摘要:

The characteristic locus method, in addition to its role in the assessment of system stability, is also an effective approach for evaluating tracking accuracy, interaction and dynamic/static loop performance. However, with regard to the evaluation of the latter three, there exists a degree of ambiguity and personal judgement, in addition to a requirement for aprioriassumptions about proper problem conditioning. This paper stipulates quantitative bounds/measures of accuracy, interaction and loop assessment that identify the pertinent open-loop system factors affecting the particular performance aspect, and thus provide unambiguous measures to aid system design. Additionally, eigenvector scaling is introduced in order to tighten the bounds.

ISSN:0020-7179    

DOI:10.1080/00207178808906065

出版商:Taylor & Francis Group    

年代:1988

数据来源: Taylor

摘要:

For a general singular pencilsF—G∈ℝm×n[s] the right, left characteristic sequences 𝒞r(F,G), 𝒞l(F,G) are defined and they are shown to be of the piecewise arithmetic progression type. The sets of column, row minimal indices ℐc(F,G), ℐr(F,G) are defined by a singular points analysis of 𝒞r(F,G), ℐl(F,G) respectively. Thus, ℐc(F,G), ℐr(F,G) emerge as numerical invariants of the ordered pair (F,G) and they may be computed by rank tests on Toeplitz matrices defined on (F,G).

ISSN:0020-7179    

DOI:10.1080/00207178808906066

出版商:Taylor & Francis Group    

年代:1988

数据来源: Taylor

摘要:

In this paper we demonstrate a connection between the pole assignment technique and the linear optimal regulator technique, namely there exist common Lyapunov functions between two closed-loop systems synthesized by each method. From this property, we derive a useful theorem for pole assignment.

ISSN:0020-7179    

DOI:10.1080/00207178808906067

出版商:Taylor & Francis Group    

年代:1988

数据来源: Taylor

摘要:

A design method to construct an LQ regulator for a system with time-delay is proposed. It is based on a new technique for composing the solution of an infinite-dimensional Riccati equation using the spectral decomposition of the hamiltonian. This design method has the feature that the closed-loop poles can be calculated apriorifrom the open-loop poles and one scalar design parameter, so that we can choose the parameter to obtain an optimal control law which results in the desired degree of exponential stability of the closed-loop system.

ISSN:0020-7179    

DOI:10.1080/00207178808906068

出版商:Taylor & Francis Group    

年代:1988

数据来源: Taylor

摘要:

A unified approach is proposed here to solve three problems simultaneously, namely H ∞ -optimization, Hankel approximation and minimal balanced realization for scalar systems. It is shown that all these problems can be reduced to the same standard form wherein merely the singular-value decomposition (SVD) of an easily constructed square matrix is required to establish the final solutions. The formulation is solely in terms of the coefficients of a transfer function and the resulting algorithm can be performed in a very computationally efficient way.

ISSN:0020-7179    

DOI:10.1080/00207178808906069

出版商:Taylor & Francis Group    

年代:1988

数据来源: Taylor

摘要:

An explicit computational form is given for the estimation of the smallest upper bound on the sensitivity function over an operating band for a given stability margin in a single right half-plane (RHP) zero scalar system. A stable minimum-phase weighting function is constructed to meet such an infimum and the given stability margin with the H∞criterion. A numerical example is presented to demonstrate the optimal sensitivity estimation, weighting function construction and controller design.

ISSN:0020-7179    

DOI:10.1080/00207178808906070

出版商:Taylor & Francis Group    

年代:1988

数据来源: Taylor

摘要:

In this note we generalized the results of Foias and Tannenbaum (1986 a) and Foiaset al. (1987) on optimal sensitivity minimization to a much broader class of distributed plants. We relate our results to certain kinds of Hankel norm approximation problems that frequently appear in control engineering applications.

ISSN:0020-7179    

DOI:10.1080/00207178808906071

出版商:Taylor & Francis Group    

年代:1988

数据来源: Taylor

摘要:

The problem of designing precompensators necessary for decoupling invertible systems having weak inherent coupling is considered. A simple algorithm, which is directly applicable to systems characterized by square proper transfer function matricesT(s), is proposed for such design. In the proposed algorithm the concept ‘row reducedness at infinity’ ofT(s)is utilized as an alternative, necessary and sufficient condition for the decouplability ofT(s). In addition to the simple design procedure of the compensator, the algorithm yields the infinite-zero structure of bothT(s)and the resulting compensated system.

ISSN:0020-7179    

DOI:10.1080/00207178808906072

出版商:Taylor & Francis Group    

年代:1988

数据来源: Taylor

摘要:

Mita (1977) presented a simple method for the invariant zeros of a linear multivariable system by constructing its maximal unobservable subspace (MUS). However, the results of this method have been restricted to a class of controllable and right invertible systems satisfying the diagonal decoupling condition and described by the triple (A,B,C). Here, a general method based on the theory of decoupling extends Mita's results for all classes of systems described by the 4-tuples (A, B, C, E) with no special requirements. The MUS of all classes of such systems is simply constructed and its order is given in terms of the system order n and the infinite zeros or the minimal observability indices of the system. In addition, the invariant zeros and zero directions are found from the eigenvalues and eigenvectors of a reduced-order matrix having the dimension of the MUS obtained directly from the state-space parameters. Algorithms are presented for the invariant zeros and zero directions of invertible decouplable (or non-decouplable) systems S(A, B, C, E) and all classes of invertible and non-invertible systems S(A, B, C, E). Numerical examples demonstrate the generality and feasibility of the proposed method.

ISSN:0020-7179    

DOI:10.1080/00207178808906073

出版商:Taylor & Francis Group    

年代:1988

数据来源: Taylor

摘要:

In practical control problems it is often desirable to make setpoint changes without overshoot. In this paper necessary and sufficient conditions are presented which guarantee that linear state-space models will not exhibit overshoot. Several examples illustrate the utility and simplicity of the new results

ISSN:0020-7179    

DOI:10.1080/00207178808906074

出版商:Taylor & Francis Group    

年代:1988

数据来源: Taylor