Data Mining-Based Abnormal State Diagnosis Method for Active Distribution Networks

The existing methods for diagnosing abnormal states in power systems suffer from issues such as prolonged diagnosis time and inaccurate classification. With the development of big data, data analysis methods have been applied to various areas of the power system. In this paper, focusing on the field of abnormal state detection in distribution networks, a diagnostic method based on data mining techniques is proposed. Firstly, an improved K-means clustering algorithm is employed to classify data in the database into clusters. Subsequently, the FP-Growth association rule algorithm is used to associate clusters with abnormal state types, thereby establishing a database for abnormal state diagnosis in active distribution networks. Finally, a case study simulation is conducted using the IEEE 9 bus system to validate the feasibility of the proposed abnormal state diagnosis method for active distribution networks.