Performance Study of Various Machine Learning Classifiers for Arc Fault Detection in AC Microgrid. Issue 1 (April 2021)

• Record Type: Journal Article
• Authors: Joga, S Ramana Kumar; Sinha, Pampa; Maharana, Manoj Kumar
• Is Part Of: IOP conference series. Volume 1131:Issue 1(2021)
• Publication Date: 2021-04
• Subjects: Arc Fault -- Machine Learning -- Wavelet -- Microgrid -- Fault Detection -- Discrete Wavelet Transform -- PQ disturbances; Materials science -- Periodicals; 620.1105
• DOI: 10.1088/1757-899X/1131/1/012012
• Languages: English
• ISSNs: 1757-8981

Abstract

Abstract: Fault is the abnormal condition in power system, which must be detected as early as possible. It is very important to detect the fault as quick as possible to reduce the effects of fault like equipment damage, property loss and human loss. Arc faults have high power discharge property between two conductors; this property causes damage to the conductors which leads to electric fire between the conductors. It is very necessary to detect these faults immediately to avoid fire accidents. There are various methods to detect these arc faults in microgrid. In this paper voltage and current signals are measured through instrumental transformers and voltage signal is decomposed by the discrete wavelet transform signal processing technique. The decomposed signals are further processed in various machines learning classifier's for detecting the arc fault. The Proposed methodology studies the performance of various machine learning classifiers to detect arc fault in microgrid and it is carried out in MATLAB/Simulink Software.