On-site remote patrol method and device for real-time monitoring and abnormity insight

Abstract

The invention relates to an on-site remote patrol method and device for real-time monitoring and abnormity insight, and relates to the technical field of artificial intelligence. The method comprises the following steps: obtaining service quality data corresponding to each stage of service in a power grid enterprise platform system, and extracting a to-be-checked data sample from the service quality data; obtaining a preset quality rule in the quality control technology standard library, determining a qualified data sample meeting the quality rule, calculating the similarity between the to-be-checked data sample and the qualified data sample, and determining a remote patrol result corresponding to each stage of business according to the similarity; analyzing reasons for quality abnormity of each target device, and determining quality tracing information corresponding to each target device; the quality tracing information is used for determining source equipment causing quality abnormity of the target equipment. By adopting the method, the patrol efficiency of the quality abnormity monitoring process of the power equipment can be improved, and the patrol effect is improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for on-site remote inspection with real-time monitoring and abnormality insight. Background Art

[0002] The quality of equipment in the power grid directly affects the operational efficiency and safety and reliability of the power grid. In recent years, accidents related to equipment risks in the power grid industry have occurred from time to time, and equipment quality hazards and family defects have been discovered during internal maintenance and repair. The lack of transparency in the entire chain of data in each business link leads to insufficient information for management decisions, and various subjective interests drive the reporting of data to have confidence intervals. People feel helpless in the face of massive data in unrestored scenarios, and it is difficult to assess the competence of employees under the influence of complex manufacturing scenarios.

[0003] It can be seen that the current quality abnormality monitoring process of power equipment has the problems of low inspection efficiency and poor inspection effect. Summary of the invention

[0004] Based on this, it is necessary to provide an on-site remote inspection method, device, computer equipment, computer-readable storage medium and computer program product that can improve the inspection efficiency of the quality abnormality monitoring process of power equipment, improve real-time monitoring and abnormality insight of inspection results, and address the above-mentioned technical problems.

[0005] In a first aspect, an embodiment of the present application provides a method for on-site remote inspection with real-time monitoring and abnormal insight. The method comprises:

[0006] Obtaining service quality data corresponding to services at each stage in the power grid enterprise platform system, and extracting data samples to be verified from the service quality data; the service quality data is used to characterize the quality of the operation status of the services at each stage;

[0007] Acquire the quality rules preset in the quality control technology standard library, determine the qualified data samples that meet the quality rules, calculate the similarity between the data samples to be verified and the qualified data samples, and determine the remote inspection results corresponding to the business at each stage according to the similarity; the remote inspection results are used to characterize the target equipment corresponding to the business at each stage with abnormal quality;

[0008] Based on the entire life cycle of the target equipment corresponding to the business at each stage, the causes of quality abnormalities in each target equipment are analyzed, and quality traceability information corresponding to each target equipment is determined; the quality traceability information is used to determine the source equipment that causes the quality abnormalities in the target equipment.

[0009] In one embodiment, the method further comprises:

[0010] Based on the pre-built equipment life cycle quality data abnormality warning model, combined with the project technical specification information, the data of each business link of the equipment life cycle are compared to determine the quality data of the target equipment;

[0011] When the quality data does not conform to the data regulations in the equipment life cycle quality data abnormality warning model, quality data abnormality warning information is generated.

[0012] In one embodiment, the method further comprises:

[0013] According to the differences in equipment categories, the equipment feature values ​​are extracted to build a feature library corresponding to the equipment;

[0014] Build the knowledge graph corresponding to the device based on machine learning information extraction technology based on sequence annotation;

[0015] According to the feature library corresponding to the device and the knowledge graph corresponding to the device, a full-chain traceability model for quality and safety is constructed; the full-chain traceability model for quality and safety is used to generate the quality traceability information.

[0016] In one embodiment, the quality traceability information includes a quality traceability result of the power equipment, and the method further includes:

[0017] Based on the forward chain multi-step reasoning method, the system queries and displays the events of defect failures that have occurred in the knowledge base and determines the root causes in a goal-oriented manner;

[0018] Based on the station information in the geographic wiring diagram, the station internal wiring diagram is constructed according to the model data and power grid topology data stored in the graph database;

[0019] A multi-step query analysis is performed based on the tags in the form of fault description text to determine the quality tracing result of the power equipment.

[0020] In one embodiment, the quality traceability information includes a suspected familial defect analysis result, and the method further includes:

[0021] Based on the high-frequency subgraph mining method, the correlation between the defect failure information and the problem root cause information corresponding to each equipment product is analyzed;

[0022] Analyze events of the same type as the target event to achieve probability ranking information for suspected family defects of the equipment; the target event is a fault or defect event with a determined cause;

[0023] The suspected familial defect analysis result is determined according to the probability ranking information.

[0024] In one embodiment, the calculating the similarity between the data sample to be verified and the qualified data sample includes:

[0025] By representing text as word vectors, a combination of word vectors is used to represent text;

[0026] Encode the labels through Huffman coding and extract text features;

[0027] Based on the word frequency in the statistical corpus, the similarity between the data sample to be verified and the qualified data sample is calculated.

[0028] In a second aspect, the present application also provides a remote on-site inspection device for real-time monitoring and abnormal insight. The device comprises:

[0029] A data sample acquisition module is used to obtain the service quality data corresponding to the services at each stage in the power grid enterprise platform system, and extract the data samples to be verified from the service quality data; the service quality data is used to characterize the quality of the operation status of the services at each stage;

[0030] A remote inspection result determination module is used to obtain the quality rules preset in the quality control technology standard library, determine the qualified data samples that meet the quality rules, calculate the similarity between the data samples to be verified and the qualified data samples, and determine the remote inspection results corresponding to the business at each stage according to the similarity; the remote inspection results are used to characterize the target equipment corresponding to the business at each stage with abnormal quality;

[0031] The quality traceability module is used to analyze the causes of quality abnormalities of each target device based on the entire life cycle of the target device corresponding to the business at each stage, and determine the quality traceability information corresponding to each target device; the quality traceability information is used to determine the source device that causes the quality abnormality of the target device.

[0032] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0033] Obtaining service quality data corresponding to services at each stage in the power grid enterprise platform system, and extracting data samples to be verified from the service quality data; the service quality data is used to characterize the quality of the operation status of the services at each stage;

[0034] Acquire the quality rules preset in the quality control technology standard library, determine the qualified data samples that meet the quality rules, calculate the similarity between the data samples to be verified and the qualified data samples, and determine the remote inspection results corresponding to the business at each stage according to the similarity; the remote inspection results are used to characterize the target equipment corresponding to the business at each stage with abnormal quality;

[0035] Based on the entire life cycle of the target equipment corresponding to the business at each stage, the causes of quality abnormalities in each target equipment are analyzed, and quality traceability information corresponding to each target equipment is determined; the quality traceability information is used to determine the source equipment that causes the quality abnormalities in the target equipment.

[0036] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0037] Obtaining service quality data corresponding to services at each stage in the power grid enterprise platform system, and extracting data samples to be verified from the service quality data; the service quality data is used to characterize the quality of the operation status of the services at each stage;

[0038] Acquire the quality rules preset in the quality control technology standard library, determine the qualified data samples that meet the quality rules, calculate the similarity between the data samples to be verified and the qualified data samples, and determine the remote inspection results corresponding to the business at each stage according to the similarity; the remote inspection results are used to characterize the target equipment corresponding to the business at each stage with abnormal quality;

[0039] Based on the entire life cycle of the target equipment corresponding to the business at each stage, the causes of quality abnormalities in each target equipment are analyzed, and quality traceability information corresponding to each target equipment is determined; the quality traceability information is used to determine the source equipment that causes the quality abnormalities in the target equipment.

[0040] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0041] Obtaining service quality data corresponding to services at each stage in the power grid enterprise platform system, and extracting data samples to be verified from the service quality data; the service quality data is used to characterize the quality of the operation status of the services at each stage;

[0042] Acquire the quality rules preset in the quality control technology standard library, determine the qualified data samples that meet the quality rules, calculate the similarity between the data samples to be verified and the qualified data samples, and determine the remote inspection results corresponding to the business at each stage according to the similarity; the remote inspection results are used to characterize the target equipment corresponding to the business at each stage with abnormal quality;

[0043] Based on the entire life cycle of the target equipment corresponding to the business at each stage, the causes of quality abnormalities in each target equipment are analyzed, and quality traceability information corresponding to each target equipment is determined; the quality traceability information is used to determine the source equipment that causes the quality abnormalities in the target equipment.

[0044] The above-mentioned on-site remote inspection method, device, computer equipment, storage medium and computer program product for real-time monitoring and abnormal insight first obtain the quality data of the business at each stage from the power grid enterprise platform system to evaluate the business operation status and extract the data samples to be verified; then, according to the quality rules in the quality control standard library, the qualified data samples are screened out, and the similarity between the samples to be verified and the qualified samples is calculated, so as to determine the remote inspection results of the business at each stage and identify the potential quality abnormal equipment; then, based on the whole life cycle of the target equipment, the abnormal cause is analyzed and the quality traceability information is extracted, and the key features are identified by using the text similarity algorithm; finally, a full-chain traceability model for the quality and safety of key equipment is constructed, and the equipment quality traceability, defect analysis and timing analysis of faults are realized by combining the graph database and the knowledge graph; through the above-mentioned technical scheme, the intelligence level of equipment management is improved, and the accuracy and real-time performance of abnormality detection are enhanced, thereby effectively ensuring the safe operation of the power grid, improving the inspection efficiency of the quality abnormality monitoring process of power equipment, and improving the inspection effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0046] FIG1 is a diagram showing an application environment of a method for on-site remote inspection with real-time monitoring and abnormal insight in one embodiment;

[0047] Figure 2 A flowchart of a method for on-site remote inspection with real-time monitoring and abnormal insight in one embodiment;

[0048] Figure 3A flowchart of an on-site remote inspection method for real-time monitoring and abnormality insight in another embodiment;

[0049] Figure 4 A structural block diagram of an on-site remote inspection device for real-time monitoring and abnormal insight in one embodiment;

[0050] Figure 5 is an internal structure diagram of a computer device in one embodiment;

[0051] Figure 6 FIG. 4 is a diagram showing the internal structure of a computer device in another embodiment. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application 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 application and are not used to limit the present application.

[0053] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0054] The on-site remote inspection method for real-time monitoring and abnormal insight provided by the embodiment of the present application can be applied in the application environment shown in Figure 1. The terminal 102 communicates with the server 104 through a network.

[0055] The terminal 102 may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, IoT devices, and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, etc. The portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers.

[0056] In one embodiment, Figure 2 As shown, a method for on-site remote inspection with real-time monitoring and abnormal insight is provided, which is described by taking the method applied to the terminal 102 in FIG. 1 as an example, and includes the following steps:

[0057] S201, obtaining the service quality data corresponding to the services at each stage in the power grid enterprise platform system, and extracting data samples to be verified from the service quality data.

[0058] Among them, service quality data is used to characterize the quality of the operation of services at each stage; specifically, service quality data is important information used to measure the operation of services at each stage of power grid enterprises. This type of data includes equipment performance indicators (such as load, efficiency), fault records and detection results, service response time, etc., which can reflect the health of the business in different links. These data help managers identify potential quality problems in a timely manner, so that they can take corresponding improvement measures to improve the efficiency and security of the entire business process.

[0059] S202, obtain the quality rules preset in the quality control technology standard library, determine the qualified data samples that meet the quality rules, calculate the similarity between the data samples to be verified and the qualified data samples, and determine the remote inspection results corresponding to the business at each stage based on the similarity.

[0060] Among them, the remote inspection results are used to characterize the target equipment corresponding to each stage of the business with quality abnormalities. Specifically, the remote inspection results are based on the judgment drawn from the analysis of business quality data, aiming to identify whether there are quality abnormalities. By calculating the similarity between the data samples to be verified and the qualified data samples, the abnormal conditions of the equipment or business processes within a specific time period are determined. This result can provide an important basis for on-site management, allowing relevant personnel to quickly locate the problem area, conduct targeted inspections and maintenance, and ensure the normal operation of equipment and systems.

[0061] S203, based on the entire life cycle of the target equipment corresponding to the business at each stage, analyze the causes of quality abnormalities of each target equipment and determine the quality traceability information corresponding to each target equipment.

[0062] Among them, quality traceability information is used to determine the source equipment that causes quality abnormalities in the target equipment; specifically, quality traceability information is data obtained by in-depth analysis of the quality of the equipment throughout its life cycle. It not only includes analysis of the cause of the failure, but also involves related historical quality records, initiated maintenance measures and subsequent effects. This type of information enables companies to track the quality performance of equipment, understand the root causes of potential defects, and form a complete quality management closed loop to ensure the safety and stability of the overall power grid. Through this traceability information, companies can conduct more comprehensive risk assessments and fault prevention, and improve service quality and customer satisfaction.

[0063] In the above-mentioned on-site remote inspection method for real-time monitoring and abnormal insight, first, the quality data of the business at each stage is obtained from the power grid enterprise platform system to evaluate the business operation status and extract the data samples to be verified; then, according to the quality rules in the quality control standard library, qualified data samples are screened out, and the similarity between the samples to be verified and the qualified samples is calculated, so as to determine the remote inspection results of the business at each stage and identify potential quality abnormal equipment; then, based on the entire life cycle of the target equipment, the cause of the abnormality is analyzed and the quality traceability information is extracted, and the key features are identified using the text similarity algorithm; finally, a full-chain traceability model for the quality and safety of key equipment is constructed, combined with the graph database and knowledge graph to realize equipment quality traceability, defect analysis and timing analysis of faults; through the above-mentioned technical solution, the intelligence level of equipment management is improved, and the accuracy and real-time performance of abnormality detection are enhanced, thereby effectively ensuring the safe operation of the power grid, improving the inspection efficiency of the quality abnormality monitoring process of power equipment, and improving the inspection effect.

[0064] In one embodiment, the method also includes: based on a pre-built equipment life cycle quality data anomaly warning model, combined with the technical specification information of the engineering project, comparing the data of each business link of the equipment life cycle to determine the quality data of the target equipment; when the quality data does not meet the data requirements in the equipment life cycle quality data anomaly warning model, generating quality data anomaly warning information.

[0065] For example, the equipment life cycle quality data anomaly warning model is used, with engineering projects as the main line and combined with engineering project technical specifications, to conduct full-process and full-data flow record comparison of equipment data in each business link of engineering design, bidding and procurement, production supervision, quality inspection, and equipment operation and maintenance, to achieve in-depth analysis of the root causes and influencing factors of quality problems in the whole life cycle of key equipment. When the quality data collected in each business link violates the model regulations, the quality data anomaly warning mechanism is triggered.

[0066] In this embodiment, based on the pre-built equipment life cycle quality data abnormality warning model, the data of each business link of the engineering project is systematically compared to identify the quality data of the target equipment. First, the relevant quality standards are obtained from the project technical specifications, and the abnormality warning model is established in combination with the historical operation data of the equipment; then, the data of equipment operation is monitored in real time, and these data are compared with the standards set in the model to detect whether there are abnormalities; when it is found that the quality data does not meet the preset regulations of the model, the system automatically generates quality data abnormality warning information and notifies relevant managers to take necessary measures; through this method, enterprises can identify quality deviations in a timely manner, reduce the probability of failures, improve equipment management efficiency, and at the same time enhance the monitoring of equipment safety status, and promote continuous improvement and the implementation of operational best practices.

[0067] In one embodiment, the method also includes: extracting equipment feature values ​​according to the differences in equipment categories to construct a feature library corresponding to the equipment; constructing a knowledge graph corresponding to the equipment based on machine learning information extraction technology based on sequence annotation; constructing a full-chain traceability model for quality and safety based on the feature library corresponding to the equipment and the knowledge graph corresponding to the equipment; the full-chain traceability model for quality and safety is used to generate quality traceability information.

[0068] Exemplarily, the knowledge model is constructed through the following steps:

[0069] Feature library construction: According to the differences in device categories, device feature values ​​are sorted and extracted. At the same time, for new devices entering the network, extraction methods are performed while the device quality control is in progress.

[0070] Knowledge graph construction: Direct manual extraction based on rules to meet the needs of specific project information extraction; machine learning information extraction based on sequence annotation. Sequence annotation is a relatively basic task in natural language processing technology. In essence, word segmentation, part-of-speech tagging, named entity recognition, keyword extraction, semantic role labeling and slot extraction all belong to the category of sequence annotation. The feature sequence indexing method based on word sequence is adopted to eliminate the limitations of word segmentation for indexing, which can ensure a higher feature entity extraction result, that is, Labelset={B, M, E, S}, where: B is the starting character of the word; M is the middle character of the word; E is the ending character of the word; S is a single word.

[0071] In this embodiment, equipment feature values ​​are extracted based on the differences in equipment categories to construct an equipment feature library. First, a database containing equipment features is established by analyzing the key parameters and performance indicators of different categories of equipment. Then, based on the machine learning information extraction technology of sequence annotation, a corresponding knowledge graph of the equipment is constructed to integrate equipment features with relevant knowledge information. Subsequently, the equipment feature library is combined with the knowledge graph to construct a full-chain traceability model for quality and safety, which is used to trace equipment quality and safety records. Through this model, quality traceability information can be generated in real time to help managers gain an in-depth understanding of the quality status, failure history and risk factors of the equipment, thereby improving decision-making capabilities and emergency response efficiency. This comprehensive technical solution helps to realize the intelligent and information-based management of equipment and enhance the safety and reliability of equipment operation.

[0072] In one embodiment, the quality traceability information includes the quality traceability results of power equipment, and the method also includes: based on a forward chain multi-step reasoning method, in a goal-oriented manner, querying and displaying events in the knowledge base that have occurred and determined the root causes of defects and faults; based on the site information in the geographic wiring diagram, constructing a site wiring diagram based on the model data and power grid topology data stored in the graph database; performing multi-step query analysis based on the labels in the form of fault description text to determine the quality traceability results of the power equipment.

[0073] Exemplarily, a forward chain multi-step reasoning method is used to trace the quality of power equipment. It is goal-oriented and queries and displays an event in the knowledge base where a defect fault has occurred and the root cause has been determined. Through a certain station on the geographic wiring diagram, the wiring diagram within the station can be automatically constructed according to the CIM / E model data and the power grid topology data stored in the graph database. At the same time, the specific quality information, connection status, operating parameters and other information of the equipment will be automatically displayed. According to the label in the form of the fault description text, it is possible to realize multi-step query and analysis from the fault phenomenon to the cause based on the information such as the external environment of the equipment, the grid operating load, the equipment fault component, the fault phenomenon, the fault cause, the fault qualitative and the fault nature, so that the user can review the event and finally deduct points for the equipment quality.

[0074] In this embodiment, based on the forward chain multi-step reasoning method, the fault event is taken as the target, and its root cause is determined by querying and displaying the defect fault events in the knowledge base; first, the fault cases stored in the knowledge base are used to identify related events and their causal relationships through the reasoning mechanism; then, combined with the site information in the geographic wiring diagram, the power grid topology data is extracted from the graph database, and the wiring diagram within the site is constructed to ensure the visualization of the logical relationship; finally, multi-step query analysis is performed based on the text label of the fault description, and the fault information and electrical wiring diagram data are integrated to determine the quality traceability results of the power equipment; this solution effectively improves the accuracy and efficiency of fault diagnosis, allowing managers to quickly trace the root cause of the fault and its impact, and better prevent and maintain faults, thereby improving the safety and reliability of the power system.

[0075] In one embodiment, the quality traceability information includes the results of suspected familial defect analysis, and the method further includes: based on the high-frequency subgraph mining method, analyzing the correlation between the defect fault information and the problem root cause information corresponding to each equipment product; analyzing events of the same type as the target event to achieve probability ranking information for suspected familial defects of the equipment; the target event is a fault or defect event with a determined cause; and determining the suspected familial defect analysis results based on the probability ranking information.

[0076] For example, the high-frequency subgraph mining method is used to analyze suspected family defects, analyze the close relationship between the products of manufacturers and parts suppliers and defect failure phenomena, root causes of problems, etc., analyze similar cause events for failures and defect events with determined causes, and realize the probability of suspected family defects of equipment and sort them, so that users can judge the potential defects and failures of other related equipment. It is also an important basis for supplier evaluation. Suspected family defect analysis can be displayed on relative geographic coordinates (or maps) through the distribution of sites of a certain type of equipment of a certain manufacturer. According to the specific description of the quality event, extract keywords and generate quality event labels, such as oil leakage, temperature anomaly, deliquesce discoloration, etc.

[0077] In this embodiment, based on the high-frequency subgraph mining method, it aims to reveal the correlation between the defect fault information and the root cause of the problem of different equipment products; first, by constructing a knowledge graph containing fault and defect cases, the high-frequency subgraph mining algorithm is applied to extract frequently occurring patterns and relationships between events; then, cluster analysis is performed on events of the same type as the target event (fault or defect event with a determined cause), so as to achieve the probability ranking of suspected familial defects; by calculating the similarity of each event with the target event at the root cause and integrating the results, a ranking information of suspected familial defects is formed; finally, based on these probability ranking results, events with similar fault characteristics are identified, and analysis conclusions of suspected familial defects are provided; this method effectively improves the accuracy of fault location, helps enterprises quickly identify and solve potential equipment defect problems, thereby reducing the risk of faults and improving the reliability and safety of equipment operation.

[0078] In one embodiment, the similarity between the data sample to be verified and the qualified data sample is calculated, including: representing the text as a vector of words and using a combination of word vectors to represent the text; encoding the label through Huffman coding and extracting text features; and calculating the similarity between the data sample to be verified and the qualified data sample based on the word frequency in the statistical corpus.

[0079] Exemplarily, the implementation idea of ​​text similarity recognition adopts the fastText architecture, which includes three parts: model architecture, hierarchical SoftMax and N-gram features. The fastText model uses the hierarchical Softmax technique, which is based on Huffman coding to encode labels, which can greatly reduce the number of model prediction targets. FastText uses the fact of category imbalance to establish a tree structure for characterizing categories by using the Huffman algorithm. The depth of the tree structure of frequently appearing categories is smaller than that of the tree structure of infrequently appearing categories, which also makes further calculations more efficient. In text feature extraction, an algorithm based on a language model, n-gram features, is used, based on an assumption that the appearance of the nth word is related to the previous n-1 words, but not to any other words. The probability of the entire sentence appearing is equal to the product of the probabilities of each word appearing. The probability of each word can be obtained by statistical calculation in the corpus.

[0080] In this embodiment, first, the text is converted into word vectors, and the word vectors are combined to form an overall representation of the text to capture the semantic information of the text; then, Huffman coding is applied to encode the labels to achieve efficient text feature extraction, thereby reducing data redundancy and improving storage efficiency; finally, based on the word frequency information of the statistical corpus, the similarity between the data sample to be verified and the qualified data sample is calculated. This process takes into account the frequency and position of the vocabulary to ensure the accuracy and relevance of the similarity measurement; through this solution, the efficiency of data quality review can be effectively improved, and reliable data support can be provided for subsequent analysis and decision-making, thereby reducing potential error risks and improving the overall intelligence level of the system.

[0081] In another embodiment, Figure 3 As shown, a method for on-site remote inspection with real-time monitoring and abnormal insight is provided, including the following steps:

[0082] S301, based on the pre-built equipment life cycle quality data abnormal warning model, combined with the project technical specification information, compare the data of each business link of the equipment life cycle to determine the quality data of the target equipment.

[0083] S302, when the quality data does not comply with the data requirements in the equipment life cycle quality data abnormality warning model, generate quality data abnormality warning information.

[0084] S303, based on the forward chain multi-step reasoning method, is goal-oriented and queries and displays events in the knowledge base where defect failures have occurred and the root causes have been determined.

[0085] S304, based on the station information in the geographic wiring diagram, a station internal wiring diagram is constructed according to the model data and the power grid topology data stored in the graph database.

[0086] S305, performing multi-step query analysis based on the tags in the form of fault description text to determine the quality tracing result of the power equipment.

[0087] S306, based on the high-frequency subgraph mining method, analyzing the correlation between the defect failure information and the problem root cause information corresponding to each device product.

[0088] S307, analyzing events of the same type as the target event to obtain probability ranking information for suspected familial defects of the equipment.

[0089] S308, determining the suspected familial defect analysis result according to the probability ranking information.

[0090] It should be noted that the specific limitations of the above steps can be found in the above specific limitations of a real-time monitoring and abnormal insight on-site remote inspection method, which will not be repeated here.

[0091] In some embodiments, (1) data has the problem of multi-source heterogeneous data. In addition to the equipment manufacturer, the entire life cycle of the equipment also involves the equipment's raw material and parts suppliers, design units, supervision units, testing units, on-site installation and construction units, operation and maintenance service units, transportation monitoring units, etc. However, due to the differences in system architecture, data format and management methods of different units, the data is diverse and complex. (2) The problem of full-domain application of physical coding. The introduction of a physical coding system requires a large amount of capital to purchase and maintain related equipment, and the production process needs to be transformed to adapt to the new coding system. These cost investments may be a considerable burden for many companies. Therefore, before fully verifying the benefits of the physical coding system, companies tend to be cautious about its actual application.

[0092] Based on this, the embodiment of the present application provides a real-time monitoring and abnormal insight on-site remote inspection method, and the real-time monitoring and abnormal insight on-site remote inspection method is described in detail with a specific embodiment below. It is worth understanding that the following description is only an exemplary description, and not a specific limitation of the application.

[0093] The on-site remote inspection method with real-time monitoring and abnormal insight provided in this application is built around remote on-site remote inspection with real-time monitoring and abnormal insight, to avoid or reduce the occurrence of equipment quality accidents, and to provide timely feedback and presentation of equipment quality inspection results, causes of occurrence, and closed-loop problem handling.

[0094] The on-site remote inspection method with real-time monitoring and abnormal insight provided by the present application constructs quality elements of the whole-chain quality data model including procurement, production scheduling, manufacturing supervision, factory testing, material sampling, delivery acceptance, installation and commissioning, operation and maintenance, and decommissioning of the design equipment, and uses the quality control technical standards for remote inspection specifications to achieve the purpose of auxiliary inspection, real-time monitoring and abnormal insight through direct data collection from the supplier, remote consultation, AI intelligent diagnosis and other means; outlines the overall quality performance of the supplier's supply categories through the quality status of the equipment; and achieves automatic marking of equipment quality problems, automatic verification of factory reports and supervision summary reports through labeling means, and integrates stage quality data integration, process quality problem monitoring, and early insight into abnormal problems into a quality defect intelligent inspection system.

[0095] Description of a method for on-site remote inspection with real-time monitoring and abnormal insight:

[0096] The main technical means used in this application are text similarity recognition, big data analysis, artificial intelligence, and blockchain. By building an on-site remote inspection system with real-time monitoring and abnormal insight, the business covered includes the power industry supply chain, infrastructure, systems and other fields. Through direct data acquisition from the supplier side, remote consultation, AI intelligent diagnosis and other means, the purpose of auxiliary inspection, real-time monitoring and abnormal insight is achieved. Through the quality of the equipment, the overall quality performance of the supplier's supply category is outlined, and through labeling means, the automatic labeling of equipment quality problems, automatic verification of factory reports and supervision summary reports are achieved. The intelligent inspection system for quality defects integrates stage quality data integration, process quality problem monitoring, and early insight into abnormal problems.

[0097] 1. Construction of special quality model

[0098] The special quality data model of this application is based on the business involved in the daily operations in the power grid industry. It collects and analyzes the characteristics of business quality data at each stage, uses the text similarity fastText algorithm to extract the text features of each data, and cleans, preprocesses and extracts features from these data to ensure the quality and availability of the data.

[0099] Then, build a model that meets the requirements based on actual business needs and rules, and continuously adjust the model parameters and hyperparameters to optimize the model's performance.

[0100] Through the special quality data verification and analysis model, the "man-machine-material-method cycle" related to product design, procurement, production scheduling, supervision, sampling inspection, acceptance, installation and commissioning, operation and maintenance, and decommissioning related to quality control equipment is used to determine whether there is a trend of related quality abnormalities or even quality accidents.

[0101] The data supply chain of each link is obtained through a unified platform, and the corresponding quality data samples for verification are extracted. Through the text similarity fastText algorithm, the consistency is compared with the rules defined in the quality control technology standard library to form the corresponding remote inspection results. The quality control technology standard library includes the definition of rule items, rule content, and the upper and lower limits of the technical parameters of the rules. The definition of the rule library must be preset in advance according to the specific business. The construction of the special quality model is to provide dynamic evaluation and quality performance services for suppliers and supervision and inspection agencies in the future.

[0102] This application adopts the means of text similarity recognition. The content extracted from the power grid enterprise platform system includes the business involved in daily operations, and collects and analyzes the business quality data of each stage, such as procurement information (bidding, process planning, purchase orders and other procurement information), production scheduling information (including production scheduling plan, production scheduling equipment and other scheduling information), manufacturing and supervision (including production process records, supervision witness information, quality warning information), factory testing (including factory testing records, supervision witness information, quality warning information), material sampling (including sampling sheets, test records, quality defect information), delivery acceptance (delivery acceptance records, quality warning information), installation and commissioning (including installation and commissioning records, quality warning information), operation and maintenance (operation and maintenance records, defect sheets, maintenance and test rules), decommissioning and scrapping (weight difference, material difference and other data), etc., extract the corresponding verification data samples (such as: key data features of safety measures, etc.), and obtain the key information of quality inspection through the text similarity algorithm.

[0103] The implementation idea of ​​text similarity recognition adopts the fastText architecture, which includes three parts: model architecture, hierarchical SoftMax and N-gram features. The fastText model uses the hierarchical Softmax technique, which is based on Huffman coding to encode labels, which can greatly reduce the number of model prediction targets. FastText uses the fact of category imbalance to establish a tree structure for characterizing categories by using the Huffman algorithm. The depth of the tree structure of frequently appearing categories is smaller than that of the tree structure of infrequently appearing categories, which also makes further calculations more efficient. In text feature extraction, an algorithm based on a language model, n-gram features, is used, based on an assumption that the appearance of the nth word is related to the previous n-1 words, but not to any other words. (This is also an assumption in hidden Markov.) The probability of the entire sentence is equal to the product of the probabilities of each word. The probability of each word can be obtained by statistical calculation in the corpus. Assuming that the sentence T is composed of a word sequence w1, w2, w3...wn, the N-Gram language model is expressed as follows:

[0104] P(T)=P(w1)*p(w2)*p(w3)***p(wn)=p(w1)*p(w2|w1)*p(w3|w1w2)***p(wn|w1w2w3...). The probability is calculated as follows: P(w1|begin)=all sentences starting with w1 / total number of sentences; p(w2|w1)=number of times w1 and w2 appear at the same time / number of times w1 appears.

[0105] 2. Research on the whole chain traceability model of quality and safety

[0106] On the basis of building special quality models, based on physical ID, big data, blockchain and other technologies as well as the entire life cycle of equipment, it provides information support for defect analysis and quality problem handling.

[0107] Solve the problem of heterogeneous source data and realize the application of physical coding in the entire process of equipment safety production.

[0108] Based on the multi-dimensional equipment life cycle management, the equipment is traced throughout the entire chain, taking into account the quality management requirements of the equipment throughout its life cycle, as well as the defects and quality problems in the supervision during the quality control management process and the inspection process at each stage.

[0109] Data is the core of the process. Data is collected, managed and analyzed throughout the equipment life cycle. From raw material procurement to manufacturing, and then to the testing and delivery of the final product, relevant data is recorded in detail at every stage. These data include but are not limited to raw material batches, production process parameters, equipment operating status, quality inspection results, etc. Only by comprehensively and accurately collecting these data can a reliable basis be provided for subsequent quality traceability.

[0110] At the same time, in-depth mining and analysis of the collected data can help discover the root causes and potential risks of quality problems, providing strong support for quality improvement and preventive measures.

[0111] Specifically include:

[0112] The core of the research on the full-chain traceability model of quality and safety is to build a full-chain traceability model based on graph database, knowledge graph and full life cycle equipment quality and safety to realize the traceability of equipment quality, analysis of family defects and timing analysis of failures. Based on the construction foundation of the special quality model, the procurement, production supervision, testing, logistics, operation, maintenance and other links of power equipment are expanded to play a phased role in the performance of quality defects in the entire life cycle management chain. Organizing various types of data in the chain, integrating the production, manufacturing, logistics information of internal and external related equipment and components with open social network information, can fully explore the potential knowledge associations that affect the quality and cost of power equipment, provide reasonable and accurate quality assessments, and form customized equipment standards for special operating environments such as regional climate, natural disasters, and large load fluctuations, thereby extending the service life of the equipment and reducing the number of maintenance times;

[0113] (1) Knowledge model construction

[0114] The knowledge model is constructed through the following steps:

[0115] Feature library construction: According to the differences in device categories, device feature values ​​are sorted and extracted. At the same time, for new devices entering the network, extraction methods are performed while the device quality control is in progress.

[0116] Knowledge graph construction: Direct manual extraction based on rules to meet the needs of specific project information extraction; machine learning information extraction based on sequence annotation. Sequence annotation is a relatively basic task in natural language processing technology. In essence, word segmentation, part-of-speech tagging, named entity recognition, keyword extraction, semantic role labeling and slot extraction all belong to the category of sequence annotation. The feature sequence indexing method based on word sequence is adopted to eliminate the limitations of word segmentation for indexing, which can ensure a higher feature entity extraction result, that is, Labelset={B, M, E, S}, where: B is the starting character of the word; M is the middle character of the word; E is the ending character of the word; S is a single word.

[0117] (2) Model application

[0118] Use the community method to classify power equipment: classify the faults and defect events that occur in certain equipment that are specified based on multiple conditions such as equipment type, voltage level, and operating environment, and explore the potential relationship between the root cause of the problem and the phenomenon, so that users can quickly analyze equipment defects and locate the root cause of the problem. All faults and defect events covered in the database can be filtered according to the equipment's commissioning time, equipment type, manufacturer, and event type, and the site where the quality event occurred can be displayed on the relative geographic coordinates (or map). The filtered results are secondary classified according to the classification of quality event failures and non-stops and the degree of general, serious, and critical quality defects to help users optimize equipment operation and maintenance and repair strategies.

[0119] The forward chain multi-step reasoning method is used to trace the quality of power equipment: with a goal-oriented approach, a certain event in the knowledge base where a defect fault has occurred and the root cause has been determined is queried and displayed. Through a certain station on the geographic wiring diagram, the wiring diagram within the station can be automatically constructed based on the CIM / E model data and the power grid topology data stored in the graph database. At the same time, the specific quality information, connection status, operating parameters and other information of the equipment will be automatically displayed. According to the label in the form of the fault description text, it is possible to realize multi-step query and analysis from the fault phenomenon to the cause based on the external environment of the equipment, the grid operating load, the equipment fault component, the fault phenomenon, the fault cause, the fault qualitative and the fault nature, so that users can review the event and finally deduct points for the equipment quality.

[0120] Use high-frequency subgraph mining method to analyze suspected family defects: analyze the close relationship between the products of manufacturers and parts suppliers and defect failure phenomena, root causes of problems, etc., analyze similar cause events for failure and defect events with confirmed causes, realize the probability of suspected family defects of equipment and sort them, so that users can judge the potential defect failure of other related equipment, and it is also an important basis for supplier evaluation. Suspected family defect analysis can be displayed on relative geographic coordinates (or maps) through the distribution of sites of a certain type of equipment of a certain manufacturer. Extract keywords based on the specific description of the quality event and generate quality event labels, such as oil leakage, temperature anomaly, deliquescence discoloration, etc.

[0121] 3. Research on the whole chain traceability model of quality and safety

[0122] The research on abnormal warning model of quality data is based on the ID coding and technical requirements of physical power. Through the research on data of various business links in the whole life cycle and the research on quality characteristic data, firstly, the artificial intelligence technologies such as statistical analysis, correlation analysis and machine learning are used for fusion analysis and deep mining, and the multi-perspective and multi-dimensional key equipment status characteristic data are extracted to build the historical data model framework of key equipment; then, the correlation and change rules of various status data are analyzed, and a large amount of sample data is used for intelligent learning to revise, supplement and improve the existing physical and mathematical models of status evaluation and fault diagnosis. The construction of abnormal warning model of quality data of key equipment throughout the life cycle driven by multi-source data based on technical requirements, historical data and sustainable iteration is realized.

[0123] On the basis of building special quality models, based on physical ID, big data, blockchain and other technologies as well as the entire life cycle of equipment, it provides information support for defect analysis and quality problem handling.

[0124] Solve the problem of heterogeneous source data and realize the application of physical coding in the entire process of equipment safety production.

[0125] Based on the multi-dimensional equipment life cycle management, the equipment is traced throughout the entire chain, taking into account the quality management requirements of the equipment throughout its life cycle, as well as the defects and quality problems in the supervision during the quality control management process and the inspection process at each stage.

[0126] Data is the core of the process. Data is collected, managed and analyzed throughout the equipment life cycle. From raw material procurement to manufacturing, and then to the testing and delivery of the final product, relevant data is recorded in detail at every stage. These data include but are not limited to raw material batches, production process parameters, equipment operating status, quality inspection results, etc. Only by comprehensively and accurately collecting these data can a reliable basis be provided for subsequent quality traceability.

[0127] At the same time, in-depth mining and analysis of the collected data can help discover the root causes and potential risks of quality problems, providing strong support for quality improvement and preventive measures.

[0128] Specifically, it includes the use of an abnormal early warning model for quality data throughout the life cycle of equipment. Taking engineering projects as the main line and combining them with the technical specifications of engineering projects, the data of equipment in each business link of engineering design, bidding and procurement, production supervision, quality inspection, and equipment operation and maintenance are compared in the whole process and all data stream records, so as to achieve in-depth analysis of the root causes and influencing factors of quality problems in the whole life cycle of key equipment. When the quality data collected in each business link violates the model regulations, the quality data abnormal early warning mechanism is triggered.

[0129] The key points of the on-site remote inspection method for real-time monitoring and abnormal insight provided by this application are:

[0130] The data has the problem of multi-source heterogeneous data. In addition to the equipment manufacturer, the entire life cycle of the equipment also involves the equipment's raw material and parts suppliers, design units, supervision units, testing units, on-site installation and construction units, operation and maintenance service units, transportation monitoring units, etc. However, due to the differences in system architecture, data format and management methods of different units, the data is diverse and complex.

[0131] The difficulty of applying physical coding in the whole domain is that the introduction of physical coding system requires a large amount of money to purchase and maintain related equipment, and the production process needs to be transformed to adapt to the new coding system. These cost investments may be a considerable burden for many companies, so before fully verifying the benefits of physical coding system, companies tend to be cautious about its actual application.

[0132] The advantages of the on-site remote inspection method for real-time monitoring and abnormal insight provided by this application are:

[0133] Text similarity recognition is used to extract data samples corresponding to quality inspections (such as key data features of safety measures, etc.), and key information for quality inspections is obtained through text similarity algorithms.

[0134] Research on the full-chain traceability model for quality and safety, build a full-chain traceability model for key equipment quality and safety based on graph database, knowledge graph and the entire life cycle of equipment, to achieve equipment quality traceability, family defect analysis and fault timing analysis.

[0135] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0136] Based on the same inventive concept, the embodiment of the present application also provides a real-time monitoring and abnormal insight on-site remote inspection device for implementing the real-time monitoring and abnormal insight on-site remote inspection method involved in the above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more real-time monitoring and abnormal insight on-site remote inspection device embodiments provided below can refer to the limitations of the real-time monitoring and abnormal insight on-site remote inspection method above, and will not be repeated here.

[0137] In one embodiment, Figure 4 As shown, a field remote inspection device for real-time monitoring and abnormal insight is provided, including: a data sample acquisition module 401, a remote inspection result determination module 402, and a quality tracing module 403, wherein:

[0138] The data sample acquisition module 401 is used to obtain the service quality data corresponding to the services at each stage in the power grid enterprise platform system, and extract the data samples to be verified from the service quality data; the service quality data is used to characterize the quality of the operation status of the services at each stage;

[0139] The remote inspection result determination module 402 is used to obtain the quality rules preset in the quality control technology standard library, determine the qualified data samples that meet the quality rules, calculate the similarity between the data samples to be verified and the qualified data samples, and determine the remote inspection results corresponding to the business at each stage according to the similarity; the remote inspection results are used to characterize the target equipment corresponding to the business at each stage with quality abnormalities;

[0140] The quality traceability module 403 is used to analyze the causes of quality anomalies of each target device based on the entire life cycle of the target device corresponding to each stage of business, and determine the quality traceability information corresponding to each target device; the quality traceability information is used to determine the source device that causes the quality anomaly of the target device.

[0141] In one embodiment, the device is used to: based on a pre-built equipment life cycle quality data abnormality warning model, combined with the technical specification information of the engineering project, compare the data of each business link of the equipment life cycle to determine the quality data of the target equipment; when the quality data does not meet the data requirements in the equipment life cycle quality data abnormality warning model, generate quality data abnormality warning information.

[0142] In one embodiment, the device is used to: extract equipment feature values ​​according to the differences in equipment categories to construct a feature library corresponding to the equipment; construct a knowledge graph corresponding to the equipment based on machine learning information extraction technology based on sequence annotation; construct a full-chain traceability model for quality and safety based on the feature library corresponding to the equipment and the knowledge graph corresponding to the equipment; the full-chain traceability model for quality and safety is used to generate quality traceability information.

[0143] In one embodiment, the device is used to: based on a forward-chain multi-step reasoning method, in a goal-oriented manner, query and display events in the knowledge base that have occurred and determined the root causes of defects and faults; based on the site information in the geographic wiring diagram, construct a site wiring diagram based on the model data and power grid topology data stored in the graph database; perform multi-step query analysis based on the labels in the form of fault description text to determine the quality tracing results of power equipment.

[0144] In one embodiment, the device is used to: analyze the correlation between the defect fault information and the problem root cause information corresponding to each equipment product based on the high-frequency subgraph mining method; analyze the events of the same type as the target event to realize the probability ranking information for the suspected familial defects of the equipment; the target event is a fault or defect event with a determined cause; and determine the suspected familial defect analysis results based on the probability ranking information.

[0145] In one embodiment, the device is used to: represent text as vectors of words and use a combination of word vectors to represent text; encode tags through Huffman coding to extract text features; and calculate the similarity between the data sample to be verified and the qualified data sample based on the word frequency in the statistical corpus.

[0146] Each module in the above-mentioned on-site remote inspection device for real-time monitoring and abnormal insight can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0147] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a field remote inspection method with real-time monitoring and abnormal insight is realized.

[0148] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 6As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a method for on-site remote inspection with real-time monitoring and abnormal insight is realized. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0149] Those skilled in the art will understand that Figure 5 and Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0150] In one embodiment, a computer device includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, the steps in the above method embodiments are implemented.

[0151] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0152] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0153] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0154] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0155] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for on-site remote inspection with real-time monitoring and abnormal insight, characterized in that: The method comprises: Obtaining service quality data corresponding to services at each stage in the power grid enterprise platform system, and extracting data samples to be verified from the service quality data; the service quality data is used to characterize the quality of the operation status of the services at each stage; Acquire the quality rules preset in the quality control technology standard library, determine the qualified data samples that meet the quality rules, calculate the similarity between the data samples to be verified and the qualified data samples, and determine the remote inspection results corresponding to the business at each stage according to the similarity; the remote inspection results are used to characterize the target equipment corresponding to the business at each stage with abnormal quality; Based on the entire life cycle of the target equipment corresponding to the business at each stage, the causes of quality abnormalities in each target equipment are analyzed, and quality traceability information corresponding to each target equipment is determined; the quality traceability information is used to determine the source equipment that causes the quality abnormalities in the target equipment.

2. The method according to claim 1, characterized in that The method further comprises: Based on the pre-built equipment life cycle quality data abnormality warning model, combined with the project technical specification information, the data of each business link of the equipment life cycle are compared to determine the quality data of the target equipment; When the quality data does not conform to the data regulations in the equipment life cycle quality data abnormality warning model, quality data abnormality warning information is generated.

3. The method according to claim 1, characterized in that The method further comprises: According to the differences in equipment categories, the equipment feature values ​​are extracted to build a feature library corresponding to the equipment; Build the knowledge graph corresponding to the device based on machine learning information extraction technology based on sequence annotation; According to the feature library corresponding to the device and the knowledge graph corresponding to the device, a full-chain traceability model for quality and safety is constructed; the full-chain traceability model for quality and safety is used to generate the quality traceability information.

4. The method according to claim 1, characterized in that: The quality traceability information includes a quality traceability result of the power equipment, and the method further includes: Based on the forward chain multi-step reasoning method, the system queries and displays the events of defect failures that have occurred in the knowledge base and determines the root causes in a goal-oriented manner; Based on the station information in the geographic wiring diagram, the wiring diagram within the station is constructed according to the model data and power grid topology data stored in the graph database; A multi-step query analysis is performed based on the tags in the form of fault description text to determine the quality tracing result of the power equipment.

5. The method according to claim 1, characterized in that The quality traceability information includes a suspected familial defect analysis result, and the method further includes: Based on the high-frequency subgraph mining method, the correlation between the defect failure information and the problem root cause information corresponding to each equipment product is analyzed; Analyze events of the same type as the target event to achieve probability ranking information for suspected family defects of the equipment; the target event is a fault or defect event with a determined cause; The suspected familial defect analysis result is determined according to the probability ranking information.

6. The method according to claim 1, characterized in that The calculating the similarity between the data sample to be verified and the qualified data sample includes: By representing text as word vectors, a combination of word vectors is used to represent text; Encode the labels through Huffman coding and extract text features; Based on the word frequency in the statistical corpus, the similarity between the data sample to be verified and the qualified data sample is calculated.

7. A remote inspection device for real-time monitoring and abnormal insight, characterized in that: The device comprises: A data sample acquisition module is used to obtain the service quality data corresponding to the services at each stage in the power grid enterprise platform system, and extract the data samples to be verified from the service quality data; the service quality data is used to characterize the quality of the operation status of the services at each stage; A remote inspection result determination module is used to obtain the quality rules preset in the quality control technology standard library, determine the qualified data samples that meet the quality rules, calculate the similarity between the data samples to be verified and the qualified data samples, and determine the remote inspection results corresponding to the business at each stage according to the similarity; the remote inspection results are used to characterize the target equipment corresponding to the business at each stage with abnormal quality; The quality traceability module is used to analyze the causes of quality abnormalities of each target device based on the entire life cycle of the target device corresponding to the business at each stage, and determine the quality traceability information corresponding to each target device; the quality traceability information is used to determine the source device that causes the quality abnormality of the target device.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images