SACO WORK VELOCITY ANALYSIS ON MANEUVERING CHARGES AT NGURAH RAI AIRPORT USING ARTIFICIAL NEURAl, NETWORK

SKRIPSI

Abstract

By : Dewa Ngakan Made Barel

Email : barel_dewa@yahoo.com

Faculties : Fakultas Teknik

Department : S1 Teknik Elektro

Ngurah Rai Airport as one of the burdens in supply with two feeders, feeder Gayatri as the feeder main and feeder Bandara as a feeder backup, SACO is used as an automatic switch that can change the supply load of feeder Gayatri to feeder Bandara when feeder Gayatri impaired, airports permit limits the maximum voltage drop is 0.5 kV feeders of nominal voltage value, and therefore the analysis is the calculation of the voltage drop and short circuit on a feeder Gayatri. This analysis uses the program ANN (Artificial Neural Network) that is in it. Input data consists of impedance, short circuit 3-phase, 2-phase and single phase to ground, while its output in the form of the value of the voltage drop, the value of the time delay short circuit 3-phase, 2-phase, and the first phase to ground. Parameters used include the number of iterations is 5000 epoch, the learning speed of 0.3 and a hidden layer of 40 hidden layer, where in the target value of the test is 0.00001. Data is divided into two for its testing process, the training data and test data, of the training process and testing conducted voltage drop values obtained at 100% feeder is 274.7 volts with a value of MSE (Mean Squarred Error) is 45.5 , Value fastest time delay is 0.3 seconds at the point of interruption 5%, and the MSE with the best result is 0.000321126 for the first phase to ground short circuit.

Keyword : Drop voltage, Short-circuit currents, and Artificial Neural Network (ANN)

Download : File 1 File 2 File 3