Desain Optimasi Lokasi, Jumlah dan Kinerja Dynamic Voltage Restorer Menggunakan Artificial Neural Network (STUDI KASUS : PT. PLN (Persero) UP3 Sibolga Penyulang SB 02)

Abstract

The increasing demand for electrical energy each year is a consistent trend, making the distribution system in the power system need to be considered for its stability and reliability. A low stability levels can cause disturbances and result in losses. Improving stability and reliability in the system can be done by installing Custom Power Devices (CPD) such as Dynamic Voltage Restorer (DVR). In this thesis, optimization of the location, quantity, and performance of DVR is carried out using Artificial Neural Network based on voltage stability of the distribution network in the Sibolga SB02 feeder area. From the research results, bus 2, bus 12, bus 24, bus 27, and bus 35 are the optimal locations for placing DVRs with a total of 5 DVRs placed in the system. Short circuit 3-phase simulations were conducted to observe the effect of DVR performance on feeder stability. The research results show that the voltage magnitude drops to 0.1770 pu during disturbances, and the DVR successfully restored the voltage by adding 0.8073 pu, which is above the normal voltage limit of ≥ 0.9 pu.