基于LSTM方法能源需求预测方法及其在
能源调度中的应用
本文研究基于长短期记忆网络（LSTM）的能源需求预测方法，解决传统预测方法难以处理复杂非线性和长期依赖性的问题。通过美国PJM电力市场数据实验验证，LSTM模型在预测准确性和稳定性上显著优于传统方法。提出的模型在能源调度中展现出提升电力系统效率与稳定性的潜力，特别在智能化能源管理和可再生能源利用方面。未来研究将优化算法并结合更多影响因素，以增强其在复杂能源系统中的适应性和应用范围。
[1] 徐佳，沈泺成，罗．J. OPTICS, and O. TECHNOLOGY, “基于自动响应控制器的负荷高峰期负载均衡研究，” vol. 20, no. 4, pp. 166-171, 2022.
[2] 于贵瑞，郝天象，and 朱．J. 中国科学院院刊，“中国碳达峰，碳中和行动方略之探讨，” vol. 37, no. 4, pp. 423-434, 2022.
[3]S. F. Rafique, Z. J. I. G. Jianhua, Transmission, and Distribution, “Energy management system, generation and demand predictors: a review,” vol. 12, no. 3, pp. 519-530, 2018.
[4]P. Vrablecová, A. B. Ezzeddine, V. Rozinajová, S. Šárik, A. K. J. C. Sangaiah, and E. Engineering, “Smart grid load forecasting using online support vector regression,”vol. 65, pp. 102-117, 2018.
[5]A. Gasparin, S. Lukovic, and C. J. C. T. o. I. T. Alippi, “Deep learning for time series forecasting: The electric load case,” vol. 7, no. 1, pp. 1-25, 2022.
[6]I. K. Nti, M. Teimeh, O. Nyarko-Boateng, A. F. J. J. o. E. S. Adekoya, and I. Technology, “Electricity load forecasting: a systematic review,” vol. 7, pp. 1-19, 2020.
[7]Y. Yu, X. Si, C. Hu, and J. J. N. c. Zhang, “A review of recurrent neural networks: LSTM cells and network architectures,” vol. 31, no. 7, pp. 1235-1270, 2019.
[8]T. J. A. J. o. A. S. Fadziso and Engineering, “Overcoming the vanishing gradient problem during learning recurrent neural nets (RNN),” vol. 9, no. 1, pp. 197-208, 2020.
[9]K. Albeladi, B. Zafar, A. J. I. J. o. A. C. S. Mueen, and Applications, “Time Series Forecasting using LSTM and ARIMA,” vol. 14, no. 1, pp. 313-320, 2023.
[10]K. Xie, H. Yi, G. Hu, L. Li, and Z. J. N. Fan, “Short-term power load forecasting based on Elman neural network with particle swarm optimization,” vol. 416, pp.