Optimal design of excitation signal for identification of nonlinear ship manoeuvring model

Optimal design of excitation signal for identification of nonlinear ship manoeuvring model
复制标题

非线性船舶操纵模型辨识激励信号优化设计

DOI:
10.1016/j.oceaneng.2019.106778
复制
发表时间:
2020-01
期刊:
影响因子:
5
通讯作者:
Zao-Jian Zou
Zao-Jian Zou
中科院分区:
工程技术2区
文献类型:
--
作者:
Zi-Hao Wang;Carlos GuedesSoares;Zao-Jian Zou

文献摘要

参考文献

相似文献

为了提高船舶操纵模型辨识的稳定性和准确性,提出了一种激励信号的优化设计方案,以确定能够提供最大动态信息的训练数据。选择一个多电平伪随机序列作为优化目标,以最大限度地覆盖非线性动态特性。信息量用D-最优性准则量化,最优解用蚁群算法计算。对优化后的信号进行了蒙特卡罗分析和泛化验证。通过与广泛使用的之字形操纵信号的比较,证明了优化激励信号的优越性。在存在噪声干扰的情况下,优化训练数据的应用降低了参数估计的方差,提高了辨识模型的泛化能力,尤其是对涌浪方程的辨识。
An optimal design scheme of excitation signals is presented to determine the training data that provides the maximum dynamic information to improve the stability and accuracy of the identification of ship manoeuvring models. A multi-level pseudo-random sequence is selected as the optimized object for covering the maximum nonlinear dynamic characteristics. The amount of information is quantified by the D-optimality criterion, and the optimal solution is calculated by the ant colony optimization algorithm. Monte Carlo analysis and generalization validation are conducted to evaluate the optimized signals. The superiority of the optimized excitation signal is demonstrated by comparing with the widely used zigzag manoeuvre signal. In the presence of noise interference, the application of the optimized training data reduces the variance of parameter estimation and improves the generalization ability of the identified model, especially for the surge equation.
DOI: 10.1007/978-3-642-01020-0_2
发表时间: 2009-04
期刊: --
影响因子: --
作者:
T. Stützle
通讯作者: T. Stützle
DOI: 10.1007/978-1-4899-7687-1_22
发表时间: 2017
期刊: --
影响因子: --
作者:
Marco Dorigo;M. Birattari
通讯作者: Marco Dorigo;M. Birattari
DOI: 10.1145/2001858.2002122
发表时间: 2011-07
期刊: Proceedings of the 13th annual conference companion on Genetic and evolutionary computation
影响因子: --
作者:
Marco Dorigo;M. Birattari
通讯作者: Marco Dorigo;M. Birattari
DOI: 10.1109/joe.2015.2460871
发表时间: 2016-07
影响因子: 4.1
作者:
L. Perera;P. Oliveira;C. Guedes Soares
通讯作者: L. Perera;P. Oliveira;C. Guedes Soares
DOI: 10.1007/978-3-540-39930-8_5
发表时间: 2004
期刊: --
影响因子: --
作者:
V. Maniezzo;L. Gambardella;Fabio de Luigi
通讯作者: V. Maniezzo;L. Gambardella;Fabio de Luigi