Authors:
Carlos E. de Souza,
Alexandre Trofino,
José de Oliveira,
Volume: 1, Page 3194 Paper number 9505
Abstract:
This paper addresses the problems of robust H-infinity performance
analysis and control synthesis for linear systems subject to uncertain
real time-varying parameters. The uncertain parameters enter affinely
in the matrices of the system state-space model and the admissible
values of the parameters and their rates of variation are supposed
to belong to a given polytope. New LMI based methods of robust H-infinity
analysis and control design via state feedback are proposed using parameter-dependent
Lyapunov functions.
Authors:
Toru Asai,
Volume: 1, Page 3200 Paper number 1825
Abstract:
This paper deals with the constantly scaled H-infinity control problem.
In general, the solvability condition for this problem is non-convex.
However, it is known that the state feedback (SF) and the full information
(FI) problems can be reduced to convex problems. In this paper, we
consider a class of problems including the SF and the FI problems as
the special cases. The class is characterized by an additional assumption
called a pseudo full information (PFI) condition. Assuming the PFI
condition, we first derive a solvability condition. This condition
involves fewer number of variables than the standard solutions for
the constantly scaled H-infinity control problem. Based on this condition,
we further give a convex sufficient condition for the solvability.
The convex solvability condition for the SF and the FI problems can
be regarded as the extreme cases of the derived sufficient condition.
Moreover, we show that there exists a simple formula of a possible
controller, if the PFI condition is assumed.
Authors:
Olof Lindgärde,
Bengt Lennartson,
Volume: 1, Page 3206 Paper number 1939
Abstract:
Feedback control using both continuous-time and discrete-time measurements
is considered in this paper. A design method including both H-infinity-synthesis
and frequency analysis is suggested. It includes a strategy for selecting
weighting filters in the control design. The presented frequency analysis
method is then a powerful tool in the design process.
Authors:
Muralidhar Ravuri,
Chung-Yao Kao,
Alexander Megretski,
Volume: 1, Page 3212 Paper number 1915
Abstract:
In this paper, we consider the H_(infinity) optimization problem for
systems with a hierarchical structure. We propose an iterative algorithm
to treat the problem. In each iteration of our algorithm, we need to
solve a problem of H_(infinity) optimization feature, but with a much
smaller dimension. We also show that the solution of the optimization
problem in each iteration can be obtained by solving a set of Linear
Matrix Inequalities (LMIs).
Authors:
Carl-Magnus Fransson,
Bengt Lennartson,
Claes Breitholtz,
Anders Bondeson,
Yueqiang Liu,
Volume: 1, Page 3218 Paper number 2071
Abstract:
A constrained, two step optimization procedure based on H-inf loop
shaping is suggested for highly unstable plants with large parametric
uncertainties. With the proposed controller synthesis, optimality is
inherited from the H-inf design, and robustness to uncertainties is
guaranteed by use of Horowitz bounds. This implies that specifications
can be made for the worst case of the sensitivity, the complementary
sensitivity, and the controller sensitivity functions. The method is
applied to a Tokamak fusion reactor for two different types of sensors,
and in this example, optimization of one plant design parameter and
one controller tuning parameter will illustrate the procedure. The
results are validated by theory for performance limitations of unstable
plants, as well as with a manual search over the two dimensional design
and tuning parameter space.
Authors:
Sang Hyun Cho,
Ki Tae Kim,
Hong Bae Park,
Volume: 1, Page 3224 Paper number 1855
Abstract:
This paper describes the synthesis of robust and non-fragile H-infinity
state feedback controllers for linear time varying systems with affine
parameter uncertainties and static state feedback controller with polytopic
uncertainty. The sufficient condition of controller existence, the
design method of robust and non-fragile H-infinity static state feedback
controller, and the region of controllers which satisfies non-fragility
are presented. Also using some change of variables and Schur complements,
the obtained condition can be rewritten as parameterized Linear Matrix
Inequalities (PLMIs), that is, LMIs whose coefficients are functions
of a parameter confined to a compact set. We show that the resulting
controller guarantees the asymptotic stability and disturbance attenuation
of the closed loop system in spite of controller gain variations within
a resulted polytopic region.
Authors:
Youngjin Choi,
Wan Kyun Chung,
Youngil Youm,
Volume: 1, Page 3230 Paper number 2164
Abstract:
Suboptimal H_(infinity) controller is said to internally stabilize
the closed loop transfer matrix. However, the stability of controller
is not assured in conventional H_(infinity) control theory. Unstable
controller can damage the whole system when the sensor fails or actuator
saturates. The sufficient condition for the existence of stable H_(infinity)
controller is suggested. This paper presents the stable n-dimensional
H_(infinity) controller and its parameterization based on the suggested
sufficient condition for the linear time invariant systems using doubly
coprime factorization. The stability of H_(infinity) controller and
closed-loop stability are guaranteed if the positive semi-definite
solutions for the suggested three Riccati equations exist.
|