Robot team collaborative management method and system for high-risk operation of power station

Abstract

The invention discloses a robot team cooperative management method and system for power station high-risk operation, and relates to the technical field of intelligent management, and the method comprises the steps: segmenting a local model from a power station BIM model, aligning the local model with a monitoring region, fitting a routing inspection path through the combination of heterogeneous team configuration, assigning unmanned aerial vehicles to carry out multi-vehicle cooperative scanning, and carrying out multi-vehicle cooperative scanning; and acquiring multi-modal monitoring data. And generating a dynamic risk map through marginal center analysis of the data, performing task matching in combination with a quadruped robot detection capability matrix, and outputting an inspection task sequence. And completing risk collaborative investigation by compensating and optimizing the task sequence. The technical problems that in the prior art, traditional power station inspection depends on manpower or single equipment, monitoring blind areas exist, the inspection efficiency is low, and risks cannot be evaluated in real time are solved, and the technical effects that heterogeneous robots work cooperatively, high-risk areas are covered comprehensively, the inspection efficiency is improved, and dynamic risk evaluation and timely response are achieved are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent management, and particularly to a method and system for collaborative management of robot teams for high-risk operations in power plants. Background Art

[0002] With the continuous expansion of the scale and increasing complexity of power plant equipment, traditional manual inspection and single monitoring methods are no longer sufficient to meet the safety monitoring requirements of high-risk operation areas. Traditional methods usually rely on fixed cameras and manual operations, which have problems such as monitoring blind spots, low efficiency, missed detections, and false detections. Especially in high-risk operation areas of power plants, the viewing angle and coverage of traditional monitoring equipment are limited, and all risk hazards cannot be captured in real time. In addition, manual inspections are slow, unable to respond to sudden risks in a timely manner, and cannot handle complex operating environments, seriously affecting the safety management and operation efficiency of power plants. To solve these problems, inspection systems based on robots, drones, and intelligent data analysis have gradually been applied, but existing technologies still face problems such as insufficient path planning, task allocation, and risk assessment. Summary of the Invention

[0003] This application provides a method and system for collaborative management of robot teams for high-risk operations in power plants, which are used to solve the technical problems in the prior art that traditional power plant inspections rely on manual labor or single equipment, have monitoring blind spots, low inspection efficiency, and cannot evaluate risks in real time.

[0004] In the first aspect of this application, a method for collaborative management of robot teams for high-risk operations in power plants is provided. The method includes: splitting a first local BIM model from the power plant BIM model, where the first local BIM model is horizontally aligned with the first monitoring area; fitting an inspection coverage path according to the heterogeneous team configuration of the first monitoring area and the first local BIM model to obtain a first set of monitoring coverage paths; H drones in the heterogeneous team configuration perform multi-aircraft collaborative scanning of the first monitoring area along the first set of monitoring coverage paths to obtain multi-modal monitoring data; the first edge center receives and performs risk coupling analysis based on the multi-modal monitoring data to generate a dynamic risk map; constructing M detection capability matrices for M quadruped robots in the heterogeneous team configuration, and performing a capability-task matching on the dynamic risk map and the M detection capability matrices to output M inspection task sequences; performing adjacent area inspection compensation optimization on the M inspection task sequences to output M compensation task sequences for performing risk collaborative investigation of the dynamic risk map.

[0005] The second aspect of the present application provides a robot team-based collaborative management system for high-risk operations in power plants. The system includes: a model segmentation module for segmenting a first local BIM model from a power plant BIM model, where the first local BIM model is horizontally aligned with a first monitoring area; an inspection coverage path fitting module for performing inspection coverage path fitting based on the heterogeneous team configuration of the first monitoring area and the first local BIM model to obtain a first set of monitoring coverage paths; a multi-robot collaborative scanning module for performing multi-robot collaborative scanning of the first monitoring area along the first set of monitoring coverage paths by H unmanned aerial vehicles in the heterogeneous team configuration to obtain multi-modal monitoring data; a risk coupling analysis module for performing risk coupling analysis based on the multi-modal monitoring data received by a first edge center and generating a dynamic risk map; a task matching module for constructing M detection capability matrices for M quadruped robots in the heterogeneous team configuration, and performing capability-task matching on the dynamic risk map and the M detection capability matrices to output M inspection task sequences; a risk collaborative investigation module for performing adjacent area inspection compensation optimization on the M inspection task sequences, outputting M compensation task sequences, and performing risk collaborative investigation of the dynamic risk map.

[0006] One or more technical solutions provided in the present application have at least the following technical effects or advantages: The robot team-based collaborative management method and system for high-risk operations in power plants provided by the present application relate to the technical field of intelligent management. By segmenting a local model from a power plant BIM model, aligning it with a monitoring area, fitting an inspection path in combination with a heterogeneous team configuration, assigning unmanned aerial vehicles to scan to obtain multi-modal monitoring data, analyzing and generating a dynamic risk map, and performing task matching and compensation optimization in combination with the capability matrix of quadruped robots, it ensures efficient inspection and risk assessment of high-risk operation areas in power plants. It solves the technical problems in the prior art that traditional power plant inspections rely on manual labor or single equipment, there are monitoring blind spots and low inspection efficiency, and risks cannot be evaluated in real time, and realizes the technical effects of comprehensively covering high-risk areas through heterogeneous robot collaborative operations, improving inspection efficiency, and realizing dynamic risk assessment and timely response. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0008] Figure 1 It is a schematic flowchart of the robot team-based collaborative management method for high-risk operations in power plants provided by the embodiments of the present application; Figure 2 This is a schematic structural diagram of a robotic team-based collaborative management system for high-risk operations in power plants provided by an embodiment of the present application.

[0009] Explanation of reference numerals: Model segmentation module 11, inspection coverage path fitting module 12, multi-robot collaborative scanning module 13, risk coupling analysis module 14, task matching module 15, risk collaborative investigation module 16. Detailed implementation manners

[0010] The present application provides a robotic team-based collaborative management method and system for high-risk operations in power plants, which is used to solve the technical problems in the prior art that traditional power plant inspections rely on manual labor or single equipment, there are monitoring blind spots, the inspection efficiency is low, and the risk cannot be evaluated in real time.

[0011] Next, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0012] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned accompanying drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules not clearly listed or inherent to these processes, methods, products, or devices.

[0013] Embodiment 1, as Figure 1 shown, the present application provides a robotic team-based collaborative management method for high-risk operations in power plants, and the method includes: P10: Segment the first local BIM model from the power plant BIM model, where the first local BIM model is horizontally aligned with the first monitoring area.

[0014] Specifically, first, the overall BIM (Building Information Model) of the power station needs to be processed to split out the first local BIM model. The BIM model is the abbreviation of Building Information Model, which creates a three-dimensional model of the building and its surrounding environment through digital means and contains various data such as the geometric shape, physical properties, and construction information of the building. Through the splitting operation, a sub-model of a specific area within the power station, that is, the first local BIM model, is extracted, and this model is used for further monitoring and risk analysis. The extraction of the local BIM model can be achieved through the technology that combines Geographic Information System (GIS) and BIM, ensuring the accurate extraction of data of a specific area from the large-scale data of the power station for subsequent operations.

[0015] The core feature of the first local BIM model is its ability to represent the spatial information of a specific operation area of the power station. Compared with the BIM model of the entire power station, the spatial scope of the local model is more specific and narrow, focusing on the specific area that needs to be monitored. This enables subsequent work such as monitoring path planning and risk identification to be carried out on a more targeted and efficient basis.

[0016] After splitting out the first local BIM model, it is necessary to further accurately align it with the actual first monitoring area in the horizontal space. The first monitoring area is the first monitoring region, and the subsequent "area" specifically refers to the region. Horizontal space alignment means that in the two-dimensional plane, the coordinate system of the local BIM model is exactly the same as the geographic coordinate system of the monitoring area, ensuring that each point in the model can accurately correspond to the specific position in the actual monitoring area. The alignment process involves coordinate transformation and Geographic Information System (GIS) technology. By matching the coordinate system of the BIM model with the geographic coordinate system in GIS and using coordinate transformation algorithms (such as affine transformation, projection transformation, etc.), the point, line, and surface elements in the BIM model are mapped into the geographic coordinate space of GIS. In this process, it is necessary to use high-precision positioning devices (such as GPS, RTK, etc.) to obtain the actual geographic coordinate data of the monitoring area as the basis for alignment.

[0017] By completing this step, the first local BIM model of the power station can accurately reflect the spatial layout of the monitoring area and provide an accurate spatial basis for subsequent work such as multi-machine collaborative scanning and dynamic risk map generation.

[0018] P20: According to the heterogeneous team configuration of the first monitoring area and the first local BIM model, perform inspection coverage path fitting to obtain the first set of monitoring coverage paths.

[0019] Furthermore, step P20 of the embodiment of the present application further includes: P21: Locally call the overall control information of the power station; P22: Based on the overall control information of the power station, perform collaborative perception area positioning in the first local BIM model to obtain a collaborative perception virtual boundary; P23: Extract the building height distribution information of the collaborative perception virtual boundary from the first local BIM model; P24: Use the heterogeneous team configuration as the equipment scheduling constraint, and use the building height distribution information and the collaborative perception virtual boundary as the monitoring range constraint to perform flight trajectory fitting, and output the first set of monitoring coverage paths.

[0020] It should be understood that according to the first monitoring area division of the power station and the first local BIM model, combined with the heterogeneous team configuration, the inspection coverage path is fitted to generate an optimized inspection path to ensure that the robot can effectively cover the monitoring area.

[0021] First, based on the heterogeneous team configuration of the first monitoring area division and the first local BIM model, start the fitting work of the inspection coverage path. The heterogeneous team configuration refers to the combination and configuration of different types of monitoring equipment (such as drones and quadruped robots) within the monitoring area division. Each piece of equipment has its unique performance parameters, such as flight altitude, endurance, sensor type, etc. These parameters will be used as equipment scheduling constraint conditions for the subsequent path planning process. The first local BIM model provides detailed building structure and spatial layout information within the monitoring area division, including the location of equipment and facilities, obstacle distribution, etc., providing basic geographical data support for path planning.

[0022] Furthermore, locally call the overall control information of the power station. The overall control information of the power station covers aspects such as the overall operation status of the power station, equipment distribution, and operation plan. This information is crucial for understanding the role and importance of the monitoring area division in the entire power station and also provides macro-level guidance for the subsequent path planning.

[0023] Next, based on the called overall control information of the power station, perform collaborative perception area positioning within the first local BIM model to obtain a collaborative perception virtual boundary. The collaborative perception area refers to the area range where different equipment needs to work collaboratively during the monitoring process. By positioning these areas, the range and boundary where the equipment needs to cooperate with each other can be clarified, and then the collaborative perception virtual boundary can be determined. This boundary will be used as one of the monitoring range constraint conditions for the subsequent flight trajectory fitting process.

[0024] Next, extract the building height distribution information of the collaborative perception virtual boundary from the first local BIM model. The building height distribution information reflects the height changes of the buildings within the monitored area, which is particularly important for the flight height planning of the UAV. When the UAV is performing the monitoring task, it needs to adjust its flight height according to the height of the buildings to avoid collisions and ensure the accuracy of the monitoring data.

[0025] Finally, use the heterogeneous team configuration as the equipment scheduling constraint, and the building height distribution information and the collaborative perception virtual boundary as the monitoring scope constraint to perform flight trajectory fitting. This process comprehensively considers the performance parameters of the equipment, the spatial layout within the monitored area, and the collaborative work requirements. Through the path planning algorithm, the first set of monitoring coverage paths is output. These paths will guide the UAV and the quadruped robot to efficiently and safely perform the monitoring task within the monitored area, ensuring full coverage of the entire area while meeting the collaborative work requirements between the equipment.

[0026] Through the above steps, the planning process of the entire inspection path not only considers the actual layout and monitoring requirements of the power station, but also combines factors such as the height of the buildings and the robot configuration to ensure the accuracy and executability of the path planning.

[0027] Furthermore, step P22 of the embodiment of the present application further includes: P22-1: Decompose the overall power station control information to obtain the power station monitoring layout information and the power station airspace restriction information; P22-2: According to the power station monitoring layout information, frame the monitoring scope distribution in the first local BIM model, and inversely deduce and screen the first compensation blind area distribution based on the framing result; P22-3: According to the power station airspace restriction information, screen and frame the collaborative perception virtual boundary in the first compensation blind area distribution. Among them, the power station monitoring layout information includes the deployment data of fixed cameras, temperature and humidity sensors, and gas detectors.

[0028] Optionally, in order to more accurately determine the collaborative perception virtual boundary, further refinement can be carried out. First, decompose the overall control information of the power station to extract the power station monitoring layout information and the airspace restriction information. The power station monitoring layout information includes the deployment data of the static monitoring equipment already deployed in the power plant, such as the deployment positions and quantities of fixed cameras, temperature and humidity sensors, and gas detectors. These devices are used to monitor the environment and equipment conditions in the power station to ensure that key parameter data can be obtained in real time. The airspace restriction information includes the no-fly airspace within the power station area, such as the demarcation of the no-fly area, the flight height limit of the UAV, and the flight range. These information are the restrictive conditions that must be followed in the UAV inspection task.

[0029] Next, based on the data obtained from the power station monitoring layout information, the monitoring range distribution is demarcated in the first local BIM model, that is, the coverage range of the static monitoring devices is mapped into the BIM model to form a visual representation of the monitoring area. The purpose of demarcating the monitoring range is to clarify which areas are within the effective coverage range of the deployed monitoring devices and ensure that these areas can be effectively monitored in real time. According to the results of the monitoring layout, the first compensation blind area distribution is further deduced and screened out, that is, the areas not covered by the existing monitoring devices. These blind areas may have monitoring dead spots and cannot be effectively monitored by the existing devices due to the limitations of the device layout. Therefore, the screening of the compensation blind areas provides an important basis for the subsequent planning of drone and robot tasks.

[0030] Next, based on the power station airspace restriction information, the collaborative perception virtual boundary is screened out and demarcated in the first compensation blind area distribution, that is, the area that belongs to the monitoring blind area and does not belong to the no-fly zone. This virtual boundary not only needs to cover the compensation blind area but also needs to avoid conflicts with the no-fly airspace of the power station. The no-fly zone refers to the area where drones cannot enter, which may be established due to safety, privacy protection, or other restrictions. Therefore, through the airspace restriction information, it can be ensured that the range of the collaborative perception virtual boundary includes both the blind areas that need to be monitored and does not violate the flight restrictions, thus ensuring the safety and legality of the drones when performing tasks.

[0031] Through these steps, the system can, on the basis of making full use of the existing monitoring devices, accurately determine the uncovered blind areas and plan feasible flight paths, making the inspection tasks more comprehensive and efficient, while avoiding interference from flight conflict areas, so as to achieve efficient power station safety monitoring and management.

[0032] Furthermore, step P22-2 of the embodiment of the present application further includes: P22-21: According to the fixed camera deployment data, temperature and humidity sensor deployment data, and gas detector deployment data, perform coverage mapping in the first local BIM model to obtain a multi-modal monitoring range distribution boundary; P22-22: Solve the intersection of the multi-modal monitoring range distribution boundaries and output an effective monitoring range distribution boundary; P22-23: Locate the first monitoring boundary in the first local BIM model; P22-24: Remove the effective monitoring range distribution boundary from the first monitoring boundary to obtain the first compensation blind area distribution.

[0033] Optionally, in order to accurately identify the monitoring blind spots in the power station operation area that have not been covered by static monitoring devices, first, based on the deployment data of fixed cameras, temperature and humidity sensors, and gas detectors, coverage mapping is carried out in the first local BIM model. Specifically, the system constructs the perception volume model of each type of monitoring device in three-dimensional space according to parameters such as the installation coordinates, monitoring angles, effective sensing distances, and coverage radii of various monitoring devices, and projects it onto the first local BIM model, thereby forming a complete set of multi-modal monitoring range distribution boundaries. The boundaries formed by each type of device reflect its real-time monitoring ability range in space and are the prerequisite basis for subsequent compensation blind spot identification.

[0034] After completing the perception range mapping, further solve the intersection of the spatial boundary regions formed by different types of sensors. The purpose of the intersection is to identify the regions jointly covered by multiple sensors. These overlapping regions not only have higher perception stability and data redundancy capabilities but also can be regarded as the spatial regions that currently have sufficient risk monitoring capabilities. The effective monitoring range distribution boundary output by the intersection operation, as the total space already covered by the current static devices, lays a spatial basis for the subsequent dissection of monitoring dead spots.

[0035] On this basis, locate the complete boundary required to be covered by this inspection task in the first local BIM model, that is, the first monitoring boundary. This boundary is usually preset by the task planning system according to the risk level, power station operation plan, and operation and maintenance requirements, representing the global coverage target of the heterogeneous robot team for this operation. Subsequently, the system takes the first monitoring boundary as the universal set and performs a set difference operation to exclude the identified effective monitoring range distribution boundary from it. The finally obtained region is the first compensation blind spot distribution, which reflects the key spatial regions in the task scope that have not been covered by the current static devices.

[0036] These compensation blind spots will be used as the key target areas for subsequent dynamic inspections and real-time perceptions by drones or mobile robots. The entire identification process is based on the high-precision spatial data and device deployment parameters in the BIM model, combined with operations such as geometric modeling, spatial superposition, and calculation of the difference set of layout boundaries, ensuring the accuracy and executability of the extracted blind spots and providing rigorous data support for the scientific determination of the collaborative perception virtual boundary.

[0037] Furthermore, step P24 of the embodiment of the present application further includes: P24-1: Fit the flight trajectory according to the building height distribution information and the collaborative perception virtual boundary, and output P reference flight trajectories; P24-2: Extract the H flight ability matrices of the H UAVs from the heterogeneous team configuration; P24-3: Allocate the P reference flight trajectories according to the H flight ability matrices, and output H groups of reference flight trajectories, where the flight ability matrix includes the flight height limit and the flight endurance characteristics; P24-4: Smoothly connect the H groups of reference flight trajectories, output H monitoring coverage paths, and form the first group of monitoring coverage paths.

[0038] Specifically, to generate the first group of monitoring coverage paths more precisely, first, extract the building height distribution information and the collaborative perception virtual boundary from the first local BIM model. The building height distribution information details the building height changes in the monitoring area, including the highest point, the lowest point, and the changes in intermediate heights of the buildings. The collaborative perception virtual boundary defines the area range that the UAVs need to focus on monitoring, and these areas are usually monitoring blind spots and not no-fly zones. Using this information, generate P reference flight trajectories through a path planning algorithm. The path planning algorithm needs to consider height restrictions to ensure that the UAVs fly above the building heights to avoid collisions; at the same time, it is necessary to ensure that the flight trajectories cover all key areas within the collaborative perception virtual boundary, and optimize the flight paths to reduce unnecessary detours and improve the monitoring efficiency.

[0039] Next, extract the flight ability matrices of the H UAVs from the heterogeneous team configuration. Each flight ability matrix contains key parameters such as the flight height limit, the flight endurance characteristics, the load capacity, and the flight speed. These parameters will be important bases for subsequent task allocation to ensure that each UAV's task matches its performance.

[0040] Then, according to the H flight ability matrices, allocate the P reference flight trajectories. The allocation process needs to comprehensively consider factors such as height matching, endurance matching, load matching, and task priorities. Through a reasonable allocation algorithm, such as the greedy algorithm, the genetic algorithm, or other optimization algorithms, allocate the P reference flight trajectories to the H UAVs to generate H groups of reference flight trajectories. Ensure that the flight trajectories allocated to each UAV are within its flight height limit range, and the flight trajectory length is within the UAV's endurance range, and at the same time, the UAV can carry the sensors and other equipment required to complete the task. Prioritize the allocation of flight tasks in key areas to ensure that high-risk areas are monitored first.

[0041] Finally, smooth connection processing is performed on the reference flight trajectories assigned to each drone. The purpose of smooth connection is to eliminate sudden changes and discontinuities between trajectories, ensure that the drone can make a smooth transition when performing tasks, and avoid flight instability or discontinuous monitoring data caused by sudden trajectory changes. Through mathematical smoothing algorithms, such as Bezier curves, spline curves, etc., the stitched trajectories are smoothed to ensure the smoothness and continuity of the trajectories. At the same time, safety checks are performed on the smoothed trajectories to ensure that they meet flight safety requirements, such as avoiding buildings and obstacles. Finally, H smooth monitoring coverage paths are generated. These paths not only cover all key risk points in the monitoring area but also ensure the safety and stability of the drone during flight. The H monitoring coverage paths together constitute the first group of monitoring coverage paths, providing a scientific and reasonable flight path for subsequent monitoring tasks.

[0042] Through the above steps, through reasonable flight trajectory fitting, extraction and task assignment of the flight ability matrix, and smooth connection of trajectories, the finally obtained first group of monitoring coverage paths can efficiently and safely guide the drone to complete the inspection task in the high-risk operation area of the power station.

[0043] P30: In the heterogeneous team configuration, H drones perform multi-aircraft collaborative scanning of the first monitoring area along the first group of monitoring coverage paths to obtain multi-modal monitoring data.

[0044] Furthermore, step P30 of the embodiment of the present application further includes: P31: The H drones perform multi-aircraft collaborative scanning of the first monitoring area along the H monitoring coverage paths to obtain H groups of multi-modal data sequences. Each group of multi-modal data sequences includes thermodynamic profile data, spatial point cloud sequences, and gas spatio-temporal distribution matrices; P32: According to the acquisition windows of the H groups of multi-modal data sequences, local multi-modal monitoring storage is segmented to obtain multi-modal distribution data; P33: Spatially and temporally align the H groups of multi-modal data sequences and the multi-modal distribution data to obtain the multi-modal monitoring data.

[0045] It should be understood that the H drones in the heterogeneous team configuration perform collaborative scanning along the pre-planned first group of monitoring coverage paths to ensure coverage of the entire first monitoring area and obtain multi-modal monitoring data through collaborative work. Each drone collects relevant multi-modal monitoring data according to its flight path. These data cover multiple dimensions, including thermodynamic profile data, spatial point cloud sequences, and gas spatio-temporal distribution matrices. Among them, the thermodynamic profile data is used to monitor temperature distribution and abnormal heat sources, the spatial point cloud sequences help create a three-dimensional model of the area, and the gas spatio-temporal distribution matrix provides information on the change of gas concentration, especially for the leakage monitoring of harmful gases.

[0046] After obtaining the monitoring data of each drone, local multi-modal monitoring storage is segmented according to the acquisition window of each group of data, that is, the time period and spatial range of data acquisition. The acquisition window refers to the time period during which the drone collects data within a specific time range. By segmenting the monitored data stored locally according to these acquisition windows, it can be ensured that the obtained data is consistent with the data collected by the drone in terms of time, and different types of monitored data can be effectively stored and organized according to their temporal and spatial correlations.

[0047] Next, spatio-temporal alignment is performed on all H groups of multi-modal data sequences and the organized multi-modal distribution data. Spatio-temporal alignment refers to calibrating data collected from different sources and at different times in terms of time and space so that they can be analyzed and processed within a unified spatio-temporal framework. This process involves precisely matching and adjusting the timestamps and spatial coordinates of the data to eliminate data inconsistencies caused by factors such as differences in acquisition devices and deviations in flight trajectories. After spatio-temporal alignment, the obtained multi-modal monitoring data will be highly accurate and consistent, providing reliable data support for subsequent risk coupling analysis and the generation of dynamic risk maps.

[0048] Through these steps, not only is the collaborative scanning of the drone along the monitoring coverage path achieved, obtaining rich multi-modal monitoring data, but also through the segmentation of the data acquisition window and spatio-temporal alignment, the quality and usability of the data are ensured, providing a solid data foundation for the effective implementation of the entire monitoring management method.

[0049] P40: The first edge center receives and performs risk coupling analysis based on the multi-modal monitoring data to generate a dynamic risk map.

[0050] Furthermore, step P40 of the embodiment of the present application further includes: P41: Decompose the multi-modal monitoring data to obtain thermodynamic field data, three-dimensional geometric field data, and gas concentration field data; P42: Pre-define multi-modal risk characteristics, where the multi-modal risk characteristics consist of a temperature anomaly threshold, a structural displacement safety threshold, and a gas concentration anomaly threshold; P43: Use the temperature anomaly threshold and the gas concentration anomaly threshold to map and traverse the thermodynamic field data and the gas concentration field data to obtain a temperature anomaly node distribution and a gas anomaly node distribution; P44: After spatially aligning the three-dimensional geometric field data to the first local BIM model, use the structural displacement safety threshold to judge structural deformation to obtain a structural anomaly node distribution; P45: Perform risk coupling analysis on the temperature anomaly node distribution, the gas anomaly node distribution, and the structural anomaly node distribution to generate the dynamic risk map.

[0051] Optionally, the first edge center receives multimodal monitoring data from multiple drones, conducts risk coupling analysis based on this data, and finally generates a dynamic risk map. This map will provide real-time risk assessment for the high-risk operation areas of the power station, facilitating the timely adoption of response measures.

[0052] When performing this task, first decompose the multimodal monitoring data. For example, based on the different physical properties of the monitoring data, split the complex data set into subsets that are easier to process and analyze, obtaining different types of data, including thermodynamic field data, three-dimensional geometric field data, and gas concentration field data. Thermodynamic field data mainly reflects the temperature changes in the area, three-dimensional geometric field data provides the spatial layout and morphological information of the power station area, while gas concentration field data reflects the gas distribution, especially the concentration of harmful gases.

[0053] Next, according to the monitoring requirements and safety standards, pre-define multimodal risk characteristics. These characteristics include temperature anomaly threshold, structural displacement safety threshold, and gas concentration anomaly threshold. These thresholds are set according to the safety requirements of the power station facilities and are used to determine what degree of change or anomaly will be regarded as a potential safety risk. For example, when the temperature exceeds the preset threshold, it may mean overheating or malfunction of the equipment, abnormal gas concentration may indicate a leakage risk, and structural displacement exceeding the safety limit may predict damage to the building or equipment.

[0054] Furthermore, use the temperature anomaly threshold and gas concentration anomaly threshold to traverse the thermodynamic field data and gas concentration field data, map and identify the temperature anomaly node distribution and gas anomaly node distribution. By comparing with the preset thresholds, determine which areas have exceeded the temperature standard and which areas have gas concentration exceeding the safety standard, thus marking the potential risk areas and providing a basis for further risk assessment.

[0055] At the same time, align the three-dimensional geometric field data spatially to the first local BIM model to ensure that this data matches the first local BIM model of the power station. The aligned data can accurately reflect the actual structure of the power station, facilitating subsequent detailed structural deformation judgment. Using the structural displacement safety threshold, the deformation conditions of the power station equipment and buildings can be checked to obtain the structural anomaly node distribution, and these nodes may be safety hazards caused by structural damage or movement.

[0056] Finally, conduct risk coupling analysis on the temperature anomaly node distribution, gas anomaly node distribution, and structural anomaly node distribution, synthesize different types of risk data, evaluate the mutual relationships between various anomaly nodes, and generate a comprehensive dynamic risk map. This map can reflect the risk status of the power station operation area in real time and be dynamically adjusted according to different risk sources, providing a decision-making basis for power station management personnel to ensure timely and effective safety measures are taken in a high-risk operation environment.

[0057] Through these steps, it is possible to combine various monitoring data, accurately evaluate the risk status of the power station, and achieve real-time monitoring and management of high-risk areas through the generated dynamic risk map, greatly enhancing the security guarantee ability of power station operations.

[0058] Furthermore, step P45 of the embodiment of the present application further includes: P45-1: Predefine the abnormal node fusion distance; P45-2: After spatially aligning the temperature abnormal node distribution, gas abnormal node distribution, and structure abnormal node distribution based on the IPC algorithm, use the abnormal node fusion distance for abnormal node clustering to obtain the coupled risk node distribution; P45-3: Align the coupled risk node distribution to the first local BIM model to locate O power station devices; P45-4: According to the risk conduction characteristics of the O power station devices, construct the risk conduction topology of the O coupled risk nodes in the coupled risk node distribution, and output the dynamic risk map. Among them, the O coupled risk nodes in the coupled risk node distribution have O abnormal feature identifiers, and each abnormal feature identifier includes an abnormal device ID, a temperature abnormal quantity and / or a concentration abnormal quantity and / or a deformation abnormal quantity.

[0059] In a possible embodiment of the present application, the process of risk coupling analysis can be further refined to generate a more accurate and practical dynamic risk map.

[0060] When performing risk coupling analysis, first predefine the abnormal node fusion distance. This fusion distance is a key parameter, which is preset based on the layout of power station devices and the requirements of monitoring accuracy, and is used to measure the spatial relationship between different abnormal nodes, especially the mutual relationship between multiple risk types (such as temperature abnormality, gas abnormality, and structure abnormality). The abnormal node fusion distance is used to determine which nodes are spatially close and which nodes should be regarded as the same group of risk sources. In the risk assessment of a power station, a reasonable fusion distance can help the system aggregate multiple interrelated risk points into a larger risk area, thereby improving the monitoring and response efficiency.

[0061] Next, the IPC algorithm is used for spatial alignment to align the temperature anomaly node distribution, gas anomaly node distribution, and structural anomaly node distribution to ensure that these data can be compared and analyzed in the same spatial coordinate system. The IPC algorithm (Iterative Closest Point) is an algorithm commonly used for point cloud data registration and spatial alignment. It can align data from different sources to a common reference frame through iterative optimization to ensure accurate matching between data. The aligned data is then clustered according to the predefined abnormal node fusion distance, and nodes that are close in space and have similar abnormal characteristics are clustered together to obtain the coupled risk node distribution. These coupled risk nodes represent the intersection of multiple risk factors in the power station, which may be a complex area of excessive temperature, excessive gas concentration, or structural damage.

[0062] Next, the obtained coupled risk node distribution is aligned to the first local BIM model to ensure that the spatial coordinates of all risk nodes are consistent with the actual facility layout of the power station. Through alignment, the system can accurately locate the O power station equipment in the power station and associate these equipment with the corresponding risk nodes to ensure that the status and location of each equipment can be reflected in the risk map.

[0063] Finally, based on the risk transmission characteristics of the O located power station equipment, the risk transmission topology of O coupled risk nodes is constructed. Risk transmission characteristics refer to the possibility and path of risk transmission between equipment. These characteristics are predefined based on information such as the physical connection, functional association, and historical fault data of the equipment. By analyzing the working principle and risk transmission path of the equipment, a topological structure reflecting the risk diffusion is constructed to show the possibility and path of risk transmission from one node to other nodes. This can intuitively show the transmission path and impact range of risks between power station equipment, providing a scientific basis for risk management and decision-making.

[0064] In the output dynamic risk map, the O coupled risk nodes will be marked with corresponding abnormal feature identifiers, each of which includes abnormal equipment ID, abnormal temperature, abnormal gas concentration and / or abnormal deformation. These feature identifiers provide specific abnormal information to help operation and maintenance personnel quickly locate and handle problems. Through this series of steps, the system can not only identify risk hotspots in the power plant, but also clarify the relationship between each risk node and the risk transmission path, providing comprehensive risk monitoring and analysis capabilities for the power plant.

[0065] P50: Construct M detection capability matrices of the M quadruped robots in the heterogeneous team configuration, perform capability-task matching on the dynamic risk map and the M detection capability matrices, and output M inspection task sequences.

[0066] Furthermore, step P50 of the embodiment of the present application also includes: P51: Calculate the similarity between the O abnormal feature identifiers and the M detection ability matrices, and based on the calculation results, perform ability-task matching for the O coupling risk nodes and the M quadruped robots, and output M groups of abnormal detection tasks, where the detection ability matrix includes the device coverage type and the defect coverage type; P52: Serialize the M groups of abnormal detection tasks according to the dynamic risk map, and output the M inspection task sequences.

[0067] Optionally, the task allocation process of the quadruped robots can be further refined to ensure efficient and accurate risk detection. First, for the M quadruped robots in the heterogeneous team configuration, construct their respective detection ability matrices. The detection ability matrix is an important quantitative description of the performance of the quadruped robot, including two key dimensions: the device coverage type and the defect coverage type. The device coverage type refers to the types of power station equipment that the robot can detect, such as transformers, transmission lines, etc.; the defect coverage type involves the types of defects that the robot can identify, such as temperature anomalies, gas leaks, structural deformations, etc. By constructing the detection ability matrix, the advantages and limitations of each quadruped robot in risk detection can be clearly understood.

[0068] Next, calculate the similarity between the O abnormal feature identifiers and the M detection ability matrices. These abnormal feature identifiers include risk nodes that may exist in the power station, such as equipment failures, high temperatures, gas leaks, etc. By calculating the similarity between these identifiers and the detection ability matrix, the system can determine whether each robot is suitable for performing a specific task. Multiple algorithms can be used to calculate the similarity, such as cosine similarity and Euclidean distance based on feature vectors. The specific choice depends on the expression forms of the abnormal feature identifiers and the detection ability matrix. For example, some robots may be more sensitive to detecting high temperatures or harmful gases, while others are more suitable for performing equipment appearance inspections. According to the calculation results, the system will match the most suitable robot for each risk node to ensure the rationality of task allocation. The output M groups of abnormal detection tasks will assign specific tasks to each robot, ensuring that the robot performs tasks within its capabilities while maximizing the overall operation efficiency.

[0069] Finally, serialize the M groups of abnormal detection tasks according to the dynamic risk map, and output the M inspection task sequences. The dynamic risk map not only shows the distribution of coupling risk nodes but also contains information about the risk conduction relationship and priority. Through the serialization process, the abnormal detection tasks are arranged in a certain order to form an inspection task sequence. The basis for serialization can be factors such as the severity of the risk, the urgency of the task, and the location of the robot, to ensure that the quadruped robot can efficiently cover all key risk points when performing tasks.

[0070] Through the above steps, the system can achieve efficient inspection task allocation based on the matching of robot capabilities and tasks, ensuring that each quadruped robot can work in its most proficient field, maximizing the utilization rate of resources and improving the overall inspection efficiency.

[0071] P60: Perform adjacent zoning inspection compensation optimization on the M inspection task sequences, output M compensated task sequences, and conduct risk collaborative investigation of the dynamic risk map.

[0072] Specifically, after the preliminary planning of the M inspection task sequences, it is necessary to perform adjacent zoning inspection compensation optimization on these task sequences. The purpose of this optimization process is to solve the problems of task duplication or omission that may occur between adjacent monitoring zones, ensuring that the quadruped robots can seamlessly connect when performing tasks and avoiding blind spots in risk investigation caused by zone boundaries.

[0073] First, analyze the task distribution in the M inspection task sequences to identify the task boundaries between adjacent monitoring zones. Since there may be some overlap or gap in the division of monitoring zones, special attention needs to be paid to these boundary areas. By analyzing the position information and task content in the task sequences, determine which tasks are located at the junction of adjacent zones.

[0074] Next, perform compensation optimization on the tasks located at the junction of adjacent zones. This process involves adjusting the task order and task content in the task sequences to ensure that the quadruped robots can cover all key risk points when performing tasks while avoiding duplicate execution of the same task. Specific methods for compensation optimization can include task splitting, task merging, task reallocation, etc. For example, if a certain task is involved in both of two adjacent zones, it can be split into two subtasks and executed by the quadruped robots in the two zones respectively; if there are duplicate tasks in two adjacent zones, they can be merged into one task and completed by one quadruped robot.

[0075] After completing the compensation optimization, output M compensated task sequences. These compensated task sequences are optimized and can better adapt to the task requirements between adjacent monitoring zones, ensuring that the quadruped robots can efficiently and accurately complete risk investigation when performing tasks. The compensated task sequences not only consider the coverage of tasks but also optimize the task execution order and resource allocation, improving the overall efficiency of task execution.

[0076] Finally, according to the M compensation task sequences, a risk collaborative investigation of the dynamic risk map is carried out. The dynamic risk map provides the real-time distribution and changes of risks within the monitoring area. When the quadruped robot executes tasks according to the compensation task sequences, it can conduct collaborative investigations based on the risk information in the map. During the collaborative investigation process, the quadruped robots can share monitoring data in real time, discover and handle newly emerging risks in a timely manner, and ensure that the risks within the monitoring area are comprehensively and promptly controlled.

[0077] In summary, the embodiments of the present application at least have the following technical effects: In the present application, the first local BIM model is segmented from the power station BIM model and aligned with the monitoring area division. Combined with the heterogeneous team configuration, the inspection coverage path is fitted, and drones are assigned for multi-robot collaborative scanning to obtain multi-modal monitoring data. The first edge center performs risk coupling analysis based on these data to generate a dynamic risk map. Subsequently, a detection ability matrix of the quadruped robot is constructed, the ability-task matching is carried out, and the inspection task sequence is output. Through the optimization of the adjacent area inspection compensation, the compensation task sequence is finally output to complete the risk collaborative investigation of the dynamic risk map, ensuring efficient and comprehensive power station inspection and risk assessment.

[0078] It achieves the technical effects of comprehensively covering high-risk areas through the collaborative operation of heterogeneous robots, improving the inspection efficiency, and realizing dynamic risk assessment and timely response.

[0079] Embodiment 2, based on the same inventive concept as the robot team-based collaborative management method for high-risk operations in power stations in the foregoing embodiment, as Figure 2 shown, the present application provides a robot team-based collaborative management system for high-risk operations in power stations. The system in the embodiments of the present application and the method embodiments are based on the same inventive concept. Among them, the system includes: A model segmentation module 11, configured to segment the first local BIM model from the power station BIM model, where the first local BIM model is horizontally aligned with the first monitoring area division.

[0080] An inspection coverage path fitting module 12, configured to perform inspection coverage path fitting according to the heterogeneous team configuration of the first monitoring area division and the first local BIM model to obtain the first set of monitoring coverage paths.

[0081] A multi-robot collaborative scanning module 13, configured to perform multi-robot collaborative scanning of the first monitoring area division along the first set of monitoring coverage paths by H drones in the heterogeneous team configuration to obtain multi-modal monitoring data.

[0082] A risk coupling analysis module 14, configured to perform risk coupling analysis based on the multi-modal monitoring data received by the first edge center and generate a dynamic risk map.

[0083] The task matching module 15 is configured to construct M detection capability matrices for M quadruped robots in the heterogeneous workgroup configuration, perform capability-task matching on the dynamic risk map and the M detection capability matrices, and output M inspection task sequences.

[0084] The risk collaborative investigation module 16 is configured to perform adjacent area inspection compensation optimization on the M inspection task sequences, output M compensated task sequences, and perform risk collaborative investigation on the dynamic risk map.

[0085] Furthermore, the inspection coverage path fitting module 12 is further configured to perform the following steps: Locally call the power station global control information; perform collaborative perception area positioning in the first local BIM model based on the power station global control information to obtain a collaborative perception virtual boundary; extract the building height distribution information of the collaborative perception virtual boundary from the first local BIM model; use the heterogeneous workgroup configuration as an equipment scheduling constraint, and use the building height distribution information and the collaborative perception virtual boundary as monitoring range constraints to perform flight trajectory fitting, and output the first set of monitoring coverage paths.

[0086] Furthermore, the inspection coverage path fitting module 12 is further configured to perform the following steps: By decomposing the power station global control information, obtain power station monitoring layout information and power station airspace restriction information; frame the monitoring range distribution in the first local BIM model according to the power station monitoring layout information, and inversely deduce and screen the first compensation blind area distribution based on the framed result; frame and screen the collaborative perception virtual boundary in the first compensation blind area distribution according to the power station airspace restriction information. The power station monitoring layout information includes fixed camera deployment data, temperature and humidity sensor deployment data, and gas detector deployment data.

[0087] Furthermore, the inspection coverage path fitting module 12 is further configured to perform the following steps: Perform flight trajectory fitting based on the building height distribution information and the collaborative perception virtual boundary, and output P reference flight trajectories; extract the H flight ability matrices of the H UAVs from the heterogeneous team configuration; allocate the P reference flight trajectories according to the H flight ability matrices, and output H groups of reference flight trajectories, where the flight ability matrix includes the flight height limit and the flight endurance characteristics; smoothly connect the H groups of reference flight trajectories, output H monitoring coverage paths, and form the first group of monitoring coverage paths. Perform flight trajectory fitting based on the building height distribution information and the collaborative perception virtual boundary, and output P reference flight trajectories; extract the H flight ability matrices of the H UAVs from the heterogeneous team configuration; allocate the P reference flight trajectories according to the H flight ability matrices, and output H groups of reference flight trajectories, where the flight ability matrix includes the flight height limit and the flight endurance characteristics; smoothly connect the H groups of reference flight trajectories, output H monitoring coverage paths, and form the first group of monitoring coverage paths.

[0088] Further, the inspection coverage path fitting module 12 is further configured to perform the following steps: According to the fixed camera deployment data, the temperature and humidity sensor deployment data, and the gas detector deployment data, perform coverage range mapping on the first local BIM model to obtain a multi-modal monitoring range distribution boundary; solve the intersection of the multi-modal monitoring range distribution boundaries, and output an effective monitoring range distribution boundary; locate the first monitoring boundary in the first local BIM model; remove the effective monitoring range distribution boundary from the first monitoring boundary to obtain the first compensation blind area distribution.

[0089] Further, the multi-UAV collaborative scanning module 13 is further configured to perform the following steps: The H UAVs perform multi-UAV collaborative scanning of the first monitoring area along the H monitoring coverage paths to obtain H groups of multi-modal data sequences, where each group of multi-modal data sequences includes thermodynamic profile data, a spatial point cloud sequence, and a gas spatio-temporal distribution matrix; according to the acquisition windows of the H groups of multi-modal data sequences, segment the local multi-modal monitoring storage to obtain multi-modal distribution data; perform spatio-temporal alignment on the H groups of multi-modal data sequences and the multi-modal distribution data to obtain the multi-modal monitoring data.

[0090] Further, the risk coupling analysis module 14 is further configured to perform the following steps: Decompose the multi-modal monitoring data to obtain thermodynamic field data, three-dimensional geometric field data, and gas concentration field data; predefined multi-modal risk characteristics, where the multi-modal risk characteristics consist of a temperature anomaly threshold, a structural displacement safety threshold, and a gas concentration anomaly threshold; use the temperature anomaly threshold and the gas concentration anomaly threshold to map and traverse the thermodynamic field data and the gas concentration field data to obtain a temperature anomaly node distribution and a gas anomaly node distribution; after spatially aligning the three-dimensional geometric field data to the first local BIM model, use the structural displacement safety threshold to perform structural deformation judgment to obtain a structural anomaly node distribution; perform risk coupling analysis on the temperature anomaly node distribution, the gas anomaly node distribution, and the structural anomaly node distribution to generate the dynamic risk map.

[0091] Further, the risk coupling analysis module 14 is further configured to perform the following steps: Predefine an abnormal node fusion distance; after spatially aligning the temperature anomaly node distribution, the gas anomaly node distribution, and the structural anomaly node distribution based on the IPC algorithm, use the abnormal node fusion distance to perform abnormal node clustering to obtain a coupled risk node distribution; align the coupled risk node distribution to the first local BIM model to locate O power station devices; according to the risk conduction characteristics of the O power station devices, construct a risk conduction topology of the O coupled risk nodes in the coupled risk node distribution, and output the dynamic risk map. Wherein, the O coupled risk nodes in the coupled risk node distribution have O abnormal feature identifiers, and each abnormal feature identifier includes an abnormal device ID, a temperature anomaly amount and / or a concentration anomaly amount and / or a deformation anomaly amount.

[0092] Further, the task matching module 15 is further configured to perform the following steps: Calculate the similarity between the O abnormal feature identifiers and the M detection ability matrices, and based on the calculation results, perform ability-task matching on the O coupled risk nodes and the M quadruped robots, and output M groups of abnormal detection tasks, where the detection ability matrix includes a device coverage type and a defect coverage type; serialize the M groups of abnormal detection tasks according to the dynamic risk map, and output the M inspection task sequences.

[0093] It should be noted that the above sequence of embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. And the above describes specific embodiments of this specification. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or may be advantageous.

[0094] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.

[0095] This specification and the drawings are merely exemplary descriptions of the present application and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. A robot team-based collaborative management method for high-risk operations in power plants, characterized in that The method includes: Segmenting a first local BIM model from the power station BIM model, where the first local BIM model is horizontally aligned with the first monitoring area; Performing inspection coverage path fitting based on the heterogeneous team configuration of the first monitoring area and the first local BIM model to obtain a first set of monitoring coverage paths; In the heterogeneous team configuration, H unmanned aerial vehicles perform multi-aircraft collaborative scanning of the first monitoring area along the first set of monitoring coverage paths to obtain multi-modal monitoring data; The first edge center receives and performs risk coupling analysis based on the multi-modal monitoring data to generate a dynamic risk map; Constructing M detection capability matrices for the M quadruped robots in the heterogeneous team configuration, and performing capability-task matching on the dynamic risk map and the M detection capability matrices to output M inspection task sequences; Performing adjacent area inspection compensation optimization on the M inspection task sequences, outputting M compensation task sequences, and performing risk collaborative investigation of the dynamic risk map.

2. The robot team-based collaborative management method for high-risk operations in power plants according to claim 1, wherein, Performing inspection coverage path fitting based on the heterogeneous team configuration of the first monitoring area and the first local BIM model to obtain a first set of monitoring coverage paths, and the method includes: Locally invoking the power station global control information; Performing collaborative perception area positioning in the first local BIM model based on the power station global control information to obtain a collaborative perception virtual boundary; Extracting the building height distribution information of the collaborative perception virtual boundary from the first local BIM model; Using the heterogeneous team configuration as an equipment scheduling constraint, and using the building height distribution information and the collaborative perception virtual boundary as monitoring range constraints to perform flight trajectory fitting, and outputting the first set of monitoring coverage paths.

3. The robot team-based collaborative management method for high-risk operations in power plants according to claim 2, wherein Performing collaborative perception area positioning in the first local BIM model based on the power station global control information to obtain a collaborative perception virtual boundary, and the method includes: Decomposing the power station global control information to obtain power station monitoring layout information and power station airspace restriction information; Framing the monitoring range distribution in the first local BIM model according to the power station monitoring layout information, and inversely inferring and screening the first compensation blind area distribution based on the framing result; Screening and framing the collaborative perception virtual boundary in the first compensation blind area distribution according to the power station airspace restriction information.

4. The robot team-based collaborative management method for high-risk operations in power plants according to claim 2, characterized in that Using the heterogeneous team configuration as an equipment scheduling constraint, and using the building height distribution information and the collaborative perception virtual boundary as monitoring range constraints to perform flight trajectory fitting, and outputting the first set of monitoring coverage paths, and the method includes: Performing flight trajectory fitting based on the building height distribution information and the collaborative perception virtual boundary to output P reference flight trajectories; Extracting H flight capability matrices of the H unmanned aerial vehicles from the heterogeneous team configuration; Allocating the P reference flight trajectories according to the H flight capability matrices to output H groups of reference flight trajectories, where the flight capability matrix includes flight height limit and flight endurance characteristics; Smoothing and connecting the H groups of reference flight trajectories to output H monitoring coverage paths, which constitute the first set of monitoring coverage paths.

5. The robot team-based collaborative management method for high-risk operations in power plants according to claim 4, characterized in that, In the heterogeneous team configuration, H drones perform multi-aircraft collaborative scanning of the first monitoring area along the first group of monitoring coverage paths to obtain multimodal monitoring data. The method includes: The H drones perform multi-aircraft collaborative scanning of the first monitoring area along the H monitoring coverage paths to obtain H groups of multimodal data sequences. Each group of multimodal data sequences includes thermodynamic profile data, spatial point cloud sequences, and gas spatio-temporal distribution matrices. According to the acquisition windows of the H groups of multimodal data sequences, local multimodal monitoring storage is segmented to obtain multimodal distribution data. The H groups of multimodal data sequences and the multimodal distribution data are spatio-temporally aligned to obtain the multimodal monitoring data.

6. The robot team-based collaborative management method for high-risk operations in power plants according to claim 2, wherein, The first edge center receives and performs risk coupling analysis based on the multimodal monitoring data to generate a dynamic risk map. The method includes: The multimodal monitoring data is decomposed to obtain thermodynamic field data, three-dimensional geometric field data, and gas concentration field data. Multimodal risk characteristics are predefined. The multimodal risk characteristics are composed of a temperature anomaly threshold, a structural displacement safety threshold, and a gas concentration anomaly threshold. The temperature anomaly threshold and the gas concentration anomaly threshold are used to map and traverse the thermodynamic field data and the gas concentration field data to obtain a temperature anomaly node distribution and a gas anomaly node distribution. After spatially aligning the three-dimensional geometric field data to the first local BIM model, the structural displacement safety threshold is used to judge structural deformation to obtain a structural anomaly node distribution. Risk coupling analysis is performed on the temperature anomaly node distribution, the gas anomaly node distribution, and the structural anomaly node distribution to generate the dynamic risk map.

7. The robot team-based collaborative management method for high-risk operations in power plants according to claim 6, characterized in that Risk coupling analysis is performed on the temperature anomaly node distribution, the gas anomaly node distribution, and the structural anomaly node distribution to generate the dynamic risk map. The method includes: An abnormal node fusion distance is predefined. After spatially aligning the temperature anomaly node distribution, the gas anomaly node distribution, and the structural anomaly node distribution based on the IPC algorithm, the abnormal node fusion distance is used for abnormal node clustering to obtain a coupled risk node distribution. The coupled risk node distribution is aligned to the first local BIM model to locate O power station devices. According to the risk conduction characteristics of the O power station devices, a risk conduction topology of the O coupled risk nodes in the coupled risk node distribution is constructed, and the dynamic risk map is output.

8. The robot team-based collaborative management method for high-risk operations in power plants according to claim 7, characterized in that The O coupled risk nodes in the coupled risk node distribution have O abnormal feature identifiers. Each abnormal feature identifier includes an abnormal device ID, a temperature anomaly amount, and / or a concentration anomaly amount, and / or a deformation anomaly amount.

9. The robot team-based collaborative management method for high-risk operations in power plants according to claim 8, wherein, An M detection ability matrix of M quadruped robots in the heterogeneous team configuration is constructed, and an ability-task matching is performed on the dynamic risk map and the M detection ability matrices to output M inspection task sequences. The method includes: The similarity between the O abnormal feature identifiers and the M detection ability matrices is calculated, and based on the calculation results, an ability-task matching is performed on the O coupled risk nodes and the M quadruped robots to output M groups of abnormal detection tasks. The detection ability matrix includes a device coverage type and a defect coverage type. Serialize the M groups of anomaly detection tasks according to the dynamic risk map, and output the M inspection task sequences.

10. The robot team-based collaborative management method for high-risk operations in power plants according to claim 3, wherein The power station monitoring layout information includes fixed camera deployment data, temperature and humidity sensor deployment data, and gas detector deployment data.

11. The robot team-based collaborative management method for high-risk power plant operations according to claim 10, characterized in that, According to the power station monitoring layout information, frame the monitoring range distribution in the first local BIM model, and inversely deduce and screen the first compensation blind area distribution based on the framing result. The method includes: According to the fixed camera deployment data, temperature and humidity sensor deployment data, and gas detector deployment data, perform coverage range mapping in the first local BIM model to obtain the multi-modal monitoring range distribution boundary; Solve the intersection of the multi-modal monitoring range distribution boundaries and output the effective monitoring range distribution boundary; Locate the first monitoring boundary in the first local BIM model; Remove the effective monitoring range distribution boundary from the first monitoring boundary to obtain the first compensation blind area distribution.

12. The robot team-based collaborative management system for high-risk operations in power plants is characterized in that, The system includes: A model segmentation module for segmenting the first local BIM model from the power station BIM model, where the first local BIM model is horizontally aligned with the first monitoring area; An inspection coverage path fitting module for performing inspection coverage path fitting according to the heterogeneous team configuration of the first monitoring area and the first local BIM model to obtain the first group of monitoring coverage paths; A multi-aircraft collaborative scanning module for performing multi-aircraft collaborative scanning of the first monitoring area along the first group of monitoring coverage paths by H unmanned aerial vehicles in the heterogeneous team configuration to obtain multi-modal monitoring data; A risk coupling analysis module for performing risk coupling analysis according to the multi-modal monitoring data received by the first edge center and generating a dynamic risk map; A task matching module for constructing M detection ability matrices of M quadruped robots in the heterogeneous team configuration, and performing ability-task matching on the dynamic risk map and the M detection ability matrices to output M inspection task sequences; A risk collaborative investigation module for optimizing the adjacent area inspection compensation of the M inspection task sequences, outputting M compensation task sequences, and performing risk collaborative investigation of the dynamic risk map.

Images