MELISSA: An LLM-Powered Smart Home Energy Consumption Monitoring Framework

• Gabriel Arquelau Pimenta Rodrigues UnB http://orcid.org/0000-0002-4502-2153
• Matheus Noschang de Oliveira UnB https://orcid.org/0009-0001-0769-2826
• André Luiz Marques Serrano UnB https://orcid.org/0000-0001-5182-0496
• Geraldo Pereira Rocha Filho UnB / UESB https://orcid.org/0000-0001-6795-2768
• Guilherme Fay Vergara UnB https://orcid.org/0000-0002-4551-2240
• Letícia Rezende Mosquéra UnB https://orcid.org/0009-0004-7544-0059
• Vinícius Pereira Gonçalves UnB https://orcid.org/0000-0002-3771-2605

DOI: https://doi.org/10.5753/sbcup.2025.7610

Resumo

This work introduces MELISSA, a multi-agent system that uses Large Language Models (LLMs) to optimize household energy consumption by integrating historical analysis and meteorological inputs, acting as an intelligent Home Energy Management System (HEMS) for smart spaces. Through a twostage process that condenses data by approximately 99%, the system identifies consumption patterns and anomalies. The Gunning Fog Index indicates that the outputs are easily readable by the target audience, with a moderate Self-BLEU score. Thus, MELISSA offers an effective residential energy management solution, using LLMs to communicate with the end-users. Future enhancements include integrating energy generation data.

Referências

Badar, A. Q. and Anvari-Moghaddam, A. (2022). Smart home energy management system–a review. Advances in Building Energy Research, 16(1):118–143.

Condon, F., Martínez, J. M., Eltamaly, A. M., Kim, Y.-C., and Ahmed, M. A. (2022). Design and implementation of a cloud-iot-based home energy management system. Sensors, 23(1):176.

Giudici, M., Padalino, L., Paolino, G., Paratici, I., Pascu, A. I., and Garzotto, F. (2024). Designing home automation routines using an llm-based chatbot. Designs, 8(3):43.

Gunning, R. (1969). The fog index after twenty years. Journal of Business Communication, 6(2):3–13.

Hosseini, S., Kelouwani, S., Agbossou, K., Cardenas, A., and Henao, N. (2017). A semi-synthetic dataset development tool for household energy consumption analysis. In 2017 IEEE International Conference on Industrial Technology (ICIT), pages 564–569. IEEE.

King, E., Yu, H., Lee, S., and Julien, C. (2024). Sasha: Creative goal-oriented reasoning in smart homes with large language models. Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., 8(1).

Liu, X., Ding, Y., Tang, H., and Xiao, F. (2021). A data mining-based framework for the identification of daily electricity usage patterns and anomaly detection in building electricity consumption data. Energy and Buildings, 231:110601.

Lo, L. S. (2023). The clear path: A framework for enhancing information literacy through prompt engineering. The Journal of Academic Librarianship, 49(4):102720.

Michelon, F., Zhou, Y., and Morstyn, T. (2025). Large language model interface for home energy management systems. arXiv preprint arXiv:2501.07919.

Pullinger, M., Kilgour, J., Goddard, N., Berliner, N., Webb, L., Dzikovska, M., Lovell, H., Mann, J., Sutton, C., Webb, J., et al. (2021). The ideal household energy dataset, electricity, gas, contextual sensor data and survey data for 255 uk homes. Scientific Data, 8(1):146.

Serrano, A. L. M., Rodrigues, G. A. P., Martins, P. H. d. S., Saiki, G. M., Filho, G. P. R., Gonçalves, V. P., and Albuquerque, R. d. O. (2024). Statistical comparison of time series models for forecasting brazilian monthly energy demand using economic, industrial, and climatic exogenous variables. Applied Sciences, 14(13):5846.

Vogelsang, A. (2024). From specifications to prompts: On the future of generative large language models in requirements engineering. IEEE Software, 41(5):9–13.

Wang, T., Zhao, Q., Gao, W., and He, X. (2024). Research on energy consumption in household sector: a comprehensive review based on bibliometric analysis. Frontiers in Energy Research, 11.

Yonekura, H., Tanaka, F., Mizumoto, T., and Yamaguchi, H. (2024). Generating human daily activities with llm for smart home simulator agents. In 2024 International Conference on Intelligent Environments (IE), pages 93–96.