Showing papers by "Kang-Zhi Liu published in 2003"
01 Sep 2003
TL;DR: A simple derivation of the Riccati equation based solutions to the standard H∞ control problem, namely the well-known Glover-Doyle solution and DGKF solution is given based on LMI solution.
Abstract: In this paper, a simple derivation of the Riccati equation based solutions to the standard H ∞ control problem, namely the well-known Glover-Doyle solution and DGKF solution is given based on LMI solution. It is hoped that this will be helpful for ordinary control engineers in deepening the understanding of Riccati equation solutions.
9 citations
01 Jan 2003
TL;DR: In this article, an adaptive backstepping control system is proposed for a 2-mass-spring system having large load uncertainty, which is able to adapt to a wide range of loads without on site tuning and significant performance deterioration.
Abstract: In this paper, an adaptive backstepping control system is proposed for a 2-mass-spring system having large load uncertainty. The purpose is to construct a control system that is able to adapt to a wide range of loads without on site tuning and significant performance deterioration. Theoretically the global asymptotic tracking of step reference is guaranteed for arbitrarily large load uncertainty. It is verified in simulations that the response speed slows down as the load inertia moment increases or the shaft becomes less stiff. However, no serious vibration is observed.
7 citations
25 Jun 2003
TL;DR: In this paper, a nonlinear output feedback control method is proposed for a magnetic bearing system which has a strong nonlinearity in the magnetic bearing systems and a magnetic actuator.
Abstract: In this paper, a nonlinear output feedback control method is proposed for a magnetic bearing system which has a strong nonlinearity in the magnetic bearing system which has a strong nonlinearity in the magnetic actuator. The basic idea is to make full use of the feature that the dynamics part is linear and hence can be stabilized by output of the actuator is regarded as a virtual input. Then it is shown that the backstepping and completing square techniques enable the construction of the construction of the real input using measured output only.
6 citations
01 Sep 2003
TL;DR: It is shown that the backstepping and completing square techniques enable the construction of the real input using measured output only and can be stabilized by output feedback when the output of the actuator is regarded as a virtual input.
Abstract: In this paper, a nonlinear output feedback control method is proposed for a magnetic bearing system which has a strong nonlinearity in the magnetic actuator. The basic idea is to make full use of the feature that the dynamics part is linear and hence can be stabilized by output feedback when the output of the actuator is regarded as a virtual input. Then it is shown that the backstepping and completing square techniques enable the construction of the real input using measured output only.
4 citations
01 Sep 2003
TL;DR: A simple control approach is proposed for the multi-chain systems based on a switching algorithm of two sets of smooth steering laws, which is very flexible in the sense that the requirement on the generator is very weak.
Abstract: In this paper, a simple control approach is proposed for the multi-chain systems. This method is based on a switching algorithm of two sets of smooth steering laws. Global exponential convergence of all states is guaranteed. More importantly, the convergence rate is given explicitly. Two concrete methods are proposed for the design of steering laws. One is based on the backstepping design, another is based on the use of Riccati equation. Extension of this approach to some driftless systems is also discussed. This approach is very flexible in the sense that the requirement on the generator is very weak. Therefore, there is a great freedom in the determination of the generator which may be used to satisfy other control specifications.
2 citations
Proceedings Article•
01 Jan 2003TL;DR: In this article, a new method with control in the radial direction was proposed to transfer the cargo to the target point as quickly as possible while suppressing the swaying and to quickly minimize the residual sway at the target points.
Abstract: The basic motions of a rotational crane as a construction machine are classified into cargo windup, boom derricking and boom slewing The purpose of controlling the crane is to transfer the cargo to the target point as quickly as possible while suppressing the swaying and to quickly minimize the residual sway at the target point In this paper, we propose a new method with control in the radial direction
1 citations
Proceedings Article•
01 Jan 2003TL;DR: In this paper, a nonlinear output feedback control method is proposed for a magnetic bearing system which has a strong nonlinearity in the magnetic actuator, which can be stabilized by output feedback when the output of the actuator is regarded as a virtual input.
Abstract: In this paper, a nonlinear output feedback control method is proposed for a magnetic bearing system which has a strong nonlinearity in the magnetic actuator. The basic idea is to make full use of the feature that the dynamics part is linear and hence can be stabilized by output feedback when the output of the actuator is regarded as a virtual input. Then it is shown that the backstepping and completing square techniques enable the construction of the real input using measured output only.
1 citations
01 Jan 2003
TL;DR: In this paper, the authors proposed a new method to design discontinuous stabilizing controllers for chained systems, which yield exponential convergence of the states to the origin, where the closed loop system is formulated by a linear differential equation with a regular singular point at z=0.
Abstract: In this paper, we propose a new method to design discontinuous stabilizing controllers for chained systems. The proposed controllers yield exponential convergence of the states to the origin. In our method, the closed loop system of a chained system is formulated by a linear differential equation with a regular singular point at z=0. It is shown that the controller design method proposed by Astolfi, in which /spl sigma/ process is used for a coordinate transformation, is included in our method as a special case where the linear differential equation with a regular singular point at z=0 is limited to its subclass, namely an Euler differential equation. This method can be easily extended to apply to multiple chained systems.
1 citations
09 Dec 2003
TL;DR: In this article, a simple derivation of the Riccati equation based solutions to the standard H/sub /spl infin// control problem, namely the well-known Glover-Doyle solution and DGKF solution is given based on LMI solution.
Abstract: In this paper, a simple derivation of the Riccati equation based solutions to the standard H/sub /spl infin// control problem, namely the well-known Glover-Doyle solution and DGKF solution is given based on LMI solution. It is hoped that this will be helpful for ordinary control engineers in deepening the understanding of Riccati equation solutions.