Jin-Zhi Wang

Publications (2)1.55 Total impact

• Source

  Available from: amss.ac.cn

  Article: Global stability of systems with amplitude and rate saturation compensation

  Jin-Zhi Wang, C. W. Chan, Ji-Feng Zhang
  [show abstract] [hide abstract]
  ABSTRACT: Rate and amplitude saturation of the actuator is common in practical control systems. When the actuator is rate and amplitude saturated, the control performance can deteriorate rapidly, and in the worse case, the closed-loop system can become unstable. It is therefore important that both types of saturations are properly compensated. Following the approach for compensating amplitude constraints, a scheme for compensating systems with both rate and amplitude saturation is proposed in this paper. The conditions for the compensated system to be globally stable are derived, and from this result, a procedure for designing the rate and amplitude saturation compensators is devised. As it is difficult to design both the rate and the amplitude saturation compensators simultaneously, a two-step approach is adopted. In the proposed compensator design procedure, the amplitude saturation compensator is designed first, followed by the rate saturation compensator. As the compensators designed using the proposed procedure satisfy the conditions for global stability, the compensated system is therefore globally stable. It is also shown that these compensators can be designed using the LMI technique. The implementation of the design procedure is demonstrated by a simulation example. Copyright © 2004 John Wiley & Sons, Ltd.
  International Journal of Robust and Nonlinear Control 03/2005; 15(4):155 - 170. · 1.55 Impact Factor
• Conference Proceeding: On the minimal-order of dynamic output regulators

  Jin-Zhi Wang, Ji-Feng Zhang
  [show abstract] [hide abstract]
  ABSTRACT: The minimal-order problem of the dynamic output regulators of linear time-invariant systems is considered, and a lower bound for the minimal-order is given. Unlike in previous works, the lower bound presented is a simple function of the system dimension and ranks of the input and output matrices, computation involves neither a rank minimization problem nor a semi-definite programming problem. An example is given to illustrate the non-conservativeness of the lower bound
  Intelligent Control and Automation, 2000. Proceedings of the 3rd World Congress on; 02/2000