Power Distribution Network Disconnection Fault Diagnosis Using Data Mining Method

Aiming at the problem that the 10kV distribution network branch line ungrounded disconnection fault is difficult to be detected, using the multi-source data from distribution network, the method of fault diagnosis through data mining is proposed. Firstly, based on the existing data sources, a fault diagnosis feature set with both electrical features and non- electrical features is established. Then, the Principal Component Analysis algorithm is used for feature reduction. Finally, based on the FP-Growth algorithm, the association rules are mined to determine whether the branch line ungrounded disconnection fault has happened. The case analysis using historical operation data shows that the proposed method can quickly and accurately diagnose the occurrence of the 10kV distribution network branch line ungrounded disconnection fault, and has good error tolerance.