There has been a great deal of excitement recently over the development of a theory for explicitly linearizing the input-output response of a nonlinear system using state feedback. One shortcoming of this theory is th...
详细信息
There has been a great deal of excitement recently over the development of a theory for explicitly linearizing the input-output response of a nonlinear system using state feedback. One shortcoming of this theory is the inability to deal with non-minimum phase nonlinear systems. Highly maneuverable jet aircraft, such as the V/STOL Harrier, belong to an important class of a slightly non-minimum phase nonlinear systems. The non-minimum phase character of these aircraft is clue in part to a slight coupling between rolling moments and lateral accelerations. In this paper, we show that, while straightforward application of the linearization theory to a non-minimum phase system results in a system with a linear input-output response but unstable internal dynamics, designing a feedback control based on a minimum phase approximation to the true system results in a system with desirable properties such as bounded tracking and asymptotic stability.
In this paper several techniques are given for the identification of stable LSI discrete time systems from input-output data. Explicit H ∞ norm error bounds are given and convergence in the noise free and the unifor...
详细信息
In this paper several techniques are given for the identification of stable LSI discrete time systems from input-output data. Explicit H ∞ norm error bounds are given and convergence in the noise free and the uniformly bounded deterministic noise case are established. The assumptions made on the unknown system are minimal and are limited throughout the paper to a lower bound on the decay rate of the unknown system and an upper bound on the gain of the unknown system. Given this information an experiment and a construction are specified: the experiment involves obtaining a specified number of frequency measurements of the unknown systems at a set of specified frequencies; the construction uses this experimental data to generate an identified model with prescribed H ∞ norm error tolerance to the unknown system. The resulting model identification process is highly efficient from a computational point of view.
This paper is a condensed version of recent work [17] wherein several fundamental practical concerns pertaining to the use of linear, time-invariant, distributed parameter systems as models of physical processes are i...
详细信息
This paper is a condensed version of recent work [17] wherein several fundamental practical concerns pertaining to the use of linear, time-invariant, distributed parameter systems as models of physical processes are investigated. In particular, it is argued here that certain common distributed parameter systems exhibit properties which make them unsuitable for use as plant models in the design of feedback compensation. Classes of distributed parameter systems exhibiting these properties are defined and are said to be ill-posed for use in feedback compensator design. The structure of these classes is investigated within an extremely general mathematical framework which imposes no a priori restrictions on the class of distributed parameter systems under consideration. Within this framework necessary and sufficient conditions are given which completely characterise the classes of ill-posed models and establish their equivalence. These conditions are concretely illustrated by way of applications to several well-known flexible structure models. Surprisingly, several of these models are determined to be ill-posed for use in feedback compensator design.
The GE multidisciplinary expert-aided analysis and design (MEAD) project, which involves the integration of several computer-aided controlengineering (CACE) packages under a supervisor which coordinates the execution...
详细信息
The GE multidisciplinary expert-aided analysis and design (MEAD) project, which involves the integration of several computer-aided controlengineering (CACE) packages under a supervisor which coordinates the execution of these packages with a database manager, an expert system, and an advanced user interface, is discussed. The principal components are, in functional terms: a supervisor, which integrates the underlying CACE packages and coordinates all activity within the GE MEAD computer program (GMCP); an expert system shell and rule bases for expert aiding of specific procedures to relieve the user from unnecessary low-level detail; a database manager for tracking system models that evolve over time along with associated results; and a user interface that facilitates access to the CACE package capabilities by permitting the user to work in several modalities, i.e. menu/forms style, using GE MEAD commands, the core packages' native commands, or the GE MEAD macro facility. The operation of the GMCP, including the user interface, expert system, and database manager, is illustrated by examples.< >
An iterative algorithm is presented to compute lower bounds for the structured singular value ( mu ). The algorithm resembles a mixture of power methods for eigenvalues and singular values, since the structured singul...
详细信息
An iterative algorithm is presented to compute lower bounds for the structured singular value ( mu ). The algorithm resembles a mixture of power methods for eigenvalues and singular values, since the structured singular value can be viewed as a generalization of both. If the algorithm converges, a lower bound for mu results. The authors prove that mu is always an equilibrium point of the algorithm. However, since in general there are many equilibrium points, some heuristic ideas to achieve convergence are presented. Extensive numerical experience with the algorithm is discussed.< >
In this paper we present a theoretical basis for an integrated approach to the design of reliable control systems. In this approach, the control and diagnostic modules of a reliable control system are designed togethe...
详细信息
In this paper we present a theoretical basis for an integrated approach to the design of reliable control systems. In this approach, the control and diagnostic modules of a reliable control system are designed together, instead of independently, thereby accounting for the interactions which occur between these two modules in a functioning reliable control system. The approach makes use of a significant generalisation of the familiar 2-parameter controller known as the 4-parameter controller. This controller has two vector inputs and not one, but two vector outputs; correspondingly, this controller is comprised of not two, but four matrix parameters. The additional controller output is monitored to detect and isolate sensor and actuator faults, thereby providing the controller with diagnostic capabilities, or "intelligence," in addition to its control capabilities. Using a parameterization of all stabilising 4-parameter controllers, the fundamental limitations and inherent tradeoffs governing the design of reliable control systems are clarly delineated, with special attention given the issues of uncertainty and control/diagnostic interaction. Several means for resolving these desig tradeoffs are proposed, and illustrated by way of non-trivial examples. These results are expected to pave the way for the achievement of improved overall control/dignostic performance in reliable control systems, and also to pave the way for the achivement of significant complexity reductions in both the design and implementation of reliable control systems.
作者:
SWENSON, ENMAHINSKE, EBSTOUTENBURGH, JSCapt. Erick N. Swenson
USNR (Ret.):is a project manager for special projects in the Surface Ship Systems Division Hughes Aircraft Company Fullerton Calif where he has been employed since his retirement from the U.S. Navy in 1975. Originally trained as an electronics technician during WWII in the Captain Eddy program he later received a BS degree in electrical engineering from the University of Rochester Rochester N. Y. in 1950. Subsequent engineering education was received at the University of Pittsburgh Pittsburgh Penn. and the Naval Postgraduate School Monterey Calif. After commissioning he was ordered to duty as the electronics division officer on the USSMissouri(BB-63) and electronics ships superintendent at Hunters Point Naval Shipyard San Francisco Calif. When the design of the Naval Tactical Data System began in the mid-1950s Lt. (j.g.) Swenson was ordered to the Bureau of Ships Navy Department Washington D.C. as the junior engineering duty only officer assigned to the project. From 1962 to 1965 LCdr. Swenson was assigned as the BuShips technical representative on the program at Remington Rand Univac St. Paul Minn. For the next ten years he returned to BuShips/NavSea/NAVSEC as the NTDS project officer. During this time the project expanded considerably foreign military sales were heavily involved and interoperability with other services and countries were established. His final effort on active duty was to instigate the redesign of the previousSpruanceclass destroyers into the newerAdmiral Kiddclass improvement program. He is a registered professional electrical engineer in the State of California listed inWho's Who in the Worldis a life member of ASNE and chairman of the Long Beach/Greater LA Section. Capt. Edmund B. Mahinske
USN (Ret.):is an alumnus of the U.S. Naval Academy the Massachusetts Institute of Technology and the Harvard Business School. His technical background is in electronics and he specialized in the management of programs involving the application of comp
A little over thirty years ago, a group of naval engineers were assembled by the Bureau of Ships to develop a new system approach to the combat information center (CIC). The CIC of World War II, with its “grease pen...
详细信息
A little over thirty years ago, a group of naval engineers were assembled by the Bureau of Ships to develop a new system approach to the combat information center (CIC). The CIC of World War II, with its “grease pencil” plots and voice telling of tactical information from sensors and other ships, could no longer provide the timely, coordinated reaction to postwar threats. This project group led the Navy into the new world of large-scale, high-speed digital electronics and into a new mode of conducting naval warfare as well. There were no off-the-shelf computers of the requisite capability, size and reliability; what were available were monstrous vacuum tube computers. There were no display equipments that were “conversant” in both the digital language of the computer and the analog language of the sensors and the weapon systems. Who ever heard, at that time, of a computer running a tactical communication net automatically? It was hard enough to find sufficient numbers of engineers who knew what a digital computer was. This paper, by three naval engineers in the implementing engineering office, depicts the evolvement of the Naval Tactical Data Systems (NTDS) as they saw it. It discusses the problems that stemmed from the transition from the old world of analog into the new digital world, the system concepts that steered the development; the key decisions that were made; new electronic equipment and processes that became necessary; and the need of the mangagement to face the real world of deadlines, ship schedules and operational requirements.
This paper addresses some fundamental concerns pertaining to the subject of practical controller design for LTI plants with possibly distributed parameters. The well-posedness of a general LTI model is introduced in t...
详细信息
This paper addresses some fundamental concerns pertaining to the subject of practical controller design for LTI plants with possibly distributed parameters. The well-posedness of a general LTI model is introduced in terms of the existence of bandlimited stabilizing compensation for the given model. The well posedness of a model is critical from an implementation standpoint for the application of any model-based controller design methodology. The problem addressed in this paper is to characterize LTI distributed parameter models that are well posed with respect to bandlimited stabilising control. The problem is posed on a framework consisting of a novel class of convolution kernels and a completely general control system configuration. These two aspects of the problem formulation place no a priori restriction on the class of LTI distributed paramater systems under consideration or the control design methodology used. Indeed, this framework allows the modelling of LTI distributed parameter systems including point sensing, boundary actuation and effects of initial conditions. Within this framework the result obtained are algebraic in nature. A characterization of well-posed models is given in the form of necessary conditions for the existence of a bandlimited stabilizing controller. Also a restriction of the class of all linear distributed parameter systems under consideration is derived which exhibits this condition.
This paper presents an overview of recent developments in the theory of singular perturbations and time scales (SPaTS) in discrete control systems. The focus is in three directions: modeling, analysis, and control. Fi...
详细信息
This paper presents an overview of recent developments in the theory of singular perturbations and time scales (SPaTS) in discrete control systems. The focus is in three directions: modeling, analysis, and control. First, sources of discrete models and the effect of the discretizing interval on the modeling are reviewed. Then the analysis of two-time scale systems is presented to bring out typical characteristic features of SPaTS. Finally, in the control of the two-time scale systems, the important issue of multirate sampling is addressed. The bibliography containing over 100 titles is included.
In this paper several fundamenal results from the theory of linear state-space systems in finite-dimensional space are extended to encompass a class of linear state-space systems in infinite-dimensional space; specifi...
详细信息
In this paper several fundamenal results from the theory of linear state-space systems in finite-dimensional space are extended to encompass a class of linear state-space systems in infinite-dimensional space; specifically, we generalize those results from the finite-dimensional theory pertaining to the relationship between input-output and internal stability, the problem of dynamic output feedback stabilization, and more generally, the concept of joint stabilizability/detectability. In the course of doing so a complete structural characterization of jointly stabilizable/detectable systems is obtained. These results are distinguished by their generality, as we consider a large class of linear state-space systems, assuming only that (i) the evolution of the state is governed by a strongly-continuous semigroup of bounded linear operators, (ii) the state-space is a Hilbert space, (iii) the input and output spaces are finite-dimensional, and (iv) the sensing and control operators are bounded. In turn, general conclusions regarding the fundamental structure of control-theoretic problems in infinite-dimensional space may be drawn from these results.
暂无评论