基于极值优化的LSTM短期风速预测
Research on LSTM Short-Term Wind Speed Prediction Based on Extreme Value Optimization
DOI: 10.12677/ORF.2023.136592, PDF,    国家自然科学基金支持
作者: 刘立夫：武汉科技大学理学院，湖北 武汉；贾世会*：武汉科技大学理学院，湖北 武汉；冶金工业过程系统科学湖北省重点实验室，湖北 武汉
关键词: 极值优化；LSTM；风速预测；神经网络；Extreme Value Optimization； LSTM； Wind Speed Prediction； Neural Network
摘要: 针对风电机组组网运行存在的功率波动性和随机性的问题，提出一种基于极值优化的长短时记忆神经网络，以提高风速预测的精度和风电机组运行的稳定性。首先对数据进行标准化处理，以得到较好的拟合并防止训练发散。利用LSTM的非线性拟合能力，将数据代入LSTM网络当中进行预测。最后针对LSTM网络所预测的风速序列在局部极值点处与实际值差别较大的情况，使用极值优化模型分别对部分极大值点和极小值点进行算术平均和几何平均。实验结果表明，相比于单纯的LSTM模型，本文提出的基于极值优化的LSTM模型在风速预测上具有更好的精度。
Abstract: Aiming at the problems of power fluctuation and randomness in wind turbine network operation, a long-short-term memory neural network based on extreme value optimization is proposed to improve the accuracy of wind speed prediction and the stability of wind turbine operation. The data is first normalized to get a better fit and prevent training divergence. Using the nonlinear fitting ability of LSTM, the data is substituted into the LSTM network for prediction. Finally, in view of the large difference between the wind speed sequence predicted by the LSTM network and the actual value at the local extreme points, the extreme value optimization model is used to perform arithmetic mean and geometric mean on some maximum and minimum points, respectively. The experimental results show that, compared with the pure LSTM model, the LSTM model based on extreme value optimization proposed in this paper has better accuracy in wind speed prediction.
文章引用：刘立夫, 贾世会. 基于极值优化的LSTM短期风速预测[J]. 运筹与模糊学, 2023, 13(6): 5959-5968. https://doi.org/10.12677/ORF.2023.136592

参考文献

[1] 杜颖, 等. 基于最小二乘支持向量机的风电场短期风速预测[J]. 电网技术, 2008, 284(15): 62-66.
[2] 王红刚. 基于融合长短时记忆网络的超短期风速预测研究[D]: [硕士学位论文]. 武汉: 武汉理工大学, 2019.
[3] Li, J. and Li, M. (2019) Prediction of Ultra-Short-Term Wind Power Based on BBO-KELM Method. Journal of Renewable and Sustainable Energy, 11, Article ID: 056104. [Google Scholar] [CrossRef] 
[4] 陈好杰, 程浩忠, 徐国栋, 马则良, 傅业盛. 基于云理论的风电场群长期出力区间预测[J]. 电力系统保护与控制, 2019, 47(3): 110-117.
[5] 黄越辉, 曲凯, 李驰, 司刚全. 基于K-meansMCMC算法的中长期风电时间序列建模方法研究[J]. 电网技术, 2019, 43(7): 2469-2476.
[6] 林涛, 刘航鹏, 赵参参, 赵成林, 马同宽. 基于SSA-PSO-ANFIS的短期风速预测研究[J]. 太阳能学报, 2021, 42(3): 128-134.
[7] 方必武, 刘涤尘, 王波, 闫秉科, 汪勋婷. 基于小波变换和改进萤火虫算法优化LSSVM的短期风速预测[J]. 电力系统保护与控制, 2016, 44(8): 37-43.
[8] 谢义超. 基于CEEMDAN分解和改进的LSTM模型的短期风速预测[D]: [硕士学位论文]. 武汉: 武汉科技大学, 2021.
[9] Liu, H., Duan, Z., Han, F.Z. and Li, Y.F. (2018) Big Multi-Step Wind Speed Forecasting Model Based on Secondary Decomposition, Ensemble Method and Error Correction Algorithm. Energy Conversion and Management, 156, 525-541. [Google Scholar] [CrossRef] 
[10] Giorgi, M., Ficarella, A. and Tarantino, M. (2011) As-sessment of the Benefits Of Numerical Weather Predictions in Wind Power Forecasting Based on Statistical Methods. Energy, 36, 3968-3978. [Google Scholar] [CrossRef] 
[11] Billinton, R., Chen, H. and Ghajar, R. (1996) A Sequential Simulation Technique for Adequacy Evaluation of Generating Systems Including Wind Energy. IEEE Transactions on Energy Conversion, 11, 728-734. [Google Scholar] [CrossRef] 
[12] Wang, Y.R., Wang, D.C. and Tang, Y. (2020) Clustered Hybrid Wind Power Prediction Model Based on ARMA, PSO-SVM and Clustering Methods. IEEE Access, 8, 17071-17079. [Google Scholar] [CrossRef] 
[13] Cardoso, C.A. and Cruz, G.L. (2016) Forecasting Natural Gas Consumption Using ARIMA Models and Artificial Neural Networks. IEEE Latin America Transactions, 14, 2233-2238. [Google Scholar] [CrossRef] 
[14] Jin, J., Wang, B., Yu, M. and Liu, Q. (2017) The Short-Term Wind Speed Prediction Based on HF-EMD and BP Neural Network Model. 2017 12th IEEE Conference on Industrial Electronics and Applications (ICIEA), Siem Reap, 18-20 June 2017, 489-494. [Google Scholar] [CrossRef] 
[15] 蒋艳凰, 赵强利. 机器学习方法[M]. 北京: 电子工业出版社, 2009: 142-182.
[16] Bossanyi, E.A. (1985) Short-Term Wind Prediction Using Kalman Filters. Wind Engi-neering, 9, 1-8.
[17] 蒋宗礼. 人工神经网络导论[M]. 北京: 高等教育出版社, 2001.
[18] 朱大奇, 史慧. 人工神经网络原理及应用[M]. 北京: 科学出版社, 2006.
[19] Mikolov, T., Karafiát, M., Burget, L., et al. (2015) Recur-rent Neural Network Based Language Model. Interspeech, Conference of the International Speech Communication Association, Makuhari, 26-30 September 2010, 1045-1048.
[20] Bengio, Y., Frasconi, P. and Simard, P. (1993) The Problem of Learning Long-Term Dependencies in Recurrent Networks. IEEE International Conference on Neural Networks, San Francisco, 28 March-1 April 1993, 1183-1188.
[21] Hochreiter, S. and Schmidhuber, J. (1997) Long Short—Term Memory. Neural Computation, 9, 1735-1780. [Google Scholar] [CrossRef] [PubMed]
[22] 黄进. 基于极值偏移优化的ARIMA的短期风速预测[J]. 制造业自动化, 2018, 40(1): 79-85.

为你推荐