Pemantauan Gangguan Pada Saluran Transmisi Berbasis Titik Koordinat Menara

Abstract

This research uses an Artificial Neural Network (ANN) method to estimate fault location based on the coordinate point. The 150 kV transmission system is modeled using the Matlab R2016a software. The modeled transmission system is the transmission from Koto Panjang (KP) bus to Garuda Sakti (GS) bus in Riau Province with 64,11017 Km length. Min – max normalization method is used to process the peak value of the voltage phase (VA, VB, VC), current (IA, IB, IC) and zero sequence current (INOL). ANN estimation is used to estimate the coordinate point of fault that occurs on electrical transmission line. The training and testing data are generated by simulating each type of short circuit in Matlab/Simulink R2016a with variation of fault location. Variation of fault location are based on distance of tower from the KP bus. The obtained result from 10 ANN estimation testing are coordinate point estimation in latitude and longitude form. The ANN model that has smallest average error percentage is the ACG fault ANN model, which is 0,004013 % and the largest is the ABG fault ANN model, which is 0,78527 %. The output of the ANN estimation then denormalized and plotted to the actual map using ArcMap 10.4.1.
Keywords: ArcMap, Artificial neural network, Coordinate point, Matlab, Short circuit, Simulink, Transmission line