing network, and a disturbance waveform response vector group of each observation point is extracted; the plurality of sensors comprise acoustic emission sensors and microseismometers; spatiotemporal coherence indexes between the observation points are calculated according to the disturbance waveform response vector group, two observation points with spatiotemporal coherence indexes exceeding a threshold value are clustered together, and a plurality of disturbance coherence maps are formed; graph neural clustering is performed on the plurality of disturbance coherence maps, the plurality of disturbance coherence maps are clustered into a plurality of blocks as potential instability blocks; risk intensities of each position in each potential instability block are calculated, the risk intensities of each position are rendered by color spectrum and superimposed on a tunnel BIM model, and are imported into a mixed reality platform for real-time display.

2. The method for analyzing the collapse of surrounding rock in tunnel construction according to claim 1, characterized in that, The acoustic energy release of each observation point is acquired in real time by the acoustic emission sensor; the first frequency offset of each observation point is acquired in real time by the acoustic emission sensor, the second frequency offset of each observation point is acquired in real time by the microseismometer, the first frequency offset and the second frequency offset of each observation point are normalized and weighted to form a final frequency offset; the signal entropy increment of each observation point is acquired in real time by the acoustic emission sensor; the acoustic energy release, the final frequency offset and the signal entropy increment of each observation point are taken as the disturbance waveform response vector group of each observation point.

3. The method of claim 1, wherein the method comprises: Before the calculation of the spatiotemporal coherence indexes between the observation points, a time sliding window operation is performed on the disturbance waveform response vector group to form a plurality of time window disturbance waveform response vector groups.

4. The method of claim 3, wherein the method is characterized by, The calculation of the spatiotemporal coherence indexes between the observation points comprises the following steps: ， wherein, is a spatial coherence index between the first observation point and the second observation point, is a number of time windows, is a dot product of a set of perturbed wavefield response vectors of the first observation point at time a set of perturbed wavefield response vectors of the second observation point at time​​​​​​​​​​​​ 5. The method of claim 4, wherein the method is characterized by: the formation of the plurality of disturbance coherence maps comprises the following steps: two observation points with spatiotemporal coherence indexes exceeding a threshold value are connected together, the nodes are the observation points, and the edge weights are the spatiotemporal coherence indexes, finally forming a plurality of disturbance coherence maps.

6. The method of claim 1, wherein the method comprises: The graph neural clustering of the plurality of disturbance coherence maps comprises the following steps: the plurality of disturbance coherence maps are clustered by a graph neural network, and regions simultaneously facing structural instability risks are identified as blocks.

7. The method for predicting and analyzing the collapse of surrounding rock during tunnel construction as described in claim 1, characterized in that, The calculation of the risk intensities of each position in each potential instability block comprises the following steps: ， wherein, is the risk intensity for the current location, is the number of blobs, is the perturbation intensity factor for the th blob, is the Gaussian kernel function, is the coordinate of the current location, is the centroid coordinate of the th blob, is the centroid coordinate of the th blob, is the blob density factor for the th blob, wherein, is the number of observation points within the blob multiplied by the average of the spatiotemporal coherence index within the blob, is the normalized value of the maximum acoustic energy release within the blob.

8. The method of claim 1, wherein the method comprises: the superimposition of the risk intensities of each position on the tunnel BIM model after color spectrum rendering comprises the following steps: a risk intensity layer is mapped by using a transparent material, and the higher the risk intensity is, the darker the color is.

9. A system for analyzing and predicting the collapse of surrounding rock in advance during tunnel construction, characterized by, The method comprises the following steps: a sensing network module is arranged to arrange a plurality of sensors in the tunnel to form a micro-disturbance sensing network, and to extract a disturbance waveform response vector group of each observation point; the plurality of sensors comprise acoustic emission sensors and microseismometers; a coherence map generation module is configured to calculate spatiotemporal coherence indexes between the observation points according to the disturbance waveform response vector group, to cluster two observation points with spatiotemporal coherence indexes exceeding a threshold value together, and to form a plurality of disturbance coherence maps; a block generation module is configured to perform graph neural clustering on the plurality of disturbance coherence maps, to cluster the plurality of disturbance coherence maps into a plurality of blocks as potential instability blocks; The identification module is used for calculating the risk intensity of each position in each potential unstable block, superimposing the risk intensity of each position after chromatographic rendering, and importing into a mixed reality platform for real-time display.

10. The analysis system for the advance prediction of rock collapse during tunnel construction according to claim 9, characterized in that, The plurality of sensors include: an acoustic emission sensor and a microseism meter; The acoustic emission sensor is used for acquiring the acoustic energy release of each observation point in real time; The acoustic emission sensor is used for acquiring the first frequency offset of each observation point in real time, and the microseism meter is used for acquiring the second frequency offset of each observation point in real time; the first frequency offset and the second frequency offset of each observation point are normalized and weighted to form a final frequency offset; The acoustic emission sensor is used for acquiring the signal entropy increment of each observation point in real time; The acoustic energy release, the final frequency offset and the signal entropy increment of each observation point are used as a disturbance waveform response vector group of each observation point.

Images