Methods, systems, and media for establishing logical model of power grid knowledge graph

Abstract

The invention provides a method, a system and a medium for establishing a logic model for a power grid knowledge graph. The method comprises the following steps: determining an entity mode for defining an entity and attributes thereof; judging a plurality of entities in the first data source as complete entities or incomplete entities; generating entity attribute information of all complete entities and incomplete entities according to an entity mode based on the first data source to generate an entity set of the logic model; determining a relationship mode for defining a relationship betweenthe source entity and the target entity; for the source entity and the target entity existing in the entity set in the second data source, generating corresponding entity relationship information at least based on the second data source according to a relationship mode so as to generate a relationship set of the logic model; and establishing a logic model including entities, attributes and relationships based on the entity set and the relationship set. By means of the scheme, the design defects of an existing model can be overcome, a more reasonable management and control model is provided fora user, and information matching of a unified data model is supported.

Description

technical field [0001] The present invention relates to knowledge graph technology, and more specifically, to a method for establishing a logic model for power grid knowledge graph, a corresponding system, and a computer-readable storage medium. Background technique [0002] With the further development of knowledge graph technology, knowledge graph has laid a foundation for large-scale knowledge base organization and intelligent application with its powerful semantic processing ability and knowledge organization ability. A knowledge graph consists of a large number of entities and entity associations. Through the knowledge graph, entities such as landmarks, people's names, cities, sports teams, buildings, geographical features, movies, celestial bodies, and works of art can be retrieved and information related to these entities can be obtained. This is the key to building intelligent applications that incorporate the collective intelligence of the network and understand th...

AI Technical Summary

Problems solved by technology

This kind of traditional information extraction methods cannot work efficiently in the face of massive relationship categories in Internet texts, that is, it is unrealistic to train an extractor for each target relationship. Network Text We cannot specify the type of relationship in advance

[0004] In addition, the current enterprise-level data model-based knowledge resource classification, intelligent search, and cross-domain knowledge fusion and representation are still in their infancy, lacking an intuitive and popular model interface for relevant managers and business personnel, and the logic of the data model Link search capabilities and static semantic analysis and evaluation capabilities are also severely limited

Data models such as the State Grid Corporation’s Enterprise Common Data Model (SG-CIM) are comprehensive abstractions of corporate-level grid, assets, and financial data. The following problems still exist in the three supporting aspects: (1) The quality of model design still needs to be improved, that is, in the current model design results, there are still some data objects with inconsistent abstraction levels, inaccurate entity relationships, incomplete data objects and (2) The model mapping rate is not high, that is, each unit conducts mapping comparisons based on different versions of physical models, resulting in a relatively low average mapping rate. (3) Lack of tool support, that is, the current data model management and control mostly adopts offline methods, the process is complicated, the communication efficiency is low, and the model design results are relatively abstract, which makes it difficult for personnel at all levels to understand the model, and the application ability is insufficient. The quality of iterative improvement cannot be guaranteed

Images