This paper is concerned with identification of continuous-time Wiener-type nonlinear model. A subspace method is applied to identify the linear dynamic element represented by δ-operator model transformed through Laguerre filters. The nonlinear part is estimated via least-squares approximation. Numerical simulation results are provided to illustrate the proposed identification method.
The problem of controlling the RTAC (Rotational/Translational Actuator) system provides a benchmark for evaluting various nonlinear control design techniques. The RTAC system, which represents a translational oscillator with an eccentric rotational proof-mass actuator, was originally studied as a simplified model of a dual-spin spacecraft to investigate the resonance capture phenomenon. This paper presents design methods of a controller that supresses oscillations of the RTAC system. The controller gives a dual-mode control that consists of an energy-based control that can effectively suppress the oscillations of the bed, and a stabilizing control that makes the whole system globally and asymptotically stable. The energy-based control is taken over by the stabilizing control when the oscillations of the bed become small to some extent. Two types of energy-based control law are proposed : one rotates the disk in normal and reverse rotation in a limited angular amplitude and the other rotates the disk in one direction. It is shown that the proposed controller achieves a performance comparable with that of the time optimal controller. Simulation experiments also show the effectiveness of the proposed methods.
This paper addresses an algorithm of J spectral factorization for discrete time polynomial matrices. Particularly, we focus on the computation of J spectral factorization for discrete time polynomial matrices. Our approach is based on two-variable polynomial matrices, which is introduced for the analysis and synthesis of discrete time dissipativeness by authors, and algebraic Riccati equations. We show that the solvablity of J spectral factorization of a given polynomial matrix is equivalent to that of an algebraic Riccati equation consisting of the coefficient matrices of the given polynomial matrix. And then we provide an algorithm of discrete time J spectral factorization. Moreover, we give an illustrative example in order to show the validity of our results.
This paper considers stability of piecewise linear second-order systems. We analyze the trajectory of the systems by using eigenvalues and eigenvectors of coefficient matrices of subsystems. For the case where the coefficient matrices have complex eigenvalues, we present the necessary and sufficient condition for the piecewise linear systems to be stable.