A large amount of data has been accumulated in the technical supervision work of power grid enterprises for many years. Considering that the family defects of power equipment have a great influence on equipment condition assessment, this paper introduces an improved hierarchical clustering method based on the analysis of the traditional hierarchical clustering method to automatically identify and cluster the family status data of power equipment. The quantitative assessment of the family defect factor is then given. The family defect assessment is combined with the MMS communication network equipment in the substation communication network, which effectively and reasonably reflects the impact of various types of family defects on the health status of the equipment and provides a strong basis for maintenance decision making.

Aiming at issues such as the inability of intelligent grid business data to be integrated across disciplines, and the inability of data resources to be intelligently analyzed and managed, a knowledge map construction method based on a unified data center for the entire business is proposed. On the basis of using big data technology to collect power grid multi-source data in the unified data center of the whole business, knowledge extraction is performed on structured, semi-structured, and unstructured data using semantic annotation methods to obtain knowledge entities, attributes, and relationships. Knowledge maps are constructed through knowledge fusion technology, and intelligent recommendation results and related information are recommended according to user search content. Experiments show that this method improves the precision and recall rate, and has better intelligent search and analysis capabilities.