Authors:
Keith Glover,
Glenn Vinnicombe,
George Papageorgiou,
Volume: 1, Page 4084 Paper number 9007
Abstract:
The gap metric and nu-gap metric have many appealing properties for
assessing the uncertainty in a plant in a feedback configuration. This
short paper addresses the question of relating these generalised stability
margins to more traditional single-loop and multi-loop robustness measures.
In particular we show that when a system is controlled by a controller
which provides a stability margin of epsilon in the gap metric for
the weighted plant then if the weighting matrices are diagonal the
plant will have robust stability to easily interpreted simultaneous
gain and phase variations at all the inputs and outputs.
Authors:
Keiji Watanabe,
Zhi-Wei Luo,
Guido Izuta,
Volume: 1, Page 4086 Paper number 57
Abstract:
This paper develops robust pole placement that satisfies mixed sensitivity
specification and places pole of the closed-loop system in pre-assigned
half plane simultaneously. The feature of the proposed method is that
the pre-assigned half plane is broad in comparison with that of affine
transformation of standard H-infinity control and multipurpose H-infinity
synthesis in LMI. Theoretical background and illustrative numerical
example are presented
Authors:
Myung-Gon Yoon,
Hidenori Kimura,
Volume: 1, Page 4092 Paper number 102
Abstract:
We study on the robustness of non-strongly stabilizable SISO plants
under small perturbations in graph topology. From a new notion of pole-zero
shifting, we show that there are two types of non-strongly stabilizable
plants: one that can be made strongly stabilizable by arbitrary small
perturbations and the other that is essentially non-strongly stabilizable.
We provide a simple criterion for this classification and examine its
relevance to the simultaneous stabilization problem.
Authors:
Aurelio Piazzi,
Antonio Visioli,
Volume: 1, Page 4098 Paper number 1109
Abstract:
In this paper we propose a method, based on dynamic inversion, for
the set-point regulation of uncertain nonminimum-phase scalar systems.
In particular, the worst-case settling time is minimized taking into
account an amplitude constraint on the control variable and limits
on the overshoot and undershoot of the output function. The application
of the devised methodology yields to the connected design of both the
controller and the reference command input. The latter is obtained
by solving a special stable inversion problem on the nominal dynamic
system that leads to a noncausal signal, causing the so-called ``preaction
control''. Eventually, an optimization problem arises and its solution
is gained by means of genetic algorithms. A simulation example shows
the effectiveness of the overall methodology, despite the inherent
difficulty of the addressed problem.
Authors:
Ülo Nurges,
Ennu Rüstern,
Volume: 1, Page 4104 Paper number 1016
Abstract:
A new version of pole placement controller design, called robust reflection
coefficients placement, is proposed for discrete-time systems. Instead
of a single stable point a stable simplex must be preselected in the
closed loop characteristic polynomial coefficient space. A constructive
procedure for generating simplexis inside the "nice stability region"
is given starting from the unit hypercube of reflection coefficients
of monic polynomials. This procedure is quite straightforward because,
first, an appropriate stable point has to be chosen and then the edges
of the stable simplex will be generated by a linear Schur invariant
transformation. The procedure of robust controller design by quadratic
programming makes use of a stability measure defined as the minimal
distance between a preselected stable simplex and vertices of the uncertain
interval plant.
Authors:
Newton G. Bretas,
Luís Fernando Costa Alberto,
Volume: 1, Page 4110 Paper number 1455
Abstract:
The object of this work is to obtain uniform estimates, with respect
to parameters, of the attractor and of the basin of attraction of a
dynamical system and to apply these results to analyze the roughness
of the synchronization of two subsystems. These estimates are obtained
through an uniform version of the Invariance Principle of La Salle
which is stated and proved in this work.
Authors:
Hakan Koç,
Dominique Knittel,
Michel de Mathelin,
Gabriel Abba,
Volume: 1, Page 4116 Paper number 2086
Abstract:
The plant is a web transport system with winder and unwinder. Due to
a wide-range variation of the radius and inertia of the rollers the
system dynamics change considerably. Two different control strategies
for web tension and linear transport velocity are presented. The first
is an H infinity robust control strategy with varying gains based on
a particularity of the plant. The second is an LPV control strategy
with smooth scheduling of controllers synthesized for different operating
points. The quadratic stability and the quadratic performance of the
closed loop system are analyzed. The LPV control strategy gives better
results on an experimental setup, for the rejection of the disturbances
introduced by velocity variations.
|