e-journal
A wavelet-support vector machine combined approach for location of transmission line faults
Abstract:
This paper presents a wavelet-SVM (support vector machine) combined approach for digital relaying. The proposed algorithm for fault location employs wavelet transform together with SVM. Computer simulations using MATLAB have been conducted for a 300 km, 400 kV line. The SVM has been trained using simulated data to give desired output for fault locations for faults at any unknown distances. Simulation results indicate that the localisation algorithm is immune to effects of faults inception angle, impedance and distance. The most significant contribution of this paper is that the proposed location algorithm has a maximum error of 2% with a computational time of about one cycle. and it performs better compared to other fault location techniques based on fuzzy inference systems (FIS), validating the efficacy of proposed method.
KEYWORDS: Wavelet transform; multi-resolution analysis (MRA); fuzzy inference system (FIS); support vector machine (SVM); fault location.
Tidak ada salinan data
Tidak tersedia versi lain