A Client Selection Approach Based on Dynamic Clustering and Reputation in Federated Learning
Resumo
A previsão de demanda elétrica enfrenta desafios de confidencialidade e de heterogeneidade dos perfis de consumo. Este trabalho propõe uma estratégia de aprendizado federado com agrupamento (clustered federated learning) que combina uma topologia de clusters revisada em tempo de execução por fusão e criação com uma função de reputação que pondera a qualidade dos modelos locais e bônus de exploração de cliente e de cluster. Os resultados demonstram que os modelos-cluster alcançaram erro quadrático médio cerca de 0,146 ordens de grandeza menor que o de um modelo global único treinado com FedAvg, além de acompanhar melhor a curvas dos participantes de seu cluster.Referências
Ahmed, S., Shamim Kaiser, M., Chaki, S., and Shawkat Ali, A. B. M. (2025). Adaptive federated learning with reinforcement learning-based client selection for heterogeneous environments. IEEE Access, 13:131671–131695.
Ali, S. S., Ali, M., Bhatti, D. M. S., and Choi, B. J. (2025). Explainable clustered federated learning for solar energy forecasting. Energies, 18(9).
Azzalini, D., Flammini, B., Emanuele, C. A., Guadagno, A., Ragaini, E., and Amigoni, F. (2025). An empirical evaluation of deep autoencoders for anomaly detection in the electricity consumption of buildings. Energy and Buildings, 327:115069.
Bandoria, L. H., Silva, W. N., Dias, B. H., and Almeida, M. C. (2026). Self-supervised learning for advanced load profile segmentation in distribution networks. Sustainable Energy, Grids and Networks, 46:102166.
Casolaro, A., Capone, V., Iannuzzo, G., and Camastra, F. (2023). Deep learning for time series forecasting: Advances and open problems. Information, 14(11):598.
Dakheel, F. and Çevik, M. (2025). Optimizing smart grid load forecasting via a hybrid long short-term memory-xgboost framework: Enhancing accuracy, robustness, and energy management. Energies, 18(11).
Duan, M., Liu, D., Ji, X., Liu, R., Liang, L., Chen, X., and Tan, Y. (2021). Fedgroup: Efficient federated learning via decomposed similarity-based clustering. In 2021 IEEE Intl Conf on parallel & distributed processing with applications, big data & cloud computing, sustainable computing & communications, social computing & networking, pages 228–237.
Fu, L., Zhang, H., Gao, G., Zhang, M., and Liu, X. (2023). Client selection in federated learning: Principles, challenges, and opportunities. IEEE Internet of Things Journal, 10(24):21811–21819.
Gholizadeh, N. and Musilek, P. (2022). Federated learning with hyperparameter-based clustering for electrical load forecasting. Internet of Things, 17:100470.
Grataloup, A., Jonas, S., and Meyer, A. (2024). A review of federated learning in renewable energy applications: Potential, challenges, and future directions. Energy and AI, 17:100375.
He, M. (2024). Spatial-temporal prediction of electric vehicle charging load based on multi-time step semi-dynamic traffic equilibrium. In 2024 5th International Conference on Clean Energy and Electric Power Engineering (ICCEPE), pages 1243–1248. IEEE.
Humiston, C., Cetin, M., and Queiroz, A. R. d. (2026). Evaluating battery degradation models in rolling-horizon bess arbitrage optimization. Energies, 19(4):1056.
Mendes, N., Mendes, J., Mohammadi, J., and Moura, P. (2024). Federated learning framework for prediction of net energy demand in transactive energy communities. Sustainable Energy, Grids and Networks, 40:101522.
Moyeen, M., Kaur, K., Agarwal, A., Manzano, S. R., Zaman, M., and Goel, N. (2025). Fed-reputed: Reputation-aware client selection in hierarchical federated learning for consumer electronics. IEEE Transactions on Consumer Electronics.
Navidi, T., El Gamal, A., and Rajagopal, R. (2023). Coordinating distributed energy resources for reliability can significantly reduce future distribution grid upgrades and peak load. Joule, 7(8):1769–1792.
Ramos, P. V. B., Silva, W. N., Bandória, L. H., Dias, B. H., Villela, S. M., and Morais, H. (2026). Federated electricity demand forecasting: A client selection mechanism based on timevae latent representations. Applied Energy, 419:128064.
Ramos, P. V. B., Villela, S. M., Silva, W. N., and Dias, B. H. (2023). Residential energy consumption forecasting using deep learning models. Applied Energy, 350:121705.
Rashid, M. M., Xiang, Y., Uddin, M. P., Tang, J., Sood, K., and Gao, L. (2025). Trustworthy and fair federated learning via reputation-based consensus and adaptive incentives. Trans. Info. For. Sec., 20:2868–2882.
Silva, W. N., Ramos, P. V. B., Dias, B. H., Almeida, M. C., Tenório, L. H., Oliveira, L. W., and Villela, S. M. (2024). Forecasting demand in data-limited environments: a benchmarking study. In Congresso Brasileiro de Automática-CBA, volume 4.
Street, P. (2024). Pecan street dataport. Acessado em: 25 mar. 2024.
Thwe, M. M., Ştefanov, A., Rajkumar, V. S., and Palensky, P. (2025). Digital twins for power systems: Review of current practices, requirements, enabling technologies, data federation and challenges. IEEE Access.
Wang, Z., Wen, M., Xu, Y., Zhou, Y., Wang, J. H., and Zhang, L. (2023). Communication compression techniques in distributed deep learning: A survey. Journal of Systems Architecture, 142.
Wu, J., Ji, H., Yi, J., and Liu, L. (2025). Optimizing client selection in federated learning base on genetic algorithm. Cluster Computing, 28(6):400.
Ali, S. S., Ali, M., Bhatti, D. M. S., and Choi, B. J. (2025). Explainable clustered federated learning for solar energy forecasting. Energies, 18(9).
Azzalini, D., Flammini, B., Emanuele, C. A., Guadagno, A., Ragaini, E., and Amigoni, F. (2025). An empirical evaluation of deep autoencoders for anomaly detection in the electricity consumption of buildings. Energy and Buildings, 327:115069.
Bandoria, L. H., Silva, W. N., Dias, B. H., and Almeida, M. C. (2026). Self-supervised learning for advanced load profile segmentation in distribution networks. Sustainable Energy, Grids and Networks, 46:102166.
Casolaro, A., Capone, V., Iannuzzo, G., and Camastra, F. (2023). Deep learning for time series forecasting: Advances and open problems. Information, 14(11):598.
Dakheel, F. and Çevik, M. (2025). Optimizing smart grid load forecasting via a hybrid long short-term memory-xgboost framework: Enhancing accuracy, robustness, and energy management. Energies, 18(11).
Duan, M., Liu, D., Ji, X., Liu, R., Liang, L., Chen, X., and Tan, Y. (2021). Fedgroup: Efficient federated learning via decomposed similarity-based clustering. In 2021 IEEE Intl Conf on parallel & distributed processing with applications, big data & cloud computing, sustainable computing & communications, social computing & networking, pages 228–237.
Fu, L., Zhang, H., Gao, G., Zhang, M., and Liu, X. (2023). Client selection in federated learning: Principles, challenges, and opportunities. IEEE Internet of Things Journal, 10(24):21811–21819.
Gholizadeh, N. and Musilek, P. (2022). Federated learning with hyperparameter-based clustering for electrical load forecasting. Internet of Things, 17:100470.
Grataloup, A., Jonas, S., and Meyer, A. (2024). A review of federated learning in renewable energy applications: Potential, challenges, and future directions. Energy and AI, 17:100375.
He, M. (2024). Spatial-temporal prediction of electric vehicle charging load based on multi-time step semi-dynamic traffic equilibrium. In 2024 5th International Conference on Clean Energy and Electric Power Engineering (ICCEPE), pages 1243–1248. IEEE.
Humiston, C., Cetin, M., and Queiroz, A. R. d. (2026). Evaluating battery degradation models in rolling-horizon bess arbitrage optimization. Energies, 19(4):1056.
Mendes, N., Mendes, J., Mohammadi, J., and Moura, P. (2024). Federated learning framework for prediction of net energy demand in transactive energy communities. Sustainable Energy, Grids and Networks, 40:101522.
Moyeen, M., Kaur, K., Agarwal, A., Manzano, S. R., Zaman, M., and Goel, N. (2025). Fed-reputed: Reputation-aware client selection in hierarchical federated learning for consumer electronics. IEEE Transactions on Consumer Electronics.
Navidi, T., El Gamal, A., and Rajagopal, R. (2023). Coordinating distributed energy resources for reliability can significantly reduce future distribution grid upgrades and peak load. Joule, 7(8):1769–1792.
Ramos, P. V. B., Silva, W. N., Bandória, L. H., Dias, B. H., Villela, S. M., and Morais, H. (2026). Federated electricity demand forecasting: A client selection mechanism based on timevae latent representations. Applied Energy, 419:128064.
Ramos, P. V. B., Villela, S. M., Silva, W. N., and Dias, B. H. (2023). Residential energy consumption forecasting using deep learning models. Applied Energy, 350:121705.
Rashid, M. M., Xiang, Y., Uddin, M. P., Tang, J., Sood, K., and Gao, L. (2025). Trustworthy and fair federated learning via reputation-based consensus and adaptive incentives. Trans. Info. For. Sec., 20:2868–2882.
Silva, W. N., Ramos, P. V. B., Dias, B. H., Almeida, M. C., Tenório, L. H., Oliveira, L. W., and Villela, S. M. (2024). Forecasting demand in data-limited environments: a benchmarking study. In Congresso Brasileiro de Automática-CBA, volume 4.
Street, P. (2024). Pecan street dataport. Acessado em: 25 mar. 2024.
Thwe, M. M., Ştefanov, A., Rajkumar, V. S., and Palensky, P. (2025). Digital twins for power systems: Review of current practices, requirements, enabling technologies, data federation and challenges. IEEE Access.
Wang, Z., Wen, M., Xu, Y., Zhou, Y., Wang, J. H., and Zhang, L. (2023). Communication compression techniques in distributed deep learning: A survey. Journal of Systems Architecture, 142.
Wu, J., Ji, H., Yi, J., and Liu, L. (2025). Optimizing client selection in federated learning base on genetic algorithm. Cluster Computing, 28(6):400.
Publicado
19/10/2026
Como Citar
FONSECA, Gabriel A.; BRITO, Arthur A.; RAMOS, Paulo Vitor B.; DIAS, Bruno H.; VILLELA, Saulo M.; MORAIS, Hugo; PEREIRA, Lucas.
A Client Selection Approach Based on Dynamic Clustering and Reputation in Federated Learning. In: ENCONTRO NACIONAL DE INTELIGÊNCIA ARTIFICIAL E COMPUTACIONAL (ENIAC), 23. , 2026, Cuiabá/MT.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 3322-3333.
ISSN 2763-9061.
DOI: https://doi.org/10.5753/eniac.2026.31331.
