Intelligent disaster risk prevention and control system for water-rich weathered granite gneiss tunnel construction

Abstract

The invention discloses a water-rich weathered granite gneiss tunnel construction disaster risk intelligent prevention and control system, and relates to the field of tunnel construction, and the system comprises a collection module which is used for collecting image data in a tunnel, and storing the image data in the tunnel; the feature extraction module is used for receiving the image data stored in the acquisition module and extracting a feature image of the image data; the iteration module is used for acquiring feature images of the image data extracted by the feature extraction module, and transmitting the feature images to the acquisition module so as to iterate the image data corresponding to the feature images; through a visual monitoring technology, after image data in a construction tunnel are collected in a targeted manner, the contour of the image data is identified, then disaster risks in the construction tunnel are monitored in real time based on contour image variability analysis, early warning is synchronously sent based on an identification result, it is guaranteed that constructors know the disaster risk state in the construction tunnel in real time, and the construction efficiency is improved. And stable tunnel construction under the geological scene of water-rich weathered granite gneiss is maintained.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel construction, and specifically to an intelligent prevention and control system for construction disaster risks of water-rich weathered granite gneiss tunnels. Background Technique

[0002] The construction of water-rich weathered granite gneiss tunnels is extremely challenging. Due to the water-rich rocks, it is easy to cause water inrush, collapse and other situations. During construction, it is necessary to conduct detailed geological surveys and adopt appropriate methods. For example, when using the drill and blast method, the blasting parameters must be strictly controlled. At the same time, advance support, waterproof and drainage treatment should be done well, and real-time monitoring and measurement should be carried out to ensure construction safety and project quality in all aspects.

[0003] The invention patent application with the application number 202310843463.3 discloses an intelligent analysis method for preventing and controlling geological disasters in tunnel construction; the method includes: constructing a knowledge graph model for preventing and controlling geological disasters in tunnel construction; structurally representing the knowledge related to disaster prevention and control: disaster prevention and control correlation analysis; based on the constructed knowledge graph model, mining the potential correlation between prevention and control countermeasures and geological disasters in tunnel construction through visual graph analysis technology; and optimizing prevention and control countermeasures based on the correlation relationship. This application aims to solve the problem that "under complex geological conditions, geological disasters in tunnel construction are affected by the coupling effect of multiple disaster sources, the disaster evolution mechanism is extremely complex, and the elements such as principles, measures, timing, materials, equipment, processes, etc. related to prevention and control countermeasures are closely related to the engineering geological, hydrogeological conditions and characteristics such as disaster type, location, scale, shape, nature, etc. of the disaster occurrence. Selecting prevention and control measures based on past experience or standards and specifications has great uncertainty and blindness".

[0004] However, for the construction of water-rich weathered granite gneiss tunnels, the construction difficulty is greater than that of tunnels with ordinary geological structures;

[0005] Therefore, an intelligent prevention and control system for construction disaster risks of water-rich weathered granite gneiss tunnels is proposed. Summary of the Invention

[0006] Aiming at the above-mentioned shortcomings of the existing technology, the present invention provides an intelligent prevention and control system for construction disaster risks of water-rich weathered granite gneiss tunnels, which can effectively solve the problems of the existing technology.

[0007] To achieve the above objectives, the present invention is realized through the following technical solutions;

[0008] The present invention discloses an intelligent prevention and control system for construction disaster risks of water-rich weathered granite gneiss tunnels, including:

[0009] A collection module, which is used to collect tunnel image data and store the tunnel image data; a feature extraction module, which is used to receive the image data stored in the collection module and extract the feature image of the image data; an iterative module, which is used to obtain the feature image of the image data extracted by the operation of the feature extraction module and transmit the feature image to the collection module to iterate each image data corresponding to each feature image; an analysis module, which is used to retrieve the feature image stored based on the iterative operation in the collection module and analyze the risk of tunnel construction disasters based on the feature image; a warning module, which is used to receive the analysis result of the risk of tunnel construction disasters in the analysis module and decide whether to send a warning message to the mobile computer device of the system-side user based on the analysis result:

[0010] Furthermore, a construction unit and a selection unit are arranged at the lower level of the collection module. The construction unit is used to upload the detection information of water-rich weathered granite gneiss in the tunnel construction area, construct a distribution model of water-rich weathered granite gneiss based on the detection information, and synchronously configure a real coordinate for the distribution model of water-rich weathered granite gneiss, so that the position coordinate of any position on the distribution model of water-rich weathered granite gneiss is consistent with its real position coordinate. The selection unit is used to traverse the distribution model of water-rich weathered granite gneiss and select image data collection points on the distribution model of water-rich weathered granite gneiss;

[0011] Among them, the detection information of water-rich weathered granite gneiss in the tunnel construction area is the distribution position information of water-rich weathered granite gneiss. The closed three-dimensional figure obtained by connecting the distribution position information of water-rich weathered granite gneiss is recorded as the distribution model of water-rich weathered granite gneiss;

[0012] Among them, the detection information is obtained by comprehensively detecting the tunnel construction area with a high-precision geological radar.

[0013] Furthermore, after constructing the distribution model of water-rich weathered granite gneiss, the construction unit synchronously divides the distribution model of water-rich weathered granite gneiss based on the preset tunnel construction area, so that the corresponding area of the tunnel construction area in the distribution model of water-rich weathered granite gneiss is separated in the distribution model of water-rich weathered granite gneiss. The system-side user performs the selection operation of the image data collection point in the selection unit on the distribution model of water-rich weathered granite gneiss after the separation operation is completed;

[0014] Among them, the system-side user presets an image data collection frequency in the collection module. The collection module performs continuous image data collection operations at the collection points based on the image data collection frequency. When the collection module stores the image data, it stores the image data separately based on the collection points where the image data comes from and sorts and stores the image data based on the collection time when the image data comes from.

[0015] Furthermore, the image data collection points also include:

[0016] Customize the cross-section capture spacing. On the water-rich weathered granite gneiss distribution model after the separation operation, based on the cross-section capture spacing, start capturing the cross-sections of the tunnel construction area from one end of the tunnel construction area, and divide the water-rich weathered granite gneiss distribution model based on the cross-sections to obtain several divided surfaces. Calculate the areas of each divided surface, sort the divided surfaces in descending order based on the area calculation results, and select the image data collection points again on the divided surfaces corresponding to the latter half of the descending order queue;

[0017] Among them, in the stage of setting the cross-section capture spacing, it follows that: the larger the volume of the water-rich weathered granite gneiss distribution model, the larger the cross-section capture spacing; the smaller the volume of the water-rich weathered granite gneiss distribution model, the smaller the cross-section capture spacing. And the prediction model based on machine learning uses the volume of the water-rich weathered granite gneiss distribution model, the geological structure complexity, historical construction disaster data, etc. as input features, and establishes a non-linear relationship model between the volume and the cross-section capture spacing through training and learning to complete the final setting of the cross-section capture spacing.

[0018] Furthermore, the calculation logic of the divided surface area is:

[0019] S split =S out -S in ;

[0020] In the formula: S split is the area of the divided surface; S out is the area of the region defined by the outer contour of the divided surface; S in is the cross-section area of the tunnel construction area;

[0021] Among them, when selecting the image data collection points on the divided surface, identify the distances from each point on the cross-section contour of the tunnel construction area to each point on the inner contour of the divided surface, and use the point on the cross-section contour of the tunnel construction area corresponding to the recognition result with the shortest distance as the image data collection point. And when calculating the area of the divided surface, use a method that combines pixel statistics and geometric calculation to convert it into a pixel matrix for pixel statistics to obtain an approximate area, and then apply geometric algorithms for precise correction; when selecting the collection points, use the object detection algorithm based on the deep neural network to automatically identify the cross-section contour of the tunnel construction area and the inner contour of the divided surface, and determine the shortest distance point by calculating the Euclidean distance between the point cloud data, which is used as the basic parameter in the calculation of the divided surface area.

[0022] Furthermore, during the acquisition stage of the image data collected at the same acquisition point, the acquisition distance and angle are kept consistent, and the feature image extracted by the feature extraction module from the image data is the contour image;

[0023] In the extraction stage of the feature image, the image data is first optimized and then the extraction operation of the feature image is performed;

[0024]

[0025] Where: O(x,y) is the optimized image data, that is, the image data after contour highlighting; I(x,y) is the original image data; α is the weight coefficient; is the Laplacian operator; * represents the convolution operation; K σ (x,y) is the Gaussian blur kernel function;

[0026] Among them, the weight coefficient α > 0, (x,y) represents the coordinate position in the kernel function, and σ represents the standard deviation of the Gaussian distribution;

[0027] Among them, in the optimization process, an adaptive weight adjustment mechanism is introduced. The adaptive weight adjustment mechanism dynamically adjusts the weight coefficient according to features such as the gray distribution and texture complexity of the image by constructing a weight prediction model based on a convolutional neural network.

[0028] Furthermore, the feature extraction module runs continuously. Each time the feature extraction module runs, the target of the received image data is all the image data stored in the same storage interval in the acquisition module;

[0029] In the stage of retrieving the feature image by the analysis module, the target of retrieving the feature image is all the feature images stored in the same storage interval in the acquisition module.

[0030] Furthermore, the tunnel construction disaster risk analysis logic in the analysis module is expressed as:

[0031]

[0032] Where: D is the difference between two feature images, that is, used to represent the tunnel construction disaster risk represented by the two feature images; ω1, ω2 are weights; S is the number of successfully matched key point pairs; M is the set of matched point pairs; (x 1i , y 1i ) is the coordinate of k1i in the feature image; (x 2i , y 2i ) is the coordinate of k 2i in the feature image; D curve is the contour curve similarity distance between two feature images;

[0033] Among them, two feature images extract a set of key points based on the SIFT algorithm, denoted as k1 = {k 11 , k 12 ,... k 1n}, k2 = {k 21 , k 22 ,... k 2n} and a set of matching point pairs M = {(k 1i , k 2i ) | i = 1,..., s; j = 1,..., s} is obtained through key point matching;

[0034] The contour curves of the two feature images are respectively discretized into point sets, and D is calculated through the dynamic time warping algorithm curve ;

[0035] Based on the above formula, the difference is calculated pairwise for all feature images corresponding to each acquisition point, the maximum difference value in the recognition calculation result is identified, a risk determination threshold is set, and the maximum difference value is compared with the risk determination threshold. The acquisition points where the maximum difference value is greater than or equal to the risk determination threshold are recorded as construction disaster risk points.

[0036] Furthermore, the content of the warning information issued by the warning module is the coordinates of the construction disaster risk points. The warning information is displayed in real time through the system interface and pushed to the system - end users through a preset communication method;

[0037] Among them, when the warning information is pushed, a message push strategy based on priority is adopted. The priority of the warning information is determined according to the risk level. The warning information with a high - risk level is pushed first, and at least one communication method is used for redundant push.

[0038] Furthermore, a construction unit and a selection unit are connected to the lower - level of the acquisition module through wireless network interaction. The acquisition module is connected to a feature extraction module and an iteration module through wireless network interaction. The acquisition module is connected to an analysis module through wireless network interaction, and the analysis module is connected to a warning module through wireless network interaction.

[0039] Adopting the technical solution provided by the present invention, compared with the known prior art, it has the following beneficial effects:

[0040] The present invention provides an intelligent prevention and control system for construction disaster risks in a water - rich weathered granite gneiss tunnel. During the operation of the system, through visual monitoring technology, after specifically collecting image data in the construction tunnel, the contour of the image data is recognized. Furthermore, based on the analysis of the variability of the contour image, the disaster risks in the construction tunnel are monitored in real time, and a warning is issued synchronously based on the recognition result, ensuring that construction personnel are aware of the disaster risk status in the construction tunnel in real time, maintaining the stable progress of tunnel construction in the water - rich weathered granite geological scenario, and ensuring the stable progress of tunnel construction. Brief Description of the Drawings

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0042] Figure 1 It is a schematic structural diagram of an intelligent prevention and control system for construction disaster risks in a water-rich weathered granite gneiss tunnel;

[0043] Figure 2 It is a schematic diagram of the area plane of the segmentation surface of the water-rich weathered granite gneiss distribution model in the present invention. Detailed implementation manners

[0044] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0045] The following further describes the present invention with reference to the embodiments.

[0046] Embodiment:

[0047] The intelligent prevention and control system for construction disaster risks in the water-rich weathered granite gneiss tunnel of this embodiment, as Figure 1 shown, includes:

[0048] An acquisition module, configured to acquire image data in the tunnel and store the image data in the tunnel;

[0049] A construction unit and a selection unit are arranged at a lower level of the acquisition module. The construction unit is configured to upload the detection information of water-rich weathered granite gneiss in the tunnel construction area, construct a water-rich weathered granite gneiss distribution model based on the detection information, and synchronously configure real coordinates for the water-rich weathered granite gneiss distribution model, so that the position coordinates of any position on the water-rich weathered granite gneiss distribution model are consistent with its real position coordinates. The selection unit is configured to traverse the water-rich weathered granite gneiss distribution model and select image data acquisition points on the water-rich weathered granite gneiss distribution model;

[0050] Among them, the detection information of water-rich weathered granite gneiss in the tunnel construction area is the distribution position information of water-rich weathered granite gneiss. The closed three-dimensional figure obtained by connecting the distribution position information of water-rich weathered granite gneiss is denoted as the water-rich weathered granite gneiss distribution model;

[0051] Among them, the detection information is obtained by comprehensively detecting the tunnel construction area through a high-precision geological radar;

[0052] After constructing the distribution model of water-rich weathered granite gneiss, the construction unit synchronously divides the distribution model of water-rich weathered granite gneiss based on the preset tunnel construction area, so that the corresponding area of the tunnel construction area in the distribution model of water-rich weathered granite gneiss is separated in the distribution model of water-rich weathered granite gneiss. The system-end user performs the selection operation of the image data acquisition point in the selection unit on the distribution model of water-rich weathered granite gneiss after the separation operation is completed;

[0053] Among them, the system-end user presets an image data acquisition frequency in the acquisition module. The acquisition module performs continuous image data acquisition operations at the acquisition points based on the image data acquisition frequency. When the acquisition module stores the image data, it stores them separately based on the acquisition points where the image data comes from and sorts and stores them based on the acquisition time of the image data source;

[0054] The image data acquisition points also include:

[0055] Custom cross-section capture spacing. On the distribution model of water-rich weathered granite gneiss after the separation operation is completed, based on the cross-section capture spacing, start capturing the cross-section of the tunnel construction area from one end of the tunnel construction area, divide the distribution model of water-rich weathered granite gneiss based on the cross-section to obtain several divided surfaces, calculate the area of each divided surface, sort the divided surfaces in descending order based on the area calculation results, and select the image data acquisition points again on the divided surfaces corresponding to the latter half of the descending order queue;

[0056] Among them, in the stage of setting the cross-section capture spacing, it follows that: the larger the volume of the distribution model of water-rich weathered granite gneiss, the larger the cross-section capture spacing; the smaller the volume of the distribution model of water-rich weathered granite gneiss, the smaller the cross-section capture spacing. And the prediction model based on machine learning uses the volume of the distribution model of water-rich weathered granite gneiss, the geological structure complexity, historical construction disaster data, etc. as input features, and establishes a non-linear relationship model between the volume and the cross-section capture spacing through training and learning to complete the final setting of the cross-section capture spacing;

[0057] The calculation logic of the divided surface area is:

[0058] S split =S out -S in ;

[0059] In the formula: S split is the area of the divided surface; S out is the area of the region defined by the outer contour of the divided surface; S in is the cross-sectional area of the tunnel construction area;

[0060] Among them, when selecting the image data acquisition points on the segmentation plane, the distances from each point on the cross-section contour of the tunnel construction area to each point on the inner contour of the segmentation plane are identified, and the points on the cross-section contour of the tunnel construction area corresponding to the recognition result with the shortest distance are used as the image data acquisition points. When calculating the area of the segmentation plane, a method that combines pixel statistics and geometric calculation is used to convert it into a pixel matrix for pixel statistics to obtain an approximate area, and then a geometric algorithm is applied for precise correction; when selecting the acquisition points, an object detection algorithm based on a deep neural network is used to automatically identify the cross-section contour of the tunnel construction area and the inner contour of the segmentation plane, and the shortest distance points are determined by calculating the Euclidean distance between the point cloud data, which serves as the basic parameter for calculating the segmentation plane area;

[0061] A feature extraction module, which is used to receive the image data stored in the acquisition module and extract the feature image of the image data;

[0062] During the acquisition stage of the image data collected at the same acquisition point, the acquisition distance and angle are kept consistent, and the feature image extracted by the feature extraction module from the image data is the contour image;

[0063] During the extraction stage of the feature image, the image data is first optimized and then the extraction operation of the feature image is performed;

[0064]

[0065] In the formula: O(x,y) is the optimized image data, that is, the image data after the contour is highlighted; I(x,y) is the original image data; α is the weight coefficient; is the Laplace operator; * represents the convolution operation; K σ (x,y) is the Gaussian blur kernel function;

[0066] Among them, the weight coefficient α > 0, (x,y) represents the coordinate position in the kernel function, and σ represents the standard deviation of the Gaussian distribution;

[0067] Among them, during the optimization process, an adaptive weight adjustment mechanism is introduced. The adaptive weight adjustment mechanism dynamically adjusts the weight coefficient by constructing a weight prediction model based on a convolutional neural network according to features such as the gray distribution and texture complexity of the image;

[0068] Through the above logical formula calculation, the image data collected at the acquisition point is optimized to highlight the contour information in the image data, thereby improving the accuracy of the system operation output result;

[0069] The feature extraction module runs continuously. Each time the feature extraction module runs, the target of the received image data is all the image data stored in the same storage interval in the acquisition module;

[0070] In the stage of retrieving the feature images by the analysis module, the target of retrieving the feature images is all the feature images stored in the same differentiated storage range in the acquisition module;

[0071] An iteration module, which is used to obtain the feature images of the image data extracted by the feature extraction module during operation, and transmit the feature images to the acquisition module to iterate each image data corresponding to each feature image;

[0072] An analysis module, which is used to retrieve the feature images stored based on the iterative operation in the acquisition module and analyze the risk of tunnel construction disasters based on the feature images;

[0073] The analysis logic of the risk of tunnel construction disasters in the analysis module is expressed as:

[0074]

[0075] In the formula: D is the difference between two feature images, that is, it is used to represent the risk of tunnel construction disasters shown by the two feature images; ω1, ω2 are weights; S is the number of successfully matched key point pairs; M is the set of matched point pairs; (x 1i , y 1i ) is the coordinate of k 1i in the feature image; (x 2i , y 2i ) is the coordinate of k 2i in the feature image; D curve is the similarity distance of the contour curves of the two feature images;

[0076] Among them, the key point sets are extracted from the two feature images based on the SIFT algorithm, denoted as k1 = {k 11 , k 12 ,... k 1n}, k2 = {k 21 , k 22 ,... k 2n}, and the set of matched point pairs is obtained through key point matching, M = {(k 1i , k 2i )|i = 1,..., s; j = 1,..., s};

[0077] The contour curves of the two feature images are respectively discretized into point sets, and D curve is calculated through the dynamic time warping algorithm;

[0078] Through the above logical formula, the calculation logic of the difference between feature images in the analysis module is further defined, providing support for the further operation of the warning module in this embodiment.

[0079] Based on the above formula, perform differential calculations in pairs for all feature images corresponding to each acquisition point, identify the maximum difference value in the calculation results, set a risk judgment threshold, compare the maximum difference value with the risk judgment threshold, and record the acquisition points where the maximum difference value is greater than or equal to the risk judgment threshold as construction disaster risk points;

[0080] An early warning module, which is used to receive the analysis results of tunnel construction disaster risk in the analysis module and decide whether to send early warning information to the mobile computer devices of system-end users based on the analysis results.

[0081] The content of the early warning information issued by the early warning module is the coordinates of the construction disaster risk points. The early warning information is displayed in real time through the system interface and pushed to system-end users through a preset communication method;

[0082] Among them, when pushing the early warning information, a message pushing strategy based on priority is adopted. The priority of the early warning information is determined according to the risk level. The early warning information with a high risk level is pushed first, and at least one communication method is used for redundant pushing;

[0083] The lower level of the acquisition module is connected with a construction unit and a selection unit through wireless network interaction. The acquisition module is connected with a feature extraction module and an iteration module through wireless network interaction. The acquisition module is connected with an analysis module through wireless network interaction. The analysis module is connected with an early warning module through wireless network interaction.

[0084] In this embodiment, the acquisition module runs to acquire image data in the tunnel, stores the image data in the tunnel, the construction unit synchronously uploads the detection information of water-rich weathered granite gneiss in the tunnel construction area, constructs a water-rich weathered granite gneiss distribution model based on the detection information, and synchronously configures real coordinates for the water-rich weathered granite gneiss distribution model, so that the position coordinates of any position on the water-rich weathered granite gneiss distribution model are consistent with its position coordinates in reality. The selection unit traverses the water-rich weathered granite gneiss distribution model in real time, selects image data acquisition points on the water-rich weathered granite gneiss distribution model, and the feature extraction module runs later to receive the image data stored in the acquisition module and extract the feature images of the image data. The feature extraction module further receives the image data stored in the acquisition module and extracts the feature images of the image data. Then, the analysis module retrieves the feature images stored based on iterative operations in the acquisition module, analyzes the tunnel construction disaster risk based on the feature images, and finally the early warning module receives the analysis results of the tunnel construction disaster risk in the analysis module and decides whether to send early warning information to the mobile computer devices of system-end users;

[0085] Through the operation of the system in the above embodiment, it brings intelligent prevention and control of disaster risks to the construction scenario of water-rich weathered granite gneiss tunnels, ensuring the stable progress of the construction of water-rich weathered granite gneiss tunnels.

[0086] See Figure 2 As shown, the shaded part in this figure represents the area of the dividing surface of the water-rich weathered granite gneiss distribution model.

[0087] In summary, during the operation of the system in the above embodiments, through visual monitoring technology, after specifically collecting image data in the construction tunnel, the contour of the image data is recognized. Then, based on the analysis of the contour image variability, the disaster risk in the construction tunnel is monitored in real time, and an early warning is issued synchronously based on the recognition result, ensuring that construction workers are aware of the disaster risk status in the construction tunnel in real time, maintaining the stable progress of tunnel construction in the geological scenario of water-rich weathered granite gneiss, and ensuring the stable progress of tunnel construction.

[0088] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent prevention and control system for construction disaster risks of water-rich weathered granite gneiss tunnels, characterized in that, Including: A collection module for collecting tunnel image data and storing the tunnel image data; A construction unit and a selection unit are arranged at a lower level of the collection module. The construction unit is used to upload the detection information of water-rich weathered granite gneiss in the tunnel construction area, construct a distribution model of water-rich weathered granite gneiss based on the detection information, and synchronously configure real coordinates for the distribution model of water-rich weathered granite gneiss, so that the position coordinates of any position on the distribution model of water-rich weathered granite gneiss are consistent with its real position coordinates. The selection unit is used to traverse the distribution model of water-rich weathered granite gneiss and select image data acquisition points on the distribution model of water-rich weathered granite gneiss; A feature extraction module for receiving the image data stored in the collection module and extracting the feature image of the image data; An iteration module for obtaining the feature image of the image data extracted by the operation of the feature extraction module and transmitting the feature image to the collection module to iterate each image data corresponding to each feature image; An analysis module for retrieving the feature image stored based on the iteration operation in the collection module and analyzing the risk of tunnel construction disasters based on the feature image; An early warning module for receiving the analysis result of the risk of tunnel construction disasters in the analysis module and deciding whether to send an early warning message to the mobile computer device of the system-end user based on the analysis result.

2. The intelligent prevention and control system for construction disaster risks of water-rich weathered granite gneiss tunnels according to claim 1, characterized in that The detection information of water-rich weathered granite gneiss in the tunnel construction area is the distribution position information of water-rich weathered granite gneiss. The closed three-dimensional figure obtained by connecting the distribution position information of water-rich weathered granite gneiss is denoted as the distribution model of water-rich weathered granite gneiss; Among them, the detection information is obtained by comprehensively detecting the tunnel construction area with a high-precision geological radar.

3. The intelligent prevention and control system for construction disaster risks of water-rich weathered granite gneiss tunnels according to claim 2, wherein, After constructing the distribution model of water-rich weathered granite gneiss, the construction unit synchronously divides the distribution model of water-rich weathered granite gneiss based on the preset tunnel construction area, so that the corresponding area of the tunnel construction area in the distribution model of water-rich weathered granite gneiss is separated in the distribution model of water-rich weathered granite gneiss. The system-end user performs the selection operation of the image data acquisition point in the selection unit on the distribution model of water-rich weathered granite gneiss after the separation operation is completed; Among them, the system-end user presets an image data collection frequency in the collection module. The collection module performs continuous image data collection operations at the collection points based on the image data collection frequency. When the collection module stores the image data, it stores the image data separately based on the collection points where the image data comes from and sorts and stores the image data based on the collection time of the image data source.

4. The intelligent prevention and control system for construction disaster risks of water-rich weathered granite gneiss tunnels according to claim 1, characterized in that, The image data acquisition points also include: Custom cross-section capture spacing. On the distribution model of water-rich weathered granite gneiss after the separation operation, based on the cross-section capture spacing, start capturing the cross-section of the tunnel construction area from one end of the tunnel construction area, divide the distribution model of water-rich weathered granite gneiss based on the cross-section to obtain several divided surfaces, calculate the area of each divided surface, sort the divided surfaces in descending order based on the area calculation result, and select image data acquisition points again on the divided surfaces corresponding to the latter half of the descending order queue; Among them, in the stage of setting the cross-section capture spacing, it follows that: the larger the volume of the water-rich weathered granite gneiss distribution model, the larger the cross-section capture spacing; the smaller the volume of the water-rich weathered granite gneiss distribution model, the smaller the cross-section capture spacing. And the prediction model based on machine learning takes the volume of the water-rich weathered granite gneiss distribution model, the complexity of geological structures, historical construction disaster data, etc. as input features, and establishes a non-linear relationship model between the volume and the cross-section capture spacing through training and learning to complete the final setting of the cross-section capture spacing.

5. The intelligent prevention and control system for construction disaster risks of water-rich weathered granite gneiss tunnels according to claim 4, wherein The calculation logic of the split surface area is as follows: S split = S out - S in ; Where: S split is the area of the dividing surface; S out is the area of the region bounded by the outer contour of the dividing surface; S in is the cross-sectional area of the tunnel construction area; Among them, when selecting the image data acquisition points on the split surface, the distances from each point on the cross-section contour of the tunnel construction area to each point on the inner contour of the split surface are identified, and the point on the cross-section contour of the tunnel construction area corresponding to the recognition result with the shortest distance is used as the image data acquisition point. And when calculating the split surface area, a method that combines pixel statistics and geometric calculation is used to convert it into a pixel matrix for pixel statistics to obtain an approximate area, and then a geometric algorithm is applied for precise correction; when selecting the acquisition points, an object detection algorithm based on a deep neural network is used to automatically identify the cross-section contour of the tunnel construction area and the inner contour of the split surface, and the Euclidean distance between the point cloud data is calculated to determine the shortest distance point, which is used as the basic parameter in the calculation of the split surface area.

6. The intelligent prevention and control system for construction disaster risks of water-rich weathered granite gneiss tunnels according to claim 1, wherein During the acquisition stage of the image data collected at the same acquisition point, the acquisition distance and angle are kept consistent, and the feature image extracted by the feature extraction module from the image data is the contour image; During the extraction stage of the feature image, the image data is first optimized, and then the extraction operation of the feature image is performed; Where: O(x, y) is the image data after optimization processing, that is, the image data after contour highlighting; I(x, y) is the original image data; α is the weight coefficient; is the Laplacian operator; * represents the convolution operation; K σ (x, y) is the Gaussian blur kernel function; The weight coefficient α > 0, (x, y) represents the coordinate position in the kernel function, and σ represents the standard deviation of the Gaussian distribution; Among them, during the optimization process, an adaptive weight adjustment mechanism is introduced. The adaptive weight adjustment mechanism dynamically adjusts the weight coefficient by constructing a weight prediction model based on a convolutional neural network according to features such as the gray distribution and texture complexity of the image.

7. The intelligent prevention and control system for construction disaster risks of water-rich weathered granite gneiss tunnels according to claim 1, wherein The feature extraction module runs continuously. Each time the feature extraction module runs, the target of the received image data is all the image data stored in the same differentiated storage interval in the acquisition module; During the stage when the analysis module retrieves the feature image, the target of retrieving the feature image is all the feature images stored in the same differentiated storage interval in the acquisition module.

8. The intelligent prevention and control system for construction disaster risks of water-rich weathered granite gneiss tunnels according to claim 1, wherein The risk analysis logic of tunnel construction disasters in the analysis module is expressed as: Where: D is the difference between two feature images, that is, it is used to represent the tunnel construction disaster risk presented by the two feature images; ω1, ω2 are weights; S is the number of successfully matched key point pairs; M is the set of matched point pairs; (x 1i , y 1i ) is the coordinate of k 1i in the feature image; (x 2i , y 2i ) is the coordinate of k 2i in the feature image; D curve is the similarity distance of the contour curves of the two feature images; Among them, key point sets are extracted from two feature images based on the SIFT algorithm, denoted as k1 = {k 11 , k 12 ,... k 1n}, k2 = {k 21 , k 22 ,... k 2n}. After key point matching, a set of matching point pairs is obtained, M = {(k 1i , k 2i ) | i = 1,..., s; j = 1,..., s}; The contour curves of two feature images are respectively discretized into point sets, and D is obtained by calculating through the dynamic time warping algorithm curve , and D curve is involved in the process of tunnel construction disaster risk analysis; Based on the above formula, the difference is calculated pairwise for all the feature images corresponding to each acquisition point, and the maximum difference value in the calculation results is identified. A risk determination threshold is set, and the maximum difference value is compared with the risk determination threshold. The acquisition points where the maximum difference value is greater than or equal to the risk determination threshold are recorded as construction disaster risk points.

9. The intelligent prevention and control system for construction disaster risks of water-rich weathered granite gneiss tunnels according to claim 1, characterized in that, The content of the warning information sent by the warning module is the coordinates of the construction disaster risk points. The warning information is displayed in real time through the system interface and is pushed to the system-end users through a preset communication method; Among them, when pushing the warning information, a message pushing strategy based on priority is adopted. The priority of the warning information is determined according to the risk level. The warning information with a high risk level is pushed first, and at least one communication method is used for redundant pushing.

10. The intelligent prevention and control system for construction disaster risks of water-rich weathered granite gneiss tunnels according to claim 1, wherein, The lower level of the acquisition module is connected with a construction unit and a selection unit through wireless network interaction. The acquisition module is connected with a feature extraction module and an iterative module through wireless network interaction. The acquisition module is connected with an analysis module through wireless network interaction. The analysis module is connected with an early warning module through wireless network interaction.

Images