Wind speed inversion of high frequency radar based on artifical neural network
-
摘要: 风速是重要的海洋状态参数之一，对海面风速的准确提取是实现海洋环境监测和沿海工程应用的重要保证。目前，作为新兴海洋环境监测设备，高频雷达在风速提取方面仍然存在挑战。本文提出了一种基于人工神经网络的风速提取方法，利用历史浮标测量海态数据训练风速提取网络，实现风速与有效波高、波周期、风向及时间因素之间的非线性映射。测试结果表明了这一网络在时间和空间上的稳定性；进而将已训练的网络应用到便携式高频地波雷达OSMAR-S的风速反演中，得到的风速与浮标测量风速间的相关系数达到0.849，均方根误差为2.11 m/s。这一结果明显优于常规由浪高反演风速的SMB方法，验证了该方法在高频雷达风速反演中的可行性。Abstract: Wind speed is one of the important ocean state parameters. Accurate extraction of sea surface wind speed is an important guarantee for achieving marine environmental monitoring and coastal engineering applications. At present, as an emerging marine environment monitoring device, high frequency radar still has challenges in wind speed extraction. This paper proposes a wind speed extraction method based on artificial neural network which can be trained by historical sea state data measured by buoys to achieve non-linear mapping among wind and effective wave height, wave period, wind direction, and time. The test results show the stability of the trained network both in time and space and the trained network was applied to the wind speed inversion of the high frequency surface wave radar, OSMAR-S. The correlation coefficient between the inversion wind speed and the measured wind speed of the buoy reaches 0.849, and the root mean square error is 2.11 m/s. This result is significantly better than the conventional SMB method which inverts the wind speed from wave height, and verifies the feasibility of this method in high frequency radar wind speed inversion.
-
图 3 人工神经网络风速反演结果
a. ANN反演风速与浮标测量风速序列的对比；b. ANN反演风速与浮标测量风速散点图
Fig. 3 Wind speed inversion results of artificial neural network
a. Comparison of wind speed inverted by ANN and wind speed measured by buoys; b. scatter plots of wind speed inverted by ANN and wind speed measured by buoys
图 4 SMB风速反演结果
a. SMB反演风速与浮标测量风速序列的对比；b. SMB反演风速与浮标测量风速散点图
Fig. 4 Wind speed inversion results of SMB
a. Comparison of wind speed inverted by SMB and wind speed measured by buoys; b. scatter plots of wind speed inverted by SMB and wind speed measured by buoys
图 5 两种网络的风速反演均方根误差和相关系数对比
Fig. 5 Comparison of wind speed root mean square error and correlation coefficient inverted by two networks
图 7 C浮标处ANN风速反演结果
a. ANN反演风速与浮标测量风速序列的对比；b. ANN反演风速与浮标测量风速散点图
Fig. 7 Wind speed inversion result of ANN at C buoy
a. Comparison of wind speed inverted by ANN and wind speed measured by buoys; b. scatter plots of wind speed inverted by ANN and wind speed measured by buoys
-
[1] 王其松, 邓家泉, 刘诚, 等. 叠加风场在南海台风浪数值后报中的应用研究[J]. 海洋学报, 2017, 39(7): 70−79. doi: 10.3969/j.issn.0253-4193.2017.07.007Wang Qisong, Deng Jiaquan, Liu Cheng, et al. Application of superimposed wind fields to the hindcast modelling of typhoon-induced waves in the South China Sea[J]. Haiyang Xuebao, 2017, 39(7): 70−79. doi: 10.3969/j.issn.0253-4193.2017.07.007 [2] 郑静静, 刘桂梅, 高姗, 等. 风和径流量对长江口缺氧影响的数值模拟[J]. 海洋学报, 2018, 40(9): 1−17. doi: 10.3969/j.issn.0253-4193.2018.09.001Zheng Jingjing, Liu Guimei, Gao Shan, et al. The impact of river discharge and wind on the hypoxia off the Changjiang Estuary: A numerical modeling study[J]. Haiyang Xuebao, 2018, 40(9): 1−17. doi: 10.3969/j.issn.0253-4193.2018.09.001 [3] Barrick D E. Extraction of wave parameters from measured HF radar sea-echo Doppler spectra[J]. Radio Science, 1977, 12(3): 415−424. doi: 10.1029/RS012i003p00415 [4] Lipa B, Barrick D. Least-squares methods for the extraction of surface currents from CODAR crossed-loop data: application at ARSLOE[J]. IEEE Journal of Oceanic Engineering, 1983, 8(4): 226−253. doi: 10.1109/JOE.1983.1145578 [5] Zeng Yuming, Zhou Hao, Lai Yeping, et al. Wind-direction mapping with a modified wind spreading function by broad-beam high-frequency radar[J]. IEEE Geoscience and Remote Sensing Letters, 2018, 15(5): 679−683. doi: 10.1109/LGRS.2018.2809558 [6] Bretschneider C L. Forecasting relations for wave generation[J]. Look Lab Hawaii, 1970, 1(3): 31−34. [7] Kinsman B. Wind Waves: Their Generation and Propagation on the Ocean Surface[M]. New York: Dover, 1984. [8] Hasselmann K, Barnett T P, Bouws E, et al. Measurements of wind-wave growth and swell decay during the Joint North Sea Wave Project (JONSWAP)[R]. Hamburg: Deutsches Hydrographisches Institut, 1973. [9] Mathew T E, Deo M C. Inverse estimation of wind from the waves measured by high-frequency radar[J]. International Journal of Remote Sensing, 2012, 33(10): 2985−3003. doi: 10.1080/01431161.2011.624131 [10] Shen Wei, Gurgel K W, Voulgaris G, et al. Wind-speed inversion from HF radar first-order backscatter signal[J]. Ocean Dynamics, 2012, 62(1): 105−121. doi: 10.1007/s10236-011-0465-9 [11] Zeng Yuming, Zhou Hao, Roarty H, et al. Wind speed inversion in high frequency radar based on neural network[J]. International Journal of Antennas and Propagation, 2016, 2016: 2706521. [12] Haykin S. Neural Networks: A Comprehensive Foundation[M]. New York: Prentice Hall PTR, 1994. [13] Rumelhart D E, Hinton G E, Williams R J. Learning internal representations by error propagation[R]. California: California University San Diego La Jolla Institute for Cognitive Science, 1985. [14] Heron S F, Heron M L. A comparison of algorithms for extracting significant wave height from HF radar ocean backscatter spectra[J]. Journal of Atmospheric and Oceanic Technology, 1998, 15(5): 1157−1163. doi: 10.1175/1520-0426(1998)015<1157:ACOAFE>2.0.CO;2 -
下载:
点击查看大图
计量
- 文章访问数: 333
- HTML全文浏览量: 25
- PDF下载量: 193
- 被引次数: 0
