負載預測在能源規劃與運行中扮演著至關重要的角色。隨著能源需求的增長和環境議題的日益嚴重，準確地預測用電負載成為電力用戶和能源供應商共同關注的焦點。對企業而言，準確的需量預測可以幫助其制定有效的生產計畫和預算，從而降低生產成本並提高效率，即時的用電負載預測成為了管理電費支出的不可或缺方法。本文採用一種將類神經網路與整合學習算法(Ensemble Learning)相結合的混合預測模型，以預測每15分鐘為週期的短期用電負載，提供某鋼鐵廠內供電中心用電管理的依據，以避免超約用電問題，減少維運成本。為了驗證模型準確性與適合性，本文使用某鋼鐵廠內負載用電量和發電機發電量作相關性分析，並與單一模型之間比較預測準確度。實驗結果顯示，混合模型的準確度比單一模型提高約13%-24%，本研究所提出的用電負載預測混合模型能有效地提高預測的準確度。

Load forecasting plays a crucial role in energy planning and operations. With the growing demand for energy and the increasing severity of environmental issues, accurately predicting electricity load has become a shared focus for power users and energy suppliers. For enterprises, accurate demand forecasting helps develop effective production plans and budgets, reducing production costs and improving efficiency. Real-time electricity load forecasting has become an indispensable method for managing electricity expenses. This study employs a hybrid forecasting model that combines neural networks with ensemble learning algorithms to predict short-term electricity loads in 15-minute intervals. The model provides a basis for electricity management at a steel plant's power supply center, helping to avoid exceeding contractual electricity limits and reducing maintenance and operational costs. To validate the model's accuracy and suitability, this study analyzes the correlation between the electricity load and generator output at a steel plant and compares the forecasting accuracy of the hybrid model with that of single models. Experimental results show that the hybrid model improves accuracy by approximately 13%-24% compared to single models. The proposed hybrid model for electricity load forecasting effectively enhances prediction accuracy. The model’s improved accuracy and reliability thus offer energy managers powerful tools to optimize procurement strategies and prevent unexpected costs.

[1] M. A. Hammad, B. Jereb, B. Rosi, and D. Dragan, “Methods and Models for Electric Load Forecasting: A Comprehensive Review,” Logist. Sustain. Transp., vol. 11, no. 1, pp. 51–76, Feb. 2020.
[2] S. Akhtar et al., “Short-Term Load Forecasting Models: A Review of Challenges, Progress, and the Road Ahead,” Energies, vol. 16, no. 10, p. 4060, May 2023.
[3] A. Román-Portabales, M. López-Nores, and J. J. Pazos-Arias, “Systematic Review of Electricity Demand Forecast Using ANN-Based Machine Learning Algorithms,” Sensors, vol. 21, no. 13, p. 4544, Jul. 2021.
[4] A. Azeem, I. Ismail, S. M. Jameel, and V. R. Harindran, “Electrical Load Forecasting Models for Different Generation Modalities: A Review,” IEEE Access, vol. 9, pp. 142239–142263, 2021.
[5] G. Dudek and P. Pełka, “Pattern similarity-based machine learning methods for mid-term load forecasting: A comparative study,” Applied Soft Computing, vol. 104, p. 107223, Jun. 2021.
[6] G. Dudek, “Pattern similarity-based methods for short-term load forecasting – Part 1: Principles,” Applied Soft Computing, vol. 37, pp. 277–287, Dec. 2015.
[7] G. Dudek, “Pattern similarity-based methods for short-term load forecasting – Part 2: Models,” Applied Soft Computing, vol. 36, pp. 422–441, Nov. 2015.
[8] K. Goswami, A. Ganguly, and A. K. Sil, “Day Ahead Forecasting and Peak Load Management using Multivariate Auto Regression Technique,” in 2018 IEEE Applied Signal Processing Conference (ASPCON), Kolkata, India: IEEE, pp. 279–282, Dec. 2018.
[9] S. H. Rafi, Nahid-Al-Masood, and M. M. Mahdi, “A Short-Term Load Forecasting Technique Using Extreme Gradient Boosting Algorithm,” in 2021 IEEE PES Innovative Smart Grid Technologies - Asia (ISGT Asia), Brisbane, Australia: IEEE, pp. 1–5, Dec. 2021.
[10] G. Dudek, “A Comprehensive Study of Random Forest for Short-Term Load Forecasting,” Energies, vol. 15, no. 20, p. 7547, Oct. 2022.
[11] A. A. Mamun, Md. Sohel, N. Mohammad, Md. S. Haque Sunny, D. R. Dipta, and E. Hossain, “A Comprehensive Review of the Load Forecasting Techniques Using Single and Hybrid Predictive Models,” IEEE Access, vol. 8, pp. 134911–134939, 2020.
[12] M. Tan, S. Yuan, S. Li, Y. Su, H. Li, and F. H. He, “Ultra-Short-Term Industrial Power Demand Forecasting Using LSTM Based Hybrid Ensemble Learning,” IEEE Trans. Power Syst., vol. 35, no. 4, pp. 2937–2948, Jul. 2020.
[13] S. H. Rafi, Nahid-Al-Masood, S. R. Deeba, and E. Hossain, “A Short-Term Load Forecasting Method Using Integrated CNN and LSTM Network,” IEEE Access, vol. 9, pp. 32436–32448, 2021.
[14] Nur Shakirah Md Salleh, Azizah Suliman, and Bo Nørregaard Jørgensen, “Comparison of Electricity Load Prediction Errors Between Long Short-Term Memory Architecture and Artificial Neural Network on Smart Meter Consumer,” In Advances in Visual Informatics: 7th International Visual Informatics Conference, IVIC 2021, vol. 13051, pp. 600-609, Nov. 2021.
[15] F. M. Bianchi, E. Maiorino, M. C. Kampffmeyer, A. Rizzi, and R. Jenssen, “Recurrent Neural Networks for Short-Term Load Forecasting: An Overview and Comparative Analysis, ” in SpringerBriefs in Computer Science. Cham: Springer International Publishing, 2017.
[16] C. Olah, “Understanding LSTM Networks,” Aug. 2015. [Online]. Available: https://colah.github.io/posts/2015-08-Understanding-LSTMs/
[17] Y. Lecun, L. Bottou, Y. Bengio and P. Haffner, "Gradient-based learning applied to document recognition," in Proceedings of the IEEE, vol. 86, no. 11, pp. 2278-2324, Nov. 1998.
[18] Z. Lin, L. Cheng, and G. Huang, “Electricity consumption prediction based on LSTM with attention mechanism,” IEEJ Trans. Electr. Electron. Eng., vol. 15, no. 4, pp. 556–562, Apr. 2020.
[19] D. H. Vu, K. M. Muttaqi, and A. P. Agalgaonkar, “Assessing the influence of climatic variables on electricity demand,” in 2014 IEEE PES General Meeting | Conference & Exposition, National Harbor, MD, USA: IEEE, pp. 1–5, Jul. 2014.
[20] M. N. Fekri, H. Patel, K. Grolinger, and V. Sharma, “Deep learning for load forecasting with smart meter data: Online Adaptive Recurrent Neural Network,” Applied Energy, vol. 282, p. 116177, Jan. 2021.