基于支持向量回归的破损船舶横摇运动快速预报

doi:10.16183/j.cnki.jsjtu.2023.431

上海交通大学学报 ›› 2025, Vol. 59 ›› Issue (7): 1041-1049.doi: 10.16183/j.cnki.jsjtu.2023.431

基于支持向量回归的破损船舶横摇运动快速预报

刘涵, 苏焱(), 张国强

中山大学海洋工程与技术学院,广东珠海 519000

收稿日期:2023-08-31 修回日期:2023-11-19 接受日期:2023-12-08 出版日期:2025-07-28 发布日期:2025-07-22
通讯作者: 苏焱 E-mail:suyan23@mail.sysu.edu.cn
作者简介:刘涵(2001—),硕士生,从事船舶运动研究.
基金资助:
国家重点研发计划(2021YFC2800700)

Fast Prediction for Roll Motion of a Damaged Ship Based on SVR

LIU Han, SU Yan(), ZHANG Guoqiang

School of Ocean Engineering and Technology, Sun Yat-Sen University, Zhuhai 519000, Guangdong, China

Received:2023-08-31 Revised:2023-11-19 Accepted:2023-12-08 Online:2025-07-28 Published:2025-07-22
Contact: SU Yan E-mail:suyan23@mail.sysu.edu.cn

摘要/Abstract

摘要：

基于ANSYS-AQWA求解破损舰船DTMB5415在多个工况下的横摇运动响应,通过与文献结果对比验证数值模型的有效性,并基于数值结果构建破损船舶横摇运动响应数据库;采用支持向量回归算法对横摇运动数据库进行辨识建模,探究工况要素与横摇运动方程系数之间的关系,构建横摇运动响应快速预报模型并进行验证.该方法相较于传统计算流体力学模型,预报效率显著提高.

关键词: 快速预报, 支持向量回归, 破损船舶, 横摇运动

Abstract:

ANSYS-AQWA is applied to analyze the rolling motion response of the damaged ship DTMB5415 under various working conditions. The results are compared with those in exiting literature to validate the practicality of the hydrodynamic model. Additionly, the rolling motion response database for the damaged ship is constructed. The support vector regression (SVR) algorithm is used to model the rolling motion database for identification, exploring the relationship between the operating condition factors and coefficients in the equation of roll motion. Finally, a fast prediction model for rolling motion is constructed and validated, offering a significant improvement in the prediction efficiency compared with traditional computational fluid dynamics models.

Key words: fast prediction, support vector regression (SVR), damaged ship, rolling motion

中图分类号:

U661

刘涵, 苏焱, 张国强. 基于支持向量回归的破损船舶横摇运动快速预报[J]. 上海交通大学学报, 2025, 59(7): 1041-1049.

LIU Han, SU Yan, ZHANG Guoqiang. Fast Prediction for Roll Motion of a Damaged Ship Based on SVR[J]. Journal of Shanghai Jiao Tong University, 2025, 59(7): 1041-1049.

图/表 14

图1

表1

图2

图3

图4

图5

图6

图7

图8

图9

表2

表3

图10

图11

参考文献 16

[1]	XU S, GAO Z, XUE W. CFD database method for roll response of damaged ship during quasi-steady flooding in beam waves[J]. Applied Ocean Research, 2022, 126: 103282.
[2]	欧珊, 毛筱菲, 刘祖源, 等. 基于OpenFOAM的破损船舶横摇阻尼[J]. 上海交通大学学报, 2019, 53(3): 305-314. doi: 10.16183/j.cnki.jsjtu.2019.03.007
	OU Shan, MAO Xiaofei, LIU Zuyuan, et al. Roll damping of damaged ship based on OpenFOAM[J]. Journal of Shanghai Jiao Tong University, 2019, 53(3): 305-314.
[3]	The Stability in Waves Committee. Final report and recommendations to the 28th ITTC[C]// Proceedings of the 28th ITTC. Wuxi, China: International Towing Tank Conference, 2017(III):275-335.
[4]	The Specialist Committee on Stability in Waves. Final report and recommendations to the 25th ITTC[C]// Proceedings of the 25th ITTC. Fukuoka, Japan: International Towing Tank Conference, 2008(II):605-639.
[5]	ACANFORA M, BEGOVIC E, DE LUCA F. A fast simulation method for damaged ship dynamics[J]. Journal of Marine Science and Engineering, 2019, 7(4): 111-119.
[6]	MENG Y, ZHANG X, ZHU J. Parameter identification of ship motion mathematical model based on full-scale trial data[J]. International Journal of Naval Architecture and Ocean Engineering, 2022, 14: 100437.
[7]	CHEN H, LI Q, WANG Z. Improved maximum likelihood method for ship parameter identification[C]// 2018 37th Chinese Control Conference. Wuhan, China: IEEE, 2018: 1614-1621.
[8]	ZHENG J, YAN M, LI Y, et al. An online identification approach for a nonlinear ship motion model based on a receding horizon[J]. Transactions of the Institute of Measurement and Control, 2021, 43(13): 3000-3012.
[9]	徐锋, 邹早建, 宋鑫. 基于支持向量机的水下运载器平面操纵运动建模[J]. 上海交通大学学报, 2012, 46(3): 358-362.
	XU Feng, ZOU Zaojian, SONG Xin. Modeling of underwater vehicles’ planar maneuvering motion based on support vector machines[J]. Journal of Shanghai Jiao Tong University, 2012, 46(3): 358-362.
[10]	张心光. 基于支持向量回归机的船舶操纵运动在线辨识建模[J]. 船舶工程, 2019, 41(3): 98-101.
	ZHANG Xinguang. Online identification modeling of ship manoeuvring motion using support vector regression[J]. Ship Engineering, 2019, 41(3): 98-101.
[11]	WANG Z, XU H, XIA L, et al. Kernel-based support vector regression for nonparametric modeling of ship maneuvering motion[J]. Ocean Engineering, 2020, 216: 107994.
[12]	ZHU M, WEN Y, SUN W, et al. A novel adaptive weighted least square support vector regression algorithm-based identification of the ship dynamic model[J]. IEEE Access, 2019, 7: 128910-128924.
[13]	BEGOVIC E, DAY A H, INCECIK A. An experimental study of hull girder loads on an intact and damaged naval ship[J]. Ocean Engineering, 2017, 133: 47-65.
[14]	黄柏刚, 邹早建. 基于固定网格小波神经网络的不规则波中船舶横摇运动在线预报[J]. 船舶力学, 2020, 24(6): 693-705.
	HUANG Baigang, ZOU Zaojian. Online prediction of ship roll motion in irregular waves using a fixed grid wavelet network[J]. Journal of Ship Mechanics, 2020, 24(6): 693-705.
[15]	沈文君, 赵志娟, 刘利琴, 等. 波浪周期对小型船舶动力响应的影响研究[J]. 船舶力学, 2022, 26(3): 342-352.
	SHEN Wenjun, ZHAO Zhijuan, LIU Liqin, et al. Research of wave period effect on the dynamic response characteristics of a small ship[J]. Journal of Ship Mechanics, 2022, 26(3): 342-352.
[16]	HAN H, WANG W. A hybrid BPNN-GARF-SVR prediction model based on EEMD for ship motion[J]. CMES-Computer Modeling in Engineering & Sciences, 2023, 134(2): 1353-1370.

参数	实船数据	参数	实船数据
总长,L_OA/m	153.3	棱形系数,C_P	0.616
垂线间长,L_PP/m	142.2	中横剖面系数,C_M	0.815
设计水线宽,B_WL/m	19.082	重心纵向位置,L_CG/m	70.137
型深,D/m	12.47	横摇惯性半径,K_xx/m	6.932
吃水,d/m	6.15	纵摇惯性半径,K_yy/m	36.802
排水量,Δ/t	8635	艏摇惯性半径,K_zz/m	36.802
方形系数,C_B	0.505

组别	工况			超参数
组别	H/m	T/s	β/rad	BoxConstraint	KernelScale	Epsilon
1	1	7	0	957.411 1	39.963 6	0.000 5
2	1	8	0	80.412 5	3.440 8	0.003 5
3	1	9	π/4	391.721 7	31.774 4	0.003 2
4	1	10	π/4	863.240 9	560.921 6	0.001 3
5	1	9	π/2	692.254 4	16.115 8	0.000 9
6	1.5	6	π/4	998.172 2	78.043 7	0.006 9
7	1.5	9	π	593.945 7	51.515 5	0.000 4
8	1.5	10	π	876.905 7	850.494 3	0.055 9

组别	工况			辨识建模结果
组别	H/m	T/s	β/rad	d₁	d₃	r₁	r₃	f₄₁	f₄₂
1	1	7	0	0.099 9	0.003 1	0.056 9	0.007 6	-0.159 1	0.283 3
2	1	8	0	0.099 4	0.001 9	0.066 2	-0.001 3	-0.352 3	0.128 8
3	1	9	π/4	0.118 5	0.000 0	0.067 2	0.000 0	-0.634 8	0.719 7
4	1	10	π/4	0.071 6	0.000 0	0.061 6	0.000 0	-0.572 1	1.192 8
5	1	9	π/2	0.021 2	0.000 0	0.042 0	0.001 4	1.279 9	1.522 9
6	1.5	6	π/4	0.404 3	-0.003 2	0.105 8	-0.000 8	-0.202 9	1.428 4
7	1.5	9	π	0.076 7	0.000 1	0.063 4	0.000 0	-0.473 7	-0.405 6
8	1.5	10	π	0.074 1	0.000 0	0.056 5	0.000 0	-0.821 0	-0.590 9

基于支持向量回归的破损船舶横摇运动快速预报

Fast Prediction for Roll Motion of a Damaged Ship Based on SVR

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 14

参考文献 16

相关文章 15

编辑推荐

Metrics

本文评价

[1]	钟一鸣1, 于曹阳1, 2, 向先波2, 3, 连琏1, 4. 基于级联滤波与误差触发支持向量回归的海洋航行器运动预测研究[J]. 海洋工程装备与技术, 2025, 12(1): 133-140.
[2]	刘玉川1，李浩1，唐宇龙1，梁杜娟2，谭佳3，符玥1，李勇明4. 基于互信息-支持向量回归的阿尔兹海默症磁共振影像脑年龄检测[J]. J Shanghai Jiaotong Univ Sci, 2025, 30(1): 130-135.
[3]	李芬, 孙凌, 王亚维, 屈爱芳, 梅念, 赵晋斌. 基于CEEMDAN-GSA-LSTM和SVR的光伏功率短期区间预测[J]. 上海交通大学学报, 2024, 58(6): 806-818.
[4]	张博, 李克庆, 胡亚飞, 吉坤, 韩斌. 基于灰狼优化算法改进支持向量回归的充填体强度预测研究[J]. J Shanghai Jiaotong Univ Sci, 2023, 28(5): 686-694.
[5]	刘钇汛, 刘志浩, 高钦和, 黄通, 马栋. 基于周向应变分析的重载轮胎垂向力估计算法[J]. 上海交通大学学报, 2023, 57(10): 1273-1281.
[6]	朱城昊, 王晗, 孙国歧, 魏晓宾, 王富文, 蔡旭. 一种并网逆变器直流电容容值辨识方法[J]. 上海交通大学学报, 2022, 56(6): 693-700.
[7]	徐宏东, 高海波, 徐晓滨, 林治国, 盛晨兴. 基于证据推理规则CS-SVR模型的锂离子电池SOH估算[J]. 上海交通大学学报, 2022, 56(4): 413-421.
[8]	卢地华, 陈自强. 基于双充电状态的锂离子电池健康状态估计[J]. 上海交通大学学报, 2022, 56(3): 342-352.
[9]	王会峰, 黄福祥, 蒋扬, 阴炳钢, 田立锋, 李隶辉. 半潜平台湿拖作业时横摇水动力特性数值研究 [J]. 海洋工程装备与技术, 2022, 9(2): 29-37.
[10]	王宏健，徐金龙，刘向波，李娟，张爱华. 基于支持向量回归机的自适应差分滤波算法研究[J]. 上海交通大学学报（自然版）, 2014, 48(07): 929-935.
[11]	王晓龙，蔡琦，陈玉清. 基于多参数耦合的船舶核动力装置参数重构[J]. 上海交通大学学报（自然版）, 2014, 48(02): 210-213.
[12]	刘长卿, 车驰东, 朱瑞仪. 城市交通高架轨道振动衰减的快速预报[J]. 上海交通大学学报（自然版）, 2013, 47(02): 242-247.
[13]	胡友涛, 胡昌华. 一种基于遗传算法优化小波支持向量回归机的实时寿命预测方法[J]. 上海交通大学学报（自然版）, 2011, 45(08): 1216-1220.
[14]	张宏伟, 宋执环. 基于彩色图像特征的铜成分软测量模型[J]. 上海交通大学学报（自然版）, 2011, 45(08): 1211-1215.
[15]	任会峰, 阳春华, 周璇, 桂卫华, 鄢锋. 基于最小二乘支持向量回归机的矿浆酸碱度鲁棒软测量[J]. 上海交通大学学报（自然版）, 2011, 45(08): 1136-1139.