The week's pick
Reseach Article
Development of an Energy Management System in Buildings based on Edge Computing and Machine Learning
| International Journal of Computer Applications |
| Foundation of Computer Science (FCS), NY, USA |
| Volume 187 - Number 100 |
| Year of Publication: 2026 |
| Authors: Jorge Pedro, Wendley Souza da Silva, Geraldo Eufrazio Martins Júnior |
10.5120/ijca77f4b325cae6
|
Jorge Pedro, Wendley Souza da Silva, Geraldo Eufrazio Martins Júnior . Development of an Energy Management System in Buildings based on Edge Computing and Machine Learning. International Journal of Computer Applications. 187, 100 ( Apr 2026), 18-31. DOI=10.5120/ijca77f4b325cae6
Abstract
The exponential increase in demand for electricity, especially in urban centers, has put the sustainability and efficiency of energy supply systems in check [1]. This paper presents the development and experimental validation of an energy management and optimization system in Buildings Based on Edge Computing, using the ESP32-S3 microcontroller and embedded Machine Learning (ML) algorithms. The system consists of a central processing unit and distributed sensing and actuation modules, communicating via the ESP-NOW wireless protocol without the need for an external router. The voltage and current sensors, based on the IC ATM90E36 with resistive splitters and ZMCT103C current transformers, were validated in the laboratory with R² greater than 0.9998 and average errors of ±0.17% and ±0.30%, respectively — performance higher in current than reported by [2]. ESP-NOW communication at 20 meters with masonry obstacles achieved a 99.5% success rate and 5.8 ms average latency (250 ms interval). The Random Forest Regressor model, trained on 718 hourly samples of 30 days of continuous operation, achieved MAE of 70.71 W (22.7% reduction compared to baseline) and was integrated directly into the ESP32-S3 firmware with an inference time of 20 ms and no dependence on external connectivity.
References
- A. A. Sônego, R. Marcelino, and V. Gruber, "The Internet of Things applied to the concept of energy efficiency: a quantitative-qualitative analysis of the state of the art literature," AtoZ: new practices in information and knowledge, vol. 5, no. 2, p. 80, 2017, doi: 10.5380/atoz.v5i2.47860.
- N. E. Guevara, Y. H. Bolaños, J. P. Diago, and J. M. Segura, “HardwareX Development of a low-cost IoT system based on LoRaWAN for monitoring variables related to electrical energy consumption in low voltage networks,” HardwareX, vol. 12, p. e00330, 2022, doi: 10.1016/j.ohx.2022.e00330.
- FCCC, "Paris Agreement," Framework on Climate Change, vol. CP/2015/L. 2015.
- D. A. Verde Romero, E. Villalvazo Laureano, R. O. Jiménez Betancourt, and E. Navarro Álvarez, “An open source IoT edge-computing system for monitoring energy consumption in buildings,” Results Eng., vol. 21, no. February, 2024, doi: 10.1016/j.rineng.2024.101875.
- J. dos S. Costa and L. M. L. de Andrade Junior, "Energy efficiency applied to electricity consumption: A literature review study," Res. Soc. Dev., vol. 10, no. 4, p. e26210414085, 2021, doi: 10.33448/rsd-v10i4.14085.
- L. Da Xu, W. He, and S. Li, “Internet of things in industries: A survey,” IEEE Trans. Ind. Informatics, vol. 10, no. 4, pp. 2233–2243, 2014, doi: 10.1109/TII.2014.2300753.
- H. Hua, Y. Li, T. Wang, N. Dong, W. Li, and J. Cao, “Edge Computing with Artificial Intelligence: A Machine Learning Perspective,” ACM Comput. Surv., vol. 55, no. 9, pp. 1–35, Sep. 2023, doi: 10.1145/3555802.
- H. J. El-Khozondar et al., “A smart energy monitoring system using ESP32 microcontroller,” e-Prime - Adv. Electr. Eng. Electron. Energy, vol. 9, no. June, p. 100666, 2024, doi: 10.1016/j.prime.2024.100666.
- T. M. SERRANO, DESIGN AND DEVELOPMENT OF AN IOT SOCKET WITH ADVANCED ELECTRIC ENERGY MEASUREMENT AND MONITORING, vol. 2507, no. February. CAMPINAS, 2020.
- M. I. Joha, M. S. Nazim, and M. I. Zubair, “Smart Energy Metering, Control, and Protection System for Making a Smart Home Using IoT,” 2023 26th Int. Conf. Comput. Inf. Technol. ICCIT 2023, no. December, pp. 1–6, 2023, doi: 10.1109/ICCIT60459.2023.10441201.
- U. Ali, M. U. Ramzan, W. Ali, M. E. Rana, and A. Qayyum, “IoT-Driven Smart Energy Monitoring: Real-time Insights and AI-Based Unit Predictions,” 2023 IEEE 21st Student Conf. Res. Dev. SCOReD 2023, pp. 672–677, 2023, doi: 10.1109/SCOReD60679.2023.10563825.
- Francisco Borges Carreiro;, Júlia Ferreira Nascimento, and Guilherme Mateus Lopes Sousa, "Modeling and Development of an IoT and Digital Twin Solution for Residential Energy Consumption Measurement," Procedings do XXIV Congr. Bras. Automatic, pp. 1513–1520, 2022, doi: 10.20906/cba2022/3380.
- M. Khan, J. Seo, and D. Kim, “Towards energy efficient home automation: A deep learning approach,” Sensors (Switzerland), vol. 20, no. 24, pp. 1–18, 2020, doi: 10.3390/s20247187.
- D. Alfonso, V. Romero, and E. V. Laureano, "An open-source IoT edge computing system for monitoring energy consumption in buildings," vol. 21, 2024.
- K. E. N. Peffers, T. Tuunanen, and M. A. Rothenberger, “A Design Science Research Methodology for Information Systems Research,” vol. 24, no. 3, pp. 45–77, 2008, doi: 10.2753/MIS0742-1222240302.
- A. Dresch, A. Proença, J. Antonio, and V. Antunes, "Design Science Research : research method for production engineering," pp. 741–761, 2013.
- M. BUSCHMANN, Frank; MEUNIER, regine; ROHNERT, Hans; SOMMERLAND, Peter; STAL, SOFT WARE ARCHITECTURE. New York, 1996.
- Hik, "DIGITAL MULTIMETER," vol. 189, no. L, pp. 1–4, 2000.
- MICROCHIP TECHNOLOGY IN, “Atmel M90E36A.” San Jose, 2016.
- S. Kapoor and A. Narayanan, “Article Leakage and the reproducibility crisis in machine- learning-based science Leakage and the reproducibility crisis in machine-learning-based science,” Patterns, vol. 4, no. 9, p. 100804, 2023, doi: 10.1016/j.patter.2023.100804.
Index Terms
Keywords