“Estimasi Jenis dan Lokasi Titik Gangguan pada Saluran Distribusi Bawah Tanah PT. Pertamina RU II Dumai dengan Menggunakan Metode Adaptive Neuro Fuzzy Inference System (ANFIS)”

This paper describes a method of Adaptive Neuro Fuzzy Inference System (ANFIS) for identifying and locating short circuit faults in underground power distribution line at PT. Pertamina RU II Dumai. The underground power distribution line is modeled using ATP-EMTP software. The first ANFIS is used to determine the type of fault such as single line to ground faults (A-G), double line to ground faults (A-B-G), double line faults (A-B), and 3 phase faults (A-B-C). The second ANFIS is used to determine the fault location. The input for identifying faults are the fundamental component of three phase currents and the zero sequence current. The input for locating faults are the fundamental component of three phase currents and voltages. The simulation's results show that ANFIS is able to determine the type of fault with RMSE of A-G is 6,8840.10-8, A-B-G is 6,6182.10-6, A-B is 1,2309.10-5, and A-B-C is 2,6709.10-6. Furthermore, this method is also able to determine the fault location with RMSE of A-G is 0,0024, A-B-G is 0,0009, A-B is 7,3304.10-5, and A-B-C is 0,0161