A Combined Model Based on ISSA-CNN-BiGRU-MH-Attention and Its Application in Power Load Forecasting

Yixiang Ding

doi:10.54097/47xh9k21

Authors

Yixiang Ding

DOI:

https://doi.org/10.54097/47xh9k21

Keywords:

Electricity Load Short-term Forecasting, BiGRU, MH-Attention, Sparrow Search Algorithm.

Abstract

Existing prediction algorithms face challenges in terms of accuracy and training speed, which hinders high-efficiency, high-accuracy power load forecasting. This paper proposes a combined prediction model to address these issues. To overcome the Bidirectional Gated Recurrent Unit (BiGRU)'s limitations in capturing long-term dependencies and handling complex time-series data, a Convolutional Neural Network (CNN) module is introduced to extract local features and enhance the model's feature representation. Additionally, a Multi-Head Attention (MH-Attention) module is incorporated to dynamically assign weights to different time steps, improving adaptivity and focus on key features. For hyperparameter optimization, an Improved Sparrow Search Algorithm (ISSA) is proposed, which addresses traditional SSA’s tendency to fall into local optima and slow convergence by incorporating an adaptive update mechanism and hybrid heuristic strategy. The model is validated using a power plant dataset from Quanzhou, with results showing excellent forecasting ability: R2=0.9955, RMSE=56.9596, and MAE=34.6080. Comparison with other models demonstrates improved performance, with R2 increasing by 0.3%-0.65%, RMSE decreasing by 6.38%-35.96%, and MAE reducing by 26.54%-49.52%. These results confirm the model’s effectiveness and superiority.

Downloads

Download data is not yet available.

References

[1] Tan X Y, Ao G, Qian G C, et al. Research on power load forecasting using deep neural network and wavelet transform [J]. International Journal of Information Technologies and Systems Approach (IJITSA), 2023, 16 (2): 1-13.

[2] Wang S B, Luo W H, Yin S X, et al. Interpretable state estimation in power systems based on the kolmogorov-arnold networks [J]. Electronics, 2025, 14 (2): 320-339.

[3] Mohammad T A, Shohani A M A W. Short-Term prediction of the solar photovoltaic power output using nonlinear autoregressive exogenous inputs and artificial neural network techniques under different weather conditions [J]. Energies, 2024, 17 (23): 6153-6168.

[4] Sanjeev, Rohit K, Ajay S, et al. Development of seasonal ARIMA model to predict wholesale price of rice in Delhi market [J]. Current Journal of Applied Science and Technology, 2022, 155-161.

[5] Dabeeruddin S, Haitham R A, Ali G, et al. Household-level energy forecasting in smart buildings using a novel hybrid deep learning model [J]. IEEE ACCESS, 2021 (9): 33498-33511.

[6] Shi C Y, Jiang H F, Zhao F Z, et al. Blood metal levels predict digestive tract cancer risk using machine learning in a U.S. cohort [J]. Scientific Reports, 2025, 15 (1): 1285-1298.

[7] Hua Q, Fan Z, Mu W, et al. A short-term power load forecasting method using CNN-GRU with an attention mechanism [J]. Energies, 2024, 18 (1): 106-122.

[8] Hardanee A F O, Demirel H. Hydropower station status prediction using RNN and LSTM algorithms for fault detection [J]. Energies, 2024, 17 (22): 5599-5621.

[9] Yang G, Du S, Duan Q, et al. Short-term price forecasting method in electricity spot markets based on Attention-LSTM-mTCN [J]. Journal of Electrical Engineering & Technology, 2022, 17 (2): 1009-1018.

[10] Lai Y B, Wang Q F, Chen G, et al. VMD-BiGRU for short-term power load forecasting with energy valley optimizer enhancement [J]. Journal of Physics: Conference Series, 2024, 2868 (1): 012004-012012.

[11] Konyrbaev N, Nikitenko Y, Shtanko V, et al. Evaluation and optimization of the naive bayes algorithm for intrusion detection systems using the USB- IDS-1 dataset [J]. Eastern-European Journal of Enterprise Technologies, 2024, 6 (2): 74-82.

[12] Aqueel A, Kumar A Y, Achhaibar S. Enhancing waste cooking oil biodiesel yield and characteristics through machine learning, response surface methodology, and genetic algorithms for optimal utilization in CI engines [J]. International Journal of Green Energy, 2024, 21 (6): 1345-1365.

[13] Jiang Q K, Wang H Y. Risk assessment and hybrid algorithm transportation path optimization model for road transport of dangerous goods [J]. IATSS Research, 2025, 49 (1): 72-80.

[14] Mohammad B S, Mohsen T. Response estimation of reinforced concrete shear walls using artificial neural network and simulated annealing algorithm [J]. Structures, 2021, 34: 1155-1168.

[15] Li Y, Qian M, Dai D J, et al. Flow control of flow boiling experimental system by Whale Optimization Algorithm (WOA) improved single neuron PID [J]. Actuators, 2024, 14 (1): 5-22.

[16] Tang F. Short-term wind power prediction based on improved sparrow search algorithm optimized long short-term memory with peephole connections [J]. Wind Engineering, 2025, 49 (1): 71-90.

[17] Xu L L, Li F Y, Chang S. A fiber recognition framework based on multi-head attention mechanism [J]. Textile Research Journal, 2024, 94 (23-24): 2629-2640.

[18] Cai H, Lu K Z, Wu Q R, et al. Adaptive classification algorithm for concept drift data stream [J]. Journal of Computer Research and Development, 2022, 59 (03): 633-646.

[19] Ai S Y, Song Z M, Shen X. Branching heuristic strategy based on variable mixing features [J]. Computer Systems & Applications, 2020, 29 (03): 200-205.