An extreme learning machine based fast and accurate adaptive distance relaying scheme. (December 2015)

• Record Type: Journal Article
• Authors: Dubey, Rahul; Samantaray, S.R.; Panigrahi, B.K.
• Is Part Of: International journal of electrical power & energy systems. Volume 73(2015:Dec.)
• Page Start: 1002
• Page End: 1014
• Publication Date: 2015-12
• Subjects: Adaptive distance relaying scheme (ADRS) -- Fast adaptive distance relaying scheme (FADRS) -- Power swing -- Artificial neural networks (ANNs) -- Extreme learning machine (ELM); Electrical engineering -- Periodicals; Electric power systems -- Periodicals; Électrotechnique -- Périodiques; Réseaux électriques (Énergie) -- Périodiques; Electric power systems; Electrical engineering; Periodicals; 621.3
• DOI: 10.1016/j.ijepes.2015.06.024
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: An ELM based adaptive and intelligent relaying scheme is proposed. The impact of shunt capacitance and mutual coupling is studied. The proposed relaying scheme is fast and accurate. Provides highly improved and reliable performance compared to existing scheme. Abstract: The ideal trip characteristics of the distance relay is greatly affected by pre-fault system conditions, ground fault resistance, shunt capacitance and mutual coupling of transmission network. This paper presents an extreme learning machine (ELM) based fast and accurate adaptive relaying scheme for stand-alone distance protection of transmission network. The proposed ELM based fast adaptive distance relaying scheme (FADRS) is extensively validated on the two terminal transmission lines with complex mutual coupling and shunt capacitance and, the performance is compared with the conventional artificial neural networks (ANNs) based adaptive distance relaying scheme (ADRS). The simulation results show significant improvement in the performance indices such as relay speed and selectivity. Further, the performance of proposed FADRS is tested for stressed condition such as power swing and found to be effective and reliable.