PYI: Complexity Uncertainty, Information, and Organization in Control System Design for Modern Engineering Systems

PYI：现代工程系统控制系统设计中的复杂性不确定性、信息和组织

• 批准号: 9058397
• 负责人: Carl Nett
• 金额: $ 18万
• 依托单位: Georgia Tech Research Corporation
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1990
• 资助国家: 美国
• 起止时间: 1990-09-01 至 1996-01-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9058397&HistoricalAwards=false
• 关键词: PYI; Complexity; Uncertainty; Information; Organization

The overall goal of our research program will be to make progress toward the development of control system design methodologies which support the following control system design paradigm: Minimize control system complexity subject to the achievement of a specified control accuracy in the face of system/environmental uncertainty. A key focus of this research program will be the role and use of information and organization for the achievement of this paradigm, as the nature and use of available information, and in particular how it is organized, largely determine how effectively the control system reduces the effects of system/environmental uncertainty, and what level of complexity is required to do so. The task in our research program will include the development of control-oriented system identification methods. In this work we have developed methods for identification H-infinity; in other words, the identified model converges to the unknown system in the H-infinity norm, and at each step an explicit bound is given on the H-infinity norm of the error between the identified model and the unknown system. This bound is precisely the uncertainty specification required in most currently popular approaches to robust control system design. Applications feedback will be essential for the achievement of our immediate goal of better bridging the gap that currently exists between the areas of robust control and system identification. For this purpose we have in mind an important application; namely, active control of fan/compressor stall phenomena in advanced variable-cycle aircraft gas turbine engines.

我们研究计划的总体目标是在控制系统设计方法的开发方面取得进展，该方法支持以下控制系统设计范式：在面对系统/环境不确定性时实现指定的控制精度，最大限度地降低控制系统的复杂性。 该研究计划的一个重点是信息和组织在实现这一范式中的作用和使用，因为可用信息的性质和使用，特别是信息的组织方式，在很大程度上决定了控制系统如何有效地减少系统/环境不确定性的影响，以及需要什么程度的复杂性。 我们研究计划的任务将包括开发面向控制的系统识别方法。 在这项工作中，我们开发了识别 H-无穷大的方法；换句话说，识别模型在 H 无穷范数下收敛到未知系统，并且在每一步都在识别模型和未知系统之间的误差的 H 无穷范数上给出显式界限。 这个界限正是当前最流行的鲁棒控制系统设计方法所需的不确定性规范。 应用程序反馈对于实现我们的近期目标至关重要，即更好地弥合当前鲁棒控制和系统识别领域之间存在的差距。 为此，我们想到了一个重要的应用；即先进可变循环飞机燃气涡轮发动机中风扇/压缩机失速现象的主动控制。