2692-4048ISSN (Online)
2770-0070ISSN (Print)
Volume 21 , Issue 2 , PP: 283-305, 2026 | Cite this article as | XML | Html | PDF | Full Length Article
El-Sayed M. El-kenawy 1 *
Doi: https://doi.org/10.54216/FPA.210218
This study addresses the challenge of smart-home energy forecasting across multiple appliances under varying temperature and seasonal regimes, aiming to improve demand planning and household energy efficiency. The analysis leverages a 100,000-row dataset from Kaggle, encompassing appliance type, time of consumption, outdoor temperature, season, and household size. The study benchmarks several recurrent neural network models, including Long Short-Term Memory (LSTM), Bidirectional LSTM (BiLSTM), Gated Recurrent Unit (GRU), and Bidirectional RNN (BiRNN), as well as a feedforward Artificial Neural Network (ANN). A novel enhancement, the Evolutionary Attention-based LSTM (EALSTM), is introduced, and its hyperparameters are optimized using the Greylag Goose Optimization (GGO) algorithm. The performance of GGO-optimized EALSTM is compared to other metaheuristics, such as Differential Evolution (DE), Genetic Algorithm (GA), Quantum-Inspired Optimization (QIO), JAYA, Bat Algorithm (BA), and Stochastic Fractal Search (SFS). The results indicate that GGO-optimized EALSTM outperforms all other models, achieving superior accuracy across multiple metrics, including MSE, RMSE, MAE, r, R2 , RRMSE, NSE, and WI. Key contributions of the paper include (i) the establishment of an appliance- and season-aware forecasting benchmark, (ii) a comprehensive optimizer comparison for EALSTM using GGO, and (iii) the provision of actionable visual analytics to enhance the understanding of energy demand patterns and model errors.
Smart-home energy forecasting , Evolutionary Attention-based LSTM , Greylag Goose Optimization , Appliance-level prediction , Metaheuristic optimization
[1] S. Miller et al., “Energy consumption and demand response: A smart grid perspective,” Energy Reports, vol. 2, pp. 1–10, 2016.
[2] M. Ku et al., “Residential energy consumption modeling and forecasting,” Renewable and Sustainable Energy Reviews, vol. 81, pp. 1–12, 2018.
[3] J. Anderson et al., “Impact of smart grids on residential energy consumption,” IEEE Transactions on Smart Grid, vol. 11, no. 1, pp. 123–132, 2020.
[4] A. Martin et al., “Forecasting household energy consumption,” Energy Economics, vol. 64, pp. 1–10, 2017.
[5] Y. Liu et al., “A smart home energy consumption forecasting method based on machine learning,” Renewable Energy, vol. 139, pp. 1–9, 2019.
[6] H. Wang et al., “Household-level energy forecasting with behavioral insights,” Applied Energy, vol. 276, pp. 115–126, 2020.
[7] Z. Liu et al., “Short-term appliance-level load forecasting using hybrid models,” Energy, vol. 223, pp. 120–133, 2021.
[8] F. Gao et al., “Residential energy consumption forecasting using machine learning techniques,” Energy, vol. 141, pp. 1–10, 2017.
[9] A. Joshi et al., “Energy consumption forecasting in smart homes using machine learning,” Energy Reports, vol. 5, pp. 1–10, 2019.
[10] M. Marzband et al., “Forecasting energy demand in smart homes with machine learning,” Applied Energy, vol. 255, pp. 113–120, 2019.
[11] V. Ferraro et al., “Machine learning for household energy forecasting: A case study,” Energy and Buildings, vol. 209, pp. 109–118, 2020.
[12] Y. Li et al., “Time-series modeling for residential energy consumption forecasting,” Energy Reports, vol. 3, pp. 1–10, 2017.
[13] L. Dan et al., “Hybrid forecasting models for smart home energy use,” Energy Reports, vol. 7, pp. 1–10, 2021.
[14] D. Petrovic et al., “Modeling seasonal variability in household energy forecasting,” Energy Reports, vol. 6, pp. 1–10, 2020.
[15] R. Gonzalez et al., “Multi-scale energy forecasting under distributional drift,” Energy Reports, vol. 8, pp. 1–10, 2022.
[16] J. Liang et al., “Detecting abrupt transitions in residential load forecasting,” Energy Reports, vol. 7, pp. 1–10, 2021.
[17] A. Wicks et al., “Modeling peak household energy usage periods,” Energy Reports, vol. 8, pp. 1–10, 2022.
[18] P. Rao et al., “Noise-robust models for smart-home energy forecasting,” Energy Reports, vol. 4, pp. 1–10, 2018.
[19] Y. Huang et al., “Selective attention for temporal interactions in appliance forecasting,” Energy Reports, vol. 6, pp. 1–10, 2020.
[20] L. Xu et al., “Deep sequence models for energy forecasting,” Energy Reports, vol. 7, pp. 1–10, 2021.
[21] F. Gao et al., “Deep sequence models for energy forecasting,” Energy Reports, vol. 5, pp. 1–10, 2019.
[22] K. Chu et al., “Limitations of recurrent neural networks for energy forecasting,” Neurocomputing, vol. 361, pp. 94–103, 2019.
[23] M. Al et al., “Regularization in deep sequence forecasting of residential energy,” IEEE Access, vol. 8, pp. 22190–22201, 2020.
[24] J. Zhou et al., “Attention-based recurrent models for residential load forecasting,” IEEE Transactions on Smart Grid, vol. 11, no. 5, pp. 4207–4216, 2020.
[25] L. Cao et al., “Adaptive attention mechanisms for household energy prediction,” Energy Reports, vol. 7, pp. 243–252, 2021.
[26] Q. Zhao et al., “Lightweight attention-enabled forecasting of appliance loads,” Sustainable Energy, Grids and Networks, vol. 26, pp. 100–111, 2021.
[27] R. Jones et al., “Recurrent neural attention for residential energy consumption,” Applied Energy, vol. 269, pp. 114–124, 2020.
[28] A. Muhammad et al., “Computational challenges in hyperparameter optimization for smart-home forecasting,” Journal of Renewable and Sustainable Energy, vol. 13, no. 3, p. 033701, 2021.
[29] S. Lim et al., “Scalable optimization strategies for energy demand forecasting,” Renewable Energy, vol. 189, pp. 332–344, 2022.
[30] G. Sun et al., “Greylag goose optimization: A novel evolutionary algorithm,” Expert Systems with Applications, vol. 168, pp. 114–128, 2021.
[31] C. Chou et al., “Cross-feature dependency modeling in smart-home forecasting,” IEEE Transactions on Industrial Informatics, vol. 15, no. 6, pp. 3312–3320, 2019.
[32] M. Bux et al., “A public dataset for appliance-level residential energy consumption,” Data in Brief, vol. 25, pp. 104–114, 2019.
[33] J. Kim et al., “Smart home energy datasets for load forecasting,” Scientific Data, vol. 7, no. 1, p. 23, 2020.
[34] Y. Tang et al., “Wrapper-based hyperparameter optimization for neural energy forecasters,” Applied Energy, vol. 283, pp. 116–127, 2021.
[35] P. Johnson et al., “Comprehensive evaluation metrics for energy forecasting,” Energy Reports, vol. 6, pp. 53–62, 2020.
[36] H. Zhu et al., “Skill measures for forecasting in energy systems,” Renewable and Sustainable Energy Reviews, vol. 135, pp. 110–118, 2021.
[37] K. Cho et al., “Recurrent neural network baselines for smart-home load forecasting,” Neural Processing Letters, vol. 50, no. 3, pp. 2025–2036, 2019.
[38] B. Matthiesen et al., “Comparison of recurrent neural models for energy forecasting,” Energy, vol. 231, pp. 120–137, 2021.
[39] D. Yu et al., “Baseline deep learning approaches for appliance-level forecasting,” Applied Energy, vol. 275, pp. 115–126, 2020.
[40] X. Bi et al., “Evaluation of bidirectional recurrent models for energy forecasting,” Energy Reports, vol. 8, pp. 75–85, 2022.
[41] Y. Dai et al., “Optimization algorithms for load forecasting in smart homes,” Energy and Buildings, vol. 241, pp. 110–120, 2021.
[42] X. Nie et al., “Comparative study of optimization algorithms for deep forecasting models,” Expert Systems with Applications, vol. 160, pp. 113–121, 2020.
[43] L. Zhang et al., “Evolutionary tuning for attention-based recurrent networks,” Neural Networks, vol. 127, pp. 74–85, 2020.
[44] X. Li et al., “Attention-augmented sequence models for load forecasting,” Applied Energy, vol. 285, pp. 116–127, 2021.
[45] J. Liu et al., “Lightweight neural architectures for real-time residential energy forecasting,” Energy, vol. 209, pp. 118–128, 2020.
[46] Y. Xie et al., “Efficient deep learning models for smart-home energy prediction,” Energy Reports, vol. 7, pp. 243–253, 2021.
[47] R. Gao et al., “Advances in attention-based forecasting for smart grids,” IEEE Access, vol. 9, pp. 12839– 12849, 2021.
[48] M. A. Soliman, A. Abdelmgeed, and A. M. Zaki, “Comparative analysis of machine learning techniques for energy consumption prediction in smart homes,” Energy Reports, vol. (8), no. 2, pp. 120–130, 2023. Publisher: Elsevier.
[49] B. Yildiz, J. I. Bilbao, J. Dore, and A. Sproul, “Household electricity load forecasting using historical smart meter data with clustering and classification techniques,” in 2018 IEEE Innovative Smart Grid Technologies - Asia (ISGT Asia), pp. 873–879, May 2018.
[50] S. Mahjoub, S. Labdai, L. Chrifi-Alaoui, B. Marhic, and L. Delahoche, “Short-Term Occupancy Forecasting for a Smart Home Using Optimized Weight Updates Based on GA and PSO Algorithms for an LSTM Network,” Energies, vol. 16, p. 1641, Jan. 2023. Publisher: Multidisciplinary Digital Publishing Institute.
[51] H. Youssef, S. Kamel, M. H. Hassan, and L. Nasrat, “Optimizing energy consumption patterns of smart home using a developed elite evolutionary strategy artificial ecosystem optimization algorithm,” Energy, vol. 278, p. 127793, Sept. 2023.
[52] S. Balavignesh, C. Kumar, R. Sripriya, and T. Senjyu, “An enhanced coati optimization algorithm for optimizing energy management in smart grids for home appliances,” Energy Reports, vol. 11, pp. 3695– 3720, June 2024.
[53] J. Byun, I. Hong, B. Kang, and S. Park, “A smart energy distribution and management system for renewable energy distribution and context-aware services based on user patterns and load forecasting,” IEEE Transactions on Consumer Electronics, vol. 57, pp. 436–444, May 2011.
