留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于SSA-CNN模型的双排开孔圆筒防波堤透射系数预测

邓斌 王玲 何军 尹龙斌 蒋昌波 陈杰 伍志元

邓斌,王玲,何军,等. 基于SSA-CNN模型的双排开孔圆筒防波堤透射系数预测[J]. 海洋学报,2024,46(4):122–132 doi: 10.12284/hyxb2024035
引用本文: 邓斌,王玲,何军,等. 基于SSA-CNN模型的双排开孔圆筒防波堤透射系数预测[J]. 海洋学报,2024,46(4):122–132 doi: 10.12284/hyxb2024035
Deng Bin,Wang Ling,He Jun, et al. Prediction of transmission coefficient of double-row perforated cylinder breakwater based on SSA-CNN model[J]. Haiyang Xuebao,2024, 46(4):122–132 doi: 10.12284/hyxb2024035
Citation: Deng Bin,Wang Ling,He Jun, et al. Prediction of transmission coefficient of double-row perforated cylinder breakwater based on SSA-CNN model[J]. Haiyang Xuebao,2024, 46(4):122–132 doi: 10.12284/hyxb2024035

基于SSA-CNN模型的双排开孔圆筒防波堤透射系数预测

doi: 10.12284/hyxb2024035
基金项目: 国家重点研发项目( 2021YFB2601100);国家自然科学基金项目( 51979015,51839002);水利工程仿真与安全国家重点实验室开放基金项目(HESS-2114);湖南省科技创新计划项目(2020RC3037,20hnkj019)。
详细信息
    作者简介:

    邓斌(1985—),男,湖南省衡南县人,教授,主要从事河流、海岸动力过程及其数值模拟研究。E-mail:dengbin07@csust.edu.cn

    通讯作者:

    何军(1981—),男,安徽省和县人,高级工程师,主要从事水运工程管理与研究。E-mail:hejun@pdiwt.com.cn

  • 中图分类号: P753

Prediction of transmission coefficient of double-row perforated cylinder breakwater based on SSA-CNN model

  • 摘要: 双排开孔圆筒防波堤是一种新型环境友好型防波堤,对其消浪特性的研究具有重要工程意义。随着人工智能的发展,基于机器学习技术求解防波堤水动力学问题成了一种新的研究范式。本文提出基于麻雀搜索算法(Sparrow Search Algorithm, SSA)优化卷积神经网络(Convolutional Neural Network, CNN)模型,实现对双排开孔圆筒防波堤透射系数的智能优化预测。结果表明:(1)确定波高、波周期、波长、波速、排间距、开孔率、水深为影响透射系数的关键因子;(2)当SSA-CNN模型的种群数量为10时,对波浪透射系数预测的R2值达到0.9909,平均相对误差相比单一的CNN模型降低了5.07%。研究成果为利用神经网络研究波浪透射问题提供了一种新的优化预测模型。
  • 图  1  模型布置图

    Fig.  1  Layout of the model

    图  2  麻雀搜索算法计算流程

    Fig.  2  SSA calculation process

    图  3  SSA-CNN模型

    Fig.  3  SSA-CNN model

    图  4  CNN网络结构

    Fig.  4  Structure of the convolutional neural network

    图  5  SSA-CNN和CNN模型的预测值和真实值结果对比

    Fig.  5  Comparison of predicted values and real values of SSA-CNN and CNN models

    图  6  透射系数随相对排间距的变化情况(e = 46.22%)

    Fig.  6  The change of Kt with B/D (e = 46.22%)

    图  7  透射系数随开孔率的变化情况(H/D = 9)

    Fig.  7  The change of Kt with e (H/D = 9)

    图  8  SSA-CNN适应度曲线

    Fig.  8  Fitness curve of SSA-CNN

    图  9  SSA-CNN模型预测误差分布

    Fig.  9  Prediction error distribution of SSA-CNN model

    图  10  CNN模型预测误差分布

    Fig.  10  Prediction error distribution of CNN

    图  11  预测值与CFD计算值对比

    Fig.  11  Comparison between predicted values and CFD calculation values

    图  12  预测误差分布

    Fig.  12  Prediction error distribution

    表  1  数值模拟工况

    Tab.  1  Numerical simulation working condition

    工况
    波高,H/m 0.06 0.07 0.09 0.11
    波周期,T/s 1.4 1.5 1.6 1.8
    水深,d/m 0.5
    排间距,B/m 1.0 1.2 1.4 1.8
    开孔率,e 23.11% 34.67% 46.22%
    圆筒直径,D/m 0.2
    开孔直径,D1/m 0.04
    下载: 导出CSV

    表  2  SSA-CNN算法中的超参数

    Tab.  2  Hyperparameters in the SSA-CNN algorithm

    参数 超参数 初始化范围
    x1 批大小 1~130
    x2 学习率 0.001~0.01
    x3 conv_1 核大小 1~5
    x4 conv_1 核数量 1~40
    x5 conv_2核大小 1~5
    x6 conv_2核数量 1~40
    x7 全连接层神经元数量 1~10
    x8 正则化系数 0.0001~0.1
    下载: 导出CSV

    表  3  互信息值

    Tab.  3  Mutual information value

    序号 参数 MI
    1 H 1.194
    2 T 0.592
    3 L 0.592
    4 u 0.592
    5 d 0
    6 B 0.586
    7 e 0.473
    8 D1 0
    9 D 0
    下载: 导出CSV

    表  4  模型预测效果对比

    Tab.  4  Comparison of model prediction effect

    模型 R2 MAE MSE RMSE MRE
    CNN 0.8724 0.0276 0.0012 0.0345 0.0687
    SSA-CNN 0.9909 0.0071 0 0.0092 0.0180
    下载: 导出CSV

    表  5  平均相对误差结果分析

    Tab.  5  Analysis of mean relative error results

    序号种群数量平均相对误差
    153.19%
    2100.71%
    3151.33%
    4201.04%
    5250.82%
    6300.98%
    下载: 导出CSV
  • [1] 季则舟, 吴波, 李超, 等. 双排圆筒透空式防波堤: 201920286464.1[P]. 2019−03−07.

    Ji Zezhou, Wu Bo, Li Chao, et al. Double-row cylinder permeable breakwater: 201920286464.1[P]. 2019−03−07.
    [2] Stoker J J. Water Waves: The Mathematical Theory with Applications[M]. New York: Interscience Publishers, 1957, 83.
    [3] Suvarna P S, Sathyanarayana A H, Umesh P, et al. Laboratory investigation on hydraulic performance of enlarged pile head breakwater[J]. Ocean Engineering, 2020, 217: 107989. doi: 10.1016/j.oceaneng.2020.107989
    [4] Liu H.W., Ghidaoui M.S., Huang Z.H., et al. Numerical investigation of the interactions between solitary waves and pile breakwaters using BGK-based methods[J]. Computers and Mathematics with Applications, 2010, 61(12), 3668–3677.
    [5] 徐天宇. 基于卷积神经网络的近海工程水动力学特征预测[D]. 杭州: 浙江大学, 2020.

    Xu Tianyu. Prediction of hydrodynamic characteristics of offshore engineering based on convolutional neural network[D]. Hangzhou: Zhejiang University, 2020.
    [6] 周飞燕, 金林鹏, 董军. 卷积神经网络研究综述[J]. 计算机学报, 2017, 40(6): 1229−1251. doi: 10.11897/SP.J.1016.2017.01229

    Zhou Feiyan, Jin Linpeng, Dong Jun. Review of convolutional neural network[J]. Chinese Journal of Computers, 2017, 40(6): 1229−1251. doi: 10.11897/SP.J.1016.2017.01229
    [7] 赵西增, 徐天宇, 谢玉林, 等. 基于卷积神经网络的涵洞式直立堤波浪透射预测[J]. 力学学报, 2021, 53(2): 330−338. doi: 10.6052/0459-1879-20-235

    Zhao Xizeng, Xu Tianyu, Xie Yulin, et al. Prediction of wave transmission of culvertbreakwater based on CNN[J]. Chinese Journal of Theoretical and Applied Mechanics, 2021, 53(2): 330−338. doi: 10.6052/0459-1879-20-235
    [8] Montes-Atenas G, Seguel F, Valencia A, et al. Predicting bubble size and bubble rate data in water and in froth flotation-like slurry from computational fluid dynamics (CFD) by applying deep neural networks (DNN)[J]. International Communications in Heat and Mass Transfer, 2016, 76: 197−201. doi: 10.1016/j.icheatmasstransfer.2016.05.031
    [9] Wei Zhangping, Davison A. A convolutional neural networkbased model to predict nearshore waves and hydrodynamics[J]. Coastal Engineering, 2022, 171: 104044. doi: 10.1016/j.coastaleng.2021.104044
    [10] Yao Z, Wang Z, Wang D, et al. An ensemble CNN-LSTM and GRU adaptive weighting model based improved sparrow search algorithm for predicting runoff using historical meteorological and runoff data as input[J]. Journal of Hydrology, 2023, 625: 129977.
    [11] Formentin S M, Zanuttigh B, van der Meer J W. A neural network tool for predicting wave reflection, overtopping and transmission[J]. Coastal Engineering Journal, 2017, 59(1): 1750006.
    [12] 赵沛泓, 孙大鹏, 吴浩. 采用JADE-SVR方法研究波浪和开孔沉箱相互作用[J]. 水道港口, 2021, 42(2): 166−173. doi: 10.3969/j.issn.1005-8443.2021.02.003

    Zhao Peihong, Sun Dapeng, Wu Hao. Investigating on the wave interaction with perforated caisson based on JADE-SVR machine[J]. Journal of Waterway and Harbor, 2021, 42(2): 166−173. doi: 10.3969/j.issn.1005-8443.2021.02.003
    [13] Zanuttigh B, Formentin S M, van der Meer J W. Prediction of extreme and tolerable wave overtopping discharges through an advanced neural network[J]. Ocean Engineering, 2016, 127: 7−22. doi: 10.1016/j.oceaneng.2016.09.032
    [14] Kim T, Lee W D, Kwon Y, et al. Prediction of wave transmission characteristics of low crested structures using artificial neural network[J]. Journal of Ocean Engineering and Technology, 2022, 36(5): 313−325. doi: 10.26748/KSOE.2022.024
    [15] Li Qi, Shi Yaru, Lin Ruiqi, et al. A novel oil pipeline leakage detection method based on the sparrow search algorithm and CNN[J]. Measurement, 2022, 204: 112122.
    [16] Chen Gonggui, Zhu Mengyuan, Huang Jing, et al. Short-term wind speed prediction with master-slave performance based on CNN-LSTM and improved POABP[J]. Engineering Letters, 2023, 31(2): 848−861.
    [17] Li Xinhong, Guo Mengmeng, Zhang Renren, et al. A data-driven prediction model for maximum pitting corrosion depth of subsea oil pipelines using SSA-LSTM approach[J]. Ocean Engineering, 2022, 261: 112062. doi: 10.1016/j.oceaneng.2022.112062
    [18] 王军, 马小越, 张宇航, 等. 基于SSA-LSTM模型的黄河水位预测研究[J]. 人民黄河, 2023, 45(9): 65−69. doi: 10.3969/j.issn.1000-1379.2023.09.011

    Wang Jun, Ma Xiaoyue, Zhang Yuhang, et al. Research on the prediction of the Yellow River water level based on SSA-LSTM model[J]. Yellow River, 2023, 45(9): 65−69. doi: 10.3969/j.issn.1000-1379.2023.09.011
    [19] Xue Jiankai, Shen Bo. A novel swarm intelligence optimization approach: sparrow search algorithm[J]. Systems Science & Control Engineering, 2020, 8(1): 22−34.
    [20] Fan Guofeng, Li Yun, Zhang Xinyan, et al. Short-term loadforecasting based on a generalized regression neural network optimized by an improved sparrow search algorithm using the empirical wavelet decomposition method[J]. Energy Science & Engineering, 2023, 11(7): 2444−2468.
    [21] Zhu Yucheng, Cao Xudong, Han Zhongting. An improved CNN employing SSA and its application in bearing fault diagnosis[C]//Proceedings of the IEEE 5th International Conference on Automation, Electronics and Electrical Engineering. Shenyang: IEEE, 2022: 714–718.
    [22] 张新生, 贺凯璐. 基于SSA-CNN的长距离矿浆管道临界流速预测[J]. 安全与环境学报, 2022, 22(5): 2524−2531.

    Zhang Xinsheng, He Kailu. Prediction of critical velocity of long-distance slurry pipeline based on SSA-CNN[J]. Journal of Safety and Environment, 2022, 22(5): 2524−2531.
    [23] 邓斌, 尹龙斌, 黄姣凤, 等. 波浪与新型双排开孔圆筒防波堤相互作用三维数值模拟[J]. 力学学报, 2023, 55(4): 845−857. doi: 10.6052/0459-1879-22-545

    Deng Bin, Yin Longbin, Huang Jiaofeng, et al. Three dimensional numerical simulation of wave interaction with a new type of doublerow perforated cylinder breakwater[J]. Chinese Journal of Theoretical and Applied Mechanics, 2023, 55(4): 845−857. doi: 10.6052/0459-1879-22-545
    [24] Wang Ran, Zhao Jianhui, Yang Huijin, et al. Inversion of soil moisture on farmland areas based on SSA-CNN using multi-source remote sensing data[J]. Remote Sensing, 2023, 15(10): 2515. doi: 10.3390/rs15102515
    [25] Zhang R, Su J, Feng J. An extreme learning machine model based on adaptive multi-fusion chaotic sparrow search algorithm for regression and classification[J]. Evolutionary Intelligence, 2023: 1−20.
    [26] Fathy A, Alanazi T M, Rezk H, et al. Optimal energy management of micro-grid using sparrow search algorithm[J]. Energy Reports, 2022, 8: 758−773.
    [27] Liu Yang, Lu Yutong, Wang Yueqing, et al. A CNN-basedshock detection method in flow visualization[J]. Computers & Fluids, 2019, 184: 1−9.
    [28] Yu Dingjun, Wang Hanli, Chen Peiqiu. Mixed pooling for convolutional neural networks[C]//Proceedings of the 9th International Conference. Shanghai: Springer, 2014: 364−375.
    [29] Hinton G E, Srivastava N, Krizhevsky A, et al. Improving neural networks by preventing coadaptation of feature detectors[J]. Comput. ENCE 2012, 3: 212–223.
    [30] Srivastava N, Hinton G, Krizhevsky A, et al. Dropout: a simple way to prevent neural networks from overfitting[J]. The Journal of Machine Learning Research, 2014, 15(1): 1929−1958.
    [31] Gal Y, Ghahramani Z. A theoretically grounded application of dropout in recurrent neural networks[C]//Proceedings of the 30th International Conference on Neural Information Processing Systems. Barcelona: Curran Associates Inc. , 2016: 1027−1035.
    [32] 周长春, 姜杰, 李谦, 等. 基于融合特征选择算法的钻速预测模型研究[J]. 钻探工程, 2022, 49(4): 31−40.

    Zhou Changchun, Jiang Jie, Li Qian, et al. Research on drilling rate prediction model based on fusion feature selection algorithm[J]. Drilling Engineering, 2022, 49(4): 31−40.
  • 加载中
图(12) / 表(5)
计量
  • 文章访问数:  100
  • HTML全文浏览量:  49
  • PDF下载量:  12
  • 被引次数: 0
出版历程
  • 收稿日期:  2024-01-11
  • 修回日期:  2024-03-21
  • 网络出版日期:  2024-05-20
  • 刊出日期:  2024-06-30

目录

    /

    返回文章
    返回