Open AccessAnalysis of Artificial Neural Network in Predicting the Fuel Consumption by Type of Power Plant
Author(s) -
Widodo Saputra,
P Poningsih,
Muhammad Ridwan Lubis,
Sundari Retno Andani,
Irfan Sudahri Damanik,
Anjar Wanto
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1255/1/012069
Subject(s) - artificial neural network , fuel efficiency , power consumption , architecture , power station , power (physics) , consumption (sociology) , electric power , computer science , automotive engineering , artificial intelligence , engineering , electrical engineering , geography , archaeology , social science , physics , quantum mechanics , sociology
The electric is one of the needs for human, the growth of the electric power in Indonesia is very increased. There are 13 types of power plants in Indonesia, including to require the fuel in its operational. Fuel consumption needs to be recorded at regular intervals so that the needs of fuel for the power plant remain fulfilled. This research discusses about the predictions fuel consumption based on the type of power plants. The method used is the Artificial Neural Network with the back propagation algorithm. This method is good enough to use in predicting the data. The data used is fuel consumption data based on the type of power plant of the year 2014-2016. The best architecture is 8-23-1 that gets results accuracy of 88%, epoch 6016 iteration and MSE 0.0005509801. From these results then predict using the 8-23-1 architecture is the best architecture.