Tunnel working face geological risk prediction method and device based on multi-source information fusion

Abstract

The invention provides a tunnel working face geological risk prediction method and device based on multi-source information fusion, and the method comprises the steps: obtaining a tunnel excavation working face image, and carrying out the preprocessing of the tunnel excavation working face image, and obtaining a tunnel excavation working face target image; inputting the target image of the tunnel excavation working face into a trained groundwater flow state recognition model to obtain a groundwater flow state classification result corresponding to the image of the tunnel excavation working face; inputting the target image of the tunnel excavation working face into a trained fracture crushing degree identification model to obtain a rock mass crushing degree classification result corresponding to the image of the tunnel excavation working face; and obtaining a geological condition comprehensive score of the tunnel excavation working face image based on an underground water flow state classification result and the rock mass crushing degree classification result, and determining a geological risk level corresponding to the tunnel excavation working face image based on the geological condition comprehensive score. According to the invention, the accuracy and efficiency of tunnel working face geological risk prediction can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of geological risk prediction, and in particular to a method and device for predicting geological risks of a tunnel working face based on multi-source information fusion. Background Art

[0002] Tunnel engineering belongs to underground structures. Underground structures are diverse, and the construction methods and technologies for building underground structures are also diverse; the construction methods of tunnels are related to the characteristics of underground structures. The characteristics of tunnel construction are: high concealment, many unknown factors; limited working space, narrow working surface, and large interference in construction process; strong cyclicity of construction process operations, because tunnel engineering is longitudinal, construction strictly follows a certain sequence of cyclic operations, such as excavation must be cyclic in the order of "drilling-charging-blasting-ventilation-slag removal"; the comprehensiveness of construction operations is strong, and multiple processes (excavation, support, lining, etc.) are carried out in the same working environment; the geological mechanical state of the construction process is changing, and the physical and mechanical properties of the surrounding rock are also changing, so the construction is dynamic; the working environment is harsh, the working space is narrow, the construction noise is loud, dust, smoke, humidity, dark light, poor geological conditions and safety issues bring unfavorable working environment to the construction personnel; the operation risk is high, and the risk is related to concealment and dynamism. During the construction process, the construction personnel must pay attention to the risk of tunnel construction at any time.

[0003] In the process of tunnel construction, groundwater is a common risk factor, especially when passing through aquifers or rock formations with developed fissures, water inrush may cause serious safety hazards. At present, the risks of water inrush and fissures in tunnels to tunnel construction are mainly judged by manual observation combined with experience. This judgment method is highly subjective, has high lag and limited coverage, which makes the geological risk prediction results of the tunnel working face low in accuracy and the risk prediction is not timely; therefore, how to timely and accurately predict the geological risks of the tunnel working face is a technical problem that needs to be solved urgently. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a method and device for predicting geological risks of a tunnel working face based on multi-source information fusion, so as to eliminate or improve one or more defects existing in the prior art.

[0005] One aspect of the present invention provides a method for predicting geological risks of a tunnel working face based on multi-source information fusion, the method comprising the following steps:

[0006] Acquire a tunnel excavation working face image, and pre-process the tunnel excavation working face image to obtain a tunnel excavation working face target image;

[0007] Inputting the tunnel excavation working face target image into a trained groundwater flow pattern recognition model to obtain a groundwater flow pattern classification result corresponding to the tunnel excavation working face image;

[0008] Inputting the tunnel excavation working face target image into a trained fracture fragmentation degree recognition model to obtain a rock mass fragmentation degree classification result corresponding to the tunnel excavation working face image;

[0009] A comprehensive geological condition score of the tunnel excavation working face image is obtained based on the groundwater flow state classification result and the rock mass fragmentation degree classification result, and a geological risk level corresponding to the tunnel excavation working face image is determined based on the comprehensive geological condition score.

[0010] In some embodiments of the present invention, preprocessing the tunnel excavation working face image to obtain a tunnel excavation working face target image includes:

[0011] De-noising the tunnel excavation working face image based on a non-local mean denoising method;

[0012] Adaptive histogram equalization is performed on the denoised tunnel excavation working face image to obtain a tunnel excavation working face image with enhanced image contrast;

[0013] The tunnel excavation working face image after image contrast enhancement is geometrically corrected to obtain the tunnel excavation working face target image.

[0014] In some embodiments of the present invention, the groundwater flow regime identification model includes a U2Net network and a classifier.

[0015] In some embodiments of the present invention, the target image of the tunnel excavation working face is input into a trained fracture fragmentation degree recognition model to obtain a rock mass fragmentation degree classification result corresponding to the tunnel excavation working face image, including:

[0016] Determining cracks in the target image of the tunnel excavation working face based on an edge detection algorithm;

[0017] Dividing the tunnel excavation working face target image into a plurality of grids, and calculating the length, width and number of cracks in each grid;

[0018] Performing cluster analysis on the plurality of grids by using a clustering algorithm based on the length, width and number of cracks in each grid;

[0019] The rock mass fragmentation degree classification result corresponding to the tunnel excavation working face image is determined based on the cluster analysis result.

[0020] In some embodiments of the present invention, obtaining a comprehensive geological condition score of the tunnel excavation working face image based on the groundwater flow state classification result and the rock mass fragmentation degree classification result includes:

[0021] Determining a water flow state score based on the groundwater flow state classification result;

[0022] Determining a degree of fragmentation score based on the rock mass fragmentation degree classification result;

[0023] A comprehensive score of the geological conditions of the tunnel excavation working face image is calculated based on the water flow state score and the fragmentation degree score.

[0024] In some embodiments of the present invention, the calculation formula for the comprehensive geological condition score is:

[0025] S total =W water ×S water +W fracture ×S fracture；

[0026] Among them, S total S is the comprehensive score of geological conditions. water S is the water flow rate score, S fracture The degree of fragmentation is scored, W water is the water flow state weight, W fracture is the weight of the degree of fragmentation.

[0027] In some embodiments of the present invention, preprocessing the tunnel excavation working face image to obtain a tunnel excavation working face target image further includes:

[0028] Obtain tunnel excavation face images at different resolutions and scales;

[0029] The multiple tunnel excavation working face images with different resolutions and different scales are fused.

[0030] In some embodiments of the present invention, the groundwater flow state classification result is dry, water-containing or flowing water; and / or,

[0031] The rock mass fragmentation degree classification results are broken, relatively broken or complete.

[0032] Another aspect of the present invention provides a tunnel working face geological risk prediction system based on multi-source information fusion, including a processor, a memory and a computer program stored in the memory, the processor is used to execute the computer program, and when the computer program is executed, the system implements the steps of the method described in any of the above embodiments.

[0033] Another aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method described in any of the above embodiments when executed by a processor.

[0034] The method and device for predicting geological risks of tunnel working faces based on multi-source information fusion of the present invention identifies groundwater flow state classification results based on a groundwater flow state recognition model, and identifies rock mass fragmentation degree classification results based on a fracture fragmentation degree recognition model, and finally determines the final geological risk level based on the groundwater flow state classification results and the rock mass fragmentation degree classification results. The method and device use advanced computer vision technology to comprehensively monitor the tunnel working face, and can predict the potential geological risks of the tunnel working face more timely and accurately, so that effective prevention and response measures can be taken in a timely manner to reduce the accident rate and improve the construction efficiency of the tunnel.

[0035] Additional advantages, purposes, and features of the present invention will be described in part in the following description, and will become apparent to those skilled in the art after studying the following, or may be learned from the practice of the present invention. The purposes and other advantages of the present invention may be achieved and obtained by the structures specifically indicated in the specification and the accompanying drawings.

[0036] Those skilled in the art will appreciate that the objectives and advantages that can be achieved with the present invention are not limited to the above specific description, and the above and other objectives that can be achieved by the present invention will be more clearly understood from the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The drawings described herein are used to provide a further understanding of the present invention, constitute a part of this application, and do not constitute a limitation of the present invention. The components in the drawings are not drawn to scale, but are only for illustrating the principles of the present invention. In order to facilitate the illustration and description of some parts of the present invention, the corresponding parts in the drawings may be enlarged, that is, they may become larger relative to other components in the exemplary device actually manufactured according to the present invention. In the drawings:

[0038] Figure 1 It is a flow chart of a method for predicting geological risks of a tunnel working face based on multi-source information fusion in one embodiment of the present invention.

[0039] Figure 2 It is a schematic diagram of the architecture of a tunnel working face geological risk prediction system based on multi-source information fusion in one embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0041] It should also be noted that, in order to avoid obscuring the present invention due to unnecessary details, only structures and / or processing steps closely related to the solutions according to the present invention are shown in the accompanying drawings, while other details that are not closely related to the present invention are omitted.

[0042] It should be emphasized that the term “include / comprises” when used herein refers to the presence of features, elements, steps or components, but does not exclude the presence or addition of one or more other features, elements, steps or components.

[0043] It should also be noted that, unless otherwise specified, the term “connection” herein may refer not only to a direct connection but also to an indirect connection involving an intermediate.

[0044] At present, the traditional geological risk prediction of tunnel working faces mainly relies on manual observation and experience judgment. As a result, due to the different experience and cognition of different observers, there are differences in the judgment of the same phenomenon, resulting in inconsistent early warning results of geological risks; in addition, the existing manual observation method cannot achieve real-time monitoring, and it is easy to miss the best early warning time; and manual observation is usually only for local areas, and it is difficult to fully monitor the entire working face of the tunnel; based on the above situation, it can be seen that the existing traditional geological risk prediction method of tunnel working faces has the problems of low prediction accuracy and untimely early warning, which leads to low safety and efficiency of tunnel construction. In order to solve the above technical problems of the geological risk prediction method of tunnel working faces in the prior art, the present application provides a method and device for predicting geological risks of tunnel working faces based on multi-source information fusion. The method and device use advanced computer vision technology to comprehensively monitor the tunnel working face during the tunnel construction process, including the identification of groundwater flow state and the identification of the degree of fragmentation of the rock structure surface; by combining these two aspects of information, a comprehensive scoring mechanism is established, and then a comprehensive early warning system is formed, which ultimately realizes real-time and accurate early warning of geological risks of tunnel working faces.

[0045] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the accompanying drawings, the same reference numerals represent the same or similar components, or the same or similar steps.

[0046] Figure 1 FIG. 1 is a flow chart of a method for predicting geological risks of a tunnel working face based on multi-source information fusion according to an embodiment of the present application. Figure 1As shown, the method for predicting geological risks of a tunnel working face based on multi-source information fusion includes at least steps S10 to S40.

[0047] Step S10: Acquire a tunnel excavation working face image, and pre-process the tunnel excavation working face image to obtain a tunnel excavation working face target image.

[0048] In this step, high-resolution camera equipment suitable for the tunnel environment can be used to collect images of the tunnel excavation working face to ensure that clear images can be collected in low-light environments. In addition, considering the environment inside the tunnel, an industrial-grade camera with waterproof and dustproof characteristics can be used to collect images of the tunnel excavation working face.

[0049] In addition, the preprocessing of the tunnel excavation working face image may include image denoising, image enhancement, etc.; in one embodiment, the preprocessing of the tunnel excavation working face image to obtain the tunnel excavation working face target image may include: denoising the tunnel excavation working face image based on the non-local mean denoising method; performing adaptive histogram equalization on the denoised tunnel excavation working face image to obtain the tunnel excavation working face image after image contrast enhancement; performing geometric correction on the tunnel excavation working face image after image contrast enhancement to obtain the tunnel excavation working face target image. In this embodiment, an advanced denoising algorithm - Non-Local Means algorithm is used to reduce noise points in the image and enhance image quality; adaptive histogram equalization (CLAHE) and other technologies are used to improve the image contrast and highlight the characteristics of groundwater flow; in addition, in order to correct the image distortion caused by the camera angle or the equipment installation position and ensure the accuracy of the image data, the tunnel excavation working face image collected by the image acquisition device is further geometrically corrected. It can be understood that the image preprocessing method and steps listed in this embodiment are merely examples. In other embodiments, specific tunnel excavation working face images can be preprocessed based on actual needs, as long as the target image of the tunnel excavation working face obtained after preprocessing can meet the requirements of the groundwater flow state recognition model and the fracture fragmentation degree recognition model for the input image.

[0050] In another embodiment, the preprocessing of the tunnel excavation working face image includes not only denoising, geometric correction, etc., but also the following steps: obtaining tunnel excavation working face images of different resolutions and scales; fusing multiple tunnel excavation working face images of different resolutions and scales. This embodiment further fuses images of different resolutions and scales, thereby improving the ability to express image details, thereby better extracting feature information from the image.

[0051] Step S20: inputting the tunnel excavation working face target image into a trained groundwater flow pattern recognition model to obtain a groundwater flow pattern classification result corresponding to the tunnel excavation working face image.

[0052] In this step, the groundwater flow state in the target image of the tunnel excavation working face is identified based on the pre-trained groundwater flow state recognition model. The groundwater flow state can be divided into three types: dry, water-containing or flowing. Therefore, the groundwater flow state classification result corresponding to the target image of the tunnel excavation working face obtained in this step is dry, water-containing or flowing.

[0053] In one embodiment, the groundwater flow pattern recognition model includes a U2Net network and a classifier. The U2Net network is a powerful image segmentation network that can accurately segment the area of ​​interest under complex backgrounds. In addition, the output of the U2Net network is connected to the classifier, and the area of ​​interest information output by the U2Net network is further sent to the classifier for classification, thereby obtaining the groundwater flow pattern classification result to which the target image of the tunnel excavation working face belongs.

[0054] Furthermore, the classifier may be a softmax classifier, that is, the groundwater flow state recognition model includes a U2Net network and a softmax classifier. In this embodiment, the target image of the tunnel excavation working face is analyzed using computer vision technology based on the U2Net network to identify the groundwater flow state, and the groundwater flow state category is obtained based on the softmax classifier. The groundwater flow state category may be dry, water-containing or flowing water, etc.

[0055] In addition, in order to obtain a trained groundwater flow state identification model, a tunnel working face geological risk prediction method of an embodiment may also include the following steps: constructing an initial groundwater flow state identification model and a first sample data set, and each sample data in the first sample data set includes a tunnel excavation working face target image sample and a groundwater flow state result sample; further pre-training the initial groundwater flow state identification model based on the constructed first sample data set to update the model parameters of the initial groundwater flow state identification model to obtain a trained groundwater flow state identification model.

[0056] Step S30: inputting the tunnel excavation working face target image into a trained fracture fragmentation degree recognition model to obtain a rock mass fragmentation degree classification result corresponding to the tunnel excavation working face image.

[0057] In this step, the degree of rock fragmentation in the target image of the tunnel excavation working face is identified based on the pre-trained fracture fragmentation degree recognition model. The degree of rock fragmentation can be divided into three levels: broken, relatively broken and complete. Therefore, the rock fragmentation degree classification result corresponding to the target image of the tunnel excavation working face obtained in this step is broken, relatively broken or complete.

[0058] In one embodiment, the target image of the tunnel excavation working face is input into a trained crack fragmentation degree recognition model to obtain a classification result of the rock mass fragmentation degree corresponding to the tunnel excavation working face image, including: determining the cracks of the target image of the tunnel excavation working face based on an edge detection algorithm; dividing the target image of the tunnel excavation working face into multiple grids, calculating the length, width and number of cracks in each grid; clustering the multiple grids based on the length, width and number of cracks in each grid through a clustering algorithm; and determining the classification result of the rock mass fragmentation degree corresponding to the tunnel excavation working face image based on the clustering analysis results. In this embodiment, the target image of the tunnel excavation working face is first processed based on an edge detection algorithm to identify the cracks on the structural surface, and further the crack length is statistically analyzed based on a gridding method to evaluate the degree of rock mass fragmentation.

[0059] Exemplarily, the edge detection algorithm used may be specifically the Laplace edge detection algorithm, which is a classic edge detection method that can effectively identify edge features in an image. Furthermore, the target image of the tunnel excavation working face processed by the Laplace edge detection algorithm is segmented according to a certain grid size to form structured statistical units. In each grid unit, the length, width and number of cracks are counted to quantify the degree of rock mass fragmentation. In this example, the Laplace edge detection algorithm is used to process the target image of the tunnel excavation working face. The Laplace edge detection algorithm identifies the edge by calculating the grayscale change rate of the image, and can effectively detect the cracks; it can be understood that in other embodiments, other edge detection algorithms besides the Laplace edge detection algorithm can also be used to identify cracks in the image.

[0060] In addition, when calculating the length and width of the cracks in each grid, an image processing algorithm can also be used for statistical calculation; after the length, width and number of the cracks in each grid unit are obtained by statistics, a clustering algorithm can be further used based on the statistical results to classify the degree of rock mass fragmentation, so as to identify whether the rock mass fragmentation degree of the target image of the tunnel excavation working face is broken, relatively broken or complete. Specifically, for multiple grid units, multiple grid units can be clustered according to the length, width and number of cracks in each grid unit, and grid units with similar length, width and number of cracks are classified into one category, and finally the category containing the largest number of grids is used as the category of rock mass fragmentation degree of the target image of the tunnel excavation working face; for example, when the number of grid units classified into the broken category is the largest, it is judged that the rock mass fragmentation degree of the target image of the tunnel excavation working face is broken. Exemplarily, the clustering algorithm used in this embodiment may be a K-means algorithm, that is, the degree of rock fragmentation in the target image of the tunnel excavation working face is classified based on the K-means algorithm; it can be understood that the clustering algorithm listed in this example is only an example, and in other embodiments, other clustering algorithms may also be used to achieve the classification of the degree of rock fragmentation.

[0061] Similarly, in order to obtain a trained fracture degree recognition model, a tunnel working face geological risk prediction method of one embodiment may also include the following steps: constructing an initial fracture degree recognition model and a second sample data set, and each sample data in the second sample data set includes a tunnel excavation working face target image sample and a rock mass fracture degree classification result sample; further pre-training the initial fracture degree recognition model based on the constructed second sample data set to update the model parameters of the initial fracture degree recognition model to obtain a trained fracture degree recognition model.

[0062] Step S40: obtaining a comprehensive geological condition score of the tunnel excavation working face image based on the groundwater flow state classification result and the rock mass fragmentation degree classification result, and determining a geological risk level corresponding to the tunnel excavation working face image based on the comprehensive geological condition score.

[0063] In this step, a comprehensive risk assessment system is established to combine the groundwater flow state classification results and the rock mass fragmentation degree classification results to comprehensively score the tunnel excavation working face image and obtain a geological risk level that matches the comprehensive score. In addition, before determining the geological risk level, the corresponding relationship between the comprehensive score and the geological risk level can be pre-constructed.

[0064] In one embodiment, obtaining a comprehensive geological condition score of the tunnel excavation working face image based on the groundwater flow state classification result and the rock mass fragmentation degree classification result includes: determining a water flow state score based on the groundwater flow state classification result; determining a fragmentation degree score based on the rock mass fragmentation degree classification result; and calculating a comprehensive geological condition score of the tunnel excavation working face image based on the water flow state score and the fragmentation degree score. Similarly, in order to determine a water flow state score and a fragmentation degree score based on the groundwater flow state classification result and the rock mass fragmentation degree classification result, respectively, the corresponding relationship between the groundwater flow state classification result and the water flow state score, and the corresponding relationship between the rock mass fragmentation degree classification result and the fragmentation degree score can be pre-constructed.

[0065] Exemplarily, the calculation formula for the comprehensive geological condition score is:

[0066] S total =W water ×S water +W fracture ×S fracture；

[0067] Among them, S total S is the comprehensive score of geological conditions. water S is the water flow rate score, S fracture The degree of fragmentation is scored, W water is the water flow state weight, W fracture is the weight of the degree of fragmentation.

[0068] In the above embodiment, different weights are set for groundwater flow patterns and structural surface fragmentation according to actual engineering requirements, and the water flow pattern score and the fragmentation degree score are combined based on the comprehensive scoring formula to calculate the comprehensive score of the geological conditions of the rock mass. When determining the water flow pattern weight and the fragmentation degree weight, the optimal weight distribution can be determined based on expert experience or historical data analysis; in addition, the comprehensive scoring model can be dynamically adjusted based on real-time data based on a machine learning algorithm to optimize the accuracy of the comprehensive score.

[0069] Furthermore, based on the correspondence between the comprehensive score and the geological risk specification, multiple warning levels (the warning level can also be understood as the risk level) can be set, such as a low warning level, a medium warning level and a high warning level. Each warning level corresponds to a different comprehensive score. After the geological condition comprehensive score of the tunnel excavation working face image is determined based on the above embodiment, the calculated geological condition comprehensive score can be mapped to the warning level based on the fuzzy logic algorithm. In addition, a warning value can be further set. When the warning level corresponding to the geological condition comprehensive score reaches the warning value, the warning of the geological risk of the tunnel working face is realized. In addition, in order to further improve the reliability of the warning module, in one embodiment, the warning value can be optimized in real time based on historical data and a method based on big data analysis.

[0070] Through the above embodiments, it can be found that the tunnel working face geological risk prediction method based on multi-source information fusion of the present application adopts the advanced U2Net network to realize the identification of groundwater flow state, and adopts the Laplace edge detection algorithm to realize the identification of cracks, which improves the accuracy and reliability of geological risk identification and reduces human errors. Through comprehensive scoring and early warning indicators, it can provide scientific and objective risk assessment and early warning, help construction management personnel take timely measures to prevent accidents; in addition, the method also realizes automated monitoring and early warning, reduces manual intervention, and improves work efficiency and safety.

[0071] Correspondingly, the present invention also provides a tunnel working face geological risk prediction system based on multi-source information fusion, including a processor, a memory and a computer program stored in the memory, the processor is used to execute the computer program, and when the computer program is executed, the system implements the steps of the method described in any of the above embodiments.

[0072] For example, Figure 2 As shown, the system also includes at least an image acquisition module 100, a water flow state recognition module 200, a fragmentation degree recognition module 300, and a geological risk assessment module 400. The image acquisition module 100 is used to obtain a tunnel excavation working face image, and pre-process the tunnel excavation working face image to obtain a tunnel excavation working face target image; the water flow state recognition module 200 is used to obtain a groundwater flow state classification result corresponding to the tunnel excavation working face image based on a trained groundwater flow state recognition model; the fragmentation degree recognition module 300 is used to obtain a rock mass fragmentation degree classification result corresponding to the tunnel excavation working face image based on a trained fracture fragmentation degree recognition model; and the geological risk assessment module 400 is used to obtain a comprehensive score of the geological conditions of the tunnel excavation working face image based on the groundwater flow state classification result and the rock mass fragmentation degree classification result, and determine the geological risk level corresponding to the tunnel excavation working face image based on the comprehensive score of the geological conditions.

[0073] The geological risk prediction system of the above-mentioned embodiment can monitor the groundwater flow state and the degree of structural surface fragmentation of the tunnel working face in real time, realize all-weather and full-coverage risk monitoring, and through the comprehensive scoring of groundwater flow state and the degree of structural surface fragmentation, more comprehensively and accurately evaluate the geological risk of the tunnel working face.

[0074] The embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the aforementioned edge computing server deployment method are implemented. The computer-readable storage medium can be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the technical field.

[0075] It should be understood by those skilled in the art that the exemplary components, systems and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software or a combination of the two. Whether it is performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link via a data signal carried in a carrier.

[0076] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.

[0077] In the present invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with features of other embodiments or replace features of other embodiments.

[0078] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the embodiments of the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for predicting geological risks of tunnel working faces based on multi-source information fusion, characterized in that: The method comprises the following steps: Acquire a tunnel excavation working face image, and pre-process the tunnel excavation working face image to obtain a tunnel excavation working face target image; Inputting the tunnel excavation working face target image into a trained groundwater flow pattern recognition model to obtain a groundwater flow pattern classification result corresponding to the tunnel excavation working face image; Inputting the tunnel excavation working face target image into a trained fracture fragmentation degree recognition model to obtain a rock mass fragmentation degree classification result corresponding to the tunnel excavation working face image; A comprehensive geological condition score of the tunnel excavation working face image is obtained based on the groundwater flow state classification result and the rock mass fragmentation degree classification result, and a geological risk level corresponding to the tunnel excavation working face image is determined based on the comprehensive geological condition score.

2. The method for predicting geological risks of tunnel working faces based on multi-source information fusion according to claim 1 is characterized in that: Preprocessing the tunnel excavation working face image to obtain a tunnel excavation working face target image includes: De-noising the tunnel excavation working face image based on a non-local mean denoising method; Adaptive histogram equalization is performed on the denoised tunnel excavation working face image to obtain a tunnel excavation working face image with enhanced image contrast; The tunnel excavation working face image after image contrast enhancement is geometrically corrected to obtain the tunnel excavation working face target image.

3. The method for predicting geological risks of tunnel working faces based on multi-source information fusion according to claim 1, characterized in that: The groundwater flow state identification model includes a U2Net network and a classifier.

4. The method for predicting geological risks of tunnel working faces based on multi-source information fusion according to claim 1, characterized in that: The target image of the tunnel excavation working face is input into the trained fracture fragmentation degree recognition model to obtain the rock mass fragmentation degree classification result corresponding to the tunnel excavation working face image, including: Determining cracks in the target image of the tunnel excavation working face based on an edge detection algorithm; Dividing the tunnel excavation working face target image into a plurality of grids, and calculating the length, width and number of cracks in each grid; Performing cluster analysis on the plurality of grids by using a clustering algorithm based on the length, width and number of cracks in each grid; The rock mass fragmentation degree classification result corresponding to the tunnel excavation working face image is determined based on the cluster analysis result.

5. The method for predicting geological risks of tunnel working faces based on multi-source information fusion according to claim 1, characterized in that: A comprehensive score of the geological conditions of the tunnel excavation working face image is obtained based on the groundwater flow state classification result and the rock mass fragmentation degree classification result, including: Determining a water flow state score based on the groundwater flow state classification result; Determining a degree of fragmentation score based on the rock mass fragmentation degree classification result; A comprehensive score of the geological conditions of the tunnel excavation working face image is calculated based on the water flow state score and the fragmentation degree score.

6. The method for predicting geological risks of tunnel working faces based on multi-source information fusion according to claim 5 is characterized in that: The calculation formula for the comprehensive score of geological conditions is: S total =W water ×S water +W fracture ×S fracture； Among them, S total S is the comprehensive score of geological conditions. water S is the water flow rate score, S fracture The degree of fragmentation is scored, W water is the water flow state weight, W fracture is the weight of the degree of fragmentation.

7. The method for predicting geological risks of tunnel working faces based on multi-source information fusion according to claim 2, characterized in that: Preprocessing the tunnel excavation working face image to obtain a tunnel excavation working face target image also includes: Obtain tunnel excavation face images at different resolutions and scales; The multiple tunnel excavation working face images with different resolutions and different scales are fused.

8. The method for predicting geological risks of a tunnel working face based on multi-source information fusion according to any one of claims 1 to 7, characterized in that: The groundwater flow state classification result is dry, water-bearing or flowing water; and / or, The rock mass fragmentation degree classification results are broken, relatively broken or complete.

9. A tunnel working face geological risk prediction system based on multi-source information fusion, comprising a processor, a memory and a computer program stored in the memory, characterized in that: The processor is used to execute the computer program. When the computer program is executed, the system implements the steps of the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images