Collaborative Research: Integrated Parameter and Control Design

合作研究：综合参数与控制设计

• 批准号: 0300568
• 负责人: Mustafa Khammash
• 金额: $ 23.44万
• 依托单位: University of California-Santa Barbara
• 依托单位国家: 美国
• 项目类别: Continuing grant
• 财政年份: 2003
• 资助国家: 美国
• 起止时间: 2003-04-15 至 2006-03-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0300568&HistoricalAwards=false
• 关键词: Collaborative; Research; Integrated; Parameter; Control

Project SummaryMotivated by design problems i anotechnology and by key theoretical problems of interest,we consider a paradigm whereby the plant description depends on one or several parameters thatare subject to design.We propose a general framework for I tegrated Parameter and Controller(IPC)design that allows multiple performance requirements.We propose to address the problemof designing these parameters in conjunction with the feedback controller i order to achieve one ormore objectives.We then propose to address and solve the underlying optimization problem.Thetheoretical underpinnings will rely o ew results in the non-convex optimization literature thatutilize convex relaxation i order to get global optimal solutions.The potential of our approachto yield e .ective solutions to the resulting problems is demonstrated by our recent work thatapplies the proposed methods to the robust synthesis problem in 1 giving the only global optimalsolution to this important problem.Aside from its contribution to the theory of control design,the IPC design tools to be developed will be used to address several practical problems where thesimultaneous parameter and controller design can have a large and measurable impact.Speci .cally,we will address a signi .cant nanotechnology design problem i volving ultra-high resolution micro-cantilevers.This problem will take adva tage of existing expertise i the laboratory of the PIs andwill be conducted i close collaboration with the relevant industry.Based o the interdisciplinarytools eeded for IPC design theory,we also address educational goals that are made possible bythe proposed research.The intellectual merit of this proposal stems from the novel approach for a well recognizedunsolved problem of integrated parameter and controller (IPC)design.The proposal will contributesigni .cantly to optimization based methodologies in control theory solving hitherto unaddressedproblems that are of importance for practical control design.The proposal lays particular emphasison applying the theory developed to micro-cantilever based technologies in the context of IPCdesign.The theoretical paradigm is well suited and timely for furthering this technology.Apart from the theoretical impact the related technology to be developed will be applicableto practical problems i the two hundred million dollar scanning probe industry.The PIs havesigni .cant collaboration with industry and the technical knowhow developed will be guided andtransferred e .ectively to the industry.E .ort will be placed on integrating the software developedon a typical Matlab platform.This will enable wide use of the tools developed by the relatedcommunities.The main topic of the proposal provides a unique opportunity to expose the studentto the importance of plant structure that facilitates good overall plant-controller performance.Anovel course that builds o the ideas and results of this project to motivate the general philosophyand techniques of integrated parameter and control design will be developed.1

项目摘要受设计问题和感兴趣的关键理论问题的启发，我们考虑了一个范例，即对象的描述取决于一个或几个参数是受设计。我们提出了一个通用的框架，集成参数和控制器（IPC）。设计，允许多个性能要求。我们建议解决的问题，设计这些参数与反馈控制器i，以实现一个或多个目标。然后，我们提出解决和解决潜在的优化问题。理论基础将依赖于新的结果，在非-凸优化文献利用凸松弛来获得全局最优解。我们最近的工作证明了我们的方法产生所产生的问题的有效解决方案的潜力。我们最近的工作将所提出的方法应用于1中的鲁棒综合问题，除了对控制设计理论的贡献外，即将开发的IPC设计工具还将用于解决几个实际问题，其中参数和控制器设计的同时进行可能会产生很大且可测量的影响。特别是，我们将解决一个重大的纳米技术设计问题，涉及超高分辨率的微，这个问题将利用PI实验室现有的专业知识，并将与相关行业密切合作。基于IPC设计理论的跨学科工具，我们还提出了通过拟议的研究可能实现的教育目标。这一建议的智力价值源于一种新颖的方法，用于解决一个公认的未解决的综合问题。参数和控制器设计。该建议将对控制理论中基于最优化的方法学作出重大贡献，解决迄今为止尚未解决的问题，这些问题对实际控制设计具有重要意义。该建议特别强调将所发展的理论应用于微控制器设计。悬臂梁技术在IPC设计的背景下。理论范式是非常适合和及时的进一步这项技术。除了理论的影响，开发的相关技术将应用于价值2亿美元的扫描探针工业的实际问题。PI与工业界有着重要的合作，开发的技术诀窍将有效地引导和转移到工业界。E .ort将放在集成在典型的Matlab平台上开发的软件上。这将使相关社区开发的工具得到广泛使用。主要的该项目的主题提供了一个独特的机会，让学生了解工厂结构的重要性，这有助于良好的整体工厂控制器性能。将开发一门基于该项目的思想和结果的新课程，以激发综合参数和控制设计的一般哲学和技术。1