Improving the Stability Behavior of Limiting Zeros for Multivariable Systems Based on Multirate Sampling

作     者：Zeng, Cheng Liang, Shan Xiang, Shuwen 

作者机构：Guizhou Univ Coll Comp Sci & Technol Guiyang 550025 Guizhou Peoples R China Guizhou Inst Technol Sch Sci Guiyang 550003 Guizhou Peoples R China Chongqing Univ Coll Automat Chongqing 400044 Peoples R China 

出 版 物：《INTERNATIONAL JOURNAL OF CONTROL AUTOMATION AND SYSTEMS》 (国际控制与自动化系统杂志)

年 卷 期：2018年第16卷第6期

页      面：2621-2633页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 08[工学] 0811[工学-控制科学与工程] 

基　　金：National Natural Science Foundation of China Joint Funds of the Natural Science Foundation Project of Guizhou [LH7362] 

主　　题：Limiting zeros multirate input and hold multivariable systems sampled-data models stability 

摘      要：It is well-known that existence of unstable sampling zeros is recognized as a major barrier in many control problems, and stability of sampling zeros, in general, depends on the type of hold circuit used to generate the continuous-time system input. This paper is concerned with stability of limiting zeros, as the sampling period tends to zero, of multivariable sampled-data models composed of a generalized sample hold function (GSHF), a continuous-time plant with the relative degrees being two and three, and a sampler in cascade. In particular, the main focus of the paper is how to preserve the stability of limiting zeros when at least one of the relative degrees of a multivariable system is more than two. In this case, the asymptotic properties of the limiting zeros on the basis of normal form representation of continuous-time systems are analyzed and approximate expressions for their stability are discussed as power series expansions with respect to a sufficiently small sampling period. More importantly, unstable sampling zeros of the sampled-data models mentioned above can be avoided successfully through the contribution of this paper, whereas a zero-order hold (ZOH) or a fractional-order hold (FROH) fails to do so. It is a further extension of previous results for single-input single-output cases to multivariable systems.