Application of probabilistic neural network with transmission and distribution protection schemes for classification of fault types on radial, loop, and underground structures

摘要: This paper proposes the development of transmission and distribution protection schemes to classify faults along systems. The systems under consideration are composed a 500-kV radial (two-bus single circuit), loop (three-bus double circuit) structure systems, 115-kV underground system. complex system shows advantage proposed method. A decision algorithm based on discrete wavelet transform (DWT) probabilistic neural network is investigated for inclusion in Fault signals each case extracted several scales DWT decompose high-frequency components from fault using mother daubechies4. maximum coefficients at 1/4 cycle that can detect used as input patterns training process algorithm. In addition, technique back-propagation also compared with this paper. Moreover, real signal an experimental set-up was investigated. Based accurate results simulation signal, it be concluded adequate other power different line models applied actual even if accuracy slightly reduced by effect noise Thus, overall show faulty bus types satisfactory results.