Analysis and implementation of a structural vibration control algorithm based on an IIR adaptive filter
Analysis and implementation of a structural vibration control algorithm based on an IIR adaptive filter
复制标题
基于IIR自适应滤波器的结构振动控制算法分析与实现
DOI：
10.1088/0964-1726/22/8/085008
复制
发表时间：
2013
影响因子：
4.1
通讯作者：
Wang Xiaohua
中科院分区：
文献类型：
--
作者：
Huang Quanzhen;Luo Jun;Li Hengyu;Wang Xiaohua
随着大型柔性结构在航天器上的广泛应用，此类结构的振动控制问题已成为重要的设计问题。滤波-X最小均方(FXLMS)算法是目前自适应滤波振动主动控制中最常用的算法。它假定干扰源是可以测量的，并且干扰源被认为是输入到控制器的参考信号。然而，在实际控制系统中，这种假设并不准确，因为它没有考虑参考信号对输出反馈信号的影响。提出了一种基于无限冲激响应(IIR)滤波器结构的自适应振动主动控制算法(FULMS，Filtered-U Least Mean Square)。该算法基于FXLMS算法框架，用IIR滤波器代替有限冲激响应(FIR)滤波器。本文重点介绍了控制器的结构设计、FULMS滤波控制方法的过程、实验模型对象的设计以及整个控制系统的实验平台搭建。对FXLMS算法和FULMS算法进行了理论分析和实验验证。结果表明，FULMS算法收敛速度快，控制效果好。FULMS控制器的设计是可行和有效的，在航天工程的实际应用中具有较大的应用价值。
With the wide application of large-scale flexible structures in spacecraft, vibration control problems in these structures have become important design issues. The filtered-X least mean square (FXLMS) algorithm is the most popular one in current active vibration control using adaptive filtering. It assumes that the source of interference can be measured and the interference source is considered as the reference signal input to the controller. However, in the actual control system, this assumption is not accurate, because it does not consider the impact of the reference signal on the output feedback signal. In this paper, an adaptive vibration active control algorithm based on an infinite impulse response (IIR) filter structure (FULMS, filtered-U least mean square) is proposed. The algorithm is based on an FXLMS algorithm framework, which replaces the finite impulse response (FIR) filter with an IIR filter. This paper focuses on the structural design of the controller, the process of the FULMS filtering control method, the design of the experimental model object, and the experimental platform construction for the entire control system. The comparison of the FXLMS algorithm with FULMS is theoretically analyzed and experimentally validated. The results show that the FULMS algorithm converges faster and controls better. The design of the FULMS controller is feasible and effective and has greater value in practical applications of aerospace engineering.