基于PZT的悬臂梁变负载自适应振动主动控制研究

发布时间：2019-04-25 12:56

【摘要】：由于柔性机械臂具有质量轻、负载自重比高和灵活性好等优点，从而在航天等领域得到了越来越广泛的应用。但是柔性臂的弹性振动问题，严重影响其快速性和准确性，这对振动抑制研究提出了挑战。在实际应用中，柔性臂的负载往往不能确定，而是在一定范围内变化，从而导致在非标称模型条件下振动抑制效果降低。为了能够在变负载情况下快速地消除柔性臂的残余振动，本课题采用Euler-Bernoulli梁作为柔性臂模型，研究了基于振动主动控制的抑振策略，，并且能够根据其末端负载的变化调整控制参数。首先，以悬臂梁为基本模型，分析了在负载自重比分别为0(即空载)、10%和20%三种情况下的振型，并利用MATLAB和ANSYS计算出前四阶固有频率值。同时，选用压电陶瓷作为致动器，建立了致动器作用下悬臂梁振动的模态运动方程。根据这一数学模型，提出了致动器位置优化配置的准则，并利用遗传算法计算出了最优位置。其次，为了使控制器中的参数能够跟随由负载改变引起的固有频率参数变化，实现自适应控制，需要根据控制器的输入和输出对悬臂梁的固有频率参数进行辨识。本文采用递推最小二乘法(RLS)设计了固有频率辨识算法，阐述了该算法辨识系统固有频率的基本原理，并推导得出了计算公式。然后，根据致动器作用下悬臂梁振动的数学模型，设计了积分振动控制器(IRC)，建立了闭环控制系统模型，优化了控制参数，并从理论上说明了该控制策略能够增大系统阻尼。通过对整个闭环系统的鲁棒性分析，说明了进行自适应控制的必要性，并建立了整个闭环系统现实自适应控制的框图。最后，搭建了基于MATLAB/Simulink和dSPACE的实时仿真控制系统，对固有频率辨识算法的准确性和IRC控制模型的有效性以及控制参数的相对最优性进行了验证。实验结果表明，固有频率辨识算法具有较高的精度，与理论分析值的误差可控制在5%以内；IRC控制模型能够比较有效地抑制三种负载自重比情况下悬臂梁的振动，使振动系统的阻尼比至少增大一倍，而且所优化的控制器参数相对最优。
[Abstract]:Due to the advantages of light weight, high load-to-weight ratio and good flexibility, flexible manipulator has been more and more widely used in aerospace and other fields. However, the elastic vibration of the flexible arm seriously affects its rapidity and accuracy, which poses a challenge to the study of vibration suppression. In practical applications, the load of the flexible arm can not be determined, but it changes within a certain range, which results in the reduction of the vibration suppression effect under the condition of non-nominal model. In order to eliminate the residual vibration of flexible arm rapidly under variable load, Euler-Bernoulli beam is used as the model of flexible arm, and the vibration suppression strategy based on active vibration control is studied in this paper. And the control parameters can be adjusted according to the change of the end load. Firstly, taking the cantilever beam as the basic model, the mode shapes of the first four order natural frequencies are calculated by using MATLAB and ANSYS when the load gravity ratio is 0 (i.e., no load), 10% and 20%, respectively. At the same time, using piezoelectric ceramics as actuators, the modal motion equations of cantilever beam vibration under actuators are established. According to this mathematical model, a criterion for optimal placement of actuators is proposed, and the optimal position is calculated by genetic algorithm. Secondly, it is necessary to identify the natural frequency parameters of the cantilever beam according to the input and output of the controller in order to make the parameters in the controller change with the natural frequency parameters caused by the load change and realize the adaptive control. In this paper, the recursive least square method (RLS) is used to design the natural frequency identification algorithm, and the basic principle of the algorithm to identify the natural frequency of the system is expounded, and the calculation formula is derived. Then, according to the mathematical model of cantilever beam vibration under the action of actuator, the integral vibration controller (IRC), is designed and the closed-loop control system model is established, and the control parameters are optimized. It is shown theoretically that the control strategy can increase the damping of the system. Through the robustness analysis of the whole closed-loop system, the necessity of the adaptive control is explained, and the block diagram of the real adaptive control of the whole closed-loop system is established. Finally, a real-time simulation control system based on MATLAB/Simulink and dSPACE is built to verify the accuracy of the natural frequency identification algorithm, the effectiveness of the IRC control model and the relative optimality of the control parameters. The experimental results show that the natural frequency identification algorithm has high accuracy, and the error between the natural frequency identification algorithm and the theoretical analysis value can be controlled less than 5%. The IRC control model can effectively suppress the vibration of the cantilever beam under three load-weight ratios, and increase the damping ratio of the vibration system by at least one time, and the optimal parameters of the controller are relatively optimal.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP241;TB535

【参考文献】