Production and scheduling graph model data difference resolution and homologous mapping method based on power grid business rules

Abstract

The invention provides a power grid business rule-based production and scheduling graph model data difference resolution and homologous mapping method, which relates to the technical field of data processing, and comprises the following steps of: performing access and standardization processing on multi-source heterogeneous graph model data of a design state, a production side and a scheduling side to obtain standardized graph model data; converting the standardized graph model data into a unified data entity set, and constructing a reference data plane through the unified data entity set; defining a reference data range on the reference data plane, and performing hierarchical division on the reference data range to obtain a plurality of initial data partitions; performing boundary optimization processing on each initial data partition to obtain an accurate boundary range of each data partition; and mapping each data entity to a corresponding boundary data partition. According to the method, one-time modeling multi-party multiplexing of the graph model data is realized, and efficient coordination of one graph of a power grid is supported.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power system and automation, in particular to a production and dispatch graph model data difference elimination and homologous mapping method based on power grid business rules. BACKGROUND

[0002] In the production and dispatch management of the power system, as the core basis supporting the power grid planning, operation control and fault handling, the graph model data has long been facing a series of problems caused by the separation of production and dispatch business processes: the repeated modeling of the production and dispatch departments leads to manual multiple remodelling, the models are not interconnected, the time effectiveness conflict between the production side which updates the graph model at a monthly or yearly level according to the operation and maintenance rhythm and the dispatch side which updates the graph model at a second or minute level according to the real-time operation is prominent, the standards of the graph model are not unified between the SVG of the production side which needs to draw the physical path of the device to the terminal to the connection line and the OCS system of the dispatch side which needs to simplify the graph model into a logical topology containing virtual nodes, the network parameters are maintained in a scattered manner, and the data interaction between the systems relies on manual connection through multiple interfaces, which aggravates the data separation and seriously restricts the efficient cooperation of the power grid one map.

[0003] For example, a test station updates the state to under maintenance in the production system due to maintenance, but the dispatch system does not change synchronously and still shows running, so manual intervention is needed to check and correct before the business accuracy can be restored, which directly exposes the natural difference between the production and dispatch business rules that the current technology has not systematically eliminated, the lack of standardized difference identification, cooperation and automatic synchronization mechanism, and the high dependence of data consistency on manual checking, which leads to low efficiency and easily causes accuracy deviation of core business applications such as power transmission dynamic capacity increase and intelligent alarm of the substation. SUMMARY

[0004] The technical problem to be solved by the present application is to provide a production and dispatch graph model data difference elimination and homologous mapping method based on power grid business rules, which realizes the reuse of graph model data modeling by multiple parties and supports the efficient cooperation of the power grid one map.

[0005] To solve the above technical problems, the technical solution of the present application is as follows: In a first aspect, a production and dispatch graph model data difference elimination and homologous mapping method based on power grid business rules is provided, which comprises: Accessing and standardizing the multi-source heterogeneous graph model data of the design state, the production side and the dispatch side to obtain standardized graph model data; converting the standardized graph model data into a unified data entity set, and constructing a reference data plane through the unified data entity set; According to the reference data plane, defining a reference data range, hierarchically dividing the reference data range to obtain a plurality of initial data partitions, and through boundary optimization processing of each initial data partition, obtaining the accurate boundary range of each data partition; mapping each data entity to the corresponding boundary data partition; Based on the data characteristics of each data partition, adjustment factors are extracted; the map data is then collaboratively optimized using these adjustment factors to obtain the optimized map data. Based on the optimized graph data, identify the attribute differences, topology differences, and naming differences between the production side and the scheduling side; Based on the identified attribute differences, topology differences, and naming differences, difference resolution processing is performed according to power grid business rules to obtain unified graph model data after resolution. Based on the unified graph data after resolution, a common source mapping is performed to obtain unified graph data for all power grid services.

[0006] Furthermore, multi-source heterogeneous graphical model data from the design, production, and scheduling sides are accessed and standardized to obtain standardized graphical model data. This standardized graphical model data is then converted into a unified set of data entities, and a baseline data plane is constructed using this unified set of data entities, including: Access design-state data, production-side diagram data, and scheduling-side diagram data; The multi-source heterogeneous graph model data were standardized separately to obtain standardized multi-source heterogeneous graph model data. The standardized multi-source heterogeneous graph data is converted into data entities with a unified structure and identifier, resulting in a unified set of data entities. Based on the unified set of data entities, a baseline data plane is constructed at the data processing layer.

[0007] Furthermore, based on the defined baseline data range on the baseline data plane, the baseline data range is hierarchically divided to obtain multiple initial data partitions; by performing boundary optimization processing on each initial data partition, the precise boundary range of each data partition is obtained; each data entity is mapped to the corresponding boundary data partition, including: Based on the benchmark data plane, a benchmark data range covering the target power grid area is defined, and the benchmark data range is hierarchically divided to obtain multiple initial data partitions with preliminary boundaries. For each initial data partition, the electrical distance, connection density, and device type distribution characteristics of the internal data entities are comprehensively considered. The boundary optimization algorithm is used to eliminate the overlap and gap between partitions and obtain the boundary range of each data partition. Based on the boundary range of each data partition, each data entity in the unified data entity set is mapped to the corresponding boundary data partition according to its spatial location attributes, electrical connection relationship and equipment type.

[0008] Furthermore, based on the data characteristics of each data partition, adjustment factors are extracted; the map data is then collaboratively optimized using these adjustment factors to obtain optimized map data, including: Based on the boundary data partitions that have completed the mapping of data entities, analyze and extract the internal data features; Based on the extracted data features and combined with power grid business rules, the adjustment factor for data collaborative optimization is calculated; The adjusted factors obtained from the calculation are used to perform collaborative optimization processing on the map data mapped to each boundary data partition; After completing the collaborative optimization process, the optimized data within each boundary data partition is integrated to obtain globally consistent optimized map data.

[0009] Furthermore, based on the optimized graph data, the differences in attributes, topology, and naming between the production and scheduling sides are identified, including: Based on the optimized graph data, the attribute differences between corresponding devices on the production side and the scheduling side are identified. The attribute conflict is determined by comparing whether the numerical deviation of the core attributes exceeds the preset threshold, and the compatibility analysis of non-core attributes is performed to obtain the attribute difference identification results. Based on the attribute difference identification results, the structural differences between the physical topology of the production side and the logical topology of the scheduling side are identified, and the topology difference identification results are obtained. Based on the topology difference identification results, the naming differences between the production side and the scheduling side are finally identified. By using a pre-built keyword mapping table and ID association rules, naming format conflicts and encoding rule inconsistencies are identified, and the naming difference identification results are obtained. Based on the results of attribute difference identification, topological difference identification, and naming difference identification, the results are classified and integrated to obtain a structured difference report containing complete difference information.

[0010] Furthermore, based on the identified attribute differences, topology differences, and naming differences, difference resolution processing is performed according to power grid business rules to obtain unified graph model data after resolution, including: Based on the attribute differences identified in the structured difference report, and according to the authoritative data source priority rules defined in the power grid business rules, the core attribute conflicts between the production side and the dispatch side are arbitrated and unified, and the relationships between non-core attribute differences are established to obtain intermediate graph model data with unified attributes. Based on the intermediate graph model data with unified attributes, for the topology differences identified in the structured difference report, according to the topology conversion specifications defined in the power grid business rules, the deletion of virtual nodes and reconstruction of physical connections, the merging of buses and the segmentation of topology shrinkage, and the standardized splitting and combination of multi-port devices are performed to obtain topology-consistent intermediate graph model data. Based on topology-consistent intermediate graph data, for the naming differences identified in the structured difference report, according to the naming conversion rules and ID mapping specifications defined in the power grid business rules, the format unification and semantic conversion of equipment naming are performed, and a bidirectional association mapping between production asset ID and scheduling topology node ID is established, finally obtaining the unified graph data after resolution.

[0011] Furthermore, based on the resolved unified graph model data, a source mapping is performed to form unified graph model data for all power grid services, including: Based on the unified graph model data after decomposition, a device-level mapping relationship is constructed by using GisID as the unique primary index. A bidirectional association graph between production asset ID and scheduling topology node ID is established through the graph database to obtain the device-level homogeneous mapping relationship. Based on the device-level homogeneous mapping relationship, an attribute-level mapping system is constructed. By establishing a production and scheduling attribute mapping table and defining the field correspondence and data type conversion rules, the attribute-level homogeneous mapping relationship is obtained. Based on attribute-level homogeneous mapping relationships, a unified topology model is finally constructed. By integrating physical and logical topologies, an event-driven topology synchronization mechanism is established, and unified graph model data for all power grid services is obtained based on the unified topology model.

[0012] Secondly, a system for resolving differences and mapping common sources in production and scheduling diagram data based on power grid business rules includes: The acquisition module is used to access and standardize multi-source heterogeneous graph model data from the design, production, and scheduling sides to obtain standardized graph model data; it then converts the standardized graph model data into a unified set of data entities and constructs a baseline data plane using this unified set of data entities. The mapping module is used to hierarchically divide the baseline data range according to the baseline data range defined on the baseline data plane, and obtain multiple initial data partitions; by performing boundary optimization processing on each initial data partition, the precise boundary range of each data partition is obtained; each data entity is mapped to the corresponding boundary data partition; based on the data characteristics of each data partition, adjustment factors are extracted; and the map model data is collaboratively optimized using adjustment factors to obtain optimized map model data. The processing module is used to identify attribute differences, topology differences, and naming differences between the production side and the scheduling side based on the optimized graph model data; based on the identified attribute differences, topology differences, and naming differences, it performs difference resolution processing according to power grid business rules to obtain unified graph model data after resolution; based on the unified graph model data after resolution, it performs source mapping to obtain unified graph model data for all power grid businesses.

[0013] Thirdly, a computing device includes: One or more processors; A storage device for storing one or more programs that, when executed by one or more processors, cause the one or more processors to implement the method.

[0014] Fourthly, a computer-readable storage medium storing a program that, when executed by a processor, implements the method.

[0015] The above-described solution of the present invention has at least the following beneficial effects: Because it adopts standardized processing of multi-source data access based on CIM standard instantiation model, unified graphic format according to SVG specifications, core attribute deviation threshold judgment, equipment ID and terminal number topology matching algorithm, keyword mapping table multi-dimensional difference identification, authoritative data source arbitration with equipment factory documents as the highest priority, ReconnectTerminal algorithm to delete virtual nodes, business rule-driven difference resolution of regular expression naming conversion template, and a three-level homogeneous mapping mechanism of equipment, attributes and topology with GisID as the unique master index, it overcomes the data timeliness conflicts caused by repeated modeling between production and dispatch departments in the power system and inconsistent graphic model update cycles, the difficulties in cross-system adaptation caused by inconsistent graphic model standards and fragmented naming and ID encoding, and the inefficiency of decentralized maintenance of multi-source data and reliance on manual intervention. It achieves improved graphic format adaptation efficiency, significantly improved parameter uniformity, improved difference identification accuracy, improved naming conflict resolution efficiency, significantly reduced topology conversion manual cost, and shortened cross-system data association time. It realizes one-time modeling and multi-party reuse of graphic model data, supports efficient collaboration of a single power grid map, and improves the accuracy of dispatch instructions. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a method for resolving differences in production and scheduling model data and mapping to common sources based on power grid business rules, provided by an embodiment of the present invention.

[0017] Figure 2 This is a schematic diagram of a system for resolving differences in production and scheduling model data and mapping to common sources based on power grid business rules, provided by an embodiment of the present invention. Detailed Implementation

[0018] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0019] like Figure 1As shown, an embodiment of the present invention proposes a method for resolving differences and mapping common sources in production and scheduling model data based on power grid business rules. The method includes the following steps: Step 1: Access and standardize the multi-source heterogeneous graph model data from the design, production, and scheduling sides to obtain standardized graph model data; convert the standardized graph model data into a unified set of data entities, and construct a baseline data plane through the unified set of data entities; Step 2: Based on the baseline data range defined on the baseline data plane, the baseline data range is hierarchically divided to obtain multiple initial data partitions; by performing boundary optimization processing on each initial data partition, the precise boundary range of each data partition is obtained; and each data entity is mapped to the corresponding boundary data partition. Step 3: Extract adjustment factors based on the data characteristics of each data partition; perform collaborative optimization processing on the map data using the adjustment factors to obtain optimized map data; Step 4: Based on the optimized graph data, identify the attribute differences, topology differences, and naming differences between the production side and the scheduling side; Step 5: Based on the identified attribute differences, topology differences, and naming differences, perform difference resolution processing according to power grid business rules to obtain unified graph model data after resolution; Step 6: Based on the unified graph data after resolution, perform source mapping to obtain unified graph data for all power grid services.

[0020] In this embodiment of the invention, because it employs the technical means of accessing and standardizing multi-source heterogeneous graph model data from the design state, production side, and dispatch side, and converting it into a unified data entity set to construct a benchmark data plane, hierarchically dividing and optimizing the boundaries according to the benchmark data range, mapping the data entities to the corresponding partitions, extracting adjustment factors based on data partition features to collaboratively optimize the graph model data, identifying differences in production and dispatch side attributes, topology, and naming based on the optimized data, resolving these differences according to power grid business rules, and then performing homogeneous mapping based on the unified graph model data, it overcomes the problems of heterogeneous and fragmented multi-source graph model data in the power system without a unified processing benchmark, lack of ordered partitioning and precise boundaries in data management, insufficient graph model data quality, difficulty in locating differences between production and dispatch side data, data inconsistency due to lack of standardized resolution of differences, and data silos formed by cross-business data. This achieves standardized integration and unified benchmark construction of multi-source graph model data, ordered and precise mapping of data management, improved graph model data quality, accurate identification and standardized resolution of differences between production and dispatch side data, and ultimately obtains unified graph model data for all power grid businesses to support power grid single-map collaboration. In a preferred embodiment of the invention, step 1 above may include: Step 1.1 involves accessing design-state data, production-side schematic data, and dispatch-side schematic data. Specifically, this includes connecting to three core business systems within the power system: the design platform, the production management system, and the dispatch operation system. Design-state data, production-side schematic data, and dispatch-side schematic data are acquired respectively. When acquiring design-state data from the design platform, it is necessary to collect the unique geographic information identifier for the entire equipment lifecycle, a graphical file showing the equipment layout and connections drawn using scalable vector graphics specifications, and key parameters determined during the initial equipment design phase, including core planning and design information such as rated capacity, voltage level, and equipment model. When acquiring production-side schematic data from the production management system, it is necessary to extract the equipment asset ledger built based on the public information model. The ledger includes the equipment model. The system includes full lifecycle management information such as procurement time and installation location; it also extracts topology data reflecting the actual physical installation and hard connections of the equipment, which must record in detail the correspondence between equipment and terminals, cable types, and laying paths; in addition, it needs to collect complete records of all equipment maintenance; when obtaining dispatch side diagram data from the dispatch operation system, it needs to collect simplified logical topology data for power grid operation calculations, which includes virtual nodes and equivalent lines set to improve calculation efficiency; it also needs to collect real-time operating parameters of the equipment, covering dynamic data such as voltage, current, and power that reflect the current operating conditions of the equipment; and it also needs to obtain equipment naming codes that conform to dispatch management specifications, which should facilitate dispatch personnel to quickly identify the region and function type of the equipment.

[0021] Step 1.2 involves standardizing the multi-source heterogeneous graphic model data to obtain standardized multi-source heterogeneous graphic model data. Specifically, this includes standardizing the multi-source heterogeneous graphic model data obtained from the design platform, production management system, and scheduling operation system from two dimensions: graphic format and model structure. In terms of graphic format standardization, the format requirements of all graphic files are strictly unified according to the Scalable Vector Graphics Specification, the graphic file header structure is unified, and the Extensible Markup Language version, encoding format, and independent attributes are clearly defined. At the same time, the namespace, extended link namespace, power grid-specific namespace, and graphic outline position and size of the Scalable Vector Graphics are defined, and the export system identifier of the graphic data is marked. Secondly, the organization of graphic content should be standardized, dividing graphics into two parts: graphic definition and graphic rendering. The graphic definition part needs to predefine the font, color, and other style attributes and primitive symbols required for graphic display. The graphic rendering part needs to standardize the way primitives are referenced or graphics are drawn using basic elements. Graphic exchange rules should be unified, and domain data import should take precedence over graphic import. During the import process, the uniqueness of the association between graphic objects and domain data should be checked. After import, issues such as inconsistencies between graphics and domain data, such as missing domain data, should be reported in a clear manner. In terms of model structure standardization, various models should be instantiated according to the Common Information Model Standard. For the plant / station model, it is necessary to ensure that the plant / station class includes the plant / station name, a globally unique identifier, and a hierarchical relationship. The voltage level class is associated with the hierarchical relationship to clarify the inclusion relationship between the plant / station and the voltage level. For the equipment model, taking the transformer as an example, the transformer class needs to be split into winding classes, and the rated capacity, rated voltage, and other attributes of the windings need to be clarified. The windings need to be associated with topology nodes through the terminal class to ensure that the electrical connection relationship of the equipment is clear. For the topology container model, the voltage level and bay classes need to be instantiated according to the hierarchical structure of plant / station, voltage level, bay, and equipment to ensure that the topology container structure is consistent across different systems. At the same time, the parameter field names and data types of all models should be consistent. For example, the rated capacity of the equipment recorded on the production side should be consistent with the rated capacity attribute field name in the public information model on the dispatch side.

[0022] Step 1.3 involves converting the standardized multi-source heterogeneous graphical model data into data entities with a unified structure and identifier, resulting in a unified set of data entities. Specifically, this includes converting the design-state data, production-side graphical model data, and scheduling-side graphical model data, after standardization of graphical format and model structure, according to a pre-defined unified data structure to obtain data entities with a consistent structure and unique identifier. Each data entity must contain four basic elements: a globally unique identifier, a business name, core attributes, and relationships. The globally unique identifier uses a universally unique identification code format to ensure that equipment identifiers are not duplicated throughout the design, production, and scheduling processes. The business name must follow the rules of voltage level, equipment type, and serial number, such as 220 kV circuit breaker 1, ensuring logical consistency in the naming of equipment across different systems. The core attribute part needs to integrate design, production, and scheduling... The key parameters of each stage are converted. The entities corresponding to the design-side data need to include initial parameters such as rated capacity and voltage level determined in the design stage. The entities corresponding to the production-side data need to include equipment models, installation locations and the most recent maintenance time in the asset ledger and maintenance records. The entities corresponding to the scheduling-side data need to include the current voltage, current and node association information in the logical topology in the real-time operating parameters. The association part needs to establish the corresponding association between the data entities and the original data. The design-side data entities are associated with geographic information identifiers and scalable vector graphics files. The production-side data entities are associated with terminal connection information in the physical topology. The scheduling-side data entities are associated with virtual node information in the logical topology. Through the conversion process, the multi-source data that was originally scattered in different systems and had different formats are integrated into a unified set of data entities with a unified structure, standardized identification and complete information.

[0023] Step 1.4: Based on the unified data entity set, construct a baseline data plane at the data processing layer. Specifically, this includes: building a baseline data plane at the data processing layer, using the obtained unified data entity set as core data support for data storage deployment; storing the unified data entity set in a distributed database; employing a master-slave replication mechanism to ensure data storage reliability; partitioning data by device type or geographical region using a sharding strategy to improve data read / write efficiency; and constructing a multi-dimensional data indexing system, establishing indexes for key dimensions such as device type, geographical region, data source, and globally unique identifiers to support rapid querying and locating of data entities in business operations. For example, all transformer types can be quickly filtered out using the device type index. Data entities, located through geographic region indexing, encompass all device data within a specific area. Integrated data verification functionality performs dual verification on unified data entities stored in the baseline data plane. Uniqueness verification ensures no duplicate data entities by comparing globally unique identifiers. Integrity verification checks each data entity for missing critical information, such as equipment rated capacity and voltage level, using preset core attribute templates. If missing information is detected, an alarm is triggered and a request for supplementation is sent. A data update mechanism is established, clearly defining the update trigger conditions and process for data entities in the baseline data plane. When data changes during design, production, or scheduling, the corresponding unified data entities are updated synchronously. The data in the baseline data plane remains consistent with the business system, ultimately providing sufficient support.

[0024] In this embodiment of the invention, by using access design-state data, production-side diagram model data, and dispatch-side diagram model data, the multi-source heterogeneous diagram model data is standardized. The standardized data is then converted into data entities with a unified structure and identifier, resulting in a unified set of data entities. Finally, a benchmark data plane is constructed at the data processing layer based on the set. This overcomes the problem that the design-state, production-side, and dispatch-side diagram model data in the power system are scattered, have heterogeneous formats, and lack unified structures and identifiers, making it difficult to integrate and communicate the data and lacking a unified processing foundation. It achieves effective aggregation and standardized processing of multi-source heterogeneous diagram model data, providing a unified benchmark for diagram model data difference identification, difference resolution, and homogeneous mapping, breaking the fragmented state of multi-source data.

[0025] In a preferred embodiment of the present invention, step 2 above may include: Step 2.1: Based on the baseline data plane, define the baseline data range covering the target power grid area. Divide the baseline data range hierarchically to obtain multiple initial data partitions with preliminary boundaries. Specifically, based on the constructed baseline data plane, clarify the boundary of the target power grid area. The power grid area must completely cover the power grid assets to be managed, including all substations, transmission lines, distribution equipment, and associated design, production, and dispatching-side graphical models within the target area. For example, the graphical models of all equipment in a provincial power grid, including 500kV hub substations, 220kV regional substations, 110kV terminal substations, and the transmission lines and distribution switches connecting these substations. When defining the baseline data range, it is necessary to associate the geographical boundaries and administrative jurisdiction of the power grid. The range should not omit key power grid assets, nor include redundant data from irrelevant areas.

[0026] The defined baseline data range is hierarchically divided. The division logic must align with the actual management architecture and business process requirements of the power grid to avoid deviating from real-world application scenarios. The first layer is divided by geographical region, splitting the target power grid area into several geographical zones at the city, county, or power supply area levels. The provincial power grid is further divided into zones such as City A, City B, etc. Power grid assets within each geographical zone share similar operation and maintenance management affiliations. The second layer, based on geographical partitioning, is divided by voltage level, further subdividing the map data within each geographical zone into 500kV, 220kV, 110kV, and 10kV voltage level zones, etc., because different voltage levels of equipment have different business rules in production, operation, maintenance, and dispatch management. Significant differences exist, and layering allows for tailored processing. The third layer, based on voltage level partitioning, is divided according to equipment function type. For example, the 220kV voltage level partition is split into 220kV main transformer equipment partition, 220kV busbar equipment partition, and 220kV switchgear partition. Equipment diagrams and data of the same function type are centrally managed. Through this three-layer progressive hierarchical division, multiple initial data partitions with preliminary boundaries are obtained. Each initial data partition corresponds to a specific combination of geographical region, voltage level, and equipment function. For example, the initial partition for 220kV main transformer equipment in City A, the initial partition for 110kV busbar equipment in City B, etc. The boundaries of each initial partition are tentatively defined as the logical range under the corresponding level division, and the correlation between equipment has not yet been precisely calibrated.

[0027] Step 2.2: For each initial data partition, considering the electrical distance, connection density, and equipment type distribution characteristics of the internal data entities, a boundary optimization algorithm is used to eliminate overlaps and gaps between partitions, thus obtaining the boundary range of each data partition. Specifically, for each initial data partition, the key features of all data entities within it are collected and analyzed one by one, and the boundary optimization is based on clear criteria. Among them, the electrical distance feature requires the statistical analysis of the actual electrical connection distance between every two equipment data entities within the partition, such as the cable length between the main transformer and the bus, the transmission distance between the substations at both ends of the line, etc., and the electrical connection method between the equipment is also marked. Connection density characteristics require calculating the number of connections of equipment data entities per unit area or unit voltage level within a partition. For example, this involves counting the average number of switches, transformers, and other equipment associated with each main transformer within a 220kV main transformer equipment partition to determine the tightness of connections between devices. Equipment type distribution characteristics require analyzing the proportion and distribution location of different types of equipment data entities within a partition. For example, this involves counting the number of main transformers, switches, busbars, transformers, and other equipment within an initial partition, as well as the geographically concentrated areas of each type of equipment. Boundary optimization algorithms are then used to adjust the boundaries of each initial data partition. The algorithm execution process must focus on eliminating overlaps and gaps between partitions to ensure the safety of equipment. The issue revolves around two core objectives: integrity of relationships. Regarding overlapping partitions, if two adjacent initial partitions have overlapping equipment data entities, their affiliation must be determined based on the electrical connection priority between the devices. For example, if a busbar is directly hard-connected to the 220kV main transformer in City A, and the busbar's operating status depends on the main transformer parameters for scheduling calculations, then the busbar data entity should be assigned to the main transformer equipment partition. Simultaneously, the busbar should be removed from the initial boundary of the busbar equipment partition to avoid overlap. Regarding gaps between partitions, if two adjacent initial partitions contain equipment data entities that are not assigned to any partition, their affiliation must be supplemented based on their electrical connection relationships and functional attributes. For zone boundaries, if the cable directly connects the main transformer and the busbar, and the main transformer belongs to the main transformer equipment zone and the busbar belongs to the busbar equipment zone, then based on the physical laying path and electrical dependencies of the cable, the cable is assigned to the main transformer equipment zone, and the boundary of the main transformer equipment zone is extended to the location of the cable to eliminate gaps. During the optimization process, it is also necessary to repeatedly verify whether the adjusted boundary of each zone conforms to the power grid business rules. For example, ensure that the equipment data entities within the same electrical circuit belong to the same zone as much as possible, and avoid splitting the diagram data of a complete transmission line into multiple zones; ensure that the equipment entities corresponding to the dispatch side logical topology data and the production side physical topology data belong to the same zone.

[0028] Step 2.3, based on the boundary range of each data partition, maps each data entity in the unified data entity set to the corresponding boundary data partition according to its spatial location attributes, electrical connection relationships, and equipment type. Specifically, this includes: identifying the three core attributes of each data entity in the unified data entity set to provide a basis for the mapping operation. The first attribute is the spatial location attribute, which includes the geographical coordinates and installation address of the equipment corresponding to each data entity in the actual power grid. This information is extracted from the geographic information identifiers in the design data and the equipment asset ledger in the production data to ensure accurate location of the equipment's physical location. The second attribute is the electrical connection relationship... The first attribute is the electrical connection link between the equipment and other equipment corresponding to each data entity, such as the terminal connection relationship between the main transformer and the switch, and the bus connection relationship between the line and the substation. This information is integrated from the physical topology data on the production side and the logical topology data on the dispatch side to clarify the association logic between the equipment. The second attribute is the equipment type attribute, which is the specific type of equipment corresponding to each data entity, such as 500kV main transformer, 220kV switch, 110kV bus, 10kV distribution transformer, etc. This information is extracted from the equipment design parameters in the design state data and the equipment asset ledger in the production side data, and the equipment type classification is accurate.

[0029] Following the order of spatial location matching, electrical connection verification, and finally equipment type adaptation, each data entity is mapped to its corresponding boundary data partition. Based on the spatial location attribute of the data entity, if the geographical coordinates of the equipment corresponding to a data entity are within the boundary range of the 220kV main transformer equipment partition in City A, and the installation address is the main transformer bay within the 220kV substation in City A, then the data entity is initially assigned to that partition. The rationality of the assignment is verified through the electrical connection relationship attribute. If the data entity is a 220kV main transformer in City A, its electrical connection is verified. If the switches, transformers, and other equipment connected to the main transformer have all been assigned to the 220kV main transformer equipment zone in City A, then the assignment of the main transformer data entity is confirmed. If it is found that a switch connected to the main transformer is assigned to another zone, then the boundary optimization results of step 2.2 are reviewed to confirm whether there is a boundary division deviation. If the boundary division is correct, it means that the electrical connection relationship between the switch and the main transformer needs to be focused on in the difference handling. The main transformer is still assigned to the original zone, and finally adapted through the equipment type attribute. The equipment type of the data entity is 220kV main transformer, which is consistent with the 220kV main transformer in City A. If the voltage level and equipment function of the main transformer equipment zone are perfectly matched, the mapping is complete. If there is a mismatch between the equipment type and the zone location, the spatial location attribute is rechecked to confirm whether the matching deviation is caused by an error in the geographical coordinate input. After correction, it is mapped to the corresponding 110kV main transformer equipment zone. For different data entities corresponding to the same equipment in the design, production, and dispatching phases, it is necessary to ensure that the three are mapped to the same boundary data zone. For example, the design, production, and dispatching data entities of a 220kV main transformer must all be mapped to the geographical area-voltage level-equipment function zone where the main transformer is located. This avoids the multiple source data of the same equipment being scattered in different zones. During the mapping process, the mapping result of each data entity must be recorded in real time, including the name of the zone to which it belongs and the mapping basis, to obtain a mapping list for easy verification and adjustment. All data entities in the unified data entity set are accurately mapped to the corresponding boundary data zone. Each zone gathers full-dimensional graphic model data of design, production, and dispatch under a specific geographical area-voltage level-equipment function in the target power grid area.

[0030] In this embodiment of the invention, because a reference data range covering the target power grid area is defined based on a reference data plane and hierarchically divided into multiple initial data partitions with preliminary boundaries is obtained, the electrical distance, connection density, and equipment type distribution characteristics of data entities within each initial data partition are combined with boundary optimization algorithms to eliminate overlaps and gaps between partitions, thereby determining the precise boundary range of each data partition, and then mapping data entities to the corresponding boundary data partitions based on the boundary range of each data partition and combined with the spatial location attributes, electrical connection relationships, and equipment types of each data entity in a unified data entity set, this invention overcomes the technical means of the lack of clear and orderly management partitions in the reference data plane of the power system, the ambiguity of initial partition boundaries, and the easy overlap or gaps between partitions leading to chaotic data entity affiliation, making it difficult to carry out targeted data processing according to region or feature, affecting the efficiency of collaborative optimization and difference identification of map and model data. This invention achieves the goal of establishing an orderly and accurate partition management system for map and model data of the target power grid area, clarifying the affiliation partition of each data entity, avoiding data management chaos, providing a clear data partitioning basis for extracting data features based on partitions, carrying out collaborative optimization of map and model data, and multi-dimensional difference identification, and improving the overall targeting and efficiency of data processing.

[0031] In a preferred embodiment of the present invention, step 3 above may include: Step 3.1: Based on the completed data entity mapping of each boundary data partition, analyze and extract internal data features. Specifically, this includes: systematically analyzing and extracting internal data features for each boundary data partition that has completed data entity mapping, such as a 220kV main transformer equipment partition in a city-level area, a 110kV busbar equipment partition in a county-level area, and a 10kV distribution line partition in an urban area. The extraction process needs to closely focus on the core pain points of power production and dispatch model data, covering four key dimensions: data integrity, data consistency, data timeliness, and data association density. In terms of data integrity feature extraction, it traverses all data entities within the partition, including design-state equipment parameter entities and production-side assets. The system checks for missing attributes in various categories, including ledger entities, dispatch-side operating parameter entities, and topology entities. For core attributes—parameters directly affecting grid operation safety and dispatch calculations, such as transformer rated capacity, rated voltage, equipment model, line conductor model, rated current, bus reference voltage, and insulation class—the system statistically analyzes the percentage of data entities lacking this attribute within a given zone. For example, if three transformers in a 220kV transformer zone lack the rated capacity attribute, the percentage and corresponding equipment number must be specified. For non-core attributes—parameters serving the entire equipment lifecycle management, such as equipment installation location, procurement time, maintenance team, and dispatch monitoring team—the system records the specific types of missing attributes and the distribution of corresponding data entities. For example, in a certain area, none of the eight 110kV switches had maintenance team information marked. The focus was on verifying the integrity of the connection between design and production data. For instance, whether the number of equipment terminals and connection paths marked in the design-side SVG graphics were consistent with the terminal numbers and cable laying paths recorded in the production-side physical topology. This was to avoid deviations in subsequent topology optimization due to gaps in design and production data. In terms of data consistency feature extraction, the focus was on parameter matching and format consistency of the same equipment across multiple data sources. Taking a 220kV main transformer within a zone as an example, the comparison was made between the rated capacity recorded in the production-side asset ledger and the rated capacity value in the dispatch-side logical topology model to check for discrepancies. The equipment status marked in the production-side maintenance records, such as under maintenance, was also examined. 1. Whether the equipment status displayed in real-time operating parameters on the dispatch side is synchronized with the equipment connection relationship drawn in the design-state SVG graphic, such as whether the terminal connection between the main transformer and the switchgear matches the terminal association relationship recorded in the physical topology on the production side; 2. Analyze the consistency between the naming format and ID encoding, such as whether there is a conflict between the naming format used by the production side for the main transformer, the asset ID containing the TX-001 field, and the naming format of the No. 1 main transformer on the dispatch side, and the topology ID containing the Region-TX-001 field. 3. Analyze the number and degree of deviation of statistical parameter values, format differences of data entities. For example, if the rated capacity of a certain main transformer on the production side is 100MVA and on the dispatch side is 105MVA, the deviation and the equipment involved need to be recorded.

[0032] In terms of data timeliness feature extraction, the latest update time of each data entity within the partition is recorded, and the update cycle differences are statistically analyzed according to the data source. For production-side data entities, the last modification time of equipment asset ledger parameters and the submission time of maintenance records are extracted to clarify their monthly or grade-level update patterns. For example, the last modification of a main transformer asset ledger was 2 months ago, and the latest maintenance record was 1 month ago. For dispatch-side data entities, the acquisition time of real-time operating parameters and the adjustment time of logical topology are extracted to clarify their second-level or minute-level update patterns. For example, the real-time current parameters of a line were acquired 8 seconds ago, and the logical topology was last adjusted 3 minutes ago. For design-state data entities, the finalization time of design parameters and the generation time of SVG graphics are extracted to clarify their project cycle-level update patterns. By comparing the update times of multiple source data entities of the same equipment, the time difference between production-side and dispatch-side data is calculated. For example, if the production side updated 3 hours ago, and the dispatch-side data is still stuck at the 6-hour operation time during maintenance of a main transformer, the number of such timeliness conflicts and the range of time differences need to be recorded.

[0033] In terms of data association density feature extraction, the analysis examines the tightness of electrical connections between equipment data entities within a partition and the completeness of cross-business links. For electrical connection association, the analysis checks whether there are any breaks in the complete links of equipment, terminals, and connecting lines in the physical topology of the production side. For example, whether the cable connection between a switch and the busbar is clearly associated with the terminal number in the topology. At the same time, the analysis checks whether the correspondence between virtual nodes in the logical topology of the dispatching side and physical equipment on the production side is clear. For example, whether the virtual nodes of a ring network are clearly associated with the physical equipment combination of 2 main transformers and 4 lines in the ring network. For business association analysis, the analysis examines the completeness of the transmission of design state data to production side data, such as whether the equipment parameters marked on the design drawings are completely recorded. The system tracks the synchronization effectiveness of production-side data to scheduling-side data in the production asset ledger, such as whether the production-side equipment maintenance plan is synchronized to the scheduling side for adjusting operating strategies. It also counts the number of data entity pairs with broken connection links or missing business associations. For example, if the design parameters of a certain line are not transmitted to the production ledger, both production and scheduling parameters will be missing the conductor type of that line. After completing the feature extraction of a single partition, a structured partition data feature report is generated, which clarifies the specific performance of the partition in four dimensions, such as a core attribute missing rate of 7%, an entity proportion with parameter deviations exceeding the allowable range of 11%, a time difference of up to 4 hours for entities with timeliness conflicts, and 12 pairs of entities with broken connections. The system also marks a list of data entities that need to be prioritized for optimization.

[0034] Step 3.2, based on the extracted data features and combined with power grid business rules, calculate the adjustment factor for data collaborative optimization. Specifically, this includes: reviewing the power grid business rules supporting data collaborative optimization to ensure that the adjustment factor calculation meets the actual needs of power production and dispatch. These rules include the following four categories: First, parameter allowable error rules: based on power equipment industry standards and dispatch system measurement accuracy requirements, clarify the allowable deviation range of core parameters. For example, based on the standard transformer rated capacity, the allowable deviation is ±5%; based on SCADA system technical specifications, the real-time current measurement error is less than or equal to 2%; and the bus reference voltage allowable deviation is ±2%. Second, data source priority rules: clarify the authority level of different data sources, with priority from high to low, based on the original parameters in the equipment factory documents. The following are the criteria for data synchronization delay thresholds: 1) Verification of parameters from third-party testing reports before equipment commissioning; 2) Dynamic correction of real-time operation and maintenance data on the production side after commissioning; 3) Feedback on the operational status of real-time measurement data on the scheduling side; 4) Topology association specifications require that virtual nodes in the scheduling side logical topology must be clearly associated with combinations of physical devices on the production side, prohibiting independent virtual nodes without corresponding physical devices. Additionally, the relationships between devices, terminals, and connecting lines in the production side physical topology must be consistent with the connection logic of the design-state SVG graphics; 5) Data synchronization delay thresholds stipulate that after changes to the production-side model data, it must be synchronized to the scheduling side system within 10 minutes. The collection interval for real-time operating parameters on the scheduling side must not exceed 5 minutes. After the design-state model data is finalized, it must be synchronized to the production and scheduling systems within 24 hours.

[0035] Based on the obtained partitioned data feature report and the aforementioned power grid business rules, adjustment factors for data collaboration optimization are calculated according to four dimensions: data integrity, data consistency, data timeliness, and data association density. The adjustment factor values ​​range from 0 to 1; higher values ​​indicate more severe data problems in that dimension and require priority optimization. For data integrity adjustment factor calculation, the core attribute missing rate is the primary criterion. If the core attribute missing rate within a partition exceeds the 5% warning threshold, the adjustment factor is doubled based on the missing rate; for example, when the missing rate is 10%, the adjustment factor is 0.2. If the core attribute missing rate is below 5%, the impact of non-core attribute missing rates on business collaboration is considered; for example, if non-core attribute missing rates hinder cross-departmental collaboration, the adjustment factor is increased by 0.1 to 0.3. For data consistency adjustment factor calculation, the percentage of data entities with parameter deviations exceeding the allowable range is statistically analyzed. If the percentage exceeds 8%, the adjustment factor is 0.3 to 0.5. Simultaneously, the percentage of data entities with naming format and ID encoding conflicts is statistically analyzed; if the percentage exceeds 10%, the adjustment factor is additionally increased by 0.1 to 0.2. For example, if 20% of the main transformers in a partition have naming format conflicts, the consistency adjustment factor is increased by [amount missing] on top of the base value. 0.2. Regarding the calculation of the data timeliness adjustment factor, the maximum time difference between multiple data sources from the same device is used as the basis. If the time difference exceeds 10 minutes, the synchronization delay threshold adjustment factor is 0.4 to 0.6; if the time difference is within 5 to 10 minutes, the adjustment factor is 0.2 to 0.4; if the time difference is less than 5 minutes, the adjustment factor is 0.1 to 0.2. For example, if the production side update status of a main transformer is "under maintenance" and the dispatch side has not synchronized for 30 minutes, then the timeliness adjustment factor is 0.6. Regarding the calculation of the data association density adjustment factor, the proportion of data entity pairs with broken electrical connection links is statistically analyzed. If the proportion... For data entities with missing data in the statistical business context exceeding 12%, the adjustment factor is 0.3 to 0.5. If the proportion exceeds 15%, the adjustment factor is increased by an additional 0.1 to 0.2. For example, if 18% of the lines in a certain partition have missing data transmission from design to production, the correlation density adjustment factor is increased by 0.2 on the base value. The weighted sum of the adjustment factors of the four dimensions is used to obtain the comprehensive adjustment factor for each boundary data partition. The weights are set according to the priority of power grid business: data consistency and timeliness have the highest weights, each accounting for 30%; data integrity has a weight of 25%; and data correlation density has a weight of 15%.

[0036] Step 3.3 involves using the calculated adjustment factors to perform collaborative optimization on the graphical data mapped to each boundary data partition. Specifically, this includes: optimizing the graphical data mapped to each boundary data partition based on the calculated comprehensive adjustment factor and the adjustment factors of each dimension, prioritizing data with higher adjustment factors. This optimization process closely addresses the pain points of graphical data mentioned in the background technology, prioritizing the completion of missing core attributes with higher adjustment factors in terms of data integrity optimization. For data entities lacking core attributes such as rated capacity and rated voltage, corresponding parameters are extracted from the equipment's factory documentation; if the factory documentation is missing, parameters from a third-party testing report before equipment commissioning are used. For data entities lacking core attributes such as equipment model and conductor model, corresponding information is extracted from the production-side asset ledger and synchronously updated to the scheduling-side logical topology model. For non-core attribute missing issues, information such as installation location and maintenance team is supplemented from production-side maintenance records and equipment installation files, and a relationship is established between non-core and core attributes. For example, binding maintenance teams to equipment numbers facilitates cross-departmental collaborative queries and verification of design and production-side data. If the terminal numbers in the design-state SVG graphic are inconsistent with the terminal numbers in the production-side physical topology, the terminal labels in the design-state SVG graphic will be corrected according to the actual installation records on the production side to ensure complete connection between the two. Regarding data consistency optimization, multi-source data parameters will be unified according to data source priority rules. If there are conflicts between core parameters on the production side and the dispatch side (e.g., the rated capacity of the main transformer is recorded as 100MVA on the production side and 105MVA on the dispatch side), the parameters in the equipment's factory documentation will be used to synchronously update the parameters in the production-side asset ledger and the dispatch-side logical topology model. If the factory documentation is missing, the parameters in the third-party testing report will be used. For inconsistent equipment status (e.g., the production side records it as under maintenance while the dispatch side records it as running), the latest maintenance record on the production side will be used to synchronously update the equipment status in the real-time operating parameters on the dispatch side. For conflicts in naming formats and ID encoding, the naming conventions in the power grid business rules will be unified, for example, converting the naming of the main transformer on the production side to the format of the No. 1 main transformer on the dispatch side, while establishing the association between the production asset ID and the dispatch topology node ID to ensure that the identifiers of the same equipment are interoperable across multiple systems.

[0037] To optimize data timeliness, an automatic data change synchronization mechanism is established. When changes occur in the production-side model data, such as updating equipment status to "under maintenance" or modifying asset ledger parameters, the system automatically receives a change notification and pushes it to the scheduling-side data platform via a message queue. After receiving the notification, the scheduling-side data platform verifies the data integrity. If the data format and range meet the requirements, it immediately updates the scheduling-side logical topology and real-time operating parameters. If the update fails, a rollback mechanism is triggered to restore the state before the update and a retry strategy is initiated, retrying once every 1 minute, up to a maximum of 3 times. At the same time, an error log is recorded for maintenance personnel to troubleshoot. The scheduling-side real-time operating parameters are automatically updated and synchronized to the production-side system at a collection interval of 5 minutes to ensure that the production side can obtain the equipment operating status in real time. In terms of optimizing data association density, electrical connection links and business association links are supplemented. For broken links in the physical topology of the production side, such as those between equipment, terminals, and connecting lines, information such as cable type, laying path, and terminal number should be supplemented based on the construction records at the production site to ensure the integrity of the link. For virtual nodes in the logical topology of the dispatching side, the associated combination of physical equipment on the production side should be clearly defined. For example, a virtual node in a ring network is associated with 2 main transformers and 4 lines, and the association relationship should be marked in the diagram data. For business association breakpoints from the design state to the production side and from the production side to the dispatching side, a data transmission verification mechanism should be established. After the design state data is finalized, the system should automatically verify whether the production side asset ledger is completely entered and the design parameters are entered. After the production side data is changed, the system should automatically verify whether the dispatching side is updated synchronously. If there is a breakpoint, an alarm should be triggered to prompt the maintenance personnel to complete the transmission link.

[0038] Step 3.4: After completing the collaborative optimization process, integrate the optimized data within each boundary data partition to obtain globally consistent optimized graph model data. Specifically, this includes: after completing the collaborative optimization process for the graph model data within each boundary data partition, initiating the global data integration process to ensure consistency of the integrated graph model data across the entire power grid; supporting efficient collaboration of a single power grid map; establishing partitioned data integration standards; and clarifying that the data format, parameter units, naming conventions, and topology association rules during the integration process must be consistent with the power grid public information model specifications and SVG graphics exchange specifications. For example, the rated capacity of equipment should uniformly use MVA as the unit; equipment naming should uniformly use voltage level and equipment type and serial number formats; and topology should uniformly use equipment... The combination of physical mapping logic for terminals and connecting lines with simplified logic containing virtual nodes ensures that the optimized data of each partition can be integrated according to a unified standard. Secondly, cross-partition data correlation verification is carried out. For map data with correlation between different boundary data partitions, such as the connection of a 220kV line in partition A and a 220kV bus in partition B, it is necessary to verify whether the connection terminal number of the line in partition A is consistent with the connection terminal number of the bus in partition B, and whether the rated parameters of the line match the reference voltage of the bus. If there is a mismatch, the process is traced back to the collaborative optimization stage of the corresponding partition to check whether there are deviations in the parameter settings and topology correlation during the optimization process. After correction, the correlation verification is carried out again until the cross-partition data correlation is consistent.

[0039] A unified global graph model data system is formed by integrating multi-source data dimensions. This system links and integrates SVG graphics and design parameters in the design state, asset ledgers and physical topology on the production side, and logical topology and real-time operating parameters on the dispatch side, using unique device identifiers such as GisIDs. This ensures centralized querying and management of multi-source data for the same device. For example, the integrated data for a 220kV main transformer needs to include the SVG graphics and rated parameters in the design state, installation location and maintenance records on the production side, and logical topology nodes and real-time operating current on the dispatch side. Data from each dimension is bound by GisIDs to avoid data fragmentation. A global data consistency check is initiated, using a graph theory-based loop detection algorithm to traverse the global topology graph to check for isolated nodes and loop conflicts, ensuring a reasonable topology structure. A parameter integrity check algorithm checks for missing core parameters in the global graph model data according to a preset core attribute template. If missing parameters are found, an alarm is triggered to prompt for completion. Simultaneously, a subset of device data is manually verified to ensure the consistency and accuracy of the multi-source data. For example, 10 main transformers are randomly selected to verify whether their rated capacity and equipment status on the production and dispatch sides are consistent, and whether the topology connections on the design and production sides match. After verification, the globally consistent optimized map and model data is stored in the unified data base of the power grid. This data can be used by business applications such as power grid planning, operation control, and fault handling to completely solve the problem of fragmented production and scheduling map and model data, and support efficient collaboration of the power grid with a single map.

[0040] In this embodiment of the invention, by employing a technical approach that involves analyzing and extracting the internal data features of each boundary data partition based on the completed data entity mapping, calculating adjustment factors for data collaborative optimization in conjunction with power grid business rules, and using these adjustment factors to collaboratively optimize the map model data mapped to each boundary data partition, and finally integrating the optimized data within each boundary data partition, this invention overcomes the technical problems of multi-source map model data in power systems being difficult to optimize in a targeted manner due to unclear data features within partitions, lack of quantitative adjustment basis supported by power grid business rules leading to blind optimization processes, independent optimization of data in each partition easily resulting in global data inconsistencies, and insufficient map model data quality affecting the accuracy of difference identification and resolution. This achieves targeted optimization of map model data in each boundary data partition, ensures that the optimization process aligns with the actual needs of power grid production and scheduling, achieves consistency of map model data globally, and improves the quality of map model data, laying a reliable data foundation for the accurate identification and standardized resolution of subsequent attribute, topology, and naming differences.

[0041] In a preferred embodiment of the present invention, step 4 above may include: Step 4.1: Based on the optimized graph data, identify the attribute differences between corresponding equipment on the production side and the dispatch side. Determine attribute conflicts by comparing whether the numerical deviations of core attributes exceed preset thresholds, and perform compatibility analysis on non-core attributes to obtain attribute difference identification results. Specifically, this includes: clarifying the attribute classification of corresponding equipment on the production side and the dispatch side; core attributes are parameters that directly affect power grid operation safety and dispatch calculations, including equipment rated capacity, rated voltage, equipment model, conductor model, rated current, bus reference voltage, etc. In the background technology, numerical deviations in these attributes can easily lead to dispatch decision-making errors; non-core attributes serve the entire lifecycle management of equipment but do not directly participate in dispatch calculations. The parameters include equipment installation location, procurement time, maintenance team, dispatch monitoring team, and equipment commissioning time. While these attributes do not affect dispatch calculations, they require cross-departmental collaboration. The process involves extracting attribute information of corresponding equipment on the production and dispatch sides from the optimized graphical model data. For example, for a 220kV main transformer, the process involves extracting the rated capacity equipment status from the production-side asset ledger and the rated capacity equipment status from the dispatch-side logical topology model; extracting the maintenance team from the production-side maintenance records and the monitoring team from the dispatch-side system; and identifying core attribute differences. For each core attribute, the numerical deviation between the production-side and dispatch-side parameters is calculated, and the deviation is compared with a preset threshold. The preset threshold is then combined with power equipment... The measurement standards are determined based on industry standards and dispatching systems. For example, the rated capacity threshold is set at 5% according to the allowable deviation of transformer capacity in GB / T1094.1, the real-time current threshold is set at 2% according to the measurement error requirements of the SCADA system, and the bus reference voltage threshold is set at 2% according to the power grid topology connection rules. If the numerical deviation exceeds the preset threshold, it is judged as an attribute conflict. For example, if the rated capacity of the main transformer is recorded as 100MVA on the production side and 106MVA on the dispatching side, the deviation of 6% exceeds the 5% threshold and is judged as a rated capacity attribute conflict. If the numerical deviation is within the threshold range, it is marked as an acceptable difference, and a non-core attribute compatibility analysis is carried out: verifying whether the differences in the description of non-core attributes between the production side and the dispatching side are acceptable. Impacts on cross-departmental information sharing: For example, although the production-side maintenance team A and the dispatch-side monitoring team B have different names, they both correspond to the same equipment maintenance responsibility unit. By establishing a relationship between the maintenance team and the monitoring team, information sharing can be achieved, which is determined to be a compatibility difference. If the installation location on the production side is marked as bay 1 in the east area of ​​the substation, but the installation location on the dispatch side is not marked, it will be impossible to quickly locate the physical location when a fault occurs, which is determined to be an incompatible difference. It is necessary to clearly mark this in the difference identification results, and organize all core attribute conflicts, acceptable differences, and non-core attribute compatible / incompatible differences to obtain the attribute difference identification results, including equipment number, attribute name, production-side attribute value, dispatch-side attribute value, and difference type.

[0042] Step 4.2: Based on the attribute difference identification results, identify the structural differences between the physical topology of the production side and the logical topology of the scheduling side to obtain the topology difference identification results. Specifically, this includes: based on the obtained attribute difference identification results, selecting equipment with no conflicting core attributes or clearly defined difference types to ensure that the topology difference identification is based on equipment with relatively consistent attribute information. If a device has a conflicting rated voltage attribute, it needs to be temporarily stored until the attribute difference is resolved before conducting topology analysis to avoid misjudgment of topology differences due to attribute errors. Extract the topology structure information of the corresponding equipment on the production side and the scheduling side respectively: the production side topology is the physical topology, and the complete physical connection relationship of equipment-terminal-connection line needs to be extracted, including the association between equipment number and terminal number, cable type and laying path, and the correspondence between hard connections between terminals. This information is obtained from the optimized production side physical topology data; the scheduling side topology is the logical topology, and the simplified connection relationship including virtual nodes needs to be extracted, including virtual node number, physical equipment combination associated with virtual node, equivalent line parameters, merged bus number and connected equipment. This information is obtained from the optimized scheduling side logical topology data. Node matching, link comparison, and structural analysis are used. The process identifies topology differences by using equipment number and terminal number as keywords to match physical topology nodes on the production side with logical topology nodes on the dispatching side. For example, a matching relationship is established between terminal 1 of main transformer T1 on the production side and the corresponding logical node of main transformer T1 on the dispatching side. The connection links between the matched nodes are compared. If the production side has a complete physical link between main transformer T1 terminal 1, switchgear terminal 3, bus M1 terminal 2, and the dispatching side has a simplified link between main transformer T1 logical node - virtual node V1 bus M4 logical node, this is marked as a difference between complete physical link and simplified logical link. The rationality of special structures in the dispatching side logical topology is analyzed. If the dispatching side... Virtual nodes without corresponding physical devices on the degree side are marked as redundant virtual node differences. If multiple bus voltage levels are consistent on the production side, but not merged into equivalent buses on the dispatch side, they are marked as bus unmerged differences. If the main transformer on the production side is modeled as a whole, and the main transformer on the dispatch side is split into high, medium, and low voltage windings and modeled separately, they are marked as equipment splitting differences. The identified topology differences are classified as virtual node differences, bus merging or splitting differences, link simplification differences, and equipment modeling differences. The equipment number, topology node number, production side topology description, and dispatch side topology description involved in the differences are recorded to obtain the topology difference identification results.

[0043] Step 4.3: Based on the topology difference identification results, the naming differences between the production side and the scheduling side are finally identified. Through a pre-built keyword mapping table and ID association rules, naming format conflicts and encoding rule inconsistencies are identified to obtain the naming difference identification results. Specifically, this includes: using the obtained topology difference identification results as a basis, determining the scope of devices requiring naming difference identification; only performing naming analysis on devices with clearly defined topologies, such as lines where the correspondence between physical and logical links has been confirmed, further analyzing their naming and ID differences; if a device has redundant virtual node topology differences, the physical devices associated with the virtual nodes must first be identified before naming is performed on these physical devices. Analysis was conducted to construct a keyword mapping table and ID association rules. The keyword mapping table was developed based on the differences in naming rules between the production and dispatch sides, covering common differences in the expression of equipment serial number, site name, and equipment type, such as AND sign, substation and station, kV and kilovolt, switch and circuit breaker, while also pre-setting special character conversion rules. The ID association rules were developed based on the encoding logic of the two types of IDs. Production asset IDs focus on the full life cycle management of equipment, and the encoding includes the year of purchase, plant area code, and equipment type code. Dispatch topology node IDs focus on the power grid operation logic, and the encoding includes the power grid zone, voltage level, and serial number. The rules clearly define the correspondence between equipment type code, voltage level, plant area, or zone in the two types of IDs.

[0044] Naming format conflict identification: Extract the names of corresponding equipment on the production side and the dispatch side, and compare the differences in expression through a keyword mapping table. For example, the descriptions of the No. 2 main transformer on the production side and the No. 2 main transformer on the dispatch side are consistent after the mapping table conversion, and it is determined that there is no format conflict. The No. 3 switch of the 10kV Chengdong substation on the production side and the No. 3 circuit breaker of the 10kV Chengdong station on the dispatch side have differences in the transformer or station number switch or circuit breaker before the mapping table conversion, and it is determined to be a naming format conflict. Record the conflict type and corresponding keywords. ID coding rule inconsistency identification: Compare the coding structure of the production asset ID and the dispatch topology node ID. If the equipment type code and voltage level of the two types of IDs do not match, it is determined to be an ID coding logic conflict. If the coding structures of the two types of IDs match but the field descriptions are different, a corresponding relationship can be established through ID association rules, and it is determined to be an ID coding format difference. The naming format conflict, ID coding logic conflict, and ID coding format difference are sorted out, including equipment number, production side naming, dispatch side naming, production asset ID, dispatch topology node ID, and difference type, to form the naming difference identification result.

[0045] Step 4.4: Based on the attribute difference identification results, topology difference identification results, and naming difference identification results, classify and integrate them to obtain a structured difference report containing complete difference information. Specifically, this includes: determining the core framework of the structured difference report, which must cover five modules: basic equipment information, attribute difference details, topology difference details, naming difference details, and difference impact assessment. This ensures the report not only fully presents the difference information but also provides a clear basis for subsequent difference resolution. Simultaneously, it addresses the lack of standardization in difference identification and the low efficiency of manual verification in the background technology. Through structured design, it reduces the workload of manually sorting out differences and classifies the difference information: attribute difference identification results, topology difference identification results, and naming difference identification results are linked and integrated according to equipment number, ensuring that the three types of differences for the same equipment are presented together. For example, for main transformer T1, the report first fills in the basic equipment information, and then... Fill in attribute differences, topology differences, and naming differences, and supplement the difference impact assessment: Combine the core business applications mentioned in the background technology to analyze the degree of impact of each type of difference on the business. For example, conflicts in the rated capacity attribute of the main transformer will cause deviations in the power flow calculation on the dispatch side, affecting the setting of dynamic capacity expansion limits for transmission, and are marked as high impact; missing installation locations of non-core attributes will prolong fault location time and affect fault handling efficiency, and are marked as medium impact; naming format conflicts only affect the consistency of information expression and do not affect the issuance of dispatch instructions, and are marked as low impact. Optimize report readability and traceability: Use a unified identifier for difference types to facilitate maintenance personnel to quickly focus on key differences; add difference source traceability notes at the end of the report to indicate the identification basis for each type of difference; and attach an index table of equipment numbers and corresponding difference types to support quick querying of all equipment involved in a certain type of difference, ultimately obtaining a structured difference report containing complete difference information.

[0046] In this embodiment of the invention, because it uses optimized graph model data, it first determines attribute conflicts by comparing whether the deviation of core attribute values ​​exceeds a preset threshold and performs compatibility analysis on non-core attributes to identify attribute differences. Then, it combines the attribute difference identification results to identify structural differences between the physical topology of the production side and the logical topology of the dispatching side. Next, it identifies naming format conflicts and coding rule inconsistencies between the production side and the dispatching side through a pre-built keyword mapping table and ID association rules. Finally, it integrates the three types of difference identification results—attribute, topology, and naming—to obtain a structured difference report containing complete difference information. Therefore, it overcomes the technical problems of lack of clear judgment basis for attribute conflicts in graph model data of the production and dispatching sides in the power system, difficulty in accurately locating topological differences, difficulty in difference identification due to fragmented naming and coding rules, and the scattered and disordered nature of various difference information, which makes it difficult to manage uniformly, resulting in missed or misjudged differences, low efficiency of cross-system data adaptation, and high cost of manual verification. It achieves accurate identification of the three core differences—attribute, topology, and naming—of graph model data of the production and dispatching sides, avoids difference omissions and misjudgments, and makes difference information structured and clear. This provides an accurate basis for subsequent difference resolution according to power grid business rules and improves the efficiency of cross-departmental graph model data collaborative processing.

[0047] In a preferred embodiment of the present invention, step 5 above may include: Step 5.1: Based on the attribute differences identified in the structured difference report, and according to the authoritative data source priority rules defined in the power grid business rules, arbitrate and unify the core attribute conflicts between the production side and the dispatch side, and establish relationships for non-core attribute differences to obtain intermediate graph model data with unified attributes. Specifically, this includes: extracting all attribute difference information from the structured difference report and classifying it into two categories: core attribute conflicts and non-core attribute differences. Core attribute conflicts include situations where the numerical deviations between the production side and the dispatch side in parameters such as rated capacity, rated voltage, equipment model, conductor model, and rated current exceed preset thresholds. For example, the rated capacity of a 220kV main transformer is recorded as 100MVA on the production side and 106MVA on the dispatch side; the conductor model of a 110kV line is recorded as LGJ-240 on the production side and LGJ-300 on the dispatch side. Non-core attribute differences include installation location, maintenance team, etc. In cases where parameters such as group, dispatch and monitoring team, and procurement time are described differently or missing, for example, a switch on the production side is labeled as maintenance team 1, while the dispatch side labels the monitoring team as control team 1; a main transformer's installation location is recorded as bay 2 in the west substation area on the production side, but not on the dispatch side. Based on the authoritative data source priority rules defined in the power grid business rules, the arbitration basis for various attributes is clarified. The priority rules are formulated in combination with the pain point of data consistency highly relying on manual verification in the background technology. The priority from high to low is: equipment factory documents, third-party inspection reports before equipment commissioning, real-time operation and maintenance data on the production side, and measurement data on the dispatch side. Arbitration and unification are carried out for core attribute conflicts: for each core attribute conflict, the corresponding parameters in the equipment factory documents are retrieved first. If the factory documents are complete, the parameters therein are used as the benchmark to synchronously update the parameters in the production side asset ledger and the dispatch side logical topology model. For example, if the factory documentation for a main transformer indicates a rated capacity of 100MVA, then the record of 106MVA on the dispatching side will be corrected to 100MVA, and the production side ledger will be checked to ensure consistency with the factory documentation. If the factory documentation is missing, then the parameters in the third-party test report will be retrieved as a benchmark, and the data on both sides will be updated synchronously. If the test report is also missing, then the latest operation and maintenance record on the production side will be used as the standard to correct the data on the dispatching side.For conflicts in core attributes related to equipment status, the status recorded in the latest maintenance work order on the production side shall prevail, and the equipment status in the real-time operating parameters on the dispatch side shall be updated synchronously. For differences in non-core attributes, a correlation relationship shall be established: for non-core attributes with different descriptions, a corresponding relationship shall be established through the non-core attribute correlation table defined in the power grid business rules. For example, the production side maintenance shift 1 and the dispatch side control group 1 shall be bound together, and both shall be marked as the maintenance and monitoring responsibility units for the same equipment. For missing non-core attributes, they shall be supplemented from high-priority data sources. For example, if the installation location of a certain main transformer is missing on the dispatch side, the No. 2 bay of the substation's west plant area shall be extracted from the production side asset ledger and added to the dispatch side. A correlation mapping between the installation location on the production side and the location label on the dispatch side shall be established. After all core attribute arbitration and non-core attribute correlation are completed, the data shall be integrated to form an intermediate model data with unified attributes. In this data, the core attributes of the corresponding equipment on the production side and the dispatch side shall be completely consistent, and the non-core attributes shall be mutually queried through the correlation relationship.

[0048] Step 5.2: Based on the attribute-unified intermediate graph model data, for the topology differences identified in the structured difference report, according to the topology transformation specifications defined in the power grid business rules, perform virtual node deletion and physical connection reconstruction, bus merging and segmentation topology contraction processing, and standardized splitting and combining of multi-port devices to obtain topology-consistent intermediate graph model data. Specifically, this includes: based on the obtained attribute-unified intermediate graph model data, extracting all topology difference information from the structured difference report, simplifying differences in virtual node differences, bus merging or splitting differences in links, and modeling differences in multi-port devices. The four categories include, for example, redundant virtual nodes V3 on the dispatch side without corresponding physical equipment; three 10kV low-voltage busbars with the same voltage level in a substation on the production side but not merged on the dispatch side; a line on the production side is a complete physical link of switch A-terminal 1-cable-terminal 2-busbar B, which is simplified on the dispatch side to switch A logical node-virtual node V5-busbar B logical node; a main transformer on the production side is modeled as a whole, while on the dispatch side it is split into high-voltage winding, medium-voltage winding, and low-voltage winding modeled separately. Based on the topology conversion specifications defined in the power grid business rules, the handling methods for various topology differences are clarified. This specification addresses the pain point of inconsistent standards between the physical topology on the production side and the logical topology on the dispatch side in the background technology. The core principle is to adapt the production side physical topology to the operational calculation needs of the dispatch side.

[0049] Processing is performed one by one according to the type of topology difference: Deletion of virtual nodes and reconstruction of physical connections: For redundant virtual nodes on the scheduling side, the deletion operation is performed directly; for virtual nodes on the scheduling side used to simplify calculations and associated with physical equipment on the production side, the terminal information of its upstream and downstream equipment is traversed through the ReconnectTerminal algorithm. For example, if virtual node V5 is associated with switch A terminal 1 and bus B terminal 2, then after deleting V5, the physical connection relationship between switch A terminal 1 and bus B terminal 2 is directly established. The physical information such as cable type and laying path is clearly marked in the diagram data to ensure that the reconstructed link is consistent with the actual installation on the production side. Topology shrinkage processing of bus merging and segmentation: For multiple bus segments on the production side with the same voltage level and similar electrical distance, merging processing is performed according to the specifications. For example, if there are three 10kV busbars M1, M2, and M3 in a substation, all with a voltage level of 10.5kV and a spacing of 30 meters between sections, they can be merged into one equivalent busbar M4. The base voltage and equivalent impedance of the merged busbar are marked in the diagram data, and the original connection records between each busbar section and the equipment are retained for easy traceability. In the case of a single busbar on the production side and an incorrect split into multiple sections on the dispatching side, the merging is performed according to the physical topology on the production side, and the logical topology on the dispatching side is corrected.

[0050] Standardized decomposition and combination of multi-port devices: For devices that are modeled as a whole on the production side and decomposed into multiple components on the dispatch side, standardized decomposition is performed according to the CIM standard. For example, if the production side of a main transformer records equipment information as a whole, and the dispatch side decomposes it into high-voltage winding, medium-voltage winding, and low-voltage winding, then based on the dispatch side decomposition logic, detailed winding information is added to the production side model data to ensure that the model after decomposition on the production side is consistent with the logical model structure on the dispatch side. For devices that are decomposed on the production side and modeled as a whole on the dispatch side, the dispatch side model is modified according to the production side decomposition structure. After completing all topology difference processing, it is integrated to form intermediate model data with consistent topology. In this data, the connection relationship between the physical topology of the production side and the logical topology of the dispatch side, the equipment modeling method are completely coordinated, and there are no redundant virtual nodes, chaotic bus merging or decomposition, or inconsistent equipment modeling.

[0051] Step 5.3: Based on the topology-consistent intermediate graph model data, for the naming differences identified in the structured difference report, according to the naming conversion rules and ID mapping specifications defined in the power grid business rules, perform format unification and semantic conversion of equipment naming, and establish a bidirectional association mapping between production asset IDs and scheduling topology node IDs, finally obtaining the unified graph model data after resolution. Specifically, this includes: extracting all naming difference information from the structured difference report and classifying and organizing it into two categories: naming format conflict and inconsistent ID encoding rules. Naming format conflicts include differences in equipment naming between the production side and the dispatching side. For example, the production side uses the format "1 main transformer 2 main transformers" while the dispatching side uses "No. 1 main transformer 2 main transformers". Similarly, the production side uses the format "East City Substation, West City Substation" while the dispatching side uses "East City Station, West City Station" for substations. The production side uses the format "110kV East City Line" while the dispatching side uses "110kV East City Line". Inconsistent ID encoding rules include differences in the encoding structure and field meanings of production asset IDs and dispatching topology node IDs. Based on the naming conversion rules and ID mapping specifications defined in the power grid business rules, specific methods for unifying naming and associating IDs are clarified. The naming conversion rules address the pain point of low efficiency in manual verification due to chaotic naming in the background technology. They include three types of rules: keyword mapping, format unification, and semantic conversion. Keyword mapping rules clarify the correspondence between common expressions such as # and number, substation and station, kV and kilovolt, switch and circuit breaker. Format unification rules clarify that equipment naming must follow the structure of voltage level + equipment type + serial number + affiliated station.

[0052] Unifying the format and semantic conversion of device naming: For naming format conflicts, based on the keyword mapping and format unification requirements in the naming conversion rules, the names on the production side or the scheduling side are converted into a unified format. Referring to the naming format of the corresponding equipment on the production side, the names are completed by combining voltage level, equipment type, and the site to which they belong. A bidirectional association mapping between production asset ID and scheduling topology node ID is established: According to the ID mapping specification, for each device, the corresponding fields in the production asset ID and scheduling topology node ID are extracted, and a key-value pair relationship is established. After completing the naming unification and ID association, the unified graph model data after resolution is obtained. In this data, the attributes, topology, and naming of the corresponding devices on the production side and the scheduling side are completely consistent, which can directly support cross-departmental business collaboration without the need for manual verification and correction.

[0053] In this embodiment of the invention, by employing a structured difference report, arbitrating and unifying core attribute conflicts between the production and dispatch sides based on the priority rules of authoritative data sources in power grid business rules, and establishing non-core attribute associations, the invention overcomes the technical challenges of lacking standard arbitration basis for core attribute conflicts in production and dispatch side diagram data in power systems, difficulty in coordinating differences due to inconsistent physical and logical standards in topology, fragmented naming and ID encoding without interoperability mechanisms, and high dependence on manual verification for data consistency, achieving complete unification of core attributes in production and dispatch side diagram data, consistent topology structure, and bidirectional association querying of equipment names and IDs. This results in unified diagram data that resolves all differences, significantly reducing manual intervention costs, improving cross-departmental data collaboration efficiency, providing consistent data support for power grid planning, operation control, fault handling, and other businesses, and facilitating efficient collaboration with a single diagram in the power grid.

[0054] In a preferred embodiment of the present invention, step 6 above may include: Step 6.1: Based on the unified graph model data after deconstruction, a device-level mapping relationship is constructed by using GisID as the unique primary index. A bidirectional association graph between production asset IDs and scheduling topology node IDs is established through the graph database to obtain device-level homogeneous mapping relationships. Specifically, this includes: determining the rules for obtaining GisIDs to ensure their uniqueness and relevance; GisIDs need to integrate the device's geospatial information, device type characteristics, and full lifecycle identifiers. Specifically, this is determined by the latitude and longitude coordinates of the actual installation location of the device processed by a hash algorithm, the device type code, and the device's geospatial information, device type characteristics, and full lifecycle identifiers. The GisID consists of three parts: the commissioning time stamp, the installation coordinates, and the network configuration. For example, if a 220kV main transformer is installed at 118.5°E and 31.2°N, its hash value is H123, the equipment type code is TX, and the commissioning time stamp is 20231001143000, then its GisID is H123-TX-20231001143000. After obtaining the GisID, a corresponding GisID is assigned to each device in the unified schema data after decomposition, ensuring that each device corresponds to only one GisID, and that the GisID is consistent between the production side and the dispatch side. To maintain complete consistency within the system, a unique master index for device-level mapping is constructed. A graph database supporting complex relationship storage and querying is selected. Three types of core nodes are created in the graph database: GisID nodes, production asset ID nodes, and scheduling topology node ID nodes. Then, relationships between nodes are established: with the GisID node as the core, relationships are established between the GisID node and the production asset ID node (related to production ID), and between the GisID node and the scheduling topology node ID node (related to scheduling ID). A bidirectional query relationship is also established between the production asset ID node and the scheduling topology node ID node, ensuring that any node can be traced back to the other two nodes. This completes the initialization and verification of the relationship graph: the GisID, production asset ID, and scheduling topology node ID of all devices and their corresponding relationships are imported into the graph database to form a complete bidirectional relationship graph. The validity of the relationships is verified through test cases. For example, inputting the production asset ID of a device allows for quick querying of the corresponding GisID and scheduling topology node ID, and inputting the scheduling topology node ID allows for reverse querying of the production asset ID and GisID.

[0055] Step 6.2: Based on the equipment-level homogeneous mapping relationship, construct an attribute-level mapping system. This is achieved by establishing a production and scheduling attribute mapping table and defining field correspondences and data type conversion rules. Specifically, this includes: based on the established equipment-level homogeneous mapping relationship, sorting out all attribute fields in the production-side and scheduling-side graph model data, and classifying them into three categories: core business attributes, operation and maintenance management attributes, and operation and monitoring attributes. Core business attributes include attributes that directly affect power grid operation calculations, such as rated capacity, rated voltage, equipment model, and conductor model. Operation and maintenance management attributes include attributes that support production and operation and maintenance, such as maintenance status, maintenance time, installation location, and maintenance team. Operation and monitoring attributes include attributes that are monitored in real time by the scheduling side, such as real-time current, real-time voltage, and power factor. Establish a production and scheduling attribute mapping table, with the vertical axis representing production-side attribute fields and the horizontal axis representing scheduling-side attribute fields, clearly defining the scheduling-side attribute fields corresponding to each production-side attribute field. For example, the maintenance status field on the production side corresponds to the equipment operation status field on the dispatching side; the rated capacity field on the production side corresponds to the rated capacity field on the dispatching side; and the installation location field on the production side corresponds to the physical location information field on the dispatching side. For attribute fields with different data types, explicit data type conversion rules are defined: if the rated capacity field on the production side is a string type and the field on the dispatching side is a numeric type, the conversion rule is to extract the numeric part from the string and remove the unit identifier; if the maintenance status field on the production side is a text description and the field on the dispatching side is an enumeration type, the conversion rule is to establish a correspondence between the text description and the enumeration value, and then apply this correspondence to the attribute. The mapping table is validated and optimized: Select multiple typical devices, extract their production-side and scheduling-side attribute data, and perform attribute matching and transformation through the attribute mapping table and transformation rules. Check whether the transformed data meets the business calculation requirements of the scheduling side and the operation and maintenance management requirements of the production side. If a production-side attribute is found to have no corresponding scheduling-side field, mark it as a production-unique attribute in the attribute mapping table and record its purpose. If a scheduling-side attribute is found to have no corresponding production-side field, mark it as a scheduling-unique attribute and supplement the association description. After the validation is completed, the final attribute-level homogeneous mapping relationship is formed to ensure that the production and scheduling-side attribute information can be interacted without obstacles.

[0056] Step 6.3: Based on attribute-level homogeneous mapping relationships, a unified topology model is finally constructed. By integrating physical and logical topologies, an event-driven topology synchronization mechanism is established, and unified graph model data for all power grid services is obtained based on the unified topology model. Specifically, this includes: merging physical and logical topologies: using the production-side physical topology while incorporating the simplified requirements of the dispatch-side logical topology. For example, in a substation, the connection between multiple 10kV distribution transformers and the busbar is described. The production-side physical topology records the independent cable connection path between each transformer and the busbar. During fusion, the terminal connection relationship of each transformer is retained, and the busbar segments corresponding to these transformers are merged into an equivalent value. The busbars, in a unified topology model, are marked with the physical connection relationships between equivalent buses and each distribution transformer. To accommodate the simplified calculation needs of the dispatching side, an event-driven topology synchronization mechanism is established: Topology change event types are defined, including events triggered by the production side and events triggered by the dispatching side. Event triggering conditions and processing flows are set: When a topology change occurs on the production side, the system automatically generates a topology change event. The event includes the GisID of the changed device, the change type, and the topology structure information before and after the change. This event is pushed to the dispatching side's topology management module via a message queue. After receiving the event, the dispatching side module verifies the completeness of the change information based on attribute-level homogeneous mapping relationships. If the information is complete, it proceeds according to the topology... The system updates the logical topology on the scheduling side using the fusion rules, and simultaneously generates a synchronization confirmation event to be fed back to the production side. If the production side does not receive confirmation feedback, the system automatically initiates a retry mechanism, retrying once every 2 minutes, with a maximum of 3 retries. If a retry fails, an exception log is recorded and an operation and maintenance alarm is triggered. When a topology change occurs on the scheduling side, it is also synchronized to the production side according to the same process, ensuring that topology changes on both sides are consistent in real time. This solves the problems of topology update timeliness conflicts and reliance on manual connection in the background technology. Based on a unified topology model, a unified graph model data for all power grid services is obtained: the fused topology structure, attribute-level homogeneous mapping relationships, and device-level homogeneous mapping relationships are integrated to form a coverage design state. The system integrates full-dimensional graph and model data from the production and scheduling sides. It performs global consistency checks on the integrated data, examining for anomalies in device ID associations, attribute conversions, and topology connections. For example, it checks whether a device's production asset ID and scheduling topology node ID are correctly associated via GisID, whether the conversion result of a certain attribute conforms to business rules, and whether a certain topology connection simultaneously meets the requirements of production physical paths and scheduling logic calculations. After successful verification, the unified graph and model data for all business operations is stored on the unified power grid data platform, enabling access for all business processes, including power grid planning, operation control, and fault handling. This completely breaks down data barriers between the production and scheduling sides, supporting efficient collaboration across the power grid using a single graph.

[0057] In this embodiment of the invention, because it adopts unified graph model data after decomposition, it constructs device-level mapping relationships with GisID as the unique primary index and establishes a bidirectional association graph between production asset IDs and scheduling topology node IDs through a graph database. Based on the device-level mapping, it constructs an attribute-level mapping system, establishes production and scheduling attribute mapping tables, defines field correspondences and data type conversion rules, and finally constructs a unified topology model based on attribute-level mapping, integrates physical and logical topologies, and establishes an event-driven topology synchronization mechanism. Therefore, it overcomes the difficulties in cross-system querying caused by the lack of a unique association index for device IDs on the production and scheduling sides in the power system, the information interaction barriers caused by inconsistent attribute fields and data types, and the difficulty in maintaining topology consistency due to the fragmentation of physical and logical topology standards and the lack of an automatic synchronization mechanism. This results in data silos restricting the collaboration of all power grid services. It achieves bidirectional interoperability querying of device IDs on the production and scheduling sides, barrier-free interaction of attribute information, and dynamic synchronization and consistency of topology structure. Ultimately, it obtains unified graph model data for all power grid services, completely breaks down the data barriers between production and scheduling, supports efficient collaboration of all business links such as power grid planning, operation control, and fault handling with a single graph, and reduces the cost of cross-system data adaptation.

[0058] like Figure 2 As shown, embodiments of the present invention also provide a system for resolving differences and mapping common sources in production and scheduling model data based on power grid business rules, comprising: The acquisition module is used to access and standardize multi-source heterogeneous graph model data from the design, production, and scheduling sides to obtain standardized graph model data; it then converts the standardized graph model data into a unified set of data entities and constructs a baseline data plane using this unified set of data entities. The mapping module is used to hierarchically divide the baseline data range according to the baseline data range defined on the baseline data plane, and obtain multiple initial data partitions; by performing boundary optimization processing on each initial data partition, the precise boundary range of each data partition is obtained; each data entity is mapped to the corresponding boundary data partition; based on the data characteristics of each data partition, adjustment factors are extracted; and the map model data is collaboratively optimized using adjustment factors to obtain optimized map model data. The processing module is used to identify attribute differences, topology differences, and naming differences between the production side and the scheduling side based on the optimized graph model data; based on the identified attribute differences, topology differences, and naming differences, it performs difference resolution processing according to power grid business rules to obtain unified graph model data after resolution; based on the unified graph model data after resolution, it performs source mapping to obtain unified graph model data for all power grid businesses.

[0059] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for resolving differences and mapping common sources in production and scheduling diagram data based on power grid business rules, characterized in that, The method includes: Multi-source heterogeneous graphical model data from the design, production, and scheduling sides are accessed and standardized to obtain standardized graphical model data; the standardized graphical model data is then converted into a unified set of data entities, and a baseline data plane is constructed using this unified set of data entities. Based on the baseline data range defined on the baseline data plane, the baseline data range is hierarchically divided to obtain multiple initial data partitions; by performing boundary optimization processing on each initial data partition, the precise boundary range of each data partition is obtained; and each data entity is mapped to the corresponding boundary data partition. Based on the data characteristics of each data partition, adjustment factors are extracted; the map data is then collaboratively optimized using these adjustment factors to obtain the optimized map data. Based on the optimized graph data, identify the attribute differences, topology differences, and naming differences between the production side and the scheduling side; Based on the identified attribute differences, topology differences, and naming differences, difference resolution processing is performed according to power grid business rules to obtain unified graph model data after resolution. Based on the unified graph data after resolution, a common source mapping is performed to obtain unified graph data for all power grid services.

2. The method for resolving differences and mapping similar data in production and scheduling diagrams based on power grid business rules according to claim 1, characterized in that, Multi-source heterogeneous graphical model data from the design, production, and scheduling sides are accessed and standardized to obtain standardized graphical model data; Standardized map data is transformed into a unified set of data entities, and a baseline data plane is constructed using this unified set of data entities, including: Access design-state data, production-side diagram data, and scheduling-side diagram data; The multi-source heterogeneous graph model data were standardized separately to obtain standardized multi-source heterogeneous graph model data; The standardized multi-source heterogeneous graph data is converted into data entities with a unified structure and identifier, resulting in a unified set of data entities. Based on the unified set of data entities, a baseline data plane is constructed at the data processing layer.

3. The method for resolving differences and mapping similar data in production and scheduling diagrams based on power grid business rules according to claim 2, characterized in that, Based on the definition of the reference data range on the reference data plane, the reference data range is hierarchically divided to obtain multiple initial data partitions; By performing boundary optimization on each initial data partition, the precise boundary range of each data partition is obtained; Map each data entity to its corresponding boundary data partition, including: Based on the benchmark data plane, a benchmark data range covering the target power grid area is defined, and the benchmark data range is hierarchically divided to obtain multiple initial data partitions with preliminary boundaries. For each initial data partition, the electrical distance, connection density, and device type distribution characteristics of the internal data entities are comprehensively considered. The boundary optimization algorithm is used to eliminate the overlap and gap between partitions and obtain the boundary range of each data partition. Based on the boundary range of each data partition, each data entity in the unified data entity set is mapped to the corresponding boundary data partition according to its spatial location attributes, electrical connection relationship and equipment type.

4. The method for resolving differences and mapping similar data in production and scheduling diagrams based on power grid business rules according to claim 3, characterized in that, Based on the data characteristics of each data partition, adjustment factors are extracted; the map model data is then collaboratively optimized using these adjustment factors to obtain optimized map model data, including: Based on the boundary data partitions that have completed the mapping of data entities, analyze and extract the internal data features; Based on the extracted data features and combined with power grid business rules, the adjustment factor for data collaborative optimization is calculated; The adjusted factors obtained from the calculation are used to perform collaborative optimization processing on the map data mapped to each boundary data partition; After completing the collaborative optimization process, the optimized data within each boundary data partition is integrated to obtain globally consistent optimized map data.

5. The method for resolving differences and mapping similar data in production and scheduling diagrams based on power grid business rules according to claim 4, characterized in that, Based on the optimized graph data, attribute differences, topology differences, and naming differences between the production side and the scheduling side are identified, including: Based on the optimized graph data, the attribute differences between corresponding devices on the production side and the scheduling side are identified. The attribute conflict is determined by comparing whether the numerical deviation of the core attributes exceeds the preset threshold, and the compatibility analysis of non-core attributes is performed to obtain the attribute difference identification results. Based on the attribute difference identification results, the structural differences between the physical topology of the production side and the logical topology of the scheduling side are identified, and the topology difference identification results are obtained. Based on the topology difference identification results, the naming differences between the production side and the scheduling side are finally identified. By using a pre-built keyword mapping table and ID association rules, naming format conflicts and encoding rule inconsistencies are identified, and the naming difference identification results are obtained. Based on the results of attribute difference identification, topological difference identification, and naming difference identification, the results are classified and integrated to obtain a structured difference report containing complete difference information.

6. The method for resolving differences and mapping similar data in production and scheduling diagrams based on power grid business rules according to claim 5, characterized in that, Based on the identified attribute differences, topology differences, and naming differences, difference resolution processing is performed according to power grid business rules to obtain unified graph model data after resolution, including: Based on the attribute differences identified in the structured difference report, and according to the authoritative data source priority rules defined in the power grid business rules, the core attribute conflicts between the production side and the dispatch side are arbitrated and unified, and the relationships between non-core attribute differences are established to obtain intermediate graph model data with unified attributes. Based on the intermediate graph model data with unified attributes, for the topology differences identified in the structured difference report, according to the topology conversion specifications defined in the power grid business rules, the deletion of virtual nodes and reconstruction of physical connections, the merging of buses and the segmentation of topology shrinkage, and the standardized splitting and combination of multi-port devices are performed to obtain topology-consistent intermediate graph model data. Based on topology-consistent intermediate graph data, for the naming differences identified in the structured difference report, according to the naming conversion rules and ID mapping specifications defined in the power grid business rules, the format unification and semantic conversion of equipment naming are performed, and a bidirectional association mapping between production asset ID and scheduling topology node ID is established, finally obtaining the unified graph data after resolution.

7. The method for resolving differences and mapping similar data in production and scheduling diagrams based on power grid business rules according to claim 6, characterized in that, Based on the unified graph model data after deconstruction, a common-source mapping is performed to form unified graph model data for all power grid services, including: Based on the unified graph model data after decomposition, a device-level mapping relationship is constructed by using GisID as the unique primary index. A bidirectional association graph between production asset ID and scheduling topology node ID is established through the graph database to obtain the device-level homogeneous mapping relationship. Based on the device-level homogeneous mapping relationship, an attribute-level mapping system is constructed. By establishing a production and scheduling attribute mapping table and defining the field correspondence and data type conversion rules, the attribute-level homogeneous mapping relationship is obtained. Based on attribute-level homogeneous mapping relationships, a unified topology model is finally constructed. By integrating physical and logical topologies, an event-driven topology synchronization mechanism is established, and unified graph model data for all power grid services is obtained based on the unified topology model.

8. A system for resolving differences and mapping common sources in production and scheduling diagram data based on power grid business rules, wherein the system implements the method as described in any one of claims 1 to 7, characterized in that, include: The acquisition module is used to access and standardize multi-source heterogeneous graph model data from the design state, production side, and scheduling side to obtain standardized graph model data. Standardized map data is converted into a unified set of data entities, and a baseline data plane is constructed using this unified set of data entities. The mapping module is used to hierarchically divide the reference data range according to the reference data range defined on the reference data plane, and obtain multiple initial data partitions. By performing boundary optimization on each initial data partition, the precise boundary range of each data partition is obtained; each data entity is mapped to the corresponding boundary data partition; adjustment factors are extracted based on the data characteristics of each data partition; and the graph model data is collaboratively optimized using the adjustment factors to obtain the optimized graph model data. The processing module is used to identify attribute differences, topology differences, and naming differences between the production side and the scheduling side based on the optimized graph model data; based on the identified attribute differences, topology differences, and naming differences, it performs difference resolution processing according to power grid business rules to obtain unified graph model data after resolution; based on the unified graph model data after resolution, it performs source mapping to obtain unified graph model data for all power grid businesses.

9. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images