RESEARCH ON WIND FARM WIND MEASUREMENT DATA INTERPOLATION METHOD BASED ON HYBRID NEURAL NETWORK

Xing Zuoxia; Chou Jiaming; Guo Shanshan; Chen Mingyang; Chen Liang; Liu Yang

doi:10.19912/j.0254-0096.tynxb.2024-0029

PDF(1430 KB)

Acta Energiae Solaris Sinica ›› 2025, Vol. 46 ›› Issue (5) : 458-464. DOI: 10.19912/j.0254-0096.tynxb.2024-0029

RESEARCH ON WIND FARM WIND MEASUREMENT DATA INTERPOLATION METHOD BASED ON HYBRID NEURAL NETWORK

Xing Zuoxia¹, Chou Jiaming¹, Guo Shanshan¹, Chen Mingyang¹, Chen Liang², Liu Yang¹

Author information +

History +

Abstract

This paper presents a wind measurement data interpolation model for wind farms based on a hybrid neural network, the hyperparameters of this model （CNN-LSTM-SA） are optimized using the PSO-GWO optimization algorithm, and then the wind measurement data is interpolated. First, the wind measurement data at two adjacent heights under the height to be interpolated, the mesoscale data, and the wind speed data at other time periods at the height to be interpolated are selected to establish a regression model with "three features and one target data." Then, the target interpolated data is predicted using this model to achieve the purpose of interpolation. In this paper, the wind measurement data of a certain wind farm in Liaoning is used for simulation verification. The simulation results show that the method has an NMSE error of 0.0021 and a power generation of 1143732 kWh, which is both superior to the interpolation results of commonly used methods in engineering. It has certain reference significance for the practical work of the project.

Key words

wind farm / wind resource assessment / interpolation / neural network / optimization algorithm / hyperparameters

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

Xing Zuoxia, Chou Jiaming, Guo Shanshan, Chen Mingyang, Chen Liang, Liu Yang. RESEARCH ON WIND FARM WIND MEASUREMENT DATA INTERPOLATION METHOD BASED ON HYBRID NEURAL NETWORK[J]. Acta Energiae Solaris Sinica. 2025, 46(5): 458-464 https://doi.org/10.19912/j.0254-0096.tynxb.2024-0029

References

[1] NB/T 31147—2018, 风电场工程风能资源测量与评估技术规范(2018版)[S].
NB/T 31147—2018, Technical specification for measurement and evaluation of wind energy resources in wind farm projects(2018edition)[S].
[2] 张雪婷, 陈正洪, 许杨, 等. 复杂山地下测风塔缺失测风数据插补订正方法的比较分析[J]. 风能, 2015(1): 82-86.
ZHANG X T, CHENG Z H, XU Y, et al.Comparative analysis of interpolation and correction methods for missing wind data of anemometer tower under complex mountain area[J]. Wind energy, 2015(1): 82-86.
[3] ROGERS A L, ROGERSJ W, MANWELL J F.Comparison of the performance of four measure-correlate-predict algorithms[J]. Journal of wind engineering &industrial aerodynamics, 2005, 93(3): 243-264.
[4] CHEN XI, WANG BIN, YU MIN, et al.The interpolation of missing wind speed data based on optimized LSSVM model[A]//Power Electronics and Motion Control Conference. Hefei, China, 2016: 1448-1451.
[5] 高洁, 胡已坤, 刘雨佳, 等. 再分析资料在风速数据插补延长中的应用方法研究[J]. 风能, 2019(7): 62-68.
GAO J, HU Y K, LIU Y J, et al.Research on the application method of reanalysis data in interpolation and extension of wind speed data[J]. Wind energy, 2019(7): 62-68.
[6] 尹旭, 利昭仪, 朱雨露, 等. 风电场测风数据插补的不确定度研究[J]. 可再生能源, 2023, 41(5): 625-630.
YIN X, LI Z Y, ZHU Y L, et al.Study on uncertainty of Wind data interpolation of wind farm[J]. Renewable energy resources, 2023, 41(5): 625-630.
[7] 于佳鹤, 崔杰, 王风涛. 风电场测风数据插补方法研究及插补结果偏差分析[J]. 太阳能, 2021(2): 26-35.
YU J H, CUI J, WANG F T.Research on interpolation method of wind data in wind farm and deviation analysis of interpolation results[J]. Solar energy, 2021(2): 26-35.
[8] 付淑仪. 不同测风时长的数据插补对风能参数的影响研究[J]. 风能, 2020(4): 54-58.
FU S Y.Study on the influence of data interpolation with different wind measuring time on wind energy parameters[J]. Wind energy, 2020(4): 54-58.
[9] 刘崇波, 文雯, 盛科. 风电场测风数据插补方法适用性研究[J]. 风能, 2021(7): 64-70.
LIU C B, WEN W, SHENG K.Study on the applicability of interpolation method for wind measurement data of wind farm[J]. Wind energy, 2021(7): 64-70.
[10] GRUBER K, KLOCKL C, REGNER P, et al.Assessing the Global Wind Atlas and local measurements for bias correction of wind power generation simulated from MERRA-2 in Brazil[J]. Energy, 2019, 189: 116212.
[11] KHATIBI A, KRAUTER S.Validation and performance of satellite meteorological dataset MERRA-2 for solar and wind applications[J]. Energies, 2021, 14(4): 882.
[12] ASIM M, HUSSAIN A, HAYAT N, et al.Comparing MERRA and MERRA-2 reanalysis datasets with mast measured wind data for Karachi ,Pakistan[J]. Pakistan Journal of engineering and applied sciences , 2020, 26(1): 55-62.
[13] 陈卓, 郭军红, 亢朋朋, 等. 5套再分析资料在阿勒泰地区风资源评估中的适用性研究[J]. 太阳能学报, 2023, 44(4): 60-66.
CHEN Z, GUO J H, KANG P P, et al.Applicability of five sets of reanalysis data in wind resource assessmeant in altay prefecture[J]. Acta energiae solaris sinica, 2023, 44(4): 60-66.
[14] 余建星, 傅一钦, 余杨, 等. 渤海地区双参数风资源评估方法研究[J]. 太阳能学报, 2021, 42(7): 325-332.
YU J X, FU Y Q, YU Y, et al.Research on two-parameters wind resource assessment methods in bohai sea.[J]. Acta energiae solaris sinica, 2021, 42(7): 325-332.
[15] 徐昊, 王永生, 许志伟, 等. 基于生成对抗网络多变量风电时间序列异常值处理[J]. 太阳能学报, 2022, 43(12): 300-311.
XU H, WANG Y S, XU Z W, et al.Outlier processing of multivariable wind power time series based on generative adversarial network[j]. Acta energiae solaris sinica, 2022, 43(12): 300-311.
[16] 李辰, 李建勋. 卷积神经网络的正交性特征提取方法及其应用[J]. 上海交通大学学报, 2021, 55(10): 1320-1329.
LI C, LI J X.Orthogonal feature extraction method of convolutional neural network and its application[J]. Journal of Shanghai Jiaotong University, 2021, 55(10): 1320-1329.