长短期记忆神经网络（LSTM）对风暴潮数值模拟的优化应用-海洋预报

长短期记忆神经网络（LSTM）对风暴潮数值模拟的优化应用

单位：1. 自然资源部海洋环境探测技术与应用重点实验室, 广东广州 510301;
2. 国家海洋局汕尾海洋环境监测中心站, 广东汕尾 516600;
3. 国家海洋局深圳海洋环境监测中心站, 广东深圳 518000

分类号：P731.23

出版年·卷·期（页码）：2024·41·第四期（1-10）

摘要：

利用长短期记忆神经网络和数值模式相结合的方法，设计了两套针对粤东遮浪海洋站点台风风暴潮增水的预报优化方案。与实测资料对比结果显示，长短期记忆神经网络方法可以显著改善数值模式模拟结果的准确性，最大增水和主振过程中增水后报结果的平均绝对误差、平均相对误差和平均改善幅度分别为 7.1 cm、8.2%、74% 和 16.1 cm、34.7%、33%。进一步分析表明，利用台风信息预测数值模拟结果的订正值可以有效改善神经网络方法的不稳定性，比直接预测风暴潮增水值更加准确、可靠。

Using a combination of Long Short-Term Memory (LSTM) neural network and numerical model, two sets of prediction schemes for typhoon storm surge at the Zhelang marine station in eastern Guangdong have been designed. Compared with the measured data, the LSTM neural network can significantly improve the accuracy of the numerical model results. The average absolute error, average relative error and average improvement amplitude of the prediction results for the maximum surge and the main oscillation process are 7.1 cm, 8.2%, 74% and 16.1 cm, 34.7%, 33%, respectively. Further analysis shows that predicting the corrected value of numerical results using typhoon information can effectively limit the instability of neural network, which is more accurate and reliable in comparison with predicting the storm surge level directly.

参考文献：

[1] 自然资源部. 2013—2022年中国海洋灾害公报[R]. 北京, 2013— 2022. Ministry of Natural Resources. Bulletin of China marine disaster from 2013 to 2022[R]. Beijing, 2013-2022.
[2] 刘秋兴, 董剑希, 于福江, 等. 覆盖中国沿海地区的精细化台风风暴潮模型的研究及适用[J]. 海洋学报, 2014, 36(11): 30-37. LIU Q X, DONG J X, YU F J, et al. A high-resolution typhoon storm surge forecast model covering the whole China's coastal areas and its application[J]. Acta Oceanologica Sinica, 2014, 36(11): 30-37.
[3] CHAO W T, YOUNG C C, HSU T W, et al. Long-lead-time prediction of storm surge using artificial neural networks and effective typhoon parameters: revisit and deeper insight[J]. Water, 2020, 12(9): 2394.
[4] HAQUE A U, MANDAL P, MENG J L, et al. Wind speed forecast model for wind farm based on a hybrid machine learning algorithm [J]. International Journal of Sustainable Energy, 2015, 34(1): 38- 51.
[5] KUMAR N K, SAVITHA R, AL MAMUN A. Regional ocean wave height prediction using sequential learning neural networks [J]. Ocean Engineering, 2017, 129: 605-612.
[6] 周水华, 洪晓, 梁昌霞, 等. 基于人工神经网络的台风浪高快速计算方法[J]. 热带海洋学报, 2020, 39(4): 25-33. ZHOU S H, HONG X, LIANG C X, et al. A method of tropical cyclone wave height calculation based on Artificial neural network [J]. Journal of Tropical Oceanography, 2020, 39(4): 25-33.
[7] ZHENG G, LI X F, ZHANG R H, et al. Purely satellite data-driven deep learning forecast of complicated tropical instability waves[J]. Science Advances, 2020, 6(29): eaba1482.
[8] 薛彦广, 沙文钰, 徐海斌, 等. 人工神经网络在风暴潮增水预报中的应用[J]. 海洋预报, 2005, 22(2): 33-37. XUE Y G, SHA W Y, XU H B, et al. Application of the artificial neural network in storm surge forecast[J]. Marine Forecasts, 2005, 22(2): 33-37.
[9] LEE T L. Back-propagation neural network for the prediction of the short-term storm surge in Taichung harbor, Taiwan[J]. Engineering Applications of Artificial Intelligence, 2008, 21(1): 63-72.
[10] CHEN W B, LIU W C, HSU M H. Predicting typhoon-induced storm surge tide with a two-dimensional hydrodynamic model and artificial neural network model[J]. Natural Hazards and Earth System Sciences, 2012, 12(12): 3799-3809.
[11] KIM S, MATSUMI Y, MASE H, et al. Development of real time storm surge forecasting using artificial neural network[C]// Proceedings of the 11th International Conference on Hydroscience & Engineering. 2014.
[12] KIM S W, MELBY J A, NADAL-CARABALLO N C, et al. A time-dependent surrogate model for storm surge prediction based on an artificial neural network using high-fidelity synthetic hurricane modeling[J]. Natural Hazards, 2015, 76(1): 565-585.
[13] KIM S, MATSUMI Y, PAN S Q, et al. A real-time forecast model using artificial neural network for after-runner storm surges on the Tottori coast, Japan[J]. Ocean Engineering, 2016, 122: 44-53.
[14] 卢君峰, 李少伟, 袁方超. 基于BP神经网络的厦门沿海风暴潮预报应用[J]. 海洋预报, 2016, 33(4): 9-16. LU J F, LI S W, YUAN F C. Application of storm surge forecasting by BP artificial neural network off coast of Xiamen [J]. Marine Forecasts, 2016, 33(4): 9-16.
[15] 张娟, 周水华, 黄宝霞, 等. 人工神经网络在台风风暴潮模拟中的解释应用[J]. 海洋预报, 2016, 33(2): 60-65. ZHANG J, ZHOU S H, HUANG B X, et al. Interpretation of numerical storm surge model results using the artificial neural network[J]. Marine Forecasts, 2016, 33(2): 60-65.
[16] SAHOO B, BHASKARAN P K. Prediction of storm surge and coastal inundation using Artificial Neural Network - A case study for 1999 Odisha Super Cyclone[J]. Weather and Climate Extremes, 2019, 23: 100196.
[17] 周寅杰, 刘强, 张晓琪. 基于TSA-BP模型的温州站台风风暴潮增水预测[J]. 海洋环境科学, 2022, 41(5): 807-812. ZHOU Y J, LIU Q, ZHANG X Q. Prediction of typhoon storm surge at Wenzhou station based on TSA-BP model[J]. Marine Environmental Science, 2022, 41(5): 807-812.
[18] 雷森, 史振威, 石天阳, 等. 基于递归神经网络的风暴潮增水预测[J]. 智能系统学报, 2017, 12(5): 640-644. LEI S, SHI Z W, SHI T Y, et al. Prediction of storm surge based on recurrent neural network[J]. CAAI Transactions on Intelligent Systems, 2017, 12(5): 640-644.
[19] 薛明, 李醒飞, 成方林. 基于多种神经网络的风暴潮增水预测方法的比较分析[J]. 海洋通报, 2019, 38(3): 290-295. XUE M, LI X F, CHENG F L. Comparative analysis of storm surge water prediction methods based on multiple neural networks [J]. Marine Science Bulletin, 2019, 38(3): 290-295.
[20] 刘媛媛, 张丽, 李磊, 等. 基于多变量LSTM神经网络模型的风暴潮临近预报[J]. 海洋通报, 2020, 39(6): 689-694. LIU Y Y, ZHANG L, LI L, et al. Storm surge nowcasting based on multivariable LSTM neural network model[J]. Marine Science Bulletin, 2020, 39(6): 689-694.
[21] WAIBEL A, HANAZAWA T, HINTON G, et al. Phoneme recognition using time-delay neural networks[J]. IEEE Transactions on Acoustics, Speech, and Signal Processing, 1989, 37(3): 328-339.
[22] HOCHREITER S, SCHMIDHUBER J. Long short-term memory [J]. Neural Computation, 1997, 9(8): 1735-1780.
[23] WANG B, LIU S C, WANG B, et al. Multi-step ahead short-term predictions of storm surge level using CNN and LSTM network [J]. Acta Oceanologica Sinica, 2021, 40(11): 104-118.
[24] 苗庆生, 徐珊珊, 杨锦坤, 等. 长短期记忆神经网络在厦门风暴潮预报中的应用[J]. 中国海洋大学学报, 2022, 52(9): 10-19. MIAO Q S, XU S S, YANG J K, et al. Application of long shortterm memory neural network in Xiamen storm surge forecast[J]. Periodical of Ocean University of China, 2022, 52(9): 10-19.
[25] BAJO M, UMGIESSER G. Storm surge forecast through a combination of dynamic and neural network models[J]. Ocean Modelling, 2010, 33(1-2): 1-9.
[26] WANG Q, CHEN J H, HU K L. Storm surge prediction for Louisiana coast using artificial neural networks[C]//Proceedings of the 23rd International Conference on Neural Information Processing. Kyoto: Springer, 2016: 396-405.
[27] CHEN C S, BEARDSLEY R C, COWLES G, et al. An unstructured grid, finite-volume community ocean model: FVCOM user manual[R]. 2013.
[28] 张余得, 商少平, 谢燕双, 等. 基于强风圈半径的台风风场模型[J]. 厦门大学学报(自然科学版), 2014, 53(2): 252-256. ZHANG Y D, SHANG S P, XIE Y S, et al. Typhoon wind field model based on the radii of wind circle[J]. Journal of Xiamen University (Natural Science), 2014, 53(2): 252-256.
[29] 林小刚, 罗荣真, 张娟, 等. 浪流耦合对汕尾港台风风暴潮模拟的影响[J]. 海洋预报, 2020, 37(4): 30-37. LIN X G, LUO R Z, ZHANG J, et al. Effects of wave-current interaction on storm surge simulation in Shanwei Port[J]. Marine Forecasts, 2020, 37(4): 30-37.
[30] KAREVAN Z, SUYKENS J A K. Transductive LSTM for timeseries prediction: an application to weather forecasting[J]. Neural Networks, 2020, 125: 1-9.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】