&nbsp;基于三参数单纯形法的船舶操纵Norrbin模型参数辨识

PDF(12111 KB)

大连海事大学学报 ›› 2026, Vol. 52 ›› Issue (1) : 1-11.

基于三参数单纯形法的船舶操纵Norrbin模型参数辨识

孙庆杰^1a,1b，任俊生^*1a,1b，李晴昊²

作者信息 +

Parameter identification of Norrbin model for ship maneuvering based on three-parameter simplex method

SUN Qingjie^1a,1b，REN Junsheng^*1a,1b，LI Qinghao²

Author information +

文章历史 +

摘要

为有效估计船舶操纵运动模型参数，本文提出一种基于三参数单纯形法(SA)的非线性辨识方法。该方法以试验数据为基础构建精度指标函数，确立模型输出与试验数据之间的差异评判准则；通过在三维参数空间中执行反射、扩张、收缩等几何操作，对参数进行迭代寻优。算法经不断迭代直至满足收敛准则，最终获得使指标函数最小化的全局最优参数组合。应用该SA算法对Mariner船的Norrbin模型参数进行辨识，并分别采用支持向量机(SVM)与扩展卡尔曼滤波(EKF)算法作为对比方法，系统开展了辨识性能的对比分析。通过计算各方法参数的相对误差可知，SA算法的辨识精度显著优于其他两种方法：其参数相对误差均为0.022%，最大误差不超过0.041%；EKF算法的参数相对误差范围为8.150%~32.068%，而SVM算法的参数相对误差均为11.2%，部分参数误差甚至超过20%。结果表明，采用SA算法辨识得到的模型更符合船舶的实际操纵性能。此外，在算法结构方面，SA方法相较于SVM与EKF算法更为简洁，所需预设参数更少，因而具备更好的实用性与工程适用性。

Abstract

To effectively estimate the parameters of ship maneuvering motion models, this paper proposes a nonlinear identification method based on the threeparameter simplex algorithm (SA). A precision indicator function is constructed based on experimental data to establish a discrepancy criterion between model outputs and experimental data. It performs geometric operations such as reflection, expansion, and contraction in a threedimensional parameter space to achieve iterative optimization of parameters. The algorithm undergoes continuous iteration until it meets the convergence criterion, ultimately obtaining the globally optimal parameter combination that minimizes the indicator function. The proposed SA algorithm is applied to identify the parameters of the Norrbin model for the Mariner ship, and support vector machine (SVM) and extended Kalman filter (EKF) algorithms are used as comparison methods to systematically analyze identification performance. It can be seen from the calculation of the relative errors of the parameters obtained by each method that the identification accuracy of the SA algorithm is significantly superior to that of the other two methods: the relative error of its parameters is 0.022% for all cases, with the maximum error not exceeding 0.041%; the relative error of the EKF algorithm ranges from 8.150% to 32.068%, while the relative error of the SVM algorithm is 11.2% for all parameters, and the error of some parameters even exceeds 20%. The results show that the model identified by the SA algorithm is more consistent with the actual maneuvering performance of the ship. Furthermore, in terms of algorithm structure, the SA method is more concise than the SVM and EKF algorithms, requires fewer preset parameters, and therefore exhibits better practicality and engineering applicability.

导出引用

孙庆杰, 任俊生, 李晴昊. 基于三参数单纯形法的船舶操纵Norrbin模型参数辨识[J]. 大连海事大学学报. 2026, 52(1): 1-11

SUN Qingjie, REN Junsheng, LI Qinghao. Parameter identification of Norrbin model for ship maneuvering based on three-parameter simplex method [J]. Journal of Dalian Maritime University. 2026, 52(1): 1-11

参考文献

［1］张炜灵,蔡烽,霍龙.船舶操纵响应模型参数辨识的发展趋势分析［J］.舰船电子工程,2019,39(11):4-5.
ZHANG W L, CAI F, HUO L. Development trend of parameter identification of ship maneuvering response equation［J］. Ship Electronic Engineering, 2019,39(11):4-5. (in Chinese)
［2］朱曼,文元桥,孙吴强,等.船舶运动模型参数辨识研究综述［J］.交通信息与安全,2022,40(5):1-11.
ZHU M, WEN Y Q, SUN W Q, et al. A review of parameter identification methods for ship dynamic models［J］. Journal of Transport Information and Safety, 2022,40(5):1-11. (in Chinese)
［3］郑文龙,肖昌润.船舶操纵非线性KT方程参数的辨识［J］.舰船科学技术,2017,39(17):129-132.
ZHENG W L, XIAO C R. The parameter identification of KT equations on ship maneuverability［J］. Ship Science and Technology, 2017,39(17):129-132. (in Chinese)[ZK)]
［4］王雪刚,邹早建.基于果蝇优化算法的船舶操纵响应模型的辨识［J］.大连海事大学学报,2012,38(3):1-4.
WANG X G, ZOU Z J. Identification of ship manoeuvring response model based on fruit fly optimization algorithm［J］. Journal of Dalian Maritime University, 2012,38(3):1-4. (in Chinese)
［5］TOMERA M. Swarm intelligence applied to identification of nonlinear ship steering model［C］//2015 IEEE 2nd International Conference on Cybernetics (CYBCONF). Gdynia:IEEE, 2015: 133-139.
［6］张心光,邹早建.基于支持向量回归机的船舶操纵响应模型辨识［J］.上海交通大学学报,2011,45(4):501-504.
ZHANG X G, ZOU Z J. Identification of response models of ship manoeuvring motion using support vector regression［J］. Journal of Shanghai Jiao Tong University, 2011,45(4):501504. (in Chinese)
［7］LUO W L, ZOU Z J. Parametric identification of ship maneuvering models by using support vector machine［J］. Journal of Ship Research. 2009, 53: 19-30.
［8］宋利飞,王毓清,彭伟,等.基于SVR的船舶简化分离型模型水动力系数辨识研究［J］.中国舰船研究,2025,20(1):65-75.
SONG L F,WANG Y Q,PENG W, et al. Hydrodynamic coefficients identification of ship simplified modular model based on support vector regression［J］. Chinese Journal of Ship Research, 2025,20(1):6575. (in Chinese).
［9］祝添权,初秀民,田国昊,等.基于LSSVM的船舶操纵响应模型参数辨识及应用［J］.武汉理工大学学报,2021,43(12):28-36.
ZHU T Q, CHU X M, TIAN G H, et al. Parameter identification and application of ship maneuvering response model based on LS-SVM［J］. Journal of Wuhan University of Technology, 2021,43(12):2836. (in Chinese)
［10］褚式新,俞剑,姜文,等.基于遗传算法的水面无人艇操纵响应模型参数辨识仿真研究［J］.中国电子科学研究院学报,2023,18(10):930-935.
CHU S X, YU J, JIANG W, et al. Simulation research of parameter identification of maneuvering response model of unmanned surface vessel based on genetic algorithms［J］. Journal of China Academy of Electronics and Information Technology, 2023,18(10):930935. (in Chinese)
［11］谢朔,陈德山,初秀民,等.改进多新息卡尔曼滤波法辨识船舶响应模型［J］.哈尔滨工程大学学报,2018, 39(2):282-289.
XIE S, CHEN D S, CHU X M, et al. Identification of ship response model based on improved multiinnovation extended Kalman filter［J］. Journal of Harbin Engineering University, 2018,39(2):282289. (in Chinese)
［12］李晴昊,任俊生,华焱.基于均方根容积卡尔曼滤波的船舶操纵运动响应模型参数辨识［J］.系统仿真学报,2024,36(8):1790-1799.
LI Q H, REN J S, HUA Y. Parametric identification of ship maneuvering motion response model based on square root cubature Kalman filtering［J］. Journal of System Simulation, 2024, 36(8): 17901799. (in Chinese)
［13］MENG Y, ZHANG X K, ZHANG X F, et al. Maneuverability indices estimation of nonlinear response model based on full-scale trial data and square root unscented Kalman filter［J］. Ocean Engineering, 2023, 286:115699.
［14］MENG Y, ZHANG X F, ZHU J X. 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.
［15］黄小斌,杨盐生.船舶操纵运动数学模型预报精度的研究［J］.大连海事大学学报,1997,23(2):27-31.
HUANG X B, YANG Y S. Improvement on precise prediction of ship manoeuvring mathematical model［J］. Journal of Dalian Maritime University, 1997,23(2):27-31. (in Chinese)
［16］任俊生.船舶运动与控制［M］.大连:大连海事大学出版社,2021.
REN J S. Ship motion and control［M］.Dalian：Dalian Maritime University Press, 2021. (in Chinese)
［17］Fossen T I. Handbook of marine craft hydrodynamics and motion control［M］. John Wiley & Sons, 2011.