Identification of Inrush Current Using a GSA-BP Network

Ensuring a stable and efficient transformer operation is very crucial task nowadays, especially with the integration of modern sensitive electrical equipment appliances down line. However, maloperation still cannot be completely avoided, particularly existence inrush current that possess similar characteristics as fault currents when occurred. Thus, this paper proposes an enhanced method for identification based on backpropagation (BP) network, optimized using genetic simulated annealing algorithms. The proposed has ability to find global optimal solution while avoiding local optima, increased accuracy low calculation complexity. Through extensive simulations, it was found have differences in their harmonic contents, which can exploited those method. annealing–BP (GSA-BP) algorithm make use 200 samples improve detection from 80% 97.5%. Comparative studies performed against existing methods show GSA-BP network superior efficiency being practical real-life application protection system.