Forecast of maximum water increase in typhoon storm surge based on random forest model

doi:10.13474/j.cnki.11-2246.2021.375

Abstract

Abstract: Using 98 typhoon storm surges that landed in Guangdong, Fujian, and Zhejiang provinces from 1989 to 2018.Seven input factors of minimum central pressure, maximum wind speed, central pressure at landing, wind speed at landing, typhoon intensity at landing, landing location, and landing direction are selected, and a random forest regression model is used to predict the maximum water increase duing to typhoon storm surge, and it is verified with real data. The model is applied to the prediction of typhoon storm surges in the coming years, and the practicability of the model is illustrated by taking Typhoon Lichma in 2019 as an example. The maximum water increase at tide gauge stations simulated by Mike21 software is used to construct a prediction model for the maximum water increase at tide gauge stations, which is planned to be applied to the prediction of the maximum water increase at the port station and Beijin station in Guangdong province. The results show that it is feasible.

Key words: random forest, typhoon storm surge, maximum water increase, tide gauge station

CLC Number:

P208

ZHU Peijing, LUO Nianxue, ZHAO Qiansheng. Forecast of maximum water increase in typhoon storm surge based on random forest model[J]. Bulletin of Surveying and Mapping, 2021, 0(12): 71-74,82.

References

[1] 高佳,潘嵩,王慧,等.基于ADCIRC+SWAN耦合模型的风暴潮数值模拟研究——以深圳西部海域为例[J].海洋通报,2019,38(6):662-674.
[2] 沙文钰.风暴潮、浪数值预报[M].北京:海洋出版社,2004.
[3] 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.
[4] 蔡煜东,姚林声.预报台风风暴潮极值的人工神经网络方法[J].海洋预报,1994,11(1):77-81.
[5] 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.
[6] XU S D, HUANG W R. Artificial neural network model application on long term water level predictions of south Florida coastal waters[C]//Proceeding of 2010 International Conference on Computational Intelligence and Software Engineering. Wuhan:IEEE, 2010:1-4.
[7] 王如云,雷磊,占飞,等.基于高低潮的风暴增水人工神经网络预报模型[J].海洋预报,2014,31(6):23-27.
[8] 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.
[9] 李侠男. 基于随机森林算法的房地产项目风险评价研究[D].天津:天津大学,2017.
[10] 李丽. 基于随机森林算法的企业信用风险评价研究——以我国电力生产行业上市公司为例[D].成都:西南财经大学,2012.
[11] 罗庆. Research progress on application of MIKE21 hydro-dynamic model[J]. Advances in Envionmental Protection, 2020, 10(4):510-515.
[12] 许婷.MIKE21 HD计算原理及应用实例[J].港工技术,2010,47(5):1-5.
[13] 刘晓琴,刘国龙,王振.MIKE系列模型在蓄滞洪区洪水模拟中的应用研究[J].中国农村水利水电,2020(6):10-15,20.
[14] 孙玲玲.基于MIKE21的水库洪水期洪水演进数值模拟[J].工程技术研究,2020,5(11):246-248.
[15] YANG L H, LI J Z, KANG A Q, et al. The effect of nonstationarity in rainfall on urban flooding based on coupling SWMM and MIKE21[J]. Water Resources Management, 2020, 34:1535-1551.