Three-dimensional posture detection method and system for posture deviation anomaly of heading machine

Abstract

The application discloses a three-dimensional posture detection method and system for posture deviation of a heading machine, and relates to the technical field of heading machine posture detection. The method comprises the following steps: recognizing core space constraint coordinates of a heading task; performing periodic node segmentation and coordinate alignment on the heading task to establish standard space coordinates; connecting a heading machine monitoring device during the heading process to obtain real-time heading monitoring data and construct a monitoring data network; using the monitoring data network to predict the projection of the heading posture into the standard space coordinates, perform alignment matching, and identify posture constraint abnormalities; and positioning the posture constraint abnormalities in the standard space coordinates in three dimensions to obtain a three-dimensional detection result of the abnormal posture. The application solves the technical problems of heading machine posture monitoring lag, insufficient control accuracy, and low abnormality recognition efficiency in the prior art, achieves real-time monitoring of heading machine posture deviation, and improves the accuracy of heading posture control and the abnormality recognition efficiency.

Description

Technical Field

[0001] This invention relates to the field of tunneling machine attitude detection technology, specifically to a three-dimensional attitude detection method and system for detecting abnormal attitude deviations in tunneling machines. Background Technology

[0002] In tunneling engineering, the attitude control accuracy of the tunnel boring machine (TBM) directly affects construction safety and project quality. Traditional attitude monitoring methods mostly rely on data from a single sensor, which suffers from limited monitoring dimensions and data error accumulation, making it difficult to capture attitude deviations in complex geological conditions in real time. At the same time, existing technologies mostly identify attitude anomalies after the fact, lacking the ability to predict in advance, which can easily lead to risks such as over-excavation, under-excavation, or equipment collisions. Furthermore, the insufficient accuracy of anomaly location makes subsequent adjustments difficult, seriously restricting the improvement of tunneling efficiency and construction safety.

[0003] Existing technologies suffer from technical problems such as lagging attitude monitoring of tunneling machines, resulting in insufficient control accuracy and low efficiency in anomaly identification. Summary of the Invention

[0004] This application provides a three-dimensional attitude detection method and system for tunneling machine attitude deviation anomalies, which is used to address the technical problems of insufficient control accuracy and low anomaly identification efficiency caused by the lag in tunneling machine attitude monitoring in the prior art.

[0005] In view of the above problems, this application provides a three-dimensional attitude detection method and system for tunneling machine attitude deviation anomalies.

[0006] The first aspect of this application provides a three-dimensional attitude detection method for tunnel boring machine attitude deviation anomalies, the method comprising:

[0007] The core spatial constraint coordinates of the tunneling task are identified based on the tunneling area range. Using these core spatial constraint coordinates as a reference, the tunneling task is segmented into periodic nodes and aligned with coordinates to establish a standard spatial coordinate system, which includes the full-cycle tunneling spatial constraints. During tunneling, monitoring equipment is connected to the tunneling machine to acquire real-time tunneling monitoring data. Data interaction compensation is performed according to the deployment location of the monitoring equipment and the characteristics of the collected data to construct a monitoring data network. The tunneling posture is predicted and projected onto the standard spatial coordinate system using the monitoring data network. Alignment matching is performed based on the tunneling monitoring cycle and the spatial constraint cycle to identify posture constraint anomalies. Three-dimensional offset positioning is performed within the standard spatial coordinate system based on the posture constraint anomalies to obtain the three-dimensional detection results of the abnormal posture, including posture angle, position coordinates, offset direction, and offset amplitude.

[0008] In one possible approach, based on the risk value of the constraint impact of the posture constraint anomaly, the identification step size and detection granularity are configured; the offset positioning engine is driven to perform three-dimensional offset search positioning in the standard spatial coordinates according to the identification step size and detection granularity, and the abnormal posture detection parameters matching the risk level are output to obtain the three-dimensional detection result of the abnormal posture.

[0009] In one possible manner, the type of spatial constraint corresponding to the attitude constraint anomaly is obtained, and a type risk value is determined based on the constraint type; according to the spatial anomaly offset distance aligned with the tunneling monitoring work cycle and the spatial constraint cycle, combined with the constraint tolerance threshold of the anomaly constraint type, the anomaly constraint risk level is calculated, and the anomaly constraint risk level is used to characterize the degree of constraint violation of the current attitude anomaly; the constraint impact risk value is obtained by quantifying the type risk value and the anomaly constraint risk level.

[0010] In one possible approach, a dynamic parameter fuzzy list is set, which includes the constraint risk type, the impact risk value, and the corresponding recognition step size and detection granularity. The recognition step size is the smallest spatial interval unit in which the detection engine moves or scans in standard spatial coordinates during each movement in the process of performing 3D offset localization or attitude anomaly search. The detection granularity is the detection accuracy of attitude angle changes or the smallest identifiable unit of the spatial search unit during attitude offset detection. Using the constraint impact risk value and constraint type as input, the recognition step size and detection granularity are obtained by matching in the dynamic parameter fuzzy list.

[0011] In one possible manner, the position coordinates and corresponding spatial coordinates of each regional node within the tunneling area are obtained; a multi-level spatial coordinate system is constructed, including a geographic coordinate system, a tunneling design coordinate system, and a tunneling machine equipment body coordinate system; the position coordinates and corresponding spatial coordinates of each regional node are projected into the multi-level spatial coordinate system to obtain coordinate differences; coordinate transformation and alignment are performed based on the coordinate differences, and spatial coordinate transformation and projection are performed according to the tunneling constraint parameters of each regional node to obtain the core spatial constraint coordinates.

[0012] In one possible approach, the position coordinates and spatial coordinate differences of each regional node in each coordinate system are calculated. Based on the coordinate difference values, a coordinate transformation and alignment operation is performed to establish a transformation relationship matrix between each coordinate system. According to the tunneling constraint parameters corresponding to each regional node, including the allowable deviation range of attitude, the tolerance of tunneling path, and the boundary information of risk areas, the aligned coordinates are spatially transformed and projected to obtain core spatial constraint coordinates of a unified standard, which are used as a spatial reference benchmark for attitude detection and anomaly identification.

[0013] In one possible approach, the tunneling task is decomposed into periodic nodes. Based on the propulsion logic and attitude changes between nodes, the attitude relationship type is identified, and the attitude relationship type between nodes is determined, including alignment and succession relationships or distribution order relationships. According to the attitude relationship type and tunneling constraint parameters, each node is segmented by attitude constraints and spatial region is labeled to form a full-cycle tunneling coordinate chain. Based on the full-cycle coordinate chain, multi-level coordinate transformation and attitude relationship alignment are performed with the core spatial constraint coordinates as the reference coordinates to generate standard spatial coordinates that include full-cycle tunneling attitude constraints.

[0014] In one possible approach, a monitoring radar chart for each monitoring device is constructed based on its deployment location and monitoring parameters. Abnormal offset data of each monitoring device is analyzed based on the relationship between the monitoring device and the tunneling environment, combined with anomaly sample analysis, and then added to its monitoring radar chart. The monitoring radar charts of all monitoring devices are overlaid in multiple dimensions to obtain associated interactive monitoring data groups and interactive data compensation and correction relationships. The associated interactive monitoring data groups are corrected according to the interactive data compensation and correction relationships, and a spatial network is constructed based on the corrected monitoring radar charts of each monitoring device to build the monitoring data network.

[0015] In one possible approach, based on the time-series attitude data collected by each monitoring device in the monitoring data network, the attitude change trajectory of the tunneling machine within the current working cycle is constructed; using the historical attitude angle change trend within the time window, the attitude state of the upcoming cycle node is estimated through curve fitting; the estimated attitude state is projected onto the corresponding target node position in the standard spatial coordinates and compared with the attitude constraint parameters marked on the node to calculate the predicted offset, including angle deviation, position drift and its direction component; if the offset exceeds the set attitude tolerance threshold, it is determined that there is an abnormal attitude constraint trend, and the attitude constraint abnormality is output.

[0016] A second aspect of this application provides a three-dimensional attitude detection system for tunnel boring machine attitude deviation anomalies, the system comprising:

[0017] The system includes the following modules: a constraint coordinate identification module for identifying the core spatial constraint coordinates of the tunneling task based on the tunneling area; a standard spatial coordinate establishment module for segmenting and aligning the tunneling task into periodic nodes based on the core spatial constraint coordinates, and establishing standard spatial coordinates that include the full-cycle tunneling spatial constraints; a monitoring data network construction module for connecting the tunneling machine monitoring equipment during tunneling to acquire real-time tunneling monitoring data, and performing data interaction compensation according to the deployment location of the monitoring equipment and the characteristics of the collected data to construct a monitoring data network; an attitude constraint anomaly identification module for predicting the tunneling attitude and projecting it into the standard spatial coordinates using the monitoring data network, and identifying attitude constraint anomalies based on the alignment and matching between the tunneling monitoring work cycle and the spatial constraint cycle; and a three-dimensional detection result acquisition module for performing three-dimensional offset positioning within the standard spatial coordinates based on the attitude constraint anomalies to obtain the three-dimensional detection results of the abnormal attitude, including attitude angle, position coordinates, offset direction, and offset amplitude.

[0018] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0019] The core spatial constraint coordinates of the tunneling task are identified based on the tunneling area. Using these core spatial constraint coordinates as a reference, the tunneling task is periodically segmented and aligned to establish a standard spatial coordinate system. During tunneling, monitoring equipment is connected to the tunneling machine to acquire real-time monitoring data. Data interaction compensation is performed according to the deployment location of the monitoring equipment and the characteristics of the collected data to construct a monitoring data network. The tunneling posture is predicted and projected onto the standard spatial coordinate system using this monitoring data network. Alignment matching is performed based on the tunneling monitoring cycle and the spatial constraint cycle to identify posture constraint anomalies. Three-dimensional offset positioning is performed within the standard spatial coordinate system based on the posture constraint anomalies to obtain the three-dimensional detection results of the abnormal posture. This achieves the technical effect of real-time monitoring of tunneling machine posture offset anomalies, improving the accuracy of tunneling posture control and the efficiency of anomaly identification. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating the three-dimensional attitude detection method for tunneling machine attitude deviation anomalies provided in an embodiment of this application.

[0022] Figure 2This is a schematic diagram of the structure of a three-dimensional attitude detection system for tunneling machine attitude deviation anomalies provided in an embodiment of this application.

[0023] Figure labeling: Constraint coordinate recognition module 10, standard space coordinate establishment module 20, monitoring data network construction module 30, attitude constraint anomaly recognition module 40, 3D detection result acquisition module 50. Detailed Implementation

[0024] This application provides a three-dimensional attitude detection method and system for tunneling machine attitude deviation anomalies, which is used to address the technical problems of insufficient control accuracy and low anomaly identification efficiency caused by the lag in tunneling machine attitude monitoring in the prior art.

[0025] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0026] Example 1, as Figure 1 As shown, this application provides a three-dimensional attitude detection method for tunnel boring machine attitude deviation anomalies, the method comprising:

[0027] Step S100: Identify the core spatial constraint coordinates of the tunneling task based on the tunneling area range.

[0028] Specifically, the location coordinates and corresponding spatial coordinates of each regional node within the tunneling area are obtained, and a multi-level spatial coordinate system is constructed, including a geographic coordinate system, a tunneling design coordinate system, and a tunneling machine equipment body coordinate system. These coordinates are projected into this system to obtain coordinate differences. Based on the differences, the transformation relationship matrix required for coordinate transformation and alignment is calculated. Then, combined with the tunneling constraint parameters of each regional node (such as the allowable deviation range of attitude, tunneling path tolerance, and risk area boundary information), the aligned coordinates are spatially transformed and projected to finally obtain a unified standard core spatial constraint coordinate, which serves as a spatial reference benchmark for subsequent attitude detection and anomaly identification.

[0029] Step S200: Based on the core spatial constraint coordinates as the reference coordinates, the tunneling task is divided into periodic nodes and the coordinates are aligned to establish standard spatial coordinates, which include the full-cycle tunneling spatial constraints.

[0030] Specifically, using the core spatial constraint coordinates as the reference coordinates, the tunneling task is decomposed into periodic nodes. Based on the propulsion logic and attitude changes between nodes, the attitude relationship type (including alignment and succession relationship or distribution order relationship) is identified. According to the identified attitude relationship type and tunneling constraint parameters, each periodic node is segmented by attitude constraint and spatial region is labeled to form a full-cycle tunneling coordinate chain. Based on this full-cycle coordinate chain, multi-level coordinate transformation and attitude relationship alignment are performed to finally generate a standard spatial coordinate containing full-cycle tunneling spatial constraints, providing a unified spatial reference framework for subsequent monitoring of tunneling attitude and anomaly identification.

[0031] Step S300: Connect the tunneling machine monitoring equipment during the tunneling process, obtain real-time tunneling monitoring data, and perform data interaction compensation according to the deployment location of the monitoring equipment and the characteristics of the collected data to build a monitoring data network.

[0032] Specifically, during the tunneling process, monitoring equipment is connected to the tunneling machine to acquire real-time tunneling monitoring data; based on the deployment location and monitoring parameters of each monitoring device, a corresponding monitoring radar map is constructed; considering the influence relationship between the monitoring devices and the tunneling environment, abnormal offset data of each device is analyzed through abnormal samples and added to the corresponding monitoring radar map; the monitoring radar maps of all monitoring devices are overlaid in multiple dimensions to obtain related interactive monitoring data groups and interactive data compensation and correction relationships; after correcting the related interactive monitoring data groups according to the compensation and correction relationships, spatial networking is performed according to the corrected monitoring radar maps of each device, ultimately forming a monitoring data network.

[0033] Step S400: The tunneling posture is predicted by the monitoring data network and projected into the standard spatial coordinates. Alignment and matching are performed based on the tunneling monitoring cycle and the spatial constraint cycle to identify posture constraint anomalies.

[0034] Specifically, based on the time-series attitude data collected by each monitoring device in the monitoring data network, the attitude change trajectory of the tunneling machine within the current working cycle is constructed; using the historical attitude angle change trend within the time window, the attitude state of the upcoming cycle node is predicted through curve fitting; the predicted attitude state is projected onto the corresponding target node position in the standard spatial coordinate system and compared with the attitude constraint parameters marked on the node to calculate the predicted offset (including angle deviation, position drift and its direction components); if the offset exceeds the set attitude tolerance threshold, it is determined that there is an abnormal attitude constraint trend, and the attitude constraint anomaly is output.

[0035] Step S500: Based on the abnormal attitude constraint, perform three-dimensional offset positioning within the standard spatial coordinates to obtain the three-dimensional detection results of the abnormal attitude, including attitude angle, position coordinates, offset direction, and offset amplitude.

[0036] Specifically, for identified attitude constraint anomalies, the type of the corresponding spatial constraint is first determined to obtain the type risk value. The anomaly constraint risk level is calculated by combining the spatial anomaly offset distance and the constraint tolerance threshold of the anomaly constraint type. The type risk value and risk level are then fused and quantified to obtain the constraint impact risk value. Based on this risk value, the identification step size (the minimum spatial interval of each movement of the detection engine) and detection granularity (the detection accuracy of attitude angle changes) are matched in the dynamic parameter fuzzy list. The offset positioning engine is then driven to perform three-dimensional offset search and positioning within the standard spatial coordinates, outputting anomaly attitude detection parameters that match the risk level, and obtaining a three-dimensional detection result including attitude angle, position coordinates, offset direction, and offset amplitude. Simultaneously, this three-dimensional detection result is fed back to the tunneling equipment control module, which automatically triggers attitude adjustment commands (including track fine-tuning control, cutterhead pitch angle correction, or advance rate adjustment) based on the anomaly type and risk level. This achieves attitude adaptive compensation control to reduce tunneling offset risk, or triggers an early warning when the offset degree is high.

[0037] In one possible implementation, step S500 further includes:

[0038] Step S510: Based on the constraint impact risk value of the posture constraint anomaly, configure the identification step size and detection granularity.

[0039] Step S520: Drive the offset positioning engine to perform three-dimensional offset search positioning in the standard spatial coordinates according to the recognition step size and detection granularity, output abnormal posture detection parameters that match the risk level, and obtain the three-dimensional detection result of the abnormal posture.

[0040] Specifically, based on the constraint impact risk value of anomalies in attitude constraints (this risk value comprehensively considers the parameter risk of the constraint; for example, in a coal mining scenario, if the tunneling deviation may trigger an area with risks such as leakage or explosion, the corresponding constraint impact risk value is high due to the high danger of such situations), a preset dynamic parameter fuzzy list is invoked (this list contains the constraint risk type, impact risk value, and corresponding recognition step size and detection granularity, where the recognition step size is the smallest spatial interval unit in which the detection engine moves or scans in standard spatial coordinates each time, and the detection granularity is the detection accuracy of attitude angle changes or the smallest identifiable unit of the spatial search unit); by inputting the constraint impact risk value and the corresponding constraint type into the dynamic parameter fuzzy list for matching, for high-risk situations (such as those involving dangerous areas such as explosion-proof areas), a smaller recognition step size and a finer detection granularity are configured to ensure detection reliability, while for low-risk situations, a larger recognition step size and a coarser detection granularity are configured to improve efficiency, thereby completing the configuration of the recognition step size and detection granularity.

[0041] Using the configured recognition step size (the smallest spatial interval unit in which the detection engine moves or scans in standard spatial coordinates) and detection granularity (the smallest recognizable unit for detecting changes in attitude angle or spatial search units) as parameters, the offset positioning engine is driven to initiate three-dimensional offset search and positioning within standard spatial coordinates. The engine performs spatial movement scanning according to the set recognition step size, while simultaneously using the detection granularity as a standard to finely identify changes in attitude angle and differences in spatial position. During this process, based on the risk level corresponding to the attitude constraint anomaly, matching abnormal attitude detection parameters are output (e.g., under high risk level, more precise angle deviation components are output based on finer granularity, and under low risk level, core position drift data is output based on a larger step size). Finally, these parameters are integrated to form an abnormal attitude three-dimensional detection result that includes attitude angle, position coordinates, offset direction, and offset amplitude.

[0042] In one possible implementation, step S510 further includes:

[0043] Step S511: Obtain the type of spatial constraint corresponding to the attitude constraint anomaly, and determine the type risk value based on the constraint type.

[0044] Step S512: Calculate the abnormal constraint risk level based on the spatial anomaly offset distance aligned with the tunneling monitoring work cycle and the spatial constraint cycle, combined with the constraint tolerance threshold of the abnormal constraint type. The abnormal constraint risk level is used to characterize the degree of constraint violation of the current attitude anomaly.

[0045] Step S513: Quantify the risk value of the type by fusing it with the risk level of the abnormal constraint to obtain the risk value of the constraint impact.

[0046] Specifically, by matching the location information of the identified attitude constraint anomalies in standard spatial coordinates with a preset spatial constraint type database, the spatial constraint type corresponding to the anomaly is determined (such as explosion-proof risk zones, gas leakage warning zones, and conventional rock strata tunneling zones that may be involved in coal mining scenarios). A preset constraint type-risk value mapping model is invoked (this model presets basic risk parameters based on the inherent hazard attributes, severity of accident consequences, and safety control levels of different spatial constraint types) to quantify and assign risk values ​​to the matched spatial constraint types, and output the corresponding type risk values ​​(for example, explosion-proof risk zones correspond to high type risk values, and conventional tunneling zones correspond to low type risk values).

[0047] Based on the time-space matching relationship after aligning the tunneling monitoring work cycle with the spatial constraint cycle, the actual distance of the attitude constraint anomaly from the preset constraint boundary in the standard spatial coordinates is extracted as the spatial anomaly offset distance. The constraint tolerance threshold corresponding to the constraint type to which the anomaly belongs (i.e., the maximum allowable offset limit for this type of constraint, such as the constraint tolerance threshold for explosion-proof areas is usually smaller than that for regular areas) is called. By calculating the ratio of the spatial anomaly offset distance to the constraint tolerance threshold, and combining it with the preset grading rules (e.g., ratio ≤20% is low level, 20%~60% is medium level, and >60% is high level), the risk level is divided to obtain the anomaly constraint risk level. This level quantitatively characterizes the actual degree of damage to the spatial constraint boundary caused by the current attitude anomaly (the higher the level, the more serious the degree of constraint damage).

[0048] The abnormal constraint risk level is converted into a quantitative score according to a preset level-value conversion rule (e.g., low level corresponds to 1-3 points, medium level corresponds to 4-6 points, and high level corresponds to 7-9 points). Then, based on the safety control priority of the constraint type, the fusion weight of the type risk value and the abnormal constraint risk quantitative score is set (e.g., in high-risk types such as explosion-proof, the type risk value weight is set to 0.6, and the abnormal constraint risk quantitative score weight is set to 0.4; in the regular type, the weights of both are set to 0.5). The weighted calculation formula "Constraint impact risk value = type risk value × type weight + abnormal constraint risk quantitative score × abnormal weight" is used to calculate the quantitative result of 0-10 points (the higher the score, the higher the comprehensive risk). This realizes the fusion quantification of the two risk parameters, and the final output is a constraint impact risk value that can be directly used to configure the identification step size and detection granularity.

[0049] In one possible implementation, step S510 further includes:

[0050] Step S514: Set a dynamic parameter fuzzy list, which includes the constraint risk type, the impact risk value and the corresponding recognition step size and detection granularity. The recognition step size is the smallest spatial interval unit in which the detection engine moves or scans in the standard spatial coordinates during the execution of 3D offset positioning or attitude anomaly search. The detection granularity is the detection accuracy of attitude angle changes or the smallest recognizable unit of the spatial search unit during attitude offset detection.

[0051] Step S515: Using the constraint influence risk value and constraint type as input, match in the dynamic parameter fuzzy list to obtain the recognition step size and detection granularity.

[0052] Specifically, a dynamic parameter fuzzy list is pre-constructed, containing multiple sets of associated parameter data. Each set of data corresponds to a specific constraint risk type (such as explosion-proof risk zones, gas leakage early warning zones, and conventional rock strata tunneling zones in coal mining), a numerical range of constraint impact risk values ​​(such as 0~3 points, 4~6 points, 7~10 points), and identification step size and detection granularity parameters matching the type and range. The identification step size is defined as the smallest spatial interval unit for each movement or scan when the detection engine performs 3D offset positioning or attitude anomaly search in standard spatial coordinates (it is clear that a smaller step size allows for more detailed searching and is suitable for anomaly detection in high-precision, high-risk areas, while a larger step size results in higher search efficiency and is suitable for low-risk or coarse positioning scenarios). The detection granularity is defined as the detection accuracy of attitude angle changes during attitude offset detection (it is clear that a finer granularity can detect smaller offsets or attitude changes and is suitable for areas with high attitude accuracy requirements or close to risk boundaries, while a coarser granularity is suitable for areas with lower accuracy requirements to improve processing efficiency) or the smallest identifiable unit of the spatial search unit. Through this parameter association setting, a dynamic configuration system covering different risk levels and accuracy requirements is formed.

[0053] The obtained constraint impact risk value and the determined constraint type are used as joint search conditions and input into a preset dynamic parameter fuzzy list. By filtering entries in the list that completely match the constraint type and locating the numerical range of the constraint impact risk value, the corresponding parameter configuration is called to obtain the recognition step size (the smallest spatial interval unit in which the detection engine moves or scans in standard spatial coordinates) and detection granularity (the smallest recognizable unit of the detection accuracy for attitude angle changes or the smallest recognizable unit of the spatial search unit). For example, if the constraint type is an explosion-proof risk area and the constraint impact risk value is in the range of 7 to 10 points, then the corresponding small recognition step size (e.g., 0.1 meters) and fine detection granularity (e.g., 0.1°) in the matching list are used. If it is a conventional tunneling area and the risk value is in the range of 0 to 3 points, then a larger recognition step size (e.g., 0.5 meters) and coarser detection granularity (e.g., 0.5°) are used to match, thereby obtaining parameters that are adapted to the current risk state.

[0054] In one possible implementation, step S100 further includes:

[0055] Step S110: Obtain the position coordinates and corresponding spatial coordinates of each regional node within the tunneling area.

[0056] Step S120: Construct a multi-level spatial coordinate system, including a geographic coordinate system, a tunneling design coordinate system, and a tunneling machine equipment body coordinate system.

[0057] Step S130: Project the position coordinates and corresponding spatial coordinates of each region node into the multi-level spatial coordinate system to obtain the coordinate differences.

[0058] Step S140: Perform coordinate transformation and alignment based on the coordinate differences, and perform spatial coordinate transformation and projection according to the tunneling constraint parameters of each regional node to obtain the core spatial constraint coordinates.

[0059] Specifically, within the designated tunneling area, the actual coordinates of all key nodes (including but not limited to tunneling start point, end point, path turning point, risk area boundary point, and support structure location point) within the area are collected through on-site surveying, analysis of engineering design drawings, or pre-set geological survey data. At the same time, the theoretical spatial coordinates of these nodes in the tunneling engineering design scheme (such as coordinates on the design path, coordinates of the planned support location, etc.) are obtained, forming a coordinate dataset containing the correspondence between the actual location and theoretical space of nodes in each area. This provides basic data support for the subsequent construction of a multi-level spatial coordinate system and coordinate transformation.

[0060] A multi-level spatial coordinate system is constructed, consisting of a geographic coordinate system, a tunneling design coordinate system, and a tunneling machine equipment coordinate system. The geographic coordinate system, referencing geodetic benchmarks (such as a nationally unified geographic coordinate system), accurately represents the absolute geographical location of each node within the tunneling area, ensuring consistency with the external macro-geographic space. The tunneling design coordinate system, established based on the tunneling project's design drawings and planned paths, uses the design starting point as a reference, reflecting the ideal spatial layout of the tunneling path, target cross-section, and key control points, directly corresponding to the expected trajectory of the engineering design. The tunneling machine equipment coordinate system uses key components of the tunneling machine itself (such as the cutterhead center, the machine's rotation center, or positioning reference points) as its origin, with coordinate axes related to the equipment's structural characteristics (such as the machine's longitudinal, lateral, and vertical directions), used to describe the tunneling machine's attitude (such as pitch angle and yaw angle) and relative position in real time. These three levels of coordinate systems are interconnected, collectively forming a complete spatial positioning system covering macro-geography, design planning, and the equipment itself.

[0061] The geographic coordinate system parameters (such as latitude, longitude, and altitude) and the tunneling design coordinate system parameters (such as relative coordinates of the design path) corresponding to the location coordinates of each regional node are clearly defined. Then, the location coordinates are projected from the geographic coordinate system to the tunneling design coordinate system and the tunneling machine equipment body coordinate system through a coordinate transformation algorithm, and the spatial coordinates are projected from the tunneling design coordinate system to the geographic coordinate system and the tunneling machine equipment body coordinate system. Subsequently, the projection results of the same node under different coordinate systems are numerically compared, and the differences of coordinate components (X, Y, Z axes) and three-dimensional spatial distance deviations are calculated to form a coordinate difference dataset, which includes the specific deviation values ​​of each node between the geographic coordinate system and the tunneling design coordinate system, the tunneling design coordinate system and the equipment body coordinate system, and the geographic coordinate system and the equipment body coordinate system.

[0062] Based on the obtained coordinate difference dataset, a transformation matrix (including rotation parameters, translation parameters, and scaling factors) is constructed between the geographic coordinate system, the tunneling design coordinate system, and the tunneling machine equipment body coordinate system. Matrix operations are used to transform and align the coordinates of each coordinate system, eliminating systematic deviations between different coordinate systems and ensuring that the coordinates of nodes in each region remain consistent under a unified benchmark. Subsequently, the preset tunneling constraint parameters of each regional node (such as the boundary coordinate tolerance of risk areas, the attitude deviation threshold of the critical path, and the positional accuracy requirements of the support structure) are called. Based on the aligned coordinates, spatial transformation and projection are performed on the coordinates according to the constraint parameters (such as correcting the offset of coordinates that exceed the constraint range and enhancing the accuracy of high-risk node coordinates). Finally, the core spatial constraint coordinates that can reflect the constraint requirements of the entire region are integrated to serve as the benchmark for subsequent tunneling task cycle segmentation and coordinate alignment.

[0063] In one possible implementation, step S140 further includes:

[0064] Step S141: Calculate the position coordinates and spatial coordinates of each region node in each coordinate system. Based on the coordinate difference values, perform coordinate transformation and alignment operations to establish the transformation relationship matrix between each coordinate system.

[0065] Step S142: According to the tunneling constraint parameters corresponding to each regional node, including the allowable deviation range of attitude, the tolerance of tunneling path, and the boundary information of risk areas, perform spatial transformation and projection on the aligned coordinates to obtain the core spatial constraint coordinates of a unified standard, which are used as the spatial reference benchmark for attitude detection and anomaly identification.

[0066] Specifically, the numerical differences (including differences in X, Y, and Z axis components and three-dimensional spatial distance deviations) between the location coordinates and spatial coordinates of each regional node in the geographic coordinate system, the tunneling design coordinate system, and the tunneling machine equipment body coordinate system are calculated to obtain coordinate difference values. Based on these difference values, coordinate transformation algorithms (such as the seven-parameter transformation method) are used to solve for the rotation matrix, translation vector, and scaling factor between coordinate systems, constructing the transformation relationship matrix between coordinate systems at each level. Through matrix operations, coordinates under different coordinate systems are transformed to the same reference, completing the coordinate transformation and alignment operation, and ensuring the coordinate consistency of each regional node in the multi-level coordinate system.

[0067] The system calls upon the tunneling constraint parameters corresponding to each regional node, including the allowable attitude deviation range (such as the maximum allowable offset values ​​of pitch angle and yaw angle), tunneling path tolerance (such as the allowable deviation width on both sides of the design path), and risk area boundary information (such as the coordinate boundaries and buffer zone range of explosion-proof zones and high-gas zones). Based on the aligned coordinates, the system performs spatial transformation and projection on the coordinates according to the constraint parameters (such as enhancing the accuracy of node coordinates near the risk area boundary and correcting the offset of coordinates exceeding the path tolerance). All node coordinates are uniformly transformed to a preset standard coordinate framework, ultimately obtaining the core spatial constraint coordinates. These coordinates serve as the spatial reference benchmark for subsequent attitude detection and anomaly identification, accurately reflecting the constraint requirements and safety boundaries of each region.

[0068] In one possible implementation, step S200 further includes:

[0069] Step S210: Decompose the tunneling task into periodic nodes, identify the attitude relationship type based on the propulsion logic and attitude changes between nodes, and determine the attitude relationship type between nodes, including alignment and connection relationship or distribution order relationship.

[0070] Step S220: According to the attitude relationship type and tunneling constraint parameters, perform attitude constraint segmentation and spatial region labeling on each node to form a full-cycle tunneling coordinate chain.

[0071] Step S230: Based on the full-cycle coordinate chain, and using the core spatial constraint coordinates as the reference coordinates, perform multi-level coordinate transformation and attitude relationship alignment to generate standard spatial coordinates that include full-cycle tunneling attitude constraints.

[0072] Specifically, based on the overall task cycle of the tunneling project (such as the total length of tunnel excavation and the division of coal seam mining operation stages), the tunneling task is decomposed into several continuous periodic nodes according to the construction rhythm (for example, each 5 meters of tunnel excavation is set as a node, or daily operation nodes are divided according to shifts); based on the advancement logic between each periodic node (such as continuous tunneling in straight sections, advancement in curved sections according to the designed curvature, and adjustment of direction in turning sections according to the planned angle, etc.), combined with the attitude change data of adjacent nodes during the tunneling process (including continuous recording of parameters such as azimuth angle, pitch angle, and deflection angle), the attitude relationship type between nodes is analyzed and identified. Among them, the attitude alignment and continuity relationship means that the attitude of the subsequent node must strictly and continuously inherit the attitude of the previous node, maintaining the continuity of direction and attitude angle (e.g., when tunneling a straight section, the subsequent node must maintain the same direction as the preceding node, and the angle deviation must be controlled within the allowable range to avoid deviation); the attitude distribution order relationship means that there are structural changes in the attitude between nodes and there is no need for continuous continuity (e.g., when tunneling a curved section or turning section, the attitude of each node must be gradually adjusted according to the design curvature, and there is an angle difference within a preset range between adjacent nodes, or in coal seam mining, the tunneling direction needs to be adjusted due to changes in geological conditions, and the attitude between nodes shows a phased change according to the planned order). Through this analysis, the specific attitude correlation attributes between each node are clarified.

[0073] Based on the determined attitude relationship type (alignment and connection relationship or distribution order relationship), and combined with the tunneling constraint parameters corresponding to each cycle node (such as the attitude continuous deviation threshold of alignment and connection nodes, the independent constraint boundary of distribution order nodes, the coordinate range of risk areas, etc.), attitude constraint segmentation is performed on each node. For nodes with alignment and connection relationships, continuous attitude constraint intervals are set (ensuring seamless connection of constraint parameters between nodes, such as angle change gradient not exceeding 1° / meter). For nodes with distribution order relationships, independent constraint boundaries are defined (avoiding overlap of constraint ranges of different nodes, such as setting a 0.5-meter boundary buffer zone). At the same time, in the core spatial constraint coordinate system, spatial regions are labeled according to the constraint attributes of nodes (such as straight segments, turning segments, high-risk areas, etc.), and the constraint information of all nodes is sequentially linked in the tunneling advancement order to form a complete tunneling full-cycle coordinate chain covering the starting node to the ending node. This coordinate chain includes the attitude constraint range, spatial boundary, and correlation of each node, providing structured data support for subsequent coordinate transformation and alignment.

[0074] Based on the full-cycle coordinate chain, and using the core spatial constraint coordinates as a unified benchmark, this method considers the coordinate transformation relationships caused by differences in benchmark positions during data acquisition for coordinate data corresponding to different stages of the tunneling process (such as straight sections, curved sections, and turning sections). It calls a pre-built transformation matrix (containing rotation, translation, and scaling parameters) to perform multi-level coordinate switching transformations. First, the local coordinates of each stage are transformed to the tunneling design coordinate system, then mapped to the geographic coordinate system, and finally unified to the core spatial constraint coordinate system. Simultaneously, attitude alignment is performed based on the attitude relationship types between nodes: aligning nodes with continuity relationships, and eliminating transformation deviations through attitude parameter interpolation to ensure continuity of attitude trajectories between stages; for nodes with distributed order relationships, adjusting coordinate associations according to preset structural evolution rules to ensure that attitude changes meet design curvature or turning requirements; finally, integrating all transformation and alignment results generates standard spatial coordinates containing attitude constraint parameters for each stage within the full cycle (such as angle thresholds, path boundaries, and risk area coordinates), providing a unified benchmark for full-process attitude detection.

[0075] In one possible implementation, step S300 further includes:

[0076] Step S310: Based on the deployment location and monitoring parameters of the monitoring equipment, construct a monitoring radar map for each monitoring device.

[0077] Step S320: Based on the relationship between the monitoring equipment and the tunneling environment, and combined with the analysis of abnormal samples, analyze the abnormal offset data of each monitoring equipment, and add the abnormal offset data to its monitoring radar chart.

[0078] Step S330: Overlay the monitoring radar images of all monitoring devices in multiple dimensions to obtain the associated interactive monitoring data groups and the interactive data compensation and correction relationships.

[0079] Step S340: Correct the associated interactive monitoring data group according to the interactive data compensation and correction relationship, and construct the monitoring data network by performing location spatial networking according to the corrected monitoring radar map of each monitoring device.

[0080] Specifically, the precise deployment locations of each monitoring device within the tunneling area are clearly defined (e.g., near the tunneling machine cutterhead, at the top support structure of the roadway, at the boundary of the risk area, etc.), and these are used as the origin coordinates of the radar chart. Then, based on the monitoring parameter types of the equipment (e.g., three-dimensional attitude angle, vibration acceleration, gas concentration, temperature, etc.), each parameter is set as a radial dimension of the radar chart, and each dimension is divided into scale intervals according to the normal value range of the parameter (e.g., the attitude angle dimension has an upper limit of ±5°, and the concentration dimension has a safety threshold as the critical line). Finally, using polar coordinate plotting, with the origin as the center and each parameter dimension as the radius direction, a monitoring radar chart representing the monitoring coverage range and parameter benchmark threshold of a single device is drawn, so that the monitoring capability and normal state boundary of the equipment can be intuitively presented in the chart.

[0081] First, the influence relationship between each monitoring device and the tunneling environment is analyzed (such as the frequency response characteristics of vibration sensors affected by the impact of tunneling machine operation, the sensitivity changes of gas detectors affected by roadway airflow disturbances, and the measurement deviation patterns of attitude sensors affected by geological structures). Then, combined with historical abnormal samples (such as tunneling machine attitude deviation exceeding the standard, gas concentration surge records, and abnormal equipment vibration data), the specific deviation data of different devices under abnormal conditions are analyzed (including the magnitude, duration, and spatial distribution characteristics of parameters deviating from normal thresholds, such as abnormal attitude angle deviation of ±3°, concentration exceeding the threshold by 20%). Subsequently, these abnormal deviation data are added to the monitoring radar chart of the corresponding device in a differentiated visual form (such as red warning lines, shaded areas exceeding the benchmark range, etc.), so that the radar chart can not only reflect the normal range of parameters, but also intuitively show the degree and distribution of deviation under abnormal conditions, forming a monitoring radar chart containing both benchmark and abnormal information.

[0082] Based on the spatial layout coordinates of each monitoring device, the monitoring radar images of all devices are superimposed in three dimensions (covering the synchronization of spatial positions on the X, Y, and Z axes and the time dimension). The overlapping areas of different radar images are identified through the spatial intersection algorithm (such as the area monitored by multiple devices within 5 meters around the tunneling machine), and the associated parameter data in these areas (such as the attitude angle and vibration value of a spatial point at the same time) are extracted to form an associated interactive monitoring data group. At the same time, the monitoring differences of the same parameters in the overlapping areas are compared (such as the deviation of gas concentration measurement values ​​of different devices at the same location). Combining the characteristics of the equipment and environmental influencing factors (such as differences in sensor accuracy and the influence of airflow on concentration detection), the data compensation coefficient (such as adding 0.02% to the data of a certain device for correction) and the correction formula are calculated to establish the interactive data compensation correction relationship and clarify the calibration logic between the data of different devices.

[0083] Based on the compensation coefficients and correction formulas determined in the interactive data compensation and correction relationship, a data fusion algorithm (such as Kalman filtering) is used to iteratively correct the associated interactive monitoring data group. For the monitoring values ​​of multiple devices with the same parameter in the overlapping area, weighted fusion is performed according to the device weight allocation to eliminate random errors. For device data with deviations, directional calibration is performed according to the correction formula. After the correction is completed, the spatial coordinate boundaries and effective monitoring ranges of each monitoring device in the corrected monitoring radar map are extracted. The devices are mapped to the three-dimensional coordinate system according to their actual deployment positions through a spatial topology algorithm. A wireless communication link between devices is established using Mesh self-organizing network technology (such as low-latency data interaction based on the ZigBee protocol), and a data verification mechanism is embedded (such as triggering cross-validation when the parameter deviation of adjacent devices exceeds the limit). Finally, a monitoring data network with real-time data interaction, dynamic error compensation, and full-area coverage capabilities is constructed.

[0084] In one possible implementation, step S400 further includes:

[0085] Step S410: Based on the time-series attitude data collected by each monitoring device in the monitoring data network, construct the attitude change trajectory of the tunneling machine within the current working cycle.

[0086] Step S420: Using the historical attitude angle change trend within the time window, the attitude state of the upcoming period node is predicted by curve fitting.

[0087] Step S430: Project the estimated attitude state onto the corresponding target node position in the standard spatial coordinates, and compare it with the attitude constraint parameters marked on the node to calculate the predicted offset, including angle deviation, position drift and its direction component.

[0088] Step S440: If the offset exceeds the set attitude tolerance threshold, it is determined that there is an abnormal attitude constraint trend, and the attitude constraint abnormality is output.

[0089] Specifically, by monitoring the real-time data interface of the data network, time-series data (including 3D coordinates, pitch angle, yaw angle, and timestamp information at 30 frames per second) collected by various attitude sensors, GNSS positioning modules, and inertial measurement units are retrieved. The multi-source data is synchronized to millisecond-level accuracy using timestamp alignment technology, data noise is eliminated using the Kalman filter algorithm, and discrete data points are smoothed using polynomial interpolation. The processed data is then mapped to 3D space by combining standard spatial coordinate system parameters, ultimately generating a continuous attitude change trajectory curve with time as the horizontal axis and spatial position and attitude angle as the vertical axis. The trajectory contains attitude feature values ​​and corresponding spatial coordinate markers at each moment.

[0090] A fixed time window (e.g., the first 30 minutes or the first 3 completed cycle nodes) is set. Historical attitude angle (including pitch, yaw, and azimuth) data sequences within this window are extracted from the constructed attitude change trajectory. The rate of change and trend characteristics of the attitude angles over time (e.g., uniform angle increase, periodic fluctuation, or phased stability) are calculated using the sliding window analysis method. Based on the extracted trend characteristics, a suitable curve fitting model is selected (e.g., linear fitting for uniform change, exponential fitting for accelerated change, and sine curve fitting for periodic change). The historical attitude angle data is substituted into the model for parameter solving and curve optimization, ensuring the fitted curve best matches the historical change trend. Subsequently, based on the extension pattern of the fitted curve, the attitude angle value corresponding to the upcoming cycle node (e.g., the end point of the next work segment) on the time axis is calculated, completing the prediction of the attitude state at that node. The prediction result includes specific angle values ​​and trend reliability indicators.

[0091] The estimated attitude state (including the estimated attitude angle and corresponding spatial position) is projected into a standard spatial coordinate system through a coordinate transformation algorithm, accurately locating the corresponding target period node position within the system. Subsequently, the attitude constraint parameters of the target node pre-labeled in the standard spatial coordinate system (such as the maximum allowable pitch angle ±1.5°, yaw angle ±2°, and position boundary coordinate range) are retrieved. Each parameter of the estimated attitude state is compared with the constraint parameters one by one, and the predicted offset is calculated through spatial vector operations. The angle deviation is the difference between the estimated attitude angle and the upper / lower limit of the constraint angle (e.g., the deviation between the estimated pitch angle of 3° and the upper limit of the constraint of 2° is 1°). The position drift is the three-dimensional straight-line distance between the estimated position and the reference position of the target node. The directional component is obtained by decomposing the drift vector to obtain the offset values ​​on the X, Y, and Z axes (e.g., a positive offset of 0.3 meters on the X-axis and a negative offset of 0.1 meters on the Y-axis), fully presenting the magnitude and direction of the offset.

[0092] The calculated predicted offset (including angle deviation, position drift and its directional component) is compared with the preset attitude tolerance threshold (such as angle deviation threshold ±1°, position drift threshold ±0.5m, and the directional component must be within the safe area). If any parameter in the offset (such as angle deviation reaching 1.2°, position drift reaching 0.6m, or the directional component pointing to a non-planned area) exceeds the corresponding tolerance threshold, it is determined that the tunneling machine has an abnormal attitude constraint trend. The abnormal response mechanism is then triggered, and an attitude constraint abnormality report containing the abnormality type (angle exceeding limit, position deviation or directional deviation), specific offset value, associated target node information and risk level is output, providing a clear basis for subsequent adjustment operations.

[0093] Example 2, based on the same inventive concept as the three-dimensional attitude detection method for tunneling machine attitude deviation anomalies in the previous examples, such as... Figure 2As shown, this application provides a three-dimensional attitude detection system for tunnel boring machine attitude deviation anomalies. The system and method embodiments in this application are based on the same inventive concept. The system includes:

[0094] The constraint coordinate recognition module 10 is used to identify the core spatial constraint coordinates of the tunneling task based on the tunneling area range.

[0095] The standard spatial coordinate establishment module 20 is used to perform periodic node segmentation and coordinate alignment of the tunneling task based on the core spatial constraint coordinates as the reference coordinates, and to establish standard spatial coordinates, which include full-cycle tunneling spatial constraints.

[0096] The monitoring data network construction module 30 is used to connect the tunneling machine monitoring equipment during the tunneling process, acquire real-time tunneling monitoring data, and perform data interaction compensation according to the deployment location of the monitoring equipment and the characteristics of the collected data to construct a monitoring data network.

[0097] The attitude constraint anomaly identification module 40 is used to predict the tunneling attitude by using the monitoring data network and project it into the standard spatial coordinates, and to identify attitude constraint anomalies by aligning and matching the tunneling monitoring work cycle with the spatial constraint cycle.

[0098] The three-dimensional detection result acquisition module 50 is used to perform three-dimensional offset positioning in the standard spatial coordinates according to the abnormal posture constraint, and obtain the three-dimensional detection result of the abnormal posture, including posture angle, position coordinates, offset direction and offset amplitude.

[0099] Furthermore, the system is also used to implement the following functions:

[0100] Based on the risk value of the constraint impact of the posture constraint anomaly, the recognition step size and detection granularity are configured; according to the recognition step size and detection granularity, the offset positioning engine is driven to perform three-dimensional offset search positioning in the standard spatial coordinates, and the abnormal posture detection parameters matching the risk level are output to obtain the three-dimensional detection result of the abnormal posture.

[0101] Furthermore, the system is also used to implement the following functions:

[0102] Obtain the type of spatial constraint corresponding to the attitude constraint anomaly, and determine the type risk value based on the constraint type; calculate the anomaly constraint risk level according to the spatial anomaly offset distance aligned with the tunneling monitoring work cycle and the spatial constraint cycle, combined with the constraint tolerance threshold of the anomaly constraint type, and the anomaly constraint risk level is used to characterize the degree of constraint violation of the current attitude anomaly; use the type risk value and the anomaly constraint risk level to fuse and quantify, and obtain the constraint impact risk value.

[0103] Furthermore, the system is also used to implement the following functions:

[0104] A dynamic parameter fuzzy list is set up, which includes constraint risk type, impact risk value, and corresponding recognition step size and detection granularity. The recognition step size is the smallest spatial interval unit in which the detection engine moves or scans in standard spatial coordinates during the execution of 3D offset localization or attitude anomaly search. The detection granularity is the detection accuracy of attitude angle changes or the smallest identifiable unit of spatial search unit during attitude offset detection. Using the constraint impact risk value and constraint type as input, the recognition step size and detection granularity are obtained by matching in the dynamic parameter fuzzy list.

[0105] Furthermore, the system is also used to implement the following functions:

[0106] Within the tunneling area, obtain the position coordinates and corresponding spatial coordinates of each regional node;

[0107] A multi-level spatial coordinate system is constructed, including a geographic coordinate system, a tunneling design coordinate system, and a tunneling machine equipment body coordinate system. The position coordinates and corresponding spatial coordinates of each regional node are projected into the multi-level spatial coordinate system to obtain coordinate differences. Coordinate transformation and alignment are performed based on the coordinate differences, and spatial coordinate transformation and projection are performed according to the tunneling constraint parameters of each regional node to obtain the core spatial constraint coordinates.

[0108] Furthermore, the system is also used to implement the following functions:

[0109] Calculate the position coordinates and spatial coordinate differences of each regional node in each coordinate system. Based on the coordinate differences, perform coordinate transformation and alignment operations to establish a transformation relationship matrix between each coordinate system. According to the tunneling constraint parameters corresponding to each regional node, including the allowable deviation range of attitude, tunneling path tolerance, and risk area boundary information, perform spatial transformation and projection on the aligned coordinates to obtain a unified standard core spatial constraint coordinate, which is used as a spatial reference benchmark for attitude detection and anomaly identification.

[0110] Furthermore, the system is also used to implement the following functions:

[0111] The tunneling task is decomposed into periodic nodes. Based on the propulsion logic and attitude changes between nodes, the attitude relationship type is identified, and the attitude relationship type between nodes is determined, including alignment and succession relationship or distribution order relationship. According to the attitude relationship type and tunneling constraint parameters, each node is segmented into attitude constraints and spatial regions are labeled to form a full-cycle tunneling coordinate chain. Based on the full-cycle coordinate chain, multi-level coordinate transformation and attitude relationship alignment are performed with the core spatial constraint coordinates as the reference coordinates to generate standard spatial coordinates that include full-cycle tunneling attitude constraints.

[0112] Furthermore, the system is also used to implement the following functions:

[0113] Based on the deployment location and monitoring parameters of the monitoring equipment, a monitoring radar chart for each monitoring device is constructed. Based on the relationship between the monitoring equipment and the tunneling environment, and combined with anomaly sample analysis, abnormal offset data of each monitoring device is analyzed and added to its monitoring radar chart. The monitoring radar charts of all monitoring devices are overlaid in multiple dimensions to obtain associated interactive monitoring data groups and interactive data compensation and correction relationships. The associated interactive monitoring data groups are corrected according to the interactive data compensation and correction relationships, and spatial networking is performed according to the corrected monitoring radar charts of each monitoring device to construct the monitoring data network.

[0114] Furthermore, the system is also used to implement the following functions:

[0115] Based on the time-series attitude data collected by each monitoring device in the monitoring data network, the attitude change trajectory of the tunneling machine within the current working cycle is constructed; using the historical attitude angle change trend within the time window, the attitude state of the upcoming cycle node is estimated through curve fitting; the estimated attitude state is projected onto the corresponding target node position in the standard spatial coordinates and compared with the attitude constraint parameters marked on the node to calculate the predicted offset, including angle deviation, position drift and its direction component; if the offset exceeds the set attitude tolerance threshold, it is determined that there is an abnormal attitude constraint trend, and the attitude constraint abnormality is output.

[0116] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0117] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0118] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.

Claims

1. A three-dimensional posture detection method for posture deviation anomaly of a roadheader, characterized by, The method comprises the following steps: identifying core space constraint coordinates of a tunneling task based on a tunneling region range; performing period node segmentation and coordinate alignment on the tunneling task according to the core space constraint coordinates as reference coordinates, and establishing standard space coordinates, wherein the standard space coordinates comprise full-period tunneling space constraints; connecting tunneling machine monitoring devices during tunneling to obtain real-time tunneling monitoring data, and performing data interaction compensation according to the layout positions of the monitoring devices and the characteristics of the collected data to construct a monitoring data network; projecting a tunneling posture into the standard space coordinates using the monitoring data network, aligning and matching based on the tunneling monitoring work period and the space constraint period, and identifying posture constraint abnormalities; performing three-dimensional offset positioning of the posture constraint abnormalities in the standard space coordinates to obtain three-dimensional detection results of the abnormal postures, including posture angles, position coordinates, offset directions, and offset amplitudes; The method of identifying core space constraint coordinates of a tunneling task based on a tunneling region range comprises the following steps: obtaining position coordinates and corresponding space coordinates of each regional node within the tunneling region range; constructing a multi-level space coordinate system, including a geographic coordinate system, a tunneling design coordinate system, and a tunneling machine device body coordinate system; projecting the position coordinates and corresponding space coordinates of each regional node into the multi-level space coordinate system to obtain coordinate differences; performing coordinate conversion alignment according to the coordinate differences, and performing space coordinate conversion projection according to the tunneling constraint parameters of each regional node to obtain the core space constraint coordinates; wherein the tunneling constraint parameters include posture allowable deviation range, tunneling path tolerance, and risk region boundary information; The method of performing period node segmentation and coordinate alignment on the tunneling task according to the core space constraint coordinates as reference coordinates, and establishing standard space coordinates, wherein the standard space coordinates comprise full-period tunneling space constraints, comprises the following steps: performing period node decomposition of the tunneling task, identifying posture relationship types based on the propulsion logic and posture changes between nodes, and determining the posture relationship types between nodes, including alignment connection relationship or distribution order relationship; performing posture constraint segmentation and space region labeling on each node according to the posture relationship types and tunneling constraint parameters to form a full-period tunneling coordinate chain; performing multi-level coordinate conversion and posture relationship alignment based on the full-period tunneling coordinate chain and the core space constraint coordinates as reference coordinates to generate standard space coordinates containing full-period tunneling posture constraints.

2. The three-dimensional attitude detection method for attitude offset anomaly of a heading machine according to claim 1, characterized in that, The method of performing three-dimensional offset positioning of the posture constraint abnormalities in the standard space coordinates to obtain three-dimensional detection results of the abnormal postures comprises the following steps: configuring identification step length and detection granularity based on the constraint influence risk value of the posture constraint abnormalities; driving an offset positioning engine to perform three-dimensional offset search positioning in the standard space coordinates according to the identification step length and detection granularity, outputting abnormal posture detection parameters matched with risk levels, and obtaining the three-dimensional detection results of the abnormal postures.

3. The three-dimensional attitude detection method for attitude offset anomaly of a heading machine according to claim 2, characterized by, Before the step of configuring identification step length and detection granularity based on the constraint influence risk value of the posture constraint abnormalities, the method comprises the following steps: obtaining the type of space constraint corresponding to the posture constraint abnormalities, and determining type risk values based on the constraint type. According to the space anomaly offset distance aligned with the excavation monitoring work period and the space constraint period, and in combination with the constraint tolerance threshold of the anomaly constraint type, an anomaly constraint risk level is calculated, and the anomaly constraint risk level is used to represent the constraint damage degree of the current attitude anomaly; The type risk value and the anomaly constraint risk level are fused and quantized to obtain a constraint impact risk value.

4. The three-dimensional attitude detection method for attitude offset anomaly of a heading machine according to claim 3, characterized by, Based on the constraint impact risk value of the attitude constraint anomaly, a recognition step and a detection granularity are configured, including: A dynamic parameter fuzzy list is set, which includes a constraint risk type, an impact risk value, and corresponding recognition steps and detection granularities. The recognition step is the minimum space interval unit of the detection engine in the standard space coordinates in the three-dimensional offset positioning or attitude anomaly search process. The detection granularity is the detection accuracy of the attitude angle change or the minimum identifiable unit of the space search unit in the attitude offset detection process. The constraint impact risk value and the constraint type are used as inputs to match in the dynamic parameter fuzzy list to obtain the recognition step and the detection granularity.

5. The three-dimensional pose detection method for pose offset anomaly of a heading machine according to claim 1, characterized in that, According to the coordinate difference, the coordinate conversion alignment is performed, and the space coordinate conversion projection is performed according to the excavation constraint parameters of each regional node to obtain the core space constraint coordinates, including: The coordinate difference values of the position coordinates and the space coordinates of each regional node in each layer coordinate system are calculated, and based on the coordinate difference values, the coordinate conversion alignment operation is performed to establish the transformation relationship matrix between each coordinate system; According to the corresponding excavation constraint parameters of each regional node, the space transformation projection is performed on the aligned coordinates to obtain the unified standard core space constraint coordinates, which are used as the space reference benchmark for attitude detection and anomaly recognition.

6. The three-dimensional attitude detection method for attitude offset anomaly of a heading machine according to claim 1, characterized by, According to the layout position of the monitoring device and the characteristics of the collected data, a monitoring data network is constructed, including: According to the layout position of the monitoring device and the monitoring parameters, a monitoring radar chart of each monitoring device is constructed; According to the abnormal offset data of each monitoring device, the abnormal offset data is added to the monitoring radar chart of the monitoring device according to the relationship between the monitoring device and the excavation environment and the abnormal sample analysis; All monitoring radar charts of the monitoring devices are overlapped in multiple dimensions to obtain an associated interactive monitoring data group and an interactive data compensation correction relationship; According to the interactive data compensation correction relationship, the associated interactive monitoring data group is corrected, and the position space network is constructed according to the corrected monitoring radar chart of each monitoring device to construct the monitoring data network.

7. The three-dimensional pose detection method for pose offset anomaly of a heading machine according to claim 1, characterized in that, The monitoring data network is used to predict the excavation attitude projection into the standard space coordinates, and based on the alignment matching of the excavation monitoring work period and the space constraint period, the attitude constraint anomaly is identified, including: Based on the time series attitude data collected by each monitoring device in the monitoring data network, an attitude change trajectory in the current work period of the excavator is constructed; The attitude state of the upcoming period node is estimated by curve fitting based on the historical attitude angle change trend in the time window. The estimated attitude state is projected into a corresponding target node position in the standard space coordinate, and compared with the attitude constraint parameter marked by the node to calculate a pre-judgment offset, including an angle deviation, a position drift and a direction component thereof; If the offset exceeds a set attitude tolerance threshold, it is determined that there is an attitude constraint abnormal trend, and an attitude constraint abnormality is output.

8. A three-dimensional posture detection system for posture deviation anomaly of a roadheader, characterized by, The system is used to implement the three-dimensional attitude detection method for attitude offset abnormality of the heading machine according to any one of claims 1-7, and the system comprises: A constraint coordinate identification module (10) is configured to identify core space constraint coordinates of a heading task based on a heading region range; A standard space coordinate establishment module (20) is configured to take the core space constraint coordinates as reference coordinates, perform periodic node segmentation and coordinate alignment on the heading task, and establish standard space coordinates, wherein the standard space coordinates comprise full-period heading space constraints; A monitoring data network construction module (30) is configured to connect a heading machine monitoring device in a heading process, acquire real-time heading monitoring data, and construct a monitoring data network according to a layout position of the monitoring device and a data collection feature for data interaction compensation; An attitude constraint abnormality identification module (40) is configured to project the heading attitude into the standard space coordinates by using the monitoring data network, align and match based on a heading monitoring work period and a space constraint period, and identify an attitude constraint abnormality; A three-dimensional detection result acquisition module (50) is configured to perform three-dimensional offset positioning of the attitude constraint abnormality in the standard space coordinates, and acquire a three-dimensional detection result of an abnormal attitude, including an attitude angle, a position coordinate, an offset direction and an offset amplitude.

Images