基于非静压模型数值模拟与卷积神经网络的滑坡涌浪水动力特性预测

doi:10.11978/2024071

热带海洋学报 ›› 2025, Vol. 44 ›› Issue (2): 187-195.doi: 10.11978/2024071CSTR: 32234.14.2024071

基于非静压模型数值模拟与卷积神经网络的滑坡涌浪水动力特性预测

王傲宇¹(), 屈科¹^,²^,³(), 王旭¹, 高榕泽¹, 门佳¹

1.长沙理工大学水利与环境工程学院, 湖南长沙 410114
2.洞庭湖水环境治理与生态修复湖南省重点实验室, 湖南长沙 410114
3.水沙科学与水灾害防治湖南省重点实验室, 湖南长沙 410114

收稿日期:2024-03-27 修回日期:2024-05-30 出版日期:2025-03-10 发布日期:2025-04-11
通讯作者: 屈科
作者简介:
王傲宇(1999—), 男, 湖南省常德市人, 硕士研究生, 主要从事深度学习及波浪水动力方面的研究。email: aoyuwang@stu.csust.edu.cn
基金资助:
国家重点研发计划课题项目(2022YFC3103601)

The hydrodynamic characteristics prediction of landslide-induced surge waves based on non-hydrostatic model numerical simulation and convolutional neural network

WANG Aoyu¹(), QU Ke¹^,²^,³(), WANG Xu¹, GAO Rongze¹, MEN Jia¹

1. School of Hydraulic and Environmental Engineering, Changsha University of Science & Technology, Changsha 410114, China
2. Key Laboratory of Dongting Lake Aquatic Eco-Environmental Control and Restoration of Hunan Province, Changsha 410114, China
3. Key Laboratory of Water-Sediment Sciences and Water Disaster Prevention of Hunan Province, Changsha 410114, China

Received:2024-03-27 Revised:2024-05-30 Online:2025-03-10 Published:2025-04-11
Contact: QU Ke
Supported by:
National Key Research and Development Program of China(2022YFC3103601)

摘要/Abstract

摘要：

海底滑坡作为一种破坏力巨大并且在全世界范围广泛分布的自然灾害, 往往会给人类的生命安全产生巨大的威胁。滑坡产生的巨大涌浪会对海洋建筑物造成严重破坏, 因此迅速预测和评估海底滑坡所能产生的涌浪大小是防灾减灾工作的关键部分, 对海洋资源的开发利用以及人民生命财产安全至关重要。文章以非静压模型(non-hydrostatic wave model, NHWAVE)进行了滑坡涌浪的数值模拟, 得到了不同滑坡产生涌浪的数据, 并以一维卷积神经网络(1-dimensional convolutional neural network, CONV1D)为基础, 训练了滑坡产生涌浪的预测模型。该模型使用了不同测点和不同类型的滑坡数据集进行训练, 并使用平均绝对误差等评价指标对卷积神经网络的预测结果进行评估。在使用少量数据集的条件下, 卷积神经网络能很好地学习到滑坡涌浪的规律, 并且对于数据集中不存在的特征也能预测得到不错的结果, 具有较好的泛化能力。模型训练好之后, 只要实时输入滑坡发生位置自由表面的水位数据, 神经网络就能在短时间内预测出未来下游测点涌浪的时程曲线。通过神经网络预测, 可以提前对灾害进行评估, 从而采取及时有效的应对措施。

关键词: 海底滑坡, 卷积神经网络, 时序预测, 灾害预警

Abstract:

Submarine landslides, as a natural disaster with immense destructive potential and widespread distribution worldwide, often pose significant threats to the safety of human lives. The massive waves generated by these landslides can cause severe damage to marine structures. Therefore, rapidly predicting and assessing the size of the waves generated by underwater landslides is a crucial part of disaster prevention and mitigation efforts, essential for the development and utilization of marine resources, as well as for the safety of human lives and properties. In this study, numerical simulations of landslide-generated waves were conducted using the non-hydrostatic wave model NHWAVE (a non-hydrostatic wave model) to obtain wave data for different types of landslides. Subsequently, a CONV1D (1-dimensional convolutional neural network) was trained as the prediction model for landslide-generated waves. The model was trained using datasets for various monitoring points and different types of landslides, and evaluation metrics such as mean squared error were employed to assess the prediction performance of the convolutional neural network. The results indicate that, under the condition of using a limited amount of data, the convolutional neural network can effectively learn the patterns of landslide-generated waves. Moreover, it can predict reasonably well even for features that are not uniquely present in the dataset, demonstrating good generalization capability. Once the model is trained, inputting real-time water level data from the location of the landslide occurrence enables the neural network to predict the temporal wave profiles at downstream monitoring points in a short time. By using neural networks for prediction, it is possible to assess disasters in advance and take timely and effective response measures.

Key words: submarine landslide, convolutional neural network, temporal prediction, disaster warning

中图分类号:

P751

王傲宇, 屈科, 王旭, 高榕泽, 门佳. 基于非静压模型数值模拟与卷积神经网络的滑坡涌浪水动力特性预测[J]. 热带海洋学报, 2025, 44(2): 187-195.

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.

图/表 7

图1

图2

图3

图4

图5

图6

图7

参考文献 24

[1]	戴晨, 夏非, 张永战, 等, 2015. 南黄海辐射沙脊群海域历史地震的浅地层记录[J]. 海洋地质与第四纪地质, 35(5): 77-85.
	DAI CHEN, XIA FEI, ZHANG YONGZHAN, 2015. Shallow seismic stratigraphic records of earthquakes in the radial sand ridge field of South Yellow Sea. Marine Geology & Quaternary Geology, 35(5): 77-85 (in Chinese with English abstract).
[2]	高榕泽, 屈科, 任兴月, 等, 2024. 卷积神经网络方法在岛礁类海啸波水动力特性演变的应用[J]. 热带海洋学报, 43(4): 68-75.
	GAO RONGZE, QU KE, REN XINGYUE, et al, 2024. Application of convolutional neural network methods in the evolution of hydrodynamic characteristics of tsunamis like-wave over fringing reef[J]. Journal of Tropical Oceanography, 43(4): 68-75 (in Chinese with English abstract).
[3]	胡光海, 2010. 东海陆坡海底滑坡识别及致滑因素影响研究[D]. 青岛: 中国海洋大学.
	HU GUANGHAI, 2010. Identification of submarine landslides along the continental slope of the East China Sea and analysis of factors causing submarine landslides[D]. Qingdao: Ocean University of China (in Chinese with English abstract).
[4]	贾永刚, 单红仙, 2000. 黄河口海底斜坡不稳定性调查研究[J]. 中国地质灾害与防治学报, 11(1): 1-5.
	JIA YONGGANG, SHAN HONGXIAN, 2000. Investigation and study of slope unstability of subaqueous delta of modern Yellow River[J]. The Chinese Journal of Geological Hazard and Control, 11(1): 1-5 (in Chinese with English abstract).
[5]	贾永刚, 王振豪, 刘晓磊, 等, 2017. 海底滑坡现场调查及原位观测方法研究进展[J]. 中国海洋大学学报, 47(10): 61-72.
	JIA YONGGANG, WANG ZHENHAO, LIU XIAOLEI, et al, 2017. The research progress of field investigation and in-situ observation methods for submarine landslide[J]. Periodical of Ocean University of China, 47(10): 61-72 (in Chinese with English abstract).
[6]	刘保华, 李西双, 赵月霞, 等, 2005. 冲绳海槽西部陆坡碎屑沉积物的搬运方式: 滑塌和重力流[J]. 海洋与湖沼, 36(1): 1-9.
	LIU BAOHUA, LI XISHUANG, ZHAO YUEXIA, et al, 2005. Debris transport on the western continental slope of the Okinawa trough: slumping and gravity flowing[J]. Oceanologia et Limnologia Sinica, 36(1): 1-9 (in Chinese with English abstract).
[7]	马云, 2014. 南海北部陆坡区海底滑坡特征及触发机制研究[D]. 青岛: 中国海洋大学.
	MA YUN, 2014. Study of submarine landslides and trigger mechanism along the continental slope of the Northern South China Sea[D]. Qingdao: Ocean University of China (in Chinese with English abstract).
[8]	杨作升, 陈卫民, 陈彰榕, 等, 1994. 黄河口水下滑坡体系[J]. 海洋与湖沼, 25(6): 573-581.
	YANG ZUOSHENG, CHEN WEIMIN, CHEN ZHANGRONG, et al, 1994. Subaqueous landslide system in the Huanghe River (Yellow River) delta[J]. Oceanologia et Limnologia Sinica, 25(6): 573-581 (in Chinese with English abstract).
[9]	BRADFORD S F, AFFILIATIONS A, 2011. Nonhydrostatic model for surf zone simulation[J]. Journal of Waterway, Port, Coastal, and Ocean Engineering, 137(4): 163-174.
[10]	ENET F, GRILLI S T, 2007. Experimental study of tsunami generation by three-dimensional rigid underwater landslides[J]. Journal of Waterway, Port, Coastal, and Ocean Engineering, 133(6): 442-454.
[11]	GEE M J R, UY H S, WARREN J, et al, 2007. The Brunei slide: a giant submarine landslide on the North West Borneo Margin revealed by 3D seismic data[J]. Marine Geology, 246(1): 9-23.
[12]	GRAVES A, 2012. Long short-term memory[M]//GRAVES A, Supervised sequence labelling with recurrent neural networks. Berlin Heidelberg: Springer: 37-45.
[13]	HSU S K, KUO J, LO C L, et al, 2008. Turbidity currents, submarine landslides and the 2006 Pingtung earthquake off SW Taiwan[J]. Terrestrial, Atmospheric and Oceanic Sciences, 19(6): 767-772.
[14]	JIANG MINGJING, SUN CHAO, CROSTA G B, et al, 2015. A study of submarine steep slope failures triggered by thermal dissociation of methane hydrates using a coupled CFD-DEM approach[J]. Engineering Geology, 190: 1-16.
[15]	KINGMA D P, BA J, 2014. Adam: a method for stochastic optimization[J]. arXiv preprint arXiv: 1412. 6980.
[16]	KRIZHEVSKY A, SUTSKEVER I, HINTON G E, 2012. ImageNet classification with deep convolutional neural networks[C]// Proceedings of the 26th international conference on neural information processing systems. Lake Tahoe: 1097-1105.
[17]	MA GANGFENG, SHI FENGYAN, KIRBY J T, 2012. Shock-capturing non-hydrostatic model for fully dispersive surface wave processes[J]. Ocean Modelling, 43-44: 22-35.
[18]	MA GANGFENG, KIRBY J T, SHI FENGYAN, 2013. Numerical simulation of tsunami waves generated by deformable submarine landslides[J]. Ocean Modelling, 69: 146-165.
[19]	MAKINOSHIMA F, OISHI Y, YAMAZAKI T, et al, 2021. Early forecasting of tsunami inundation from tsunami and geodetic observation data with convolutional neural networks[J]. Nat Commun, 12: 2253. doi: 10.1038/s41467-021-22348-0 pmid: 33859177
[20]	MURPHY K, SCHÖLKOPF B, SRIVASTAVA N, et al, 2014. Dropout: a simple way to prevent neural networks from overfitting[J]. The Journal of Machine Learning Research, 15(1): 1929-1958.
[21]	NAMEKAR S, YAMAZAKI Y, CHEUNG K F, 2009. Neural network for tsunami and runup forecast[J]. Geophysical Research Letters, 36: L08604.
[22]	NIAN T, GUO X, ZHENG D, et al, 2019. Susceptibility assessment of regional submarine landslides triggered by seismic actions[J]. Applied Ocean Research, 93: 101964.
[23]	RASYIF T M, KATO S, SYAMSIDIK, et al, 2019. Numerical simulation of morphological changes due to the 2004 tsunami wave around Banda Aceh, Indonesia[J]. Geosciences, 9(3): 125.
[24]	ZHAO T, UTILI S, CROSTA G B, 2016. Rockslide and impulse wave modelling in the Vajont reservoir by DEM-CFD analyses[J]. Rock Mechanics and Rock Engineering, 49(6): 2437-2456.

基于非静压模型数值模拟与卷积神经网络的滑坡涌浪水动力特性预测

The hydrodynamic characteristics prediction of landslide-induced surge waves based on non-hydrostatic model numerical simulation and convolutional neural network

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 7

参考文献 24

相关文章 3

Metrics

本文评价

推荐阅读 0

[1]	高榕泽, 屈科, 任兴月, 王旭. 卷积神经网络方法在岛礁类海啸波水动力特性演变的应用[J]. 热带海洋学报, 2024, 43(4): 68-75.
[2]	翁少佳, 蔡锦海, 庞运禧, 罗荣真. 卷积神经网络在近岸表层海温预报中的应用[J]. 热带海洋学报, 2024, 43(1): 40-47.
[3]	陈梅, 夏真, 刘文涛, 何健, 马胜中. 南海宣德环礁东南海域地质灾害特征[J]. 热带海洋学报, 2022, 41(2): 103-111.