Adaptive Fuzzy System for Discrimination of Fault from Non-fault Switching in Over-current Protection

• Published in: Electric power components and systems Vol. 35; no. 12; pp. 1367 - 1384
• Main Authors: Faiz, Jawad, Lotfi-Fard, Saeed
• Format: Journal Article
• Language: English
• Published: Taylor & Francis Group 01-12-2007
• Subjects: ANFIS; fault; motor starting; over-current protection; transformer energizing
• ISSN: 1532-5008; 1532-5016
• DOI: 10.1080/15325000701426138

Power systems switching lead to short duration transients in the voltage and current. These transients have the potential to affect end-user equipment and it may equally affect the correct operation of protection relays close to the component being switched. In the case of non-fault switching events, such transients should not lead to over-current relays operation; therefore, a reliable relay response becomes a critical matter. Meanwhile, proper techniques must be used to prevent mal-trip of such relays, due to transient currents in the power system. In this article, a method is presented to discriminate fault from non-fault events using features extracted from current signal. The proposed method is designed based on extracting the fundamental frequency damping and ratio of the amplitude of the 2nd harmonic to amplitude of the fundamental harmonic and feeding them to an adaptive network-based fuzzy inference system (ANFIS) for discriminating fault from non-fault switching. The performance of this algorithm is demonstrated by simulation of different faults and switching conditions on a power system using EMTDC /PSCAD software.