Prediction of Electricity Usage with Back-propagation Neural Network

Keywords: Back-propagation Neural Network, Electricity Usage, MSE, RMSE


The use of electricity has become a need that is increasing day by day. So it is not surprising that the problem of using electricity has attracted the attention of many researchers to research it. Electricity users make various efforts and ways to save on the use of electrical energy. One of them is saving electricity usage by electricity users using electrical energy-efficient equipment. That is why the previous research confirms the need for interventions to reduce the use of electrical energy. Therefore, this study aims to predict electricity use and measure the performance of the anticipated results of electricity use. This study uses the back-propagation method in predicting the use of electricity. This study concluded that the backpropagation architectural model with better performance is the six hidden layer architecture, 0.4 learning rate, and the Root Means Square Error (RMSE) value of 0.203424. Meanwhile, the training data test results get the best architectural model on hidden layer 8 with a learning rate of 0.3 with an RMSE performance value of 0.035811. The prediction results show that the prediction of electricity consumption is close to the actual data of actual electricity consumption.


[1] R. Kasperowicz, “Electricity consumption and economic growth: Evidence from Poland,” Journal of International
Studies, vol. 7, no. 1, pp. 46–57, 2014.
[2] S. Mahfoudh and M. Ben Amar, “The Importance Of Electricity Consumption In Economic Growth: The Example
Of African Nations,” The Journal of Energy and Development, vol. 40, no. 1, pp. 99–110, Jan. 2014, [Online].
[3] Y.-R. Zeng, Y. Zeng, B. Choi, and L. Wang, “Multifactor-influenced energy consumption forecasting using
enhanced back-propagation neural network,” Energy, vol. 127, pp. 381–396, 2017, doi:
[4] N. Tewathia, “Determinants of the household electricity consumption: A case study of Delhi,” International Journal
of Energy Economics and Policy, vol. 4, no. 3, pp. 337–348, 2014.
[5] K. Gram-Hanssen, “Households’ energy use-which is the more important: efficient technologies or user practices?,”
[6] V. Lesic, W. B. De Bruin, M. C. Davis, T. Krishnamurti, and I. M. L. Azevedo, “Consumers’ perceptions of energy use and energy savings: A literature review,” Environmental Research Letters, vol. 13, no. 3, pp. 1–13, 2018, doi:
[7] B. Karlin, N. Davis, A. Sanguinetti, K. Gamble, D. Kirkby, and D. Stokols, “Dimensions of Conservation: Exploring
Differences Among Energy Behaviors,” Environment and Behavior, vol. 46, no. 4, pp. 423–452, Dec. 2012, doi:
[8] J. L. Reyna and M. V Chester, “Energy efficiency to reduce residential electricity and natural gas use under climate
change,” Nature Communications, vol. 8, no. 1, pp. 1–12, 2017, doi: 10.1038/ncomms14916.
[9] J. Joseph, N. James, S. Sachin, V. Vincent, and S. Izudheen, “Electricity Load Prediction Using Artificial Neural
Network By Back Propagation,” International Journal of Innovate and Research Technology, vol. 2, no. 9, pp. 298–
301, 2017.
[10] Y. Wu, L. Deng, G. Li, J. Zhu, and L. Shi, “Spatio-Temporal Backpropagation for Training High-Performance
Spiking Neural Networks,” Frontiers in Neuroscience, vol. 12, 2018, doi: 10.3389/fnins.2018.00331.
[11] M. Buscema, “Back propagation neural networks.,” Substance use & misuse, vol. 33, no. 2, pp. 233–270, Jan. 1998,
doi: 10.3109/10826089809115863.
[12] L. M. Yan and O. H. Choon, “Neural network forecasting on electricity consumption in Malaysia,” Innere Medizin,
no. mindestens 6, p. 154, 2005.
[13] N. I. Nwulu and O. P. Agboola, “Modelling and predicting electricity consumption using artificial neural networks,”
in 2012 11th International Conference on Environment and Electrical Engineering, 2012, pp. 1059–1063, doi:
[14] A. Anggrawan, S. Satuang, and M. N. Abdillah, “Sistem Pakar Diagnosis Penyakit Ayam Broiler Menggunakan
Forward Chaining dan Certainty Factor,” MATRIK : Jurnal Manajemen, Teknik Informatika dan Rekayasa
Komputer, vol. 20, no. 1, pp. 97–108, 2020, doi: 10.30812/matrik.v20i1.847.
[15] Y. Lu, W. Gao, S. Kuroki, and J. Ge, “Household characteristics and electricity end-use under dynamic pricing in
the collective housing complex of a Japanese smart community,” Journal of Asian Architecture and Building
Engineering, pp. 1–16, Oct. 2021, doi: 10.1080/13467581.2021.1987244.
[16] A. Anggrawan, C. Satria, C. K. Nuraini, L. -, N. G. A. Dasriani, and M. -, “Machine Learning for Diagnosing Drug
Users and Types of Drugs Used,” International Journal of Advanced Computer Science and Applications, vol. 12,
no. 11, pp. 111–118, 2021, doi: 10.14569/ijacsa.2021.0121113.
[17] A. Anggrawan, N. Ibrahim, S. Muslim, and C. Satria, “Interaction between learning style and gender in mixed
learning with 40% face-to-face learning and 60% online learning,” International Journal of Advanced Computer
Science and Applications, vol. 10, no. 5, pp. 407–413, 2019, doi: 10.14569/ijacsa.2019.0100550.