基于ν-支持向量机的波浪中船舶操纵运动辨识建模  

Identification modeling of ship maneuvering motion in waves based on ν-support vector machine

作　　者：张艳云 欧阳子路 邹早建[1,2] ZHANG Yan-yun;OUYANG Zi-lu;ZOU Zao-jian(School of Naval Architecture,Ocean and Civil Engineering,Shanghai Jiao Tong University,Shanghai 200240,China;State Key Laboratory of Ocean Engineering,Shanghai Jiao Tong University,Shanghai 200240,China)

机构地区：[1]上海交通大学船舶海洋与建筑工程学院,上海200240 [2]上海交通大学海洋工程国家重点实验室,上海200240

出　　处：《船舶力学》2023年第5期637-645,共9页Journal of Ship Mechanics

基　　金：国家自然科学基金资助项目(51979165)。

摘　　要：本文采用系统辨识方法对船舶在波浪中的三自由度操纵运动进行黑箱建模。以KCS船为研究对象,建模所需的数据由规则波中的船模回转试验获得。分别利用原始数据及经低通滤波处理后的数据,应用ν-支持向量机进行黑箱建模。进一步地,利用所建立的模型进行运动预报,并将预报结果与真实的试验值进行比较,对所提出的建模方法进行验证。结果表明,ν-支持向量机是一种可进行波浪中的船舶操纵运动辨识建模的有效方法。A black-box model was established for three-degrees-of-freedom ship maneuvering motion in waves using system identification method in this paper.With the KCS ship taken as the study object,the data needed for identification modeling were obtained from the turning circle maneuver of free-running ship model in regular waves.ν-support vector machine was applied to the black-box modeling with the original data and the low-pass filtered data,respectively.Further the established black-box models were used for ship motion prediction,and the predicted results were compared with the free-running model test data to verify the proposed modeling method.The results show that the ν-support vector machine is an effective method for identification modeling of ship maneuvering motion in waves.

关 键 词：船舶在波浪上的操纵运动 系统辨识 黑箱建模 ν-支持向量机 

分 类 号：U661.33[交通运输工程—船舶及航道工程]