ML-Based Intermittent Fault Detection, Classification, and Branch Identification in a Distribution Network

The accurate detection and identification of intermittent cable faults are helpful in improving the reliability distribution system. This paper proposes fault for networks based on machine-learning (ML) techniques. For this reason, IEEE 33 bus system is simulated radial mesh topologies by considering all possible single- three-phase electrical limitations to collect high-resolution voltage current waveforms. Moreover, simulation investigates considers various cases including low-impedance (LIFs) high-impedance (HIFs) with a short long duration. collected data from used detection, classification, branch using eight supervised learning methods. A comparison between accuracy error these ML classifiers shows that gradient booster (GB) K-nearest neighbors (KNN) have best performance three objectives. However, GB has very high computation time compared KNN.