摘要:

This paper is concerned with the problem of identifying parameters in stochastic systems described by single-input single-output linear discrete-time equations. A stable estimation error system is developed by using the extended Kalman filter technique and the concept of strictly positive real transfer function. The identifier corresponding to the estimation error system is constructed, and the convergence of parameter estimates to the exact values is proved under some bounded conditions. It is shown that the recursive maximum likelihood identifier and the recursive extended least squares identifier are obtained by neglecting the correction terms in the proposed identifier. Numerical examples for a fourth-order system are presented to illustrate the effectiveness of the proposed method.

ISSN:0020-7721    

DOI:10.1080/00207728008967086

出版商:Taylor & Francis Group    

年代:1980

数据来源: Taylor

摘要:

In this paper, the asymptotic properties of tracking systems incorporating linear multivariable plants which are amenable to fast-sampling error-actuated control (Bradshaw and Porter 1980) are characterized in terms of the eigenstructure of the closed-loop plant matrices. It is shown that the closed-loop eigenstructure is such that the tracking behaviour of systems incorporating fast-sampling error-actuated controllers designed in accordance with the synthesis technique of Bradshaw and Porter (1980) is increasingly dominated by the modes associated with characteristic roots of the form λ=l+σ as the sampling frequency is increased and that increasingly ‘tight’ control is therefore achieved. These theoretical results are illustrated in the case of a closed-loop system incorporating a third-order plant and a fast-sampling error-actuated controller by the presentation of the computed closed-loop eigenstructure and of the results of computer simulation studies.

ISSN:0020-7721    

DOI:10.1080/00207728008967087

出版商:Taylor & Francis Group    

年代:1980

数据来源: Taylor

摘要:

The purpose of this paper is to provide a unified framework to investigate the meanings of decision principles. A decision principle is a guiding rule to define a preference order over the set of alternatives for a given decision problem. Although many decision principles have been proposed as rational ones, it is uniformly agreed that no decision principle is uniformly best and it is an important subject for decision theory to explore the meaning of a decision principle to specify when it is reasonable. This problem has been considered in various ways. This paper, first, characterizes & rational decision principle as a rule which satisfies the two conditions, Pareto consistency and the similarity condition. Pareto consistency has been accepted as an essential condition for a rational decision principle. The similarity condition, which is an original idea of this paper, requires that if two decision problems are structurally similar, then their preference orders induced by a rational decision principle should be similar. Decision principles are, then, represented in the most natural way as functors from the category of decision problems into the category of ordered sets. We demonstrate that almost all the decision principles proposed as rational satisfy this formulation and that furthermore, the max-min principle is completely characterized by these two conditions. This final result shows that the similarity condition is an essential property of rational principles and the present category-theoretic formulation can be a unified framework for the study of decision principles.

ISSN:0020-7721    

DOI:10.1080/00207728008967088

出版商:Taylor & Francis Group    

年代:1980

数据来源: Taylor

摘要:

The main objective of this paper is to demonstrate the application of filtering techniques in econometric models for the purpose of obtaining bayesian estimates of the reduced form coefficients and their covariance matrix.

ISSN:0020-7721    

DOI:10.1080/00207728008967089

出版商:Taylor & Francis Group    

年代:1980

数据来源: Taylor

摘要:

An iterative procedure is presented for the optimization of the estimated parameters of non-linear systems which can be described by a combined Wiener-Hammerstein model, usually referred to as the general model. The procedure enables the efficient and fairly accurate estimation of the parameters from a short input/output record.

ISSN:0020-7721    

DOI:10.1080/00207728008967090

出版商:Taylor & Francis Group    

年代:1980

数据来源: Taylor

摘要:

In this paper we develop a new approach to the design of near-optimal decentralized controllers for large-scale linear interconnected dynamical systems. All calculations in the design stage are done on subsystems. This is achieved by using a simple reduced order model for the interactions. The resulting controller is decentralized, independent of initial conditions and capable of accommodating constant unknown disturbances. The approach is illustrated on a 22nd order river pollution control example. In addition, it is shown that by a simple modification it is possible to ensure that the controller is asymptotically stable with a pre-specified degree of stability and we show that the modified controller is also connectively asymptotically stable under structural perturbations.

ISSN:0020-7721    

DOI:10.1080/00207728008967091

出版商:Taylor & Francis Group    

年代:1980

数据来源: Taylor

摘要:

The simultaneous state and parameter estimation problem for a linear discrete-time system with unknown noise statistics is treated as a large-scale optimization problem. Thea posterioriprobability density function is maximized directly with respect to the states and parameters subject to the constraint of the system dynamics. The resulting optimization problem is too large for any of the standard non-linear programming techniques and hence an hierarchical optimization approach is proposed. It turns out that the states can be computed at thefirst levelfor given noise and system parameters. These, in turn, are to be modified at thesecond level.The states are to be computed from a large system of linear equations and two solution methods are considered for solving these equations, limiting the horizon to a suitable length. The resulting algorithm is a filter-smoother, suitable for off-line as well as on-line state estimation for given noise and system parameters. The second level problem is split up into two, one for modifying the noise statistics and the other for modifying the system parameters. An adaptive relaxation technique is proposed for modifying the noise statistics and a modified Gauss-Newton technique is used to adjust the system parameters.

ISSN:0020-7721    

DOI:10.1080/00207728008967092

出版商:Taylor & Francis Group    

年代:1980

数据来源: Taylor

摘要:

A very general and numerically quite robust algorithm has been proposed by Sastry and Gauvrit (1980) for system identification. The present paper takes it up and examines its performance on a real test example. The example considered is the lateral dynamics of an aircraft. This is used as a vehicle for demonstrating the performance of various aspects of the algorithm in several possible modes.

ISSN:0020-7721    

DOI:10.1080/00207728008967093

出版商:Taylor & Francis Group    

年代:1980

数据来源: Taylor