为进一步提高超短期风速预测的准确性,提出一种基于样本熵的双分解和麻雀搜索算法（SSA）改进长短时记忆神经网络（ILSTM）的耦合模型（简记为DILSTM）。首先,利用变分模态分解（VMD）分解原始序列并通过样本熵量化各子序列的复杂性;其次,利用完全自适应噪声集合经验模态分解（CEEMDAN）分解复杂度最高的子序列进一步提取特征过滤噪声;最后,将双分解得到的子序列分别建立DILSTM预测模型,并对所有子序列的预测结果叠加得到最终风速预测结果。真实风场数据实验结果表明,所提模型与LSTM、随机森林（RF）、卷积神经网络（CNN）单一模型相比,R2提升约25%,RMSE降低约65%;同时,与现有同类研究对比,也证实了所提DILSTM模型的优势,该文为提高超短期风速预测精准度提供了一种新的方法。
Abstract
To improve the accuracy of ultra-short-term wind speed prediction, a coupled model named DILSTM is proposed, which combines sample entropy, dual decomposition, and sparrow search algorithm optimized long short-term memory neural network （LSTM）. Firstly, the original sequence is decomposed using variational mode decomposition （VMD） and the complexity of each sub sequence is quantified through sample entropy; the subsequence with the highest complexity is further decomposed using complete ensemble empirical mode decomposition with adaptive noise （CEEMDAN）, further extracting features and filtering noise; Finally, the sub sequences obtained from the double decomposition are used to establish DILSTM prediction models, and the final wind speed prediction results are obtained by overlaying the prediction results of all sub sequences. The experimental results of real wind field data show the proposed model outperforms single models such as LSTM, random forest （RF）, and convolutional neural network （CNN）, with an approximately 25% increase in R2 and a 65% decrease in RMSE; Meanwhile, compared with existing similar studies, the advantages of the DILSTM model proposed in this paper have been confirmed. This study provides a new method for improving the accuracy of ultra-short-term wind speed prediction.
关键词
风速 /
神经网络 /
预测 /
变分模态分解 /
样本熵 /
麻雀搜索算法
Key words
wind speed /
neural network /
forecasting /
variational mode decomposition /
sample entropy /
sparrow search algorithm
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] SUN N, ZHANG S, PENG T, et al.A composite uncertainty forecasting model for unstable time series: application of wind speed and streamflow forecasting[J]. IEEE access, 2020, 8: 209251-209266.
[2] SUN N, ZHOU J Z, CHEN L, et al.An adaptive dynamic short-term wind speed forecasting model using secondary decomposition and an improved regularized extreme learning machine[J]. Energy, 2018, 165: 939-957.
[3] XU L, OU Y X, CAI J J, et al.Offshore wind speed assessment with statistical and attention-based neural network methods based on STL decomposition[J]. Renewable energy, 2023, 216: 119097.
[4] 赵泽妮, 云斯宁, 贾凌云, 等. 基于统计模型的短期风能预测方法研究进展[J]. 太阳能学报, 2022, 43(11): 224-234.
ZHAO Z N, YUN S N, JIA L Y, et al.Recent progress in short-term forecasting of wind energy based on statistical models[J]. Acta energiae solaris sinica, 2022, 43(11): 224-234.
[5] YAN B W, SHEN R F, LI K, et al.Spartoi-temporal correlation for simultaneous ultra-short-term wind speed prediction at multiple locations[J]. Energy, 2023, 284: 128418.
[6] 邱文智, 张文煜, 郭振海, 等. 基于二次分解和乌鸦搜索算法优化组合模型的超短期风速预测[J].太阳能学报, 2024: 45(3): 73-82.
QIU W Z, ZHANG W Y, GUO Z H, et al.Ultra-short-term wind speed forecasting based on optimal combination model of secondary decomposition and crow search algorithm[J]. Acta energiae solaris sinica, 2024, 45(3): 73-82.
[7] 张琰妮, 史加荣, 李津, 等. 融合残差与VMD-ELM-LSTM的短期风速预测[J]. 太阳能学报, 2023, 44(9): 340-347.
ZHANG Y N, SHI J R, LI J, et al.Short-term wind speed prediction based on residual and VMD-ELM-LSTM[J]. Acta energiae solaris sinica, 2023, 44(9): 340-347.
[8] YANG Q S, HUANG G Q, LI T, et al.A novel short-term wind speed prediction method based on hybrid statistical-artificial intelligence model with empirical wavelet transform and hyperparameter optimization[J]. Journal of wind engineering and industrial aerodynamics, 2023, 240: 105499.
[9] 孙娜, 张楠, 张帅, 等. 一种基于物理核函数高斯过程回归的月径流预报模型及其应用[J]. 水电能源科学, 2023, 41(4): 39-43.
SUN N, ZHANG N, ZHANG S, et al.Monthly runoff prediction model and its application based on GPR with physically composite kernel[J]. Water resources and power, 2023, 41(4): 39-43.
[10] RICHMAN J S.Multivariate neighborhood sample entropy: a method for data reduction and prediction of complex data[J]. Methods in enzymology, 2011, 487: 397-408.
[11] LI K, SHEN R F, WANG Z G, et al.An efficient wind speed prediction method based on a deep neural network without future information leakage[J]. Energy, 2023, 267: 126589.
[12] SUN N, ZHANG S, PENG T, et al.Multi-variables-driven model based on random forest and Gaussian process regression for monthly streamflow forecasting[J]. Water, 2022, 14(11): 1828.
[13] CHEN B Y, XIE D T, HUANG R W, et al.Research on IGBT aging prediction method based on adaptive VMD decomposition and GRU-AT model[J]. Energy reports, 2023, 9: 1432-1446.
[14] 马卫东, 刘子全, 姚楠, 等. AOA-CEEMDAN和融合特征在齿轮箱故障诊断中的应用[J]. 机电工程, 2024, 41(5): 817-826.
MA W D, LIU Z Q, YAO N, et al.AOA-CEEMDAN and fusion features and its application in gearbox fault diagnosis[J]. Journal of mechanical & electrical engineering, 2024, 41(5): 817-826.
[15] YAN Y, WANG X R, REN F, et al.Wind speed prediction using a hybrid model of EEMD and LSTM considering seasonal features[J]. Energy reports, 2022, 8: 8965-8980.
[16] 兰小机, 贺永兰, 武帅文. 基于RF-BiLSTM模型的河流水质预测[J]. 长江科学院院报, 2024, 41(7): 57-63, 71.
LAN X J, HE Y L, WU S W.River water quality prediction based on RF-BiLSTM model[J]. Journal of Changjiang River scientific research institute, 2024, 41(7): 57-63, 71.
[17] ZHANG Z G, YIN J C.Spatial-temporal offshore wind speed characteristics prediction based on an improved purely 2D CNN approach in a large-scale perspective using reanalysis dataset[J]. Energy conversion and management, 2024, 299: 117880.
[18] ZHANG H C, PENG Z Y, TANG J J, et al.A multi-layer extreme learning machine refined by sparrow search algorithm and weighted mean filter for short-term multi-step wind speed forecasting[J]. Sustainable energy technologies and assessments, 2022, 50: 101698.
[19] LIANG T, CHAI C J, SUN H X, et al.Wind speed prediction based on multi-variable Capsnet-BILSTM-MOHHO for WPCCC[J]. Energy, 2022, 250: 123761.
[20] 毕贵红, 黄泽, 赵四洪, 等. 基于混合分解和PCG-BiLSTM的风速短期预测[J]. 太阳能学报, 2024, 45(1): 159-170.
BI G H, HUANG Z, ZHAO S H, et al.Short-term prediction of wind speed based on hybrid decomposition and PCG-BiLSTM[J]. Acta energiae solaris sinica, 2024, 45(1): 159-170.
基金
江苏省高校自然科学基金（20KJD480003）; 江苏省双创计划（JSSCBS（2020）31035）; 江苏省自然科学基金（BK20201069）
PDF(1782 KB)
