Nonlinear Repetitive Control

非线性重复控制

• 批准号: 9800294
• 负责人: Brad Paden
• 金额: --
• 依托单位: University of California-Santa Barbara
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1998
• 资助国家: 美国
• 起止时间: 1998-04-15 至 2001-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9800294&HistoricalAwards=false
• 关键词: Nonlinear; Repetitive; Control

High-speed diamond turning of eyeglass lenses, noncircular machining of camshafts and pistons, high-speed scanning and vibratory probing in atomic force microscopy (AFM), and metal rolling are examples of industrial processes which can benefit from repetitive control algorithms. Motivated by the nonlinearities in optical turning and AFM control, this research project addresses basic challenges in nonlinear repetitive control. In addition to research on basic theory involving the existence of periodic motions, stability results, and design methods, specific nonlinear problems in optical turning and AFM control are addressed. The project involves collaborations with DAC Vision, Inc. to understand the diamond turning of polycarbonate eyeglass lenses. The process requires quasi-periodic tool trajectories and suffers from significant actuator nonlinearly. Following the initial investigations into the nature of nonlinear periodic motions in single-input single-output (SISO) repetitive control systems, extension to MIMO systems, possible connections with normally hyperbolic manifold theory, and the onset of chaos in nonlinear repetitive control will be investigated. Modeling and simulation of the DAC Vision diamond turning process will be accomplished in this same period followed by control design and implementation of controllers on DAC's machines. Instructional development in UCSB's graduate control lab in collaboration with Magnetic Moments, a supplier of educational control experiments is also a part of this project. The PI will incorporate repetitive control design into this teaching lab for the outbalancing of a magnetically suspended rotor. Later in this program, repetitive control problems in atomic force microscopy (AFM) are to be addressed with the aim of improving the imaging speed of these devices. The specific problem to be addressed is the control of AFM's in tapping mode. At high tapping and scan rates, AFM's can enter into chaotic motions that degrade image quality significantly. One of the aims of this research is to control the onset of chaos and thus improve system performance.

眼镜镜片的高速金刚石车削、凸轮轴和活塞的非圆加工、原子力显微镜（AFM）中的高速扫描和振动探测以及金属轧制都是可以受益于重复控制算法的工业过程的例子。受光学车削和AFM控制中的非线性的启发，本研究项目解决了非线性重复控制的基本挑战。除了涉及周期运动的存在，稳定性结果和设计方法的基础理论研究，在光学车削和AFM控制的具体非线性问题得到解决。该项目涉及与DAC Vision，Inc.的合作。了解聚碳酸酯镜片的金刚石车削。该过程需要准周期的工具轨迹，并遭受显着的致动器非线性。在对单输入单输出（SISO）重复控制系统中的非线性周期运动的性质进行初步研究之后，将研究MIMO系统的扩展，与正常双曲流形理论的可能联系，以及非线性重复控制中的混沌开始。DAC视觉金刚石车削过程的建模和仿真将在同一时期完成，随后是DAC机器上控制器的控制设计和实施。UCSB的研究生控制实验室与Magnetic Moments（一家教育控制实验供应商）合作的教学开发也是该项目的一部分。PI将把重复控制设计纳入这个教学实验室的磁悬浮转子的平衡。在本计划的后期，原子力显微镜（AFM）中的重复控制问题将得到解决，目的是提高这些设备的成像速度。要解决的具体问题是在轻敲模式下AFM的控制。在高的轻敲和扫描速率下，AFM可能会进入混沌运动，从而显着降低图像质量。本研究的目的之一是控制混沌的发生，从而提高系统的性能。