Electric quantity perspective recommendation method and system based on graph convolutional neural network

Abstract

The invention discloses an electric quantity view angle recommendation method and system based on a graph convolutional neural network, and relates to the technical field of electric power systems. The method comprises the following steps: 1, generating an electric quantity statistics view angle based on an electric quantity graph database, generating a corresponding spectrogram sample through a query path of a historical search record, and constructing a view angle graph database; 2, based on the constructed graph database, training an intelligent visual angle recommendation network based on a graph volume neural network, and learning a mapping relation between entities and search results in the graph; and step 3, based on the trained intelligent view angle recommendation network, calculating recommendation values of all statistical view angles, sorting the recommendation values of all the statistical view angles, generating a recommendation model, extracting statistical view angle features through a structured electricity map database, learning an association relationship between the electricity statistical map features and the statistical view angles by using a map convolutional neural network, and performing recommendation on the electricity statistical map features and the statistical view angles. And a targeted view angle recommendation function based on the daily electric quantity statistical service is realized.

Description

technical field [0001] The invention belongs to the technical field of electric power systems, and in particular relates to a method and system for recommending a power perspective based on a graph convolutional neural network. Background technique [0002] As the total amount of data and knowledge in the power system continues to grow, it is difficult to accurately express the association and synergy between knowledge with the traditional manual knowledge management method, and it is also difficult to realize the extraction, management and utilization of heterogeneous knowledge, which greatly affects the Knowledge management and utilization efficiency. In order to achieve effective knowledge organization and management, the knowledge graph technology, which has been widely used in the computer field in recent years, has been introduced into the power system field. The power system knowledge graph inherits the advantages of ontology and semantic web technologies in heteroge...

AI Technical Summary

Problems solved by technology

[0003] The existing electricity statistics perspectives are basically manually selected and calculated by the planners. Not only will the statistics of the electricity perspective consume high time and labor costs, but the choice of statistical perspectives is often limited by the historical experience of the planners, who have insufficient experience. The testers usually face the problem that it is difficult to effectively choose a reasonable statistical perspective for analysis

Images