船舶与海洋工程

基于双曲正切函数的非线性新息辨识算法

  • 宋纯羽 ,
  • 张显库 ,
  • 张国庆
展开
  • (大连海事大学  航海学院,辽宁 大连 116026)
宋纯羽(1996 — ),男,博士生,E-mail:scy_@dlmu.edu.cn;张显库*(1968 — ),男,博士,教授,博士生导师,E-mail:zhangxk@dlmu.edu.cn

收稿日期: 2021-10-08

  修回日期: 2021-11-03

  网络出版日期: 2021-11-03

基金资助

国家自然科学基金资助项目(51679024);大连市重点研究领域创新团队支持计划(2020RT08);中央高校基本科研业务费专项资金资助项目(3132021139)

Nonlinear innovation identification algorithm based on hyperbolic tangent function

  • SONG Chun-yu ,
  • ZHANG Xian-ku ,
  • ZHANG Guo-qing
Expand
  • (Navigation College, Dalian Maritime University, Dalian  116026, China)

Received date: 2021-10-08

  Revised date: 2021-11-03

  Online published: 2021-11-03

摘要

以非线性Nomoto模型作为船舶运动数学模型。同时,通过线性回归模型,类比引入遗忘因子的随机梯度算法并结合推导出的随机梯度算法,以汽车运输船“Hual Trooper”号为例进行仿真实验并对该算法进行推导和验证。结合双曲正切函数,得到非线性新息辨识理论,利用双曲正切函数处理新息。通过Matlab编程比较辨识效果,并通过计算得到辨识准确的情况。在模型建立后,可以通过观察响应比较来估计船舶动态。该设计方案具有显著的计算优势,辨识速度更快,辨识效率更高。

本文引用格式

宋纯羽 , 张显库 , 张国庆 . 基于双曲正切函数的非线性新息辨识算法[J]. 大连海事大学学报, 2022 , 48(1) : 83 -88 . DOI: 10.16411/j.cnki.issn1006-7736.2022.01.009

Abstract

The nonlinear Nomoto model was used as the mathematical model of ship motion, at the same time, by using the linear regression model, the random gradient algorithm with forgetting factor was analogized, and by combining with the derived random gradient algorithm, the simulation study was carried out in the case of the automobile transport ship “Hual Trooper” as example to deduce the algorithm and verify. By combining with the hyperbolic tangent function, the nonlinear innovation identification theory was obtained, and the hyperbolic tangent function was used to handle the innovation. The identification effect was compared by MATLAB programming, and the identification accuracy was obtained by calculation. Based on the model, the ship dynamics can be estimated by observing the response comparison. The design scheme has significant computational advantages, faster identification speed and higher identification efficiency.

文章导航

/