摘要:

In this paper a scheme has been proposed for the identification and self-optimization of a non-linear control system, in which controllers have been designed to yield the error output between the non-linear system and the equivalent linear model. The output of the controller is utilized (i) to modify the output of the model for identification and (ii) to modify the output of the non-linear system for self-optimization or adaptive control. In each case the corresponding output is so modified as to minimize the error between the model output and the system output.

ISSN:0020-7721    

DOI:10.1080/00207727708942046

出版商:Taylor & Francis Group    

年代:1977

数据来源: Taylor

摘要:

In this paper we discuss the property of reachability and controllability for a linear discrete time system,

ISSN:0020-7721    

DOI:10.1080/00207727708942047

出版商:Taylor & Francis Group    

年代:1977

数据来源: Taylor

摘要:

An adaptive scheme is proposed for obtaining the steady-state Kalman gain matrix for o discrete-time system withouta prioriknowledge of the noise covariance matrices. It is based on combining an algorithm proposed recently by Carew and Bélanger with an algorithm based on stochastic approximation. Results of simulation are given comparing the proposed method with earlier algorithms.

ISSN:0020-7721    

DOI:10.1080/00207727708942048

出版商:Taylor & Francis Group    

年代:1977

数据来源: Taylor

摘要:

The solution of the linear matrix equation XC – BX = D is investigated as an eigen-value problem via anassociated matrix. A set of principal vectors of the associated matrix is shown to constitute the required solution. The properties of this set and the uniqueness of the solution are investigated as well as considering the special case of the Lyapunov matrix equation.

ISSN:0020-7721    

DOI:10.1080/00207727708942049

出版商:Taylor & Francis Group    

年代:1977

数据来源: Taylor

摘要:

A first-order simulation algorithm for multivariable non-linear stochastic integral equations of the Ito type is derived, which utilizes the second-order properties of incremental Brownian motion. Various questions of stability and convergence of the algorithm are investigated. By way of application the algorithm is used in evaluating models and optimal control laws for biological watewater treatment plants of the activated sludge type.

ISSN:0020-7721    

DOI:10.1080/00207727708942050

出版商:Taylor & Francis Group    

年代:1977

数据来源: Taylor

摘要:

The present paper is concerned with the study of Asymptotic Stability With Probability One (ASVVPI) of a class of linear time-varying-constraint stochastic systems by means of the stochastic version of Lyapunov's Second Method.

ISSN:0020-7721    

DOI:10.1080/00207727708942051

出版商:Taylor & Francis Group    

年代:1977

数据来源: Taylor

摘要:

The new theory of optimal control with a finite cost for each switch of control is applied to a specific, though quite generally useful, example. The results are explicit and, in part, surprising.

ISSN:0020-7721    

DOI:10.1080/00207727708942052

出版商:Taylor & Francis Group    

年代:1977

数据来源: Taylor

摘要:

A growing cell is characterized by the two-dimensional state vector consisting of cell volume and nuclear (chromatin) volume (mass). The state dynamics is governed by a system of linear differential equations and controlled additively by a set of genes. We assume that for the fixed cell type under consideration, the cell volume and chromatin volume at which mitosis begins is known. This point in the cell volume-nuclear volume phase plane is referred to as the target. For certain types of malignant cells the target is reached in the minimum time possible, given all the constraints of the system. The time-optimal requirement forces the active genes to be ‘ fully on ’, and the resulting slight increase from the ‘ normal on ’ activity level is shown to cause a significant decrease in the generation time of the cell type.

ISSN:0020-7721    

DOI:10.1080/00207727708942053

出版商:Taylor & Francis Group    

年代:1977

数据来源: Taylor

摘要:

The motion of a growing cell is modelled by a one-dimensional diffusion processXthaving 0 moan and a diffusion coefficient which is a function of the cell volume. When a cell is born, it is not moving and thus the diffusion coefficient is 0. As the cell grows, the coefficient increases until a critical volumeVis reached; pastV, the coefficient diminishes until it is again 0 at the point where the cell has stopped in order to give birth. The diffusion coefficient as a function of volume can be estimated experimentally by observing the growth and motion of a collection of cells of a given type. Once it is known, the volume growth curves, defined by the solution of ordinary linear differential equations, act as controllers on the processXt, which is represented as a stochastic integral.

ISSN:0020-7721    

DOI:10.1080/00207727708942054

出版商:Taylor & Francis Group    

年代:1977

数据来源: Taylor

摘要:

A signed digraph model, which has been used to assess strategies intended to stabilize models of energy use and air pollution, is extended in a simple manner to include the effects of time lags on stability, and on the growth rate of instabilities. It is shown that in an analogous model using differential equations, the effect of time lag, involving integration of a variable over all earlier times, can be incorporated equally simply, if the weighting factor in the integral (memory function) is chosen in a specific form. The results of this procedure are compared in some simple models with the results of using discrete time lag, and of numerically solving the integro-differential equation with a square memory function.

ISSN:0020-7721    

DOI:10.1080/00207727708942055

出版商:Taylor & Francis Group    

年代:1977

数据来源: Taylor