Power equipment management method and system based on Internet of Things

Abstract

This application provides a power equipment management method and system based on the Internet of Things. Through technical means such as state dense semantic mining and bidirectional vector annotation, it can not only deeply mine the rich semantic information in the operation and maintenance report, but also realize the precise positioning and linkage analysis of operation and maintenance alarm messages. This innovative technical solution will greatly improve the intelligent level of power equipment operation and maintenance, and provide an effective technical approach to solve the various drawbacks of traditional operation and maintenance methods.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to an Internet of Things-based power equipment management method and system. Background Art

[0002] In the field of power equipment operation and maintenance, the traditional way of processing operation and maintenance reports often relies on manual interpretation and analysis, which is not only inefficient, but also prone to misinterpretation or omission of information due to human factors. Especially when processing operation and maintenance alarm messages, due to the lack of effective automated tools, operation and maintenance personnel often find it difficult to quickly and accurately locate problems, thus affecting the stable operation of the power system.

[0003] In recent years, with the development of natural language processing and machine learning technologies, automated processing of power equipment operation and maintenance reports has become possible. However, most existing automated processing methods can only extract shallow information from reports and cannot deeply explore the rich semantics in reports, resulting in an insufficient understanding of the operation and maintenance status. In addition, when processing operation and maintenance alarm messages, existing methods can often only analyze a single alarm message and cannot achieve linkage positioning between alarm messages, which greatly limits the improvement of operation and maintenance efficiency. Summary of the invention

[0004] The present application at least provides a method and system for managing power equipment based on the Internet of Things.

[0005] The technical solution of the present application is achieved through at least some of the following embodiments.

[0006] An Internet of Things-based power equipment management method, applied to a power Internet of Things management system, the method comprising:

[0007] Performing state dense semantic mining on the target power equipment operation and maintenance report to obtain a dense semantic set of edge-side system operation and maintenance state of the target power equipment operation and maintenance report; the target power equipment operation and maintenance report includes a target operation and maintenance alarm message;

[0008] Extracting the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message and the basic keyword vector of the operation and maintenance alarm message keyword according to the dense semantic set of the equipment operation and maintenance status; the operation and maintenance alarm message keyword is a representative field representing the target operation and maintenance alarm message;

[0009] Performing vector annotation on the basic operation and maintenance alarm message vector based on the basic keyword vector to obtain a target operation and maintenance alarm message vector, and performing vector annotation on the basic keyword vector based on the basic operation and maintenance alarm message vector to obtain a target keyword vector;

[0010] Based on the target operation and maintenance alarm message vector and the target keyword vector, the target operation and maintenance alarm message in the target power equipment operation and maintenance report is linked to the operation and maintenance alarm message.

[0011] Preferably, the extracting the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message and the basic keyword vector of the operation and maintenance alarm message keyword according to the dense semantic set of the equipment operation and maintenance status includes:

[0012] Performing distribution vector extraction on the dense semantic set of the equipment operation and maintenance status to determine a local state semantic relationship network corresponding to the target operation and maintenance alarm message in the dense semantic set of the equipment operation and maintenance status;

[0013] Extracting the operation and maintenance alarm message vector set corresponding to the local state semantic relationship network from the equipment operation and maintenance state dense semantic set;

[0014] The distribution vector is refined on the operation and maintenance alarm message vector set to obtain a basic operation and maintenance alarm message vector of the target operation and maintenance alarm message and a basic keyword vector of the operation and maintenance alarm message keyword.

[0015] Preferably, the extracting distribution vectors of the dense semantic set of the equipment operation and maintenance status to determine the local state semantic relationship network corresponding to the target operation and maintenance alarm message in the dense semantic set of the equipment operation and maintenance status includes:

[0016] Performing state dense semantic mining on the dense semantic set of the equipment operation and maintenance state, identifying x initial semantic masks corresponding to the dense semantic set of the equipment operation and maintenance state and the confidence weights corresponding to each initial semantic mask; x is a positive integer;

[0017] Determine the initial semantic mask with the largest confidence weight among the x initial semantic masks as the basic local state semantic relationship network;

[0018] The semantic feature recognition degree of the basic local state semantic relationship network is updated to determine the local state semantic relationship network.

[0019] Preferably, the step of performing vector marking on the basic operation and maintenance alarm message vector based on the basic keyword vector to obtain a target operation and maintenance alarm message vector, and performing vector marking on the basic keyword vector based on the basic operation and maintenance alarm message vector to obtain a target keyword vector includes:

[0020] Integrating the basic operation and maintenance alarm message vector and the basic keyword vector to obtain a semantic feature commonality score between the basic operation and maintenance alarm message vector and the basic keyword vector;

[0021] Obtaining an operation and maintenance alarm message enhancement vector from the basic keyword vector according to the semantic feature commonality score, and performing vector annotation on the basic operation and maintenance alarm message vector according to the operation and maintenance alarm message enhancement vector to obtain a target operation and maintenance alarm message vector corresponding to the basic operation and maintenance alarm message vector;

[0022] A field enhancement vector is obtained from the basic operation and maintenance alarm message vector according to the semantic feature commonality score, and a vector annotation is performed on the basic keyword vector according to the field enhancement vector to obtain a target keyword vector corresponding to the basic keyword vector.

[0023] Preferably, the integrating the basic operation and maintenance alarm message vector and the basic keyword vector to obtain a semantic feature commonality score between the basic operation and maintenance alarm message vector and the basic keyword vector includes:

[0024] Performing feature mapping on the basic operation and maintenance alarm message vector and the basic keyword vector respectively to obtain an operation and maintenance alarm message mapping vector of a first specified size and a keyword mapping vector of a second specified size;

[0025] A weighted result of the operation and maintenance alarm message mapping vector and the keyword mapping vector is determined as a semantic feature commonality score between the basic operation and maintenance alarm message vector and the basic keyword vector.

[0026] Preferably, the extracting the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message and the basic keyword vector of the operation and maintenance alarm message keyword according to the dense semantic set of the equipment operation and maintenance status includes:

[0027] Based on the operation and maintenance alarm message extraction branch in the power equipment operation and maintenance report analysis network, the dense semantic set of the equipment operation and maintenance state is mined to obtain the operation and maintenance alarm message vector set corresponding to the target operation and maintenance alarm message;

[0028] Inputting the operation and maintenance alarm message vector set into the operation and maintenance alarm message positioning branch in the power equipment operation and maintenance report analysis network, and refining the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message according to the operation and maintenance alarm message positioning branch;

[0029] The operation and maintenance alarm message vector set is input into a representative field positioning branch in the power equipment operation and maintenance report analysis network, and a basic keyword vector of the operation and maintenance alarm message keyword is refined according to the representative field positioning branch.

[0030] Preferably, the performing linkage operation and maintenance alarm message positioning on the target operation and maintenance alarm message in the target power equipment operation and maintenance report based on the target operation and maintenance alarm message vector and the target keyword vector includes:

[0031] According to the operation and maintenance alarm message positioning branch in the power equipment operation and maintenance report analysis network, the target operation and maintenance alarm message vector is refined by distribution vector, the target operation and maintenance alarm message label to which the target operation and maintenance alarm message belongs and the operation and maintenance alarm text set of the target operation and maintenance alarm message are determined, and the target operation and maintenance alarm message label and the operation and maintenance alarm text set are determined as the involved text of the target operation and maintenance alarm message;

[0032] According to the representative field positioning branch in the power equipment operation and maintenance report analysis network, the target keyword vector is refined by distribution vector, the fault error reporting features corresponding to y fault error reporting topics are determined, and the y fault error reporting features are determined as keyword attributes of the operation and maintenance alarm message keyword; y is a positive integer, and the y fault error reporting topics are the discrimination possibilities of the representative field positioning branch;

[0033] Based on the involved text and the keyword attributes, a linkage operation and maintenance alarm message positioning is performed on the target operation and maintenance alarm message in the target power equipment operation and maintenance report.

[0034] Preferably, the number of keywords in the operation and maintenance alarm message is z; z is a positive integer;

[0035] The determining of the y fault error features as keyword attributes of the operation and maintenance alarm message keywords includes: based on the operation and maintenance alarm message keywords related to the y fault error topics, disassembling the fault error features corresponding to the y fault error topics into z operation and maintenance fault record sets; each operation and maintenance fault record set corresponds to an operation and maintenance alarm message keyword; weighting the fault error features in each operation and maintenance fault record set to obtain the keyword attribute corresponding to each operation and maintenance alarm message keyword;

[0036] The method of locating the target operation and maintenance alarm message in the target power equipment operation and maintenance report based on the involved text and the keyword attributes comprises: locating the target operation and maintenance alarm message in the target power equipment operation and maintenance report based on the involved text and the keyword attributes corresponding to each operation and maintenance alarm message keyword.

[0037] Preferably, the step of performing linkage operation and maintenance alarm message positioning on the target operation and maintenance alarm message in the target power equipment operation and maintenance report based on the involved text and the keyword attributes corresponding to each operation and maintenance alarm message keyword includes:

[0038] Obtaining safety hazard opinions related to the involved text;

[0039] If there is a keyword attribute matching the safety hazard viewpoint in the keyword attributes corresponding to each operation and maintenance alarm message keyword, it is determined that the target operation and maintenance alarm message in the target power equipment operation and maintenance report is a linkage operation and maintenance alarm message, and a safety hazard early warning prompt is sent to the target edge side system; the target edge side system is the edge side system that sends the target power equipment operation and maintenance report;

[0040] If the keyword attribute corresponding to each operation and maintenance alarm message keyword does not match the safety hazard viewpoint, it is determined that the target operation and maintenance alarm message in the target power equipment operation and maintenance report is a non-emergency operation and maintenance alarm message.

[0041] Preferably, the y fault error reporting features include a fault trend description and a hidden danger triggering possibility corresponding to each fault error reporting theme;

[0042] The method of performing linkage operation and maintenance alarm message positioning on the target operation and maintenance alarm message in the target power equipment operation and maintenance report based on the involved text and the keyword attribute corresponding to each operation and maintenance alarm message keyword includes:

[0043] Obtaining z keyword confidence coefficients associated with the involved text; each keyword confidence coefficient corresponds to an operation and maintenance alarm message keyword;

[0044] Obtain an emergency fault error reporting topic corresponding to each operation and maintenance alarm message keyword, and obtain a hidden danger triggering possibility corresponding to each emergency fault error reporting topic from a keyword attribute corresponding to each operation and maintenance alarm message keyword;

[0045] Adjusting the hidden danger triggering possibility corresponding to each emergency fault error reporting topic according to the z keyword confidence coefficients to obtain the urgency of the target power equipment operation and maintenance report;

[0046] If the urgency level is greater than or equal to the urgency level threshold, determining that the target operation and maintenance alarm message in the target power equipment operation and maintenance report is a linkage operation and maintenance alarm message, and sending a safety hazard warning prompt to the target edge side system; the target edge side system is the edge side system that sends the target power equipment operation and maintenance report;

[0047] If the urgency level is less than the urgency level threshold, it is determined that the target operation and maintenance alarm message in the target power equipment operation and maintenance report is a non-emergency operation and maintenance alarm message.

[0048] A power Internet of Things management system, comprising: a processor, a memory and a network interface; the processor is connected to the memory and the network interface; the network interface is used to provide a data communication function, the memory is used to store program code, and the processor is used to call the program code to execute the above method.

[0049] A computer-readable storage medium stores a computer program, which executes an Internet of Things-based power equipment management method when running.

[0050] A computer program product includes a computer program or a computer executable instruction. When the computer program or the computer executable instruction is executed by a processor, an electric power equipment management method based on the Internet of Things is implemented.

[0051] According to an embodiment of the present application, through technical means such as state dense semantic mining and bidirectional vector annotation, it is not only possible to deeply mine the rich semantic information in the operation and maintenance report, but also to achieve accurate positioning and linkage analysis of operation and maintenance alarm messages. This innovative technical solution will greatly improve the intelligent level of power equipment operation and maintenance, and provide an effective technical approach to solve the various drawbacks of traditional operation and maintenance methods.

[0052] It should be understood that the above general description and the following detailed description are merely exemplary and explanatory, and are not intended to limit the technical solutions of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present application and are used together with the specification to illustrate the technical solutions of the present application. It should be understood that the following drawings only illustrate certain embodiments of the present application and should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can also be obtained based on these drawings without creative work.

[0054] Figure 1 A schematic diagram of an electric power Internet of Things management system provided in an embodiment of the present application is shown.

[0055] Figure 2 A flow chart of an Internet of Things-based power equipment management method provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. The components of the embodiments of the present application described and shown herein can be arranged and designed in various different configurations.

[0057] Therefore, the following detailed description of the embodiments of the present application is not intended to limit the scope of the present application for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present application.

[0058] Figure 1 A schematic diagram of a power Internet of Things management system provided by an embodiment of the present application is shown, and the power Internet of Things management system 100 includes a processor 110, a memory 120, and a network interface 130. The processor 110 is connected to the memory 120 and the network interface 130. Further, the network interface 130 is used to provide a data communication function, the memory 120 is used to store program code, and the processor 110 is used to call the program code to execute the power equipment management method based on the Internet of Things.

[0059] Figure 2 The flowchart of the power equipment management method based on the Internet of Things that can implement the embodiment of the present application is shown in FIG. Figure 1 The power Internet of Things management system 100 shown implements a power equipment management method based on the Internet of Things, which exemplarily includes steps 110 to 140.

[0060] Step 110: Perform status dense semantic mining on the target power equipment operation and maintenance report to obtain a dense semantic set of edge-side system operation and maintenance status of the target power equipment operation and maintenance report; the target power equipment operation and maintenance report includes a target operation and maintenance alarm message.

[0061] Step 120: extracting the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message and the basic keyword vector of the operation and maintenance alarm message keywords according to the dense semantic set of the equipment operation and maintenance status; the operation and maintenance alarm message keywords are representative fields that characterize the target operation and maintenance alarm message.

[0062] Step 130: perform vector annotation on the basic operation and maintenance alarm message vector based on the basic keyword vector to obtain a target operation and maintenance alarm message vector; perform vector annotation on the basic keyword vector based on the basic operation and maintenance alarm message vector to obtain a target keyword vector.

[0063] Step 140: Based on the target operation and maintenance alarm message vector and the target keyword vector, perform linkage operation and maintenance alarm message positioning on the target operation and maintenance alarm message in the target power equipment operation and maintenance report.

[0064] In an exemplary complete application scenario, the power IoT management system first obtains a target power equipment operation and maintenance report. This report records in detail the equipment operation status, fault records, maintenance records, and other information of a substation in the past period of time, and includes some target operation and maintenance alarm messages, which are prompts of problems or potential risks that may occur during the operation of the equipment.

[0065] The power IoT management system first performs dense semantic mining on the report. Through natural language processing and machine learning technology, the system can deeply understand the text information in the report, extract key semantics related to the equipment operation and maintenance status, and form a dense semantic set of edge-side system operation and maintenance status. This semantic set contains all the important information about the equipment operation and maintenance status in the report, providing a rich data foundation for subsequent analysis.

[0066] Next, the power Internet of Things management system extracts the basic operation and maintenance alarm message vector and the basic keyword vector of the operation and maintenance alarm message keywords of the target operation and maintenance alarm message based on this semantic set. The basic operation and maintenance alarm message vector is the core content of the alarm message, which expresses the main characteristics and attributes of the alarm message in the form of a mathematical vector. The basic keyword vector of the operation and maintenance alarm message keywords is the representative field extracted from the alarm message, which can express the main idea of ​​the alarm message concisely and clearly.

[0067] Then, the power Internet of Things management system uses the basic keyword vector to vector-label the basic operation and maintenance alarm message vector to obtain the target operation and maintenance alarm message vector. This process is equivalent to "labeling" the operation and maintenance alarm message, allowing the system to understand and classify these messages more accurately. Similarly, the system will also use the basic operation and maintenance alarm message vector to vector-label the basic keyword vector to obtain the target keyword vector.

[0068] Finally, the power IoT management system locates the target operation and maintenance alarm message in the target power equipment operation and maintenance report based on the target operation and maintenance alarm message vector and the target keyword vector. This process is achieved by calculating the similarity between these vectors. The system can accurately find other alarm messages associated with a specific alarm message, thereby helping operation and maintenance personnel locate problems more quickly, analyze causes, and take appropriate solutions.

[0069] In this way, through the intelligent analysis and processing of the power Internet of Things management system, it is possible to deeply explore and accurately locate the alarm messages in the power equipment operation and maintenance reports, improve the operation and maintenance efficiency and quality, and provide strong technical support for ensuring the stable operation of the power system.

[0070] Based on the above content, the following is an adaptive explanation of terms for step 110 to step 140.

[0071] The target power equipment operation and maintenance report is a professional report that records the operation and maintenance of power equipment in detail. It can include important information such as the operation status, faults, maintenance records, and performance data of the power equipment over a period of time. This report is an important basis for the operation and maintenance team to manage and maintain the equipment, and also provides the management with a macro perspective of equipment operation.

[0072] State dense semantic mining is a method that uses natural language processing and machine learning techniques to conduct in-depth analysis of text data. It does not just stay on the surface meaning of the text, but can deeply mine the rich information and semantic relationships contained in the text. In the context of power equipment operation and maintenance reports, state dense semantic mining aims to extract key information and patterns related to the equipment operation and maintenance status from the report.

[0073] The edge-side system operation and maintenance status dense semantic set is a collection of key semantics about the operation and maintenance status of power equipment obtained through state dense semantic mining. It reflects all the core information related to the equipment operation and maintenance status in the report, including the performance status, potential problems, maintenance requirements, etc. of the equipment. This semantic set provides valuable information resources for subsequent data analysis and decision support.

[0074] Targeted O&M alarm messages are alarm messages that are specially marked in the power equipment O&M report and require the O&M team to pay attention to and handle. These alarm messages can indicate abnormal conditions in equipment operation, potential failures, or performance thresholds that are about to be reached, and are key prompts for the O&M team to respond and handle in a timely manner.

[0075] The basic operation and maintenance alarm message vector is the result of converting the target operation and maintenance alarm message into a vector form through mathematical methods. Each element in the vector represents a certain feature or attribute in the alarm message, so that the originally complex text information can be quantified and used for subsequent data analysis and processing.

[0076] The operation and maintenance alarm message keywords are the core words or phrases extracted from the target operation and maintenance alarm message, which can express the main content or key issues of the alarm message concisely and clearly. The extraction of keywords helps to quickly grasp the key points of the alarm message and improve the processing efficiency.

[0077] Similar to the basic operation and maintenance alarm message vector, the basic keyword vector also converts the operation and maintenance alarm message keywords into mathematical vectors. This conversion enables the similarity and correlation between keywords to be quantitatively evaluated, which facilitates subsequent data mining and classification.

[0078] In the operation and maintenance alarm message, some fields or words have particularly important representative meanings and can directly reflect the core content or key issues of the alarm message. These fields are called representative fields and are the information points that the operation and maintenance team needs to pay special attention to when processing the alarm message.

[0079] Vector annotation is a process of annotating new data (such as basic operation and maintenance alarm message vectors) using existing basic vectors (such as basic keyword vectors). Through vector annotation, additional information tags or feature descriptions can be added to new data, thereby enriching the data's expressiveness and improving the accuracy of subsequent analysis.

[0080] The terms target operation and maintenance alarm message vector and target keyword vector refer to the operation and maintenance alarm message vector and keyword vector after vector annotation processing. They not only retain the core features and information content of the original data, but also incorporate more contextual information and relevance features through the annotation process, providing a richer and more accurate basis for subsequent data mining and classification.

[0081] Linked O&M alarm message positioning is a technical method that uses the target O&M alarm message vector and the target keyword vector for intelligent matching and positioning. By calculating the similarity or correlation index between different vectors, the system can accurately find other alarm messages or related equipment information associated with a specific alarm message. This positioning method helps the O&M team quickly identify the root cause and scope of the problem and take appropriate solutions to improve the stability and reliability of the entire power system.

[0082] In the embodiment of the present application, a detailed description of step 110 to step 140 is also given to ensure that those skilled in the art can completely and clearly implement the embodiment of the present application.

[0083] In step 110, the power Internet of Things management system will conduct an in-depth analysis of the target power equipment operation and maintenance report, a process called state dense semantic mining. The core of this technology is to use natural language processing and machine learning algorithms to extract key information and patterns related to the operation and maintenance status of power equipment from a large amount of text data. First, the system reads the entire content of the target power equipment operation and maintenance report. This report can cover important information such as the equipment's operating status, historical fault records, maintenance history, and a series of target operation and maintenance alarm messages. These alarm messages are key prompts for abnormalities or potential risk points generated by the equipment during operation, and are essential for the health management and preventive maintenance of the equipment. Next, the power Internet of Things management system uses advanced natural language processing technology to parse the report. The system can recognize and understand the vocabulary, phrases, and sentence structures in the report, thereby capturing various details related to the equipment's operation and maintenance status. For example, the system can identify key information such as the type of equipment failure, the time of occurrence, the scope of the failure, and the possible causes. While parsing the text, the system also uses machine learning algorithms to conduct in-depth mining and analysis of this information. Through the well-trained model, the system can automatically identify and classify various operation and maintenance status information in the report, such as equipment performance indicators, failure modes, maintenance requirements, etc., and integrate this information into a comprehensive and detailed description of the operation and maintenance status. Finally, through state dense semantic mining, the power Internet of Things management system can generate a dense semantic set of edge-side system operation and maintenance status for the target power equipment operation and maintenance report. This semantic set not only contains the operation and maintenance status information explicitly mentioned in the report, but also infers the deep information and patterns hidden behind the text through algorithms. This information provides valuable data support for subsequent operation and maintenance decisions, and helps to improve the operating efficiency and reliability of power equipment. It can be seen that step 110 is a key link in the intelligent operation and maintenance process of the power Internet of Things management system. Through the state dense semantic mining technology, the system can deeply analyze the rich information in the power equipment operation and maintenance report, and provide strong data support for subsequent fault prediction, maintenance plan formulation and risk management.

[0084] In the power Internet of Things management system, step 120 is the key link immediately following step 110. After completing the dense semantic mining of the status of the target power equipment operation and maintenance report, the system obtains a dense semantic set containing key information of the equipment operation and maintenance status. This semantic set provides a solid data foundation for the subsequent analysis. The main task of step 120 is to extract the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message and the basic keyword vector of the operation and maintenance alarm message keywords from this semantic set. These two elements are the key to the subsequent accurate positioning and analysis of the alarm message. First, the system conducts an in-depth text analysis of the target operation and maintenance alarm message. Through natural language processing technology, the system can identify the core content of the alarm message, including key information such as fault type, occurrence time, and impact range. This information is integrated into a basic operation and maintenance alarm message vector, which mathematically represents the main features and attributes of the alarm message, which is convenient for subsequent data processing and pattern recognition. At the same time, the system will also extract the operation and maintenance alarm message keywords from the alarm message. These keywords are representative fields that characterize the subject of the alarm message and can summarize the core content of the alarm message concisely and clearly. For example, keywords such as "overtemperature" and "voltage abnormality" can quickly reflect the problems encountered by the equipment during operation. These keywords are converted into basic keyword vectors, and each element in the vector represents the importance or weight of the keyword in the alarm message. The process of refining the basic operation and maintenance alarm message vector and the basic keyword vector is actually the process of converting the alarm message from text form to mathematical form. This conversion not only facilitates the system to conduct quantitative analysis of alarm messages, but also improves the accuracy and efficiency of the analysis. Through these two vectors, the system can have a deeper understanding of the connotation and extension of the alarm message, and provide strong data support for subsequent alarm message positioning and problem solving. In this way, step 120 is a key step for in-depth processing of alarm messages in the power Internet of Things management system. By refining the basic operation and maintenance alarm message vector and the basic keyword vector, the system can convert the alarm message into a mathematical form that is easier to analyze and process, laying a solid foundation for subsequent intelligent decision-making and problem solving.

[0085] In the power Internet of Things management system, step 130 is a key link in the data processing and analysis process, which involves vector annotation of the basic keyword vector and the basic operation and maintenance alarm message vector to generate the target operation and maintenance alarm message vector and the target keyword vector. First, let's understand why vector annotation is needed. In the previous steps, the system has extracted the basic operation and maintenance alarm message vector and the basic keyword vector from the power equipment operation and maintenance report. Although these vectors contain information about alarm messages and keywords, they are independent of each other and lack direct relevance. The purpose of vector annotation is to link the two so that the system can understand the context and meaning of the alarm message more comprehensively. For example, the system uses the basic keyword vector to annotate the basic operation and maintenance alarm message vector. This means that the system "injects" the information of the keyword vector into the alarm message vector, so that the alarm message vector contains not only its own information, but also the keyword information related to it. This process is achieved through mathematical operations, such as dot product, weighted summation, etc., to ensure the accuracy and effectiveness of information fusion. The annotated operation and maintenance alarm message vector, that is, the target operation and maintenance alarm message vector, will have richer semantic information, which will help the system to understand the content and intention of the alarm message more accurately. Similarly, the system will also use the basic operation and maintenance alarm message vector to annotate the basic keyword vector. The purpose of this step is to let the keyword vector also "know" the alarm message context in which it is located. In this way, the keyword is no longer an isolated word, but is closely related to a specific alarm message, which helps the system to more accurately grasp the meaning and importance of the keyword in subsequent analysis. In this way, the vector annotation process of step 130 is a process of information fusion and enhancement. It establishes a connection between the originally independent alarm message vector and the keyword vector, and together constitutes a richer and more complete semantic space. In this space, the system can more accurately understand the meaning and context of the alarm message, and provide strong data support for subsequent problem location, cause analysis and solution formulation. The completion of this step marks that the power Internet of Things management system has taken another step forward in intelligent operation and maintenance.

[0086] In the power Internet of Things management system, step 140 is the closing link of the entire processing flow and is also a key step to achieve accurate positioning of alarm messages. Through the previous steps, the system has successfully extracted the target operation and maintenance alarm message vector and the target keyword vector from the power equipment operation and maintenance report. These vectors contain the core information and key features of the alarm message. The main task of step 140 is to use these vectors to locate the linkage operation and maintenance alarm message. The so-called "linked positioning" means that the system can accurately find other alarm messages associated with a specific alarm message based on the similarity and correlation between vectors. This positioning method not only helps the operation and maintenance personnel to quickly grasp the overall picture of the problem, but also provides strong data support for subsequent troubleshooting and solution formulation. For example, the system first calculates the similarity between the target operation and maintenance alarm message vector and the target keyword vector. This similarity calculation can be based on the vector space model (Vector Space Model) by calculating the cosine similarity or Euclidean distance between the two vectors. The higher the similarity, the stronger the correlation between the alarm messages or keywords represented by the two vectors. Next, the system sorts and clusters the alarm messages according to the calculated similarity. The purpose of sorting is to put the most relevant alarm messages in front so that the operation and maintenance personnel can handle them first. The purpose of clustering is to classify alarm messages with similar characteristics into one category, so as to help the operation and maintenance personnel quickly identify the commonalities and laws of the problem. After completing the sorting and clustering, the system can perform linkage operation and maintenance alarm message positioning. This means that when the operation and maintenance personnel view a certain alarm message, the system can automatically display other alarm messages and keywords associated with it. This linkage positioning method greatly improves the operation and maintenance efficiency, because the operation and maintenance personnel no longer need to manually search and filter relevant information, but can directly locate the problem from the associated information provided by the system. In addition, step 140 also involves the display and interaction of the positioning results. The system displays the results of linkage positioning in an intuitive and clear manner, such as presenting the association between alarm messages in the form of charts, lists or maps. At the same time, the system also provides rich interactive functions, allowing the operation and maintenance personnel to further filter, sort and view details of the positioning results as needed. Therefore, step 140 is an important link in realizing the accurate positioning of alarm messages in the power Internet of Things management system. By using the target operation and maintenance alarm message vector and the target keyword vector to perform similarity calculation, sorting, clustering and linkage positioning, the system can help operation and maintenance personnel quickly and accurately grasp the overall picture and key information of the problem, thereby improving operation and maintenance efficiency and quality, and providing strong technical support for ensuring the stable operation of the power system.

[0087] The embodiment of the present application performs dense semantic mining on the state of the target power equipment operation and maintenance report, which can deeply parse and extract rich semantic information of the edge-side system operation and maintenance status, forming a comprehensive and accurate semantic set. This not only enhances the information value of the operation and maintenance report, but also provides a solid data foundation for subsequent alarm message processing. Furthermore, by refining the basic operation and maintenance alarm message vector and keyword vector of the target operation and maintenance alarm message, the embodiment of the present application can accurately capture the core content and key features of the alarm message, and realize the refined processing of the alarm message.

[0088] More importantly, the embodiment of the present application innovatively uses the basic keyword vector to vector-label the basic operation and maintenance alarm message vector, and vice versa, thereby obtaining a more accurate target operation and maintenance alarm message vector and target keyword vector. This two-way vector labeling method not only enriches the semantic information of the vector, but also improves the correlation between vectors, making the subsequent alarm message positioning more accurate and efficient.

[0089] Finally, based on these carefully refined vectors, the embodiments of the present application can realize the linkage operation and maintenance alarm message positioning of the target operation and maintenance alarm message in the target power equipment operation and maintenance report. This positioning method is not only fast, but also can accurately identify other alarms associated with specific alarm messages, greatly improving the work efficiency of operation and maintenance personnel and the accuracy of problem positioning. In general, the beneficial effects of the embodiments of the present application are reflected in improving the intelligent level of power equipment operation and maintenance, and providing strong technical support for ensuring the stable operation of the power system.

[0090] In some replaceable embodiments, the extracting the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message and the basic keyword vector of the operation and maintenance alarm message keywords based on the dense semantic set of the equipment operation and maintenance status includes: performing distribution vector extraction on the dense semantic set of the equipment operation and maintenance status to determine the local state semantic relationship network corresponding to the target operation and maintenance alarm message in the dense semantic set of the equipment operation and maintenance status; extracting the operation and maintenance alarm message vector set corresponding to the local state semantic relationship network from the dense semantic set of the equipment operation and maintenance status; performing distribution vector extraction on the operation and maintenance alarm message vector set to obtain the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message and the basic keyword vector of the operation and maintenance alarm message keywords.

[0091] In the power Internet of Things management system, the analysis of equipment operation and maintenance status and the accurate positioning of alarm messages are crucial. In order to achieve this goal, the system performs a series of complex processing procedures. Now, one of the key steps will be elaborated in detail: how to extract the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message and the basic keyword vector of the operation and maintenance alarm message keywords based on the dense semantic set of equipment operation and maintenance status.

[0092] First, the power Internet of Things management system extracts the distribution vector of the dense semantic set of equipment operation and maintenance status. This process aims to identify and extract the local state semantic relationship network associated with the target operation and maintenance alarm message from the huge semantic set. This relationship network is actually a complex network composed of multiple interrelated semantic nodes, which reflects the specific location of the alarm message in the equipment operation and maintenance status and its relationship with other states.

[0093] Next, the system extracts the set of operation and maintenance alarm message vectors corresponding to this local state semantic relationship network from the dense semantic set of equipment operation and maintenance status. These vectors are digital representations of alarm messages, and they contain the key information and features of alarm messages. By matching these vectors with the local state semantic relationship network, the system can more accurately understand the position and role of alarm messages in the overall operation and maintenance status.

[0094] Then, the power Internet of Things management system further extracts the distribution vector of the extracted operation and maintenance alarm message vector set. The purpose of this step is to extract the most representative features from the vector set to form the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message and the basic keyword vector of the operation and maintenance alarm message keyword. These vectors and vectors not only retain the core information of the alarm message, but also facilitate the subsequent data processing and analysis of the system.

[0095] For example, the basic operation and maintenance alarm message vector is the main feature set of the alarm message, which can fully reflect the content and nature of the alarm message. The basic keyword vector of the operation and maintenance alarm message keyword is the digital representation of the key fields extracted from the alarm message. These key fields are important words or phrases that characterize the alarm message.

[0096] Through the above process, the power Internet of Things management system can accurately extract the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message and the basic keyword vector of the operation and maintenance alarm message keyword. This not only provides an accurate data basis for the subsequent alarm message positioning and analysis, but also greatly improves the efficiency and accuracy of the system in processing operation and maintenance alarm messages.

[0097] In this way, through the above implementation, the power Internet of Things management system has achieved efficient use of the dense semantic set of equipment operation and maintenance status, and accurately extracted the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message and the basic keyword vector of the operation and maintenance alarm message keyword. This not only enhances the system's ability to understand operation and maintenance alarm messages, but also provides a solid foundation for the subsequent positioning, classification and response of alarm messages. Overall, this implementation method improves the intelligence level of the power Internet of Things management system in the equipment operation and maintenance process, and provides strong technical support for ensuring the stable and efficient operation of the power system.

[0098] In some replaceable embodiments, the distribution vector extraction of the dense semantic set of the equipment operation and maintenance status is performed to determine the local state semantic relationship network corresponding to the target operation and maintenance alarm message in the dense semantic set of the equipment operation and maintenance status, including: performing state dense semantic mining on the dense semantic set of the equipment operation and maintenance status to identify x initial semantic masks corresponding to the dense semantic set of the equipment operation and maintenance status and the confidence weights corresponding to each initial semantic mask; x is a positive integer; determining the initial semantic mask with the largest confidence weight among the x initial semantic masks as the basic local state semantic relationship network; updating the semantic feature recognition of the basic local state semantic relationship network to determine the local state semantic relationship network.

[0099] In detail, when processing equipment operation and maintenance data, the power Internet of Things management system will conduct in-depth analysis of the dense semantic set of equipment operation and maintenance status to extract key information and build an accurate semantic relationship network. The following is a detailed explanation of this process.

[0100] First, the power Internet of Things management system conducts state dense semantic mining on the dense semantic set of equipment operation and maintenance status. The purpose of this step is to identify the initial semantic masks hidden in the semantic set and evaluate the confidence weight of each mask. The initial semantic mask can be understood as a semantic pattern that can represent a specific operation and maintenance status, and the confidence weight reflects the importance and accuracy of these patterns in the overall semantic set.

[0101] Through complex algorithms and a large amount of data processing, the system can identify x initial semantic masks and assign a confidence weight to each mask. Here x is a positive integer, and the specific value depends on the complexity of the semantic set and the processing power of the system.

[0102] Next, the power IoT management system selects a mask with the largest confidence weight from these initial semantic masks and determines it as the basic local state semantic relationship network. This basic network is the core of subsequent analysis, and it represents the most significant and reliable semantic relationship in the operation and maintenance status.

[0103] However, relying solely on the basic local state semantic relationship network may not be enough to capture all the key information. Therefore, the system updates the semantic feature recognition of this basic network. This update process involves adjusting the weights and connections of each node in the network to more accurately reflect the actual situation of the operation and maintenance status.

[0104] After the update, the system finally determined a complete local state semantic relationship network, which not only contains the key information in the operation and maintenance status, but also accurately reflects the relationship between these information.

[0105] Through this implementation method, the power Internet of Things management system can accurately extract and analyze the dense semantic set of equipment operation and maintenance status, so as to have a deeper understanding of the equipment operation and maintenance status and make accurate decisions. This not only improves the intelligence level of the system, but also provides a strong guarantee for the stable operation of power equipment.

[0106] Therefore, the power Internet of Things management system can efficiently extract key information from the dense semantic set of equipment operation and maintenance status and construct an accurate local state semantic relationship network. This not only enhances the system's real-time monitoring capability of equipment operation and maintenance status, but also provides strong data support for fault warning, preventive maintenance, etc. Throughout the process, the system ensures the accuracy and reliability of the extracted information through the evaluation of confidence weights and the update of semantic features, thereby greatly improving the operation and maintenance efficiency and decision-making quality of the power Internet of Things management system.

[0107] In some replaceable embodiments, the basic operation and maintenance alarm message vector is vector-annotated based on the basic keyword vector to obtain a target operation and maintenance alarm message vector, and the basic keyword vector is vector-annotated based on the basic operation and maintenance alarm message vector to obtain a target keyword vector, including: integrating the basic operation and maintenance alarm message vector and the basic keyword vector to obtain a semantic feature commonality score between the basic operation and maintenance alarm message vector and the basic keyword vector; obtaining an operation and maintenance alarm message enhancement vector from the basic keyword vector according to the semantic feature commonality score, vector-annotating the basic operation and maintenance alarm message vector according to the operation and maintenance alarm message enhancement vector to obtain a target operation and maintenance alarm message vector corresponding to the basic operation and maintenance alarm message vector; obtaining a field enhancement vector from the basic operation and maintenance alarm message vector according to the semantic feature commonality score, vector-annotating the basic keyword vector according to the field enhancement vector to obtain a target keyword vector corresponding to the basic keyword vector.

[0108] Specifically, in the power Internet of Things management system, in order to more accurately understand and respond to operation and maintenance alarm messages, the system performs in-depth processing and analysis on the basic operation and maintenance alarm message vector and the basic keyword vector. This processing process will be described in detail below.

[0109] First, the power Internet of Things management system integrates the basic operation and maintenance alarm message vector and the basic keyword vector. The purpose of this integration is to explore the commonality of semantic features between the two vectors, that is, the similarity and correlation between them at the semantic level. Through complex algorithm calculations, the system can derive a semantic feature commonality score, which reflects the strength of the common features of the two vectors when expressing operation and maintenance alarm messages.

[0110] Next, the system extracts the maintenance alarm message enhancement vector from the basic keyword vector based on the commonality score of the semantic features. This enhancement vector is actually a set of vector elements that can highlight the key features of the maintenance alarm message. The system uses this set of enhancement vectors to vector-label the basic maintenance alarm message vector, which is equivalent to adding more accurate and rich semantic information to the original vector, thereby obtaining the target maintenance alarm message vector.

[0111] Similarly, the system will extract field enhancement vectors from the basic operation and maintenance alarm message vectors based on the commonality scores of semantic features. These field enhancement vectors represent the most critical and important field information in the operation and maintenance alarm messages. Using these field enhancement vectors, the system performs vector annotation on the basic keyword vectors, further enhancing the expression ability and accuracy of the keyword vectors, and finally obtaining the target keyword vectors.

[0112] Through the above processing flow, the power Internet of Things management system can realize the refined processing of basic operation and maintenance alarm message vectors and basic keyword vectors, extract more accurate and useful information, and provide strong data support for the subsequent operation and maintenance alarm message positioning and analysis.

[0113] It can be seen that based on the embodiments of the present application, not only the semantic commonalities between the basic operation and maintenance alarm message vector and the basic keyword vector are deeply excavated, but also the accuracy and expressiveness of the target operation and maintenance alarm message vector and the target keyword vector are significantly improved by strengthening the annotation of the vector. This two-way vector annotation method enables the system to more comprehensively understand the deep meaning and key features of the operation and maintenance alarm message, thereby improving the response speed and processing efficiency of the alarm message. Overall, this technical solution enhances the intelligent processing capabilities of the power Internet of Things management system in terms of equipment operation and maintenance alarms, and provides strong technical support for ensuring the stable and safe operation of the power system.

[0114] In some replaceable embodiments, the basic operation and maintenance alarm message vector and the basic keyword vector are integrated to obtain a semantic feature commonality score between the basic operation and maintenance alarm message vector and the basic keyword vector, including: performing feature mapping on the basic operation and maintenance alarm message vector and the basic keyword vector respectively to obtain an operation and maintenance alarm message mapping vector of a first specified size and a keyword mapping vector of a second specified size; and determining a weighted result of the operation and maintenance alarm message mapping vector and the keyword mapping vector as a semantic feature commonality score between the basic operation and maintenance alarm message vector and the basic keyword vector.

[0115] In this embodiment, when processing the operation and maintenance alarm message, the power Internet of Things management system needs to perform in-depth analysis on the basic operation and maintenance alarm message vector and the basic keyword vector to evaluate the degree of semantic association between them. The implementation details of this process are as follows.

[0116] First, the power Internet of Things management system performs feature mapping on the basic operation and maintenance alarm message vector and the basic keyword vector respectively. Feature mapping is a data conversion technology that can convert raw data into a form that is easier to analyze and compare. In this process, the system uses a complex algorithm to convert the basic operation and maintenance alarm message vector into an operation and maintenance alarm message mapping vector of the first specified size, and converts the basic keyword vector into a keyword mapping vector of the second specified size.

[0117] The "first specified size" and "second specified size" in the embodiment of the present application are vectors or vector dimension sizes preset by the system, which help to unify the data format and facilitate subsequent calculations and comparisons. Through feature mapping, the system can extract the core features of operation and maintenance alarm messages and keywords, laying the foundation for subsequent analysis.

[0118] Next, the power IoT management system calculates the weighted results of the operation and maintenance alarm message mapping vector and the keyword mapping vector. The weighted result is actually a numerical value that reflects the semantic similarity or correlation between the two mapping results. The system performs a weighted summation of the operation and maintenance alarm message mapping vector and the keyword mapping vector according to the preset weight distribution scheme to obtain the semantic feature commonality score.

[0119] The higher the score, the stronger the semantic commonality between the basic operation and maintenance alarm message vector and the basic keyword vector, that is, the closer the association between the operation and maintenance alarm information and the keywords they express. Such a score helps the system understand the meaning and context of the operation and maintenance alarm message more accurately, thereby improving the response speed and processing efficiency of the alarm message.

[0120] In this way, the power Internet of Things management system can accurately evaluate the commonality of semantic features between the basic operation and maintenance alarm message vector and the basic keyword vector. This not only enhances the system's ability to understand the operation and maintenance alarm message, but also provides an important reference for subsequent alarm message classification and priority division. Overall, this technical solution improves the intelligence level and accuracy of the power Internet of Things management system in processing operation and maintenance alarm messages, and provides strong technical support for ensuring the stable operation of the power system.

[0121] In some replaceable embodiments, the extracting the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message and the basic keyword vector of the operation and maintenance alarm message keywords based on the dense semantic set of the equipment operation and maintenance status includes: performing state dense semantic mining on the dense semantic set of the equipment operation and maintenance status based on the operation and maintenance alarm message refining branch in the power equipment operation and maintenance report analysis network to obtain the operation and maintenance alarm message vector set corresponding to the target operation and maintenance alarm message; inputting the operation and maintenance alarm message vector set into the operation and maintenance alarm message positioning branch in the power equipment operation and maintenance report analysis network, and refining the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message based on the operation and maintenance alarm message positioning branch; inputting the operation and maintenance alarm message vector set into the representative field positioning branch in the power equipment operation and maintenance report analysis network, and refining the basic keyword vector of the operation and maintenance alarm message keywords based on the representative field positioning branch.

[0122] Based on this embodiment, the power Internet of Things management system will go through a series of precise and complex analysis processes when processing operation and maintenance alarm messages. The following is a detailed description of this process.

[0123] First, the power Internet of Things management system uses the power equipment operation and maintenance report to analyze the operation and maintenance alarm message extraction branch in the network and conducts in-depth mining of the dense semantic set of equipment operation and maintenance status. This semantic set contains a lot of information about the operation and maintenance status of power equipment. Through the state dense semantic mining technology, the system can extract the operation and maintenance alarm message vector set related to the target operation and maintenance alarm message. These vectors not only reflect the operating status of the equipment, but also contain possible fault or abnormal information.

[0124] Next, the system inputs the extracted operation and maintenance alarm message vector set into the operation and maintenance alarm message positioning branch in the power equipment operation and maintenance report analysis network. This positioning branch has powerful data processing and analysis capabilities. It can further refine the input vector set and finally obtain the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message. This basic operation and maintenance alarm message vector is a refined representation of the original alarm message. It retains the most critical information, which facilitates the system to respond and process quickly in the future.

[0125] At the same time, the power Internet of Things management system will also input the operation and maintenance alarm message vector set into the representative field positioning branch in the power equipment operation and maintenance report analysis network. The main task of this branch is to identify and extract the keywords in the operation and maintenance alarm messages, which are usually closely related to key information such as the equipment's fault type, location or severity. Through the processing of the representative field positioning branch, the system can obtain the basic keyword vector of the operation and maintenance alarm message keywords. This vector not only contains the semantic information of the keywords, but also reflects their importance in the operation and maintenance alarm messages.

[0126] The power Internet of Things management system deeply mines and refines the dense semantic set of equipment operation and maintenance status through different branches in the operation and maintenance report analysis network, and obtains the basic operation and maintenance alarm message vector and basic keyword vector. These two results provide strong support for the subsequent equipment fault location, warning and processing. In other words, the power Internet of Things management system can accurately extract the basic operation and maintenance alarm message vector and basic keyword vector of the target operation and maintenance alarm message. This not only improves the system's ability to parse operation and maintenance alarm messages, but also lays a solid foundation for subsequent fault warning and rapid response. The entire processing flow makes full use of the powerful functions of the power equipment operation and maintenance report analysis network, realizes the in-depth understanding and efficient processing of operation and maintenance alarm messages, thereby effectively improving the intelligence level and operation and maintenance efficiency of the power Internet of Things management system.

[0127] In some replaceable embodiments, based on the target operation and maintenance alarm message vector and the target keyword vector, the target operation and maintenance alarm message in the target power equipment operation and maintenance report is linked to the operation and maintenance alarm message positioning, including: according to the operation and maintenance alarm message positioning branch in the power equipment operation and maintenance report analysis network, the target operation and maintenance alarm message vector is subjected to distribution vector extraction, the target operation and maintenance alarm message label to which the target operation and maintenance alarm message belongs and the operation and maintenance alarm text set of the target operation and maintenance alarm message are determined, and the target operation and maintenance alarm message label and the operation and maintenance alarm text set are determined. The target operation and maintenance alarm message is the involved text; according to the representative field positioning branch in the power equipment operation and maintenance report analysis network, the target keyword vector is refined by distribution vector, and the fault error features corresponding to y fault error topics are determined, and the y fault error features are determined as the keyword attributes of the operation and maintenance alarm message keyword; y is a positive integer, and the y fault error topics are the discrimination possibilities of the representative field positioning branch; based on the involved text and the keyword attributes, the target operation and maintenance alarm message in the target power equipment operation and maintenance report is linked to the operation and maintenance alarm message positioning.

[0128] In this embodiment, the power Internet of Things management system will use the target operation and maintenance alarm message vector and the target keyword vector in combination when locating the linkage operation and maintenance alarm message. The following is a detailed explanation of this process.

[0129] First, the system uses the power equipment operation and maintenance report to analyze the operation and maintenance alarm message location branch in the network and deeply processes the target operation and maintenance alarm message vector. Specifically, the system extracts the distribution vector of this vector, which is a data representation that can reflect the core characteristics of the operation and maintenance alarm message. By analyzing this distribution vector, the system can determine the target operation and maintenance alarm message label to which the target operation and maintenance alarm message belongs. This label is an accurate classification of the alarm message type. At the same time, the system will also extract the operation and maintenance alarm text set related to this label. These texts describe the specific content and context of the alarm message in detail. These two parts of information-the target operation and maintenance alarm message label and the operation and maintenance alarm text set, together constitute the involved text of the target operation and maintenance alarm message.

[0130] Next, the power Internet of Things management system uses the representative field location branch in the power equipment operation and maintenance report analysis network to process the target keyword vector. The system extracts the distribution vector of this vector, and then determines the fault error features corresponding to the y fault error topics. These features are an in-depth characterization of keyword attributes, and they reflect the core information and characteristics of fault errors. The y fault error topics are the most likely fault types determined by the representative field location branch based on historical data and algorithm models, and y is a positive integer. These fault error features are determined as keyword attributes of the operation and maintenance alarm message keywords.

[0131] Finally, based on the involved text and keyword attributes, the power Internet of Things management system locates the target operation and maintenance alarm message in the target power equipment operation and maintenance report. This process is comprehensive. The system combines the tags, text content and keyword attributes of the operation and maintenance alarm message to conduct multi-dimensional analysis and matching, so as to accurately locate the specific operation and maintenance alarm message. This positioning method is not only highly accurate, but also can quickly respond to and handle various operation and maintenance alarm situations.

[0132] In this way, the power Internet of Things management system can efficiently and accurately locate the linkage operation and maintenance alarm messages. The system comprehensively uses the location branch of the operation and maintenance alarm message and the representative field location branch, and conducts in-depth processing and analysis of the target operation and maintenance alarm message vector and the target keyword vector. This not only improves the accuracy of alarm message positioning, but also enhances the system's adaptability to complex operation and maintenance environments. Overall, this technical solution significantly improves the operation and maintenance efficiency and response speed of the power Internet of Things management system, and provides strong technical support for ensuring the stable operation of the power system.

[0133] In some replaceable embodiments, the number of operation and maintenance alarm message keywords is z; z is a positive integer; the determining of y fault error features as keyword attributes of the operation and maintenance alarm message keywords includes: based on the operation and maintenance alarm message keywords related to the y fault error topics, the fault error features corresponding to the y fault error topics are decomposed into z operation and maintenance fault record sets; each operation and maintenance fault record set corresponds to an operation and maintenance alarm message keyword; the fault error features in each operation and maintenance fault record set are weighted to obtain the keyword attributes corresponding to each operation and maintenance alarm message keyword; the positioning of the target operation and maintenance alarm message in the operation and maintenance report of the target power equipment based on the involved text and the keyword attributes includes: positioning the target operation and maintenance alarm message in the operation and maintenance report of the target power equipment based on the involved text and the keyword attributes corresponding to each operation and maintenance alarm message keyword.

[0134] In this embodiment, when processing the operation and maintenance alarm message, the power Internet of Things management system will pay special attention to the keywords in the operation and maintenance alarm message, which are usually closely related to specific faults or problems. In this embodiment, it will be explained in detail how the system can accurately locate the operation and maintenance alarm message through keyword attributes.

[0135] First of all, it should be clear that the number of keywords in the operation and maintenance alarm message is z, where z is a positive integer. This means that in each operation and maintenance alarm message, the system will identify and extract z keywords, which are crucial for subsequent fault location and analysis.

[0136] Next, the system disassembles the fault error features corresponding to each topic based on the y fault error topics that have been determined. This disassembly process is based on the operation and maintenance alarm message keywords related to the fault error topic, with the purpose of refining complex fault error features into more specific and easier to understand operation and maintenance fault record sets. Each operation and maintenance fault record set corresponds to an operation and maintenance alarm message keyword, which ensures that each keyword has a corresponding detailed fault record.

[0137] After the disassembly is completed, the system performs weighted processing on the fault error features in each operation and maintenance fault record set. The purpose of weighting is to assign a corresponding weight value to each fault error feature based on its importance and impact. This weight value will directly affect the accuracy and effectiveness of the subsequent linkage operation and maintenance alarm message positioning. Through weighted processing, the system can more accurately characterize the keyword attributes of each operation and maintenance alarm message keyword.

[0138] Finally, based on the involved text (i.e., the target operation and maintenance alarm message label and the operation and maintenance alarm text set) and the keyword attributes corresponding to each operation and maintenance alarm message keyword, the power Internet of Things management system locates the target operation and maintenance alarm message in the target power equipment operation and maintenance report. This positioning process is comprehensive, combining multiple aspects of the operation and maintenance alarm message, including the alarm message label, text content, and keyword attributes obtained through weighted processing. Through the comprehensive analysis and comparison of this information, the system can accurately locate the specific operation and maintenance alarm message, providing strong support for subsequent fault handling.

[0139] In this way, the power Internet of Things management system can process and locate operation and maintenance alarm messages more accurately. The system breaks down complex fault error features into operation and maintenance fault record sets corresponding to the operation and maintenance alarm message keywords, and weights the features in these record sets to obtain the precise attributes of each keyword. This enables the system to comprehensively consider more dimensional information in the subsequent linkage operation and maintenance alarm message positioning, significantly improving the accuracy and efficiency of positioning. Overall, this technical solution provides more powerful and intelligent support for the operation and maintenance work of the power Internet of Things management system, which helps to improve the overall performance of the system and user satisfaction.

[0140] In some replaceable embodiments, the target operation and maintenance alarm message in the target power equipment operation and maintenance report is located as a linked operation and maintenance alarm message based on the involved text and the keyword attributes corresponding to each operation and maintenance alarm message keyword, including: obtaining the safety hazard viewpoint related to the involved text; if the keyword attributes corresponding to each operation and maintenance alarm message keyword contain a keyword attribute that matches the safety hazard viewpoint, then determining that the target operation and maintenance alarm message in the target power equipment operation and maintenance report is a linked operation and maintenance alarm message, and sending a safety hazard warning prompt to the target edge side system; the target edge side system is the edge side system that sends the target power equipment operation and maintenance report; if the keyword attributes corresponding to each operation and maintenance alarm message keyword do not match the safety hazard viewpoint, then determining that the target operation and maintenance alarm message in the target power equipment operation and maintenance report is a non-emergency operation and maintenance alarm message.

[0141] In this embodiment, the power Internet of Things management system adopts a linkage mechanism to ensure a rapid response to emergency safety hazards when processing operation and maintenance alarm messages. The core of this linkage mechanism is to accurately locate the target operation and maintenance alarm message in the target power equipment operation and maintenance report by combining the keyword attributes corresponding to the involved text and each operation and maintenance alarm message keyword.

[0142] First, the system obtains the potential safety hazards contained in the involved text. These opinions are usually extracted from the labels and texts of the operation and maintenance alarm messages, and they reflect the potential safety hazards or fault points of the power equipment.

[0143] Next, the system compares the keyword attributes corresponding to each operation and maintenance alarm message keyword with the security risk view one by one. This comparison process is delicate and complex. The system needs to analyze whether the keyword attributes contain information that matches the security risk view. If a matching keyword attribute is found, it means that the operation and maintenance alarm message may reveal a real and urgent security risk.

[0144] In this case, the power IoT management system immediately determines that the target operation and maintenance alarm message in the target power equipment operation and maintenance report is a linkage operation and maintenance alarm message, and sends a safety hazard warning prompt to the target edge side system that sent the operation and maintenance report. This warning prompt is to ensure that relevant personnel can take prompt action to investigate and deal with potential safety hazards.

[0145] However, if the keyword attributes corresponding to the keywords of each operation and maintenance alarm message do not match the safety hazard viewpoint during the comparison process, the system determines that the operation and maintenance alarm message is a non-emergency operation and maintenance alarm message. Although such messages also reflect some problems of power equipment, they do not constitute an urgent safety hazard, so they can be handled according to the normal process.

[0146] In this way, the power Internet of Things management system can achieve accurate classification and rapid response to operation and maintenance alarm messages. By comparing the keyword attributes of the operation and maintenance alarm message keywords with the safety hazard perspective, the system can quickly identify urgent safety hazards and send early warning prompts to relevant personnel, thereby ensuring that safety hazards are dealt with in a timely and effective manner. At the same time, the system can also make accurate judgments on non-emergency operation and maintenance alarm messages, avoiding overreaction to non-emergency issues. Overall, this technical solution significantly improves the power Internet of Things management system's processing capabilities and efficiency for operation and maintenance alarm messages, and provides strong support for ensuring the safe and stable operation of the power system.

[0147] In some replaceable embodiments, the y fault error features include a fault trend description and a hidden danger triggering possibility corresponding to each fault error theme; the target operation and maintenance alarm message in the target power equipment operation and maintenance report is linked to the operation and maintenance alarm message based on the involved text and the keyword attributes corresponding to each operation and maintenance alarm message keyword, including: obtaining z keyword confidence coefficients associated with the involved text; each keyword confidence coefficient corresponds to an operation and maintenance alarm message keyword; obtaining the emergency fault error theme corresponding to each operation and maintenance alarm message keyword, and obtaining each emergency fault error theme from the keyword attributes corresponding to each operation and maintenance alarm message keyword. The hidden danger triggering possibility corresponding to the topic is adjusted according to the z keyword confidence coefficients to obtain the urgency of the target power equipment operation and maintenance report; if the urgency is greater than or equal to the urgency threshold, the target operation and maintenance alarm message in the target power equipment operation and maintenance report is determined to be a linkage operation and maintenance alarm message, and a safety hazard early warning prompt is sent to the target edge side system; the target edge side system is the edge side system that sends the target power equipment operation and maintenance report; if the urgency is less than the urgency threshold, the target operation and maintenance alarm message in the target power equipment operation and maintenance report is determined to be a non-emergency operation and maintenance alarm message.

[0148] Based on this embodiment, when processing an operation and maintenance alarm message, the power Internet of Things management system will comprehensively consider multiple factors to determine the urgency of the alarm and take corresponding actions according to the urgency. The following is a detailed description of how the system locates the linkage operation and maintenance alarm message based on the attributes of the involved text and the operation and maintenance alarm message keywords.

[0149] First, the system obtains the confidence coefficients of z keywords involved in the text association. These confidence coefficients reflect the importance and credibility of each operation and maintenance alarm message keyword in the text and are the basis for subsequent analysis. Each keyword has a corresponding confidence coefficient, which is obtained through the system's analysis and learning of historical data.

[0150] Next, the system obtains the emergency fault error topic corresponding to each operation and maintenance alarm message keyword. These topics represent factors that are closely related to the keyword and may cause emergency faults. At the same time, the system also extracts the hidden danger triggering possibility corresponding to each emergency fault error topic from the keyword attributes corresponding to each operation and maintenance alarm message keyword. This hidden danger triggering possibility is a quantitative indicator used to evaluate the risk of the actual safety hidden danger caused by the fault error topic.

[0151] With these basic data, the system adjusts the potential triggering probability of each emergency fault error topic according to the confidence coefficient of the z keywords. This adjustment process is to more accurately reflect the actual situation, because different keywords have different importance in judging the degree of urgency. Through adjustment, the system can obtain a more accurate urgency assessment value that integrates all keyword information.

[0152] Finally, the system determines whether the emergency assessment value reaches or exceeds the preset emergency threshold. If the threshold is reached, it means that the target operation and maintenance alarm message in the target power equipment operation and maintenance report reveals possible emergency safety hazards. The system immediately identifies it as a linkage operation and maintenance alarm message and sends a safety hazard warning to the target edge system that sent the operation and maintenance report. In this way, relevant personnel can take quick action to investigate and deal with potential safety issues.

[0153] On the contrary, if the urgency assessment value is lower than the threshold value, the system considers that the problem reflected by the operation and maintenance alarm message is not urgent, so it will be determined as a non-urgent operation and maintenance alarm message. Such messages can be processed according to the normal process without immediate action.

[0154] In this way, the power Internet of Things management system can more accurately assess the urgency of operation and maintenance alarm messages and take corresponding countermeasures based on the assessment results. This analysis method based on the keyword confidence coefficient and the possibility of hidden danger triggering based on the emergency fault error theme not only improves the system's ability to identify emergency safety hazards, but also ensures that early warning prompts are only triggered in truly emergency situations, thereby avoiding unnecessary panic and waste of resources. Overall, this technical solution provides an efficient and accurate operation and maintenance alarm message processing method for the power Internet of Things management system, which helps to improve the safety and operation efficiency of the system.

[0155] Based on the same or similar technical concepts mentioned above, an embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored, and the computer program executes an Internet of Things-based power equipment management method when running.

[0156] Based on the same or similar technical concepts mentioned above, an embodiment of the present application also provides a computer program product, including a computer program or computer executable instructions, which, when executed by a processor, implements an Internet of Things-based power equipment management method.

[0157] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0158] The above disclosure is only a preferred embodiment of the present application, and certainly cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made based on the relevant contents of the present application are still within the scope covered by the present application.

Claims

1. A method for managing power equipment based on the Internet of Things, characterized in that: Applied to the power Internet of Things management system, the method includes: Performing state dense semantic mining on the target power equipment operation and maintenance report to obtain a dense semantic set of edge-side system operation and maintenance state of the target power equipment operation and maintenance report; the target power equipment operation and maintenance report includes a target operation and maintenance alarm message; Extracting the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message and the basic keyword vector of the operation and maintenance alarm message keyword according to the dense semantic set of the equipment operation and maintenance status; the operation and maintenance alarm message keyword is a representative field representing the target operation and maintenance alarm message; Performing vector annotation on the basic operation and maintenance alarm message vector based on the basic keyword vector to obtain a target operation and maintenance alarm message vector, and performing vector annotation on the basic keyword vector based on the basic operation and maintenance alarm message vector to obtain a target keyword vector; Based on the target operation and maintenance alarm message vector and the target keyword vector, performing linkage operation and maintenance alarm message positioning on the target operation and maintenance alarm message in the target power equipment operation and maintenance report; The step of extracting the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message and the basic keyword vector of the operation and maintenance alarm message keyword according to the dense semantic set of the equipment operation and maintenance status includes: Based on the operation and maintenance alarm message extraction branch in the power equipment operation and maintenance report analysis network, the dense semantics set of the equipment operation and maintenance state is mined to obtain the operation and maintenance alarm message vector set corresponding to the target operation and maintenance alarm message; Inputting the operation and maintenance alarm message vector set into the operation and maintenance alarm message positioning branch in the power equipment operation and maintenance report analysis network, and refining the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message according to the operation and maintenance alarm message positioning branch; Inputting the operation and maintenance alarm message vector set into the representative field positioning branch in the power equipment operation and maintenance report analysis network, and refining the basic keyword vector of the operation and maintenance alarm message keyword according to the representative field positioning branch; The step of performing linkage operation and maintenance alarm message positioning on the target operation and maintenance alarm message in the target power equipment operation and maintenance report based on the target operation and maintenance alarm message vector and the target keyword vector includes: According to the operation and maintenance alarm message positioning branch in the power equipment operation and maintenance report analysis network, the target operation and maintenance alarm message vector is refined by distribution vector, the target operation and maintenance alarm message label to which the target operation and maintenance alarm message belongs and the operation and maintenance alarm text set of the target operation and maintenance alarm message are determined, and the target operation and maintenance alarm message label and the operation and maintenance alarm text set are determined as the involved text of the target operation and maintenance alarm message; According to the representative field positioning branch in the power equipment operation and maintenance report analysis network, the target keyword vector is refined by distribution vector, the fault error reporting features corresponding to y fault error reporting topics are determined, and the y fault error reporting features are determined as keyword attributes of the operation and maintenance alarm message keyword; y is a positive integer, and the y fault error reporting topics are the discrimination possibilities of the representative field positioning branch; Based on the involved text and the keyword attributes, performing linkage operation and maintenance alarm message positioning on the target operation and maintenance alarm message in the target power equipment operation and maintenance report; Among them, the number of operation and maintenance alarm message keywords is z; z is a positive integer; the determination of y fault error features as keyword attributes of the operation and maintenance alarm message keywords includes: based on the operation and maintenance alarm message keywords related to the y fault error topics, the fault error features corresponding to the y fault error topics are decomposed into z operation and maintenance fault record sets; each operation and maintenance fault record set corresponds to an operation and maintenance alarm message keyword; the fault error features in each operation and maintenance fault record set are weighted to obtain the keyword attributes corresponding to each operation and maintenance alarm message keyword; the linkage operation and maintenance alarm message positioning of the target operation and maintenance alarm message in the operation and maintenance report of the target power equipment based on the involved text and the keyword attributes includes: linkage operation and maintenance alarm message positioning of the target operation and maintenance alarm message in the operation and maintenance report of the target power equipment based on the involved text and the keyword attributes corresponding to each operation and maintenance alarm message keyword.

2. The method according to claim 1, characterized in that The step of extracting the basic operation and maintenance alarm message vector of the target operation and maintenance alarm message and the basic keyword vector of the operation and maintenance alarm message keyword according to the dense semantic set of the equipment operation and maintenance status includes: Performing distribution vector extraction on the dense semantic set of the equipment operation and maintenance status to determine a local state semantic relationship network corresponding to the target operation and maintenance alarm message in the dense semantic set of the equipment operation and maintenance status; Extracting the operation and maintenance alarm message vector set corresponding to the local state semantic relationship network from the equipment operation and maintenance state dense semantic set; The distribution vector is refined on the operation and maintenance alarm message vector set to obtain a basic operation and maintenance alarm message vector of the target operation and maintenance alarm message and a basic keyword vector of the operation and maintenance alarm message keyword.

3. The method according to claim 2, characterized in that The extracting of distribution vectors from the dense semantic set of the equipment operation and maintenance status to determine the local state semantic relationship network corresponding to the target operation and maintenance alarm message in the dense semantic set of the equipment operation and maintenance status includes: Performing state dense semantic mining on the dense semantic set of the equipment operation and maintenance state, identifying x initial semantic masks corresponding to the dense semantic set of the equipment operation and maintenance state and the confidence weights corresponding to each initial semantic mask; x is a positive integer; Determine the initial semantic mask with the largest confidence weight among the x initial semantic masks as the basic local state semantic relationship network; The semantic feature recognition degree of the basic local state semantic relationship network is updated to determine the local state semantic relationship network.

4. The method according to claim 1, characterized in that The step of performing vector marking on the basic operation and maintenance alarm message vector based on the basic keyword vector to obtain a target operation and maintenance alarm message vector, and performing vector marking on the basic keyword vector based on the basic operation and maintenance alarm message vector to obtain a target keyword vector, includes: Integrating the basic operation and maintenance alarm message vector and the basic keyword vector to obtain a semantic feature commonality score between the basic operation and maintenance alarm message vector and the basic keyword vector; Acquire an operation and maintenance alarm message enhancement vector from the basic keyword vector according to the commonality score of the semantic features, and perform vector annotation on the basic operation and maintenance alarm message vector according to the operation and maintenance alarm message enhancement vector to obtain a target operation and maintenance alarm message vector corresponding to the basic operation and maintenance alarm message vector; A field enhancement vector is obtained from the basic operation and maintenance alarm message vector according to the semantic feature commonality score, and a vector annotation is performed on the basic keyword vector according to the field enhancement vector to obtain a target keyword vector corresponding to the basic keyword vector.

5. The method according to claim 4, characterized in that The step of integrating the basic operation and maintenance alarm message vector and the basic keyword vector to obtain a semantic feature commonality score between the basic operation and maintenance alarm message vector and the basic keyword vector includes: Performing feature mapping on the basic operation and maintenance alarm message vector and the basic keyword vector respectively to obtain an operation and maintenance alarm message mapping vector of a first specified size and a keyword mapping vector of a second specified size; A weighted result of the operation and maintenance alarm message mapping vector and the keyword mapping vector is determined as a semantic feature commonality score between the basic operation and maintenance alarm message vector and the basic keyword vector.

6. The method according to claim 1, characterized in that The method of performing linkage operation and maintenance alarm message positioning on the target operation and maintenance alarm message in the target power equipment operation and maintenance report based on the involved text and the keyword attribute corresponding to each operation and maintenance alarm message keyword includes: Obtaining safety hazard opinions related to the involved text; If there is a keyword attribute matching the safety hazard viewpoint in the keyword attributes corresponding to each operation and maintenance alarm message keyword, it is determined that the target operation and maintenance alarm message in the target power equipment operation and maintenance report is a linkage operation and maintenance alarm message, and a safety hazard early warning prompt is sent to the target edge side system; the target edge side system is the edge side system that sends the target power equipment operation and maintenance report; If the keyword attribute corresponding to each operation and maintenance alarm message keyword does not match the safety hazard viewpoint, it is determined that the target operation and maintenance alarm message in the target power equipment operation and maintenance report is a non-emergency operation and maintenance alarm message.

7. The method according to claim 1, characterized in that The y fault error reporting features include a fault trend description and a hidden danger triggering possibility corresponding to each fault error reporting topic; The method of performing linkage operation and maintenance alarm message positioning on the target operation and maintenance alarm message in the target power equipment operation and maintenance report based on the involved text and the keyword attribute corresponding to each operation and maintenance alarm message keyword includes: Obtaining z keyword confidence coefficients associated with the involved text; each keyword confidence coefficient corresponds to an operation and maintenance alarm message keyword; Obtain an emergency fault error reporting topic corresponding to each operation and maintenance alarm message keyword, and obtain a hidden danger triggering possibility corresponding to each emergency fault error reporting topic from a keyword attribute corresponding to each operation and maintenance alarm message keyword; Adjusting the hidden danger triggering possibility corresponding to each emergency fault error reporting topic according to the z keyword confidence coefficients to obtain the urgency of the target power equipment operation and maintenance report; If the urgency level is greater than or equal to the urgency level threshold, determining that the target operation and maintenance alarm message in the target power equipment operation and maintenance report is a linkage operation and maintenance alarm message, and sending a safety hazard warning prompt to the target edge side system; the target edge side system is the edge side system that sends the target power equipment operation and maintenance report; If the urgency level is less than the urgency level threshold, it is determined that the target operation and maintenance alarm message in the target power equipment operation and maintenance report is a non-emergency operation and maintenance alarm message.

8. A power Internet of Things management system, characterized in that: include: Processor, memory, and network interface; The processor is connected to the memory and the network interface; the network interface is used to provide data communication function, the memory is used to store program code, and the processor is used to call the program code to execute the power equipment management method based on the Internet of Things as described in any one of claims 1-7.

9. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is run, the method for managing electric power equipment based on the Internet of Things as described in any one of claims 1 to 7 is executed.

Images