Faulty Section Identification in Medium-Voltage Distribution Networks by Time-Frequency Feature Fusion of Transient Zero-Mode Current
Wangqing Mao, Xiuru Wang, Jian Sun, Hangyu Zhao, Rao Fu, Mengmeng Cao| Pages 546-564|

 Abstract— The accuracy of fault section identification using a single criterion is compromised under certain complex fault conditions, particularly when a single-phase grounding fault occurs in a neutral point ungrounded system, where the steady-state fault signal is weak. This paper addresses this challenge by investigating the distribution feature of zero-mode current in such systems and proposes three distinct methods for locating single-phase ground faults: the transient power method, characteristic frequency method, and transient current similarity method. Each method has its own strengths and can complement each other. Hence, fuzzy theory is utilized to establish membership functions and weight coefficients for the three methods, respectively. The weighted coefficient method is then employed to integrate the three criteria, eliminating issues such as blind areas and misjudgments in fault section identification resulting from relying solely on a single criterion. A simulation model is developed using PSCAD/EMTDC. The results obtained under various fault conditions confirm the effectiveness and robustness of the fusion method in accurately identifying fault sections.


DOI: https://doi.org/10.5455/jjee.204-1744298232