自适应主动振动控制仿真分析

摘要

In FxLMS adaptive algorithm, secondary path can affect the diagonality of the filtered self-correlation matrix and the divergence of the eigenvalues, causing a slower convergence for the algorithm. In this paper, an improved FxLMS algorithm——SOFxLMS algorithm is proposed to reduce the influence of the secondary path and increase the convergence speed in active vibration control. The active vibration control simulation models based on the plain FxLMS algorithm and SOFxLMS algorithm are set up respectively in Simulink. On the promise of ensuring the stability of the system, the two kinds of algorithms are compared by simulation analysis under the same circumstances. The results show that the SOFxLMS algorithm has better control effect than that of FxLMS algorithm under the conditions of lower order filter and larger iteration step size.%在FxLMS自适应算法中,次级通道会影响输入信号自相关矩阵的可对角化性和特征值扩散度,从而影响算法的收敛速度.为了降低次级通道对输入信号的影响,加快算法的收敛,提出将改进的FxLMS算法—SOFxLMS算法应用于振动主动控制研究中.在Simulink中分别搭建基于两种算法的主动振动控制仿真结构,基于系统稳定的前提,在相同条件下对两种算法进行仿真对比分析.仿真结果表明:在迭代步长较大、滤波器阶数较少的条件下,SOFxLMS算法具有更好的振动控制效果.