集美大学学报（自然科学版）2025，Vol.30Issue(2)：141-149,9.DOI:10.19715/j.jmuzr.2025.02.05
船舶运动模型辨识在海事事故复现中的应用
Application of Ship Motion Model Identification to Maritime Accident Reproduction
摘要
Abstract
In order to reproduce the process of maritime accidents and simulate the ship's maneuvering and motion trajectory before the accident,this article studies the collision case of the vessel Balawan,using the VDR navigation data of the accident ship as training samples,the least square support vector machine(LS-SVM)method employed for determining the hydrodynamic derivatives of MMG(ship manoeuvring mathematic model group)model.Firstly,the real ship navigation data was repaired,filtered and screened,then model train-ing samples were constructed and model parameter identification was conducted.Finally,the simulation tests of the trajectory reproduction were carried out.By comparing with the real voyage trajectory,it was verified that the LS-SVM identification method can be used as a way to reproduce maritime accidents.关键词
海事事故/船舶运动数学模型/最小二乘支持向量机回归/轨迹复现/参数辨识Key words
maritime accident/mathematical modelling of ship motion/least square support vector ma-chine/trajectory recurrence/parameter identification分类
交通运输引用本文复制引用
蒋柳杨,孙洪波,张春雨,沈莉婷..船舶运动模型辨识在海事事故复现中的应用[J].集美大学学报（自然科学版）,2025,30(2):141-149,9.基金项目
福建省自然科学基金项目"基于耦合模型的船舶锚泊安全辅助系统关键技术研究"(2023J01325) （2023J01325）
