具有迟滞非线性特性的压电陶瓷作动器的建模与控制

发布时间：2019-03-20 13:30

【摘要】：压电陶瓷作动器作为智能材料的一种,具有压电效应好、分辨率高、反应速度快以及成本低等特点,在精密仪器、微位移和微控制等领域得到了广泛的应用。压电陶瓷作动器本身所具有的迟滞非线性特性是一种强非线性,且这种迟滞非线性又是率相关的,这些都严重影响了控制精度,甚至导致控制系统发生震荡,制约着它的应用。本文首先建立压电陶瓷作动器的Hammerstein模型,然后深入研究了压电陶瓷作动器的跟踪控制问题,主要包括以下几个方面的内容。第一个部分在分析总结了国内外对于压电陶瓷作动器建模的基础上,分析各种模型的优缺点和应用范围,深入学习了 Hammerstein模型理论。基于现有的压电陶瓷作动器实物和相对应的实验平台,采集压电陶瓷作动器输入输出数据,经过系统参数辨识建立了 Hammerstein模型,且和压电作动器实物相比模型的相对误差很小,从实验上验证了建模方法的准确性与可行性。第二个部分首先是使用工业上最常用的PID控制器对系统进行控制,然后通过分析结果可以知道可能是由于参数调节不准确或者是PID控制器本身的局限性这两个原因导致控制效果不好。然后引入神经网络PID控制器实现参数的在线调节,然而跟踪控制效果依然不好。从而可以分析得出,由于PID控制器本身的局限性导致对迟滞非线性系统跟踪控制效果不好。因此在系统中加入了前馈环节,通过前馈逆补偿消除模型的迟滞非线性部分,即引入前馈逆补偿+PID控制实现对压电陶瓷作动器的跟踪控制,然后进行实验验证。第三个部分引入了自抗扰控制器,首先是将压电陶瓷作动器的模型等效为一个二阶系统,然后利用这个二阶系统设计出自抗扰控制器,对压电陶瓷作动器进行跟踪控制。最后的仿真和实验结果都表明自抗扰控制器可以比较好的跟踪控制压电陶瓷作动器,最后比较了前馈逆补偿+PID控制器和自抗扰控制器各自的优缺点和适用范围,它们各自有自己的特点,一个是控制精度高,一个是应用简单适合工业应用。
[Abstract]:As a kind of intelligent material, piezoelectric ceramic actuator has the characteristics of good piezoelectric effect, high resolution, fast reaction speed and low cost. It has been widely used in the fields of precision instruments, micro-displacement and micro-control. The hysteresis nonlinear characteristic of piezoelectric ceramic actuator itself is a kind of strong nonlinearity, and this hysteresis nonlinearity is rate dependent, which seriously affects the control accuracy and even leads to vibration of the control system, which restricts its application. In this paper, the Hammerstein model of piezoelectric actuator is established, then the tracking control problem of piezoelectric actuator is deeply studied, including the following aspects. In the first part, on the basis of analyzing and summarizing the modeling of piezoelectric ceramic actuator at home and abroad, the advantages, disadvantages and application scope of various models are analyzed, and the theory of Hammerstein model is deeply studied. Based on the existing piezoelectric actuators and corresponding experimental platforms, the input and output data of the piezoelectric actuators are collected. The Hammerstein model is established by system parameter identification, and the relative error of the model is very small compared with the piezoelectric actuators. The accuracy and feasibility of the modeling method are verified by experiments. The second part begins with the use of the most commonly used PID controllers in the industry to control the system. Then the analysis results show that the control effect may be poor due to the inaccurate parameter adjustment or the limitation of the PID controller itself. Then the neural network PID controller is introduced to realize the on-line adjustment of the parameters, however, the tracking control effect is still not good. As a result, it can be concluded that the tracking control effect of the PID controller is not good because of the limitation of the controller itself. Therefore, the feed-forward link is added to the system, and the hysteresis nonlinear part of the model is eliminated by the feed-forward inverse compensation, that is, the forward-feed inverse compensation PID control is introduced to realize the tracking control of the piezoelectric ceramic actuator, and then the experimental verification is carried out. In the third part, the auto-disturbance rejection controller is introduced. Firstly, the model of the piezoelectric ceramic actuator is equivalent to a second-order system, and then the self-disturbance rejection controller is designed with this second-order system to track and control the piezoelectric ceramic actuator. The simulation and experimental results show that the ADRC controller can well track and control the piezoelectric ceramic actuator. Finally, the advantages and disadvantages of the feedforward inverse compensation PID controller and the ADRC controller are compared. They have their own characteristics, one is high control accuracy, the other is simple application suitable for industrial applications.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM282

【参考文献】