Currently large amounts of defect records of power equipment are left in the power grid management system

which can benefit the defect management if retrieved accurately. A retrieval method for defect records of power equipment is developed based on knowledge graph technology. Firstly

a method for constructing the knowledge graph of power equipment defect with existing records is presented. Then

based on the constructed knowledge graph

a retrieval method for power equipment defect using graph search is proposed. Finally

the proposed retrieval method is compared with traditional sematic analysis methods based on machine learning. The results show that the proposed retrieval method based on knowledge graph technology can significantly improve the retrieval effect of power equipment defect records.