摘要
为了解决船舶操纵性参数辨识问题,考虑了船舶运动时域的非线性和非平稳的特点,提出了一种快速收敛迭代学习最小二乘算法,提高了参数辨识速度和精度。首先建立了船舶操纵一阶和二阶非线性运动响应模型,并将其离散化;然后采用迭代学习思想,并引入P型学习率改进了递推最小二乘算法进行参数辨识;同时分析了该算法的收敛性;最后进行了相关实验。实验结果证明了该算法的可行性和有效性。
In order to solve the problem of parameter identification of ship＇s maneuvering motion,a fast convergent iterative least square algorithm is presented,in which the nonlinear and non-stationary characteristics of the ship motion in time domain are taken into consideration.First,the first-order and second-order nonlinear response motion of the ship are estimated and discretized.Then,the parameters are identified using the proposed least square algorithm,which is improved by using iterative learning and introducing p-type learning rate.The convergence of the algorithm is also analyzed.Simulation experiments are conducted and results show that the speed and precision of parameter identification are improved using this algorithm.This demonstrates that the proposed algorithm is feasible and effective.
出处
《吉林大学学报（工学版）》
EI
CAS
CSCD
北大核心
2016年第3期897-903,共7页
Journal of Jilin University:Engineering and Technology Edition
基金
国家自然科学基金项目(51309069)
中央高校基本科研业务费项目(HEUCF150106)
关键词
计算机应用
最小二乘法
参数辨识
迭代学习
船舶操纵性
响应模型
computer application
least square method
parameter identification
interactive learning
ship maneuvering
response model
