基于卷积神经网络的涵洞式直立堤波浪透射预测

赵西增; 徐天宇; 谢玉林; 吕超凡; 姚炎明; 解静; 常江

doi:10.6052/0459-1879-20-235

基于卷积神经网络的涵洞式直立堤波浪透射预测

doi: 10.6052/0459-1879-20-235

赵西增^*,
徐天宇^*,
谢玉林^*,
吕超凡^*,
姚炎明^*,2), ,,
解静^†,
常江^**

^*浙江大学海洋学院, 浙江舟山 316021
^†交通运输部天津水运工程科学研究所, 天津 300456
^**天津水运工程勘察设计院, 天津市水运工程测绘技术重点实验室, 天津 300456

基金项目: 1) 国家自然科学基金(51679212)

详细信息

作者简介:
2) 姚炎明, 副教授, 主要研究方向: 海洋动力环境的数值模拟. E-mail: hotfireyao@163.com

通讯作者:
姚炎明

中图分类号: TV13
计量
- 文章访问数: 1870
- HTML全文浏览量: 470
- PDF下载量: 247
- 被引次数: 0
出版历程
- 收稿日期: 2020-06-30
- 刊出日期: 2021-02-10

PREDICTION OF WAVE TRANSMISSION OF CULVERT BREAKWATER BASED ON CNN

^*Zhejiang University, Ocean College, Zhoushan 316021, Zhejiang, China
^†Tianjin Research Institute for Water Transport Engineering, Tianjin 300456, China
^**Tianjin Survey and Design Institute for Water Transport Engineering, Tianjin Key Laboratory of Surveying and Mapping for Waterway Transport Engineering, Tianjin 300456, China

摘要

摘要: 涵洞式直立堤是一种具有特殊用途的海岸工程结构物,对其透浪特性的研究具有重要工程意义. 然而,目前众多学者对于涵洞式直立堤波浪透射问题的研究主要以理论分析、实验模拟及数值计算为主.随着机器学习技术的发展, 传统水动力学问题迎来了新的求解理念.机器学习算法可根据训练数据集自主学习相应的规律,以数据映射的方式建立水动力学特征预测模型,在实际应用中无需对流体运动控制方程进行求解, 具有较高的计算效率. 因此,本文基于卷积神经网络(convolutional neural network, CNN),对不同开孔条件下的涵洞式直立堤透浪特征进行预测.首先利用模型试验验证计算流体力学(computational fluid dynamics, CFD)模型的有效性,然后基于CFD模型生成相应的训练数据集, 通过训练卷积神经网络模型,建立相应的波浪透射结果之间的数据映射关系,实现在新的工况下对波浪透射系数以及透射波波形等特征的快速预测. 结果表明,经过训练的卷积神经网络可在极短时间内计算得到相应的结果, 并具有较高的准确性.研究成果可为波浪与海岸结构物相互作用的问题提供新的求解理念.
- 涵洞式直立堤 /
- 卷积神经网络 /
- 波浪透射 /
- 深度学习 /
- CFD模拟
Abstract: Culvert breakwater is a common coastal engineering structure. At the same time, the wave energy can also be transmitted into the harbor through the culvert, which will affect the hydrodynamic characteristics and mooring stability of the harbor. The study of its wave transmission characteristics is closely related to the safety of relevant production equipment and the corresponding engineering economic cost. However, many scholars mainly focus on theoretical analysis, experimental simulation and numerical calculation for wave transmission of culvert type vertical breakwater. With the development of machine learning technology, the traditional hydrodynamic problems ushered in a new solution concept which has attracted many attentions in the field of physics and engineering. At present, many scholars have applied machine learning algorithm to wave related problems. The machine learning algorithm can autonomously learn the corresponding laws according to the training data set, and establish the prediction model of hydrodynamic characteristics by data mapping. In practical application, it does not need to solve the fluid motion control equation, and has high computational efficiency. In this paper, based on the convolutional neural network (CNN), the wave transmission characteristics of the culvert breakwater under different incident and different opening conditions are predicted. The corresponding training data set is generated by a CFD model for convolution neural network training. The CFD results are compared with physical results for validation. After the data mapping relationship between different working conditions and the corresponding wave transmission results are established, the wave transmission coefficient and wave characteristics of transmission wave under the new working conditions can be predicted rapidly. The results show that the trained convolutional neural network can calculate the corresponding results within 10 milliseconds with a relatively high accuracy. This study can provide a new idea for solving the problem of interaction between waves and coastal structures, and is of importance in engineering application.
- culvert breakwater /
- CNN /
- wave transmission /
- deep learning /
- CFD modeling

HTML全文

参考文献(30)

[1]	黄蕙, 马舒文, 王定略. 涵洞式直立堤透浪特性研究. 水运工程, 2013,12:25-29 (Huang Hui, Ma Shuwen, Wang Dinglue. Wave transmission coefficients of vertical breakwater with culverts. Port & Waterway Engineering. 2013,12:25-29 (in Chinese))
[2]	吕超凡, 赵西增. 涵洞式直立堤的消波性能研究//第三十届全国水动力学研讨会暨第十五届全国水动力学学术会议, 2019: 1323-1328 (Lü Chaofan, Liu Chong, Zhao Xizeng. Wave dissipation property of the culvert type vertical breakwater//Proceedings of the 30th National Conference on Hydrodynamics & 15th National Congress on Hydrodynamics, 2019: 1323-1328 (in Chinese))
[3]	殷铭简, 赵西增. 涵洞式直立堤涵管内振荡流特性研究//第三十届全国水动力学研讨会暨第十五届全国水动力学学术会议, 2019: 1044-1049 (Yin Mingjian, Zhao Xizeng. Study on the characteristics of oscillating flow in the culvert pipe on a vertical breakwater//Proceedings of the 30th National Conference on Hydrodynamics & 15th National Congress on Hydrodynamics, 2019: 1044-1049 (in Chinese))
[4]	邓斌, 王孟飞, 黄宗伟等. 波浪作用下直立结构物附近强湍动掺气流体运动的数值模拟. 力学学报, 2020,52(2):408-419 (Deng Bin, Wang Mengfei, Huang Zongwei, et al. Numerical simulation of the hydrodynamic characteristics of violent aerated flows near vertical structure under wave action. Chinese Journal of Theoretical and Applied Mechanics. 2020,52(2):408-419 (in Chinese))
[5]	王国盛, 拾兵, 何昆等. 基于GA-BP神经网络的孤立波爬高预测. 中国海洋大学学报(自然科学版), 2018,48(S2):168-173 (Wang Guosheng, Shi Bing, He Kun, et al. Prediction of solitary wave run-up based on GA-BP neural network. Periodical of Ocean University of China. 2018,48(S2):168-173 (in Chinese))
[6]	霍政界. 基于人工神经网络的波浪发电系统输出功率预测. 电子测试, 2014(8):107-109 (Huo Zhengjie. Output power prediction of the wave power system based on artificial neural network. Electronic Test, 2014(8):107-109 (in Chinese))
[7]	金权. 基于机器学习算法对海浪波高的预测及优化研究. [博士论文]. 青岛: 自然资源部第一海洋研究所, 2019 (Jin Quan. Prediction and optimization of wave height based on machine learning algorithm. [PhD Thesis]. Qingdao: The First Institute of Oceanography, MNR, 2019 (in Chinese))
[8]	李博, 李骏旻, 李毅能等. 人工神经网络在岛屿近岸海浪模拟中的应用. 厦门大学学报(自然科学版), 2020,59(3):420-427 (Li Bo, Li Junmin, Li Yineng, et al. Application of artificial neural network to numerical wave simulation in the coastal region of island. Journal of Xiamen University (Natural Science), 2020,59(3):420-427 (in Chinese))
[9]	周飞燕, 金林鹏, 董军. 卷积神经网络研究综述. 计算机学报, 2017,6:1229-1251 (Zhou Feiyan, Jin Linpeng, Dong Jun. Review of convolutional neural network. Chinese Journal of Computers. 2017,6:1229-1251 (in Chinese))
[10]	Zhang Y, Sung W, Mavris D. Application of convolutional neural network to predict airfoil lift coefficient//AIAA/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference, 2018, 1903-1912
[11]	Guo X, Li W, Iorio F. Convolutional neural networks for steady flow approximation//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, ACM, 2016, 481-490
[12]	Krizhevsky A, Sutskever I, Hinton G. Imagenet classification with deep convolutional neural networks. Advances in Neural Information Processing Systems, 2012,25(2):1-9
[13]	Yu D, Wang H, Chen P. Mixed pooling for convolutional neural networks//Proceedings of the Rough Sets and Knowledge Technology (RSKT), 2014, 364-375
[14]	Hinton GE, Srivastava N, Krizhevsky A. et al. Improving neural networks by preventing co-adaptation of feature detectors. Computer Science, 2012,3(4):212-223
[15]	Hinton GE, Srivastava N, Krizhevsky A. et al. Dropout: A simple way to prevent neural networks from overfitting. Journal of Machine Learning Research, 2014,15(1):1929-1958
[16]	Rumelhart DE, Hinton GE, Williams RJ. Learning representations by back propagating errors. Nature, 1986,323(6088):533-536
[17]	Trischler A, Wang T, Yuan X, et al. Newsqa: a machine comprehension dataset//Proceedings of the 2nd Workshop on Representation Learning for NLP, 2017
[18]	Ganin Y, Ustinova E, Ajakan H. et al. Domain-adversarial training of neural networks. Journal of Machine Learning Research, 2017,17(1):2096-2030
[19]	王坤峰, 苟超, 段艳杰等. 生成式对抗网络GAN的研究进展与展望. 自动化学报, 2017,43(3):321-332 (Wang Kunfeng, Gou Chao, Duan Yanjie, et al. Generative adversarial networks: The state of the art and beyond. Acta Automatica Sinica. 2017,43(3):321-332 (in Chinese))
[20]	Schirrmeister RT, Gemein L, Eggensperger K. et al. Deep learning with convolutional neural networks for decoding and visualization of EEG pathology. Human Brain Mapping, 2017,38(11):5391-5420
[21]	Rozantsev A, Salzmann M, Fua P. Beyond sharing weights for deep domain adaptation//IEEE Transactions on Pattern Analysis & Machine Intelligence, 2016: 1-1
[22]	Kingma D, Ba J. Adam: a method for stochastic optimization//International Conference on Learning Representations (ICLR), 2014, 1-15
[23]	Fu Y, Zhao X, Cao F. et al. Numerical simulation of viscous flow past an oscillating square cylinder using a cip-based model. Journal of Hydrodynamics, 2017,29:96-108
[24]	Zhao X, Cheng D, Zhang D. et al. Numerical study of low-reynolds-number flow past two tandem square cylinders with varying incident angles of the downstream one using a CIP-based model. Ocean Engineering, 2016,121:414-421
[25]	Zheng K, Zhao X, Yang Z. et al. Numerical simulation of water entry of a wedge using a modified ghost-cell immersed boundary method. Journal of Marine Science and Technology, 2019,2:108-116
[26]	Zheng K, Zhao X. Numerical simulation of water exit and entry using a modified ghost cell immersed boundary method. Journal of Marine Science and Technology. 2020,3:1107-1113
[27]	Zhao X, Gao Y, Cao F. et al. Numerical modeling of wave interactions with coastal structures by a constrained interpolation profile/immersed boundary method. International Journal for Numerical Methods in Fluids. 2016,81(5):265-283
[28]	Zhao X, Cheng D, Zhang Y. et al. Experimental and numerical study on the hydrodynamic characteristics of solitary waves passing over a submerged breakwater. China Ocean Engineering, 2019,33:253-267
[29]	聂隆锋, 赵西增, 张志杭等. 基于VPM-THINC/QQ模型的波浪高保真模拟. 力学学报, 2019,51(4):1043-1053 (Nie Longfeng, Zhao Xizeng, Zhang Zhihang, et al. High-fidelity simulation of wave propagation based on VPM-THINC/QQ model. Chinese Journal of Theoretical and Applied Mechanics. 2019,51(4):1043-1053 (in Chinese))
[30]	岳杰顺, 权晓波, 叶舒然等. 水下发射水动力的多尺度预测网络研究. 力学学报, 2020,12(4):1-11 (Yue Jieshun, Quan Xiaobo, Ye Shuran, et al. A multi-scale network for the prediction of hydrodynamics in underwater. Chinese Journal of Theoretical and Applied Mechanics. 2020,12(4):1-11 (in Chinese))