EUVIMEDEUVIMED entdecken
Crossref

In Situ Wind Speed Measurement Correction for Buoy-Mounted Mechanical Anemometers: A Data-Driven Postprocessing Framework

Weiting Song, Yunzhou Li, Tongyu Xie, Yuzhe Xu, Kunwei Lin, Kaixi Zhou, Wenhai Wang, Xiao-Yu Tang, Juncheng Wang

IEEE Transactions on Instrumentation and Measurement · 2026 · 75 · 3004515-3004515

DOI: 10.1109/tim.2026.3742077 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Instrumentation and Measurement, 2026. Verfasst von Weiting Song und 8 weiteren Autor:innen. Die Arbeit nennt 53 Literaturangaben. Gefördert von: Natural Science Foundation of Zhejiang Province, China, State Key Laboratory of Industrial Control Technology, China, Fundamental Research Funds for the Central Universities.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Ocean data buoys are a critical means of measuring offshore wind speed. However, wind speed measurements obtained from buoys are influenced by platform motion, resulting in measurement errors. Critical applications like air-sea flux calculations and offshore wind turbine design demand reliable wind-speed measurements with sufficient temporal resolution. However, most buoy-measured wind data are provided as short-term averages, and high-precision correction methods are scarce for the commonly used mechanical wind anemometers, compromising the data’s utility. In this study, we introduce a specialized correction framework for buoy-measured wind speed utilizing a CNN–Wavelet–long short-term memory (LSTM)–Attention architecture with frequency attention. The dual-branch model processes features by periodicity: after augmentation, the first branch uses CNN–Wavelet–LSTM with feature-temporal attention to capture patterns in weakly periodic data; the second employs frequency attention for pronounced periodicity. The outputs of both branches are then fused to generate the final result, and kernel density estimation (KDE) is employed to establish uncertainty intervals. Validated on two in situ datasets from the Maoming sea area, our method substantially reduced measurement errors. For the first dataset, root-mean-squared error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) improved from 0.715 m/s, 0.576 m/s, and 7.308 % to 0.351 m/s, 0.275 m/s, and 3.479 %; for the second, they improved from 0.863 m/s, 0.641 m/s, and 18.080 % to 0.443 m/s, 0.290 m/s, and 7.920 %.

Förderung

  • Natural Science Foundation of Zhejiang Province, China
  • State Key Laboratory of Industrial Control Technology, China
  • Fundamental Research Funds for the Central Universities
  • National Key Research and Development Program of China
  • Key Research and Development Program of Shandong Province, China

Zitieren

APA

Song, W., Li, Y., Xie, T., Xu, Y., Lin, K., Zhou, K., Wang, W., Tang, X., & Wang, J. (2026). In Situ Wind Speed Measurement Correction for Buoy-Mounted Mechanical Anemometers: A Data-Driven Postprocessing Framework. IEEE Transactions on Instrumentation and Measurement, 75, 3004515-3004515. https://doi.org/10.1109/tim.2026.3742077

Vancouver

Song W, Li Y, Xie T, Xu Y, Lin K, Zhou K, et al. In Situ Wind Speed Measurement Correction for Buoy-Mounted Mechanical Anemometers: A Data-Driven Postprocessing Framework. IEEE Transactions on Instrumentation and Measurement. 2026;75:3004515-3004515. doi:10.1109/tim.2026.3742077

BibTeX
@article{Song2026,
  author = {Song, Weiting and Li, Yunzhou and Xie, Tongyu and Xu, Yuzhe and Lin, Kunwei and Zhou, Kaixi and Wang, Wenhai and Tang, Xiao-Yu and Wang, Juncheng},
  title = {In Situ Wind Speed Measurement Correction for Buoy-Mounted Mechanical Anemometers: A Data-Driven Postprocessing Framework},
  journal = {IEEE Transactions on Instrumentation and Measurement},
  year = {2026},
  volume = {75},
  pages = {3004515--3004515},
  doi = {10.1109/tim.2026.3742077}
}
Kostenlos starten →