A method and system for generating an electrical monitoring work ticket

Abstract

The application provides a power monitoring work ticket generation method, including the following steps: step S1, constructing a safety measure library; step S2, performing entity relationship extraction on work content to obtain a work content subgraph; the method includes using natural language processing technology to accurately convert the maintenance or repair intention of the oral or informal expression of the dispatcher into a standardized subgraph representation; step S3, searching the work content subgraph in a knowledge graph to obtain an associated work item sequence, and a risk entity set associated with the work item, the equipment, the process and the site; step S4, searching the risk entity in the knowledge graph and the safety measure library to obtain a safety measure item dictionary; and step S5, filling a work ticket according to the work item sequence, the risk entity set and the safety measure item dictionary; the application solves the problems that the traditional manual preparation of the work ticket is difficult and prone to errors, and historical data is difficult to use, and can realize the standardized generation of the work ticket.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power operation and maintenance, and in particular to a power monitoring work ticket generation method and system. BACKGROUND

[0002] The traditional manual work ticket preparation mode has problems such as low efficiency, easy to make mistakes, and difficult to adapt to the rapid updating of dispatching automation equipment, when facing a large number of complex power dispatching automation equipment operation tasks. Moreover, the identification of potential risks and the selection of countermeasures often depend on the experience and judgment of the staff, and lack of systematic and standardized protection mechanisms. This not only affects the normal operation and maintenance of power facilities, but also may pose a threat to personnel safety, and restricts the operation efficiency and service level improvement of the power industry.

[0003] The power industry has an urgent need for an automated and intelligent operation management mode. In recent years, as a powerful knowledge representation and reasoning tool, knowledge graph has been widely used in various fields, and has shown significant advantages in handling complex domain knowledge and realizing intelligent decision-making. Knowledge graph can organize multiple information such as dispatching automation equipment, operation, safety regulations, and fault cases in a structured way, and form a knowledge network that is easy for machines to understand and process through entity, attribute, and relationship modeling.

[0004] However, the depth integration and innovative application of knowledge graph technology in the specific scenario of power monitoring work ticket generation is still insufficient. Existing work ticket generation systems mostly stay at the stage of simple template filling or rule matching, and fail to fully utilize the semantic association capabilities of knowledge graph to automatically identify work content, associate risk factors, and accurately recommend safety measures in massive operation data. SUMMARY

[0005] The present application proposes a power monitoring work ticket generation method and system, which utilizes historical data and existing knowledge graphs to accurately generate the work items, risk items, and safety measures required by work tickets, solving the problems of difficulty and error-prone in traditional manual work ticket preparation, and the difficulty of utilizing historical data, and realizing the standardized generation of work tickets.

[0006] The present application adopts the following technical methods.

[0007] A power monitoring work ticket generation method, comprising the following steps:

[0008] Step S1, constructing a safety measure library;

[0009] Step S2, entity relationship extraction is performed on the work content to obtain a work content subgraph; the method comprises using natural language processing technology to accurately convert the maintenance or repair intention of the dispatch personnel in colloquial or informal expression into a standardized subgraph representation;

[0010] Step S3, the work content subgraph is retrieved in the knowledge graph to obtain an associated work item sequence, and a risk entity set of devices, processes and places associated with the work item;

[0011] Step S4, the risk entity is retrieved in the knowledge graph and the safety measure library to obtain a safety measure item dictionary;

[0012] Step S5, filling a work ticket according to the work item sequence, the risk entity set and the safety measure item dictionary.

[0013] In step S1, a safety measure library is constructed, specifically comprising the following steps:

[0014] Step S1A1, obtaining a safety measure corpus, and performing word segmentation on the safety measure corpus;

[0015] Step S1A2, using an LSTM method to determine the first information category of the safety measure corpus; determining whether the word segmentation result of the safety measure corpus contains the first information;

[0016] Step S1A3, using an LSTM method to determine the second information category of the safety measure corpus; determining whether the word segmentation result of the safety measure corpus contains the second information;

[0017] Step S1A4, adding the safety measure corpus to the safety measure library according to the first information category and the second information category;

[0018] In step S1, the safety measure library constructed has the following specific composition:

[0019] The safety measure library centrally stores the most basic and widely applicable safety measure clauses suitable for the power dispatch automation operation scenario; the library structurally organizes the measure content, ensuring that it has clear applicability and guiding significance at different time stages and for different object types;

[0020] The first information is a time dimension, which is divided into before work, during work and after work:

[0021] Before work, preventive safety measures are recorded, including environmental preparation, equipment inspection, notification to the duty station, and data backup to ensure that all conditions before work meet the safety requirements;

[0022] During work, process control safety measures are included, including operation sequence and isolation verification to ensure that risks are effectively controlled during the work process;

[0023] After the work is completed, restorative and summative safety measures are taken, including equipment restoration, site cleanup, and equipment verification, to ensure that the system returns to normal operation after the work activity is over.

[0024] The second piece of information is the object type, which is divided into software level, hardware level, and general level:

[0025] At the software level: This covers security measures for software components such as business systems, applications, processes, databases, and data tables, including system upgrades and patch management, data backup and recovery, to ensure the stable operation of the software system and information security;

[0026] Hardware level: This includes security measures for hardware equipment and infrastructure, including personal protective equipment, and refraining from tampering with other operating equipment to ensure the safe and reliable operation of hardware facilities.

[0027] General measures: These measures specify common safety requirements applicable to all work activities, including notifying the duty station. These measures do not distinguish between specific software or hardware objects.

[0028] Ideally, the constructed security measures library can be subdivided by object type into the software layer (business system layer, process layer, operating system layer, database layer, and data table layer) and the hardware layer (server device layer, network device layer, and infrastructure layer); and security measures can be divided according to the subdivided layers.

[0029] Each type of safety measure has a clearly defined applicable context and content requirements.

[0030] In steps S1A2 and S1A3, the LSTM method is specifically configured as follows:

[0031] The LSTM method consists of a vector representation layer, an LSTM layer, and a linear layer.

[0032] Input the original security measures corpus (Text) into the vector representation layer to obtain...

[0033] E=Embedding(Text)=(e1,e2,...,e n ),

[0034] Where n is the length of the corpus; Let d represent the embedding vector of the i-th word;

[0035] The vector sequence E is input into the LSTM layer, and an embedding vector e is processed at each time step t. t And based on the hidden state h of the previous time step t-1 and cell state C t-1 Update the hidden state h at the current moment. t and cell state Ct ; output the output vector O t at the current time step until the entire sequence information is processed, and the output vector O t at the last time step is obtained. t-1 t-1 T ;

[0036] The linear layer is used for classification O T as input to obtain the probability distribution of different categories.

[0037] In step S2, entity relationship extraction is performed on the work content to obtain a work content subgraph, specifically including the following steps:

[0038] Step S2A1, using an entity relationship extraction tool to extract at least 0 triples from the work content;

[0039] Step S2A2, when the number of triples is 0, the work content is taken as a single work content entity to form a subgraph with only one entity;

[0040] Step S2A3, when the number of triples is greater than 0, the extracted triples are taken as a subgraph.

[0041] In step S2A1, the entity relationship extraction method is a method based on Transformer and a dictionary rule matching method to obtain entity relationship triples;

[0042] The triples are specifically composed as follows:

[0043] The triple form is <h, r, t>, h represents the head entity, r represents the relationship, and t represents the tail entity;

[0044] Where h belongs to one of the work content entity, the fault entity, and the device entity;

[0045] Where r belongs to one of the involving relationship and the causing relationship;

[0046] Where t belongs to one of the data table entity, the place entity, the business system entity, the company entity, the role entity, the process entity, the device entity, and the fault entity.

[0047] In step S3, the work content subgraph is retrieved in the knowledge graph to obtain an associated work item sequence, specifically including the following steps:

[0048] Step S3A1, retrieving the subgraph in the knowledge graph to locate the fault entity and the work content entity, and locating to obtain the work content entity E content in the knowledge graph.

[0049] ​​Step S3A2, retrieving in the knowledge graph the work item entity that solves the relationship of E content There is a work item entity that solves the relationship, and its name is stored in the worklist;

[0050] Step S3A3, retrieving in the knowledge graph the risk entity name that has a relationship with the current work item as the head entity, and storing it in the risklist risk entity list;

[0051] Step S3A4, retrieving in the knowledge graph the work item entity that has a sequential relationship with the current work item as the head entity, storing the entity name in the worklist, and performing step S3A3 until there is no new work item;

[0052] In step S3, the knowledge graph is specifically composed as follows:

[0053] The knowledge graph includes but is not limited to the following entity relationship types:

[0054] The entity types include fault, work content, device, business system, process, work item, and safety measure entity; the relationship types include involvement, triggering, solving, and sequential relationship;

[0055] The information corresponding to the entity is stored in the form of entity attributes, and the entity attributes at least include entity id and entity name, wherein the entity id is the unique identifier of the entity, and the entity name stores the content information of the entity, wherein the entity name of the work item entity is the content of the work item;

[0056] In step S3, the risk entity includes device entity, business system entity, process entity, and site entity.

[0057] In step S3, the knowledge graph is stored in a graph database for easy updating and searching; the knowledge graph includes the association between existing account information, which is used to locate the maintenance or repair content, and specifically includes:

[0058] The entity types include personnel, site, file, and instruction entity;

[0059] The relationship types include belonging, being located, work member, and responsible relationship;

[0060] The attributes of the device entity include manufacturer and IP address;

[0061] In step S3, the risk entity includes device entity, business system entity, and process entity; the risk entity also includes instruction entity and site entity.

[0062] In step S4, the safety measure item dictionary is specifically composed as follows:

[0063] The safety measure item dictionary is stored in the form of {key:value}, where the key is one of before work, during work and after work, and the value is a list type, storing the content of the safety measure item at different times;

[0064] In step S4, the risk entity is searched in the knowledge graph and the safety measure library to obtain a safety measure item list, specifically including the following steps:

[0065] In step S4A1, the entity type is searched from the risk entity, if there is a business system and process type entity, the software type and general type safety measure content is searched in the safety measure library to join the safety measure list; in step S4A2, if there is no business system and process type entity, the hardware type and general type safety measure content is searched in the safety measure library to join the safety measure list;

[0066] In step S4A3, for the risk entity in the risk entity list, the safety measure entity with the current risk entity as the tail entity and the existing involved relationship is searched in the knowledge graph;

[0067] In step S4A4, for the safety measure entity, the risk entity set with the current safety measure entity as the head entity and the existing involved relationship is searched in the knowledge graph;

[0068] In step S4A5, if the risk entity set is a subset of the risk entity in the work ticket, the safety measure entity name is added to the safety measure list, and the time is during work.

[0069] A power monitoring work ticket generation system for implementing the above method or steps, including a memory, a processor and computer program instructions stored on the memory and capable of being executed by the processor, the computer program instructions are executed by the processor to implement the power monitoring work ticket generation method, specifically including a safety measure library, an intention recognition module, a work item retrieval module, a safety measure retrieval module and a work ticket generation module; by mining historical data and practical experience, the safety measure library; the maintenance or repair intention of the operator is recognized, and a standardized subgraph is obtained by an entity relationship extraction method; the knowledge graph is searched using the subgraph to obtain the work item and the involved risk item; the risk item is searched to obtain the safety measure item, and finally the work ticket is generated;

[0070] A safety measure library management module for managing existing formatted safety measures, adding, deleting, modifying and inquiring the safety measures;

[0071] A knowledge graph management module constructed using the interface provided by the graph database, for updating the existing knowledge graph;

[0072] The intent recognition module uses an entity relation extraction algorithm to extract triples from the text input by the dispatcher, and inputs the extracted triples into the work item retrieval module;

[0073] The work item retrieval module retrieves work items and risk items from the knowledge graph, and the order is determined by the order stored in the knowledge graph, and the dispatcher can modify the order by dragging the work item text;

[0074] The safety measure retrieval module retrieves safety measure items from the knowledge graph and the safety measure library according to the risk items, and can also modify the default order by dragging the safety measure item text;

[0075] The work order generation module generates a visual work order file according to the obtained work items, risk items and safety measure items.

[0076] Compared with the prior art, the present application has the following beneficial effects:

[0077] The present application deeply mines historical data and practical experience, and systematically constructs a standardized safety measure library. The library summarizes and organizes the safety measures previously filled in by manual, so that it becomes a rigorous and unified standard system, significantly improving the standardization and completeness of the safety measure content.

[0078] Further, the present application uses natural language processing technology to accurately convert the maintenance or repair intention expressed in colloquial or informal language by the dispatcher into a standardized subgraph representation, and based on the safety measure library and the knowledge graph, obtains the work items, risk items and safety measure items represented in a standardized manner, and then integrates to generate a complete and standardized work order. This intelligent conversion and integration process greatly simplifies the complexity of work order preparation, and ensures the accuracy and clarity of the task information.

[0079] The present application can provide fast, convenient, accurate and effective knowledge support for power dispatching automation operation, maintenance and technical management personnel, convert the maintenance or repair intention into a standardized subgraph, and use it for knowledge graph and safety measure library retrieval to obtain a standardized work order. It effectively solves the problems of difficult information acquisition, dependence on personal experience and easy omission under the traditional manual reporting mode, and effectively guarantees the safety and stability of operation and maintenance work, as well as the operation compliance of related personnel, which has great significance for improving the operation and management level and work efficiency of the entire power industry.

[0080] The application discloses a power monitoring work ticket generation method, which specifically comprises a safety measure library, an intention recognition module, a work item retrieval module, a safety measure retrieval module and a work ticket generation module; historical data and practical experience are mined to form the safety measure library; the maintenance or repair intention of an operator is recognized, and a standardized subgraph is obtained through an entity relationship extraction method; a work item and related risk items are obtained by searching a knowledge graph using the subgraph; the risk items are retrieved to obtain safety measure items, and finally a work ticket is generated.

[0081] The application utilizes a large amount of historical data and experience information to automatically generate a work ticket, effectively solves the problems of great information acquisition difficulty, dependence on personal experience and easy omission in a traditional manual reporting mode, and effectively guarantees the safety and stability of power facility operation and the operation compliance of related personnel. BRIEF DESCRIPTION OF DRAWINGS

[0082] The application will be further described in detail below in combination with the drawings and specific embodiments:

[0083] Attached Figure 1 is a schematic diagram of the principle framework of the generation method. DETAILED DESCRIPTION

[0084] The technical solutions of the application will be specifically described below in combination with the drawings.

[0085] As shown in the figure, a power monitoring work ticket generation method comprises the following steps:

[0086] Step S1, constructing a safety measure library;

[0087] Step S2, performing entity relationship extraction on work content to obtain a work content subgraph; the method comprises using natural language processing technology to accurately convert the maintenance or repair intention of a dispatcher in a colloquial or informal manner into a standardized subgraph representation;

[0088] Step S3, retrieving the work content subgraph in a knowledge graph to obtain an associated work item sequence, and a risk entity set of devices, processes and places associated with the work item;

[0089] Step S4, retrieving the risk entities in the knowledge graph and the safety measure library to obtain a safety measure item dictionary;

[0090] Step S5, filling a work ticket according to the work item sequence, the risk entity set and the safety measure item dictionary.

[0091] In step S1, the safety measure library is constructed, specifically comprising the following steps:

[0092] Step S1A1, obtaining safety measure corpus and performing word segmentation on the safety measure corpus;

[0093] Step S1A2, using the LSTM method, determine the first information category of the safety measure corpus; determine whether the safety measure corpus word segmentation result contains the first information vocabulary;

[0094] Step S1A3, using the LSTM method, determine the second information category of the safety measure corpus; determine whether the safety measure corpus word segmentation result contains the second information vocabulary;

[0095] Step S1A4, according to the first information category and the second information category, add the safety measure corpus to the safety measure library;

[0096] In step S1, the safety measure library is constructed, and the specific composition is as follows:

[0097] The safety measure library centrally stores the most basic and widely applicable safety measure clauses suitable for power dispatching automation operation scenarios; The library organizes the measure content in a structured manner to ensure its clear applicability and guidance significance at different time stages and for different object types;

[0098] The first information is time dimension, divided into before work, during work and after work:

[0099] Before work, preventive safety measures are included, including environmental preparation, equipment inspection, notification of duty station, data backup, to ensure that all conditions before work meet the safety requirements;

[0100] During work, process control safety measures are included, including operation sequence, isolation verification, to ensure that the risk is effectively controlled during the work process;

[0101] After work, recovery and summary safety measures are involved, including equipment recovery after work, site cleaning, equipment verification, to ensure that the system returns to normal state after the work activity is completed.

[0102] The second information is object type, divided into software level, hardware level and general level:

[0103] Software level: covers safety measures for business systems, application programs, processes, databases, data tables and other soft components, including system upgrade and patch management, data backup and recovery, to ensure the stable operation of software systems and information security;

[0104] Hardware level: includes safety measures for hardware devices and infrastructure, including personal protective equipment, do not move other running devices, to ensure the safe and reliable operation of hardware facilities.

[0105] General measures: specify common safety requirements applicable to all work activities, including notification of duty station, these measures do not distinguish specific software or hardware objects.

[0106] Each type of safety measure has a clearly defined applicable context and content requirements.

[0107] In steps S1A2 and S1A3, the LSTM method is specifically configured as follows:

[0108] The LSTM method consists of a vector representation layer, an LSTM layer, and a linear layer.

[0109] Input the original security measures corpus (Text) into the vector representation layer to obtain...

[0110] E=Embedding(Text)=(e1,e2,...,e n ),

[0111] Where n is the length of the corpus; Let d represent the embedding vector of the i-th word;

[0112] The vector sequence E is input into the LSTM layer, and an embedding vector e is processed at each time step t. t And based on the hidden state h of the previous time step t-1 and cell state C t-1 Update the hidden state h at the current moment. t and cell state C t Output the current output direction to O. t =LSTM(e t ,h t-1 C t-1 This process continues until all sequence information is processed, yielding the output vector O at the final time step. T ;

[0113] Linear layers are used for classification O T Using the input as the parameter, we obtain the probability distributions for different categories.

[0114] In step S2, entity relations are extracted from the work content to obtain a work content subgraph, which specifically includes the following steps:

[0115] Step S2A1: Use the entity relation extraction tool to extract entity relations from the work content to obtain at least 0 triples;

[0116] Step S2A2: When the number of triples is 0, the work content is treated as a single work content entity, forming a subgraph with only one entity;

[0117] Step S2A3: When the number of triples is greater than 0, the extracted triples are used as subgraphs.

[0118] In step S2A1, the entity relationship extraction method is a method based on a Transformer and a dictionary rule matching method to obtain entity relationship triples;

[0119] The triples are specifically constituted as follows:

[0120] The triple form is <h, r, t>, h represents a head entity, r represents a relationship, and t represents a tail entity;

[0121] Wherein h belongs to one of a work content entity, a fault entity, and a device entity;

[0122] Wherein r belongs to one of a involves relationship and a causes relationship;

[0123] Wherein t belongs to one of a data table entity, a site entity, a business system entity, a company entity, a role entity, a process entity, a device entity, and a fault entity.

[0124] In step S3, the work content subgraph is searched in the knowledge graph to obtain an associated work item sequence, specifically including the following steps:

[0125] In step S3A1, the subgraph is searched in the knowledge graph to locate a fault entity and a work content entity, and a work content entity E content is located in the knowledge graph.

[0126] In step S3A2, a work item entity that exists a solution relationship with E content is searched in the knowledge graph, and the name of the work item entity is stored in a worklist.

[0127] In step S3A3, a risk entity name that exists an involves relationship with the current work item as a head entity is searched in the knowledge graph, and the risk entity name is stored in a risk entity list.

[0128] In step S3A4, a work item entity that exists a sequence relationship with the current work item as a head entity is searched in the knowledge graph, the entity name is stored in the worklist, and step S3A3 is executed until there is no new work item.

[0129] In step S3, the knowledge graph is specifically constituted as follows:

[0130] The knowledge graph includes but is not limited to the following entity relationship types:

[0131] The entity types include a fault, a work content, a device, a business system, a process, a work item, and a safety measure entity; and the relationship types include an involves relationship, a causes relationship, a solution relationship, and a sequence relationship.

[0132] The information corresponding to the entity is stored in the form of attributes of the entity, and the attributes of the entity at least include an entity id and an entity name, wherein the entity id is a unique identifier of the entity, and the entity name stores content information of the entity, and the entity name of the work item entity is the content of the work item;

[0133] In step S3, the risk entity includes a device entity, a business system entity, a process entity, and a site entity.

[0134] In step S3, the knowledge graph is stored in a graph database for updating and searching; the knowledge graph includes associations between existing account information, is used to locate maintenance or repair content, and specifically includes:

[0135] The entity types include personnel, site, file, and instruction entity;

[0136] The relationship types include belonging, being located, working member, and responsible relationship;

[0137] The attributes of the device entity include a manufacturer and an IP address;

[0138] In step S3, the risk entity includes a device entity, a business system entity, a process entity; the risk entity also includes an instruction entity and a site entity.

[0139] In step S4, the safety measure item dictionary is specifically constituted as follows:

[0140] The safety measure item dictionary is stored in the form of {key:value}, wherein the key is one of before work, during work, and after work, and the value is a list type, and the content of the safety measure item at different times is stored;

[0141] In step S4, the risk entity is searched in the knowledge graph and the safety measure library to obtain a safety measure item list, and specifically includes the following steps:

[0142] In step S4A1, the entity type is searched from the risk entity, if there is a business system and a process type entity, the software type and the general type safety measure content in the safety measure library is searched to join the safety measure list; in step S4A2, if there is no business system and process type entity, the hardware type and general type safety measure content in the safety measure library is searched to join the safety measure list;

[0143] In step S4A3, for the risk entity in the risk entity list, the safety measure entity with the current risk entity as the tail entity and the existing involved relationship is searched in the knowledge graph;

[0144] In step S4A4, for the safety measure entity, the risk entity set with the current safety measure entity as the head entity and the existing involved relationship is searched in the knowledge graph;

[0145] Step S4A5, if the risk entity set is a subset of the risk entities in the work ticket, add the safety measure entity name to the safety measure list, and the time is in the work.

[0146] A power monitoring work ticket generation system for implementing the above method or steps, comprising a memory, a processor and computer program instructions stored on the memory and capable of being executed by the processor, the computer program instructions are executed by the processor to implement the power monitoring work ticket generation method, specifically including a safety measure library, an intention recognition module, a work item retrieval module, a safety measure retrieval module and a work ticket generation module; by mining historical data and practical experience, the safety measure library; the maintenance or repair intention of the operator is identified, and the standardized subgraph is obtained by entity relationship extraction method; the knowledge graph is searched using the subgraph, and the work item and the involved risk item are obtained; the risk item is retrieved to obtain the safety measure item, and finally the work ticket is generated;

[0147] The safety measure library management module is used for managing existing formatted safety measures, adding, deleting, modifying and inquiring the safety measures;

[0148] The knowledge graph management module is constituted using the interface provided by the graph database, and is used for updating the existing knowledge graph;

[0149] The intention recognition module uses entity relationship extraction algorithm to extract triples from the text input by the dispatcher, and inputs the extracted triples into the work item retrieval module;

[0150] The work item retrieval module retrieves the knowledge graph to obtain the work item and the risk item, and the order is determined by the order stored in the knowledge graph, and the dispatcher can modify the order by dragging the work item text;

[0151] The safety measure retrieval module retrieves the knowledge graph and the safety measure library according to the risk item to obtain the safety measure item, and can also modify the default order by dragging the safety measure item text;

[0152] The work ticket generation module generates a visual work ticket file according to the obtained work item, risk item and safety measure item.

[0153] The embodiment is aimed at the characteristics of the power monitoring work ticket, and realizes the generation operation of the work item, the risk item and the safety measure item in the work ticket. First, safety measures are summarized based on historical work tickets and historical experience to form a standardized safety measure library; entity relationship extraction is performed on the maintenance and repair intention of the dispatch personnel to obtain a subgraph; the work item sequence and the risk item set are obtained by searching the knowledge graph according to the subgraph, and the safety measure sequence is obtained by searching the knowledge graph and the safety measure library according to the risk item set, thereby realizing the automatic generation of the power monitoring work ticket. The power dispatch automation operation sequence arrangement, accurate risk identification and effective safety measure matching are realized, and the informationization and intelligentization upgrade of the dispatch operation are effectively promoted.

[0154] Those skilled in the art will understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0155] The present application is described with reference to flowcharts and / or block diagrams according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks. Figure 1 The means for performing the functions specified in one or more flows and / or blocks.

[0156] These computer program instructions can also be stored in a computer-readable memory that can guide the computer or other programmable data processing devices to work in a specific way, so that the instructions stored in the computer-readable memory produce a manufactured product including instruction means, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks. Figure 1 The means for performing the functions specified in one or more flows and / or blocks.

[0157] These computer program instructions can also be loaded into a computer or other programmable data processing device, so that a series of operation steps are performed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide a process for implementing the functions specified in the flowcharts and / or block diagrams.Figure 1 one or more processes and / or functions specified in one or more blocks Figure 1 one or more processes and / or functions specified in one or more blocks

[0158] The above descriptions are only the preferred embodiments of the present application, not intended to limit the present application to other forms described. Any person familiar with the art can make changes or modifications to the above-described technical content as equivalent embodiments. However, any simple modification, equivalent change and modification made to the above embodiments without departing from the technical solution of the present application, according to the technical essence of the present application, still belongs to the protection scope of the technical solution of the present application.

Claims

1. A method of generating a power monitoring work ticket, the method comprising: Includes the following steps; Step S1: Construct a security measures library; Step S2: Extract entity relationships from the work content to obtain a work content subgraph; the method includes using natural language processing technology to accurately convert the informal expressions of inspection or maintenance intentions of dispatchers into a standardized subgraph representation. Step S3: Search the work content subgraph in the knowledge graph to obtain the associated work item sequence and the risk entity set of the equipment, process and location associated with the work item; Step S4: Retrieve the risk entity from the knowledge graph and the security measure base to obtain a dictionary of security measure items; Step S5: Fill in the work ticket according to the work item sequence, risk entity set, and safety measure dictionary; Step S1 involves building a security measures library, which specifically includes the following steps: Step S1A1: Obtain the security measures corpus and perform word segmentation on the security measures corpus; Step S1A2: Use the LSTM method to determine the first information category of the security measures corpus; determine whether the word segmentation result of the security measures corpus contains words of the first information. Step S1A3: Use the LSTM method to determine the second information category of the security measures corpus; determine whether the word segmentation results of the security measures corpus contain words with second information. Step S1A4: Add the security measure corpus to the security measure library according to the first information category and the second information category; In step S2, entity relations are extracted from the work content to obtain a work content subgraph, which specifically includes the following steps: Step S2A1: Use the entity relation extraction tool to extract entity relations from the work content to obtain at least 0 triples; Step S2A2: When the number of triples is 0, the work content is treated as a single work content entity, forming a subgraph with only one entity; Step S2A3: When the number of triples is greater than 0, the extracted triples are used as subgraphs. In step S4, the dictionary of safety measures is specifically structured as follows: The safety measure dictionary is stored in the form of {key:value}, where the key is one of three types: before work, during work, and after work, and the value is a list type that stores the safety measure content at different times; In step S4, the risk entity is retrieved from the knowledge graph and the security measure base to obtain a list of security measures, which specifically includes the following steps: Step S4A1: Retrieve entity types from the risk entities. If there are entities of business system and process types, search for software type and general type security measures in the security measures library and add them to the security measures list. Step S4A2: If there are no entities of business system and process types, search for hardware type and general type security measures in the security measures library and add them to the security measures list. Step S4A3: For the risk entities in the risk entity list, retrieve the safety measures entities in the knowledge graph that have the current risk entity as the tail entity and that have a relationship with it. Step S4A4: For the security entity, retrieve a set of risk entities in the knowledge graph that have the current security entity as the head entity and that have related relationships; Step S4A5, if the risk entity set is a subset of the risk entity in the work ticket, the safety measure entity name is added to the safety list, and the time is in the work.

2. The method for generating a power monitoring work ticket according to claim 1, wherein: In step S1, the safety measure library is constructed, which is specifically composed as follows: The safety measure library centrally stores the most basic and widely applicable safety measure clauses suitable for power dispatching automation operation scenarios; the library structurally organizes the measure content to ensure its clear applicability and guidance significance at different time stages and for different object types; The first information is time dimension, which is divided into before work, during work and after work: Before work, preventive safety measures are recorded, including environmental preparation, equipment inspection, notification of duty station, data backup, to ensure that all conditions before work meet the safety requirements; During work, process control safety measures are included, including operation sequence, isolation verification, to ensure that the risk is effectively controlled during the work process; After work, it involves restorative and summary safety measures, including equipment recovery after work, site cleaning, equipment verification, to ensure that the system returns to normal state after the work activity is completed; The second information is object type, which is divided into software level, hardware level and general level: Software level: covering safety measures for business systems, application programs, processes, databases and data table software components, including system upgrade and patch management, data backup and recovery, to ensure the stable operation of software systems and information security; Hardware level: including safety measures for hardware devices and infrastructure, including personal protective equipment, do not move other running devices, to ensure the safe and reliable operation of hardware facilities; General measures: specify common safety requirements applicable to all work activities, including notification of duty station, these measures do not distinguish specific software or hardware objects; Each type of safety measure clearly defines its applicable situation and content requirements.

3. The method for generating a power monitoring work ticket of claim 1, wherein: In steps S1A2 and S1A3, the LSTM method is specifically composed as follows: The LSTM method is composed of a vector representation layer, an LSTM layer and a linear layer; The original safety measure corpus Text is input into the vector representation layer to obtain E = Embedding(Text) = (e1, e2,..., en) n ), where n is the corpus length; denotes the d-dimensional embedding vector for the i-th word; The vector sequence E is input to an LSTM layer, which processes one embedding vector e at each time step t t , based on the hidden state h t-1 and cell state C t-1 from the previous time step, updates the hidden state h t and cell state C t for the current time step; outputs the output vector O t for the current time step = LSTM(e t , h t-1 , C t-1 ), until the entire sequence information is processed, obtaining the output vector O T for the last time step; Linear layer for classification O T For input, get different class probability distribution.

4. The method for generating a power monitoring work ticket of claim 1, wherein: In step S2A1, the entity relationship extraction method is a method based on Transformer and a dictionary rule matching method to obtain entity relationship triples; The triple is specifically composed as follows: The triple is in the form of <h, r, t>, h represents the head entity, r represents the relationship, and t represents the tail entity; Where h belongs to one of work content entity, fault entity, and device entity; Where r belongs to one of the involving relationship and the causing relationship; Where t belongs to one of the data table entity, the site entity, the business system entity, the company entity, the role entity, the process entity, the device entity, and the fault entity.

5. The method for generating a power monitoring work ticket of claim 1, wherein: In step S3, the work content subgraph is searched in the knowledge graph to obtain the associated work item sequence, which specifically includes the following steps: Step S3A1, retrieving the sub-graph in the knowledge graph for locating the fault entity and the work content entity, and locating the work content entity E in the knowledge graph content ; Step S3A2, retrieving in the knowledge graph the E content There is a work item entity that resolves the relationship, whose name is stored in the worklist; Step S3A3, search in the knowledge graph with the current work item as the head entity and the risk entity name with the involving relationship stored in the risk entity list; Step S3A4, retrieve in the knowledge graph the work item entities with the current work item as the head entity and the existence of a sequential relationship, store the entity names in the worklist, and perform step S3A3 until there are no new work items; In step S3, the knowledge graph is specifically composed as follows: The knowledge graph includes but is not limited to the following entity relationship types: The entity types include faults, work contents, devices, business systems, processes, work items, and safety measure entities; The relationship types include involvement, triggering, solving, and sequential relationships; The information corresponding to the entity is stored in the form of entity attributes, and the entity attributes at least include an entity id and an entity name, wherein the entity id is a unique identifier of the entity, and the entity name stores the content information of the entity, wherein the entity name of the work item entity is the content of the work item; In step S3, the risk entity includes device entities, business system entities, process entities, and site entities.

6. The method of claim 5, wherein: In step S3, the knowledge graph is stored in a graph database for easy updating and searching; the knowledge graph includes the association between existing account information, is used to locate maintenance or repair content, and specifically further includes: The entity types include personnel, sites, files, and instruction entities; The relationship types include belonging, being located, work members, and responsibility relationships; The attributes of the device entity include the manufacturer and the IP address; In step S3, the risk entity includes device entities, business system entities, and process entities; the risk entity also includes instruction entities and site entities.

7. A power monitoring work ticket generation system for implementing the method as claimed in any one of claims 1-6, characterized by: The computer program instructions stored on the memory and capable of being executed by the processor to realize the generation method of the power monitoring work ticket, specifically include a safety measure library, an intention recognition module, a work item retrieval module, a safety measure retrieval module, and a work ticket generation module; the safety measure library is obtained by mining historical data and practical experience; The maintenance or repair intention of the operator is identified, and a standardized subgraph is obtained by using an entity relationship extraction method; the knowledge graph is searched using the subgraph to obtain a work item and a risk item involved; The risk item is retrieved to obtain a safety measure item, and finally a work ticket is generated; The safety measure library management module is used to manage existing formatted safety measures, and the safety measures are added, deleted, modified, and inquired; The knowledge graph management module is constituted using the interface provided by the graph database, and is used to update the existing knowledge graph; The intention recognition module uses an entity relationship extraction algorithm to extract triples from the text input by the dispatcher, and inputs the extracted triples into the work item retrieval module; The work item retrieval module retrieves the knowledge graph to obtain a work item and a risk item, and the order is determined by the order stored in the knowledge graph; the dispatcher can modify the order by dragging the work item text; The safety measure retrieval module retrieves the knowledge graph and the safety measure library according to the risk item to obtain a safety measure item, and can also modify the default order by dragging the safety measure item text; The work ticket generation module generates a visual work ticket file according to the obtained work item, risk item, and safety measure item.

Images