提出一种基于无人机搭载测风的超短期风速预测混合模型,该模型首先融合Time2Vec时间嵌入层对复杂非线性时间信息进行表征,其次采用宽深度卷积神经网络（WDCNN）,能够在风速特征准确提取的同时进一步降低高频噪声的影响,在准确特征表征的基础上,结合双向门控单元（BiGRU）与注意力机制实现短期风速的准确预测。使用无人机搭载在5~25 m获取的风速数据进行验证,所提混合模型在各高度层上对风速预测的准确性均优于对比方法,在风速波动性最大的25 m高度层,该模型风速预测的EMAE、ERMSE、EMSE分别为0.1455、0.4124和0.1700 m/s,相较于对比模型分别降低45%、25%和43%以上。
Abstract
This paper proposes a hybrid model for ultra-short term wind speed prediction based on UAV-based wind measurement. The model firstly integrates Time2Vec temporal embedding layer to extract the complex non-linear temporal information, and secondly adopts deep convolutional neural networks with wide first-layer kernels（WDCNN） to accurately extract the wind speed features while further reducing the influence of high-frequency noise, and on the basis of the accurate feature extraction, the accurate prediction of short-term wind speed is achiered by combining bi-directional gating unit （BiGRU） and the attention mechanism to achieve the accurate prediction of short-term wind speed. The accuracy of the hybrid model proposed in this paper is superior to the comparative method for wind speed prediction at all altitude layers when wind speed data acquired by a UAV mounted at 5-25 m is used for validation. Furthermore, at the 25 m altitude layer, where wind speed volatility is the greatest, the model's wind speed prediction demonstrates a lower mean absolute error （EMAE）, root mean square error （ERMSE）, and mean square error （EMSE） than that of the comparative method. The values are 0.1455, 0.4124, and 0.1700 m/s, respectively, which are lower than those of the comparison model by 45%, 25%, and 43%, respectively. These values have been reduced by more than 45%, 25%, and 43%, respectively, in comparison with the values obtained from the comparative model.
关键词
风电 /
风速 /
神经网络 /
时间序列 /
Time2Vec /
注意力机制 /
无人机搭载测风
Key words
wind power /
wind speed /
neural networks /
time series /
Time2Vec /
attentional mechanism /
unmanned aerial vehicle （UAV） wind measurement
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] CHENG L L, ZANG H X, XU Y, et al.Augmented convolutional network for wind power prediction: a new recurrent architecture design with spatial-temporal image inputs[J]. IEEE transactions on industrial informatics, 2021, 17(10): 6981-6993.
[2] HOU T H, XING H Y, GU W, et al.Wind pressure orthogonal decomposition anemometer: a wind measurement device for multi-rotor UAVs[J]. Drones, 2023, 7(6): 366.
[3] WANG Y, ZOU R M, LIU F, et al.A review of wind speed and wind power forecasting with deep neural networks[J]. Applied energy, 2021, 304: 117766.
[4] HU J M, HENG J N, WEN J M, et al.Deterministic and probabilistic wind speed forecasting with de-noising-reconstruction strategy and quantile regression based algorithm[J]. Renewable energy, 2020, 162: 1208-1226.
[5] ZHANG Y G, ZHAO Y, KONG C H, et al.A new prediction method based on VMD-PRBF-ARMA-E model considering wind speed characteristic[J]. Energy conversion and management, 2020, 203: 112254.
[6] YUNUS K, THIRINGER T, CHEN P Y.ARIMA-based frequency-decomposed modeling of wind speed time series[J]. IEEE transactions on power systems, 2016, 31(4): 2546-2556.
[7] 吴俊利, 张步涵, 王魁. 基于Adaboost的BP神经网络改进算法在短期风速预测中的应用[J]. 电网技术, 2012, 36(9): 221-225.
WU J L, ZHANG B H, WANG K.Application of Adaboost-based BP neural network for short-term wind speed forecast[J]. Power system technology, 2012, 36(9): 221-225.
[8] WANG J Z, WANG S Q, YANG W D.A novel non-linear combination system for short-term wind speed forecast[J]. Renewable energy, 2019, 143: 1172-1192.
[9] 王顺江, 范永鑫, 潘超, 等. 基于主成分约简聚类的优化ELM短期风速组合预测[J]. 太阳能学报, 2021, 42(8): 368-373.
WANG S J, FAN Y X, PAN C, et al.Short-term wind speed combined forecasting based on optimized ELM of principal component reduction clustering[J]. Acta energiae solaris sinica, 2021, 42(8): 368-373.
[10] PEI S Q, QIN H, ZHANG Z D, et al.Wind speed prediction method based on empirical wavelet transform and new cell update long short-term memory network[J]. Energy conversion and management, 2019, 196: 779-792.
[11] YILDIZ C, ACIKGOZ H, KORKMAZ D, et al.An improved residual-based convolutional neural network for very short-term wind power forecasting[J]. Energy conversion and management, 2021, 228: 113731.
[12] DING M, ZHOU H, XIE H, et al.A gated recurrent unit neural networks based wind speed error correction model for short-term wind power forecasting[J]. Neurocomputing, 2019, 365: 54-61.
[13] 刘擘龙, 张宏立, 王聪, 等. 基于序列到序列和注意力机制的超短期风速预测[J]. 太阳能学报, 2021, 42(9): 286-294.
LIU B L, ZHANG H L, WANG C, ET AL.Ultra-short-term wind speed prediction based on sequence-to-sequence and attention mechanism[J]. Acta energiae solaris sinica, 2021, 42(9): 286-294.
[14] 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.
[15] ZHAO Z N, YUN S N, JIA L Y, et al.Hybrid VMD-CNN-GRU-based model for short-term forecasting of wind power considering spatio-temporal features[J]. Engineering applications of artificial intelligence, 2023, 121: 105982.
[16] 臧海祥, 赵勇凯, 张越, 等. 基于低风速功率修正和损失函数改进的超短期风电功率预测[J]. 电力系统自动化, 2024, 48(7): 248-257.
ZANG H X, ZHAO Y K, ZHANG Y, et al.Ultra-short-term wind power prediction based on power correction under low wind speed and improved loss function[J]. Automation of electric power systems, 2024, 48(7): 248-257.
[17] 陈蕻峰, 王贺, 李岩, 等. 组合两步分解和ARIMA-LSTM的短期风速预测研究[J]. 太阳能学报, 2024, 45(2): 164-171.
CHEN H F, WANG H, LI Y, et al.Short-term wind speed prediction by combining two-step decomposition and ARIMA-LSTM[J]. Acta energiae solaris sinica, 2024, 45(2): 164-171.
[18] JIANG Z Y, CHE J X, WANG L N.Ultra-short-term wind speed forecasting based on EMD-VAR model and spatial correlation[J]. Energy conversion and management, 2021, 250: 114919.
[19] RODRIGUES MORENO S, GOMES DA SILVA R, COCCO MARIANI V, et al. Multi-step wind speed forecasting based on hybrid multi-stage decomposition model and long short-term memory neural network[J]. Energy conversion and management, 2020, 213: 112869.
[20] GENG D H, WANG B, GAO Q.A hybrid photovoltaic/wind power prediction model based on Time2Vec, WDCNN and BiLSTM[J]. Energy conversion and management, 2023, 291: 117342.
[21] KAZEMI S M, GOEL R, EGHBALI S, et al.Time2Vec: learning a vector representation of time[J]. Arxiv. 2019. arXiv:1907.05321
[22] WANG J, YANG Z S.Ultra-short-term wind speed forecasting using an optimized artificial intelligence algorithm[J]. Renewable energy, 2021, 171: 1418-1435.
[23] LIU G B, ZHOU J Z, JIA B J, et al.Advance short-term wind energy quality assessment based on instantaneous standard deviation and variogram of wind speed by a hybrid method[J]. Applied energy, 2019, 238: 643-667.
基金
云南省省市一体化专项（202202AH080009）
PDF(1785 KB)
