Heuristic based federated learning with adaptive hyperparameter tuning for households energy prediction

Abstract Federated Learning is transforming electrical load forecasting by enabling Artificial Intelligence (AI) models to be trained directly on household edge devices. However, the prediction accuracy of federated learning models tends to diminish when dealing with non-IID data highlighting the ne...

Full description

Saved in:
Bibliographic Details
Main Authors: Liana Toderean, Mihai Daian, Tudor Cioara, Ionut Anghel, Vasilis Michalakopoulos, Efstathios Sarantinopoulos, Elissaios Sarmas
Format: Article
Language:English
Published: Nature Portfolio 2025-04-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-96443-3
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850183849353412608
author Liana Toderean
Mihai Daian
Tudor Cioara
Ionut Anghel
Vasilis Michalakopoulos
Efstathios Sarantinopoulos
Elissaios Sarmas
author_facet Liana Toderean
Mihai Daian
Tudor Cioara
Ionut Anghel
Vasilis Michalakopoulos
Efstathios Sarantinopoulos
Elissaios Sarmas
author_sort Liana Toderean
collection DOAJ
description Abstract Federated Learning is transforming electrical load forecasting by enabling Artificial Intelligence (AI) models to be trained directly on household edge devices. However, the prediction accuracy of federated learning models tends to diminish when dealing with non-IID data highlighting the need for adaptive hyperparameter optimization strategies to improve performance. In this paper, we propose a novel hierarchical federated learning solution for efficient model aggregation and hyperparameter tuning, specifically tailored to household energy prediction. The households with similar energy profiles are clustered at the edge, linked, and aggregated at the fog level, to enable effective and adaptive hyperparameter tuning. The federated model aggregation is optimized using hierarchical simulated annealing optimization to prioritize updates from the better-performing models. A genetic algorithm-based hyperparameter optimization method reduces the computational load on edge nodes by efficiently exploring different configurations and using only the most promising ones for edge nodes’ cross-validation. The evaluation results demonstrate a significant improvement in average prediction accuracy and better capturing of energy patterns compared to the federated averaging approach. The impact on network traffic among nodes across different layers is kept below 30 KB. Additionally, hyperparameter tuning reduces the size of model updates and the number of communication rounds by 30%, which is particularly beneficial when network resources are limited.
format Article
id doaj-art-5cf108cf039e44a4b0983257d7907ce9
institution OA Journals
issn 2045-2322
language English
publishDate 2025-04-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-5cf108cf039e44a4b0983257d7907ce92025-08-20T02:17:13ZengNature PortfolioScientific Reports2045-23222025-04-0115111710.1038/s41598-025-96443-3Heuristic based federated learning with adaptive hyperparameter tuning for households energy predictionLiana Toderean0Mihai Daian1Tudor Cioara2Ionut Anghel3Vasilis Michalakopoulos4Efstathios Sarantinopoulos5Elissaios Sarmas6Distributed Systems Research Laboratory, Computer Science Department, Technical University of Cluj-NapocaDistributed Systems Research Laboratory, Computer Science Department, Technical University of Cluj-NapocaDistributed Systems Research Laboratory, Computer Science Department, Technical University of Cluj-NapocaDistributed Systems Research Laboratory, Computer Science Department, Technical University of Cluj-NapocaDecision Support Systems Laboratory, School of Electrical & Computer Engineering, National Technical University of AthensDecision Support Systems Laboratory, School of Electrical & Computer Engineering, National Technical University of AthensDecision Support Systems Laboratory, School of Electrical & Computer Engineering, National Technical University of AthensAbstract Federated Learning is transforming electrical load forecasting by enabling Artificial Intelligence (AI) models to be trained directly on household edge devices. However, the prediction accuracy of federated learning models tends to diminish when dealing with non-IID data highlighting the need for adaptive hyperparameter optimization strategies to improve performance. In this paper, we propose a novel hierarchical federated learning solution for efficient model aggregation and hyperparameter tuning, specifically tailored to household energy prediction. The households with similar energy profiles are clustered at the edge, linked, and aggregated at the fog level, to enable effective and adaptive hyperparameter tuning. The federated model aggregation is optimized using hierarchical simulated annealing optimization to prioritize updates from the better-performing models. A genetic algorithm-based hyperparameter optimization method reduces the computational load on edge nodes by efficiently exploring different configurations and using only the most promising ones for edge nodes’ cross-validation. The evaluation results demonstrate a significant improvement in average prediction accuracy and better capturing of energy patterns compared to the federated averaging approach. The impact on network traffic among nodes across different layers is kept below 30 KB. Additionally, hyperparameter tuning reduces the size of model updates and the number of communication rounds by 30%, which is particularly beneficial when network resources are limited.https://doi.org/10.1038/s41598-025-96443-3Federated learningEnergy predictionHyperparameters optimizationSimulated Annealing-based federated aggregationGenetic algorithm
spellingShingle Liana Toderean
Mihai Daian
Tudor Cioara
Ionut Anghel
Vasilis Michalakopoulos
Efstathios Sarantinopoulos
Elissaios Sarmas
Heuristic based federated learning with adaptive hyperparameter tuning for households energy prediction
Scientific Reports
Federated learning
Energy prediction
Hyperparameters optimization
Simulated Annealing-based federated aggregation
Genetic algorithm
title Heuristic based federated learning with adaptive hyperparameter tuning for households energy prediction
title_full Heuristic based federated learning with adaptive hyperparameter tuning for households energy prediction
title_fullStr Heuristic based federated learning with adaptive hyperparameter tuning for households energy prediction
title_full_unstemmed Heuristic based federated learning with adaptive hyperparameter tuning for households energy prediction
title_short Heuristic based federated learning with adaptive hyperparameter tuning for households energy prediction
title_sort heuristic based federated learning with adaptive hyperparameter tuning for households energy prediction
topic Federated learning
Energy prediction
Hyperparameters optimization
Simulated Annealing-based federated aggregation
Genetic algorithm
url https://doi.org/10.1038/s41598-025-96443-3
work_keys_str_mv AT lianatoderean heuristicbasedfederatedlearningwithadaptivehyperparametertuningforhouseholdsenergyprediction
AT mihaidaian heuristicbasedfederatedlearningwithadaptivehyperparametertuningforhouseholdsenergyprediction
AT tudorcioara heuristicbasedfederatedlearningwithadaptivehyperparametertuningforhouseholdsenergyprediction
AT ionutanghel heuristicbasedfederatedlearningwithadaptivehyperparametertuningforhouseholdsenergyprediction
AT vasilismichalakopoulos heuristicbasedfederatedlearningwithadaptivehyperparametertuningforhouseholdsenergyprediction
AT efstathiossarantinopoulos heuristicbasedfederatedlearningwithadaptivehyperparametertuningforhouseholdsenergyprediction
AT elissaiossarmas heuristicbasedfederatedlearningwithadaptivehyperparametertuningforhouseholdsenergyprediction