Research on an Energy Economic Demand Time Series Prediction Model Based on the Integration of XGBoost and Deep Learning
DOI:
https://doi.org/10.67541/jdc2601Keywords:
Energy economic demand forecasting; Time series prediction; XGBoost; Deep learning; Ensemble modelAbstract
To address the significant nonlinear characteristics of energy economic demand time series, the complex coupling of multiple influencing factors, and the insufficient generalization ability of traditional prediction models, this paper proposes a time series prediction model for energy economic demand based on the integration of XGBoost and deep learning. First, the energy demand data, macroeconomic indicators, and meteorological variables are uniformly preprocessed and feature-enhanced to construct a multi-dimensional input system that includes lag features, sliding window features, and periodic features. Then, an improved XGBoost is used to mine nonlinear relationships within structured features, while a deep learning network extracts temporal dynamic features and long-term dependencies. A dynamic weight fusion mechanism is designed to achieve collaborative prediction of the two models. Finally, model performance is verified through comparative, ablation, robustness, and interpretability analyses. The experimental results show that the RMSE, MAE, and MAPE on the test set reach 28.74, 21.52, and 2.83%, respectively, and the coefficient of determination (R²) reaches 0.983. Compared with the Transformer model, the MAPE is reduced by 32.8%, and compared with the traditional ARIMA model, it is reduced by 68.2%. Furthermore, under conditions of 20% outlier perturbation and 20% missing data, the model’s performance retention rates still reach 88.2% and 90.8%, respectively, demonstrating good stability and generalization ability. The research results show that the proposed integrated model can effectively improve the accuracy of energy economic demand prediction, providing reliable technical support for energy planning, load management, and smart energy decision-making.
References
[1] Alarcón, R. G., Alarcón, M. A., González, A. H., & Ferramosca, A. (2025). Artificial Neural Networks for Energy Demand Prediction in an Economic MPC‐Based Energy Management System. International Journal of Robust and Nonlinear Control, 35(2), 642-658. https://doi.org/10.1002/rnc.7671
[2] Karpavicius, T., Balezentis, T., & Streimikiene, D. (2025). Energy security indicators for sustainable energy development: Application to electricity sector in the context of state economic decisions. Sustainable Development, 33(1), 1381-1400. https://doi.org/10.1002/sd.3190
[3] Aderibigbe, A. O., Ani, E. C., Ohenhen, P. E., Ohalete, N. C., & Daraojimba, D. O. (2023). Enhancing energy efficiency with ai: a review of machine learning models in electricity demand forecasting. Engineering Science & Technology Journal, 4(6), 341-356. https://doi.org/10.51594/estj.v4i6.636
[4] Szostek, K., Mazur, D., Drałus, G., & Kusznier, J. (2024). Analysis of the effectiveness of ARIMA, SARIMA, and SVR models in time series forecasting: A case study of wind farm energy production. Energies, 17(19), 4803. https://doi.org/10.3390/en17194803
[5] Jain, R., Khairnar, A. S., & Karthikeyan, S. P. (2024, October). Forecasting Indian Conventional Generation: A Comparative Analysis of ARIMA and SARIMA Models for Improved Energy Demand Prediction. In 2024 International Conference on Sustainable Energy: Energy Transition and Net-Zero Climate Future (ICUE) (pp. 1-8). IEEE. https://doi.org/10.1109/ICUE63019.2024.10795531
[6] Abbasimehr, H., Paki, R., & Bahrini, A. (2023). A novel XGBoost-based featurization approach to forecast renewable energy consumption with deep learning models. Sustainable Computing: Informatics and Systems, 38, 100863. https://doi.org/10.1016/j.suscom.2023.100863
[7] Zhang, T., Zhang, X., Liu, Y., Chow, Y. H., Iu, H. H., & Fernando, T. (2023). Long-term energy and peak power demand forecasting based on sequential-XGBoost. IEEE Transactions on Power Systems, 39(2), 3088-3104. https://doi.org/10.1109/TPWRS.2023.3289400
[8] Zhang, L., & Jánošík, D. (2024). Enhanced short-term load forecasting with hybrid machine learning models: CatBoost and XGBoost approaches. Expert Systems with Applications, 241, 122686. https://doi.org/10.1016/j.eswa.2023.122686
[9] Usmani, M., Memon, Z. A., Zulfiqar, A., & Qureshi, R. (2024). Preptimize: Automation of time series data preprocessing and forecasting. Algorithms, 17(8), 332. https://doi.org/10.3390/a17080332
[10] Lotysh, V., Gumeniuk, L., & Humeniuk, P. (2023). Comparison of the effectiveness of time series analysis methods: SMA, WMA, EMA, EWMA, and Kalman filter for data analysis. Informatyka, automatyka, pomiary w gospodarce i ochronie środowiska, 13(3), 71-74. https://doi.org/10.35784/iapgos.3652
[11] Andrasto, T., Ristanto, R. D., Sukamta, S., Aufa, A., Hafidz, C. M., Sandyawan, B., & Alfarisi, A. F. (2023, June). Method EWMA (Exponentially Weighted Moving Average) as a filter to fine and remove noise on time series data. In IOP Conference Series: Earth and Environmental Science (Vol. 1203, No. 1, p. 012012). IOP Publishing. https://doi.org/10.1088/1755-1315/1203/1/012012
[12] Curiël, R., Alsahag, A. M. M., & Mohammadi Ziabari, S. S. (2025). Integrating Climate and Economic Predictors in Hybrid Prophet–(Q) LSTM Models for Sustainable National Energy Demand Forecasting: Evidence from The Netherlands. Sustainability, 17(19), 8687. https://doi.org/10.3390/su17198687
[13] Zulqarnain, F. N. U., & Hasan, Z. (2024). Artificial Intelligence Applications For Predicting Renewable-Energy Demand Under Climate Variability. American Journal of Scholarly Research and Innovation, 3(01), 84-116. https://doi.org/10.63125/sg0j6930
[14] Al-Suod, M., Victor, B., Valerii, T., Olha, C., Galina, S., Zannon, M., & Dmytro, Z. (2025). Forecasting energy consumption of a mining plant using artificial neural networks. IEEE Access, 13, 63237-63247. https://doi.org/10.1109/ACCESS.2025.3558445
[15] Maharana, K., Mondal, S., & Nemade, B. (2022). A review: Data pre-processing and data augmentation techniques. Global Transitions Proceedings, 3(1), 91-99. https://doi.org/10.1016/j.gltp.2022.04.020
[16] Goceri, E. (2023). Medical image data augmentation: techniques, comparisons and interpretations. Artificial intelligence review, 56(11), 12561-12605. https://doi.org/10.1007/s10462-023-10453-z
[17] Mazibuko, T., & Akindeji, K. (2025). Hybrid forecasting for energy consumption in south Africa: LSTM and XGBoost approach. Energies, 18(16), 4285. https://doi.org/10.3390/en18164285
[18] Noorunnahar, M., Chowdhury, A. H., & Mila, F. A. (2023). A tree based eXtreme Gradient Boosting (XGBoost) machine learning model to forecast the annual rice production in Bangladesh. PloS one, 18(3), e0283452. https://doi.org/10.1371/journal.pone.0283452
[19] Bai, Y., & Cai, C. X. (2024). Predicting VIX with adaptive machine learning. Quantitative Finance, 24(12), 1857-1873. https://doi.org/10.1080/14697688.2024.2439458
[20] Zhuo, H., Wang, Y., Li, S., Ma, H., Wang, Y., Li, R., & Liu, G. (2026). Enhancing Electricity Price Prediction Accuracy With an Attention Mechanism‐LSTM Hybrid Model. Engineering Reports, 8(3), e70657. https://doi.org/10.1002/eng2.70657
[21] Zeng, C., Tian, Y., Zheng, G., & Gao, Y. (2024). How much can time-related features enhance time series forecasting?. arXiv preprint arXiv:2412.01557. https://doi.org/10.48550/arXiv.2412.01557
[22] Khashei, M., Ahmadi, M., & Chahkoutahi, F. (2025). A mean weighted squared error-based neural classifier for intelligent pattern recognition in smart grids. International Journal of Electrical Power & Energy Systems, 170, 110972. https://doi.org/10.1016/j.ijepes.2025.110972
[23] Ao, X., Gong, Y., & He, A. (2025). A review of time series prediction models based on deep learning. IEEE Access. https://doi.org/10.1109/ACCESS.2025.3602791
[24] Yan, L., Peng, J., Gao, D., Wu, Y., Liu, Y., Li, H., ... & Huang, Z. (2022). A hybrid method with cascaded structure for early-stage remaining useful life prediction of lithium-ion battery. Energy, 243, 123038. https://doi.org/10.1016/j.energy.2021.123038
[25] Nong, Y. (2025). AI-Driven Predictive Cascade Failure Analysis Using Multi-Modal Environmental-Infrastructure Data Fusion Real-Time Prediction Framework for Critical Energy Infrastructure. Authorea Preprints. https://doi.org/10.36227/techrxiv.176288076.69687745/v2
[26] Gaugel, S., & Reichert, M. (2023). Industrial transfer learning for multivariate time series segmentation: a case study on hydraulic pump testing cycles. Sensors, 23(7), 3636. https://doi.org/10.3390/s23073636
[27] Liu, F., Chen, L., Zheng, Y., & Feng, Y. (2022). A prediction method with data leakage suppression for time series. Electronics, 11(22), 3701. https://doi.org/10.3390/electronics11223701
[28] Salehin, I., & Kang, D. K. (2023). A review on dropout regularization approaches for deep neural networks within the scholarly domain. Electronics, 12(14), 3106. https://doi.org/10.3390/electronics12143106
[29] Wang, M., Chen, L., & Chen, H. (2022). Multi-strategy learning boosted colony predation algorithm for photovoltaic model parameter identification. Sensors, 22(21), 8281. https://doi.org/10.3390/s22218281
[30] Wang, Y., Wu, H., Dong, J., Liu, Y., Wang, C., Long, M., & Wang, J. (2026). Deep time series models: A comprehensive survey and benchmark. IEEE Transactions on Pattern Analysis and Machine Intelligence. https://doi.org/10.1109/TPAMI.2026.3690845
[31] Ren, L., Jia, Z., Laili, Y., & Huang, D. (2023). Deep learning for time-series prediction in IIoT: progress, challenges, and prospects. IEEE transactions on neural networks and learning systems, 35(11), 15072-15091. https://doi.org/10.1109/TNNLS.2023.3291371
[32] Zhang, Q. (2026). Research on Multimodal Prediction Model of Sports Event Ticket Revenue Based on Transformer. International Scientific Technical and Economic Research, 4(2), 124-147. https://doi.org/10.71451/ISTAER2618
[33] Zhou, H. (2026). Research on the Fusion Model of DeepFM and XGBoost for Digital Consumer Behavior Prediction. International Scientific Technical and Economic Research, 4(2), 98-123. https://doi.org/10.71451/ISTAER2617
[34] Qin, Y. (2026). Research on Long Sequence Learning Behavior Modeling Based on Transformer-XL. International Scientific Technical and Economic Research, 4(2), 1-20. https://doi.org/10.71451/ISTAER2613
[35] Hodson, T. O. (2022). Root mean square error (RMSE) or mean absolute error (MAE): When to use them or not. Geoscientific Model Development Discussions, 2022, 1-10. https://doi.org/10.5194/gmd-15-5481-2022
[36] Xu, H. (2026). A Deep Reinforcement Learning Signal Control Algorithm for Traffic Carbon Emission Optimization. International Scientific Technical and Economic Research, 4(1), 200–221. https://doi.org/10.71451/ISTAER2610
[37] Chen, G. (2026). Research on algorithm improvement of ARIMA-LSTM hybrid model in time series prediction of inflation rate. International Scientific Technical and Economic Research, 4(1), 90-122. https://doi.org/10.71451/ISTAER2605
[38] Ding, Y., & Chen, G. (2026). Joint Prediction Model of Reservoir Parameters Based on Multimodal Transformer Graph Neural Operator Physical Constraint Network. International Scientific Technical and Economic Research, 4(1), 70-89. https://doi.org/10.71451/ISTAER2604
[39] Zeng, X. (2026). Cross-Border Trade Fraud Detection via Integrated Heterogeneous Graph Neural Network and XGBoost. International Scientific Technical and Economic Research, 4(1),47-69. https://doi.org/10.71451/ISTAER2603
[40] Wang, W., Shen, S., & Wang, Y. (2026). Forecasting Short-Term Export Volumes with Hybrid Models Integrating SARIMA with Attention-Based LSTM, International Scientific Technical and Economic Research, 4(1),1-22. https://doi.org/10.71451/ISTAER2601
Published
Issue
Section
License
Copyright (c) 2026 Guona Chen (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
