首页> 外文学位 >Repetitive control of hysteretic systems using robust H-infinity controller.
【24h】

Repetitive control of hysteretic systems using robust H-infinity controller.

机译:使用强大的H-infinity控制器重复控制滞后系统。

获取原文
获取原文并翻译 | 示例

摘要

Unique properties of smart materials have enabled them to successfully replace conventional actuators and sensors. However, their effectiveness and applicability in control is hindered due to the presence of inherent hysteresis. In this thesis a systematic approach is developed to model this hysteresis in order to design a robust Hinfinity controller for tracking applications in one such smart material based actuator called Thunder actuator.;The main objective of this work is to develop a controller to get rid of the delay, caused by the inherent hysteresis of the system, while following a commanded reference signal. To this end a mathematical model of the hysteretic system is developed and robust Hinfinity controller is developed for micro-position tracking control of the Thunder actuator. Tracking performance of compensated system is compared with open loop system for triangular and sinusoidal reference signals. Robustness of the developed Hinfinity controller is demonstrated and validated using experimental results.;The first half of the thesis deals with the modeling of the Thunder actuator. Classical Preisach model is used to capture the hysteresis of the Thunder actuator. Congruency property and wiping-out property are verified and a classical Preisach model is developed for Thunder actuator using experimental data from first-order reversal curves. The Preisach model developed has been experimentally validated. Using Preisach model an equivalent time-invariant uncertain second order model is developed for Thunder actuator assuming variable stiffness and damping coefficients.;The second half of the thesis involves designing a robust Hinfinity controller for the uncertain Thunder actuator model and validating the designed controller experimentally. Design of the controller is posed as a linear matrix inequality (LMI) problem and a mixed sensitivity optimal controller is calculated by minimizing the Hinfinity norm of the closed-loop system. Minimal tracking error is the performance criteria and noise rejection along with stability of the controller to time-invariant uncertainties are the robustness criteria in mixed sensitivity Hinfinity design. Full state robust controller is calculated for the nominal plant of the Thunder actuator. Robustness of the Hinfinity controller is verified through simulation and experimental studies. Experimental results demonstrate that the hysteresis loop-width of the Thunder actuator, in tracking sinusoidal and triangular signals, can be significantly reduced by using a robust- Hinfinity controller: this clearly demonstrates the effectiveness of the proposed control algorithm.
机译:智能材料的独特性能使其能够成功替代传统的执行器和传感器。然而,由于固有滞后的存在,阻碍了它们在控制中的有效性和适用性。在本文中,开发了一种系统的方法来对此滞后进行建模,以便设计一种鲁棒的Hinfinity控制器,以跟踪一种基于智能材料的智能执行器Thunder Thunder执行器中的应用。该工作的主要目的是开发一种控制器来摆脱在遵循命令的参考信号时,由系统固有的磁滞引起的延迟。为此,开发了磁滞系统的数学模型,并开发了用于迅雷执行器的微位置跟踪控制的鲁棒Hinfinity控制器。对于三角和正弦参考信号,将补偿系统的跟踪性能与开环系统进行了比较。实验结果证明了所开发的Hinfinity控制器的鲁棒性。论文的前半部分是对Thunder执行器的建模。经典的Preisach模型用于捕获Thunder执行器的磁滞。验证了一致性特性和擦除特性,并使用一阶反转曲线的实验数据为Thunder执行器开发了经典的Preisach模型。开发的Preisach模型已通过实验验证。利用Preisach模型,为刚性推力和阻尼系数可变的Thunder执行器开发了一个等效的时不变不确定二阶模型。本文的第二部分涉及为不确定的Thunder执行器模型设计一个鲁棒的Hinfinity控制器,并通过实验验证了所设计的控制器。控制器的设计被提出为线性​​矩阵不等式(LMI)问题,并且通过最小化闭环系统的Hinfinity范数来计算混合灵敏度最优控制器。最小的跟踪误差是性能标准,噪声抑制以及控制器对时不变不确定性的稳定性是混合灵敏度Hinfinity设计中的鲁棒性标准。针对Thunder执行器的标称设备计算出全状态鲁棒控制器。 Hinfinity控制器的鲁棒性已通过仿真和实验研究得到验证。实验结果表明,通过使用鲁棒的Hinfinity控制器,可以大大减少Thunder促动器在跟踪正弦和三角信号中的磁滞回线宽度:这清楚地证明了所提出的控制算法的有效性。

著录项

  • 作者

    Pasala, Dharma Theja Reddy.;

  • 作者单位

    Rice University.;

  • 授予单位 Rice University.;
  • 学科 Engineering Civil.;Engineering Mechanical.
  • 学位 M.S.
  • 年度 2010
  • 页码 104 p.
  • 总页数 104
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号