基于CNN-LSTM的高频地波雷达海流方向校正方法

doi:10.11978/2024176

Abstract

Abstract:

High-frequency surface wave radars often encounter interference from coastal topography and islands when detecting nearshore areas, leading to significant errors in the synthesized ocean current directions from radar data. Traditional high-frequency surface wave radar inversion algorithms do not account for the impact of these physical factors. To address this issue, leveraging the strengths of convolutional neural networks (CNN) and long short-term memory (LSTM) neural networks, this paper introduces a hybrid CNN-LSTM model. This model incorporates data on sea surface wind, tide, and elevation, allowing for the refinement of ocean current directions obtained from radar measurements. The experimental results show that the CNN-LSTM model, when combined with physical oceanographic factors, can effectively improve the quality of radar-detected ocean current data in areas affected by topography, significantly enhancing the accuracy of the synthesized ocean current direction. After model correction, compared to the original radar data, the directional angle error of the empirical orthogonal function ellipse decreased from 77.10° to 23.06°, the error of the ellipse’s major and minor axes decreased from 0.0896 to 0.0538, and the average flow directional angle error of the ocean currents decreased from 20.82° to 6.21°.

Key words: high-frequency surface wave radar, ocean current direction, convolutional neural network, long short-term memory neural network

CLC Number:

P714

XU Yi, WEI Jun, WEI Chunlei, YANG Fan. An ocean current direction correction method for high-frequency surface wave radars based on CNN-LSTM[J].Journal of Tropical Oceanography, 2025, 44(3): 24-35.

Figures/Tables 11

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Fig. 6

Fig. 7

Tab. 1

Tab. 2

Fig. 8

Tab. 3

References 24

[1]	蔡佳佳, 曾玉明, 周浩, 等, 2019. 基于人工神经网络的高频雷达风速反演[J]. 海洋学报, 41(11): 150-155.
	CAI JIAJIA, ZENG YUMING, ZHOU HAO, et al, 2019. Wind speed inversion of high frequency radar based on artificial neural network[J]. Haiyang Xuebao, 41(11): 150-155 (in Chinese with English abstract).
[2]	东松林, 岳显昌, 吴雄斌, 等, 2022. 基于深度学习的高频雷达射频干扰自动识别与抑制[J]. 雷达科学与技术, 20(3): 260-271.
	DONG SONGLIN, YUE XIANCHANG, WU XIONGBIN, et al, 2022. Automatic identification and suppression of radio frequency interference of HF radar based on deep learning[J]. Radar Science and Technology, 20(3): 260-271 (in Chinese with English abstract).
[3]	吴雄斌, 张兰, 柳剑飞, 2015. 海洋雷达探测技术综述[J]. 海洋技术学报, 34(3): 8-15.
	WU XIONGBIN, ZHANG LAN, LIU JIANFEI, 2015. Overview of detecting techniques using oceanographic radars[J]. Journal of Ocean Technology, 34(3): 8-15 (in Chinese with English abstract).
[4]	于彩彩, 楚晓亮, 王曙曜, 2024. 基于卷积神经网络的高频地波雷达有效波高反演[J]. 海洋科学进展, 42(1): 126-136.
	YU CAICAI, CHU XIAOLIANG, WANG SHUYAO, 2024. Significant wave height inversion of high frequency surface wave radar based on convolutional neural network[J]. Advances in Marine Science, 42(1): 126-136 (in Chinese with English abstract).
[5]	周东旭, 孙维康, 付海德, 等, 2023. 三种最新全球海潮模型在中国沿海的精度评估[J]. 海洋科学进展, 41(1): 54-63.
	ZHOU DONGXU, SUN WEIKANG, FU HAIDE, et al, 2023. Accuracy assessment of three latest global ocean tide models in coastal areas of China[J]. Advances in Marine Science, 41(1): 54-63 (in Chinese with English abstract).
[6]	BARRICK D E, EVANS M W, WEBER B L, 1977. Ocean surface currents mapped by radar[J]. Science, 198(4313): 138-144. pmid: 17755343
[7]	CROMBIE D D, 1955. Doppler spectrum of sea echo at 13. 56 Mc. /s[J]. Nature, 175(4459): 681-682.
[8]	CHEN CHANGSHENG, LIU HEDONG, BEARDSLEY R C, 2003. An unstructured grid, finite-volume, three-dimensional, primitive equations ocean model: application to coastal ocean and estuaries[J]. Journal of Atmospheric and Oceanic Technology, 20(1): 159-186.
[9]	CHEN CHANGSHENG, HUANG HAOSHENG, BEARDSLEY R C, et al, 2007. A finite volume numerical approach for coastal ocean circulation studies: Comparisons with finite difference models[J]. Journal of Geophysical Research: Oceans, 112(C3): C03018.
[10]	CHEN CHANGSHENG, GAO GUOPING, QI JIANHUA, et al, 2009. A new high-resolution unstructured grid finite volume Arctic Ocean model (AO-FVCOM): an application for tidal studies[J]. Journal of Geophysical Research: Oceans, 114(C8): C08017.
[11]	EGBERT G D, EROFEEVA S Y, 2002. Efficient inverse modeling of barotropic ocean tides[J]. Journal of Atmospheric and Oceanic Technology, 19(2): 183-204.
[12]	FUKUSHIMA K, 1980. Neocognitron: a self organizing neural network model for a mechanism of pattern recognition unaffected by shift in position[J]. Biological Cybernetics, 36(4): 193-202. doi: 10.1007/BF00344251 pmid: 7370364
[13]	GERS F A, SCHMIDHUBER J, CUMMINS F, 2000. Learning to forget: continual prediction with LSTM[J]. Neural Computation, 12(10): 2451-2471. doi: 10.1162/089976600300015015 pmid: 11032042
[14]	HOCHREITER S, SCHMIDHUBER J, 1997. Long short-term memory[J]. Neural Computation, 9(8): 1735-1780. doi: 10.1162/neco.1997.9.8.1735 pmid: 9377276
[15]	LECUN Y, BOTTOU L, BENGIO Y, et al, 1998. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 86(11): 2278-2324.
[16]	LIU YONGGANG, WEISBERG R H, MERZ C R, et al, 2010. HF radar performance in a low-energy environment: CODAR SeaSonde experience on the west Florida shelf[J]. Journal of Atmospheric and Oceanic Technology, 27(10): 1689-1710.
[17]	PADUAN J D, WASHBURN L, 2013. High-frequency radar observations of ocean surface currents[J]. Annual Review of Marine Science, 5: 115-136. doi: 10.1146/annurev-marine-121211-172315 pmid: 22809196
[18]	PEI JIFANG, YANG YU, WU ZEBIAO, et al, 2022. A sea clutter suppression method based on machine learning approach for marine surveillance radar[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 15: 3120-3130.
[19]	REN LEI, HU ZHAN, HARTNETT M, 2018. Short-term forecasting of coastal surface currents using high frequency radar data and artificial neural networks[J]. Remote Sensing, 10(6): 850.
[20]	SUN WEIKANG, ZHOU XINGHUA, ZHOU DONGXU, et al, 2022. Advances and accuracy assessment of ocean tide models in the Antarctic ocean[J]. Frontiers in Earth Science, 10: 757821.
[21]	YANG ZHIQING, LAI YEPING, ZHOU HAO, et al, 2023. Improving ship detection based on decision tree classification for high frequency surface wave radar[J]. Journal of Marine Science and Engineering, 11(3): 493.
[22]	ZHOU CHUNYE, WEI CHUNLEI, YANG FAN, et al, 2023. A quality control method for high frequency radar data based on machine learning neural networks[J]. Applied Sciences, 13(21): 11826.
[23]	ZHU LANGFENG, LU TIANYI, YANG FAN, et al, 2022. Comparisons of tidal currents in the Pearl River Estuary between high-frequency radar data and model simulations[J]. Applied Sciences, 12(13): 6509.
[24]	ZHU LANGFENG, YANG FAN, YANG YUFAN, et al, 2023. Designing theoretical shipborne ADCP survey trajectories for high-frequency radar based on a machine learning neural network[J]. Applied Sciences, 13(12): 7208.

卷积层数量	隐藏层数量	椭圆长短轴误差	椭圆方向角误差	平均流方向角误差	三个指标平均提升幅度
1	80	0.0767	25.31°	8.66°	46.65%
1	160	0.0775	25.65°	8.79°	45.99%
1	240	0.0766	25.20°	8.81°	46.49%
2	80	0.0558	24.67°	6.14°	58.73%
2	160	0.0538	23.06°	6.21°	60.06%
2	240	0.0555	24.13°	6.83°	57.97%
4	80	0.0535	23.61°	8.52°	56.24%
4	160	0.0500	23.79°	8.41°	57.64%
4	240	0.0553	23.33°	8.38°	55.91%

卷积核大小	失活率	椭圆长短轴误差	椭圆方向角误差	平均流方向角误差	三个指标平均提升幅度
2	0	0.0624	24.06°	7.47°	54.41%
3	0	0.0538	23.06°	6.21°	60.06%
4	0	0.0505	23.79°	7.29°	59.24%
3	0.25	0.0593	23.17°	9.31°	53.00%
3	0.5	0.0708	24.57°	9.87°	47.22%

实验	椭圆长短轴误差	椭圆方向角误差	平均流方向角误差
实验1	0.0540	35.57°	8.97°
实验2	0.0589	26.32°	10.29°
实验3	0.0629	28.18°	10.38°
实验4	0.0538	23.06°	6.21°
原始数据	0.0896	77.01°	20.82°

An ocean current direction correction method for high-frequency surface wave radars based on CNN-LSTM

RichHTML

PDF (PC)

Like

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 11

References 24

Related Articles 2

Metrics

Comments

Recommended 0

[1]	WANG Aoyu, QU Ke, WANG Xu, GAO Rongze, MEN Jia. The hydrodynamic characteristics prediction of landslide-induced surge waves based on non-hydrostatic model numerical simulation and convolutional neural network [J]. Journal of Tropical Oceanography, 2025, 44(2): 187-195.
[2]	GAO Rongze, QU Ke, REN Xingyue, WANG Xu. Application of convolutional neural network methods in the evolution of hydrodynamic characteristics of tsunamis like-wave over fringing reef [J]. Journal of Tropical Oceanography, 2024, 43(4): 68-75.