面向环境不确定性的船舶操纵运动建模与预报方法  

Modeling and prediction method of ship maneuvering motion facing environmental uncertainty

作　　者：陈立家[1,2,3] 周欣蔚 杨沛艺 王凯[1,2,3] 李胜为[1,2,3] CHEN Li-jia;ZHOU Xin-wei;YANG Pei-yi;WANG Kai;LI Sheng-wei(School of Navigation,Wuhan University of Technology,Wuhan 430063,Hubei,China;Hubei Key Laboratory of Inland Shipping Technology,Wuhan University of Technology,Wuhan 430063,Hubei,China;State Key Laboratory of Maritime Technology and Safety,Wuhan University of Technology,Wuhan 430063,Hubei,China)

机构地区：[1]武汉理工大学航运学院,湖北武汉430063 [2]武汉理工大学内河航运技术湖北省重点实验室,湖北武汉430063 [3]武汉理工大学水路交通控制全国重点实验室,湖北武汉430063

出　　处：《交通运输工程学报》2024年第3期279-295,共17页Journal of Traffic and Transportation Engineering

基　　金：国家重点研发计划(2019YFB1600603)。

摘　　要：为解决复杂环境因素影响下的船舶操纵运动预报精度问题,提出了一种面向环境不确定性的船舶操纵运动灰箱辨识建模与预报方法;参考分离型船舶操纵运动模型结构,考虑船舶操纵运动机理,建立了简化的灰箱模型;选取合适的试验对象,运用最小二乘支持向量机算法对建立的船舶操纵运动灰箱模型进行参数辨识,并通过旋回试验和Z形操纵试验测试了模型的泛化性;通过环境不确定性因素分析,构建了波浪作用力干扰模型、数据传输延时模型和感知设备误差模型,并以此为基础,生成了具有多种环境不确定性因素影响的船舶运动响应训练数据;通过仿真试验,验证了预报方法在环境不确定性因素干扰下的预报精度。研究结果表明:在引入环境不确定性因素影响的船舶操纵运动预报试验中,当感知设备误差由0逐渐提升至5%和10%时,除受较小的初始数量级因素影响的横摇速度外,其余船舶运动响应预报结果的均方根误差增幅均小于10%,预测模型的精度可以得到有效保证;而在感知设备误差达到20%的极端条件下,纵荡速度、横荡速度、艏摇速度预报误差相较于0时分别提升4.65%、15.97%、18.17%,误差增幅仍能有效控制在20%以下。可见,船舶操纵运动建模与预报方法可在一定程度上实现环境不确定性因素干扰下的高精度船舶操纵运动预报。In response to the issue of prediction accuracy of ship maneuvering motion under complicated environmental factors,a grey box identification modeling and prediction method for ship maneuvering motion under environmental uncertainty was proposed.The separated ship maneuvering motion model structure was referenced,the ship maneuvering motion mechanism was considered,and a simplified grey box model was developed.Suitable test subjects were selected,and parameter identification was conducted on the established ship maneuvering motion grey box model using the least squares support vector machine algorithm.The generalization ability was examined by means of the turning cycle tests and zigzag maneuvering tests.By analyzing the environmental uncertainty factors,the wave force interference model,data transmission delay model,and sensing device error model were constructed.Based on these models,the ship motion response training data affected by multiple environmental uncertainties were generated.Through the simulated tests,the prediction accuracy of the proposed method under environmental uncertainties was validated.Research results reveal that in ship maneuvering motion prediction tests with environmental uncertainty factors,when the sensing device error gradually increases from 0 to 5% and 10%,except for the rolling speed affected by a small initial magnitude,the root mean square errors(RMSEs) of other ship motion response prediction results increase by less than 10%,so the accuracy of the prediction model can be effectively guaranteed.Under the extreme condition with a 20% sensing device error,the prediction errors of surge speed,sway speed,and yawing speed increase by 4.65%,15.97%,and 18.17%,respectively compared to the 0 error level,so the error increase is effectively controlled below 20%.Thus,the ship maneuvering motion modeling and prediction method can achieve a high-precision prediction of ship maneuvering motion under the interference of environmental uncertainty factors to a certain extent.10 tabs,13 figs,40 re

关 键 词：水路运输 船舶操纵运动 环境不确定性 最小二乘支持向量机 建模与预报方法 验证试验 

分 类 号：U675.9[交通运输工程—船舶及航道工程]