基于多模型平差接力的长时序风暴增水预测方法-海洋预报

基于多模型平差接力的长时序风暴增水预测方法

单位：1. 山东科技大学测绘与空间信息学院, 山东青岛 266590;
2. 自然资源部北海预报减灾中心, 山东青岛 266061

分类号：P731.34

出版年·卷·期（页码）：2023·40·第二期（32-44）

摘要：

利用神经网络模型预测未来风暴增水时，预测精度会随预测时序的延伸不断降低。基于长短期记忆（Long Short-Term Memory）神经网络模型，以风速、风向、气压和前时序的风暴增水数据作为模型输入，利用多个模型接力预测风暴增水时间序列，减小误差随模型的迭代累积，建立基于多模型平差接力的长时序风暴增水预测方法。对比不同地点和不同台风下的风暴增水预测分析结果，模型在渤黄海北部区域每月的均方根误差为4～7 cm，在黄海中部区域可控制在10 cm以内，能够较准确预测未来24 h的风暴增水。

When predicting storm surge using classical neural network models, the prediction accuracy will continue to decrease with the extension of the time series. To alleviate this problem, we proposed a method of long-term storm surge prediction based on multi-model gradual adjustment. The method is based on the Long Short-Term Memory neural network model, and integrates multiple models to predict the storm surge time series. Data from previous time periods were fed into the model, such as wind speed, wind direction, air pressure, and storm surge data, and the errors accumulated with model iterations were considered. According to the forecast analysis of storm surge in different locations and under different typhoons, the monthly root mean square error of the prediction model is between 4 and 7 cm in the northern part of the Bohai sea and Yellow Sea, and is less than 10 cm in the central part of the Yellow Sea, which indicate the model can predict storm surge in the next 24 hours accurately.

参考文献：

[1] JIN Y D, ZHAO X, MA L N. Storm surge characteristics and extreme parameters in the Chengshantou sea area[J]. IOP Conference Series: Earth and Environmental Science, 2017, 82: 012052.
[2] 谭家辉, 潘毅, 于普兵, 等. 江苏沿海台风暴潮过程中的波浪增水分布研究[J]. 海洋预报, 2022, 39(1): 48-55. TAN J H, PAN Y, YU P B, et al. Study on the distribution of waveinduced setup along the coast of Jiangsu Province during typhoon processes[J]. Marine Forecasts, 2022, 39(1): 48-55.
[3] 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.
[4] 黄宝霞, 胡金磊, 郑淑贤, 等. 广东省徐闻沿海风暴潮数值模拟与淹没危险性分析[J]. 海洋预报, 2022, 39(1): 39-47. HUANG B X, HU J L, ZHENG S X, et al. Numerical simulation and inundation risk assessment of storm surge along the coastal areas of Xuwen, Guangdong Province[J]. Marine Forecasts, 2022, 39(1): 39-47.
[5] LIU Q R, RUAN C Q, ZHONG S, et al. Risk assessment of storm surge disaster based on numerical models and remote sensing[J]. International Journal of Applied Earth Observation and Geoinformation, 2018, 68: 20-30.
[6] 朱佩京, 罗年学, 赵前胜. 基于随机森林模型的台风风暴潮最大增水预测[J]. 测绘通报, 2021(12): 71-74, 82. ZHU P J, LUO N X, ZHAO Q S. Forecast of maximum water increase in typhoon storm surge based on random forest model[J]. Bulletin of Surveying and Mapping, 2021(12): 71-74, 82.
[7] 高强, 谷文凯, 林亚琼, 等. 山东省风暴潮灾害经济损失预测—— 基于灰色—周期外延组合模型[J]. 海洋经济, 2016, 6(1): 46-51. GAO Q, GU W K, LIN Y Q, et al. The financial loss prediction of storm surge in Shandong province based on the gray-periodic extensional combinatorial model[J]. Marine Economy, 2016, 6(1): 46-51.
[8] 李程, 王慧, 李响, 等. 近10年福建省风暴潮灾害特征分析及社会经济影响[J]. 海洋经济, 2019, 9(1): 43-46. LI C, WANG H, LI X, et al. Characteristics of storm surge disasters along Fujian coast in recent 10 Years[J]. Marine Economy, 2019, 9(1): 43-46.
[9] 林林茂. 基于ADCIRC模式的台风风暴潮数值模拟与危险性分析[D]. 哈尔滨: 哈尔滨工业大学, 2017. LIN L M. Numerical simulation and hazard analysis of typhoon storm surge based on ADCIRC model[D]. Harbin: Harbin Institute of Technology, 2017.
[10] HAN L, CHEN M X, CHEN K K, et al. A deep learning method for bias correction of ECMWF 24-240 h forecasts[J]. Advances in Atmospheric Sciences, 2021, 38(9): 1444-1459.
[11] 郝婧, 刘强. 基于SSA-ELM模型的台风风暴潮灾害损失预评估[J]. 海洋科学, 2022, 46(2): 55-63. HAO J, LIU Q. Pre-assessment of typhoon storm surge disaster loss based on the SSA-ELM model[J]. Marine Sciences, 2022, 46(2): 55-63.
[12] 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.
[13] 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.
[14] 范书鸣, 储鏖, 蒋勤. 再分析风场修正及其在渤海湾典型温带风暴潮模拟中的应用[J]. 海洋预报, 2021, 38(4): 61-68. FAN S M, CHU A, JIANG Q. Optimization of reanalysis wind field and its application in the simulation of extratropical storm surge in the Bohai Bay[J]. Marine Forecasts, 2021, 38(4): 61-68.
[15] CHEN K, KUANG C P, WANG L, et al. Storm surge prediction based on long short-term memory neural network in the East China Sea[J]. Applied Sciences, 2022, 12(1): 181.
[16] CHEN S, GE L. Exploring the attention mechanism in LSTMbased Hong Kong stock price movement prediction[J]. Quantitative Finance, 2019, 19(9): 1507-1515.
[17] LAM M W Y, CHEN X, HU S K, et al. Gaussian process Lstm recurrent neural network language models for speech recognition [C]//2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Brighton: IEEE, 2019: 7235- 7239.
[18] YU Y, SI X S, HU C H, et al. A review of recurrent neural networks: LSTM cells and network architectures[J]. Neural Computation, 2019, 31(7): 1235-1270.
[19] TIGGELOVEN T, COUASNON A, VAN STRAATEN C, et al. Exploring deep learning capabilities for surge predictions in coastal areas[J]. Scientific Reports, 2021, 11(1): 17224.
[20] ISHIDA K, TSUJIMOTO G, ERCAN A, et al. Hourly-scale coastal sea level modeling in a changing climate using long shortterm memory neural network[J]. Science of the Total Environment, 2020, 720: 137613.
[21] 吕忻, 丁骏. 基于深度学习的潮位预报订正技术研究[J]. 海洋预报, 2022, 39(2): 70-79. LYU X, DING J. Study on the correction technology of tide level forecast based on deep learning[J]. Marine Forecasts, 2022, 39(2): 70-79.
[22] CHEN D W, ZHANG J H, JIANG S X. Forecasting the short-term metro ridership with seasonal and trend decomposition using loess and LSTM neural networks[J]. IEEE Access, 2020, 8: 91181-91187.
[23] 余璇, 石绥祥, 徐凌宇, 等. 基于深度学习的赤潮发生预报方法研究[J]. 海洋通报, 2021, 40(5): 566-577. YU X, SHI S X, XU L Y, et al. Research on red tide occurrence forecast based on deep learning[J]. Marine Science Bulletin, 2021, 40(5): 566-577.
[24] 董绍江, 裴雪武, 吴文亮, 等. 基于多层降噪技术及改进卷积神经网络的滚动轴承故障诊断方法[J]. 机械工程学报, 2021, 57(1): 148-156. DONG S J, PEI X W, WU W L, et al. Rolling bearing fault diagnosis method based on multilayer noise reduction technology and improved convolutional neural network[J]. Journal of Mechanical Engineering, 2021, 57(1): 148-156.
[25] 李杰, 谭晓璇, 牛福新, 等. 天津沿海风暴潮特征及其与环境要素的关系[J]. 天津航海, 2021(1): 50-54. LI J, TAN X J, NIU F X, et al. The features of Tianjin coastal storm surge and its relations with the environmental elements[J]. Tianjin of Navigation, 2021(1): 50-54.
[26] JIA F, YANG B L. Forecasting volatility of stock index: deep learning model with likelihood-based loss function[J]. Complexity, 2021, 2021: 5511802.
[27] 代小红, 陈华江, 朱超平. 一种基于改进Faster RCNN的金属材料工件表面缺陷检测与实现研究[J]. 表面技术, 2020, 49(10): 362-371. DAI X H, CHEN H J, ZHU C P. Surface defect detection and realization of metal workpiece based on improved faster RCNN [J]. Surface Technology, 2020, 49(10): 362-371.
[28] VAN ORMONDT M, VAN DONGEREN A, ROELVINK D. A semi-empirical method for computing storm surges on open coasts during tropical cyclones[J]. Coastal Engineering, 2021, 165: 103839.
[29] 杨万康, 杨青莹, 伊小飞, 等. 典型海湾风暴潮增水特征与机理研究[J]. 海洋通报, 2017, 36(5): 532-537. YANG W K, YANG Q Y, YI X F, et al. Characteristics and mechanism of storm surge in typical bay[J]. Marine Science Bulletin, 2017, 36(5): 532-537.
[30] ENRÍQUEZ A R, WAHL T, MARCOS M, et al. Spatial footprints of storm surges along the global coastlines[J]. Journal of Geophysical Research: Oceans, 2020, 125(9): e2020JC016367.
[31] JI T, LI G S, LIU Y X Y, et al. Spatiotemporal features of storm surge activity and its response to climate change in the Southeastern coastal area of China in the Past 60 years[J]. Journal of Geophysical Research: Atmospheres, 2021, 126(4): e2020JD033234.
[32] HEO K Y, HA T, PARK K S. The effects of a typhoon-induced oceanic cold wake on typhoon intensity and typhoon-induced ocean waves[J]. Journal of Hydro-Environment Research, 2017, 14: 61-75.
[33] 李健, 刘清容, 阮成卿, 等. 南海风暴潮对风场和外围海水敏感性研究[J]. 海洋科学, 2018, 42(2): 99-107. LI J, LIU Q R, RUAN C Q, et al. Study of storm surge sensitivity with typhoon and external seawater in the South China Sea[J]. Marine Sciences, 2018, 42(2): 99-107.
[34] FENG X R, LI M J, YIN B S, et al. Study of storm surge trends in typhoon-prone coastal areas based on observations and surgewave coupled simulations[J]. International Journal of Applied Earth Observation and Geoinformation, 2018, 68: 272-278.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】