APPLICATION OF ARTIFICIAL NEURAL NETWORK BACKPROPAGATION TO PREDICT HOUSEHOLD CONSUMPTION OF ELECTRICITY IN AMBON

  • S. H. Saija Department of Mathematics, Faculty of Mathematics and Natural Sciences Pattimura University
  • Y. A. Lesnussa Department of Mathematics, Faculty of Mathematics and Natural Sciences Pattimura University
  • F. Kondolembang Department of Mathematics, Faculty of Mathematics and Natural Sciences Pattimura University
Keywords: Artificial neural networks; Backpropagation method; Forecasting of household electricity consumption

Abstract

Electricity is one of the energy most widely used in the universe. Electric power demand in Ambon city tends to increase due to the growing of population in Ambon. The necessary of electricity power in Ambon City by utilizing two systems are interconnected such as: PLTD Poka and PLTD Hative Kecil (Galala). In this research forecast the demand for household electricity consumption in 2016 based on validation data from 2011-2015 using Application of Neural Networks Backpropagation method. The validation data are using in JST-Backpropagation training, with the best network architecture that is 20 10 5 1 neurons and 0.8 learning rate, can produce the best pattern with the accuracy is 75% and the value of Mean Square Error is 0.298335. 

Downloads

Download data is not yet available.
Published
2017-07-01
How to Cite
Saija, S., Lesnussa, Y., & Kondolembang, F. (2017). APPLICATION OF ARTIFICIAL NEURAL NETWORK BACKPROPAGATION TO PREDICT HOUSEHOLD CONSUMPTION OF ELECTRICITY IN AMBON. Pattimura Proceeding: Conference of Science and Technology, 1(1), 131-138. Retrieved from https://ojs3.unpatti.ac.id/index.php/pcst/article/view/1528
Issue
2017: Proceedings of the 3rd International Seminar of Basic Science
Section
Articles

Copyright (c) 2017 Pattimura Proceeding: Conference of Science and Technology

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.