Power equipment inspection method based on AR (Augmented Reality) glasses

Abstract

The invention relates to the technical field of intelligence in the power industry, in particular to a power equipment inspection method based on AR (Augmented Reality) glasses, which comprises the following steps of: collecting power equipment data and generating equipment operation real-time data by adopting a multi-source data fusion algorithm on the basis of an Internet of Things technology. According to the invention, comprehensive collection and real-time updating of the equipment operation data are realized through the Internet of Things technology and the multi-source data fusion algorithm, and the data accuracy and timeliness are improved. And in combination with a decision tree and a neural network algorithm, the equipment state and the fault risk are deeply analyzed, and the fault prediction accuracy and the early recognition capability are improved. Through fault mode influence analysis and application of the convolutional neural network, a fault source is accurately identified, effective maintenance suggestions are provided, and the maintenance cost and the downtime are reduced. The genetic algorithm optimizes the inspection path and the work plan, and improves the inspection efficiency. Cloud computing and augmented reality technologies are combined, inspection information is visually presented, and the working efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent technology in the electric power industry, and in particular to an electric power equipment inspection method based on AR glasses. Background Art

[0002] The intelligentization of the power industry refers to the integration of intelligent technologies into the power industry to improve efficiency, quality and safety. The specific application is any method that uses programs, algorithms or automation technologies to improve the efficiency of power production, distribution and use, including automated monitoring and maintenance of power equipment.

[0003] Among them, the power equipment inspection method based on AR glasses is an advanced method of using AR glasses to inspect power equipment. Its main purpose is to improve the efficiency of power equipment inspection, ensure the safety of equipment, and improve the accuracy of inspection. On the one hand, it liberates the manual work of manual inspection and improves the efficiency and ease of use of inspection; on the other hand, it reduces inspection errors and improves the accuracy of inspection. AR glasses are usually preloaded with 3D models and related information of power equipment. During the inspection process, staff only need to check the status of the equipment through the glasses to obtain detailed parameters of the equipment, the operating status of the equipment, and whether there are potential problems. In some complex or in-depth analysis situations, AR technology can also remotely access experts for real-time guidance, effectively solving the problems of low information transmission efficiency and poor real-time performance in traditional inspection methods. Thereby greatly improving the accuracy and efficiency of power equipment inspection.

[0004] In terms of data collection and processing, existing methods for power equipment inspection rely on manual collection or simple automated systems, lacking efficient data fusion and real-time update capabilities, resulting in insufficient timeliness and accuracy of data. In terms of fault prediction and identification, there is a lack of in-depth analysis capabilities, which cannot effectively predict and identify potential faults, increasing unexpected downtime and maintenance costs. Traditional inspection route planning and work arrangements rely on experience judgment and lack scientific algorithm support, resulting in inefficient resource allocation. In terms of inspection information presentation, existing methods mostly rely on paper reports or simple electronic displays, lacking intuitive and efficient information display methods, affecting the efficiency and accuracy of inspections. Traditional methods lack intelligent means in personnel performance evaluation and route planning optimization, making it difficult to achieve continuous process improvement and personnel skills improvement. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings existing in the prior art and propose an electric power equipment inspection method based on AR glasses.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: an electric power equipment inspection method based on AR glasses, comprising the following steps:

[0007] S1: Based on the Internet of Things technology, a multi-source data fusion algorithm is used to collect power equipment data and generate real-time data on equipment operation;

[0008] S2: Based on the real-time operation data of the equipment, a decision tree and neural network algorithm are used to analyze the equipment operation status and failure risk, and generate an equipment operation status assessment report and a potential failure risk warning;

[0009] S3: Based on the potential fault risk warning, the fault mode impact analysis and convolutional neural network algorithm are used to identify the fault source and make maintenance suggestions, thereby generating a fault source identification result and maintenance suggestions;

[0010] S4: Based on the equipment operation status assessment report and the fault source identification result, a genetic algorithm is used to plan the path and plan, and an inspection path and work plan are generated;

[0011] S5: Based on the real-time data of equipment operation, equipment operation status assessment report, potential fault risk warning fault source identification results, maintenance suggestions, inspection paths and work plans, cloud computing and augmented reality technology are used to visualize the information on AR glasses to generate AR glasses display content;

[0012] S6: Based on the content displayed by the AR glasses, perform on-site inspection and maintenance work and generate an on-site maintenance work report;

[0013] S7: Based on the on-site maintenance work report, using deep learning and reinforcement learning algorithms to evaluate performance and optimize the path planning algorithm, and generate a personnel performance report and path planning algorithm optimization results;

[0014] S8: Based on the personnel performance report and the optimization result of the path planning algorithm, conduct training and update the maintenance process, and generate a personnel training and maintenance process update report;

[0015] S9: Based on the personnel training and maintenance process update report, use big data analysis and prediction algorithms to analyze the effectiveness of the training and maintenance processes, combine performance data, perform trend prediction and risk assessment, and generate a training and maintenance effectiveness analysis report;

[0016] The real-time equipment operation data specifically includes equipment operation parameters, environmental status, and operation sounds. The equipment operation status assessment report includes equipment performance indicators, operation status indications, and risk levels. The potential fault risk warning specifically includes fault points and risk levels. The fault source identification results specifically refer to faulty components and causes. Maintenance suggestions include maintenance measures and operating steps. The inspection path includes the sequence of maintenance point locations, work plans, and time arrangements. The AR glasses display content includes equipment status, warning information, fault source location, maintenance suggestions, inspection paths, and work plans. The on-site maintenance work report includes warning component inspection, maintenance, and replacement. The personnel performance report evaluation includes maintenance efficiency and quality. The path planning algorithm optimization results specifically include future inspection paths and work arrangements.

[0017] As a further solution of the present invention, based on the Internet of Things technology, a multi-source data fusion algorithm is used to collect power equipment data and generate real-time data of equipment operation. The specific steps are:

[0018] S101: Based on IoT sensor technology, data acquisition technology is used to collect power equipment data and generate original equipment monitoring data;

[0019] S102: Based on the original equipment monitoring data, a data preprocessing algorithm is used to generate processed equipment monitoring data;

[0020] S103: Based on the processed equipment monitoring data, a multi-source data fusion algorithm is used to integrate data from multiple sensors to generate fused equipment monitoring data;

[0021] S104: Based on the fused equipment monitoring data, a time series analysis method is used to monitor the equipment operation status in real time and generate real-time equipment operation data.

[0022] As a further solution of the present invention, based on the real-time data of the equipment operation, the specific steps of using decision tree and neural network algorithms to analyze the equipment operation status and failure risk, and generate an equipment operation status assessment report and potential failure risk warning are as follows:

[0023] S201: Based on the real-time data of the equipment operation, a decision tree algorithm is used to perform preliminary equipment operation status classification and risk assessment, and a preliminary risk assessment report is generated;

[0024] S202: Based on the preliminary risk assessment report, a neural network algorithm is used to deeply analyze the equipment operation mode and abnormal mode, and generate an in-depth operation status analysis report;

[0025] S203: Based on the in-depth operation status analysis report, a machine learning algorithm is used to identify potential failure modes and risk factors, and a failure mode identification result is generated;

[0026] S204: Based on the fault mode identification result, data visualization technology is used to display the equipment operation status and potential risk areas, and an equipment operation status assessment report and potential fault risk warning are generated.

[0027] As a further solution of the present invention, based on the potential fault risk warning, the fault mode impact analysis and convolutional neural network algorithm are used to identify the fault source and make maintenance suggestions. The specific steps of generating the fault source identification result and maintenance suggestions are as follows:

[0028] S301: Based on the potential failure risk warning, the failure mode impact analysis is adopted to perform failure mode analysis and evaluation, and generate a failure mode analysis report;

[0029] S302: Based on the fault mode analysis report, a convolutional neural network algorithm is used to deeply analyze the abnormal mode of the equipment and generate a fault source analysis result;

[0030] S303: Based on the fault source analysis result, a preliminary maintenance suggestion is generated using a knowledge graph method;

[0031] S304: Based on the preliminary maintenance suggestions, an expert system is used to review and optimize the maintenance suggestions, and a fault source identification result and maintenance suggestions are generated.

[0032] As a further solution of the present invention, based on the equipment operation status evaluation report and the fault source identification result, a genetic algorithm is used to plan the path and plan, and the specific steps of generating the inspection path and work plan are as follows:

[0033] S401: Based on the equipment operation status assessment report and the fault source identification result, determine the inspection and maintenance equipment and generate a target equipment list;

[0034] S402: Generate device groups based on the target device list, combined with power device characteristics and field environment information;

[0035] S403: Based on the device grouping, a preliminary inspection path is generated using a genetic algorithm;

[0036] S404: Based on the preliminary inspection path, combined with working time and equipment factors, an inspection path and a work plan are generated.

[0037] As a further solution of the present invention, based on the real-time data of equipment operation, equipment operation status assessment report, potential fault risk warning fault source identification results, maintenance suggestions, inspection paths and work plans, cloud computing and augmented reality technology are used to visualize the information on AR glasses. The specific steps of generating the AR glasses display content are as follows:

[0038] S501: Based on the real-time operation data of the equipment and the equipment operation status evaluation report, a data integration algorithm is used to generate a data integration report;

[0039] S502: Based on the data integration report, potential fault risk warning, fault source identification result and maintenance suggestion, data visualization technology is used to process information using a cloud computing platform to generate cloud visualization data;

[0040] S503: Based on the cloud visualization data, inspection path and work plan, using augmented reality technology to generate AR glasses interactive display content;

[0041] S504: Based on the AR glasses interactive display content, optimize the display format and user interaction experience to generate AR glasses display content.

[0042] As a further solution of the present invention, based on the display content of the AR glasses, the specific steps of performing on-site inspection and maintenance work and generating an on-site maintenance work report are as follows:

[0043] S601: Locate the device based on the content displayed by the AR glasses, perform a preliminary visual inspection of the device, and generate a preliminary inspection record of the device;

[0044] S602: Based on the preliminary inspection record, use a device diagnostic tool to perform in-depth device status detection and generate a device status in-depth diagnostic report;

[0045] S603: Based on the equipment status in-depth diagnosis report, perform maintenance work and generate a maintenance execution record;

[0046] S604: Based on the maintenance execution record, summarize the inspection and maintenance activities and generate an on-site maintenance work report.

[0047] As a further solution of the present invention, based on the on-site maintenance work report, the specific steps of using deep learning and reinforcement learning algorithms to evaluate performance and optimize the path planning algorithm to generate a personnel performance report and a path planning algorithm optimization result are as follows:

[0048] S701: Based on the on-site maintenance work report, a deep learning algorithm is used to analyze the work quality and efficiency to generate a maintenance work quality assessment report;

[0049] S702: Based on the maintenance work quality assessment report, a reinforcement learning algorithm is used to optimize the work path and method, and a preliminary optimization result of the path planning is generated;

[0050] S703: Based on the preliminary optimization result of the path planning, a simulated annealing algorithm is used to improve the path planning to generate a final optimization result of the path planning;

[0051] S704: Based on the final optimization result of the path planning and in combination with the work performance of the maintenance personnel, a personnel performance report and a path planning algorithm optimization result are generated.

[0052] As a further solution of the present invention, based on the personnel performance report and the optimization result of the path planning algorithm, training work and updating the maintenance process are carried out, and the specific steps of generating the personnel training and maintenance process update report are as follows:

[0053] S801: Based on the personnel performance report, identify training needs and improvement areas, design targeted training plans, and generate a training needs analysis report;

[0054] S802: Based on the training needs analysis report, actual training is conducted using case analysis and role-playing methods to generate training execution records;

[0055] S803: Based on the optimization result of the path planning algorithm, the on-site maintenance workflow is updated, and a maintenance workflow update draft is generated by using a flowchart drawing and work instruction writing method;

[0056] S804: Based on the training execution record and the maintenance process update draft, summarize and feedback are provided to generate a personnel training and maintenance process update report.

[0057] As a further solution of the present invention, based on the personnel training and maintenance process update report, big data analysis and prediction algorithms are used to analyze the training and maintenance process effects, and combined with performance data, trend prediction and risk assessment are performed to generate the training and maintenance effect analysis report in the following specific steps:

[0058] S901: Based on the personnel training and maintenance process update report, a data cleaning and format standardization method is used to generate a cleaned data set;

[0059] S902: Based on the cleaned data set, a basic data analysis report is generated by using descriptive statistical analysis and data visualization methods;

[0060] S903: Based on the basic data analysis report, a trend forecast report is generated by using time series analysis and machine learning forecasting models;

[0061] S904: Based on the trend forecast report, a risk assessment model and decision tree analysis are used to generate a training and maintenance effect analysis report.

[0062] Compared with the prior art, the advantages and positive effects of the present invention are:

[0063] In the present invention, through the application of Internet of Things technology and multi-source data fusion algorithm, the comprehensive collection and real-time update of equipment operation data are realized, and the accuracy and timeliness of the data are improved. The decision tree and neural network algorithm are combined to conduct in-depth analysis of the equipment operation status and fault risks, greatly improving the accuracy of fault prediction and early identification capabilities. The application of fault mode impact analysis and convolutional neural network can more accurately identify the source of the fault and put forward effective maintenance suggestions, reducing maintenance costs and equipment downtime. The use of genetic algorithms optimizes resource allocation in the planning of inspection paths and work plans, and improves inspection efficiency. The combination of cloud computing and augmented reality technology makes the visualization of inspection information more intuitive, and improves the work efficiency and accuracy of inspection personnel. Through deep learning and reinforcement learning to evaluate performance and optimize path planning, the intelligence and efficiency of the entire inspection process are further improved, and at the same time, personnel skills and operating standards are continuously improved through training and maintenance process updates. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 It is a schematic diagram of the main steps of the present invention;

[0065] Figure 2 It is a schematic diagram of the refinement of S1 of the present invention;

[0066] Figure 3 It is a schematic diagram of the refinement of S2 of the present invention;

[0067] Figure 4 It is a schematic diagram of the refinement of S3 of the present invention;

[0068] Figure 5 It is a schematic diagram of the refinement of S4 of the present invention;

[0069] Figure 6 It is a detailed schematic diagram of S5 of the present invention;

[0070] Figure 7 It is a detailed schematic diagram of S6 of the present invention;

[0071] Figure 8 It is a detailed schematic diagram of S7 of the present invention;

[0072] Fig. 9 It is a detailed schematic diagram of S8 of the present invention;

[0073] Fig.10 It is a detailed schematic diagram of S9 of the present invention. DETAILED DESCRIPTION

[0074] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0075] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0076] Embodiment 1

[0077] See also Figure 1 The present invention provides a technical solution: an electric power equipment inspection method based on AR glasses, comprising the following steps:

[0078] S1: Based on the Internet of Things technology, a multi-source data fusion algorithm is used to collect power equipment data and generate real-time data on equipment operation;

[0079] S2: Based on the real-time data of equipment operation, the decision tree and neural network algorithms are used to analyze the equipment operation status and failure risk, and generate equipment operation status assessment reports and potential failure risk warnings;

[0080] S3: Based on the potential fault risk warning, the fault mode impact analysis and convolutional neural network algorithm are used to identify the fault source and make maintenance suggestions, generating fault source identification results and maintenance suggestions;

[0081] S4: Based on the equipment operation status assessment report and fault source identification results, a genetic algorithm is used to plan the path and plan, and to generate the inspection path and work plan;

[0082] S5: Based on the real-time data of equipment operation, equipment operation status assessment report, potential fault risk warning fault source identification results, maintenance suggestions, inspection paths and work plans, cloud computing and augmented reality technology are used to visualize the information on AR glasses and generate AR glasses display content;

[0083] S6: Based on the content displayed by AR glasses, perform on-site inspection and maintenance work and generate on-site maintenance work reports;

[0084] S7: Based on the on-site maintenance work report, deep learning and reinforcement learning algorithms are used to evaluate performance and optimize the path planning algorithm, and generate personnel performance reports and path planning algorithm optimization results;

[0085] S8: Based on the personnel performance report and the optimization results of the path planning algorithm, conduct training and update the maintenance process, and generate personnel training and maintenance process update reports;

[0086] S9: Based on the personnel training and maintenance process update report, use big data analysis and prediction algorithms to analyze the training and maintenance process effects, combine performance data, conduct trend prediction and risk assessment, and generate training and maintenance effect analysis reports;

[0087] The real-time data of equipment operation specifically includes equipment operating parameters, environmental status, and operating sounds. The equipment operation status assessment report includes equipment performance indicators, operation status indications, and risk levels. The potential fault risk warning specifically includes the fault point and risk level. The fault source identification result specifically refers to the faulty component and cause. The maintenance suggestions include maintenance measures and operating steps. The inspection route includes the sequence of maintenance point locations, work plan, and time schedule. The content displayed by AR glasses includes equipment status, warning information, fault source location, maintenance suggestions, inspection route, and work plan. The on-site maintenance work report includes warning component inspection, maintenance, and replacement. The personnel performance report evaluation includes maintenance efficiency and quality. The path planning algorithm optimization results specifically include future inspection routes and work arrangements.

[0088] This method collects real-time data of power equipment through multi-source data fusion algorithm, which can fully understand the operating status and environmental conditions of the equipment. The application of decision tree and neural network algorithm can evaluate the operating status of the equipment and warn of potential failure risks in advance, thereby reducing the probability of equipment failure and improving the reliability and stability of the equipment.

[0089] The application of Failure Mode Effect Analysis and Convolutional Neural Network algorithms can accurately identify the source of the fault and make maintenance recommendations. This helps to quickly locate the fault point, reduce maintenance time and cost, and improve equipment availability and efficiency. The use of genetic algorithms can plan inspection routes and work plans, optimize resource utilization and schedules. By using cloud computing and augmented reality technology to visualize information on AR glasses, inspectors can intuitively obtain key information such as equipment status, warning information, fault source location, and maintenance recommendations, improving work efficiency and accuracy.

[0090] The on-site maintenance work reports generated during the execution of on-site inspections and maintenance work can provide a basis for subsequent performance evaluation and path planning algorithm optimization. The application of deep learning and reinforcement learning algorithms can evaluate personnel performance and optimize path planning algorithms to further improve inspection efficiency and quality.

[0091] The application of big data analysis and prediction algorithms can analyze the effects of training and maintenance processes and predict trends. Risk assessment can be combined with performance data to provide guidance for updating training and maintenance processes, further improving the effectiveness of personnel training and maintenance.

[0092] In summary, this method realizes the functions of comprehensive monitoring of power equipment, fault warning, fault source identification, inspection route planning and personnel performance evaluation by comprehensively utilizing Internet of Things technology, data analysis algorithm, virtual reality technology and artificial intelligence algorithm. It can improve the reliability of equipment, reduce maintenance costs, improve inspection efficiency and quality, and thus improve the operating efficiency and safety of power equipment.

[0093] See also Figure 2 ,Based on the Internet of Things technology, the specific steps of using multi-source data fusion algorithm to collect power equipment data and generate real-time data of equipment operation are as follows:

[0094] S101: Based on IoT sensor technology, data acquisition technology is used to collect power equipment data and generate original equipment monitoring data;

[0095] S102: Based on the original equipment monitoring data, a data preprocessing algorithm is used to generate processed equipment monitoring data;

[0096] S103: Based on the processed equipment monitoring data, a multi-source data fusion algorithm is used to integrate the data from multiple sensors to generate fused equipment monitoring data;

[0097] S104: Based on the fused equipment monitoring data, a time series analysis method is used to monitor the equipment operation status in real time and generate real-time equipment operation data.

[0098] Various sensors are installed on power equipment, such as temperature sensors, humidity sensors, vibration sensors, etc. These sensors send the collected raw equipment monitoring data to the data center or cloud platform through IoT sensor technology. Data acquisition technology is used to collect raw equipment monitoring data. This includes establishing a connection to communicate with the sensor, acquiring sensor data, and storing it as raw equipment monitoring data. The raw equipment monitoring data is preprocessed. The preprocessing algorithm can include processing steps such as denoising, filtering, and normalization to ensure the accuracy and reliability of the data. Multi-source data fusion algorithms are used to integrate data from multiple sensor sources. The algorithm can weightedly fuse the information from multiple data sources according to the correlation and complementarity between different sensors to generate fused equipment monitoring data. The fused equipment monitoring data is monitored in real time using time series analysis methods. The operating status and trend of the equipment can be identified, and the corresponding real-time data of equipment operation can be generated. Real-time data can be used to monitor the health of the equipment, predict potential failures, and optimize the maintenance plan of the equipment.

[0099] See also Figure 3, based on the real-time data of equipment operation, using decision tree and neural network algorithms, analyzing the equipment operation status and failure risk, and generating equipment operation status assessment reports and potential failure risk warnings. The specific steps are:

[0100] S201: Based on the real-time data of equipment operation, a decision tree algorithm is used to perform preliminary equipment operation status classification and risk assessment, and generate a preliminary risk assessment report;

[0101] S202: Based on the preliminary risk assessment report, a neural network algorithm is used to deeply analyze the equipment operation mode and abnormal mode, and generate an in-depth operation status analysis report;

[0102] S203: Based on the in-depth operation status analysis report, a machine learning algorithm is used to identify potential failure modes and risk factors, and a failure mode identification result is generated;

[0103] S204: Based on the fault mode identification results, data visualization technology is used to display the equipment operation status and potential risk areas, and generate an equipment operation status assessment report and potential fault risk warning.

[0104] First, collect the real-time data of equipment operation. Including the sensor data, operating parameters, operation records, etc. of the equipment. Ensuring the integrity and accuracy of the data is the basis for subsequent analysis. Use the decision tree algorithm to perform preliminary classification and risk assessment on the real-time data of equipment operation. The decision tree algorithm can divide the equipment into different operating status categories according to the characteristics and historical data of the equipment, and give the corresponding risk assessment results. Generate a preliminary risk assessment report to quickly understand the operating status and potential risks of the equipment. Based on the preliminary risk assessment, use the neural network algorithm to conduct a more in-depth analysis of the equipment operation mode and abnormal mode. The neural network algorithm can identify the complex patterns and associations hidden in the data by learning a large amount of equipment operation data. Generate a deep operation status analysis report to provide a more detailed description of the equipment operation status and potential problem analysis. Based on the deep operation status analysis report, use the machine learning algorithm to identify potential failure modes and risk factors. By training the model to identify known failure modes and failure-related features, the possible failure types and probabilities of the equipment can be predicted. Generate failure mode recognition results to provide a basis for subsequent maintenance and preventive measures. Use data visualization technology to display the equipment operation status assessment report and potential failure risk warning results in an intuitive way. The operating status of the equipment, potential risk areas and other information are presented in the form of charts, images, etc., to help users better understand and analyze the status of the equipment.

[0105] See also Figure 4,Based on the potential fault risk warning, the fault mode impact analysis and convolutional neural network algorithm are used to identify the fault source and make maintenance suggestions. The specific steps for generating fault source identification results and maintenance suggestions are as follows:

[0106] S301: Based on the potential failure risk warning, the failure mode impact analysis is adopted to perform failure mode analysis and evaluation, and generate a failure mode analysis report;

[0107] S302: Based on the fault mode analysis report, a convolutional neural network algorithm is used to deeply analyze the abnormal mode of the equipment and generate a fault source analysis result;

[0108] S303: Based on the fault source analysis results, a preliminary maintenance suggestion is generated using a knowledge graph method;

[0109] S304: Based on the preliminary maintenance suggestions, an expert system is used to review and optimize the maintenance suggestions, and a fault source identification result and maintenance suggestions are generated.

[0110] According to the potential fault risk warning, the failure mode impact analysis is carried out. This analysis can include the evaluation of known failure modes and their impact on equipment operation, and the generation of a failure mode analysis report. This report can help understand the impact of different failure modes on the equipment and provide a basis for subsequent fault source identification and maintenance recommendations. The abnormal mode of the equipment is deeply analyzed using the convolutional neural network algorithm. The characteristics of the normal operation mode of the equipment are learned by training the model, and compared and identified with the abnormal mode. The fault source analysis results are generated to determine the specific fault sources that may cause equipment abnormalities. Based on the fault source analysis results, the knowledge graph method is used to generate preliminary maintenance recommendations. The knowledge graph can integrate the knowledge and experience related to the equipment, and provide corresponding maintenance measures and recommendations based on the characteristics and historical data of the fault source. These recommendations may include replacing parts, adjusting parameters, and performing preventive maintenance. The expert system is used to review and optimize the preliminary maintenance recommendations. The expert system combines the knowledge and experience of domain experts to evaluate and correct the preliminary maintenance recommendations. Through the participation of the expert system, the accuracy and feasibility of the maintenance recommendations can be improved, and the final fault source identification results and maintenance recommendations can be generated.

[0111] See also Figure 5 ,Based on the equipment operation status assessment report and ,the fault source identification results, the genetic algorithm is used to ,plan the path and plan. ,The specific steps of generating the inspection path and ,work plan are as follows:

[0112] S401: Based on the equipment operation status assessment report and the fault source identification result, determine the inspection and maintenance equipment and generate a target equipment list;

[0113] S402: Generate device groups based on the target device list, combined with power device characteristics and field environment information;

[0114] S403: Generate a preliminary inspection path based on device grouping using a genetic algorithm;

[0115] S404: Based on the preliminary inspection path, combined with working time and equipment factors, an inspection path and a work plan are generated.

[0116] According to the equipment operation status assessment report and fault source identification results, determine the equipment that needs to be inspected and maintained. Generate a list of target equipment to clarify the scope of equipment that needs attention and inspection. Group the target equipment based on the characteristics of power equipment and field environment information. Divide the equipment into several groups based on factors such as the geographical location and functional characteristics of the equipment for subsequent inspection path planning and management. Based on the equipment grouping, use genetic algorithms to generate preliminary inspection paths. Genetic algorithms can optimize path planning by simulating evolutionary processes based on factors such as the distance and priority between devices. The generated preliminary inspection path can include information such as the movement order and path length between each device. According to the preliminary inspection path and working time, equipment factors and other constraints, further refine the generated inspection path and work plan. Considering the actual working time and resource constraints, adjust and optimize the preliminary inspection path to ensure that the inspection task can be completed efficiently. At the same time, formulate a detailed work plan, including specific inspection time, personnel arrangements, required tools and materials, etc.

[0117] See also Figure 6 Based on the real-time data of equipment operation, equipment operation status assessment report, potential fault risk warning fault source identification results, maintenance suggestions, inspection paths and work plans, cloud computing and augmented reality technology are used to visualize the information on AR glasses. The specific steps to generate the content displayed by AR glasses are as follows:

[0118] S501: Based on the real-time data of equipment operation and the equipment operation status evaluation report, a data integration algorithm is used to generate a data integration report;

[0119] S502: Based on the data integration report, potential fault risk warning, fault source identification results and maintenance suggestions, data visualization technology is used to process information using the cloud computing platform to generate cloud visualization data;

[0120] S503: Based on the cloud visualization data, inspection path and work plan, augmented reality technology is used to generate interactive display content for AR glasses;

[0121] S504: Based on the interactive display content of the AR glasses, optimize the display format and user interaction experience, and generate the AR glasses display content.

[0122] Based on the real-time data of equipment operation and the equipment operation status assessment report, a data integration report is generated using a data integration algorithm. The report can comprehensively analyze the equipment's operating status, potential fault risks, and fault source identification results, providing a basis for subsequent visualization. Based on the data integration report, potential fault risk warning, fault source identification results, and maintenance recommendations, the cloud computing platform is used for data processing and visualization. Through the powerful computing power and storage capacity of the cloud computing platform, a large amount of data is processed and analyzed, and cloud visualization data is generated. It can include display content in the form of charts, images, etc., which are used to intuitively present the equipment's operating status and related information. Based on the cloud visualization data, augmented reality technology is used to generate AR glasses interactive display content. AR glasses can integrate cloud visualization data with actual scenes and present information directly to users through projection or display technology. Users can see the equipment's real-time data, the location of the fault source, and related information through AR glasses, and realize real-time interaction and operation with the equipment. According to the interactive display content of AR glasses, the display format and user interaction experience are optimized to generate the final AR glasses display content. This includes adjusting the layout and display mode of data, adding interactive functions and navigation functions, etc., to improve the user's convenience and experience.

[0123] See also Figure 7 , based on the content displayed by AR glasses, perform on-site inspection and maintenance work, and generate on-site maintenance work reports. The specific steps are:

[0124] S601: Locate the device based on the content displayed by the AR glasses, perform a preliminary visual inspection of the device, and generate a preliminary inspection record of the device;

[0125] S602: Based on the preliminary inspection record, use the equipment diagnostic tool to conduct in-depth equipment status detection and generate an in-depth equipment status diagnostic report;

[0126] S603: Based on the equipment status in-depth diagnosis report, perform maintenance and servicing work and generate maintenance and servicing execution records;

[0127] S604: Based on the maintenance execution record, summarize the inspection and maintenance activities and generate an on-site maintenance work report.

[0128] According to the content displayed by AR glasses, locate the position of the equipment and conduct a preliminary visual inspection of the equipment. Through the augmented reality technology provided by AR glasses, staff can directly see the location and related information of the equipment, and conduct preliminary inspections of the appearance, connection, etc. Generate a preliminary equipment inspection record to record basic information such as the appearance and connection of the equipment. According to the preliminary equipment inspection record, use the equipment diagnostic tool to conduct in-depth status detection of the equipment. The equipment diagnostic tool can wirelessly connect to the equipment through AR glasses to obtain real-time data and status information of the equipment, and perform fault diagnosis and performance evaluation. Generate a deep equipment status diagnostic report, which records in detail the operating status, potential problems, and recommended maintenance measures of the equipment. Based on the deep equipment status diagnostic report, perform corresponding maintenance and maintenance work. According to the recommendations and requirements in the report, perform equipment repairs, replacement of parts, cleaning, etc. At the same time, record the execution of maintenance, including information such as maintenance personnel, time, specific operations, and maintenance results. Generate a maintenance and maintenance execution record for subsequent tracking and management. According to the maintenance and maintenance execution record, summarize the inspection and maintenance activities and generate a field maintenance work report. The report can include the inspection of the equipment, the execution of maintenance, the problems found, and the solutions. Reports can be used to report work progress and results to superiors and provide reference and guidance for subsequent work.

[0129] See also Figure 8 ,Based on the on-site maintenance work report, deep learning and reinforcement learning algorithms are used to evaluate performance and optimize the path planning algorithm. The specific steps for generating personnel performance reports and path planning algorithm optimization results are as follows:

[0130] S701: Based on the on-site maintenance work report, a deep learning algorithm is used to analyze the work quality and efficiency and generate a maintenance work quality assessment report;

[0131] S702: Based on the maintenance work quality assessment report, a reinforcement learning algorithm is used to optimize the work path and method, and a preliminary optimization result of the path planning is generated;

[0132] S703: Based on the preliminary optimization result of the path planning, a simulated annealing algorithm is used to improve the path planning and generate a final optimization result of the path planning;

[0133] S704: Based on the final optimization result of the path planning and combined with the work performance of the maintenance personnel, a personnel performance report and a path planning algorithm optimization result are generated.

[0134] According to the on-site maintenance work report, the quality and efficiency of the maintenance work are analyzed using a deep learning algorithm. The data in the report is learned and mined through a deep learning model, key features are extracted, and a prediction model is established. A maintenance work quality assessment report is generated to quantitatively evaluate the work performance of maintenance personnel, including maintenance accuracy, work efficiency, etc. Based on the maintenance work quality assessment report, a reinforcement learning algorithm is used to optimize the work path and method. Through interactive learning with the environment, the decision-making strategy of maintenance personnel is continuously adjusted and optimized to achieve higher work quality and efficiency. The preliminary optimization results of path planning are generated to guide maintenance personnel to choose the best path and method during inspection and maintenance. Based on the preliminary optimization results of path planning, the simulated annealing algorithm is used to further improve the path planning. The simulated annealing algorithm can gradually find the global optimal solution through random search and temperature control. According to the results of the simulated annealing algorithm, the final optimization results of path planning are generated to ensure that maintenance personnel can efficiently complete tasks during inspection and maintenance. Combined with the work performance of maintenance personnel and the final optimization results of path planning, a personnel performance report and the optimization results of the path planning algorithm are generated. The personnel performance report can comprehensively evaluate the work performance of maintenance personnel, including workload, quality, efficiency, etc. The optimization results of the path planning algorithm can be used to guide future inspection and maintenance work and improve overall work efficiency and quality.

[0135] See also Fig. 9 , based on the personnel performance report and the optimization results of the path planning algorithm, training work and updating the maintenance process are carried out. The specific steps for generating personnel training and maintenance process update reports are as follows:

[0136] S801: Based on personnel performance reports, identify training needs and improvement areas, design targeted training plans, and generate training needs analysis reports;

[0137] S802: Based on the training needs analysis report, use case analysis and role-playing methods to conduct actual training and generate training execution records;

[0138] S803: Based on the optimization results of the path planning algorithm, the on-site maintenance workflow is updated, and a maintenance workflow update draft is generated by using flowchart drawing and work instruction writing methods;

[0139] S804: Based on the training execution records and maintenance process update draft, summarize and feedback, and generate personnel training and maintenance process update reports.

[0140] Based on the personnel performance report, identify training needs and areas for improvement. By analyzing the evaluation results in the performance report, determine the deficiencies of maintenance personnel in terms of skills, knowledge, and behavior. Generate a training needs analysis report that describes in detail the training needs and key areas for improvement of each maintenance personnel. Design a targeted training plan based on the training needs analysis report. Use practical training methods such as case analysis and role-playing to combine theoretical knowledge with practical operations to help maintenance personnel improve relevant skills and knowledge. Generate training execution records to record the content, participants, time, and other information of each training. Update the on-site maintenance workflow based on the optimization results of the path planning algorithm. Adjust and improve the existing maintenance workflow based on the optimized path planning algorithm. Generate a maintenance process update draft using flowchart drawing and work instruction writing methods. Ensure that the new workflow can better adapt to the actual situation and the optimized path planning algorithm. Based on the training execution record and maintenance process update draft, provide summary feedback and generate personnel training and maintenance process update reports. Summarize the effects and results of the training and evaluate the impact of the training on the performance of maintenance personnel. At the same time, evaluate and improve the maintenance process update draft to ensure the feasibility and effectiveness of the new process. Organize the results of training and process updates into reports for reporting to superiors and subsequent reference.

[0141] See also Fig.10 Based on the personnel training and maintenance process update report, big data analysis and prediction algorithms are used to analyze the training and maintenance process effects. Combined with performance data, trend prediction and risk assessment are performed. The specific steps for generating training and maintenance effect analysis reports are as follows:

[0142] S901: Based on personnel training and maintenance process update reports, data cleaning and format standardization methods are used to generate cleaned data sets;

[0143] S902: Based on the cleaned data set, use descriptive statistical analysis and data visualization methods to generate a basic data analysis report;

[0144] S903: Generate trend forecast report based on basic data analysis report using time series analysis and machine learning forecast model;

[0145] S904: Based on the trend forecast report, use the risk assessment model and decision tree analysis to generate a training and maintenance effectiveness analysis report.

[0146] According to the personnel training and maintenance process update report, the relevant data is preprocessed using data cleaning and format standardization methods. Data cleaning includes removing duplicate values, processing missing values ​​and outliers, etc., to ensure the accuracy and completeness of the data set. A cleaned data set is generated to provide a reliable data basis for subsequent analysis. Based on the cleaned data set, descriptive statistical analysis and data visualization methods are used for basic data analysis. The effect of the training and maintenance process is preliminarily evaluated and analyzed through statistical indicators and charts. A basic data analysis report is generated, including the number of people participating in the training, the training duration, and the indicators of maintenance process improvement. Based on the basic data analysis report, time series analysis and machine learning prediction models are used for trend prediction. Through the analysis of historical data and model training, the effect change trend of the training and maintenance process in the future is predicted. A trend prediction report is generated, including the change trend of performance indicators, the improvement of maintenance efficiency, etc. Based on the trend prediction report, a risk assessment model and decision tree analysis are used to conduct a comprehensive evaluation of the training and maintenance effect. By considering the impact of different factors on the effect, the risks and potential problems of the training and maintenance process are evaluated. A training and maintenance effect analysis report is generated, including effect evaluation results, risk warnings, and improvement suggestions.

[0147] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A power equipment inspection method based on AR glasses, characterized in that: The following steps are involved: Based on the Internet of Things technology, a multi-source data fusion algorithm is used to collect power equipment data and generate real-time data on equipment operation; Based on the real-time operation data of the equipment, the decision tree and neural network algorithms are used to analyze the equipment operation status and failure risk, and generate an equipment operation status assessment report and potential failure risk warning; Based on the potential fault risk warning, the fault source is identified and maintenance suggestions are made by using the fault mode impact analysis and convolutional neural network algorithm, thereby generating a fault source identification result and maintenance suggestions; Based on the equipment operation status assessment report and fault source identification results, a genetic algorithm is used to plan the path and plan, and to generate an inspection path and a work plan; Based on the real-time data of equipment operation, equipment operation status assessment report, potential fault risk warning fault source identification results, maintenance suggestions, inspection paths and work plans, cloud computing and augmented reality technology are used to visualize the information on AR glasses to generate AR glasses display content; Based on the content displayed by the AR glasses, perform on-site inspection and maintenance work and generate on-site maintenance work reports; Based on the on-site maintenance work report, deep learning and reinforcement learning algorithms are used to evaluate performance and optimize the path planning algorithm, and generate a personnel performance report and path planning algorithm optimization results; Based on the personnel performance report and the optimization results of the path planning algorithm, conduct training work and update the maintenance process, and generate a personnel training and maintenance process update report; Based on the personnel training and maintenance process update report, use big data analysis and prediction algorithms to analyze the effectiveness of the training and maintenance process, combine performance data, conduct trend prediction and risk assessment, and generate a training and maintenance effectiveness analysis report; The real-time equipment operation data specifically includes equipment operation parameters, environmental status, and operation sounds. The equipment operation status assessment report includes equipment performance indicators, operation status indications, and risk levels. The potential fault risk warning specifically includes fault points and risk levels. The fault source identification results specifically refer to faulty components and causes. Maintenance suggestions include maintenance measures and operating steps. The inspection path includes the sequence of maintenance point locations, work plans, and time arrangements. The AR glasses display content includes equipment status, warning information, fault source location, maintenance suggestions, inspection paths, and work plans. The on-site maintenance work report includes warning component inspection, maintenance, and replacement. The personnel performance report evaluation includes maintenance efficiency and quality. The path planning algorithm optimization results specifically include future inspection paths and work arrangements.

2. The power equipment inspection method based on AR glasses according to claim 1 is characterized in that: Based on the Internet of Things technology, the specific steps of using multi-source data fusion algorithm to collect power equipment data and generate real-time data of equipment operation are as follows: Based on IoT sensor technology, data acquisition technology is used to collect power equipment data and generate original equipment monitoring data; Based on the original equipment monitoring data, a data preprocessing algorithm is used to generate processed equipment monitoring data; Based on the processed equipment monitoring data, a multi-source data fusion algorithm is used to integrate data from multiple sensors to generate fused equipment monitoring data; Based on the fused equipment monitoring data, a time series analysis method is adopted to monitor the equipment operation status in real time and generate real-time equipment operation data.

3. The power equipment inspection method based on AR glasses according to claim 1 is characterized in that: Based on the real-time data of the equipment operation, the decision tree and neural network algorithms are used to analyze the equipment operation status and failure risk, and the specific steps of generating the equipment operation status assessment report and potential failure risk warning are as follows: Based on the real-time data of the equipment operation, a decision tree algorithm is used to perform preliminary equipment operation status classification and risk assessment, and a preliminary risk assessment report is generated; Based on the preliminary risk assessment report, a neural network algorithm is used to deeply analyze the equipment operation mode and abnormal mode to generate an in-depth operation status analysis report; Based on the in-depth operation status analysis report, a machine learning algorithm is used to identify potential failure modes and risk factors, and a failure mode identification result is generated; Based on the fault mode identification results, data visualization technology is used to display the equipment operation status and potential risk areas, and generate an equipment operation status assessment report and potential fault risk warning.

4. The power equipment inspection method based on AR glasses according to claim 1 is characterized in that: Based on the potential fault risk warning, the fault mode impact analysis and convolutional neural network algorithm are used to identify the fault source and make maintenance suggestions. The specific steps of generating the fault source identification result and maintenance suggestions are as follows: Based on the potential failure risk warning, the failure mode impact analysis is adopted to conduct failure mode analysis and evaluation and generate a failure mode analysis report; Based on the fault mode analysis report, a convolutional neural network algorithm is used to deeply analyze the abnormal mode of the equipment and generate a fault source analysis result; Based on the fault source analysis results, a knowledge graph method is used to generate preliminary maintenance suggestions; Based on the preliminary maintenance suggestions, an expert system is used to review and optimize the maintenance suggestions, and to generate fault source identification results and maintenance suggestions.

5. The power equipment inspection method based on AR glasses according to claim 1 is characterized in that: Based on the equipment operation status assessment report and fault source identification results, the specific steps of using genetic algorithm to plan the path and plan and generate the inspection path and work plan are as follows: Based on the equipment operation status assessment report and fault source identification results, determine the inspection and maintenance equipment and generate a target equipment list; Based on the target device list, combined with the characteristics of the power equipment and the field environment information, generate device groups; Based on the device grouping, a preliminary inspection path is generated using a genetic algorithm; Based on the preliminary inspection path, combined with working time and equipment factors, an inspection path and a work plan are generated.

6. The power equipment inspection method based on AR glasses according to claim 1 is characterized in that: Based on the real-time data of equipment operation, equipment operation status assessment report, potential fault risk warning fault source identification results, maintenance suggestions, inspection paths and work plans, cloud computing and augmented reality technology are used to visualize the information on AR glasses. The specific steps for generating the content displayed on AR glasses are as follows: Based on the real-time data of the equipment operation and the equipment operation status evaluation report, a data integration algorithm is used to generate a data integration report; Based on the data integration report, potential fault risk warning, fault source identification results and maintenance suggestions, data visualization technology is used to process information using a cloud computing platform to generate cloud visualization data; Based on the cloud visualization data, inspection routes and work plans, augmented reality technology is used to generate interactive display content for AR glasses; Based on the AR glasses interactive display content, the display format and user interaction experience are optimized to generate AR glasses display content.

7. The power equipment inspection method based on AR glasses according to claim 1 is characterized in that: Based on the AR glasses display content, the specific steps for performing on-site inspection and maintenance work and generating an on-site maintenance work report are as follows: Based on the content displayed by the AR glasses, locate the device, perform a preliminary visual inspection of the device, and generate a preliminary inspection record of the device; Based on the preliminary inspection records, use equipment diagnostic tools to conduct in-depth equipment status detection and generate an in-depth equipment status diagnostic report; Based on the in-depth diagnosis report of the equipment status, perform maintenance and servicing work and generate maintenance and servicing execution records; Based on the maintenance execution record, inspection and maintenance activities are summarized to generate an on-site maintenance work report.

8. The power equipment inspection method based on AR glasses according to claim 1 is characterized in that: Based on the on-site maintenance work report, the specific steps of using deep learning and reinforcement learning algorithms to evaluate performance and optimize the path planning algorithm to generate a personnel performance report and path planning algorithm optimization results are as follows: Based on the on-site maintenance work report, a deep learning algorithm is used to analyze the work quality and efficiency to generate a maintenance work quality assessment report; Based on the maintenance work quality assessment report, a reinforcement learning algorithm is used to optimize the work path and method, and a preliminary optimization result of the path planning is generated; Based on the preliminary optimization result of the path planning, a simulated annealing algorithm is used to improve the path planning to generate a final optimization result of the path planning; Based on the final optimization result of the path planning and combined with the work performance of the maintenance personnel, a personnel performance report and a path planning algorithm optimization result are generated.

9. The power equipment inspection method based on AR glasses according to claim 1 is characterized in that: Based on the personnel performance report and the optimization results of the path planning algorithm, training work is carried out and the maintenance process is updated. The specific steps for generating personnel training and maintenance process update reports are as follows: Based on the personnel performance reports, identify training needs and improvement areas, design targeted training plans, and generate training needs analysis reports; Based on the training needs analysis report, actual training is conducted using case analysis and role-playing methods to generate training execution records; Based on the optimization results of the path planning algorithm, the on-site maintenance workflow is updated, and a maintenance workflow update draft is generated by using flowchart drawing and work instruction writing methods; Based on the training execution records and maintenance process update draft, summary feedback is provided to generate a personnel training and maintenance process update report.

10. The power equipment inspection method based on AR glasses according to claim 1, characterized in that: Based on the personnel training and maintenance process update report, big data analysis and prediction algorithms are used to analyze the effectiveness of the training and maintenance process. Combined with performance data, trend prediction and risk assessment are performed. The specific steps for generating a training and maintenance effectiveness analysis report are as follows: Based on the personnel training and maintenance process update report, a data cleaning and format standardization method is used to generate a cleaned data set; Based on the cleaned data set, a basic data analysis report is generated using descriptive statistical analysis and data visualization methods; Based on the basic data analysis report, a trend forecast report is generated using time series analysis and machine learning forecasting models; Based on the trend forecast report, a risk assessment model and decision tree analysis are used to generate a training and maintenance effectiveness analysis report.

Images