ABSTRACT
Forecasting is an attempt to predict what happens in the future with the data from the past, based on scientific methods and quantitative systematic review. This study reviews Neural Network backpropagation algorithms for long-term power load forecasting for every province in the Java-Bali electricity system. The purpose of this paper is to predict the power load of each province in the Java-Bali electricity system and to compare the results with the electric load forecasting using backpropagation algorithms with conventional methods for the 2015–24 Electricity Supply Business Plan RUPTL (Ren-cana Umum Penyediaan Tenaga Listrik) of the state-owned power utility firm PT Perusahaan Listrik Negara (Pesero). The results of electric power load demand calculation and simulation of long-term forecasting for 2015–24 were not much different from forecasting RUPTL data. Average annual growth of electricity load for Daerah Khusus Ibu Kota Jakarta is 8.98%, Banten Province 8.74%, West Java Province 8.76%, Central Java Province 8.73%, Daerah Istimewa Yogyakarta 8.05%, East Java Province 8.09%, and Bali Province 8.13%.
