Robust stability analysis of uncertain discrete-time systems with state delay

The sufficient conditions of stability for uncertain discrete-time systems with state delay have been proposed by some researchers in the past few years, yet these results may be conservative in application. The stability analysis of these systems is discussed, and the necessary and sufficient condition of stability is derived by method other than constructing Lyapunov function and solving Riccati inequality. The root locations of system characteristic polynomial, which is obtained by augmentation approach and Laplace expansion, determine the stability of uncertain discrete-time systems with state delay, the system is stable if and only if all roots lie within the unit circle. In order to analyze robust stability of system characteristic polynomial effectively, Kharitonov theorem and edge theorem are applied. Example shows the practicability of these methods.     

模型预测控制反馈校正系数的在线整定

模型预测控制在处理不可测干扰时有明显的局限性，特别是对于类似斜坡一样的干扰，控制作用十分缓慢。对此，提出了一种对其反馈校正系数的在线整定方法，在各种不同动态的不可测干扰存在的情况下，能显著提高预测控制的控制性能。     

Robust Stability Analysis of a Kind of Combined Integrating Control System

This paper offeres an exact study on the robust stability of a kind of combined integrating control system, and the robust stability belongs to the analysis of a kind of quasi-polynomial with two independent time delays. The parameters of stable space under time delay uncertainty are fixed after Rekasius transformation, and then a new cluster treatment of characteristic roots (CTCR) procedure is adopted to determine the stable space. By this strategy we find that the unstable space is not continuous and bot...     

On Stabilizing Sets of PI Controllers for Multiple Time Delayed Process

This paper considers the problem of stabilizing multiple time delayed processes using proportional integral (PD controller. The presented approach is based on finding all possible values of control parameters which will result in pure imaginary roots of closed loop characteristic equation under all process parameters fixed. The ergedic search of three PI control parameters are converted from the range of infinity to finite range by introducing trigonometric tangent function. After all possible stability boundaries are obtained, the Nyquist stability method is used to determine the actual stability region of the controller parameters. This method also permits design for simultaneous minimum gain and phase margin requirement. An illustrative example case is also presented.