Receptance-based robust eigenstructure assignment
Receptance-based robust eigenstructure assignment
复制标题
DOI：
10.1016/j.ymssp.2020.106697
复制
发表时间：
2020-06
影响因子：
8.4
通讯作者：
Liam J. Adamson;S. Fichera;J. Mottershead
中科院分区：
文献类型：
--
作者：
Liam J. Adamson;S. Fichera;J. Mottershead
考虑了基于接收的主动控制技术对拟合传递函数矩阵中的不确定参数的鲁棒性。由不确定的开环极点、零点和缩放参数引起的指定闭环极点的可变性通过本研究中推导的分析灵敏度公式进行量化。即使对于大量随机参数，灵敏度公式也显示出计算效率，并且仅需要测量的接收，从而保留了基于接收的技术的无模型优势。基于进化的全局优化过程用于执行特征结构分配，以便最大化由度量定义的鲁棒性。鲁棒性度量旨在衡量每个闭环极点及其各自的实部和虚部的相对重要性。所提出的技术在多自由度系统上进行了数值测试。结果表明，在单输入和多输入系统中，可以通过优化选择一组闭环极点来提高鲁棒性。然而，确定系统的闭环特征向量在不确定性传播中起着重要作用，因此，由于多输入系统可以独立地分配闭环极点和特征向量，因此多输入系统能够更大程度地减少不确定性传播。
The robustness of receptance-based active control techniques to uncertain parameters in fitted transfer function matrices is considered. Variability in assigned closed-loop poles, which arises from uncertain open-loop poles, zeros, and scaling parameters, is quantified by means of analytical sensitivity formulae, which are derived in this research. The sensitivity formulae are shown to be computationally efficient, even for a large number of random parameters, and require only measured receptances, thereby preserving the model-free superiority of receptance-based techniques. An evolution-based global optimisation procedure is used to perform eigenstructure assignment so that the robustness, as defined by a metric, is maximised. The robustness metric is designed to scale the relative importance of each closed-loop pole and their respective real and imaginary parts. The proposed technique is tested numerically on a multi-degree-of-freedom system. It is shown that, in both single- and multiple-input systems, it is possible to increase the robustness by optimally selecting a set of closed-loop poles. However, it is determined that the closed-loop eigenvectors of the system play a significant role in the propagation of uncertainty and hence, since multiple-input systems may independently assign both closed-loop poles and eigenvectors, multiple-input systems are able to reduce the uncertainty propagation to a greater extent.