摘要:

The equations of motion for a class of Tethered Satellite Systems (TSS), undergoing planar motion in a Keplerian orbit, are derived. The system consists of a rigid platform from which a point mass subsatellite can be deployed or retrieved by a flexible tether. The model incorporates offset of the tether attachment point from the platform centre of mass and its time dependent variation. The governing equations of motion are obtained using the Lagrangian procedure. The effect of system parameters on the dynamics is analysed numerically. A hybrid strategy using thruster and offset control is used to regulate the tether swing. The platform dynamics are controlled by momentum gyros. It is shown that the control of only rigid degrees of freedom is not sufficient as the flexible dynamics of the tether becomes unstable, particularly during retrieval. Passive dampers are proposed to control the flexible dynamics. Effectiveness of the control strategy is assessed through its application to several cases of practical importance.

ISSN:0020-7179    

DOI:10.1080/00207179608921623

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor

摘要:

New two-dimensional measures for reachability of linear multivariable systems are proposed and estimates for perturbations preserving reachability are derived.

ISSN:0020-7179    

DOI:10.1080/00207179608921624

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor

摘要:

A multilayer discrete-time neural net (NN) controller is presented for the direct model reference adaptive control of a class of multi-input multi-output (MIM0) nonlinear dynamical systems. The nonlinear dynamical system is assumed to be controllable and its state vector is available for measurement. The NN controller exhibits learning-while-functioning features instead of learning-then-control so that control is immediate with no explicit learning phase needed. The tracking error between the output of a nonlinear plant and an ideal linear model converges within a very short time. Persistence of excitation (PE) is not needed, linearity in the parameters is not required, and certainty equivalence is not used. This overcomes several limitations of standard adaptive control. The novel weight tuning paradigm for the NN controller is based on the well-known delta rule but includes a modification to the learning rate parameter plus a correction term. It guarantees tracking as well as bounded NN weights in non-ideal situations so that a persistency of excitation conditions on the internal signals is not needed. It is found that the constant learning rate parameter at a given layer should decrease with the number of hidden-layer neurons present in that particular layer; this is a major drawback often documented in the NN control literature. This drawback is easily overcome by employing a projection algorithm at each layer. Theoretical analysis is justified through extensive simulation studies.

ISSN:0020-7179    

DOI:10.1080/00207179608921625

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor

摘要:

The finite time horizonH∞-control problems for nonlinear time-varying systems are considered where the initial state is unknown. They are solved by using both static state feedback and dynamic output feedback controllers. A class of nonlinearH∞-controllers is parameterized as nonlinear fractional transformation on contractive nonlinear parameters. The solvability for nonlinear finite time horizonH∞-control problem requires suitable solutions to some Hamilton-Jacobi inequalities (equations).

ISSN:0020-7179    

DOI:10.1080/00207179608921626

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor

摘要:

We consider a mixed H∞/H2control problem, in which a controller is required to achieve the best possible rejection of a stochastic disturbance with given statistics, subject to an H∞-performance constraint. We assume that the H∞-constraint rules out the use of the H2-optimal (LQG) control law. As a result, the H2-performance of the controller must be degraded. Moreover, no control law can be the best with respect to all the possible measurement signals. It is thus reasonable to look for a controller that will achieve the best H2-performance, subject to the worst possible measurement signal. An optimal controller, in the sense of this paper, minimizes the worst-case ratio between the resulting H2-performance degradation and the performance degradation that would result if one had used a standard H∞-central controller. In this setting, the mixed H∞/H2problem reduces to a standard two-person, zero-sum, linear quadratic dynamic game, with a state-feedback information pattern. Both the continuous and the discrete-time cases are considered.

ISSN:0020-7179    

DOI:10.1080/00207179608921627

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor

摘要:

The problem of discrete observation of the wave equation on bounded domains in euclidean space is considered. Wallace and Wolf (1991 b) studied this problem for the heat equation using an infinite but discrete set of spatial samples for one time. They showed that the heat equation satisfied a certain type of discrete observability, which can be thought of as approximate discrete observability. In contrast to the heat equation, the wave equation has an additional initial condition that requires the use of an extended sampling scheme, and changes the structure of the problem. The conditions under which approximate discrete observability holds for a given set of samples are proven, as is the existence of samples that satisfy those conditions. Acuity of observation is established, providing general analytic bounds on the error in using only finitely many samples of the solution to the wave equation.

ISSN:0020-7179    

DOI:10.1080/00207179608921628

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor

摘要:

Feedback controllers, based on pulse-width-modulation (PWM), are derived for the average input current stabilization and tracking problems in derived DC-to-DC power supplies of the buck and boost types. The stabilization problems are solved on the basis of steady-state considerations about the current ‘ripple’ and exactly discretized nonlinear models describing the sampled PWM regulated input current trajectories. In the boost-derived converter, the stabilization problem leads to an implicit static nonlinear feedback controller, or duty ratio synthesizer, which requires online solutions of a transcendental equation at each sampling instant. The signal tracking problems are solved on the basis of discrete-time, non-kalmanian state representation models describing the average PWM regulated input current. These generalized state models naturally allow for explicit dynamical, rather than static, feedback regulators. Computer simulations, including unmodelled load variations and external stochastic perturbation inputs, are presented which test the robustness of the proposed PWM controller performances,

ISSN:0020-7179    

DOI:10.1080/00207179608921629

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor

摘要:

A new P-type iterative learning controller using the concept of the forgetting factor and current error modification is proposed in this paper for robust output tracking of uncertain nonlinear time-varying systems. The robustness, convergence and performance of this P-type learning control scheme are completely studied via a systematic approach. It is shown that uniform boundedness between the system output and the desired output is achieved in each iterate. Under some conditions on the uncertainties, the residual set of the tracking error on the final iterate can be shown to be a classKfunction of the forgetting factor. When all the uncertainties and the forgetting factor tend to zero, the system output will converge uniformly to the desired one.

ISSN:0020-7179    

DOI:10.1080/00207179608921630

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor

摘要:

A rigorous method of converting a transfer function description with any shaped uncertainty region in the complex frequency domain, to input specific, guaranteed time domain bounds, is derived. Periodic input signals are considered first, followed by an extension to aperiodic input such as a step. The results can be used with confidence in controller design techniques that make use of the complex frequency plane and require translation to the time domain.

ISSN:0020-7179    

DOI:10.1080/00207179608921631

出版商:Taylor & Francis Group    

年代:1996

数据来源: Taylor