Research on Building a Standardized Library for Auditing Heterogeneous Information Models in Distributed Data

With the advancement of smart grid transformation, the configuration audit of distribution network communication networks faces challenges such as diversified network devices, numerous manufacturers, and multi-source data management. Traditional methods exhibit significant deficiencies in processing efficiency and data consistency, failing to meet the demands of large-scale data management. This paper proposes a standard library construction method for configuration audit of distribution network data networks. By integrating knowledge graph technology with convolutional neural networks, this method establishes dynamic and static audit libraries, achieving accurate, rapid, and comprehensive management of audit data. The specific process includes the collection and preprocessing of multi-source data, construction of the knowledge graph, application of the inference mechanism, and intelligent retrieval and updating of the standard library. This method significantly improves the accuracy and completeness of data security configuration and optimizes the configuration audit process of distribution network communication networks.