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
Rosintan Fatwa
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 
Penerapan Metode Extreme Learning Machine Untuk Prediksi Konsumsi Batubara Sektor Pembangkit Listrik Tenaga Uap Rosintan Fatwa; Imam Cholissodin; Yuita Arum Sari
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 11 (2019): November 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (420.754 KB)

Abstract

PLTU is a power station that utilizes coal as fuel. The PLTU sector is a dominant sector in absorbing domestic coal. During the period 2010 - 2015, coal consumption continued to increase along with the 35,000 MW power plant project which was designed in the 2015-2019 period, 19,940 MW (56%) was a coal-fired power plant. Based on data from the Director General of Mineral and Coal at the Ministry of Energy and Mineral Resources, said that the increase in coal consumption is due to the growing PLTU and the economic development which is directly proportional to the increase in national coal consumption. Based on these problems, the prediction of coal consumption in the power plant sector is needed so that coal consumption can be controlled in accordance with its production. In this study, the prediction process is carried out in several processes, namely data normalization, prediction calculation using Extreme Learning Machine, data denormalization, and error values ​​using MAPE. Based on the results of tests conducted on daily coal consumption data for 2018 at the Tanjung Jati B PLTU Unit 1 & 2 obtained the smallest MAPE value of 6.603% with many features 2, the number of hidden neurons as much as 4, and the comparison of the percentage of training data and testing data 70 %: 30% using the Sigmoid activation function.
Co-Authors Imam Cholissodin Yuita Arum Sari