CN-122022015-A - Method and system for constructing full life cycle data model of medium-voltage power failure event

CN122022015ACN 122022015 ACN122022015 ACN 122022015ACN-122022015-A

Abstract

A method and a system for constructing a full life cycle data model of a medium-voltage outage event comprise the steps of constructing a static object layer according to a power grid topological connection relation based on ID and attribute sets of equipment objects and environment objects, recording real-time events through a pre-designed standardized outage event mode, fusing the static object layer and the real-time events by utilizing a graph database to construct a real-time relation map layer, integrating an intelligent analysis function in the real-time relation map layer, and constructing a dynamic knowledge map model. The invention realizes the complete and continuous data depiction of the power failure event, which is characterized in that the complex association between the inside and the outside of the power failure event is explicitly expressed through the knowledge graph technology, and the power failure situation can be dynamically updated in real time, so that the power failure situation is visual, clear and comprehensive, the intelligent level of the businesses such as fault location, rush repair command, power grid planning and the like is obviously improved, the power failure time is effectively shortened, and the power supply reliability and the customer satisfaction are improved.

Inventors

LI JIA
ZHANG LIN
SHI XUEFENG
GAO FEI
ZHANG YU
LI JIANFANG
LI YAJIE
ZHAO SHANSHAN
LIU JINJIE
XU DONGJIE
HAO ZEPENG
WANG QINGJIE
LIU SHUPIN
XU MENGYAO
ZHANG ZHIHUA
LEI YUHANG
QIU ZEKAI
NIU SHENGLI
YU YICHEN
LIU ZHIXIANG
DUAN XIANGJUN
FENG DEZHI
ZHAO YUMENG
XU YUANYUAN
QIAN ZHIYAN
Bai Muke
LI YUNSHUO

Assignees

中国电力科学研究院有限公司
国网陕西省电力有限公司电力科学研究院
国家电网有限公司
国网山西省电力有限公司电力科学研究院

Dates

Publication Date: 20260512
Application Date: 20251231

Claims (16)

1. The method for constructing the full life cycle data model of the medium-voltage power failure event is characterized by comprising the following steps of: based on the ID and attribute set of the equipment object and the environment object, constructing a static object layer according to the topological connection relation of the power grid; recording real-time events through a pre-designed standardized power outage event mode, wherein the standardized power outage event mode is used for recording all state changes in the whole life cycle of power outage; fusing the static object layer and the real-time event by using a graph database to construct a real-time relationship map layer; And integrating an intelligent analysis function in the real-time relationship graph layer to construct a dynamic knowledge graph model.
2. The method of claim 1, wherein the constructing a static object layer according to the grid topology connection relationship based on the ID and the attribute set of the device object and the environment object comprises: Based on the ID and attribute set of the equipment object and the environment object, modeling the relation among the objects by adopting an information technology and a data analysis method, and constructing the logic attribute and connection relation of the objects; And on the basis of the logical attribute and the connection relation of the object, carrying out deep integration on the equipment object and the geospatial data by a geographic information system technology to construct a static object layer.
3. The method of claim 1, further comprising, after the building of the static object layer according to the grid topology connection relationship based on the ID and the set of attributes of the device object and the environment object: And carrying out data cleaning and checking on the data of the static object layer by utilizing a data cleaning and checking mechanism.
4. The method of claim 1, wherein the standardized power outage event pattern comprises events and event types; The event comprises an event ID, a type, a time stamp, an associated object list, a geographic position, an event source and a key index; The event types comprise overload early warning, fault tripping, protection action, switch opening, fault line inspection starting, fault point positioning, isolation operation execution, power supply switching operation execution, repair completion and power supply recovery.
5. The method of claim 1, wherein integrating the intelligent analysis function in the real-time relationship graph layer to construct the dynamic knowledge graph model comprises: and integrating the association inquiry and calculation capability of the map in the real-time relationship map layer, rapidly positioning the root cause of the fault, generating an optimal recovery strategy, and predicting the reliability risk of the line or the area.
6. A medium voltage blackout event full life cycle data model construction system, comprising: the topology construction module is used for constructing a static object layer according to the topological connection relation of the power grid based on the IDs and attribute sets of the equipment object and the environment object; The event recording module is used for recording real-time events through a pre-designed standardized power outage event mode, wherein the standardized power outage event mode is used for recording all state changes in the whole life cycle of power outage; the map construction module is used for fusing the static object layer with the real-time event by using the map database to construct a real-time relation map layer; and the model construction module is used for integrating the intelligent analysis function in the real-time relation map layer and constructing a dynamic knowledge map model.
7. The system of claim 6, wherein the topology construction module is specifically configured to: Based on the ID and attribute set of the equipment object and the environment object, modeling the relation among the objects by adopting an information technology and a data analysis method, and constructing the logic attribute and connection relation of the objects; And on the basis of the logical attribute and the connection relation of the object, carrying out deep integration on the equipment object and the geospatial data by a geographic information system technology to construct a static object layer.
8. The system of claim 6, wherein the model building module is specifically configured to: and integrating the association inquiry and calculation capability of the map in the real-time relationship map layer, rapidly positioning the root cause of the fault, generating an optimal recovery strategy, and predicting the reliability risk of the line or the area.
9. A method for analyzing a medium voltage outage event based on a full life cycle data model, comprising: Acquiring operation information of a prediction area or a power grid; inputting the operation information of the prediction area or the power grid into a pre-constructed dynamic knowledge graph model to obtain one or more of failure reasons, optimal recovery strategies and reliability risk assessment values of the prediction area or the power grid; The pre-constructed dynamic knowledge graph model is constructed by adopting the medium-voltage power failure event full life cycle data model construction method according to any one of claims 1-5.
10. The method of claim 9, wherein inputting the operation information of the predicted area or the power grid into a pre-constructed dynamic knowledge graph model to obtain one or more of a failure cause, an optimal recovery strategy, and a reliability risk assessment value of the predicted area or the power grid, comprises: When a power failure event occurs in the operation information of the prediction area or the power grid, rapidly positioning the root cause of the fault through real-time topology analysis according to the power failure event and a pre-constructed dynamic knowledge graph model, and generating an optimal recovery strategy; When no outage event occurs in the operation information of the prediction area or the power grid, predicting the reliability risk value of the line or the area according to the historical outage event, the equipment working condition and the environmental data which are associated in the pre-constructed dynamic knowledge graph model.
11. The method of claim 10, wherein when a blackout event occurs in the operation information of the prediction area or the power grid, the method quickly locates the root cause of the fault through real-time topology analysis according to the blackout event and a pre-constructed dynamic knowledge graph model, and generates an optimal recovery strategy, which comprises: Calculating a power grid device and a user list affected by power outage according to a power outage event and a dynamic knowledge graph model which is built in advance; And positioning the root cause of the fault according to the relation between the event and the object in the pre-constructed dynamic knowledge graph model, and generating an optimal recovery strategy.
12. A system for analyzing medium voltage power outage events based on a full life cycle data model, comprising: the information acquisition module is used for acquiring the operation information of the prediction area or the power grid; The intelligent analysis module is used for inputting the operation information of the prediction area or the power grid into a pre-constructed dynamic knowledge graph model to obtain one or more of failure reasons, optimal recovery strategies and reliability risk assessment values of the prediction area or the power grid; The pre-constructed dynamic knowledge graph model is constructed by adopting the medium-voltage power failure event full life cycle data model construction method according to any one of claims 1-5.
13. The system of claim 12, wherein the intelligent analysis module comprises: The event analysis sub-module is used for rapidly positioning the root cause of the fault through real-time topology analysis according to the power failure event and a pre-constructed dynamic knowledge graph model when the power failure event occurs in the operation information of the prediction area or the power grid, and generating an optimal recovery strategy; And the risk assessment sub-module is used for predicting the reliability risk value of the line or the area according to the historical power failure event, the equipment working condition and the environmental data associated in the pre-built dynamic knowledge graph model when the power failure event does not exist in the operation information of the predicted area or the power grid.
14. The system of claim 13, wherein the event analysis submodule is specifically configured to: Calculating a power grid device and a user list affected by power outage according to a power outage event and a dynamic knowledge graph model which is built in advance; And positioning the root cause of the fault according to the relation between the event and the object in the pre-constructed dynamic knowledge graph model, and generating an optimal recovery strategy.
15. The electronic equipment is characterized by comprising at least one processor and a memory, wherein the memory and the processor are connected through a bus; the memory is used for storing one or more programs; a medium voltage outage event full life cycle data model construction method according to any one of claims 1 to 5, or a method of analyzing a medium voltage outage event based on a full life cycle data model according to any one of claims 9 to 11, when said one or more programs are executed by said at least one processor.
16. A readable storage medium having stored thereon an execution program which, when executed, implements a medium voltage outage event full life cycle data model construction method according to any one of claims 1 to 5, or a method of analyzing a medium voltage outage event based on a full life cycle data model according to any one of claims 9 to 11.

Description

Method and system for constructing full life cycle data model of medium-voltage power failure event Technical Field The invention relates to the technical field of power grid informatization and intellectualization, in particular to a method and a system for constructing a full life cycle data model of a medium-voltage power failure event. Background The power supply reliability is a core index for measuring the operation level of the power distribution network, and the medium-voltage power distribution network is a key link for influencing the power failure experience of users. Currently, management of medium voltage outage events is mostly dependent on a plurality of independent information systems, such as a Production Management System (PMS), a dispatch automation System (SCADA), an electricity consumption information acquisition system, a Geographic Information System (GIS), etc. Each of these systems records a portion of the data associated with the outage, forming a "data island". The data model in the prior art has obvious defects: the model flattening is that only basic attributes such as line name, start time, end time, outage number of users and the like of a outage event are recorded, and a causal chain and a dynamic evolution process of the outage cannot be described. The post record type model is mainly used for meeting the statistical reporting of reliability indexes (such as SAIDI and SAIFI), is a passive record and lacks supporting capability for pre-warning and in-event command. The relevance is weak, and the power failure event and key entities such as power grid equipment, an operation environment, disposal personnel and the like lack of explicit and dynamic relevance relationship, so that deep analysis is difficult to carry out. These limitations result in insufficient mining of data value, failing to meet the requirements of advanced applications such as active power grids, accurate rush repairs, and intelligent decisions. Disclosure of Invention In order to solve the problems of isolation, flatness, static state and the like of the existing medium-voltage power failure event data model, the invention provides a method for constructing a medium-voltage power failure event full life cycle data model, which comprises the following steps: based on the ID and attribute set of the equipment object and the environment object, constructing a static object layer according to the topological connection relation of the power grid; recording real-time events through a pre-designed standardized power outage event mode, wherein the standardized power outage event mode is used for recording all state changes in the whole life cycle of power outage; fusing the static object layer and the real-time event by using a graph database to construct a real-time relationship map layer; And integrating an intelligent analysis function in the real-time relationship graph layer to construct a dynamic knowledge graph model. Preferably, the constructing a static object layer according to the connection relationship of the power grid topology based on the ID and the attribute set of the device object and the environment object includes: Based on the ID and attribute set of the equipment object and the environment object, modeling the relation among the objects by adopting an information technology and a data analysis method, and constructing the logic attribute and connection relation of the objects; And on the basis of the logical attribute and the connection relation of the object, carrying out deep integration on the equipment object and the geospatial data by a geographic information system technology to construct a static object layer. Preferably, after the static object layer is built according to the network topology connection relation based on the ID and the attribute set of the device object and the environment object, the method further comprises: And carrying out data cleaning and checking on the data of the static object layer by utilizing a data cleaning and checking mechanism. Preferably, the standardized power outage event mode comprises an event and an event type; The event comprises an event ID, a type, a time stamp, an associated object list, a geographic position, an event source and a key index; The event types comprise overload early warning, fault tripping, protection action, switch opening, fault line inspection starting, fault point positioning, isolation operation execution, power supply switching operation execution, repair completion and power supply recovery. Preferably, the integrating the intelligent analysis function in the real-time relationship graph layer to construct a dynamic knowledge graph model includes: and integrating the association inquiry and calculation capability of the map in the real-time relationship map layer, rapidly positioning the root cause of the fault, generating an optimal recovery strategy, and predicting the reliability risk of the line or the area. In sti