Article Per Year (5 Year)
p-Index From 2020 - 2025
0
P-Index
This Author published in this journals
All Journal Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Anim Rofi'ah
Fakultas Ilmu Komputer, Universitas Brawijaya
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Peramalan Jumlah Pemakaian Air di PT Pembangkit Jawa Bali Unit Gresik dengan Extreme Learning Machine dan Ant Colony Optimization Anim Rofi'ah; Imam Cholissodin; Candra Dewi
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 2 No 11 (2018): November 2018
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1134.215 KB)

Abstract

PT. PJB Unit Gresik using seawater as a steam power plant. Water has advantages such as it is high availability and environmentally friendly. However, seawater requires a refining process in order to be used. Using seawater as a power plant often experiences water-reduction problems caused by certain problems, such a pipeline leakage, tempering, and removal of gases that still contain water so that additional water is required to keep the turbin working. To anticipate the lack of water that can inhibit the process, an intelligent system required to estimate the amount of water that generation process needed. One of forecasting method is Extreme Learning Machine (ELM), to maximize forecasting results with optimization algorithm Ant Colony Optimization that can be used in the optimization input weight and bias of ELM parameters. After optimization process for ELM parameters, then the next process is training and testing to get forecasting result. This study uses 103 data. Based on the research, the optimal parameter number of ants is 40, the parameter range of the input weight is 0 to 1, the using 82 of training data and 21 testing data (80%: 20%), and the maximum iteration is 500. From these parameters obtained the MAPE value for ELM-ACO is 0.170% with 3799.200 ms running time and for the ELM algorithm the MAPE value is 4.851% with 162.400 ms, so the optimization of ELM parameters can improve the forecasting results.
Co-Authors Candra Dewi Imam Cholissodin