多频未知时变扰动下的结构微振动鲁棒自适应控制

Robust adaptive control for micro-vibration under multiple unknown and time-varying disturbances

本文针对多频窄带未知和时变扰动,基于内模原理和Y-K参数化方法,提出一种反馈鲁棒自适应振动的主动控制算法。该算法通过设计PID中央鲁棒控制器,有效解决了次级通道模型未知情况下的鲁棒控制器参数设计问题。同时提出一种变步长最小均方(Variable Step Size Least Mean Square,VSSLMS)方法,可以在保证稳态误差的基础上大幅提升收敛速度,并通过系统辨识实验验证了所提VSSLMS方法相较于其他VSSLMS算法在收敛性能上的优越性。通过结构微振动主动控制实时实验,对比验证了单独采用滤波x最小均方(Least Mean Square,LMS)自适应控制算法、基于LMS算法的鲁棒自适应控制算法和基于VSSLMS算法的鲁棒自适应控制算法的抑振效果。实验结果表明,本文基于VSSLMS算法的鲁棒自适应控制算法在面向双频正弦窄带扰动以及其频谱、幅值突变情况时,都具有较好的收敛性和鲁棒性。

The paper presents a feedback robust adaptive micro-vibration control algorithm for rejection of multiple unknown and time-varying narrow band disturbances.This algorithm is based on the internal model principle and Y-K parameterization method.It solves the problem of the parameter design of center robust controller effectively when the model of secondary path is unknown.Meanwhile,a new variable step size least mean square(VSSLMS)method is proposed,and the advantages of the proposed VSSLMS method compared with other VSSLMS methods is verified by system identification.In the end,the active micro-vibra-tion real-time control experiments are carried out through an active micro-vibration control system.The effect of the filtered-x least mean square(LMS),robust adaptive control algorithm based LMS and robust adaptive control algorithm based VSSLMS are veri-fied by experimental comparison.The results of experiments show that the proposed feedback robust adaptive control algorithm has the best performance compared with the other control algorithms under various disturbances.