• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2016, Vol. 52 ›› Issue (12): 205-210.doi: 10.3901/JME.2016.12.205

• 交叉与前沿 • 上一篇    

扫码分享

PBP驱动器率相关迟滞特性研究及其线性化控制*

胡凯明, 文立华   

  1. 西北工业大学航天学院 西安 710072
  • 出版日期:2016-06-15 发布日期:2016-06-15
  • 作者简介:胡凯明(通信作者),男,1985年出生,博士研究生。主要研究方向为智能材料驱动器及其控制。E-mail:hzhkm1985@126.com;文立华,男,1964年出生,教授,博士研究生导师。主要研究方向为飞行器结构动力学分析与优化,智能材料结构振动主动控制、结构振动与噪声控制。E-mail:lhwen@nwpu.edu.cn

Research on Rate-dependent Hysteresis Characteristics of PBP Actuators and Its Linearization Control

HU Kaiming, WEN Lihua   

  1. School of Astronautic, Northwestern Polytechnical University, Xi’
    an 710072
  • Online:2016-06-15 Published:2016-06-15

摘要: 后屈曲预压缩压电双晶片(Post-buckling pre-compression, PBP)驱动器作为一种大行程压电舵机驱动器,存在着严重的率相关迟滞现象。为了使PBP驱动器能够作为具有较高控制精度的微小型飞行器舵机驱动器,利用基于Bouc-Wen模型的Hammerstein率相关迟滞模型对其进行参数识别,并通过试验验证了该模型能够较好地预测PBP驱动器的率相关迟滞特性;在此基础上为PBP驱动器设计一种具有在线自适应能力的前馈单神经元PID复合线性化控制器,在多种单复合频率信号作用下对其控制回路进行位移跟踪半实物仿真试验,并与基于径向基函数(Radial basis function, RBF)神经网络PID控制器进行对比,结果表明前馈单神经元PID控制器具有更快的响应速度和更高的控制精度。

关键词: Bouc-Wen迟滞模型, Hammerstein模型, PBP驱动器, 率相关迟滞, 前馈自适应PID复合控制, 神经网络

Abstract: As a kind of high displacement piezoelectric bimorph rudder actuator, the post-buckling pre-compression(PBP) actuators have a serious rate-dependent hysteresis phenomenon. In order to make the PBP actuators can be used as micro flight actuators with high control precision, parameter identification is carried out by the Hammerstein rate-dependent model based on Bouc-Wen hysteresis model, the experimental results show that rate-dependent hysteresis of the PBPs can be well predicted by this model. On the above foundation, a composite controller, which consisting of an adaptive single neuron PID and a feedforward controller, is designed for the PBP actuators, and then the displacement tracking hardware-in-loop experiments of control loop under various single and multi-frequency signals are carried out, and compared with the control effect of radial basis function neural net PID controller, the results show that the feedforward and single neuron PID composite controller has relatively faster response speed and relatively higher control precision.

Key words: Bouc-Wen hysteresis model, feedforward and adaptive PID composite controller, Hammerstein model, neural net, post-buckling pre-compression(PBP) actuator, rate-dependent hysteresis

中图分类号: