Prediction of Daily Electricity Use in Residential High-Rise Buildings Using Artificial Neural Networks

Ahmad Rofii, Hanif Ibrahim

2022

Abstract

The high demand for electrical energy from consumers requires producers to provide a reliable but economic supply of electrical energy. Therefore, strategies and methods are needed to match the power generation and demand. This can be achieved by planning a good and proper operation. One of the important steps in planning the operation of the electric power system is predicting the need for electrical loads. One method of predicting electrical loads is to use ANN (Artificial Neural Networks). ANN is an information processing system with characteristics similar to biological neural networks. This method uses ANN with a backpropagation algorithm, and the prediction results are obtained by adding electrical load data (KW) for the selected similar days. ANN processing using MATLAB software. The artificial neural network architecture uses 15 input layers, 15 output layers, and ten hidden layers, and the activation function used is logsig and purelin. Logsig for hidden layers and purelin for output layers. The results of the electrical load prediction using an artificial neural network with the backpropagation method, the Mean Square Error (MSE) value of network training is 0.1, and the MAPE value of data testing is 6.5%. The results of the prediction of electricity use in high-rise residential buildings in February 2023 are predicted.

Paper Citation

in Harvard Style

Rofii A. and Ibrahim H. (2022). Prediction of Daily Electricity Use in Residential High-Rise Buildings Using Artificial Neural Networks. In Proceedings of the 3rd International Seminar and Call for Paper (ISCP) UTA ’45 Jakarta - Volume 1: ISCP UTA'45 Jakarta; ISBN 978-989-758-654-5, SciTePress, pages 293-299. DOI: 10.5220/0011980100003582

in Bibtex Style

@conference{iscp uta'45 jakarta22,
author={Ahmad Rofii and Hanif Ibrahim},
title={Prediction of Daily Electricity Use in Residential High-Rise Buildings Using Artificial Neural Networks},
booktitle={Proceedings of the 3rd International Seminar and Call for Paper (ISCP) UTA ’45 Jakarta - Volume 1: ISCP UTA'45 Jakarta},
year={2022},
pages={293-299},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011980100003582},
isbn={978-989-758-654-5},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 3rd International Seminar and Call for Paper (ISCP) UTA ’45 Jakarta - Volume 1: ISCP UTA'45 Jakarta
TI - Prediction of Daily Electricity Use in Residential High-Rise Buildings Using Artificial Neural Networks
SN - 978-989-758-654-5
AU - Rofii A.
AU - Ibrahim H.
PY - 2022
SP - 293
EP - 299
DO - 10.5220/0011980100003582
PB - SciTePress