基于自整定和模糊自校正技术的改进Smith预估控制系统  

Refinements of Smith Predictor Based on Auto-tuning and Fuzzy Self-tuning

作　　者：刘红波[1] 李少远[2] 贾磊[1] 

机构地区：[1]山东大学控制科学与工程学院,山东济南250061 [2]上海交通大学自动化研究所,上海200030

出　　处：《系统仿真学报》2004年第9期2086-2090,共5页Journal of System Simulation

基　　金：国家 863 计划课题资助项目(2002AA412130);山东省科技发展计划资助项目(012050114)

摘　　要：针对常规Smith预估控制系统难于整定和对过程参数敏感的缺点,将模糊自校正与继电反馈自整定技术结合起来,提出了一种改进的Smith预估控制系统。大量仿真测试表明这种新的系统与常规系统相比可显著改善控制性能和适应能力。这种新的系统不仅对纯迟延占优势的大迟延对象有效,而且还适用于过程控制中大量存在的具有类似于纯迟延占优势对象特性的高阶对象和非最小相位过程。A new Smith predictor is proposed for industrial processes with dominant time delay, by integrating fuzzy self-tuning with relay feedback auto-tuning technique, aimed at eliminating the disadvantages that the conventional Smith predictor is difficult to tune manually and sensitive to process parameter variations. The new system has been tested on a lot of simulated plant models and demonstrated to have marked improvements in the control performance by comparison with the conventional system. This novel scheme has been found to be effective not only for dead-time-dominant processes but also for high-order or non-minimum phase processes that exhibit apparent dead-time-dominant like characteristics in their dynamics.

关 键 词：SMITH预估器 模糊自校正 自整定 热工过程控制 

分 类 号：TP273.2[自动化与计算机技术—检测技术与自动化装置] TP273.4[自动化与计算机技术—控制科学与工程]