CN-121984016-A - Scheduling decision determining method and device for power grid, storage medium and electronic equipment

CN121984016ACN 121984016 ACN121984016 ACN 121984016ACN-121984016-A

Abstract

The application discloses a scheduling decision determining method and device of a power grid, a storage medium and electronic equipment. The method comprises the steps of obtaining current power data of a target power grid, determining a current power knowledge graph of the target power grid based on the current power data, conducting semantic alignment on a current entity, a target entity, a current relation and a target relation to determine an update entity and an update relation, conducting data fusion on a current entity feature vector, a target entity feature vector, a current relation feature vector and a target relation feature vector to obtain an update entity feature vector and an update relation feature vector of the update entity, obtaining an update power knowledge graph based on the update entity, the update relation, the update entity feature vector and the update relation feature vector, determining a current scheduling decision of the target power grid based on the update power knowledge graph, and scheduling the target power grid. The application solves the technical problem of inaccurate determination result of the scheduling decision of the power grid in the related technology.

Inventors

XI SHAOQING
WANG HAIYUN
TIAN YUCHEN
WU YAN
YANG LIPING
LI YINGHAO
YAO YIDI
LU ZHONGYAN
WANG CHUNLING
Su Guojie
CHEN QIAN
ZHAO XINCHEN

Assignees

国网北京市电力公司

Dates

Publication Date: 20260505
Application Date: 20260130

Claims (10)

1. A scheduling decision determination method for an electrical network, comprising: acquiring current power data of a target power grid from a target database; Determining a current power knowledge graph of the target power grid based on the current power data, wherein the current power knowledge graph comprises a current entity, a current relation, a current entity characteristic vector and a current relation characteristic vector, wherein the current entity refers to a component part of the target power grid included in the current power data, and the current relation is used for indicating an association relation between the current entities; Performing semantic alignment on the current entity and a target entity in a target power knowledge graph of the target power grid, determining an updated entity, performing semantic alignment on the current relationship and a target relationship in the target power knowledge graph, and determining an updated relationship; Performing data fusion on the current entity feature vector and the target entity feature vector of the target entity in the target power knowledge graph to obtain an updated entity feature vector of the updated entity, and performing data fusion on the current relation feature vector and the target relation feature vector of the target relation in the target power knowledge graph to obtain an updated relation feature vector of the updated relation; updating the target power knowledge graph based on the updating entity, the updating relation, the updating entity feature vector and the updating relation feature vector to obtain an updated power knowledge graph of the target power grid; And determining a current scheduling decision of the target power grid based on the updated power knowledge graph, and scheduling the target power grid based on the current scheduling decision.
2. The method of claim 1, wherein prior to the obtaining current power data of the target grid from the target database, the method further comprises: receiving the current power data; judging whether to trigger a data updating condition based on the current power data; and storing the current power data to the target database under the condition that the data updating condition is triggered.
3. The method of claim 2, wherein the determining whether to trigger a data update condition based on the current power data comprises: determining a device parameter change rate based on current device parameter data in the current power data; determining an updated weather hazard level based on current weather data in the current power data; determining the change degree of the weather disaster based on the updated weather disaster grade; And determining to trigger the data updating condition under the condition that the equipment parameter change rate is greater than or equal to a preset change rate threshold and the weather disaster change degree is greater than or equal to a preset change degree threshold.
4. The method of claim 2, wherein storing the current power data to the target database if the data update condition is triggered comprises: Determining the data scale, service priority and stability requirement of the current power data; determining a database resource storing the current power data based on the data size, the business priority, and the stability requirement; determining a data quality of the current power data; And under the condition that the data quality meets the preset quality requirement, storing the current power data into a target database based on the database resource, wherein the target database is determined based on the database resource.
5. The method of claim 1, wherein the determining a current power knowledge-graph of the target grid based on the current power data comprises: performing entity extraction on the current power data to obtain an initial entity, and performing relation extraction on the current power data to obtain an initial relation; Determining the entity accuracy of the initial entity, and determining the initial entity as the current entity under the condition that the entity accuracy is greater than or equal to a preset accuracy threshold; determining a relationship accuracy rate and a recall rate of the initial relationship, wherein the recall rate is used for quantifying the comprehensiveness of the initial relationship determination result; determining a comprehensive evaluation value of the initial relationship based on the relationship accuracy and the recall; Determining an inter-class relationship coverage rate based on the initial relationship when the comprehensive evaluation value is greater than or equal to a preset evaluation value threshold, wherein the inter-class relationship is a relationship between two entities with different types, and the inter-class relationship coverage rate is used for quantifying the comprehensiveness of the inter-class relationship in the initial relationship determination result; determining the initial relationship as the current relationship under the condition that the coverage rate of the relationships among the classes is larger than or equal to a preset coverage rate threshold value; extracting features of the current power data, and determining the current entity feature vector and the current relation feature vector; and determining the current power knowledge graph based on the current entity, the current relation, the current entity feature vector and the current relation feature vector.
6. The method of claim 1, wherein semantically aligning the current entity with a target entity in a target power knowledge graph of the target grid, determining an updated entity, and semantically aligning the current relationship with a target relationship in the target power knowledge graph, determining an updated relationship comprises: determining an entity Euclidean distance based on a current entity semantic feature sub-vector included in the current entity feature vector and a target entity semantic feature sub-vector included in the target entity feature vector; based on the Euclidean distance and a preset distance threshold, carrying out semantic alignment on the current entity and the target entity to obtain the updated entity; Determining a relationship similarity based on a current relationship semantic feature sub-vector included in the current relationship feature vector and a target relationship semantic feature sub-vector included in the target relationship feature vector; and carrying out semantic alignment on the current relationship and the target relationship based on the relationship similarity and a preset similarity threshold value to obtain the updated relationship.
7. The method according to any one of claims 1 to 6, wherein the determining a current scheduling decision of the target grid based on the updated power knowledge-graph comprises: determining a risk entity in the updated power knowledge graph based on the updated entity feature vector; determining a risk path of a target power grid in the updated power knowledge graph based on the risk entity; Determining an alternative path to the risk path based on the updated power knowledge graph; Based on the alternative path, the current scheduling decision is determined.
8. A scheduling decision determining apparatus for an electrical network, comprising: the data acquisition module is used for acquiring current power data of the target power grid from the target database; The first determining module is configured to determine a current power knowledge graph of the target power grid based on the current power data, where the current power knowledge graph includes a current entity, a current relationship, a current entity feature vector and a current relationship feature vector, the current entity refers to a component part of the target power grid included in the current power data, and the current relationship is used to indicate an association relationship between the current entities; The semantic alignment module is used for carrying out semantic alignment on the current entity and a target entity in a target power knowledge graph of the target power grid, determining an update entity, carrying out semantic alignment on the current relationship and a target relationship in the target power knowledge graph, and determining an update relationship; The data fusion module is used for carrying out data fusion on the current entity feature vector and the target entity feature vector of the target entity in the target power knowledge graph to obtain an updated entity feature vector of the updated entity, and carrying out data fusion on the current relation feature vector and the target relation feature vector of the target relation in the target power knowledge graph to obtain an updated relation feature vector of the updated relation; The updating module is used for updating the target power knowledge graph based on the updating entity, the updating relation, the updating entity feature vector and the updating relation feature vector to obtain an updated power knowledge graph of the target power grid; and the second determining module is used for determining the current scheduling decision of the target power grid based on the updated power knowledge graph and scheduling the target power grid based on the current scheduling decision.
9. A non-volatile storage medium, characterized in that it stores a plurality of instructions adapted to be loaded by a processor and to perform the scheduling decision determining method of the electrical network according to any one of claims 1 to 7.
10. An electronic device comprising one or more processors and a memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the scheduling decision determination method of the power grid of any of claims 1-7.

Description

Scheduling decision determining method and device for power grid, storage medium and electronic equipment Technical Field The application relates to the field of power systems, in particular to a scheduling decision determining method and device of a power grid, a storage medium and electronic equipment. Background In the context of smart grids, power systems are increasingly complex, and unprecedented requirements are put forward on accuracy, instantaneity and reliability of scheduling decisions of a target grid. Therefore, real-time and accurate target power grid scheduling decision determination is realized, and the method has important significance for guaranteeing safe and stable operation of the power system. In the related art, a static model decision method is adopted to determine the scheduling decision of a target power grid, the static model decision method depends on a static power grid model and experience rules, and the models and rules may not provide accurate scheduling decision basis when the power grid state changes rapidly. Therefore, the problem of inaccurate determination results of the scheduling decisions of the power grid exists in the related technology. In view of the above problems, no effective solution has been proposed at present. Disclosure of Invention The embodiment of the application provides a scheduling decision determining method, a scheduling decision determining device, a storage medium and electronic equipment for a power grid, which at least solve the technical problem that the scheduling decision determining result of the power grid is inaccurate in the related technology. According to one aspect of the embodiment of the application, a scheduling decision determining method of a power grid is provided, which comprises the steps of obtaining current power data of the target power grid from a target database, determining a current power knowledge graph of the target power grid based on the current power data, wherein the current power knowledge graph comprises a current entity, a current relation, a current entity feature vector and a current relation feature vector, the current entity refers to a component part of the target power grid contained in the current power data, the current relation is used for indicating an association relation between the current entity, performing semantic alignment on the target entity in the target power knowledge graph of the current entity and the target power grid, determining an update entity, performing semantic alignment on the target entity in the target power knowledge graph of the current entity and the target entity feature vector in the target power knowledge graph, determining an update relation, performing data fusion on the current entity feature vector and the target entity feature vector of the target entity in the target power knowledge graph, obtaining an update entity feature vector of the update entity, and performing data fusion on the target relation feature vector of the target relation in the target power knowledge graph, obtaining an update relation feature vector of the update entity, based on the update entity, update entity feature vector, update relation, and update relation feature vector, and update target power grid scheduling decision, and power grid based on the current power grid scheduling decision, target decision, and target graph. According to another aspect of the embodiment of the application, a scheduling decision determining device of a power grid is provided, which comprises a data acquisition module, a first determining module, a data fusion module and a second determining module, wherein the data acquisition module is used for acquiring current power data of a target power grid from a target database, the first determining module is used for determining a current power knowledge graph of the target power grid based on the current power data, the current power knowledge graph comprises a current entity, a current relation, a current entity feature vector and a current relation feature vector, the current entity refers to a component part of the target power grid included in the current power data, the current relation is used for indicating an association relation between the current entity, the semantic alignment module is used for carrying out semantic alignment on the current entity and a target entity in the target power knowledge graph of the target power grid, determining an update entity, and carrying out semantic alignment on the current relation and the target feature vector of the target entity in the target power knowledge graph, and determining an update relation, the data fusion module is used for carrying out data fusion on the current entity feature vector and the target entity feature vector of the target entity in the target power knowledge graph, obtaining the update entity feature vector and the target relation feature vector in the target power knowledge graph, the update