A method, system, storage medium and electronic device for inspecting electric equipment

Abstract

A method, system, storage medium and electronic device for power equipment inspection, relating to the technical field of equipment inspection. The method comprises: collecting historical point cloud data of each power equipment, constructing a digital twin library of the equipment based on the historical point cloud data; determining the inspection tasks and target inspection routes corresponding to each inspection device in the intelligent inspection cluster; controlling the corresponding inspection device to inspect each power equipment according to the inspection tasks and the target inspection routes, and obtaining multi-source heterogeneous data of the inspection; and determining the inspection results of each power equipment based on the multi-source heterogeneous data and the digital twin library of the equipment. The implementation of the technical solution provided in the present application improves the accuracy of abnormal inspections of each power equipment.

Description

Technical Field

[0001] The present application relates to the technical field of equipment inspection, and in particular to a method, system, storage medium and electronic device for inspecting electric power equipment. Background Art

[0002] As power equipment plays an increasingly important role in modern society, ensuring its safe operation has become a vital task. As an important part of equipment maintenance and management, power equipment inspection can not only detect potential problems in a timely manner, but also prevent equipment failures and ensure the stable operation of the power system.

[0003] At present, some power equipment monitoring systems use inspection robots for automated inspections and use machine learning models to perform preliminary analysis on the acquired data to improve inspection efficiency and accuracy. However, since the actual number of abnormal data samples of power equipment collected is usually small, it is difficult to provide sufficient training sets for deep learning models, resulting in low discrimination and generalization capabilities of the model, and it is unable to accurately identify and diagnose abnormalities of various types of power equipment under complex working conditions. Summary of the invention

[0004] The present application provides a power equipment inspection method, system, storage medium and electronic device, which improve the accuracy of abnormal inspection of each power equipment.

[0005] In a first aspect, the present application provides a method for inspecting electric power equipment, the method comprising: Collect historical point cloud data of each power equipment, and build a digital twin library of the equipment based on the historical point cloud data; Determine the inspection tasks and target inspection routes corresponding to each inspection device in the intelligent inspection cluster; Control the corresponding inspection equipment to inspect each of the power equipment according to the inspection task and the target inspection route, and obtain multi-source heterogeneous inspection data; The inspection results of each of the power equipment are determined based on the multi-source heterogeneous data and the device digital twin library.

[0006] By adopting the above technical solution, by collecting historical point cloud data of each power equipment and building a digital twin library of the equipment, a wealth of prior knowledge and benchmark data is provided for subsequent anomaly detection. During actual inspections, by determining the inspection tasks and target inspection routes corresponding to each inspection device in the intelligent inspection cluster, the reasonable allocation of inspection resources and the optimization planning of inspection routes are realized, and the inspection efficiency is improved. During the inspection process, the inspection equipment collects multi-source heterogeneous data according to the inspection tasks and target inspection routes. By comparing and analyzing these multi-source heterogeneous data collected in real time with the historical data in the digital twin library of the equipment, the abnormal state of the power equipment can be more accurately identified. In addition, since the digital twin library of the equipment contains a large amount of historical point cloud data, it makes up for the problem of fewer actual abnormal samples, so that the system has better discrimination ability and generalization performance when performing anomaly detection, and improves the accuracy of abnormal inspections of various power equipment.

[0007] In a second aspect of the present application, a power equipment inspection system is provided, the system comprising: A twin library construction module, used to collect historical point cloud data of each power equipment and build a device digital twin library based on the historical point cloud data; The inspection information determination module is used to determine the inspection tasks and target inspection routes corresponding to each inspection device in the intelligent inspection cluster; An inspection data acquisition module, used to control the corresponding inspection equipment to inspect each of the power equipment according to the inspection task and the target inspection route, and obtain multi-source heterogeneous inspection data; An inspection result determination module is used to determine the inspection result of each of the power equipment based on the multi-source heterogeneous data and the device digital twin library.

[0008] In a third aspect of the present application, a computer storage medium is provided, wherein the computer storage medium stores a plurality of instructions, wherein the instructions are suitable for being loaded by a processor and executing the above method steps.

[0009] In a fourth aspect of the present application, an electronic device is provided, comprising: a processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the above-mentioned method steps.

[0010] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: This application provides rich prior knowledge and benchmark data for subsequent anomaly detection by collecting historical point cloud data of each power equipment and building a digital twin library of the equipment. During actual inspections, by determining the inspection tasks and target inspection routes corresponding to each inspection device in the intelligent inspection cluster, the reasonable allocation of inspection resources and the optimization planning of inspection routes are achieved, and the inspection efficiency is improved. During the inspection process, the inspection equipment collects multi-source heterogeneous data according to the inspection tasks and target inspection routes. By comparing and analyzing these multi-source heterogeneous data collected in real time with the historical data in the digital twin library of the equipment, the abnormal state of the power equipment can be more accurately identified. In addition, since the digital twin library of the equipment contains a large amount of historical point cloud data, it makes up for the problem of fewer actual abnormal samples, so that the system has better discrimination ability and generalization performance when performing anomaly detection, and improves the accuracy of abnormal inspections of various power equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a flowchart of a method for inspecting electric power equipment provided in an embodiment of the present application; Figure 2 It is a module schematic diagram of a power equipment inspection system provided in an embodiment of the present application; Figure 3 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application.

[0012] Description of reference numerals: 300, electronic device; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. DETAILED DESCRIPTION

[0013] In order to enable technicians in this field 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 in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0014] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0015] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0016] The following will provide a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.

[0017] Please refer to Figure 1 , a flow chart of a method for inspecting electric equipment is proposed, the method can be implemented by a computer program, can be implemented by a single-chip microcomputer, or can be run on an electric equipment inspection system, the computer program can be integrated in a computer device, or can be run as an independent tool application, specifically, the method includes steps 10 to 40, the steps are as follows: Step 10: Collect historical point cloud data of each power equipment and build a digital twin library of the equipment based on the historical point cloud data.

[0018] Among them, the power equipment in the embodiment of this application refers to various equipment in the power transmission and distribution system, including but not limited to transformers, switch cabinets, disconnectors, mutual inductors, busbars, insulators, etc. These power equipment are important components of the power grid, responsible for transmitting and distributing electricity, and their safe and reliable operation is directly related to the stability of the entire power supply system.

[0019] In the embodiments of the present application, historical point cloud data refers to a large amount of point cloud data collected by scanning the power equipment during normal operation using technologies such as 3D laser scanning. These point cloud data can more realistically reflect the geometric feature information such as the 3D structure, shape and spatial position of the power equipment. By analyzing and modeling the historical point cloud data, the digital representation of the power equipment can be reconstructed, laying the foundation for subsequent abnormality detection and fault diagnosis.

[0020] In the embodiment of this application, the equipment digital twin library refers to a virtualized digital model library constructed through digital modeling technology based on historical point cloud data and other information of power equipment. This digital twin library can reproduce the actual state, operating parameters and working environment of power equipment with high precision and in an all-round manner, realizing the digital mapping and virtual reproduction of power equipment. In this application, the equipment digital twin library serves as a reference model library for abnormal diagnosis, which is the benchmark for determining whether there is an abnormality or not. It also supports functions such as fault simulation, providing support for operation and maintenance and prediction.

[0021] Specifically, advanced point cloud acquisition equipment such as 3D laser scanners and depth cameras are used to perform all-round and high-precision scanning of power equipment when it is operating normally, and obtain a large amount of point cloud data. These point cloud data can truly reflect the geometric information such as the 3D structure, morphological details and spatial position of each power equipment. Based on the collected historical point cloud data, the first version of the equipment digital twin is constructed through digital modeling technologies such as geometric feature extraction and semantic feature analysis. The digital twin can accurately reproduce the geometric form of the power equipment. In order to improve the accuracy and authenticity of the digital twin, it will be registered and fused with the BIM model (building information model) of the power equipment to obtain the second version of the digital twin. The BIM model contains rich semantic information such as structural parameters and material properties of the power equipment. After fusion, it can make up for the deficiencies of the digital twin in details and properties. On this basis, the operating environment information of each power equipment will also be obtained, and the digital twin will be rendered and texture mapped to give it a more realistic visual effect. At the same time, various faults and defects are simulated in the digital twin, and a digital twin library of equipment in various abnormal states is generated based on the pre-built fault mode library.

[0022] Based on the above embodiment, as an optional embodiment, the step of building a device digital twin library based on historical point cloud data may further include the following steps: Step 101: Extract geometric features and semantic features from historical point cloud data, and construct a digital twin of the first device based on the geometric features and semantic features.

[0023] Specifically, the acquired point cloud data is first preprocessed, including removing noise points, filling missing areas, etc., to obtain clean and regular point cloud data. Then, the point cloud is processed using a geometric feature extraction algorithm to automatically identify and extract basic geometric shapes in the point cloud, such as planes, spheres, cylinders, etc., to obtain a preliminary geometric feature description of the point cloud, and then further extract richer geometric features, such as the fine shapes of key components such as pipelines and insulators. At the same time, based on deep learning models such as neural networks, point cloud data can be semantically segmented to identify different types of power equipment such as transformers and switchgear, and their semantic features, such as size, material, power, etc., can be extracted. After obtaining geometric features and semantic features, feature-based three-dimensional reconstruction technology will be used to fuse these feature information and reconstruct the first device digital twin. The digital twin can more realistically reproduce the overall three-dimensional form and local detail structure of the power equipment, and at the same time contains rich semantic attribute information.

[0024] Step 102: Align and fuse the first equipment digital twin with the BIM model of each power equipment to obtain a second equipment digital twin.

[0025] Specifically, the BIM model corresponding to each power equipment is obtained, and the BIM model contains rich semantic information such as the structural dimensions, material properties, and functional parameters of the equipment. Then, through the geometric registration algorithm, the spatial position and scale of the first equipment digital twin and the BIM model are automatically calibrated so that the two can be accurately aligned and overlapped. After the registration is completed, the constraint-based fusion optimization method is used to fuse the geometric form of the digital twin with the structural parameters of the BIM model. In this way, the digital twin not only retains the realistic appearance of the point cloud reconstruction, but also introduces semantic information such as the precise structural dimensions and material properties in the BIM model, thereby obtaining the second equipment digital twin.

[0026] Step 103: Render and perform fault simulation on the digital twin of the second device to obtain a device digital twin library.

[0027] Specifically, the actual operating environment information of each power equipment is obtained, including ambient lighting, background scenes, weather conditions, etc. Then, based on this environmental information, the second device digital twin is rendered by lighting simulation, texture mapping, etc., so that the digital twin not only has real three-dimensional morphology and attribute information, but also has visual effects consistent with the real environment, and the third device digital twin is obtained. Then, based on the pre-built fault mode library, various faults and defects are simulated in the third device digital twin. The fault mode library is a collection of common fault models summarized through long-term practical experience and theoretical research, including various types of power equipment faults such as insulator insulation strike, busbar contamination conduction, and transformer partial discharge. In the fault simulation process, the corresponding fault model will be called, combined with the structure and function parameters of the power equipment, the corresponding geometry and attribute changes will be automatically generated, and the real scenes of various faults and defects will be simulated. For example, when simulating the insulation strike of the insulator, the damaged geometry and electrical characteristics can be truly reproduced by changing the material and deformation of the insulator surface. Through rendering and fault simulation, a complete device digital twin library containing normal and abnormal states is finally formed. The digital twins in this library not only have realistic three-dimensional appearance and attribute information, but also simulate various possible failure scenarios, providing a comprehensive reference object for subsequent abnormal diagnosis.

[0028] Step 20: Determine the inspection tasks and target inspection routes corresponding to each inspection device in the intelligent inspection cluster.

[0029] In the embodiments of the present application, the intelligent inspection cluster refers to an automated inspection system composed of various types of inspection equipment. These inspection equipment include drones, inspection robots, mobile inspection vehicles, etc., which are equipped with sensors such as laser scanners and high-precision cameras, and can autonomously operate to detect and scan and collect power equipment.

[0030] In the embodiment of this application, the inspection task refers to the specific inspection object and work content assigned to each inspection device in the intelligent inspection cluster. According to the distribution of power equipment, the importance of equipment, the vulnerability rate and other factors, the solution will reasonably plan the target power equipment to be inspected for each inspection device, as well as the specific tasks such as data collection and measurement that need to be performed, so as to achieve efficient operation and avoid omissions and repeated inspections.

[0031] In the embodiment of the present application, the target inspection route refers to the optimal inspection path planned for each inspection device in the intelligent inspection cluster. According to the starting position of each inspection device, the distribution position of the power equipment to be inspected, and the operating capacity of the equipment itself (such as cruising range, speed, etc.), a patrol route with the shortest distance and optimal time is calculated. A reasonable target inspection route can not only maximize the operating efficiency, but also avoid the omission of inspections caused by the limitation of cruising range or operating range, thereby ensuring the comprehensiveness and quality of the inspection.

[0032] Specifically, the distribution data of power equipment in the area to be inspected is obtained, including attribute information such as the type, location, and importance of the equipment. At the same time, it is also necessary to obtain the starting position, operating capacity and other parameters of each inspection equipment in the intelligent inspection cluster, such as the cruising range of the drone and the operating radius of the inspection robot. Based on artificial experience rules and optimization algorithms, the specific inspection tasks of each inspection equipment are reasonably allocated and planned. The planning principles include: taking high-risk power equipment as the priority inspection object according to the importance and failure rate of the equipment; reasonably arranging the same inspection equipment for adjacent equipment to avoid repeated operations; and planning multiple inspection equipment for collaborative inspection for large power equipment. After determining the inspection tasks of each inspection equipment, the optimal target inspection route will be planned for each inspection equipment based on the constrained shortest path algorithm. The algorithm will comprehensively consider multiple constraints such as the starting position of the equipment, operating capacity restrictions (such as cruising range), and inspection task distribution, and calculate a route plan with the shortest distance and covering all inspection tasks.

[0033] Based on the above embodiment, as an optional embodiment, the step of determining the inspection tasks and target inspection routes corresponding to each inspection device in the intelligent inspection cluster may further include the following steps: Step 201: Obtain the current spatial location distribution of each power device, and divide the spatial location distribution into multiple inspection areas.

[0034] Specifically, the spatial location coordinate data of all power equipment in the jurisdiction can be obtained from the geographic information system of the power company, including the latitude and longitude of the equipment, altitude, etc., and the attribute parameters such as the type and power of the equipment are also required. After obtaining the spatial location distribution of each power equipment, the density clustering algorithm is used to divide the power equipment into regions. The Euclidean distance between any two power equipment is calculated based on the spatial location coordinates of the power equipment, and the distance threshold and the minimum equipment number threshold are set; then, with any power equipment as the center, the number of equipment within the neighborhood with the distance threshold as the radius is counted. When the number of equipment in the neighborhood is greater than the minimum equipment number threshold, the area is divided into an inspection area; finally, multiple inspection areas are obtained through iterative calculation, and there are overlapping areas between adjacent inspection areas to ensure the continuity of the inspection. When dividing the inspection area, factors such as the voltage level, equipment type and detection requirements of the power equipment must also be considered. For power equipment of the same voltage level, the same type and similar detection requirements, they are preferentially divided into the same inspection area. In addition, the initially divided inspection areas are adjusted and optimized based on the terrain features and obstacle distribution of the substation to ensure that the power equipment in each inspection area can be effectively reached and detected by the inspection equipment.

[0035] Step 202: Based on the current status of the power equipment in each inspection area and the inspection requirements, determine the inspection type corresponding to each inspection area.

[0036] Specifically, it is necessary to obtain the current operating status information of each power device, such as whether it is operating normally, whether there is a minor fault or a serious fault. This status information can come from the monitoring data of the device itself, or it can be obtained through previous inspection results. At the same time, it is also necessary to combine the inspection cycle requirements of each device to determine its current inspection needs, such as routine inspection, fine detection or troubleshooting. Then match the current status and inspection demand information of each power device with the pre-set inspection type corresponding rules to determine the inspection type that should be performed in each inspection area. For example, for equipment that operates normally, only a routine appearance inspection may be required, corresponding to the ordinary inspection type; for equipment that is known to have a fault, a fine detection of the fault diagnosis type needs to be performed; for equipment that has not been repaired for a long time, an in-depth inspection of the comprehensive detection type may be required.

[0037] Step 203: Match the inspection devices corresponding to each inspection area in the inspection cluster based on each inspection type.

[0038] Specifically, the intelligent inspection cluster can include different inspection equipment such as drones, land inspection robots, and pipeline vehicles, which differ in mobility, detection range, load capacity, and endurance. At the same time, it is necessary to establish a capability model for each inspection equipment and describe its various parameter indicators in detail.

[0039] Match the inspection type of each inspection area that has been determined with the capability model of each device in the inspection cluster. Different inspection types correspond to different task requirements, such as specific detection methods, operating radius, load capacity, etc. Only devices with corresponding capabilities can be competent. Through matching analysis, one or more best-matching inspection device candidates can be locked for each inspection area. For inspection areas with multiple candidate devices, further screening can be performed according to additional priority rules (such as device availability, energy consumption level, etc.) to determine the designated inspection device for the area. If there is no completely matching device in the cluster, a combination of multiple devices can also be temporarily deployed to jointly perform the inspection task of the area.

[0040] Step 204: Plan target inspection routes for inspection equipment corresponding to each inspection area according to the operating capacity of each inspection equipment.

[0041] Specifically, a virtual three-dimensional inspection scene model is established to accurately simulate various obstacles, terrain and other factors in the actual inspection environment. The operating capacity parameters of the inspection equipment assigned to each inspection area are imported into the three-dimensional scene model. These capacity parameters include the equipment's mobility, detection range, load capacity, remaining power, etc. Secondly, for each designated inspection equipment in the inspection area, its energy consumption and the detection quality score of the power equipment under different optional inspection routes are simulated and evaluated in the three-dimensional virtual scene. Energy consumption mainly depends on the route distance and terrain conditions; while the detection quality score is related to factors such as the angle and distance of the inspection route, which will affect the detection effect of the power equipment.

[0042] Based on the simulation evaluation, the optimal target inspection route of the inspection equipment in the area can be comprehensively determined through weighted calculation of energy consumption and detection quality score. This optimal route can balance energy utilization and detection quality to the greatest extent, so as to maximize the endurance and operating efficiency of the equipment.

[0043] Based on the above embodiment, as an optional embodiment, the step of planning the target inspection route of the inspection equipment corresponding to each inspection area according to the operation capacity of each inspection equipment may also include the following steps: Step 2041: Create a virtual three-dimensional inspection scene model, and import the operating capability of each inspection device into the three-dimensional inspection scene model. The operating capability at least includes mobility, detection range, and remaining power.

[0044] Specifically, a high-precision three-dimensional virtual scene model is constructed based on the actual inspection environment data (such as terrain, obstacles, layout of power facilities, etc.). This can be achieved by using point cloud data obtained by drones / ground vehicles in combination with BIM modeling technology. The constructed model needs to truly reflect the actual environmental details such as terrain, buildings, and pipeline directions. Then the operating capacity parameters of different types of inspection equipment in each inspection area are imported into the three-dimensional virtual scene model. These operating capacity parameters include the mobility of the equipment (such as maximum speed, acceleration, mobility, etc.), detection range (such as field of view, resolution, etc.), remaining power / endurance, etc. The parameter values ​​of different types of equipment will vary.

[0045] Step 2042: For any inspection device corresponding to each inspection area, in the three-dimensional inspection scene model, simulate the energy consumption of each inspection device under different inspection routes in the inspection area, and obtain the detection quality score of the power equipment.

[0046] Specifically, in a three-dimensional virtual scene, for a specific inspection area and its assigned inspection equipment (such as drones or ground inspection robots), all possible inspection route plans in the area are listed. These route plans can be automatically generated based on the spatial distribution of power equipment in the area, or some typical route options can be preset. Secondly, for each optional inspection route plan, the virtual scene is used to simulate the energy consumption of the inspection equipment when it travels along this route. The energy consumption calculation needs to comprehensively consider multiple factors such as route distance, terrain undulations, and equipment mobility parameters. Through simulation, the energy consumption value under the route plan can be obtained. At the same time, it is also necessary to simulate and evaluate the inspection coverage and inspection quality of all power equipment along the route under the route plan to obtain a detection quality score. The calculation of this score needs to consider parameters such as the equipment's detection range, resolution, relative position and angle with the power equipment. For each optional route plan, its energy consumption and detection quality score are simulated and evaluated in the above manner.

[0047] Step 2043: Determine a target inspection route among different inspection routes based on energy consumption and detection quality scores.

[0048] Specifically, for the designated inspection equipment in each inspection area, the energy consumption value and the detection quality score are normalized from all the optional inspection route plans in the area so that they are in the same numerical range. Secondly, according to the preset weight coefficient, the two indicators of energy consumption and detection quality score are weighted and summed to obtain the comprehensive score value of each route plan. The weight coefficient can be adjusted according to actual needs. If more attention is paid to energy saving, the weight of the energy consumption indicator can be appropriately increased. Then, all route plans are sorted from high to low according to the comprehensive score value, and the route with the highest score value is the optimal target inspection route for the inspection equipment in the area. If there are multiple route plans with the same score and tied for the highest, the final target route can be further screened out according to additional priority rules (such as route length, operation risk, etc.).

[0049] Step 30: Control the corresponding inspection equipment to inspect each power equipment according to the inspection task and the target inspection route, and obtain multi-source heterogeneous inspection data.

[0050] Specifically, the inspection tasks and target inspection routes corresponding to each planned inspection area are sent to the corresponding inspection equipment, such as drones, inspection robots, etc. After receiving the instructions, these devices will autonomously move to the corresponding operation area according to the planned route, and perform the corresponding detection operations according to the assigned inspection task type. Secondly, different types of inspection equipment will collect multi-source heterogeneous inspection data according to their own sensor configurations during the inspection process. For example, drones can obtain visual images and thermal imaging data through airborne high-resolution cameras, thermal infrared cameras, etc.; while ground inspection robots can use laser radars, metal detectors, etc. to collect point cloud data, metal defect signals, etc. Different types of data can reflect the operating status of power equipment in multiple dimensions such as appearance, temperature, and geometric shape. Each inspection device uploads the collected multi-source heterogeneous data to the central database of the intelligent inspection platform through wireless transmission or data exchange for data storage and management. While collecting data, some equipment can also perform simple preliminary analysis on site, such as using edge computing devices to identify defects in image data, and upload the analysis results together. For serious hidden dangers found, early warnings can also be triggered in time. Through the above process, the intelligent inspection platform can obtain heterogeneous inspection data covering all power equipment and various types, laying a data foundation for subsequent in-depth analysis, fault diagnosis, intelligent decision-making, etc. The organic integration of these heterogeneous data can fully reflect the operating status of power equipment and improve the accuracy of fault identification. At the same time, the data collected in real time based on the planned route has the advantages of good timeliness and repeatable acquisition, which is also more conducive to tracking and analyzing the dynamic changes of equipment status.

[0051] Step 40: Determine the inspection results of each power equipment based on multi-source heterogeneous data and the equipment digital twin library.

[0052] Specifically, the collected multi-source heterogeneous inspection data is preprocessed, such as data cleaning, labeling, and format unification, to ensure the quality of subsequent data analysis. Then, these heterogeneous data are fused, and the status data of the same power equipment in different dimensions are integrated together to form a holographic data view of the equipment. Secondly, the intelligent inspection platform will find the corresponding virtual twin model for each actual device based on the pre-established digital twin library of power equipment. These digital twin models can accurately simulate the operating status of the equipment under normal working conditions, including parameters and modes in various aspects such as appearance, temperature, and geometric dimensions. Then, by comparing and analyzing the holographic data of the actual equipment with its virtual twin model, the intelligent inspection platform can accurately identify the deviation between the two, that is, various anomalies and defects that may exist in the equipment. This comparative analysis based on the twin model can eliminate interference factors that affect the accuracy of the analysis to the greatest extent. On the basis of anomaly detection, the intelligent inspection platform can further analyze and diagnose the root cause, type, and degree of harm of the anomaly. This can use artificial intelligence technologies such as machine learning, combined with the historical operation and maintenance records of the equipment, domain knowledge base, etc., to intelligently identify and classify abnormal features. Finally, based on the above analysis results, an inspection report is generated for each power equipment, including the overall health status score of the equipment, a list of various defects and detailed descriptions, recommended maintenance measures, etc. For equipment with low scores or major hidden dangers, an alarm will be triggered and the on-site duty personnel will be prompted to dispatch emergency repairs in time.

[0053] Based on the above embodiment, as an optional embodiment, the step of determining the inspection result of each power equipment based on multi-source heterogeneous data and the equipment digital twin library may also include the following steps: Step 401: Based on multi-source heterogeneous data, construct a time sequence state description of each power device.

[0054] Specifically, for each heterogeneous data stream in the multi-source heterogeneous data, a state feature sequence of the device that changes over time in that dimension is extracted, and a one-dimensional temporal state description submodel of the device is constructed.

[0055] For example, for image data streams, computer vision and deep learning techniques can be used to automatically identify defect features (such as rust, cracking, etc.) on the appearance of the equipment in the image, and to build a time series model based on the changes of these defect features over time. For point cloud data streams, point cloud registration, modeling and other algorithms can be applied to extract the time series change features of the geometric shape of the equipment surface and build a corresponding time series description model. For data streams of other dimensions such as temperature and metal defects, different algorithms can also be used for time series feature extraction and modeling. The above multiple single-dimensional time series description sub-models are fused to build a multi-dimensional comprehensive time series state description model of the power equipment. This model can fully describe the dynamic change trend of the health status of the equipment in multiple dimensions such as appearance, temperature, geometry, etc.

[0056] Step 402: Perform timing similarity analysis on the timing state description and the normal mode in the equipment digital twin library to obtain abnormal power equipment with abnormal similarity.

[0057] Specifically, for each actual power equipment, the normal state mode of its corresponding virtual twin model will be retrieved from the digital twin library as a comparison. The normal state mode describes the ideal operating state of this type of equipment under the design conditions and when each component is operating normally, including characteristic data in multiple dimensions such as appearance, temperature, and geometry. Then, the actual timing state description model of the power equipment is compared with its normal state mode for timing similarity analysis, and a similarity score is given. The comparison and analysis process can use technologies such as dynamic time warping and phase alignment to eliminate the influence of factors such as time offset. The similarity score reflects the degree of deviation between the actual state of the equipment and its normal design state. If the score is lower than the preset similarity threshold, the power equipment will be preliminarily determined as a suspicious device with abnormal hidden dangers and needs further analysis. For these preliminary abnormal devices, based on their multi-source heterogeneous inspection data, artificial intelligence technologies such as machine learning are used for pattern recognition to determine whether it is a known fault type. If a known fault mode is identified, such as insulation aging, loose rivets, etc., the device is finally confirmed as an abnormal power device with abnormal similarity, and the corresponding diagnostic report and maintenance suggestions are output. However, if the recognition result of the abnormal pattern is an unknown type, it means that there may be a new form of failure. At this time, the device is temporarily classified as an "unknown abnormal device", and the digital twin library is updated based on its abnormal data to supplement the new failure mode. In the process of handling unknown abnormal devices, artificial experts can also be invited to intervene to help analyze the cause of the abnormality and annotate the new failure mode. Once the new mode is confirmed, it will be updated to the digital twin library for subsequent identification. At the same time, the intelligent inspection platform will also output the preliminary inspection result report of the unknown abnormal device, reminding the on-site duty personnel to conduct on-site inspections in a timely manner to prevent the expansion of hidden dangers.

[0058] Step 403: Determine the abnormal cause corresponding to each abnormal power equipment in the equipment digital twin library.

[0059] Specifically, the intelligent inspection platform will retrieve the virtual twin model corresponding to the abnormal power equipment from the equipment digital twin library. The model not only contains the standard state of the equipment under normal working conditions, but also describes in detail the physical structure of the equipment, the working principles of each component and their mutual relationships. Secondly, the intelligent inspection platform will input the multi-source heterogeneous inspection data of the abnormal equipment into its virtual twin model, allowing the model to simulate and reproduce the abnormal state in a virtual environment. During the simulation process, the twin model will carefully deduce the operating status of each component based on physical constraints and working mechanisms, and finally determine the location and cause of the abnormality. For example, if the abnormal data shows that the temperature of the metal insulation layer of a high-voltage transmission line is abnormally high, the intelligent inspection platform will first run the virtual twin model of the line in a computer to simulate its influence by various physical processes such as heat conduction and electromagnetic induction. After simulation, if the twin model shows that the abnormal high temperature is caused by local discharge heating caused by the cracking of a key insulator, then the intelligent inspection platform can diagnose the specific cause of the abnormality as a structural defect in the insulator.

[0060] Step 404: Based on the current operating environment of the abnormal power equipment, the cause of the abnormality is evaluated and diagnosed in a graded manner to obtain a diagnosis result.

[0061] Specifically, the intelligent inspection platform will retrieve all the operating environment data of each abnormal power equipment, including geographical location, ambient temperature and humidity, load level, atmospheric pollution and other information. These environmental parameters will affect the actual working state of the equipment, and thus have a certain correction effect on the determined abnormal causes. Secondly, the intelligent inspection platform will input the above environmental data into the digital twin model corresponding to the abnormal equipment. The digital twin model will simulate the impact of environmental parameters on the operating state of each component of the equipment in a virtual environment, and combine the preliminary abnormal causes previously diagnosed to give a corrected comprehensive abnormal diagnosis result. For example, for a transmission line that is abnormal due to defective insulators, if the ambient temperature in the area where it is located is high and the sunshine intensity is high, the digital twin model will predict that the insulator defect will accelerate local discharge and further heat. At the same time, if the actual load of the line is found to be heavy, it will further aggravate the temperature rise of the insulation layer. Therefore, based on the simulation analysis results of the digital twin model, the intelligent inspection platform will upgrade the original abnormal cause of "insulator defect" to a high-level diagnostic conclusion of "serious insulator defect, which is very likely to cause line burnout".

[0062] Step 405: Generate inspection results by combining the diagnosis results with each abnormal power device.

[0063] Specifically, after obtaining the final diagnosis result of the abnormal cause, the intelligent inspection platform will generate a personalized inspection report for each abnormal power equipment. The report not only describes the abnormal phenomenon of the equipment, the location of occurrence and the diagnosed abnormal cause in detail, but also analyzes the environmental impact, gives the development trend of the abnormality and the degree of hazard assessment. According to the hazard rating, the report will make specific repair or replacement suggestions for the abnormal equipment, and provide operation instructions and emergency plans to guide subsequent on-site operations to ensure safety and efficiency.

[0064] See also Figure 2 , is a module diagram of a power equipment inspection system provided in an embodiment of the present application, the power equipment inspection system may include: a twin library construction module, an inspection information determination module, an inspection data acquisition module and an inspection result determination module, wherein: A twin library construction module, used to collect historical point cloud data of each power equipment and build a device digital twin library based on the historical point cloud data; The inspection information determination module is used to determine the inspection tasks and target inspection routes corresponding to each inspection device in the intelligent inspection cluster; An inspection data acquisition module, used to control the corresponding inspection equipment to inspect each of the power equipment according to the inspection task and the target inspection route, and obtain multi-source heterogeneous inspection data; An inspection result determination module is used to determine the inspection result of each of the power equipment based on the multi-source heterogeneous data and the device digital twin library.

[0065] Optionally, the twin library construction module is further used to extract geometric features and semantic features from the historical point cloud data, and construct a digital twin of the first device based on the geometric features and the semantic features; Aligning and fusing the first device digital twin with the BIM model of each of the power equipment to obtain a second device digital twin; The digital twin of the second device is rendered and fault simulated to obtain a device digital twin library.

[0066] Optionally, the twin library construction module is further used to obtain operating environment information of each of the power devices, and render the second device digital twin based on the operating environment information to obtain a third device digital twin; Based on the pre-built fault mode library, various faults and defects are simulated in the third device digital twin to obtain the device digital twin library.

[0067] Optionally, the inspection information determination module is further used to obtain the current spatial position distribution of each of the electric power devices, and divide the spatial position distribution into a plurality of inspection areas; Determine the inspection type corresponding to each inspection area based on the current status and inspection requirements of the power equipment in each inspection area; Matching inspection devices corresponding to each inspection area in the inspection cluster based on each inspection type; The target inspection routes of the inspection equipment corresponding to each inspection area are planned according to the operating capacity of each inspection equipment.

[0068] Optionally, the inspection information determination module is further used to establish a virtual three-dimensional inspection scene model, and import the operating capability of each inspection device into the three-dimensional inspection scene model, wherein the operating capability at least includes mobility, detection range, and remaining power; For any inspection device corresponding to each of the inspection areas, simulating the energy consumption of each of the inspection devices under different inspection routes in the inspection area in the three-dimensional inspection scene model, and obtaining the detection quality score of the power equipment; A target inspection route is determined among different inspection routes based on the energy consumption and the detection quality score.

[0069] Optionally, the inspection result determination module is further used to construct a timing state description of each of the power devices based on the multi-source heterogeneous data; Performing a timing similarity analysis on the timing state description and the normal mode in the device digital twin library to obtain abnormal power equipment with abnormal similarity; Determining the abnormal cause corresponding to each of the abnormal power equipment in the device digital twin library; Combined with the current operating environment of the abnormal power equipment, the cause of the abnormality is evaluated and diagnosed in a graded manner to obtain a diagnosis result; The inspection result is generated by combining the diagnosis result with each abnormal power equipment.

[0070] Optionally, the inspection result determination module is further used to calculate the timing similarity score between the timing state description and the normal state mode in the device digital twin library to obtain a similarity score; Determine the power equipment whose similarity score is lower than the similarity threshold as preliminary abnormal power equipment; Performing abnormal pattern recognition on multi-source heterogeneous data of each of the preliminary abnormal power devices to obtain a recognition result; If the identification result is a known fault mode, the preliminary abnormal power equipment is confirmed as abnormal power equipment; The method further comprises: If the identification result is an unknown fault mode, the preliminary abnormal power equipment is determined as an unknown abnormal power equipment; Based on the abnormal pattern of the unknown abnormal power equipment and the corresponding multi-source heterogeneous data, the device digital twin library is updated, and the inspection results of the unknown abnormal power equipment are output.

[0071] It should be noted that: when the system provided in the above embodiment realizes its functions, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is 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 embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0072] An embodiment of the present application also provides a computer storage medium, which can store multiple instructions. The instructions are suitable for being loaded by a processor and executing a power equipment inspection method of the above embodiment. The specific execution process can be found in the specific description of the above embodiment, which will not be repeated here.

[0073] Please refer to Figure 3 The application also discloses an electronic device. Figure 3 The electronic device 300 may include: at least one processor 301 , at least one network interface 304 , a user interface 303 , a memory 305 , and at least one communication bus 302 .

[0074] The communication bus 302 is used to realize the connection and communication between these components.

[0075] The user interface 303 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0076] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0077] Among them, the processor 301 may include one or more processing cores. The processor 301 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 305, and calling data stored in the memory 305. Optionally, the processor 301 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 301 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 301, and it can be implemented separately through a chip.

[0078] Among them, the memory 305 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 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 a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 305 may optionally also be at least one storage device located away from the aforementioned processor 301. Refer to Figure 3 , the memory 305 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application program of a power equipment inspection method.

[0079] exist Figure 3In the electronic device 300 shown, the user interface 303 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 301 can be used to call an application program storing a method for inspecting electric equipment in the memory 305, and when executed by one or more processors 301, the electronic device 300 executes one or more of the methods described in the above embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited to the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0080] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0081] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0082] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0083] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0084] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, 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, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0085] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, those skilled in the art will easily think of other embodiments of the present disclosure.

[0086] This application is intended to cover any variation, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the art not described in the present disclosure. The description and examples are to be regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for inspecting electric power equipment, characterized in that: The method comprises: Collect historical point cloud data of each power equipment, and build a digital twin library of the equipment based on the historical point cloud data; Determine the inspection tasks and target inspection routes corresponding to each inspection device in the intelligent inspection cluster; Control the corresponding inspection equipment to inspect each of the power equipment according to the inspection task and the target inspection route, and obtain multi-source heterogeneous inspection data; The inspection results of each of the power equipment are determined based on the multi-source heterogeneous data and the device digital twin library.

2. The power equipment inspection method according to claim 1, characterized in that: The constructing of a device digital twin library based on the historical point cloud data includes: Extracting geometric features and semantic features from the historical point cloud data, and constructing a digital twin of the first device based on the geometric features and the semantic features; Aligning and fusing the first device digital twin with the BIM model of each of the power equipment to obtain a second device digital twin; The digital twin of the second device is rendered and fault simulated to obtain a device digital twin library.

3. The power equipment inspection method according to claim 2, characterized in that: The rendering and fault simulation of the second device digital twin are performed to obtain a device digital twin library, including: Acquire operating environment information of each of the electric power devices, and render the digital twin of the second device based on the operating environment information to obtain a digital twin of the third device; Based on the pre-built fault mode library, various faults and defects are simulated in the third device digital twin to obtain the device digital twin library.

4. The power equipment inspection method according to claim 1, characterized in that: Determining the inspection tasks and target inspection routes corresponding to each inspection device in the intelligent inspection cluster includes: Acquire the current spatial position distribution of each of the electric power devices, and divide the spatial position distribution into a plurality of inspection areas; Determine the inspection type corresponding to each inspection area based on the current status and inspection requirements of the power equipment in each inspection area; Matching inspection devices corresponding to each inspection area in the inspection cluster based on each inspection type; The target inspection routes of the inspection equipment corresponding to each inspection area are planned according to the operating capacity of each inspection equipment.

5. The power equipment inspection method according to claim 4, characterized in that: The planning of target inspection routes of the inspection equipment corresponding to each inspection area according to the operation capacity of each inspection equipment includes: Establishing a virtual three-dimensional inspection scene model, and importing the operating capability of each inspection device into the three-dimensional inspection scene model, wherein the operating capability at least includes mobility, detection range, and remaining power; For any inspection device corresponding to each of the inspection areas, simulating the energy consumption of each of the inspection devices under different inspection routes in the inspection area in the three-dimensional inspection scene model, and obtaining the detection quality score of the power equipment; A target inspection route is determined among different inspection routes based on the energy consumption and the detection quality score.

6. The power equipment inspection method according to claim 1, characterized in that: Determining the inspection result of each of the power equipment based on the multi-source heterogeneous data and the device digital twin library includes: Based on the multi-source heterogeneous data, construct a time sequence state description of each of the power devices; Performing a timing similarity analysis on the timing state description and the normal mode in the device digital twin library to obtain abnormal power equipment with abnormal similarity; Determining the abnormal cause corresponding to each of the abnormal power equipment in the device digital twin library; Combined with the current operating environment of the abnormal power equipment, the cause of the abnormality is evaluated and diagnosed in a graded manner to obtain a diagnosis result; The inspection result is generated by combining the diagnosis result with each abnormal power equipment.

7. The power equipment inspection method according to claim 6, characterized in that: The method of performing a timing similarity analysis on the timing state description and the normal mode in the device digital twin library to obtain abnormal power equipment with abnormal similarity includes: Calculate the timing similarity score between the timing state description and the normal state mode in the device digital twin library to obtain a similarity score; Determine the power equipment whose similarity score is lower than the similarity threshold as preliminary abnormal power equipment; Performing abnormal pattern recognition on multi-source heterogeneous data of each of the preliminary abnormal power devices to obtain a recognition result; If the identification result is a known fault mode, the preliminary abnormal power equipment is confirmed as abnormal power equipment; The method further comprises: If the identification result is an unknown fault mode, the preliminary abnormal power equipment is determined as an unknown abnormal power equipment; Based on the abnormal pattern of the unknown abnormal power equipment and the corresponding multi-source heterogeneous data, the device digital twin library is updated, and the inspection results of the unknown abnormal power equipment are output.

8. A power equipment inspection system, characterized in that: The system comprises: A twin library construction module, used to collect historical point cloud data of each power equipment and build a device digital twin library based on the historical point cloud data; The inspection information determination module is used to determine the inspection tasks and target inspection routes corresponding to each inspection device in the intelligent inspection cluster; An inspection data acquisition module, used to control the corresponding inspection equipment to inspect each of the power equipment according to the inspection task and the target inspection route, and obtain multi-source heterogeneous inspection data; An inspection result determination module is used to determine the inspection result of each of the power equipment based on the multi-source heterogeneous data and the device digital twin library.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: It 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 used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method as described in any one of claims 1-7.

Images