306-25

Prediction of renewable energy consumption in homes: Comparison of artificial neural network models

Doroteya Dimitrova-Angelova(1), Diego Carmona Fernández(2), Manuel Calderón Godoy(2), Miguel Ángel Jaramillo Morán(2), Juan Félix González González(1)

1. Department of Applied Physic

2. Department of Electrical Engineering, Electronics and Automation

Industrial Engineering School, University of Extremadura. Avenue Elvas s/n, 06006 Badajoz (Spain)

2025-07-25

This study analyses the prediction of renewable energy consumption in a single-family house located in Extremadura, Spain, using three neural network models: MLP, LSTM and GRU. The house has a 4.08 kWp photovoltaic installation and data recorded every five minutes during the year 2024, which allowed daily and seasonal patterns to be identified. The MLP stood out for its accuracy and computational efficiency, achieving an R2 greater than 0.99. Both the LSTM and its simplified version GRU, originally designed to capture temporal dependencies, had limited performance because the current data do not depend on the previous data. The results obtained are extrapolated to other contexts, such as commercial buildings or industrial facilities, thanks to the ability of the models to integrate universal contextual and temporal variables. The results suggest that future integrations with digital twin technology could further optimize real-time energy management. This work establishes a solid foundation for advancing the application of artificial intelligence in sustainable and adaptive energy planning for different scenarios.

Key words: Neural networks, energy consumption, renewable energies, energy efficiency.

[1] D. Dimitrova Angelova, D. Carmona Fernández, M. Calderón Godoy, J. Antonio, Á. Moreno, and J. Félix González González, “A Review on Digital Twins and Its Application in the Modeling of Photovoltaic Installations,” Energies (Basel), vol. 17, no. 5, 2024, doi: https://doi.org/10.3390/en17051227.

[2] N. U. Aningo, D. Glew, H. Tawfik, A. Hardy, and R. Wyatt-Millington, “Towards A Microgrid Based Residential Home Energy Management Using Genetic Algorithm,” in Proceedings - International Conference on Developments in eSystems Engineering, DeSE, Institute of Electrical and Electronics Engineers Inc., Dec. 2020, pp. 15–20. doi: 10.1109/DeSE51703.2020.9450730.

[3] S. Li, Y. Wang, Y. Zheng, J. Geng, and J. Zhu, “Research on Energy Saving and Environmental Protection Management Evaluation of Listed Companies in Energy Industry Based on Portfolio Weight Cloud Model,” Energies (Basel), vol. 15, no. 12, Jun. 2022, doi: 10.3390/en15124311.

[4] V. K. Saini, R. Singh, D. K. Mahto, R. Kumar, and A. Mathur, “Learning Approach for Energy Consumption Forecasting in Residential Microgrid,” in 2022 IEEE Kansas Power and Energy Conference, KPEC 2022, Institute of Electrical and Electronics Engineers Inc., 2022. doi: 10.1109/KPEC54747.2022.9814744.

[5] E. Uzun, M. S. Karabinaoğlu, and M. Ayaz, “LSTM-based estimation of energy consumption in energy-intensive facilities,” Engineering Research Express, vol. 7, no. 1, p. 015420, Mar. 2025, doi: 10.1088/2631-8695/adb378.

[6] S. Hadri, M. Najib, M. Bakhouya, Y. Fakhri, and M. El Arroussi, “Performance evaluation of forecasting strategies for electricity consumption in buildings,” Energies (Basel), vol. 14, no. 18, Sep. 2021, doi: 10.3390/en14185831.

[7] H. Çimen, N. Çetinkaya, J. C. Vasquez, and J. M. Guerrero, “A Microgrid Energy Management System Based on Non-Intrusive Load Monitoring via Multitask Learning,” IEEE Trans Smart Grid, vol. 12, no. 2, pp. 977–987, Mar. 2021, doi: 10.1109/TSG.2020.3027491.

[8] F. Wang et al., “Joint Energy Disaggregation of Behind-the-Meter PV and Battery Storage: A Contextually Supervised Source Separation Approach,” IEEE Trans Ind Appl, vol. 58, Mar. 2022, doi: 10.1109/tia.2022.3144244.

[9] W. Fan, Q. Liu, A. Ahmadpour, and S. Gholami Farkoush, “Multi-objective non-intrusive load disaggregation based on appliances characteristics in smart homes,” Energy Reports, vol. 7, pp. 4445–4459, Nov. 2021, doi: 10.1016/j.egyr.2021.07.033.

[10] T. Knayer and N. Kryvinska, “The influence of energy management systems on the progress of efficient energy use in cross-cutting technologies in companies,” Energy Effic, vol. 16, no. 3, Mar. 2023, doi: 10.1007/s12053-023-10086-9.

[11] Cristina Teba, “ENERGY 4.0: How Digital Evolution is Shaping the Future of Electricity,” Apr. 17, 2023. Accessed: Jul. 03, 2024. [Online]. Available: https://www.dexma.com/blog-en/energy-4-0-how-digital-revolution-is-shaping-the-future-of electricity/#:~:text=Overall%2C%20the%20electricity%20digital%20revolution,4.0%20in%20the%20electricity%20sector.

[12] Doroteya Dimitrova-Angelova, Diego Carmona-Fernández, Oscar Churrio, Beatriz Ledesma, Silvia Román-Suero, and Juan Félix González-González, “Digital twin as an aid tool for development cooperation,” Nov. 2024.

[13] M. Ehteram, A. N. Ahmed, P. Kumar, M. Sherif, and A. El-Shafie, “Predicting freshwater production and energy consumption in a seawater greenhouse based on ensemble frameworks using optimized multi-layer perceptron,” Energy Reports, vol. 7, pp. 6308–6326, Nov. 2021, doi: 10.1016/j.egyr.2021.09.079.

[14] H. Lee, D. Kim, and J. H. Gu, “Prediction of Food Factory Energy Consumption Using MLP and SVR Algorithms,” Energies (Basel), vol. 16, no. 3, Feb. 2023, doi: 10.3390/en16031550.

[15] S. Yadav et al., “State of the art in energy consumption using deep learning models,” AIP Adv, vol. 14, no. 6, Jun. 2024, doi: 10.1063/5.0213366.

[16] I. Amalou, N. Mouhni, and A. Abdali, “Multivariate time series prediction by RNN architectures for energy consumption forecasting,” Energy Reports, vol. 8, pp. 1084–1091, Nov. 2022, doi: 10.1016/j.egyr.2022.07.139.

[17] G. Cybenkot, “Mathematics of Control, Signals, and Systems Approximation by Superpositions of a Sigmoidal Function*,” 1989.

[18] S. Hochreiter and J. Schmidhuber, “Long Short-Term Memory,” Neural Comput, vol. 9, no. 8, pp. 1735–1780, Nov. 1997, doi: 10.1162/neco.1997.9.8.1735.

[19] scikit-learn developers, “API Reference. StandardScaler.” Accessed: Apr. 05, 2025. [Online]. Available: https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html

[20] Y. Sheng, H. Arbabi, W. O. C. Ward, M. A. Álvarez, and M. Mayfield, “City-scale residential energy consumption prediction with a multimodal approach,” Sci Rep, vol. 15, no. 1, p. 5313, Dec. 2025, doi: 10.1038/s41598-025-88603-2.

[21] International Centre for Numerical Methods in Engineering, “CIMNE Develops Advanced Energy Demand Prediction Model for Catalonia,” International Centre for Numerical Methods in Engineering. Accessed: Apr. 08, 2025. [Online]. Available: https://www.cimne.com/vnews/2/12297/cimne-develops-advanced-energy-demand-prediction-model-for-catalonia

[22] J. S. Chou, N. Q. Nguyen, and K. Srinivasan, “Forecasting Regional Energy Consumption via Jellyfish Search-Optimized Convolutional-Based Deep Learning,” Int J Energy Res, vol. 2023, 2023, doi: 10.1155/2023/3056688.