Open AccessIntelligent busbar protection scheme based on combination of support vector machine and S‐transform
Author(s) -
Gil Milad,
Abdoos Ali Akbar
Publication year - 2017
Publication title -
iet generation, transmission and distribution
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.92
H-Index - 110
eISSN - 1751-8695
pISSN - 1751-8687
DOI - 10.1049/iet-gtd.2016.1686
Subject(s) - busbar , differential protection , current transformer , support vector machine , feature vector , computer science , electronic engineering , transient (computer programming) , electric power system , signal processing , pattern recognition (psychology) , transformer , artificial intelligence , engineering , power (physics) , digital signal processing , electrical engineering , voltage , physics , quantum mechanics , operating system
Differential protection is the main protective scheme of busbar in power systems but its operation degrades during current transformer (CT) saturation conditions. In this study, the busbar differential protection scheme is improved by the addition of a new feature extracted from differential current using S‐transform analysis. S‐transform as a powerful signal processing technique gives a complete visualisation of the signal in both time and frequency domains. The new extracted feature as well as the magnitude of the differential and restraint currents are utilised to increase the security of traditional busbar differential protection scheme. With the addition of this new feature, distinctive regions are created in the feature space which can be separated by an appropriate classifier. In this study, the well‐known classifier, i.e. support vector machine is employed to effectively discriminate internal faults from external faults. To evaluate the performance of the proposed method, a part of Iranian 400‐kV power system grid is simulated in the PSCAD/EMTDC environment. The transient behaviour of CTs during saturation conditions is simulated based on the precise Jiles–Atherton model. The obtained results justify the superiority of the proposed method in terms of speed and security even in noisy conditions.