Paper Infomation
Offshore Wind Field Model Data Revision Based on U-Net
Full Text(PDF, 499KB)
Author: Jianbo Zhang, Yanxin Yao, Jiaxin Liu
Abstract: The production and daily life in coastal areas of our country are greatly affected by offshore wind fields. Currently, high-resolution gridded marine data are modeled data, primarily derived from physical numerical models, while the modeled data are sensitive to initial conditions and cannot accurately describe wind speed information near the ground and sea surface. It is necessary to consider using real-time reference data from buoys, ships, and other sources for correction. However, traditional correction methods can only correct grid values near the reference data points. Therefore, in this study, a data correction algorithm based on U-Net was proposed for wind speed correction. It uses site reference data to correct the entire target marine area based on U-Net, reducing wind speed errors after correction. Experimental results show that the Root Mean Square Error (RMSE) of 10 m wind speed in the marine target area is 1.96 m/s. This is a 16.8% reduction compared to the average error of model data. The U-Net model can effectively correct model data, providing high-quality data support for subsequent data fusion studies.
Keywords: Sea Surface Wind Field, Wind Speed Revision, Deep Learning, U-Net
References:
[1] Chou J M, Dong W J, Xu H, et al. Assessment on Economic Losses of Tropical Cyclone Disasters Affecting China’s Coastal Areas[J]. Meteorological and Environmental Sciences, 2022, 45(3): 1-10
[2] Chen L X, Xu Y L. Review of Typhoon Very Heavy Rainfall in China[J]. Meteorological and Environmental Sciences, 2017, 40(1): 3-10.
[3] Wang G S, Wang X D, Hou M, et al. Research on application of LSTM deep neural network on historical observation data and reanalysis data for sea surface wind speed forecasting[J]. Haiyang Xuebao, 2020, 42(1): 67-77
[4] Tian S, Zhang J, Chen L, et al. Random Sampling-Arithmetic Mean: A Simple Method of Meteorological Data Quality Control Based on Random Observation Thought[J]. IEEE Access, 2020, vol.8: 226999~227013
[5] 李紫丁.基于深度学习的气象要素格点预报模式订正[D].湖南:湖南师范大学,2019
[6] Wang Y. Visualization Analysis of Meteorological Big Data through Deep Learning and Network Model[C]. 2021 IEEE 3rd International Conference on Civil Aviation Safety and Information Technology (ICCASIT). Changsha, China: 2021, 601-604
[7] Jiang B R, Xiao D B, Zhang Y, et al. High Precision Air Data Estimation Algorithm with Meteorological Observation[J]. Computer Measurement & Control, 2022, 30(12): 276-283
[8] Shao C, Li H, Fang J, et al. An Error Location and Correction Method for Memory Based on Data Similarity Analysis[J] IEEE Transactions on Very Large Scale Integration (VLSI) Systems,2019, vol.27, no.10:2354-2364
[9] Ma X L, Shi Y, He J, et al. The combined descending averaging bias correction based on the Kalman filter for ensemble forecast[J]. Acta Meteorologica Sinica,2015,(5): 952-964
[10] Lv Q P, Luo J, Zhu K, et al. Experiments on Predicting Tracks of Tropical Cyclones with Climatic Persistence Method based on Support Vector Machine[J]. Marine Forecasts, 2009, 26(1): 76-83
[11] Xu L B, Wang L X, Wang C Y, et al. Research on correction method of marine environment prediction based on machine learning[J]. Marine Science Bulletin, 2020, 39(6): 695-704
[12] 刘雅忱.人工智能下深度学习在气象预报中应用综述[J].计算机产品与流通,2020,11:121+135
[13] 纪富鑫.基于深度学习的短时气象预报研究[D].合肥:安徽大学,2021
[14] Lv Q, Ding J. Study on the correction technology of tide level forecast based on deep learning[J]. Marine Forecasts, 2022, 39(2): 70-79
[15] Zhao H S, Jin L, Huang X Y, et al. ECMWF Precipitation Grading Correction Forecast Method Based on CNN and RF Algorithm[J]. Meteorological Science and Technology, 2021, 49(03): 419-426
[16] Mao K Y, Zhao C M, He J. A Research for 10 m Wind Speed Prediction based on XGBoost[J]. Journal Of Chengdu University Of Information Technology, 2020, 35(6): 604-609
[17] Zhang Y B, Song L Y, Chen M X, et al. A study of error correction for high-resolution gridded forecast based on a convolutional neural network in the Beijing-Tianjin-Hebei Region[J]. Trans Atmos Sci, 2022, 45(6): 850-862
[18] Olaf Ronneberger, Philipp Fischer, Thomas Brox. U-Net: Convolutional Networks for Biomedical Image Segmentation[J]. Medical Image Computing and Computer-Assisted Intervention – MICCAI. 2015, Vol.9351
[19] Zhang W L, Kang L, Zhou W, et al. Extreme Short-time Heavy Precipitation Forecast Based on GRAPES-MESO Model[J]. Journal of Arid Meteorology, 2021,3 9(03): 507-513