Multi-step hourly electrical load forecasting for a single household: A hybrid deep learning approach based on CNN, BiGRU, and Attention Mechanism

Year 2026, Volume: 11 Issue: 1, 255 - 274, 17.03.2026

Havva Nur Patat , Mustafa Yasin Erten , Hüseyin Aydilek

https://doi.org/10.58559/ijes.1826202

https://izlik.org/JA96YX78EW

Abstract

Accurate electrical load forecasting is crucial for the efficient management of modern power grids. However, the highly variable consumption patterns of single households present a significant challenge, especially for multi-step forecasting, which holds high practical value. This study addresses this challenge by proposing a novel hybrid deep learning model that combines Convolutional Neural Networks, a Bidirectional Gated Recurrent Unit, and an Attention mechanism. The proposed model was trained and tested to forecast the next hour in 6 steps, based on 24 hours of 10-minute interval data. The experimental results demonstrate that the proposed architecture, despite the complexity and granularity of the problem, achieves remarkably low error rates, with a Mean Absolute Error  of 0.0393, a Mean Squared Error of 0.0044, and a Root Mean Squared Error of 0.0666. This performance is competitive with results obtained by many state-of-the-art models on single-step forecasting tasks. Consequently, this study makes a unique contribution to the field by presenting an effective and robust solution to the multi-step hourly forecasting problem, a topic less explored in the literature.

Keywords

Electrical load forecasting , Multi-step forecasting , Convolutional neural networks , Bidirectional gated recurrent unit

References

• [1] IEA. World Energy Outlook 2022. International Energy Agency, Paris, France, 2022.
• [2] Mounir N, Ouadi H, Jrhilifa I. Short-term electric load forecasting using an EMD-BI-LSTM approach for smart grid energy management system. Energy and Buildings 2023; 288: 113022.
• [3] Dudek G. A Comprehensive Study of Random Forest for Short-Term Load Forecasting. Energies 2022; 15(20): 7547.
• [4] Alquthami T, Zulfiqar M, Kamran M, Milyani AH, Rasheed MB. A Performance Comparison of Machine Learning Algorithms for Load Forecasting in Smart Grid. IEEE Access 2022; 10: 48419-48433.
• [5] Ahmad N, Ghadi Y, Adnan M, Ali M. Load Forecasting Techniques for Power System: Research Challenges and Survey. IEEE Access 2022; 10: 71054-71090.
• [6] Alhussein M, Aurangzeb K, Haider SI. Hybrid CNN-LSTM Model for Short-Term Individual Household Load Forecasting. IEEE Access 2020; 8: 180544-180557.
• [7] Khan N, Haq IU, Khan SU, Rho S, Lee MY, Baik SW. DB-Net: A novel dilated CNN based multi-step forecasting model for power consumption in integrated local energy systems. International Journal of Electrical Power and Energy Systems 2021; 133: 107023.
• [8] Gonçalves R, Ribeiro VM, Pereira FL. Variable Split Convolutional Attention: A novel Deep Learning model applied to the household electric power consumption. Energy 2023; 274: 127321.
• [9] Khan SU, Iqbal E, Khan N, Zweiri Y, Abdulrahman Y. Towards net zero energy building: AI-based framework for power consumption and generation prediction. Energy & Buildings 2025; 331: 115311.
• [10] Ullah K, Akram W, Hassan A, et al. Hybrid CNN-BiGRU model with attention mechanism for enhanced short-term load forecasting. Energy Reports 2025; 14: 2570-2577.
• [11] Hua Q, Fan Z, Mu W, et al. A Short-Term Power Load Forecasting Method Using CNN-GRU with an Attention Mechanism. Energies 2025; 18: 106.
• [12] Niu D, Yu M, Sun L, Gao T, Wang K. Short-term multi-energy load forecasting for integrated energy systems based on CNN-BiGRU optimized by attention mechanism. Applied Energy 2022; 313: 118801.
• [13] Kim TY, Cho SB. Predicting residential energy consumption using CNN-LSTM neural networks. Energy 2019; 182: 72–81.
• [14] Sajjad M, Khan ZA, Ullah A, Hussain T, Ullah W, Lee MY, Baik SW. A novel CNN-GRU-based hybrid approach for short-term residential load forecasting. IEEE Access 2020; 8: 143759–143768.
• [15] Le T, Vo MT, Vo B, Hwang E, Rho S, Baik SW. Improving electric energy consumption prediction using CNN and Bi-LSTM. Applied Sciences 2019; 9(20): 4237.
• [16] Ullah FUM, Ullah A, Haq IU, Rho S, Baik SW. Short-term prediction of residential power energy consumption via CNN and multi-layer bi-directional LSTM networks. IEEE Access 2020; 8: 123369–123380.
• [17] Amirkhanloo D, Chitsaz H, Mohajerzadeh A. Hybrid Deep Learning (CNN-BiLSTM-LSTM) Model for Prediction of Short Power Consumption. In: 2024 8th International Conference on Smart Cities, Internet of Things and Applications (SCIoT), Mashhad, Iran, 2024. pp. 162–167.
• [18] Haq IU, Ullah A, Khan SU, Khan N, Lee MY, Rho S, Baik SW. Sequential learning-based energy consumption prediction model for residential and commercial sectors. Mathematics 2021; 9(6): 605.
• [19] Rajabi R, Estebsari A. Deep learning based forecasting of individual residential loads using recurrence plots. In: 2019 IEEE Milan PowerTech, Milan, Italy, 2019. pp. 1–5.
• [20] Bai Z. Residential electricity prediction based on GA-LSTM modeling. Energy Reports 2024; 11: 6223–6232.
• [21] Khan ZA, Hussain T, Ullah A, Rho S, Lee M, Baik SW. Towards efficient electricity forecasting in residential and commercial buildings: A novel hybrid CNN with a LSTM-AE based framework. Sensors 2020; 20(5): 1399.
• [22] Kim JY, Cho SB. Electric energy consumption prediction by deep learning with state explainable autoencoder. Energies 2019; 12(4): 739.
• [23] Bezzar NEH, Laimeche L, Meraoumia A. Time Series Analysis of Household Electric Consumption with XGBoost Model. In: 2022 4th International Conference on Pattern Analysis and Intelligent Systems (PAIS), Oum El Bouaghi, Algeria, 2022. pp. 1–6.
• [24] Tie R, Li M, Zhou C, Ding N. Research on the application of an improved Autoformer model integrating CNN-attention-BiGRU in short-term power load forecasting. Evolving Systems 2025; 16, 98.1-17.
• [25] Ding N, Benoit C, Foggia G, Bésanger Y, Wurtz F. Neural network-based model design for short-term load forecast in distribution systems. IEEE Transactions on Power Systems 2016; 31(1): 72-81.