A novel graph search and machine learning method to detect and locate high impedance fault zone in distribution system

High impedance fault (HIF) is difficult to detect by conventional overcurrent protection relays due the lower current values, which are normally than normal current. A fast and reliable algorithm required this type of fault. This paper proposes a novel method for detecting location HIF zone in distribution system using graph theory-based detection technique along with Random Search Multilevel Support Vector Machine (RSMSVM) classify faulted zone. Due shift in-variance property “Dual Tree Complex Wavelet Transform (DTCWT),” has been used, paper, decompose voltage/current waveform collect signature signals feed optimized RSMSVM model classifying The proposed evaluated on IEEE 33-bus also 39 bus test under noisy conditions. network integration distributed generation.