An intelligent collaborative optimization method and system for main network maintenance plan tasks

Abstract

The embodiment of the application provides an intelligent collaborative optimization method and system for a main network maintenance plan task, and belongs to the technical field of power grid maintenance. The intelligent collaborative optimization method comprises the following steps: constructing maintenance data; constructing a main and distribution network integrated topology knowledge graph based on the maintenance data; combining the knowledge graph and extracting key information according to maintenance requirements to generate at least multiple maintenance plan drafts containing power failure range description, influence distribution network lines, associated defects and hidden dangers, and forming a formal power failure plan, live working plan or start power transmission plan after confirmation or adjustment through human-computer interaction; and based on the reported plan, taking the maximization of one stop multiple uses, the minimization of frequent stop risks and the minimization of risk superposition as optimization objectives, dynamically balancing the main network and distribution network plans. The intelligent system optimization method can realize data acquisition, intelligent collaboration of main and distribution network plans, and significantly improves the efficiency, scientificity and accuracy of maintenance plan compilation.

Description

Technical Field

[0001] This invention relates to the field of power grid maintenance technology, and more specifically to an intelligent collaborative optimization method and system for main grid maintenance planning tasks. Background Technology

[0002] With the continuous expansion of the power grid and the increasing complexity of its structure, the preparation and management of main grid equipment maintenance plans face enormous challenges. Main grid maintenance plans not only involve the shutdown and reactivation arrangements for key primary equipment such as substation transformers, busbars, and transmission lines, but also require full consideration of their impact on the power supply reliability of downstream distribution networks. This is a typical multi-objective, multi-constraint, and strongly coupled collaborative decision-making problem.

[0003] When compiling maintenance plans, planners need to manually review and summarize multi-source data from multiple independent business systems, such as production management systems, dispatch automation systems, and power supply service command systems. This involves frequent switching between systems and multiple terminals, along with manual analysis, which is not only time-consuming and labor-intensive but also prone to information omissions or biases, making it difficult to guarantee the completeness and accuracy of the plan. Therefore, an intelligent collaborative optimization method and system for main grid maintenance planning tasks is needed, capable of data acquisition and intelligent collaboration between main and distribution network plans, significantly improving the efficiency, scientific rigor, and accuracy of maintenance plan compilation. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent collaborative optimization method and system for main grid maintenance planning tasks. This intelligent system optimization method can realize data acquisition and intelligent collaboration of main and distribution network plans, significantly improving the efficiency, scientificity and accuracy of maintenance plan preparation.

[0005] To achieve the above objectives, embodiments of the present invention provide an intelligent collaborative optimization method for main network maintenance planning tasks, the intelligent collaborative optimization method comprising: Extract information on potential equipment hazards, defect records, frequent outages, test information, and real-time power grid operation data to construct maintenance data; Based on the maintenance data, an integrated topology knowledge graph of the main and distribution networks is constructed. The entity nodes in the knowledge graph include substations, main network lines, distribution network lines, main transformers, circuit breakers, important users and their power supply path relationships. The attributes include the equipment's historical power outage frequency, defect status, power supply range and load characteristics. By combining the knowledge graph and extracting key information based on maintenance needs, multiple draft maintenance plans are generated, including at least a description of the power outage area, affected distribution network lines, associated defects and hidden dangers. After confirmation or adjustment through human-computer interaction, formal power outage plans, live-line work plans or power restoration plans are formed. Based on the submitted plans, the main network and distribution network plans are dynamically balanced with the optimization objectives of maximizing the use of multiple outages, minimizing the risk of frequent outages, and minimizing the superposition of risks.

[0006] Optionally, an integrated topology knowledge graph of the main and distribution networks is constructed and dynamically updated, including: The coordinates and connection relationships of the main network lines and distribution network lines are extracted from the GIS system. The topological connection relationships of substations, buses, main transformers, circuit breakers, and disconnectors, as well as the power supply path from the main network to the distribution network, are obtained from the dispatch automation system. Equipment ledgers, defect records, and hidden danger information are obtained from the production management system. Important user files, power outage history, and load data are obtained from the power supply service command system and the power consumption and procurement system. All of the above data are unified into a standardized model of equipment-attribute-relationship. Based on the data unified into a standardized model of device-attribute-relationship, entity types are defined, including: substation, main grid line, distribution network line, main transformer, busbar, circuit breaker, and important user; Based on the defined entity types and the unified data, create entity nodes, and build a unique identifier and attributes for each entity node; Construct a basic topology graph based on the topological relationships between entity nodes; For the basic topology map, locate the low-voltage busbar of each main transformer, obtain all its distribution network outgoing lines, and establish the implicit relationship between the main transformer, power supply, and distribution network lines. For important users in the basic topology map, trace their power supply points to the substation main transformer, establish dual power supply paths, and mark the power supply points on the opposite side. Treat defect records as event nodes, associate them with corresponding devices, establish the device-existence-defect relationship, and use defect level, description, etc. as attributes. In this way, an integrated topology knowledge graph of the main and distribution networks is constructed.

[0007] Optionally, by combining the knowledge graph and automatically extracting key information based on maintenance needs, multiple draft maintenance plans are generated, including at least a description of the power outage area, affected distribution network lines, associated defects, and potential hazards. Obtain maintenance requirements and perform target parsing on the maintenance requirements to extract the target equipment type, equipment number, and maintenance nature; Starting from the target device, query the set of devices that need to be de-energized, and filter the nodes of type distribution network line within the power outage range to form a list of affected distribution network lines; The system queries the relationship between the target equipment and critical equipment within the power outage area that have defects or potential hazards, and obtains the defect number, level, description, and date of discovery. Based on the power outage area and the distribution network lines involved in the list of affected distribution network lines, query the power supply-important user relationship, list the affected users and their levels, and fill the queried information into the template to form a basic draft. The template includes the power outage equipment, description of the power outage area, suggested power outage start / end time window, list of affected distribution network lines, summary of associated defects and hidden dangers, whether power transfer is required, and work content. The large language model is invoked, and prompts are input to perform compliance checks and modifications on the basic draft to obtain the maintenance plan draft.

[0008] Optionally, dynamic balancing of the main network and distribution network plans can be performed, including: Obtain the plans that need to be filled in, and obtain the set of maintenance plans for the main grid and distribution network to be balanced based on the filled-in plans. Each decision variable in the set of maintenance plans for the main grid and distribution network includes start time, construction period and merging identifier and corresponding time window, power outage range, original plans included, and associated defects. Based on the decision variables, plans with the same merging identifier can be merged into the same power outage event. Define the goal of maximizing the use of one outage for multiple purposes, minimizing the risk of frequent outages, and minimizing the risk superposition as optimization objectives. Based on the power grid topology knowledge graph, the projects with close electrical connections are pre-clustered into candidate merging groups and assigned initial merging representations. A common start time is randomly generated for each merging group and the project duration within the group is unified to generate an initial population. For each individual in the population, the constraint violation degree is calculated and the objective function is evaluated. Individuals that violate hard constraints are marked as infeasible. The constraint violation degree includes project integrity constraints, hard boundary constraints for schedule rationality, hard constraints for time windows, and hard constraints for safe topology. Based on non-dominated sorting and crowding distance, tournament selection is used to select the parent generation; Perform crossover and mutation operations on the parent individuals to generate offspring, and apply repair operations to the offspring to ensure that the constraints are met; By merging the parent and offspring generations, and performing non-dominant ranking and elite retention based on the principle of constraint and dominance, a new generation of population is formed. Repeat the above steps until the termination condition is met, and select a typical scheme from the final non-dominated solution set according to the preference weight or inflection point distance, and output the merged power outage plan list to complete the dynamic balance of the main grid and distribution network plans.

[0009] Optionally, the objective functions for maximizing the goal of multiple uses of a single stop, minimizing the risk of frequent stops, and minimizing the cumulative risk include: The objective function for maximizing the goal of "one stop, multiple uses" can be obtained through formula (1): Formula (1) in, This represents a collection of maintenance plans for the main grid and distribution network. The objective function that maximizes the goal of one stop and multiple uses. Indicates the number of distinct merge groups. Indicates the merge identifier; The preset frequency stop risk weight is obtained through formula (2): Formula (2) in, Indicates device Preset frequency stop risk weight, Indicates the importance of the equipment. Indicates the expected number of power outages, of which This indicates the number of power outages in the past 12 months. Indicates the frequency stop threshold; The objective function for minimizing the risk of frequent shutdowns can be obtained through formula (3): Formula (3) in, Let the objective function represent the goal of minimizing the risk of frequent shutdowns. Indicates the number of devices. Indicates device Should an additional power outage be added? The objective function for minimizing the risk superposition objective is obtained through formula (4): Formula (4) in, The objective function represents the goal of minimizing the superposition of risks. Indicates the first Grid-like risks include one of the following: risk penalties for complete outage of dual power sources for important users, risk penalties for heavy overload of transfer lines, risk penalties for the entire station, a single main transformer, or a single busbar, and risk penalties for multiple planned conflicts of the same transfer lines.

[0010] Optionally, the intelligent collaborative optimization method further includes: constructing corresponding intelligent agents based on the jurisdiction of the regional dispatch, distribution dispatch, and county dispatch respectively; when the main grid power outage plan affects the distribution network, initiating a collaborative request; negotiating and resolving the issue based on the shared situation and dynamic balance results; triggering a multi-department review process after reaching a consensus; and generating a maintenance application form.

[0011] Optionally, the intelligent collaborative optimization method further includes: leaving traces at each operation node throughout the entire process to form a full-cycle traceability record from data entry, balancing, collaboration, review to push execution.

[0012] On the other hand, the present invention can also provide an intelligent collaborative optimization system for main network maintenance planning tasks, the intelligent collaborative optimization system comprising: The data construction module is used to extract information on equipment hazards, defect records, frequent outage events, test information, and real-time power grid operation data to construct maintenance data. The execution module is used to execute an intelligent collaborative optimization method for main network maintenance planning tasks, as described above, based on the constructed maintenance data.

[0013] On the other hand, the present invention may also provide a processor for running a program, wherein the program is executed to perform: an intelligent collaborative optimization method for main network maintenance planning tasks as described above.

[0014] Through the above technical solution, this invention provides an intelligent collaborative optimization method and system for main grid maintenance planning tasks. By extracting information on equipment hazards, defect records, frequent outage events, test information, and real-time grid operation data, maintenance data can be constructed. Based on the constructed maintenance data, an integrated topology knowledge graph of the main and distribution networks can be built. Entity nodes in this knowledge graph can include substations, main grid lines, distribution network lines, main transformers, circuit breakers, important users, and power supply path relationships. Attributes in this knowledge graph can include historical outage frequency, defect status, power supply range, and load characteristics. Based on the acquired knowledge graph, and by extracting key information according to maintenance needs, multiple draft maintenance plans can be generated, each containing at least a description of the outage range, affected distribution network lines, associated defects, and hazards. These draft plans can be formally submitted as outage plans, live-line work plans, or power-on / off plans after confirmation or adjustment through human-computer interaction. Based on the submitted plans, the main grid and distribution network plans can be dynamically balanced with optimization objectives of maximizing multi-purpose use of outages, minimizing frequent outage risks, and minimizing risk superposition. This intelligent system optimization method enables data acquisition and intelligent coordination of main and distribution network plans, significantly improving the efficiency, scientific nature, and accuracy of maintenance plan preparation.

[0015] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate embodiments of the present invention and form part of the specification. They are used together with the following detailed description to explain the embodiments of the present invention, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of an intelligent collaborative optimization method for main network maintenance planning tasks according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating the construction of a knowledge graph for an intelligent collaborative optimization method oriented towards main network maintenance planning tasks, according to an embodiment of the present invention. Detailed Implementation

[0017] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the scope of the present invention.

[0018] In the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, they do not mean that the applicant has used or necessarily used the solution.

[0019] Figure 1 This is a flowchart of an intelligent collaborative optimization method for main network maintenance planning tasks according to an embodiment of the present invention. In this invention, the intelligent collaborative optimization process may include: In step S1, records of potential equipment hazards, defects, frequent outages, test information, and real-time power grid operation data are extracted to construct maintenance data.

[0020] In step S2, an integrated topology knowledge graph of the main and distribution networks is constructed based on the maintenance data. The entity nodes in the knowledge graph include substations, main network lines, distribution network lines, main transformers, circuit breakers, important users and their power supply path relationships. The attributes include the equipment's historical power outage frequency, defect status, power supply range and load characteristics.

[0021] In step S3, the knowledge graph is combined with key information extracted according to maintenance needs to generate multiple draft maintenance plans that include at least a description of the power outage area, the affected distribution network lines, related defects and hidden dangers. After confirmation or adjustment through human-computer interaction, a formal power outage plan, live-line work plan or power restoration plan is formed.

[0022] In step S4, based on the submitted plan, the main network and distribution network plans are dynamically balanced with the optimization objectives of maximizing multiple uses of a single outage, minimizing the risk of frequent outages, and minimizing the superposition of risks.

[0023] In this invention, during intelligent collaborative optimization, information on equipment hazards, defect records, frequent outage events, test information, and real-time power grid operation data can be extracted to construct maintenance data. Based on the constructed maintenance data, an integrated topology knowledge graph of the main and distribution networks can be built. Entity nodes in this knowledge graph can include substations, main grid lines, distribution network lines, main transformers, circuit breakers, important users, and power supply path relationships. Attributes in this knowledge graph can include historical outage frequency, defect status, power supply range, and load characteristics. Based on the acquired knowledge graph, and by extracting key information according to maintenance needs, multiple draft maintenance plans can be generated, each containing at least a description of the outage range, affected distribution network lines, associated defects, and hazards. These draft plans can be formally submitted as outage plans, live-line work plans, or power-on / off plans after confirmation or adjustment through human-computer interaction. Based on the submitted plans, the main grid and distribution network plans can be dynamically balanced with the optimization objectives of maximizing multi-use during outages, minimizing frequent outage risks, and minimizing risk superposition. This intelligent system optimization method enables data acquisition and intelligent coordination of main and distribution network plans, significantly improving the efficiency, scientific nature, and accuracy of maintenance plan preparation.

[0024] In one embodiment of the present invention, such as Figure 2 As shown, the process of constructing a knowledge graph may include: In step S5, the coordinates and connection relationships of the main network lines and distribution network lines are extracted from the GIS system; the topological connection relationships of substations, buses, main transformers, circuit breakers, disconnectors, etc., as well as the power supply path from the main network to the distribution network, are obtained from the dispatch automation system; equipment ledgers, defect records, and hidden danger information are obtained from the production management system; and important user files, power outage history, and load data are obtained from the power supply service command system and the power consumption and procurement system. The above data are then unified into a standardized model of equipment-attribute-relationship.

[0025] In step S6, based on the data unified into a standardized model of device-attribute-relationship, the entity types are defined, including: substation, main grid line, distribution network line, main transformer, busbar, circuit breaker, and important user.

[0026] In step S7, entity nodes are created based on the defined entity type and the unified data, and a unique identifier and attributes are constructed for each entity node.

[0027] In step S8, a basic topology graph is constructed based on the topological relationships between entity nodes.

[0028] In step S9, for the basic topology map, the low-voltage side bus of each main transformer is located, all its distribution network outgoing lines are obtained, and the implicit relationship between the main transformer, power supply and distribution network lines is established; for important users in the basic topology map, their power supply points are traced to the substation main transformer, a dual power supply path is established, and the power supply points on the opposite side are marked; defect records are used as event nodes, associated with the corresponding equipment, and a device-existence-defect relationship is established, and the defect level, description and other attributes are used as attributes, thereby constructing an integrated topology knowledge graph of the main and distribution networks.

[0029] In this invention, when constructing a knowledge graph, the coordinates and connection relationships of main grid lines and distribution network lines can be extracted from a GIS system. Topological connection relationships of substations, buses, main transformers, circuit breakers, and disconnectors, as well as the power supply path from the main grid to the distribution network, can also be obtained from a dispatch automation system. Equipment ledgers, defect records, and hidden danger information can be obtained from a production management system. Important user files, power outage history, and load data can be obtained from a power supply service command system and a power consumption and procurement system. All of the above data is unified into a standardized model of equipment-attribute-relationship. Based on the data unified into this standardized model, entity types can be defined, including: substations, main grid lines, distribution network lines, main transformers, buses, circuit breakers, and important users. Entity nodes can be created based on the defined entity types and the unified data, and each entity node can be assigned a unique identifier and attributes. Based on the topological relationships between entity nodes, a basic topology graph can be constructed. After constructing the knowledge graph, the low-voltage busbar of each main transformer can be located within the basic topology graph, and all its distribution network outgoing lines can be obtained, establishing the implicit relationship between the main transformer, power supply, and distribution network lines. For important users in the basic topology graph, their power supply points can be traced to the substation main transformer, establishing dual power supply paths and marking the opposite power supply points. Furthermore, defect records can be used as event nodes, associated with corresponding devices, establishing a device-existence-defect relationship, and assigning defect level and description as attributes. Through these processes, an integrated main and distribution network topology knowledge graph can be constructed.

[0030] In one embodiment of the present invention, the process of generating a draft maintenance plan may include: In step S10, maintenance requirements are obtained, and target parsing is performed on the maintenance requirements to parse out the target equipment type, equipment number, and maintenance nature.

[0031] In step S11, starting from the target device, the set of devices that need to be de-energized is queried, and nodes of type distribution network line are filtered within the power outage range to form a list of affected distribution network lines.

[0032] In step S12, the relationship between the target device and the critical equipment within the power outage range that have defects or potential hazards is queried, and the defect number, level, description, and discovery date are obtained.

[0033] In step S13, based on the power outage area and the distribution network lines involved in the list of affected distribution network lines, the power supply-important user relationship is queried, and the users and levels that may be affected are listed. The queried information is then filled into the template to form a basic draft. The template includes the power outage equipment, a description of the power outage area, a power outage start / end time suggestion window, a list of affected distribution network lines, a summary of associated defects and hidden dangers, whether power transfer is required, and the work content.

[0034] In step S14, the large language model is invoked, prompts are input, and the basic draft is checked for compliance and modified to obtain the maintenance plan draft.

[0035] In this invention, when generating a draft maintenance plan, maintenance requirements can be obtained and analyzed to identify the target equipment type, equipment number, and maintenance nature. Then, starting with the target equipment, the set of equipment requiring power outages can be queried, and nodes of type distribution network line can be filtered within the outage area to form a list of affected distribution network lines. The relationship between defects or potential hazards on the target equipment and key equipment within the outage area can be queried, and defect numbers, levels, descriptions, and discovery dates can be obtained. Based on the outage area and the affected distribution network lines, the power supply-important user relationship can be queried, and potentially affected users and their levels can be listed. The queried information is then filled into a template to form a basic draft. This template may include outage equipment, outage area description, suggested start / end times, a list of affected distribution network lines, a summary of associated defects and potential hazards, whether power transfer is required, and work content. A large language model can be invoked, prompts can be entered, and the basic draft can be checked for compliance and modified to obtain the final draft maintenance plan.

[0036] In one embodiment of the present invention, the process of performing dynamic balancing may include: In step S15, the plan that needs to be filled in is obtained, and the set of maintenance plans for the main grid and distribution network to be balanced is obtained according to the filled-in plan. Each decision variable in the set of maintenance plans for the main grid and distribution network includes start time, construction period and merging identifier and corresponding time window, power outage range, original plan included, and associated defects. Based on the decision variables, plans with the same merging identifier can be merged into the same power outage event. The optimization objectives are defined as maximizing the use of one outage for multiple purposes, minimizing the risk of frequent outages, and minimizing the risk superposition.

[0037] In step S16, based on the power grid topology knowledge graph, the plans with close electrical connections are pre-clustered into candidate merging groups and assigned initial merging representations. A common start time is randomly generated for each merging group and the construction period within the group is unified to generate an initial population.

[0038] In step S17, constraint violation degree is calculated and objective function is evaluated for each individual in the population. Individuals that violate hard constraints are marked as infeasible. Constraint violation degree includes project integrity constraints, schedule rationality hard boundary constraints, time window hard constraints, and safety topology hard constraints.

[0039] In step S18, a tournament selection method is used to select the parent generation based on the non-dominated sort and the crowding distance.

[0040] In step S19, crossover and mutation operations are performed on the parent individuals to generate offspring, and repair operations are applied to the offspring to ensure that the constraints are met.

[0041] In step S20, the parent and offspring generations are merged, and non-dominated sorting and elite retention are performed according to the constraint dominance principle to form a new generation of population.

[0042] In step S21, the above steps are repeated until the termination condition is met, and a typical scheme is selected from the final non-dominated solution set according to the preference weight or inflection point distance. The merged power outage plan list is output to complete the dynamic balance of the main grid and distribution network plans.

[0043] In this invention, during the dynamic balancing of the main grid and distribution network plans, the plans that need to be submitted can be obtained, and a set of main grid and distribution network maintenance plans to be balanced can be obtained based on the submitted plans. Each decision variable in the set of main grid and distribution network maintenance plans includes start time, duration, merging identifier and corresponding time window, outage range, original plans, associated defects, etc. Based on decision variables, plans with the same merging identifier can be merged into the same outage event. Then, the optimization objectives can be defined as maximizing the use of one outage for multiple purposes, minimizing the risk of frequent outages, and minimizing the risk superposition. Based on the power grid topology knowledge graph, plans with close electrical connections can be pre-clustered into candidate merging groups and assigned initial merging representations. A common start time can be randomly generated for each merging group, and the duration within the group can be unified to generate an initial population. For each individual in the population, constraint violation degree calculation and objective function evaluation can be performed. Individuals that violate hard constraints are marked as infeasible. Constraint violation degree can include project integrity constraints, hard boundary constraints for reasonable duration, hard constraints for time windows, and hard constraints for safe topology. Parent generations can be selected using tournament selection based on non-dominated sorting and crowding distance. Crossover and mutation operations can be performed on parent individuals to generate offspring, and repair operations can be applied to the offspring to ensure that constraints are met. Parent and offspring generations can be merged, and non-dominated sorting and elite retention can be performed according to the constraint dominance principle to form a new generation of population. The above steps are repeated until the termination condition is met, and typical schemes are selected from the final non-dominated solution set according to preference weights or inflection point distance. The merged power outage plan list is output to achieve dynamic balance of the main grid and distribution network plans.

[0044] In one embodiment of the present invention, the objective function for maximizing the goal of one-stop multi-use can be obtained by formula (1): Formula (1) in, This represents a collection of maintenance plans for the main grid and distribution network. The objective function that maximizes the goal of one stop and multiple uses. Indicates the number of distinct merge groups. This indicates a merge identifier.

[0045] The preset frequency stop risk weight can also be obtained through formula (2): Formula (2) in, Indicates equipment Preset frequency stop risk weight, Indicates the importance of the equipment. Indicates the expected number of power outages, of which This indicates the number of power outages in the past 12 months. This indicates the frequency stop threshold.

[0046] The objective function for minimizing the risk of frequent shutdowns can be obtained through formula (3): Formula (3) in, Let the objective function represent the goal of minimizing the risk of frequent shutdowns. Indicates the number of devices. Indicates equipment Should an additional power outage be added?

[0047] Finally, the objective function for minimizing the risk superposition objective can be obtained through formula (4): Formula (4) in, The objective function represents the goal of minimizing the superposition of risks. Indicates the first Grid-like risks include one of the following: risk penalties for complete outage of dual power sources for important users, risk penalties for heavy overload of transfer lines, risk penalties for the entire station, a single main transformer, or a single busbar, and risk penalties for multiple planned conflicts of the same transfer lines.

[0048] In one embodiment of the present invention, the intelligent collaborative optimization method may further include: constructing corresponding intelligent agents according to the jurisdiction of the local dispatch, distribution dispatch, and county dispatch respectively; when the main grid power outage plan affects the distribution network, a collaborative request can be initiated; negotiation and resolution can be carried out based on the results of shared situation and dynamic balance; after reaching a consensus, a multi-department review process is triggered, and a maintenance application form is generated.

[0049] In one embodiment of the present invention, the intelligent collaborative optimization method may further include: leaving traces at each operation node of the entire process to form a full-cycle traceability record from filling in, balancing, collaborating, reviewing to pushing and executing.

[0050] On the other hand, the present invention can also provide an intelligent collaborative optimization system for main network maintenance planning tasks, the intelligent collaborative optimization system may include: The data construction module is used to extract information on equipment hazards, defect records, frequent outage events, test information, and real-time power grid operation data to construct maintenance data. The execution module is used to execute an intelligent collaborative optimization method for main network maintenance planning tasks, as described above, based on the constructed maintenance data.

[0051] On the other hand, the present invention can also provide a processor for running a program, wherein the program is executed to perform: an intelligent collaborative optimization method for main network maintenance planning tasks as described above.

[0052] Through the above technical solution, this invention provides an intelligent collaborative optimization method and system for main grid maintenance planning tasks. By extracting information on equipment hazards, defect records, frequent outage events, test information, and real-time grid operation data, maintenance data can be constructed. Based on the constructed maintenance data, an integrated topology knowledge graph of the main and distribution networks can be built. Entity nodes in this knowledge graph can include substations, main grid lines, distribution network lines, main transformers, circuit breakers, important users, and power supply path relationships. Attributes in this knowledge graph can include historical outage frequency, defect status, power supply range, and load characteristics. Based on the acquired knowledge graph, and by extracting key information according to maintenance needs, multiple draft maintenance plans can be generated, each containing at least a description of the outage range, affected distribution network lines, associated defects, and hazards. These draft plans can be formally submitted as outage plans, live-line work plans, or power-on / off plans after confirmation or adjustment through human-computer interaction. Based on the submitted plans, the main grid and distribution network plans can be dynamically balanced with optimization objectives of maximizing multi-purpose use of outages, minimizing frequent outage risks, and minimizing risk superposition. This intelligent system optimization method enables data acquisition and intelligent coordination of main and distribution network plans, significantly improving the efficiency, scientific nature, and accuracy of maintenance plan preparation.

[0053] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0054] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0055] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0056] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0057] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0058] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0059] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0060] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0061] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. An intelligent collaborative optimization method for main network maintenance planning tasks, characterized in that, The intelligent collaborative optimization method includes: Extract information on potential equipment hazards, defect records, frequent outages, test information, and real-time power grid operation data to construct maintenance data; Based on the maintenance data, an integrated topology knowledge graph of the main and distribution networks is constructed. The entity nodes in the knowledge graph include substations, main network lines, distribution network lines, main transformers, circuit breakers, important users and their power supply path relationships. The attributes include the equipment's historical power outage frequency, defect status, power supply range and load characteristics. By combining the knowledge graph and extracting key information based on maintenance needs, multiple draft maintenance plans are generated, including at least a description of the power outage area, affected distribution network lines, associated defects and hidden dangers. After confirmation or adjustment through human-computer interaction, formal power outage plans, live-line work plans or power restoration plans are formed. Based on the submitted plans, the main network and distribution network plans are dynamically balanced with the optimization objectives of maximizing the use of multiple outages, minimizing the risk of frequent outages, and minimizing the superposition of risks.

2. The intelligent collaborative optimization method according to claim 1, characterized in that, Construct and dynamically update an integrated topology knowledge graph of the main and distribution networks, including: The coordinates and connection relationships of the main network lines and distribution network lines are extracted from the GIS system. The topological connection relationships of substations, buses, main transformers, circuit breakers, and disconnectors, as well as the power supply path from the main network to the distribution network, are obtained from the dispatch automation system. Equipment ledgers, defect records, and hidden danger information are obtained from the production management system. Important user files, power outage history, and load data are obtained from the power supply service command system and the power consumption and procurement system. All of the above data are unified into a standardized model of equipment-attribute-relationship. Based on the data unified into a standardized model of device-attribute-relationship, entity types are defined, including: substation, main grid line, distribution network line, main transformer, busbar, circuit breaker, and important user; Based on the defined entity types and the unified data, create entity nodes, and build a unique identifier and attributes for each entity node; Construct a basic topology graph based on the topological relationships between entity nodes; For the basic topology map, locate the low-voltage busbar of each main transformer, obtain all its distribution network outgoing lines, and establish the implicit relationship between the main transformer, power supply, and distribution network lines. For important users in the basic topology map, trace their power supply points to the substation main transformer, establish dual power supply paths, and mark the power supply points on the opposite side. Treat defect records as event nodes, associate them with corresponding devices, establish the device-existence-defect relationship, and use defect level, description, etc. as attributes. In this way, an integrated topology knowledge graph of the main and distribution networks is constructed.

3. The intelligent collaborative optimization method according to claim 1, characterized in that, By combining the knowledge graph and automatically extracting key information based on maintenance needs, multiple draft maintenance plans are generated, each containing at least a description of the power outage area, affected distribution network lines, associated defects, and potential hazards. Obtain maintenance requirements and perform target parsing on the maintenance requirements to extract the target equipment type, equipment number, and maintenance nature; Starting from the target device, query the set of devices that need to be de-energized, and filter the nodes of type distribution network line within the power outage range to form a list of affected distribution network lines; The system queries the relationship between the target equipment and critical equipment within the power outage area that have defects or potential hazards, and obtains the defect number, level, description, and discovery date. Based on the power outage area and the distribution network lines involved in the list of affected distribution network lines, query the power supply-important user relationship, list the affected users and their levels, and fill the queried information into the template to form a basic draft. The template includes the power outage equipment, description of the power outage area, suggested power outage start / end time window, list of affected distribution network lines, summary of associated defects and hidden dangers, whether power transfer is required, and work content. The large language model is invoked, and prompts are input to perform compliance checks and modifications on the basic draft to obtain the maintenance plan draft.

4. The intelligent collaborative optimization method according to claim 1, characterized in that, Dynamic balancing of the main network and distribution network plans includes: Obtain the plans that need to be filled in, and obtain the set of maintenance plans for the main grid and distribution network to be balanced based on the filled-in plans. Each decision variable in the set of maintenance plans for the main grid and distribution network includes start time, construction period and merging identifier and corresponding time window, power outage range, original plans included, and associated defects. Based on the decision variables, plans with the same merging identifier can be merged into the same power outage event. Define the goal of maximizing the use of one outage for multiple purposes, minimizing the risk of frequent outages, and minimizing the risk superposition as optimization objectives. Based on the power grid topology knowledge graph, the projects with close electrical connections are pre-clustered into candidate merging groups and assigned initial merging representations. A common start time is randomly generated for each merging group and the project duration within the group is unified to generate an initial population. For each individual in the population, the constraint violation degree is calculated and the objective function is evaluated. Individuals that violate hard constraints are marked as infeasible. The constraint violation degree includes project integrity constraints, hard boundary constraints for schedule rationality, hard constraints for time windows, and hard constraints for safe topology. Based on non-dominated sorting and crowding distance, tournament selection is used to select the parent generation; Perform crossover and mutation operations on the parent individuals to generate offspring, and apply repair operations to the offspring to ensure that the constraints are met; By merging the parent and offspring generations, and performing non-dominant ranking and elite retention based on the principle of constraint and dominance, a new generation of population is formed. Repeat the above steps until the termination condition is met, and select a typical scheme from the final non-dominated solution set according to the preference weight or inflection point distance, and output the merged power outage plan list to complete the dynamic balance of the main grid and distribution network plans.

5. The intelligent collaborative optimization method according to claim 4, characterized in that, Objective functions for maximizing the goal of multiple uses with a single stop, minimizing the risk of frequent stops, and minimizing the cumulative risk include: The objective function for maximizing the goal of "one stop, multiple uses" can be obtained through formula (1): Official (1) in, This represents a set of maintenance plans for both the main grid and distribution network. The objective function that maximizes the goal of one stop and multiple uses. Indicates the number of distinct merge groups. Indicates the merge identifier; The preset frequency stop risk weight is obtained through formula (2): Official (2) in, Indicates device Preset frequency stop risk weight, Indicates the importance of the equipment. This indicates the expected number of power outages, among which This indicates the number of power outages in the past 12 months. Indicates the frequency stop threshold; The objective function for minimizing the risk of frequent shutdowns can be obtained through formula (3): Official (3) in, The objective function representing the goal of minimizing frequent shutdown risk is: Indicates the number of devices. Indicates device Should an additional power outage be added? The objective function for minimizing the risk superposition objective is obtained through formula (4): Official (4) in, The objective function represents the goal of minimizing the superposition of risks. Indicates the first Grid-like risks include one of the following: risk penalties for complete outage of dual power sources for important users, risk penalties for heavy overload of transfer lines, risk penalties for the entire station, a single main transformer, or a single busbar, and risk penalties for multiple planned conflicts of the same transfer lines.

6. The intelligent collaborative optimization method according to claim 1, characterized in that, The intelligent collaborative optimization method further includes: constructing corresponding intelligent agents based on the jurisdiction of the local dispatch, distribution dispatch, and county dispatch respectively; when the main grid power outage plan affects the distribution network, a collaborative request is initiated; the issue is resolved through negotiation based on the shared situation and dynamic balance results; after reaching a consensus, a multi-department review process is triggered, and a maintenance application form is generated.

7. The intelligent collaborative optimization method according to claim 1, characterized in that, The intelligent collaborative optimization method also includes: leaving traces at each operation node throughout the entire process to form a full-cycle traceability record from data entry, balancing, collaboration, review to push execution.

8. An intelligent collaborative optimization system for main network maintenance planning tasks, characterized in that, The intelligent collaborative optimization system includes: The data construction module is used to extract information on equipment hazards, defect records, frequent outage events, test information, and real-time power grid operation data to construct maintenance data. The execution module is used to execute, based on the constructed maintenance data, an intelligent collaborative optimization method for main network maintenance planning tasks as described in any one of claims 1-7.

9. A processor, characterized in that, Used to run a program, wherein the program is run to execute: an intelligent collaborative optimization method for main network maintenance planning tasks as described in any one of claims 1-7.

Images