A safety early warning method and system for tunneling work

Abstract

The present invention discloses a safety warning method and system for tunneling work, which relates to the technical field of tunneling safety warning. The method includes: constructing a basic twin model of the stratum environment; identifying the position and morphological characteristics of obstacles; inputting the position and morphological characteristics of obstacles into the obstacle decoupling model library for decoupling search, and outputting a matching obstacle twin model; using the matching obstacle twin model as a plug-in, fusing it to the basic twin model of the stratum environment in the same coordinate system to obtain a fused twin model; performing obstacle area risk analysis according to the fused twin model, outputting a risk warning signal, and feeding the risk warning signal back to the tunneling equipment for risk control. It solves the technical problem of the need for separate modeling in underground tunneling work in the prior art, which leads to waste of resources and low efficiency, and achieves the technical effect of saving resources and improving the safety and efficiency of tunneling operations by sharing the underground environment twin basic model and only inserting the obstacle twin model.

Description

Technical Field

[0001] The present invention relates to the technical field of tunneling safety early warning, and in particular to a tunneling safety early warning method and system. Background Art

[0002] During tunneling construction, the frequent presence of underground structural obstacles (such as existing tunnels, underground pipelines, and the remains of civil air defense projects) presents a core challenge to construction safety. These obstacles are scattered and diverse in shape, making it difficult to fully and accurately grasp information such as their location, burial depth, and internal structure. Traditional tunneling safety early warning methods have significant limitations. Separate stratum detection and obstacle modeling are required for each tunneling area, which not only consumes significant manpower, material resources, and time, resulting in a significant waste of resources, but also hinders systematic expansion and rapid deployment. Rapidly modeling and deploying obstacles ahead of tunneling in complex and changing underground environments has become a key technical bottleneck for ensuring the safe and efficient progress of tunneling work. Summary of the Invention

[0003] The present application provides a safety warning method and system for tunneling work, which solves the technical problem in the prior art that separate modeling is required in underground tunneling work, resulting in waste of resources and low efficiency.

[0004] A first aspect of the present application provides a safety early warning method for excavation work, the method comprising:

[0005] Based on the cloud service terminal, the stratum detection modeling data of the first tunneling work area is uploaded to construct a basic twin model of the stratum environment; the position and morphological characteristics of obstacles in front of the tunneling equipment are identified using a multi-source sensing device, wherein the multi-source sensing device is embedded in the tunneling equipment, and the tunneling equipment is in the first tunneling work area; the obstacle position and morphological characteristics are input into the obstacle decoupling model library for decoupling search, and a matching obstacle twin model is output, and the obstacle decoupling model library is embedded in the cloud service terminal; the matching obstacle twin model is used as a plug-in and fused into the basic twin model of the stratum environment in the same coordinate system to obtain a fused twin model; the obstacle area risk analysis is performed according to the fused twin model, a risk warning signal is output, and the risk warning signal is fed back to the tunneling equipment for risk control.

[0006] Furthermore, the obstacle position and morphological features are input into the obstacle decoupling model library for decoupling search, and the method includes: obtaining underground obstacle samples by the cloud service terminal; identifying the parameterized model templates corresponding to the underground obstacle samples respectively, and constructing the obstacle decoupling model library; inputting the obstacle position and morphological features into the obstacle decoupling model library and performing decoupling search with each parameterized model template, and obtaining parameterized model templates whose position similarity and morphological similarity are both greater than a preset threshold.

[0007] Furthermore, an obstacle area risk analysis is performed according to the fusion twin model, and a risk warning signal is output. The method includes: performing an obstacle area risk analysis on the excavation parameters of the excavation equipment according to the fusion twin model, including collision risk, resistance mutation risk, excavation posture deviation risk and structural deformation risk; performing risk calculation according to the collision risk, resistance mutation risk, excavation posture deviation risk and structural deformation risk, and outputting an excavation risk index; determining the risk level according to the excavation risk index, and outputting a risk warning signal corresponding to the risk level.

[0008] Furthermore, the risk warning signal is fed back to the tunneling equipment for risk control, and the method includes: recording the tunneling parameters of the tunneling equipment, the tunneling parameters including the propulsion path, propulsion rate and cutterhead speed; when the tunneling equipment receives the risk warning signal, the tunneling parameters are gradient optimized and controlled to minimize the tunneling risk index to obtain optimized tunneling parameters, wherein the priority of the adaptive optimization control is that the cutterhead speed is greater than the propulsion rate and greater than the propulsion path; and the tunneling of the tunneling equipment is controlled according to the optimized tunneling parameters.

[0009] Furthermore, the cloud service terminal is distributedly connected to multiple tunneling equipment, and the method also includes: detecting multiple tunneling work areas corresponding to the multiple tunneling equipment; identifying N tunneling equipment located in the first tunneling work area, collecting N stratum detection modeling data corresponding to the N tunneling equipment, and uploading the N stratum detection modeling data to the cloud service terminal to synchronously update the stratum environment basic twin model.

[0010] Furthermore, the N stratum detection modeling data are uploaded to the cloud service terminal to synchronously update the obstacle decoupling model library. The method includes: identifying N groups of obstacle positions and morphological features corresponding to the N stratum detection modeling data; obtaining N return results corresponding to the N groups of obstacle positions and morphological features in the obstacle decoupling model library, and collecting M groups of obstacle positions and morphological features whose return results are empty, where M is a positive integer less than or equal to N; using the M groups of obstacle positions and morphological features as incremental data to update the obstacle decoupling model library.

[0011] Furthermore, the multi-source perception device includes at least any two of an inertial navigation unit, a visual camera, an infrared / structured light depth camera, and a lidar.

[0012] Furthermore, the parametric model templates corresponding to the underground obstacle samples are identified, and the method includes: collecting three-dimensional point cloud data samples of the underground obstacle samples; extracting position feature samples and morphological feature samples corresponding to the three-dimensional point cloud data samples, and the morphological features include geometric features and material features; constructing an obstacle parametric model template based on the position feature samples and morphological feature samples corresponding to the three-dimensional point cloud data samples.

[0013] Furthermore, the parameterized model templates corresponding to the underground obstacle samples are identified, and the obstacle decoupling model library is constructed. The method also includes: decomposing each underground obstacle sample according to its structural function to obtain multiple obstacle components; generating an underground component obstacle model template corresponding to the underground obstacle sample according to the multiple obstacle components; when a new underground obstacle sample is added, obtaining the required obstacle component corresponding to the new underground obstacle sample; combining the required obstacle component from the underground component obstacle model template to update the obstacle decoupling model library.

[0014] A second aspect of the present application provides a safety early warning system for excavation work, the system comprising:

[0015] A model construction module is used to upload the stratum detection modeling data of the first excavation work area based on the cloud service terminal to construct a basic twin model of the stratum environment; an identification module is used to use a multi-source sensing device to identify the position and morphological characteristics of obstacles in front of the excavation equipment, wherein the multi-source sensing device is embedded in the excavation equipment, and the excavation equipment is in the first excavation work area; a search module is used to input the obstacle position and morphological characteristics into the obstacle decoupling model library for decoupling search, and output a matching obstacle twin model, and the obstacle decoupling model library is embedded in the cloud service terminal; a fusion module is used to use the matching obstacle twin model as a plug-in to fuse it into the basic twin model of the stratum environment in the same coordinate system to obtain a fused twin model; an analysis module is used to perform obstacle area risk analysis according to the fused twin model, output a risk warning signal, and feed the risk warning signal back to the excavation equipment for risk control.

[0016] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0017] This application solves the technical problem in the existing technology that underground excavation work requires separate modeling, resulting in waste of resources and low efficiency. It achieves the technical effect of saving resources and improving the safety and efficiency of excavation operations by sharing the underground environment twin basic model and only inserting the obstacle twin model. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 A schematic flow chart of a safety warning method for excavation work provided in an embodiment of the present application;

[0020] Figure 2 A schematic structural diagram of a safety warning system for tunneling work provided in an embodiment of the present application.

[0021] Explanation of the accompanying drawings: model building module 11, recognition module 12, search module 13, fusion module 14, analysis module 15. DETAILED DESCRIPTION

[0022] This application solves the technical problem in the prior art that separate modeling is required in underground excavation work, resulting in waste of resources and low efficiency, by providing a safety warning method and system for excavation work.

[0023] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0024] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0025] Example 1, as Figure 1 As shown, the present application provides a safety early warning method for excavation work, wherein the method includes:

[0026] The stratum detection modeling data of the first tunneling work area is uploaded to the cloud service terminal to build a basic twin model of the stratum environment.

[0027] In an embodiment of the present application, through sensors and measuring equipment, the stratum detection data of the first excavation work area is collected, including various physical properties of the soil layer, such as distribution, hardness, and humidity, to generate stratum detection modeling data; after the cloud service terminal receives this data, it uploads it to the cloud for processing to generate a basic twin model of the stratum environment to reflect the basic underground environment.

[0028] Furthermore, the cloud service terminal is connected to a plurality of tunneling equipment in a distributed manner, and further includes:

[0029] Detect multiple excavation work areas corresponding to the multiple excavation equipment; identify N excavation equipment located in the first excavation work area, collect N stratum detection modeling data corresponding to the N excavation equipment, and upload the N stratum detection modeling data to the cloud service terminal to synchronously update the stratum environment basic twin model.

[0030] In an embodiment of the present application, the cloud service terminal is distributedly connected to multiple tunneling equipment, and the cloud service terminal exchanges data with multiple tunneling equipment through a wireless communication network to ensure that the working status information of each tunneling equipment and its surrounding environment data can be received in real time; each tunneling equipment collects the stratum detection data of the corresponding working area through the various sensors configured therein, including but not limited to physical parameters such as soil layer distribution, rock characteristics, water level changes, and underground structure.

[0031] When multiple tunneling equipment are operating simultaneously, the cloud service terminal detects and identifies the specific working area of ​​each tunneling equipment, ensuring that the data collected by all equipment belongs to the corresponding area. Based on the identified tunneling equipment location, the cloud service terminal assigns a corresponding working area label to each tunneling equipment and collects the corresponding stratum detection modeling data. The cloud service terminal uploads this stratum detection modeling data to the cloud for processing. After receiving the data, the cloud server performs unified data fusion and processing, analyzing key information such as the soil characteristics and obstacle locations of each tunneling equipment area. This information is then synchronized and updated to the basic twin model of the stratum environment, ensuring that the basic twin model of the stratum environment is always up to date and reflects the real-time underground working environment and its changes.

[0032] Furthermore, uploading the N stratum detection modeling data to the cloud service terminal to synchronously update the obstacle decoupling model library includes:

[0033] Identify N groups of obstacle positions and morphological features corresponding to the N formation detection modeling data; obtain N return results corresponding to the N groups of obstacle positions and morphological features in the obstacle decoupling model library, collect M groups of obstacle positions and morphological features for which the return results are empty, where M is a positive integer less than or equal to N; use the M groups of obstacle positions and morphological features as incremental data to update the obstacle decoupling model library.

[0034] The cloud service terminal receives N sets of ground exploration modeling data from multiple tunneling equipment. The data processing system analyzes this data and identifies the corresponding obstacle locations and morphological features for each set of ground exploration data. These features may include information such as the obstacle's spatial location, shape, and size. The cloud service terminal then searches the obstacle decoupling model library, finding and matching existing obstacle models based on the obstacle locations and morphological features corresponding to each set of ground exploration data. If a matching result is found in the model library, the system returns the relevant information. If no match is found, the cloud service terminal marks the data set as unmatched. M sets of obstacle locations and morphological features (where M is a positive integer less than or equal to N) with empty results are identified as new or unknown obstacle information. These M sets of obstacle locations and morphological features with no matching results are treated as incremental data and added to the obstacle decoupling model library. In this way, the obstacle decoupling model library can be dynamically updated, gradually accumulating obstacle models, and thus enhancing its ability to identify new obstacles. The updated obstacle decoupling model library will contain the latest obstacle information, enabling subsequent excavation operations to make decisions using more comprehensive and accurate obstacle data, further improving the safety and efficiency of excavation work.

[0035] A multi-source sensing device is used to identify the position and morphological characteristics of obstacles in front of the tunneling equipment, wherein the multi-source sensing device is embedded in the tunneling equipment, and the tunneling equipment is in the first tunneling work area.

[0036] Furthermore, the multi-source perception device includes at least any two of an inertial navigation unit, a visual camera, an infrared / structured light depth camera, and a lidar.

[0037] In an embodiment of the present application, multi-source sensing devices are embedded in the tunneling equipment. When the tunneling equipment is in the first tunneling work area, these embedded sensing devices use real-time data collection to identify the position and morphological characteristics of obstacles ahead. The multi-source sensing devices include at least two of an inertial navigation unit, a visual camera, an infrared / structured light depth camera, and a lidar. The inertial navigation unit monitors the dynamic state of the tunneling equipment, such as acceleration and rotation angle, and provides information on the equipment's motion trajectory and position changes, helping to determine the relative position of obstacles. The visual camera uses image acquisition and computer vision technology to identify features such as the shape, color, and texture of obstacles, providing detailed appearance information for environmental perception. The infrared / structured light depth camera uses infrared imaging or structured light technology to accurately obtain depth information of obstacles, facilitating three-dimensional spatial scanning in low-light or complex environments. The lidar generates high-precision point cloud data through laser scanning, measures the distance, shape, and position of obstacles, and provides a three-dimensional geometric description of the obstacles.

[0038] The obstacle position and morphological features are input into an obstacle decoupling model library for decoupling search, and a matching obstacle twin model is output. The obstacle decoupling model library is embedded in the cloud service terminal.

[0039] The cloud service terminal receives obstacle location and morphological feature data from multi-source sensing devices. This data captures the obstacle's spatial position, shape, and other features through sensors, providing detailed information about the obstacle within the excavation work area. The cloud service terminal inputs this data into the obstacle decoupling model library for decoupling search, using the library to analyze and match the input obstacle data. Each obstacle twin model in the obstacle decoupling model library represents the characteristics of a specific type of obstacle, including parameters such as its location, shape, size, and material. Through decoupling search, the system can quickly compare the input obstacle location and morphological features with the stored obstacle twin models and output a matching obstacle twin model.

[0040] Furthermore, the obstacle position and morphological features are input into the obstacle decoupling model library for decoupling search, including:

[0041] The cloud service terminal obtains underground obstacle samples; identifies the parameterized model templates corresponding to the underground obstacle samples, and constructs the obstacle decoupling model library; inputs the obstacle positions and morphological features into the obstacle decoupling model library and performs a decoupling search with each parameterized model template to obtain parameterized model templates whose position similarity and morphological similarity are both greater than a preset threshold.

[0042] The cloud service terminal acquires underground obstacle samples from multiple tunneling devices. These samples are obstacle data collected by sensing devices, including obstacle characteristics such as location, shape, size, and material. Through detailed analysis of the obstacle samples, the geometric features, surface characteristics, and other key parameters of the obstacles are extracted to construct parametric model templates. These templates provide a standardized model description for each obstacle type. These templates are then stored in a database, resulting in a library of obstacle decoupling models.

[0043] The cloud service terminal inputs the obstacle positions and morphological features collected in real time into the obstacle decoupling model library and performs decoupling search in combination with the parameterized model template.

[0044] The collected obstacle data is matched against each pre-built parametric model template in the obstacle decoupling model library. This involves comparing the input obstacle features with the obstacle models stored in the templates to determine the degree of match. Specifically, positional similarity is calculated by calculating the difference between the spatial position of the input obstacle and the position of the obstacle in each template. Positional similarity can be calculated using Euclidean distance or other geometric distance metrics, measuring the distance between the coordinates of the input obstacle and the coordinates of the template obstacle, where smaller distance values ​​indicate higher similarity. Morphological similarity is calculated by comparing the shape and size differences of the obstacles. Shape matching algorithms, such as contour analysis, geometric transformation, or point cloud data comparison, can be used to assess the morphological differences between the input obstacle and the template. Finally, the calculated positional and morphological similarities are compared with preset thresholds. If both similarities exceed the predetermined thresholds, the input obstacle is considered to be a close match with a template obstacle, and the corresponding parametric model template is returned.

[0045] Furthermore, identifying the parameterized model templates corresponding to the underground obstacle samples includes:

[0046] Collect three-dimensional point cloud data samples of the underground obstacle samples; extract position feature samples and morphological feature samples corresponding to the three-dimensional point cloud data samples, wherein the morphological features include geometric features and material features; and construct an obstacle parameterized model template based on the position feature samples and morphological feature samples corresponding to the three-dimensional point cloud data samples.

[0047] Using the LiDAR and structured light depth cameras on the tunneling equipment, three-dimensional point cloud data samples of underground obstacle samples are collected. These 3D point cloud data samples contain the spatial distribution and 3D morphology of the obstacles. Corresponding positional and morphological feature samples are extracted from the 3D point cloud data samples. Positional features include the center and boundary positions of the obstacle; morphological features include geometric features and material features. Geometric features include the shape, size, and surface contour of the obstacle, while material features include the material type (e.g., rock, soil, etc.) and its surface properties (e.g., hardness, roughness, etc.). Based on the positional and morphological feature samples extracted from the 3D point cloud data samples, a parametric model template for the obstacle is constructed. This template provides a standardized method for describing the obstacle by converting the geometric morphology and material properties of the obstacle into a parametric mathematical model.

[0048] Furthermore, identifying the parameterized model templates corresponding to the underground obstacle samples and constructing the obstacle decoupling model library further includes:

[0049] Each underground obstacle sample is decomposed according to its structural function to obtain multiple obstacle components; an underground component obstacle model template corresponding to the underground obstacle sample is generated according to the multiple obstacle components; when a new underground obstacle sample is added, the required obstacle component corresponding to the new underground obstacle sample is obtained; the required obstacle component is combined from the underground component obstacle model template to update the obstacle decoupling model library.

[0050] Preferably, each underground obstacle sample is decomposed according to its structural function to obtain multiple obstacle components. For example, an underground obstacle may contain multiple layers of different materials, such as rock layers, soil, pipes, or support structures. Based on the multiple extracted obstacle components, underground component obstacle model templates corresponding to the underground obstacle sample are generated. These underground component obstacle model templates describe the function and structural characteristics of each component in a standardized manner. When a new underground obstacle sample is added, the new sample is analyzed to identify the corresponding required obstacle components. If the new obstacle sample contains new components, the characteristics of these components are determined and corresponding model templates are generated. The required obstacle components are combined from the underground component obstacle model templates to update the obstacle decoupling model library. New obstacle decoupling models are generated by combining the new required obstacle components with components in the existing model library, and these updated models are stored in the obstacle decoupling model library.

[0051] The matching obstacle twin model is used as a plug-in and fused to the formation environment basic twin model in the same coordinate system to obtain a fused twin model.

[0052] In this embodiment, the matching obstacle twin model is converted into a standardized plug-in format and accurately integrated into the foundational twin model of the formation environment according to its position and size in three-dimensional space, thereby generating a fused twin model. By integrating the obstacle twin model as a plug-in with the foundational twin model of the formation environment in a unified coordinate system, a fused twin model containing all environmental and obstacle data is generated.

[0053] Obstacle area risk analysis is performed according to the fusion twin model, a risk warning signal is output, and the risk warning signal is fed back to the tunneling equipment for risk control.

[0054] Based on a fused twin model, the system comprehensively analyzes the tunneling area, identifying the location and shape of obstacles and calculating and assessing potential risks, such as collision risk, sudden change in resistance, tunneling posture deviation risk, and structural deformation risk. The system then calculates tunneling risk indicators based on the risk type, quantifies the severity of each risk, determines the risk level based on preset thresholds, and outputs corresponding risk warning signals. Finally, the risk warning signals are fed back to the tunneling equipment, and the control system adjusts tunneling parameters such as the propulsion path, propulsion rate, and cutterhead speed based on the warning signals, thereby optimizing tunneling operations, avoiding safety incidents, and ensuring the safety and efficiency of the operation.

[0055] Furthermore, the obstacle area risk analysis is performed according to the fusion twin model, and a risk warning signal is output, including:

[0056] According to the fusion twin model, an obstacle area risk analysis is performed on the excavation parameters of the excavation equipment, including collision risk, resistance mutation risk, excavation posture deviation risk and structural deformation risk; risk calculation is performed according to the collision risk, resistance mutation risk, excavation posture deviation risk and structural deformation risk, and an excavation risk index is output; the risk level is determined according to the excavation risk index, and a risk warning signal corresponding to the risk level is output.

[0057] Based on the fused twin model, the system analyzes the tunneling equipment's excavation parameters for obstacle-area risks, including collision risk, resistance mutation risk, excavation posture deviation risk, and structural deformation risk. The system utilizes the ground environment data and obstacle information integrated into the fused twin model to comprehensively assess the environment within the tunneling equipment's current working area.

[0058] Based on the ground environment and obstacle data integrated into the fused twin model, the system calculates various risks. For collision risk, the system uses a collision prediction algorithm to calculate the potential collision risk by evaluating the current distance between the tunneling equipment and the obstacle ahead, the obstacle's morphological characteristics, and its relative position to the tunneling equipment. As the equipment approaches the obstacle, the collision risk indicator dynamically adjusts based on changes in distance and morphological characteristics. For resistance mutation risk, the system calculates the risk of soil changes encountered by the tunneling equipment based on the physical properties of the stratum (such as density, hardness, and moisture). When the tunneling equipment encounters a sudden change in soil hardness or density, the system analyzes the mechanical properties of the stratum to calculate the risk of resistance mutation and generate a corresponding resistance risk indicator. For tunneling posture deviation risk, the system monitors the movement of the tunneling equipment, including parameters such as the advancement path, advancement rate, and equipment inclination. When the equipment experiences abnormal posture deviation (such as tilt or excessive vibration), the system calculates the deviation risk based on sensor data and outputs a corresponding risk value based on the deviation angle and rate. Regarding structural deformation risk, the system analyzes the potential deformation of underground structures (such as tunnel walls and support frames) during excavation. Using soil stress and strain data combined with pressure data from excavation equipment, the system calculates the risk of underground structural deformation and outputs a structural deformation risk index. By weighting the values ​​for collision risk, resistance mutation risk, excavation posture deviation risk, and structural deformation risk, the system ultimately calculates the excavation risk index, which comprehensively reflects the safety status of the current excavation operation. Higher values ​​indicate greater risk, while lower values ​​indicate safer operations.

[0059] The system determines the risk level based on the obtained excavation risk index; the risk level is divided according to the numerical range of the risk index, which may include low risk, medium risk, and high risk levels. The system matches the excavation risk index with the risk level based on the set threshold to determine the severity of the risk. The system outputs a risk warning signal corresponding to the risk level. This signal is used to indicate the risk level currently faced by the excavation equipment, and the signal content includes the specific risk type, level, and corresponding warning measures. After receiving the risk warning signal, the excavation equipment can take corresponding control measures based on the risk level, such as adjusting the propulsion path, reducing the propulsion rate, or optimizing the cutterhead speed, thereby adopting appropriate risk control strategies at different risk levels to ensure the safety and efficiency of the excavation operation.

[0060] Furthermore, feeding back the risk warning signal to the tunneling equipment for risk control includes:

[0061] Recording the excavation parameters of the excavation equipment, wherein the excavation parameters include a propulsion path, a propulsion rate, and a cutterhead speed; when the excavation equipment receives the risk warning signal, performing gradient optimization control on the excavation parameters to minimize the excavation risk index to obtain optimized excavation parameters, wherein the priority of the adaptive optimization control is that the cutterhead speed is greater than the propulsion rate and greater than the propulsion path; and controlling the excavation of the excavation equipment according to the optimized excavation parameters.

[0062] During the tunneling process, the tunneling equipment's parameters are recorded. These parameters include the tunneling path, tunneling rate, and cutterhead speed. The tunneling path refers to the route the tunneling equipment takes within the underground working area, the tunneling rate is the distance the tunneling equipment advances per unit time, and the cutterhead speed is the rotational speed of the tunneling equipment's cutterhead.

[0063] When the tunneling equipment receives a risk warning signal from the system, it initiates a gradient optimization control algorithm aimed at minimizing tunneling risk indicators. This optimization algorithm adjusts tunneling parameters based on the risk level reflected in the risk warning signal to mitigate potential risks. For example, when the risk of collision is high, the system may prioritize reducing the cutterhead speed to avoid collisions caused by high-speed rotation. Similarly, when the risk of sudden resistance changes is high, the system adjusts the propulsion rate to reduce the likelihood of the tunneling equipment encountering excessive resistance. The system gradually adjusts the propulsion path, propulsion rate, and cutterhead speed based on the severity of each risk. During the optimization process, the system sets a priority for adaptive optimization control, with cutterhead speed taking the highest priority, followed by propulsion rate, and finally propulsion path. The system controls the tunneling equipment's tunneling based on the optimized tunneling parameters. After gradient optimization, the tunneling equipment automatically adjusts its propulsion path, rate, and cutterhead speed to achieve optimal operating conditions, minimize risks, and ensure smooth tunneling operations while ensuring both safety and efficiency.

[0064] In summary, the embodiments of the present application have at least the following technical effects:

[0065] First, the cloud service terminal uploads stratum exploration modeling data from the first tunneling work area to construct a basic twin model of the stratum environment. Next, a multi-source sensing device is used to identify the location and morphological characteristics of obstacles in front of the tunneling equipment. The multi-source sensing device is embedded in the tunneling equipment, and the tunneling equipment is located in the first tunneling work area. The obstacle location and morphological characteristics are then input into the obstacle decoupling model library for decoupling search, outputting a matching obstacle twin model. The obstacle decoupling model library is then embedded in the cloud service terminal. Furthermore, the matching obstacle twin model is integrated into the basic twin model of the stratum environment as a plug-in within the same coordinate system to obtain a fused twin model. Finally, the fused twin model is used to perform a risk analysis of the obstacle area, outputting a risk warning signal, and feeding it back to the tunneling equipment for risk control. This solves the technical problem of separate modeling in underground tunneling work, which wastes resources and reduces efficiency. By sharing the basic twin model of the underground environment and inserting only the obstacle twin model, this saves resources and improves the safety and efficiency of tunneling operations.

[0066] Example 2, based on the same inventive concept as the safety warning method for tunneling work in the above embodiment, Figure 2 As shown, the present application provides a safety early warning system for tunneling work, wherein the system includes:

[0067] The model construction module 11 is used to upload the stratum detection modeling data of the first excavation work area based on the cloud service terminal to construct a basic twin model of the stratum environment; the identification module 12 is used to use a multi-source sensing device to identify the position and morphological characteristics of obstacles in front of the excavation equipment, wherein the multi-source sensing device is embedded in the excavation equipment, and the excavation equipment is in the first excavation work area; the search module 13 is used to input the obstacle position and morphological characteristics into the obstacle decoupling model library for decoupling search, and output a matching obstacle twin model, and the obstacle decoupling model library is embedded in the cloud service terminal; the fusion module 14 is used to use the matching obstacle twin model as a plug-in to fuse it into the basic twin model of the stratum environment in the same coordinate system to obtain a fused twin model; the analysis module 15 is used to perform obstacle area risk analysis according to the fused twin model, output a risk warning signal, and feed the risk warning signal back to the excavation equipment for risk control.

[0068] Furthermore, the search module 13 is used to perform the following method:

[0069] The cloud service terminal obtains underground obstacle samples; identifies the parameterized model templates corresponding to the underground obstacle samples, and constructs the obstacle decoupling model library; inputs the obstacle positions and morphological features into the obstacle decoupling model library and performs a decoupling search with each parameterized model template to obtain parameterized model templates whose position similarity and morphological similarity are both greater than a preset threshold.

[0070] Furthermore, the analysis module 15 is configured to perform the following method:

[0071] According to the fusion twin model, an obstacle area risk analysis is performed on the excavation parameters of the excavation equipment, including collision risk, resistance mutation risk, excavation posture deviation risk and structural deformation risk; risk calculation is performed according to the collision risk, resistance mutation risk, excavation posture deviation risk and structural deformation risk, and an excavation risk index is output; the risk level is determined according to the excavation risk index, and a risk warning signal corresponding to the risk level is output.

[0072] Furthermore, the analysis module 15 is configured to perform the following method:

[0073] Recording the excavation parameters of the excavation equipment, wherein the excavation parameters include a propulsion path, a propulsion rate, and a cutterhead speed; when the excavation equipment receives the risk warning signal, performing gradient optimization control on the excavation parameters to minimize the excavation risk index to obtain optimized excavation parameters, wherein the priority of the adaptive optimization control is that the cutterhead speed is greater than the propulsion rate and greater than the propulsion path; and controlling the excavation of the excavation equipment according to the optimized excavation parameters.

[0074] Furthermore, the model building module 11 is used to perform the following method:

[0075] Detect multiple excavation work areas corresponding to the multiple excavation equipment; identify N excavation equipment located in the first excavation work area, collect N stratum detection modeling data corresponding to the N excavation equipment, and upload the N stratum detection modeling data to the cloud service terminal to synchronously update the stratum environment basic twin model.

[0076] Furthermore, the model building module 11 is used to perform the following method:

[0077] Identify N groups of obstacle positions and morphological features corresponding to the N formation detection modeling data; obtain N return results corresponding to the N groups of obstacle positions and morphological features in the obstacle decoupling model library, collect M groups of obstacle positions and morphological features for which the return results are empty, where M is a positive integer less than or equal to N; use the M groups of obstacle positions and morphological features as incremental data to update the obstacle decoupling model library.

[0078] Furthermore, the identification module 12 is configured to execute the following method:

[0079] The multi-source perception device includes at least any two of an inertial navigation unit, a visual camera, an infrared / structured light depth camera, and a lidar.

[0080] Furthermore, the search module 13 is used to perform the following method:

[0081] Collect three-dimensional point cloud data samples of the underground obstacle samples; extract position feature samples and morphological feature samples corresponding to the three-dimensional point cloud data samples, wherein the morphological features include geometric features and material features; and construct an obstacle parameterized model template based on the position feature samples and morphological feature samples corresponding to the three-dimensional point cloud data samples.

[0082] Furthermore, the search module 13 is used to perform the following method:

[0083] Each underground obstacle sample is decomposed according to its structural function to obtain multiple obstacle components; an underground component obstacle model template corresponding to the underground obstacle sample is generated according to the multiple obstacle components; when a new underground obstacle sample is added, the required obstacle component corresponding to the new underground obstacle sample is obtained; the required obstacle component is combined from the underground component obstacle model template to update the obstacle decoupling model library.

[0084] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0085] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0086] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A safety early warning method for excavation work, characterized in that: The method comprises: Upload the stratum exploration modeling data of the first tunneling work area through the cloud service terminal to build a basic twin model of the stratum environment; Identifying the location and morphological characteristics of obstacles in front of the tunneling equipment using a multi-source sensing device, wherein the multi-source sensing device is embedded in the tunneling equipment, and the tunneling equipment is in the first tunneling work area; Inputting the obstacle position and morphological features into an obstacle decoupling model library for decoupling search, and outputting a matching obstacle twin model, wherein the obstacle decoupling model library is embedded in the cloud service terminal; The matching obstacle twin model is used as a plug-in to be fused with the formation environment basic twin model in the same coordinate system to obtain a fused twin model; Performing risk analysis of the obstacle area according to the fused twin model, outputting a risk warning signal, and feeding the risk warning signal back to the tunneling equipment for risk control; Inputting the obstacle position and morphological features into the obstacle decoupling model library for decoupling search, the method includes: Acquiring underground obstacle samples by the cloud service terminal; Identifying the parameterized model templates corresponding to the underground obstacle samples, and constructing the obstacle decoupling model library; Inputting the obstacle position and morphological features into the obstacle decoupling model library and performing decoupling search with each parameterized model template to obtain a parameterized model template whose position similarity and morphological similarity are both greater than a preset threshold; Identifying parameterized model templates corresponding to the underground obstacle samples and constructing the obstacle decoupling model library, the method further includes: Each underground obstacle sample is decomposed according to its structural function to obtain multiple obstacle components; generating, according to the plurality of obstacle components, an underground component obstacle model template corresponding to the underground obstacle sample, and when a new underground obstacle sample is added, obtaining a required obstacle component corresponding to the new underground obstacle sample; The required obstacle component is combined from the underground component obstacle model template to update the obstacle decoupling model library.

2. A safety early warning method for excavation work according to claim 1, characterized in that: Performing obstacle area risk analysis according to the fusion twin model and outputting a risk warning signal, the method includes: Performing obstacle area risk analysis on the tunneling parameters of the tunneling equipment based on the fused twin model, including collision risk, resistance mutation risk, tunneling posture deviation risk, and structural deformation risk; Perform risk calculations based on the collision risk, resistance mutation risk, tunneling posture deviation risk, and structural deformation risk, and output tunneling risk indicators; The risk level is determined according to the tunneling risk index, and a risk warning signal corresponding to the risk level is output.

3. A safety early warning method for excavation work according to claim 2, characterized in that: Feeding back the risk warning signal to the tunneling equipment for risk control, the method comprising: Recording excavation parameters of the excavation equipment, including excavation path, excavation rate, and cutterhead speed; When the tunneling equipment receives the risk warning signal, it performs gradient optimization control on the tunneling parameters to minimize the tunneling risk index to obtain optimized tunneling parameters, wherein the priority of the adaptive optimization control is that the cutterhead speed is greater than the propulsion rate and the propulsion path; The excavation of the excavation equipment is controlled according to the optimized excavation parameters.

4. A safety early warning method for excavation work according to claim 1, characterized in that: The cloud service terminal is connected to a plurality of tunneling equipment in a distributed manner, and the method further comprises: detecting a plurality of excavation work areas corresponding to the plurality of excavation equipment; Identify N tunneling equipment in the first tunneling work area, collect N stratum detection modeling data corresponding to the N tunneling equipment, upload the N stratum detection modeling data to the cloud service terminal, and synchronously update the stratum environment basic twin model.

5. A safety early warning method for excavation work according to claim 4, characterized in that: Uploading the N stratum detection modeling data to the cloud service terminal and synchronously updating the stratum environment basic twin model, and then synchronously updating the obstacle decoupling model library, the method includes: Identifying N groups of obstacle positions and morphological features corresponding to the N stratum detection modeling data; Obtain N return results corresponding to the N groups of obstacle positions and morphological features in the obstacle decoupling model library, and collect M groups of obstacle positions and morphological features for which the return results are empty, where M is a positive integer less than or equal to N; The M groups of obstacle positions and morphological features are used as incremental data to update the obstacle decoupling model library.

6. A safety early warning method for excavation work according to claim 5, characterized in that: The multi-source perception device includes at least any two of an inertial navigation unit, a visual camera, a depth camera, and a lidar, wherein the depth camera includes an infrared depth camera or a structured light depth camera.

7. A safety early warning method for excavation work according to claim 1, characterized in that: Identifying parameterized model templates corresponding to the underground obstacle samples, the method includes: Collecting three-dimensional point cloud data samples of the underground obstacle samples; Extracting position feature samples and morphological feature samples corresponding to the three-dimensional point cloud data samples, wherein the morphological features include geometric features and material features; A parameterized obstacle model template is constructed based on the position feature samples and the morphological feature samples corresponding to the three-dimensional point cloud data samples.

8. A safety warning system for tunneling work, characterized in that: A system for implementing a safety early warning method for excavation work according to any one of claims 1 to 7, comprising: A model building module is used to upload the stratum detection modeling data of the first tunneling work area based on the cloud service terminal to build a basic twin model of the stratum environment; an identification module configured to identify the location and morphological characteristics of obstacles in front of the tunneling equipment using a multi-source sensing device, wherein the multi-source sensing device is embedded in the tunneling equipment and the tunneling equipment is located in the first tunneling work area; A search module, configured to input the obstacle position and morphological features into an obstacle decoupling model library for decoupling search, and output a matching obstacle twin model, wherein the obstacle decoupling model library is embedded in the cloud service terminal; A fusion module is used to use the matching obstacle twin model as a plug-in to fuse it into the formation environment basic twin model in the same coordinate system to obtain a fused twin model; An analysis module is used to perform obstacle area risk analysis according to the fusion twin model, output a risk warning signal, and feed the risk warning signal back to the tunneling equipment for risk control.

Images