AR (Augmented Reality) inspection method, system and equipment applied to electric power facilities and medium

Abstract

The invention relates to the technical field of augmented reality, and discloses an AR inspection method, system and device applied to an electric power facility and a medium, and the method comprises the following steps: obtaining facility internal temperature data, radar scanning data and environment data of the electric power facility; constructing an equipment thermodynamic diagram according to the temperature data in the facility, and constructing an electrified equipment distribution diagram according to the radar scanning data; generating a dynamic safety model of the electric power facility based on the equipment thermodynamic diagram, the electrified equipment distribution diagram and the environment data; according to the dynamic safety model, planning an inspection strategy of the electric power facility, and sending the inspection strategy to the target AR glasses; and displaying the dynamic safety model and the inspection strategy to the target inspection personnel through the target AR glasses. According to the invention, by integrating temperature, radar scanning and vibration data, a dynamic safety model covering the heat distribution, space structure and operation state of the electric power facility is constructed, and the problems of incomplete environmental perception and lagging risk identification in traditional manual inspection are solved.

Description

Technical Field

[0001] This invention relates to the field of augmented reality technology, and in particular to an AR inspection method, system, equipment and medium for power facilities. Background Technology

[0002] In industrial settings such as power facilities (e.g., substations, transmission lines), inspection is a crucial step in ensuring stable equipment operation and preventing potential safety hazards.

[0003] Currently, power facility inspections mainly rely on traditional manual inspection methods. Manual inspections depend on inspectors carrying equipment such as infrared thermometers, and are completed through visual observation and manual recording. This method suffers from incomplete environmental perception and the risk of missed inspections. Summary of the Invention

[0004] In view of this, this application provides an AR inspection method, system, equipment and medium for power facilities to solve the above problems.

[0005] Firstly, an AR inspection method for power facilities is provided, which is applied to an intelligent inspection system. This method includes: Acquire temperature data within power facilities; Construct a thermal map of the equipment based on the temperature data within the facility; Acquire radar scan data of power facilities; Based on radar scan data, a distribution map of charged equipment is constructed; Acquire environmental data of power facilities, including equipment mechanical vibration data; Based on equipment heat maps, distribution maps of energized equipment, and environmental data, a dynamic safety model for power facilities is generated. Based on the dynamic safety model, an inspection strategy for power facilities is planned, and the inspection strategy is sent to the target AR glasses; Dynamic security models and inspection strategies are displayed to the target inspection personnel through AR glasses.

[0006] The above technical solutions integrate temperature, radar scanning, and vibration data to construct a dynamic safety model covering the thermal distribution, spatial structure, and operational status of power facilities. This addresses the problems of incomplete environmental perception and delayed risk identification in traditional manual inspections. Combined with AR glasses, the visual delivery of inspection strategies enhances the real-time response capabilities of inspection personnel in complex environments, reduces the risk of missed inspections due to subjective judgment and limited visibility, and significantly improves the intelligence and safety of power facility inspections.

[0007] Optionally, based on the temperature data within the facility, constructing a thermal map of the equipment includes: Acquire two-dimensional thermal imaging data of the surface temperature of equipment in a preset area of ​​a power facility; Acquire discrete temperature point data for the target region, which includes the corresponding actual spatial coordinates. The discrete temperature point data is converted into continuous temperature distribution data using an inverse distance weighted interpolation algorithm; The target device temperature data is obtained by spatiotemporal registration of continuous temperature distribution data and two-dimensional thermal imaging data. A thermal map of the equipment is obtained by mapping the temperature data of the target equipment to spatial grid cells generated based on the space where the power facility is located through 3D raster modeling.

[0008] The above technical solution utilizes the spatiotemporal registration of two-dimensional thermal imaging data with discrete temperature point clouds, combined with an inverse distance weighted interpolation algorithm, to achieve accurate reconstruction from local single-point temperature measurement to a continuous temperature field across the entire domain. By mapping temperature data to a spatial grid through three-dimensional raster modeling, a thermal map of the equipment with a three-dimensional visualization effect is generated. This map can intuitively present the temperature gradient distribution of power equipment (such as transformers and switchgear), assisting inspection personnel in quickly locating overheating defects and avoiding equipment failures caused by blind spots in temperature monitoring.

[0009] Optionally, based on radar scan data, constructing a distribution map of energized equipment includes: Determine the point cloud data for each charged device based on radar scan data; The iterative nearest point algorithm is used to stitch together multiple point cloud data to generate a point cloud map of power facilities; Equipment feature point cloud is extracted from the point cloud map of power facilities based on height segmentation; The device feature point cloud is used to generate a safety bounding box for the energized equipment using a bounding box generation algorithm. The distribution map of energized equipment is obtained based on the safety boundary box.

[0010] The above technical solution, based on the stitching and height segmentation algorithm of LiDAR point cloud data, accurately extracts the spatial features of live equipment and generates a distribution map of live equipment including safety boundary boxes. This map can clearly identify the physical boundaries and spatial layout of high-voltage equipment, effectively distinguish between static equipment and dynamic obstacles (such as workers), provide a reliable spatial benchmark for inspection path planning, and improve the safety and path planning efficiency of the inspection process.

[0011] Optionally, based on equipment heat maps, distribution maps of energized equipment, and environmental data, a dynamic safety model for power facilities can be generated, including: The equipment heat map, the distribution map of energized equipment and environmental data are denoised, time-aligned and normalized to obtain fused data; Based on the fused data, the data of live equipment is overlaid with the equipment heat map to generate a spatial risk feature map; Based on the correlation between the abnormal vibration location in the equipment's mechanical vibration data and the temperature gradient in the equipment's thermal diagram, the location of high-risk power equipment is marked. A dynamic safety model is constructed by integrating spatial risk feature maps and the locations of high-risk power equipment.

[0012] The above technical solutions eliminate sensor errors and timing deviations through denoising, alignment, and normalization of multi-source data, ensuring the reliability of data fusion. A spatial risk feature map is generated by overlaying the distribution of energized equipment with a heat map, and high-risk equipment is marked using vibration anomalies and temperature gradients, enabling coupled analysis of multi-dimensional risks in power facilities across thermal, vibration, and structural dimensions. The dynamic safety model can reflect the health status of equipment and the distribution of environmental risks in real time, providing core data support for the intelligent generation of inspection strategies.

[0013] Optional, planned inspection strategies for power facilities include: Based on the workload, current location, and power operation qualification level of all inspection personnel to be assigned, the target inspection personnel are determined. High-risk areas are identified based on equipment temperature data, equipment vibration data, and the distribution of energized equipment in the dynamic safety model. Generate inspection routes based on high-risk areas and target inspection personnel; The inspection strategy is generated by combining the inspection path with the target inspection personnel.

[0014] The above technical solution allocates tasks based on the workload, location, and qualification level of inspection personnel. It generates inspection paths by combining real-time risk data from a dynamic safety model, achieving a precise match between "personnel capability, risk level, and task requirements." This avoids assigning high-risk tasks to unqualified personnel and reduces unnecessary movement through path optimization, improving the efficiency and safety of inspection tasks and promoting the standardization and intelligentization of inspection management.

[0015] Optionally, displaying dynamic security models and inspection strategies to target inspection personnel via target AR glasses includes: The dynamic security model is converted into a preset format and AR markers are generated in conjunction with the inspection strategy. Push AR markers and inspection paths to the target AR glasses; A dynamic security model is overlaid and displayed in the field of view of the target AR glasses; Equipment in high-risk areas is marked in 3D, displaying the equipment name, current status, and safe distance; Generate visual navigation arrows within the target AR glasses' field of view; When the target inspection personnel approach the live equipment or high-risk area, a distance warning is triggered.

[0016] The above technical solution integrates complex 3D safety models and inspection paths into AR glasses through lightweight model conversion and AR marker generation, enabling real-time overlay of virtual information with the real-world environment. 3D markers and navigation arrows for high-risk equipment guide inspectors to quickly locate targets, while a distance warning mechanism prevents accidental entry into dangerous areas through real-time location monitoring. This interactive method reduces the cognitive load on inspectors and improves decision-making speed and operational safety in complex scenarios.

[0017] Optionally, the method further includes: The inspection strategy is broken down into multiple inspection tasks; The initial inspection task is issued to the target AR glasses based on the complexity of the inspection task. Acquire the working status data of the target AR glasses, including battery level, operating status, and usage time; The revised inspection task is obtained by modifying the initial inspection task based on the working status data. Update the inspection task to the target AR glasses.

[0018] The above technical solution achieves refined management of inspection work by breaking down the inspection strategy into multiple inspection tasks; it issues initial inspection tasks based on task complexity to ensure the rationality of task allocation; it acquires real-time working status data such as battery level, operating status, and usage time of the AR glasses to provide a basis for dynamic task adjustment; and it corrects and updates the initial tasks to the AR glasses based on the working status data, which can dynamically adapt to changes in device status, avoid task interruption due to insufficient device power or abnormal operation, improve the reliability and efficiency of inspection task execution, ensure the smooth progress of inspection work, and optimize resource allocation.

[0019] A second aspect of this application provides an AR inspection system for power facilities, including an acquisition module and a processing module, wherein: The acquisition module is configured to acquire temperature data within power facilities. The processing module is configured to build equipment thermal maps based on temperature data within the facility. The acquisition module is also configured to acquire radar scan data of power facilities; The processing module is also configured to construct a distribution map of charged equipment based on radar scan data; The acquisition module is also configured to acquire environmental data of power facilities, including equipment mechanical vibration data; The processing module is also configured to generate a dynamic safety model of power facilities based on equipment heat maps, distribution maps of energized equipment, and environmental data; The processing module is also configured to plan inspection strategies for power facilities based on a dynamic security model and send the inspection strategies to the target AR glasses. The processing module is also configured to display dynamic security models and inspection strategies to the target inspection personnel through the target AR glasses.

[0020] A third aspect of this application provides an electronic device including a processor, a memory, a user interface, and a network interface, wherein the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any of the foregoing.

[0021] A fourth aspect of this application provides a computer-readable storage medium storing instructions that, when executed, perform the method described in any of the preceding descriptions.

[0022] In summary, one or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: By integrating temperature, radar scan, and vibration data, a dynamic safety model covering the thermal distribution, spatial structure, and operational status of power facilities is constructed, addressing the issues of incomplete environmental perception and delayed risk identification in traditional manual inspections. Combined with AR glasses, inspection strategies are visually pushed out, enhancing inspectors' real-time response capabilities to complex environments, reducing the risk of missed inspections due to subjective judgment and limited visibility, and significantly improving the intelligence and safety of power facility inspections. Attached Figure Description

[0023] Figure 1 This is an exemplary system architecture diagram of an AR inspection method or an AR inspection system for power facilities that applies the present application. Figure 2 This is a flowchart illustrating an AR inspection method for power facilities as described in an embodiment of this application. Figure 3 This is another flowchart illustrating an AR inspection method for power facilities as described in this application embodiment; Figure 4 This is a schematic diagram of a module of an AR inspection system for power facilities in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of an electronic device disclosed in the application embodiment.

[0024] Explanation of reference numerals in the attached diagram: 100, System architecture; 101, First terminal device; 102, Second terminal device; 103, Third terminal device; 104, Network; 105, Server; 401, Acquisition module; 402, Processing module; 403, Inspection task allocation module; 501, Processor; 502, Communication bus; 503, User interface; 504, Network interface; 505, Memory. Detailed Implementation

[0025] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0026] In the description of the embodiments of this application, the words "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design that is described as "for example" or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design options. Rather, the use of the words "for example" or "for instance" is intended to present the relevant concepts in a specific manner.

[0027] In the description of the embodiments of this application, the term "multiple" means two or more. For example, multiple systems means two or more systems, and multiple screen terminals means two or more screen terminals. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0028] This embodiment discloses an AR inspection method, system, equipment, and medium applied to power facilities. Figure 1 An exemplary system architecture diagram is shown, illustrating an embodiment of an AR inspection method or an AR inspection system for power facilities to which this application can be applied.

[0029] like Figure 1 As shown, the system architecture 100 may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is used as a medium to provide communication links between the terminal devices 101, 102, 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links or fiber optic cables, etc.

[0030] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, and 103, such as model training applications, video recognition applications, web browser applications, social platform software, etc.

[0031] Terminal devices 101, 102, and 103 can be either hardware or software. When terminal devices 101, 102, and 103 are hardware, they can be various electronic devices with displays, including but not limited to AR (Augmented Reality) glasses, smartphones, tablets, e-book readers, MP3 (Moving Picture Experts Group Audio Layer III) players, MP4 (Moving Picture Experts Group Audio Layer IV) players, laptops, and desktop computers, etc. When terminal devices 101, 102, and 103 are software, they can be installed in the electronic devices listed above. They can be implemented as multiple software programs or software modules (e.g., multiple software programs or software modules used to provide distributed services) or as a single software program or software module. No specific limitations are made here.

[0032] When terminals 101, 102, and 103 are hardware devices, video capture devices can also be installed on them. These video capture devices can be various devices capable of capturing video, such as cameras, sensors, etc. Users can use the video capture devices on terminals 101, 102, and 103 to capture video.

[0033] Server 105 can be a server that provides various services, such as an intelligent inspection system that processes data displayed on terminal devices 101, 102, and 103. The intelligent inspection system can analyze and process the received data and feed back the processing results (such as identification results) to the terminal devices.

[0034] It should be noted that a server can be either hardware or software. When the server is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. When the server is software, it can be implemented as multiple software programs or software modules (e.g., multiple software programs or software modules used to provide distributed services), or as a single software program or software module. No specific limitations are made here.

[0035] It should be understood that Figure 1The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included. In particular, if the target data does not need to be obtained remotely, the above system architecture may exclude the network and include only terminal devices or servers.

[0036] Figure 2 This is a flowchart illustrating an AR inspection method for power facilities disclosed in an embodiment of this application, as shown below. Figure 2 As shown, this embodiment includes: Step S201: Obtain the temperature data inside the power facility.

[0037] For example, a multi-channel temperature sensor network deployed at key locations within power facilities collects internal temperature data. This temperature sensor network consists of several infrared sensors and thermocouple sensors. The infrared sensors are installed in hard-to-reach locations such as high-voltage switchgear and transformer casings to achieve long-distance, non-contact temperature measurement. Thermocouple sensors are placed on easily accessible surfaces of the facility to calibrate the infrared temperature measurement data. Each sensor reports the collected temperature value, sensor ID, sampling timestamp, and other information to an edge gateway device via Modbus or an industrial Ethernet port. The edge gateway performs preliminary filtering (such as applying a low-pass filter to remove transient noise) and calibration on the raw data, and transmits the processed data to an intelligent inspection system via Wi-Fi, 5G wireless networks, or a local fiber optic network to form a time-seriesed dataset of internal facility temperatures.

[0038] Step S202: Construct a thermal map of the equipment based on the temperature data within the facility.

[0039] For example, the acquired temperature data undergoes preprocessing, including: removing sensor fault points (such as temperature abrupt changes exceeding a threshold), unifying the unit (Celsius) for data from different types of sensors, and completing missing data through linear interpolation. Two-dimensional thermal imaging data of the equipment surface temperature in a predetermined area of ​​the power facility (such as distribution cabinet area, transformer area, busbar area) is acquired. This thermal imaging data is collected by a fixed or handheld infrared thermal imager, capturing the overall temperature distribution map of the equipment in this area at time T. Simultaneously, discrete temperature acquisition units (such as small infrared temperature measurement modules) are deployed within these predetermined areas to acquire discrete temperature point data, which includes the spatial coordinates (x, y, z) and the actual measured temperature value of each measurement point. Subsequently, an IDW (Inverse Distance Weighting) interpolation algorithm is used to convert the discrete temperature point data into continuous three-dimensional spatial temperature distribution data. Through this interpolation process, a continuous temperature distribution grid covering the target area is obtained. Spatiotemporal registration is performed between continuous temperature distribution data and 2D thermal imaging data: Pre-calibration using hand-eye alignment is conducted to obtain the transformation matrix between the thermal imager coordinate system and the global equipment coordinate system. Then, using the PnP (Perspective-n-Point) algorithm, the 2D thermal imaging pixel coordinates are mapped to 3D spatial coordinates via a calibration board or equipment geometric feature points, completing the spatiotemporal registration and obtaining the temperature point cloud of the target equipment in 3D space. A regular 3D grid (e.g., each grid cell with a side length of 0.1m) is constructed based on the space where the power facilities are located. The aforementioned temperature point cloud data is mapped one by one to the corresponding grid cells. The average or maximum temperature value within each grid cell is calculated to generate a thermal map of the equipment for subsequent risk analysis and AR visualization.

[0040] In one possible implementation, a thermal map of the equipment is constructed based on temperature data within the facility. Specifically, this includes: acquiring two-dimensional thermal imaging data of the surface temperature of equipment in a preset area of ​​the power facility; acquiring discrete temperature point data of the target area, wherein the discrete temperature point data includes the corresponding actual spatial coordinates; converting the discrete temperature point data into continuous temperature distribution data using an inverse distance weighted interpolation algorithm; performing spatiotemporal registration of the continuous temperature distribution data with the two-dimensional thermal imaging data to obtain the target equipment temperature data; and mapping the target equipment temperature data to spatial grid cells generated based on the space where the power facility is located through three-dimensional raster modeling to obtain the equipment thermal map.

[0041] Specifically, for example, in substations, infrared thermal imagers are installed to periodically scan equipment such as transformers and switchgear. The thermal imager captures the temperature distribution on the equipment surface in the form of two-dimensional images, providing a thermal map of each device at a given moment. For instance, the surface of a transformer may exhibit uneven temperature distribution, indicating potential hotspots. To further improve monitoring accuracy, multiple discrete temperature sensors (such as thermocouples or infrared thermometers) are deployed on the equipment surface or critical areas. These sensors provide the temperature at each measurement point and its corresponding spatial coordinates (e.g., different locations on the transformer surface). This discrete temperature data serves as a supplementary data source, helping to refine the equipment's thermal map. An inverse distance-weighted (IDW) interpolation algorithm is used to transform the discrete temperature data into a continuous temperature distribution on the equipment surface. During this process, the data at each temperature point affects the temperature estimate of a certain surrounding area. This algorithm fills in gaps in the original temperature data and generates a more accurate temperature field on the equipment surface. Finally, the two-dimensional thermal imaging data and discrete temperature data are combined for spatiotemporal registration. By using a pre-defined equipment coordinate system and registration algorithm (such as the PnP algorithm or a calibration board-based registration method), the spatial consistency between equipment surface temperature data and thermal imaging images is ensured, and the data is integrated into a unified three-dimensional temperature model. The temperature data of the target equipment is then mapped onto a three-dimensional spatial grid model of the power facility to obtain an equipment thermal map. For example, based on the three-dimensional structure inside a substation (including equipment locations, corridors, passageways, etc.), a grid system is constructed to ensure that each grid cell represents a small spatial area. Temperature data is mapped onto these grid cells to generate an equipment thermal map, displaying the spatial distribution of equipment temperature and temperature changes.

[0042] Step S203: Obtain radar scan data of power facilities.

[0043] For example, a mobile 3D LiDAR (Light Laser Detection and Ranging) scanning system is installed on-site to perform a comprehensive scan of the power facility, obtaining high-precision 3D point cloud data. The LiDAR system includes a 360° rotating scanning head and a high-frequency pulse rangefinder, with a scanning frequency of up to 200 kHz and a horizontal resolution of 0.04°. Inspection robots or personnel carrying the LiDAR equipment travel along a preset path (such as equipment corridors or power distribution room passages) to perform multiple comprehensive scans of the entire power facility. The LiDAR records the return time and reflection intensity of each laser pulse in real time, and combines this data with inertial measurement unit (IMU) data and GPS / RTK positioning signals to generate a raw point cloud sequence with attitude information. For indoor environments or environments with weak GPS signals, high-precision positioning is achieved using inertial navigation fusion positioning based on SLAM (Simultaneous Localization and Mapping) algorithms. All raw point cloud data is aggregated by an edge computing unit and synchronously uploaded to the intelligent inspection system for storage.

[0044] Step S204: Construct a distribution map of charged equipment based on radar scan data.

[0045] For example, the acquired radar point cloud data is first preprocessed, including ground point separation (extracting and removing ground points using the RANSAC (Random Sample Consensus) algorithm), denoising, and point cloud downsampling (using the Voxel Grid algorithm to unify the voxel size to 0.05 m³). Subsequently, multiple preprocessed point cloud data frames are registered: initial coarse registration based on feature descriptors (such as FPFH (Fast Point Feature Histogram) features + SAC-IA (Sample Consensus Initial Alignment) algorithm) is used, followed by fine registration using the ICP (Iterative Closest Point) algorithm, to synthesize a global point cloud map covering the entire power facility. Next, the global point cloud is processed hierarchically based on the height segmentation method: First, the Z-coordinate distribution of each point is calculated, and the point cloud above a certain threshold (e.g., above 1.5 m) is segmented into clusters that may belong to live equipment. Then, the ECE (Euclidean Cluster Extraction) algorithm is applied to separate these high-level point clouds into several clusters, each cluster corresponding to a potential device. For each cluster, geometric features (such as the bounding box size of the point cloud shape and plane fitting error) and reflection intensity features are further extracted to distinguish different types of live equipment such as transformers, switchgear, and surge arresters. Based on each live equipment cluster, its minimum three-dimensional safety bounding box is calculated, and its location, model number, and safe operating distance range are marked on the global map, ultimately forming a live equipment distribution map.

[0046] In one possible implementation, a distribution map of energized equipment is constructed based on radar scan data, specifically including: determining point cloud data for each energized device based on radar scan data; stitching together multiple point cloud data using an iterative nearest-point algorithm to generate a power facility point cloud map; extracting device feature point clouds from the power facility point cloud map based on height segmentation; generating safe bounding boxes for energized equipment from the device feature point clouds using a bounding box generation algorithm; and obtaining the distribution map of energized equipment based on the safe bounding boxes.

[0047] Specifically, for example, in substations, high-precision 3D LiDAR scanners are installed to periodically scan the facilities. LiDAR scanners obtain spatial data of surrounding objects by emitting laser beams and receiving reflected signals. This scan data is converted into point cloud data, with each point cloud representing the spatial location of a device surface or structure obtained during the radar scan. During the scan, the LiDAR system can distinguish between energized equipment (such as transformers, cables, and switches) and non-energized equipment (such as supports, ground, and walls) based on reflection intensity and distance information, identifying the point cloud data of energized equipment. Since LiDAR scanning typically covers a large area gradually, the data obtained during the scan is scattered; therefore, point cloud stitching technology is needed to register and merge multiple scan data. The ICP algorithm is used for precise stitching. This algorithm adjusts the positions of different point clouds by minimizing the distance between them based on the spatial features of each point cloud, ultimately stitching them together to form a complete point cloud map of the power facilities. This ensures accurate reproduction of the entire substation's 3D structure and equipment locations. Since substation equipment typically has different heights and sizes, a height segmentation algorithm is used to layer the point cloud data. This process effectively distinguishes equipment at different heights; for example, transformers are generally installed on the ground, while switchgear and cables may be suspended at higher positions. Height segmentation technology analyzes the Z-coordinate (height) information of the point cloud to divide the point clouds of equipment at different heights within the substation into multiple categories. For example, the point cloud above the equipment may belong to high-voltage equipment, while the point cloud below may belong to low-voltage equipment or non-energized facilities. This allows for the further extraction of feature point cloud data for each piece of equipment in the substation, providing a foundation for subsequent equipment identification and safety bounding box generation. Based on the extracted feature point cloud data, a bounding box generation algorithm (such as the minimum bounding box algorithm) is used to construct a safety bounding box for each piece of equipment. The safety bounding box is a 3D frame that tightly encloses the point cloud of the equipment, representing the actual spatial extent of the equipment. Based on the type of equipment (such as transformers, circuit breakers, etc.) and its operating voltage, the system automatically generates appropriate safety distances (such as a 2-meter safety interval) and labels the equipment type according to its physical shape and function. Finally, the safety boundary boxes of all energized equipment are integrated into a complete distribution map of energized equipment.

[0048] Step S205: Obtain environmental data of the power facilities, including equipment mechanical vibration data.

[0049] For example, triaxial accelerometers and vibration sensors (such as piezoelectric accelerometers) are installed at critical operating parts of power facilities (such as transformer rolling bearings, cooling fans, circuit breaker contacts, etc.). These sensors feature high bandwidth (0.5 Hz–10 kHz) and high sensitivity (±16 g). The sensors transmit the triaxial vibration acceleration signals, along with the sensor ID and sampling timestamp, to an edge computing node in real time via wireless (ZigBee or LoRa) or wired (RS485) networks. To ensure data quality, the original vibration signals are sampled using 32-bit floating-point sampling, and an FFT algorithm is applied for time-domain to frequency-domain conversion to extract feature values ​​(such as root mean square (RMS), peak-to-peak factor, and spectral energy distribution) to form vibration characteristic data for subsequent analysis. Furthermore, environmental data may include information such as on-site temperature and humidity, wind speed, and local noise to comprehensively assess the equipment's operating status.

[0050] Step S206: Based on the equipment heat map, the distribution map of energized equipment and environmental data, generate a dynamic safety model of the power facilities.

[0051] For example, the equipment heat map, the distribution map of energized equipment, and the environmental data are first processed for denoising, time alignment, and normalization: For temperature data, a medium-range filter (KernelSize=3×3) is applied to remove isolated hotspots; for point cloud data, outliers are filtered out; and for vibration signals, a bandpass filter (10 Hz–1 kHz) is used to remove low-frequency and high-frequency interference. Based on the timestamps of each data source, data from different sampling frequencies are interpolated to align temperature, point cloud, and vibration data on the same time axis at a fixed time step (e.g., 1 s). The temperature range is mapped to the [0, 1] interval; vibration characteristic values ​​are normalized to [0, 1]; and the point cloud cluster features (volume, shape parameters) of energized equipment are normalized according to their maximum and minimum values. After obtaining the fused data, a dynamic safety model is constructed based on the fused data: the equipment locations in the distribution map of energized equipment are overlaid with the equipment heat map to generate a spatial risk feature map. In the spatial risk feature map, each grid cell within the bounding box of energized equipment inherits the temperature value from the heatmap, using heat (normalized temperature value) as the initial weight for risk. Different priority coefficients are assigned based on the type of energized equipment (e.g., transformer > circuit breaker > surge arrester), forming a comprehensive risk score for each grid cell. Anomalies in the equipment's mechanical vibration data are analyzed: by comparing the vibration data's FFT features with the historical baseline model, any detected vibration root mean square (RMS) value or sudden jump in spectral energy (exceeding the historical average ±3σ) is marked at the corresponding spatial coordinates. Combined with the temperature gradient of the corresponding heatmap (e.g., high-temperature gradients may indicate overheating faults), these anomaly locations are identified as high-risk electrical equipment locations. The spatial risk feature map is then fused with the high-risk equipment locations: through weighted overlay or multi-dimensional matrix stitching, a dynamic safety model covering the entire power facility is obtained. The dynamic safety model includes the risk score for each grid cell (comprehensively considering temperature, energized state, vibration anomalies, environmental factors, etc.) and is updated in real-time to reflect the current operating status.

[0052] In one possible implementation, a dynamic safety model for power facilities is generated based on equipment heat maps, distribution maps of energized equipment, and environmental data. Specifically, this includes: denoising, aligning, and normalizing the equipment heat maps, distribution maps of energized equipment, and environmental data to obtain fused data; overlaying the fused data with the equipment heat maps to generate a spatial risk feature map; marking the locations of high-risk power equipment based on the correlation between abnormal vibration locations in the equipment mechanical vibration data and temperature gradients in the equipment heat maps; and constructing a dynamic safety model by fusing the spatial risk feature map and the locations of high-risk power equipment.

[0053] Specifically, in a substation, the system collects equipment heat maps (reflecting the surface temperature distribution of equipment), energized equipment distribution maps (showing the spatial location of energized equipment), and environmental data (such as equipment vibration, temperature, and humidity). For noise data in the heat maps (such as sensor errors and environmental interference), filtering algorithms (e.g., median filtering or Kalman filtering) are applied to remove invalid data. Noise removal is also performed on the vibration data to eliminate vibration interference from environmental or non-equipment factors. Since different data may have different sampling frequencies, a time alignment algorithm is needed to synchronize all data to a unified time axis for joint analysis of equipment status. For example, the timestamps of equipment heat maps and vibration data are aligned with the equipment locations to ensure that data from the same moment can be associated with the corresponding equipment. Different types of data are standardized to ensure a balanced influence from different data sources. For example, the temperature values ​​in the equipment heat maps are normalized to the 0-1 range, and the vibration data is also normalized to the maximum value for unified calculation. After the data fusion is completed, the system uses spatial coordinate information to overlay the energized status, temperature distribution, and vibration status of the equipment to generate a spatial risk feature map. The specific steps are as follows: The equipment locations on the energized equipment distribution map are overlaid with temperature data from the heat map to obtain the temperature information for each piece of equipment. Vibration data for each piece of equipment is added to the overlaid map, with higher risk weights assigned to equipment exhibiting abnormal vibration. The spatial risk feature map generated from this data clearly displays the risk levels of temperature, vibration, and energized status of different equipment in the substation, thereby determining the overall operating status of the equipment. To further enhance safety, the system analyzes the correlation between equipment vibration and temperature data. Typically, abnormal equipment vibration may be related to signs of malfunction such as overheating or overload. For example, if abnormal vibration data is observed near a piece of equipment (such as a transformer), and the temperature gradient at that location is large (indicating a high temperature in a certain part of the equipment), the equipment may be in an overheated or overloaded state, posing a risk of failure. The system marks the locations of these devices as high-risk areas and highlights them in the spatial risk feature map using different colors, markers, or icons to alert maintenance personnel to pay close attention. By integrating the spatial risk feature map and the marked high-risk equipment locations, the system constructs a dynamic safety model. This safety model integrates the equipment's operating status, spatial distribution, and environmental factors, and is dynamically updated based on real-time data. According to the safety model, all high-risk equipment and their surrounding areas will be marked as high-risk zones and warning information will be generated. As the equipment status changes (such as temperature rise, increased vibration, etc.), the safety model will automatically update the high-risk areas to ensure that inspection personnel always keep abreast of the latest risk situation of the substation. When the equipment temperature, vibration and other parameters exceed the preset threshold, the system will automatically trigger an early warning and provide visual or voice prompts to the inspection personnel to remind them to take emergency measures.

[0054] Step S207: Based on the dynamic safety model, plan the inspection strategy for the power facilities and send the inspection strategy to the target AR glasses.

[0055] For example, target inspection personnel are determined based on the workload, current location, and power operation qualification level of all inspection personnel to be assigned. Specifically: combining historical inspection task volume (including task quantity, duration, etc.) recorded by the system and the employee's specific scheduling system, the current cumulative working hours, number of completed tasks, and remaining available working hours for each inspection personnel (e.g., no more than 8 hours per day) are calculated. Their spatial coordinates are obtained in real time using positioning terminals carried by the inspection personnel (e.g., indoor UWB positioning or outdoor GPS+RTK). The system pre-enters each inspection personnel's power operation certificate level (e.g., high-voltage inspection certificate, special operation certificate, etc.), and qualified personnel are screened based on the current type of high-risk equipment (e.g., high-voltage live equipment requires T3 or higher qualification). After identifying eligible inspection personnel, high-risk areas are determined based on equipment temperature data, equipment vibration data, and the distribution of energized equipment in the dynamic safety model. The risk scores of each grid cell in the dynamic safety model are threshold-filtered; the threshold value can be dynamically adjusted (e.g., the latest statistical threshold is 0.7), selecting grid sets with risk scores greater than the threshold. Connectivity analysis is performed on the selected high-risk grids, merging adjacent grids to obtain continuous high-risk area clusters, and calculating their center location, spatial range, and the types of equipment covered. Inspection paths are generated based on the high-risk areas and the target inspection personnel: high-risk area clusters are prioritized, with the ranking considering the average risk score, area size, and equipment importance (e.g., transformers are prioritized over circuit breakers). Based on a pre-built 3D mesh model of the power facilities (including walkable nodes for passageways, stairs, and maintenance access), a 3D path planning algorithm is used to calculate the shortest path from the inspection personnel's current location to each high-risk area, generating a sequence of nodes along the path. If multiple inspection personnel participate, the system will employ a task allocation algorithm (such as the Hungarian algorithm) to assign high-risk area tasks to appropriate personnel while minimizing the total inspection time. The generated inspection path, each high-risk equipment inspection item (such as temperature threshold, vibration threshold, and accessible distance), and other preset inspection points (such as insulation condition and appearance anomalies) are packaged into an inspection strategy file (JSON format) and pushed to the target AR glasses via a secure encrypted channel (TLSv1.3) to ensure real-time data synchronization and secure transmission.

[0056] In one possible implementation, the planning of an inspection strategy for power facilities specifically includes: determining target inspection personnel based on the workload, current location, and power operation qualification level of all inspection personnel to be assigned; identifying high-risk areas based on equipment temperature data, equipment vibration data, and the distribution of energized equipment in a dynamic safety model; generating inspection paths based on the high-risk areas and target inspection personnel; and generating an inspection strategy by combining the inspection paths and target inspection personnel.

[0057] Specifically, the system collects information such as the remaining working hours of the current inspection personnel, the number of tasks completed, and their fatigue level. Based on this, the system can calculate the workload of each inspection personnel, ensuring that excessive tasks are not assigned to personnel with overburdened workloads. Through a real-time positioning system (such as GPS or UWB positioning), the system obtains the current location of each inspection personnel and, combined with the substation's facility structure, calculates the shortest path from each personnel to the target equipment. Each inspection personnel has a power operation qualification level (such as T1, T2, T3, etc.). The system will screen qualified inspection personnel based on the hazard requirements of the inspection task (e.g., high-voltage live equipment inspection requires a qualification of T2 or higher), avoiding assigning high-risk tasks to personnel without appropriate qualifications. The system extracts equipment temperature, vibration, and other status data from the dynamic safety model and, combined with the live equipment distribution map, assesses which equipment is in a high-temperature, high-vibration state and determines the failure risk of these devices. For example, if the transformer surface temperature exceeds a preset threshold or the equipment vibration exceeds safety standards, the area where these devices are located is marked as a high-risk area. By analyzing the equipment's status, the system generates a high-risk area map, identifying equipment requiring special attention and its surrounding areas. The high-risk area map considers not only the equipment's physical location but also dynamically adjusts based on its operating status and environmental factors. Based on the distribution of high-risk areas and the current location of the inspection personnel, the system uses a shortest path algorithm (such as AStar) or optimized path planning algorithm to calculate the shortest path for the inspection personnel to reach each high-risk device, while avoiding unnecessary entry into low-risk areas. The system also considers the order of equipment inspections to avoid redundant movement or path overlap. If the risk status of the equipment changes (e.g., increased temperature or abnormal vibration), the system automatically recalculates the inspection path to ensure that inspection personnel always prioritize high-risk areas. Based on the inspection path, the system generates specific inspection tasks for each inspection personnel. These tasks include the inspection requirements for each high-risk device (e.g., measuring temperature, vibration, and visual inspection) and assign an estimated completion time to each task. Inspection personnel receive a task list and execute tasks according to the actual situation. If there are live equipment or high-temperature areas along the inspection route, the system will automatically remind inspection personnel to maintain a safe distance or wear appropriate protective equipment. For high-risk equipment (such as equipment that may malfunction), the system will highlight it in the inspection task to ensure that personnel check it first. If inspection personnel find abnormal equipment status during the inspection process, or are affected by environmental factors (such as weather changes), the system will dynamically adjust the inspection task and route based on feedback information to ensure that the task is completed under optimal conditions.

[0058] Step S208: Display the dynamic security model and inspection strategy to the target inspection personnel through the target AR glasses.

[0059] For example, through the dedicated SDK interface of the target AR glasses, the dynamic safety model generated in the background is converted into a preset visualization format (such as a Unity3D scene or OpenGL rendering format), and corresponding AR anchor points are generated in combination with the inspection path information. These anchor points contain 3D coordinates, corresponding device ID, risk level, display icon or color code information. The packaged AR anchor point data and inspection path file are pushed to the target AR glasses in real time via Wi-Fi or 5G network, and the lightweight rendering engine built into the glasses dynamically updates the scene according to the received data. When the inspection personnel wearing the AR glasses move to the designated area, the AR glasses track the wearer's head posture and field of vision in real time based on the SLAM algorithm, and use the spatial mapping constructed by the depth camera and the environment to overlay and project the dynamic safety model into the inspection personnel's field of vision, displaying the 3D outline of the electrical equipment, the thermal layer of the risk area, and using a semi-transparent thermal gradient color to indicate the risk intensity of each device or area. Within the AR glasses' field of view, the locations of detected high-risk devices are marked with a 3D bounding box, and the device name (e.g., "L10# Transformer"), current temperature value (e.g., "85℃"), vibration level (e.g., "Abnormal Vibration"), and safe distance (e.g., "Recommended to maintain a distance of more than 2 m") are displayed floating above the bounding box. This information is displayed in HUD (Heads-Up Display) format and can automatically rotate with the wearer's viewpoint. 3D navigation arrows are projected onto the ground or drawn in the air to guide inspection personnel to the next high-risk device along a predetermined path. The navigation arrows have a dynamic update function; when the wearer deviates from the path, the system automatically replans and overlays the new arrow information. When the target inspection personnel approach live equipment or enter a high-risk area, the AR glasses estimate the Euclidean distance to the device in real time using a built-in distance sensor or SLAM. Once the distance falls below a preset safe threshold (e.g., 0.5 m), a dual visual and audible warning is triggered: visually, a red warning bar flashes at the edge of the field of view, and a prompt message "Approaching dangerous live equipment, please evacuate immediately" pops up in the center area; audibly, a preset alarm sound "Danger, maintain a safe distance" plays.

[0060] In one possible implementation, a dynamic safety model and inspection strategy are displayed to the target inspection personnel through the target AR glasses. Specifically, this includes: converting the dynamic safety model into a preset format and generating AR markers in conjunction with the inspection strategy; pushing the AR markers and inspection path to the target AR glasses; overlaying the dynamic safety model onto the target AR glasses' field of view; marking equipment in high-risk areas in three dimensions, displaying the equipment name, current status, and safe distance; generating a visual navigation arrow in the target AR glasses' field of view; and triggering a distance warning when the target inspection personnel approach live equipment or high-risk areas.

[0061] Specifically, the system converts the dynamic safety model generated based on parameters such as equipment temperature, vibration, and electrical status into a preset format (e.g., JSON, XML) that AR glasses can recognize. This includes information such as the equipment's spatial location, temperature, vibration status, and risk level. Combined with the inspection strategy, the system generates AR markers for each piece of equipment that needs to be inspected. Each marker represents a piece of equipment or an inspection task, containing information such as equipment name, status, risk level, and inspection requirements. The system pushes the markers and inspection path to the inspector's AR glasses based on an inspection path planning algorithm. The inspection path is automatically generated based on the equipment's risk level, location, and the inspector's current location, ensuring that inspectors inspect high-risk areas along the optimal path. After the data is pushed to the AR glasses, the inspector will see the equipment markers and path information on the glasses screen. When the inspector enters the substation wearing the AR glasses, the dynamic safety model is overlaid in real-time within the glasses' field of view, displaying the status of all equipment in the power facilities. Equipment location, temperature, vibration, and other information are visualized in the glasses' field of view in the form of colors or icons, helping inspectors to instantly obtain the equipment's health status. For example, when a transformer is overheating, the system will display its location in red and indicate the degree of overheating. For high-risk equipment (such as equipment with excessive temperature or abnormal vibration), the system will generate a 3D marker. In the AR glasses' field of view, these devices will be presented as a 3D frame or virtual model, with the device name, status (such as "high temperature" or "abnormal vibration"), and recommended safe distance (such as 2 meters) directly displayed on the marker frame. At this time, inspectors can clearly see the location, status, and required safe distance of the equipment. The system generates visual navigation arrows for inspectors based on the inspection path and displays them in real time in the AR glasses' field of view. These arrows guide inspectors to the next inspection point or high-risk area. During the inspection, the navigation arrows will be dynamically adjusted to ensure that inspectors always stay on the optimal path and avoid unnecessary detours. The system monitors the distance between inspectors and live equipment or high-risk areas in real time through sensors in the AR glasses (such as distance sensors or SLAM positioning systems). When inspectors approach these devices, the AR glasses will trigger a distance warning. Warning prompts include visual alerts (such as flashing red warning bars) and voice alerts (such as "Caution! You are approaching live equipment, please keep a safe distance"), ensuring that inspection personnel can avoid dangerous areas in a timely manner.

[0062] Figure 3 This is another schematic flowchart of an AR inspection method for power facilities disclosed in an embodiment of this application, as shown below. Figure 3 As shown, this embodiment includes: Step S301: Decompose the inspection strategy into multiple inspection tasks.

[0063] For example, the inspection strategy file (JSON format) generated in step S207 is parsed. This file contains information such as device IDs, location coordinates, risk levels, and inspection points corresponding to several high-risk areas. The system automatically decomposes the strategy into multiple specific inspection tasks through the strategy parsing module. Each inspection task corresponds to a high-risk device or a region. The decomposition process is as follows: traverse all high-risk devices in the JSON, extract their device IDs, spatial coordinates (x, y, z), risk levels, and a list of inspection items (such as temperature measurement, vibration detection, insulation detection, and appearance inspection), forming an initial set of task nodes. A unique task number (TaskID) is assigned to each task node, and the corresponding standard inspection process (working time, tool requirements, and permission requirements) is obtained from the task library according to the device type. For device tasks located in the same or adjacent areas, it is determined whether they can be merged into a composite task to reduce repetitive round trips. For example, if two transformers in the same substation have similar risk levels and similar inspection contents, they can be merged into a "substation transformer centralized inspection task". Sort all the initially disassembled and merged tasks to generate a task queue to be assigned. The queue records the priority (based on risk level), estimated execution time (e.g., 5 minutes for each equipment inspection), required qualification level, required tools (infrared thermometer, portable vibrator, insulating oil sampling tube, etc.) and spatial coordinate information of each task.

[0064] Step S302: Issue an initial inspection task to the target AR glasses based on the complexity of the inspection task.

[0065] For example, for each disassembled inspection task, the task complexity is calculated based on the following indicators: The risk score (0–1) of the corresponding device or area is obtained from the dynamic security model. A linear mapping is used to classify risk scores [0, 0.3] as low risk (Level I), [0.3, 0.6] as medium risk (Level II), and [0.6, 1.0] as high risk (Level III). Based on the complexity score of each disassembled task, the initial inspection task is sent to the target AR glasses: tasks with a complexity score of 0-0.3 are assigned as Level I tasks, labeled with simple text and icons; tasks with a score of 0.3–0.6 are assigned as Level II tasks, accompanied by a 3D model and navigation prompts; and tasks with a score of 0.6-1.0 are assigned as Level III tasks, carrying detailed inspection procedures and safety warning information. The initial inspection task is pushed to the local task management module of the target AR glasses in JSON format through a secure channel, and a task list and summary notification are generated on the AR interface.

[0066] Step S303: Obtain the working status data of the target AR glasses, including battery level, operating status, and usage time.

[0067] For example, the AR glasses' real-time operating status data is read through the embedded operating system API, mainly including: 1. Battery level: The system battery management module interface is called to obtain the current remaining battery percentage (value 0–100%), as well as real-time voltage and current data. 2. Operating status: CPU / GPU usage, memory usage, remaining storage space, and device temperature (e.g., CPU chip temperature) are obtained through the system status management interface. 3. Usage time: The cumulative runtime of the AR glasses since power-on, and the head-wearing duration of the current session are recorded. The AR glasses package the above status data with the glasses' unique device ID and report it periodically (e.g., every 30 seconds) via 5G or Wi-Fi network to the intelligent inspection system so that device availability and battery life are considered during subsequent task allocation.

[0068] Step S304: Correct the initial inspection task based on the working status data to obtain the corrected inspection task.

[0069] For example, after receiving status data reported by the AR glasses, the intelligent inspection system selects the corresponding level of tasks and resources from the issued tasks based on the current working status. For example, if the battery level is below 30%, it prioritizes filtering Level I tasks currently stored locally on the AR glasses, pauses the display of Level II and Level III tasks, and switches related resources (such as 3D models and videos) to on-demand download mode; if the CPU / GPU utilization exceeds 80%, it selects task resources from the issued tasks that do not rely on high-load rendering, such as Level I or Level II tasks with only text prompts and static charts; if the cumulative usage time exceeds 2 hours, it removes high-duration tasks (Level II / Level III) from the issued task queue based on the usage time threshold, continues to execute low-duration Level I tasks, and displays a prompt "It is recommended to restart advanced tasks after a break"; if storage space or memory resources are tight, it prioritizes retaining cached critical task resources, temporarily takes large tasks (Level III) offline, and only retains tasks with low energy and performance consumption.

[0070] Step S305: Update the corrected inspection task to the target AR glasses.

[0071] For example, the intelligent inspection system pushes the corrected tasks to the target AR glasses via an incremental update interface. For instance, the AR glasses' local task management module detects changes based on the TaskID, compares the old and new task lists, and updates the level, description, and resource links of the corrected tasks. The system triggers an AR interface refresh, with the changed tasks distinguished by new icon colors and text, while also updating navigation path priority and model loading strategies. A status bar message pops up in the AR field of view, informing the inspection personnel of specific reasons such as "Task adjusted: Low battery, some tasks simplified." The updated operation information is synchronized to the backend logs for subsequent analysis to optimize subsequent correction rules.

[0072] Figure 4 This is a schematic diagram of a module of an AR inspection system for power facilities according to an embodiment of this application, as shown below. Figure 4 As shown, the system includes: an acquisition module 401, a processing module 402, and an inspection task allocation module 403, wherein: Acquisition module 401 is configured to acquire temperature data inside power facilities; Processing module 402 is configured to construct a thermal map of the equipment based on temperature data within the facility; The acquisition module 401 is also configured to acquire radar scan data of power facilities; Processing module 402 is also configured to construct a distribution map of charged equipment based on radar scan data; The acquisition module 401 is also configured to acquire environmental data of the power facilities, including equipment mechanical vibration data; The processing module 402 is also configured to generate a dynamic safety model of the power facility based on the equipment heat map, the distribution map of the energized equipment and environmental data; The processing module 402 is also configured to plan an inspection strategy for power facilities based on a dynamic security model and send the inspection strategy to the target AR glasses. The processing module 402 is also configured to display dynamic security models and inspection strategies to the target inspection personnel through the target AR glasses.

[0073] Optionally, the processing module 402 is also configured to: Acquire two-dimensional thermal imaging data of the surface temperature of equipment in a preset area of ​​a power facility; Acquire discrete temperature point data for the target region, which includes the corresponding actual spatial coordinates. The discrete temperature point data is converted into continuous temperature distribution data using an inverse distance weighted interpolation algorithm; The target device temperature data is obtained by spatiotemporal registration of continuous temperature distribution data and two-dimensional thermal imaging data. A thermal map of the equipment is obtained by mapping the temperature data of the target equipment to spatial grid cells generated based on the space where the power facility is located through 3D raster modeling.

[0074] Optionally, the processing module 402 is also configured to: Determine the point cloud data for each charged device based on radar scan data; The iterative nearest point algorithm is used to stitch together multiple point cloud data to generate a point cloud map of power facilities; Equipment feature point cloud is extracted from the point cloud map of power facilities based on height segmentation; The device feature point cloud is used to generate a safety bounding box for the energized equipment using a bounding box generation algorithm. The distribution map of energized equipment is obtained based on the safety boundary box.

[0075] Optionally, the processing module 402 is also configured to: The equipment heat map, the distribution map of energized equipment and environmental data are denoised, time-aligned and normalized to obtain fused data; Based on the fused data, the data of live equipment is overlaid with the equipment heat map to generate a spatial risk feature map; Based on the correlation between the abnormal vibration location in the equipment's mechanical vibration data and the temperature gradient in the equipment's thermal diagram, the location of high-risk power equipment is marked. A dynamic safety model is constructed by integrating spatial risk feature maps and the locations of high-risk power equipment.

[0076] Optionally, the processing module 402 is also configured to: Based on the workload, current location, and power operation qualification level of all inspection personnel to be assigned, the target inspection personnel are determined. High-risk areas are identified based on equipment temperature data, equipment vibration data, and the distribution of energized equipment in the dynamic safety model. Generate inspection routes based on high-risk areas and target inspection personnel; The inspection strategy is generated by combining the inspection path with the target inspection personnel.

[0077] Optionally, the processing module 402 is also configured to: The dynamic security model is converted into a preset format and AR markers are generated in conjunction with the inspection strategy. Push AR markers and inspection paths to the target AR glasses; A dynamic security model is overlaid and displayed in the field of view of the target AR glasses; Equipment in high-risk areas is marked in 3D, displaying the equipment name, current status, and safe distance; Generate visual navigation arrows within the target AR glasses' field of view; When the target inspection personnel approach the live equipment or high-risk area, a distance warning is triggered.

[0078] Optionally, the system also includes an inspection task allocation module 403, configured for: The inspection strategy is broken down into multiple inspection tasks; The initial inspection task is issued to the target AR glasses based on the complexity of the inspection task. Acquire the working status data of the target AR glasses, including battery level, operating status, and usage time; The revised inspection task is obtained by modifying the initial inspection task based on the working status data. Update the inspection task to the target AR glasses.

[0079] It should be noted that the system provided in the above embodiments is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0080] This embodiment also discloses an electronic device, as shown in the reference. Figure 5 The electronic device may include: at least one processor 501, at least one communication bus 502, user interface 503, network interface 504, and at least one memory 505.

[0081] The communication bus 502 is used to enable communication between these components.

[0082] The user interface 503 may include a display screen and a camera. Optionally, the user interface 503 may also include a standard wired interface and a wireless interface.

[0083] The network interface 504 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0084] The processor 501 may include one or more processing cores. The processor 501 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 505, and by calling data stored in memory 505. Optionally, the processor 501 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 501 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content to be displayed on the screen; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 501 and may be implemented as a separate chip.

[0085] The memory 505 may include random access memory (RAM) or read-only memory. Optionally, the memory 505 may include a non-transitory computer-readable storage medium. The memory 505 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 505 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 505 may also be at least one storage device located remotely from the aforementioned processor 501. Figure 5 As shown, the memory 505, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for an AR inspection method for power facilities.

[0086] exist Figure 5In the electronic device shown, the user interface 503 is mainly used to provide an input interface for the user and to obtain the user input data; while the processor 501 can be used to call an application program stored in the memory 505 for an AR inspection method for power facilities. When executed by one or more processors 501, the electronic device performs one or more methods as described in the above embodiments.

[0087] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0088] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0089] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the shown or discussed mutual couplings or direct couplings or communication connections may be through some service interfaces; indirect couplings or communication connections between apparatuses or units may be electrical or other forms.

[0090] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0091] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0092] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory 505 and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory 505 includes various media capable of storing program code, such as a USB flash drive, external hard drive, magnetic disk, or optical disk.

[0093] The foregoing description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Other embodiments of this disclosure will be readily apparent to those skilled in the art upon consideration of the disclosure in this specification. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. An AR inspection method for power facilities, characterized in that, The method, applied to an intelligent inspection system, includes: Acquire temperature data within power facilities; Based on the temperature data within the facility, a thermal map of the equipment is constructed; Acquire radar scan data of the power facilities; Based on the radar scan data, a distribution map of the charged equipment is constructed; Acquire environmental data of the power facility, including equipment mechanical vibration data; Based on the equipment heat map, the distribution map of the energized equipment, and the environmental data, a dynamic safety model of the power facility is generated; Based on the dynamic security model, an inspection strategy for the power facilities is planned, and the inspection strategy is sent to the target AR glasses; The dynamic security model and the inspection strategy are displayed to the target inspection personnel through the target AR glasses.

2. The method according to claim 1, characterized in that, The step of constructing a thermal map of the equipment based on the temperature data within the facility includes: Two-dimensional thermal imaging data of the surface temperature of equipment in a preset area of ​​the power facility are obtained; Acquire discrete temperature point data of the target area, wherein the discrete temperature point data includes the corresponding actual spatial coordinates; The discrete temperature point data is converted into continuous temperature distribution data using an inverse distance weighted interpolation algorithm. The target device temperature data is obtained by spatiotemporal registration of the continuous temperature distribution data and the two-dimensional thermal imaging data. The temperature data of the target equipment is mapped to spatial grid cells generated based on the space where the power facility is located through three-dimensional raster modeling to obtain the thermal map of the equipment.

3. The method according to claim 1, characterized in that, The step of constructing a distribution map of charged equipment based on the radar scan data includes: The point cloud data of each charged device is determined based on the radar scan data; The iterative nearest point algorithm is used to stitch together multiple point cloud data to generate a point cloud map of power facilities; Based on height segmentation, the equipment feature point cloud of the power facility point cloud map is extracted; The device feature point cloud is used to generate a safety bounding box for the energized equipment using a bounding box generation algorithm. The distribution map of the energized equipment is obtained based on the security boundary box.

4. The method according to claim 1, characterized in that, The process of generating a dynamic safety model for the power facility based on the equipment heat map, the distribution map of the energized equipment, and the environmental data includes: The equipment heat map, the distribution map of the energized equipment, and the environmental data are denoised, time-aligned, and normalized to obtain fused data. Based on the fused data, the data of the energized equipment is overlaid with the equipment heat map to generate a spatial risk feature map; Based on the correlation between the abnormal vibration location in the equipment's mechanical vibration data and the temperature gradient in the equipment's thermal diagram, the location of high-risk power equipment is marked. A dynamic safety model is constructed by integrating the spatial risk feature map and the location of the high-risk power equipment.

5. The method according to claim 1, characterized in that, The planned inspection strategy for the power facilities includes: Based on the workload, current location, and power operation qualification level of all inspection personnel to be assigned, the target inspection personnel are determined. Based on the equipment temperature data, equipment vibration data, and distribution of energized equipment in the dynamic safety model, high-risk areas are identified. Based on the high-risk areas and the target inspection personnel, an inspection path is generated; The inspection strategy is generated by combining the inspection path with the target inspection personnel.

6. The method according to claim 5, characterized in that, The step of displaying the dynamic security model and the inspection strategy to the target inspection personnel through the target AR glasses includes: The dynamic security model is converted into a preset format and AR markers are generated in conjunction with the inspection strategy. The AR markers and the inspection path are pushed to the target AR glasses; The dynamic security model is overlaid and displayed in the field of view of the target AR glasses; The equipment in the high-risk area is marked in three dimensions, displaying the equipment name, current status, and safe distance; Generate a visual navigation arrow within the field of view of the target AR glasses; When the target inspection personnel approach the electrical equipment or the high-risk area, a distance warning is triggered.

7. The method according to claim 1, characterized in that, The method further includes: The inspection strategy is broken down into multiple inspection tasks; An initial inspection task is issued to the target AR glasses based on the complexity of the inspection task. Acquire the working status data of the target AR glasses, including battery level, operating status, and usage time; The initial inspection task is corrected based on the working status data to obtain the corrected inspection task; Update the corrected inspection task to the target AR glasses.

8. An AR inspection system for power facilities, characterized in that, It includes an acquisition module and a processing module, wherein: The acquisition module is configured to acquire the temperature data inside the power facility; The processing module is configured to construct a thermal map of the equipment based on the temperature data within the facility. The acquisition module is also configured to acquire radar scan data of the power facility; The processing module is also configured to construct a distribution map of charged equipment based on the radar scan data; The acquisition module is also configured to acquire environmental data of the power facility, including equipment mechanical vibration data. The processing module is also configured to generate a dynamic safety model of the power facility based on the equipment heat map, the distribution map of the energized equipment, and the environmental data. The processing module is also configured to plan the inspection strategy of the power facility based on the dynamic security model, and send the inspection strategy to the target AR glasses; The processing module is also configured to display the dynamic security model and the inspection strategy to the target inspection personnel through the target AR glasses.

9. An electronic device, characterized in that, The device includes a processor, a memory, a user interface, and a network interface. The memory is used to store instructions. The user interface and the network interface are both used to communicate with other devices. The processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, perform the method as described in any one of claims 1-7.

Images