Machine learning techniques applied on a nine-phase transmission line for faults classification and location
Patrick S. Pouabe Eboule, J.H.C. Pretorius, Nhlanhla Mbuli
Abstract
Nowadays, the applications of machine learning techniques have widely been implemented in various domains of the power system and especially to predict three-phase transmission line faults. This paper compares the results obtained from two powerful machine learning techniques such as Concurrent Neuro-Fuzzy and Decision Tree applied to predict faults classification and their location on a nine-phase transmission line system. The results obtained demonstrated that the utilization of these artificial intelligence techniques which have been well applied on various and diversified systems could also been applied on a future nine-phase transmission line system that carry 750 kV over 200 km to classify and locate various fault types and then, to increase the yield of the line.