节点文献
多变量内模控制的仿真研究
The Simulation Research on Multivariable Internal Model Control Algorithm
【作者】 齐菲;
【导师】 靳其兵;
【作者基本信息】 北京化工大学 , 控制理论与控制工程, 2008, 硕士
【摘要】 在现实化工过程中,多变量系统广泛存在,由于存在各回路之间的加大较大耦合、多时滞等特性,而传统的单变量PID控制的方法难于保证控制精度或难以控制。本文选取典型多变量化工过程系统为研究对象,提出了基于全极点近似的时滞系统的内模PID控制,提出了两种基于v规范型解耦的多变量内模控制设计的新方法,进行了多变量内模滤波器寻优,简单研究了主控变量法和广义逆法的胖系统多变量内模控制。具体如下:1.提出了基于全极点近似的时滞系统的内模控制,将全极点近似法引入内模控制设计,并将其应用于IMC-PID的转化过程,推导出了IMC-PID控制器参数计算的一般通式,便于仿真编程的实现,及在多变量内模控制中的直接应用。仿真结果表明,全极点近似法的近似程度高,系统的控制效果好,且仍具有很强的鲁棒性。2.提出了两类基于v规范型解耦的多变量内模控制设计的新方法:基于v规范型的新型多变量解耦内模控制和基于v规范解耦环节的多变量内模控制。分别对其进行系统性地解耦分析,详细推导控制器设计,得到控制器设计的一般通式,得到不同类型的v规范型解耦的IMC-PID转化结构,分别形成了较为完整的控制分析和设计方法的体系。3.将NLJ优化算法引入两种新型的多变量内模控制方法,以一种改进的综合跟踪特性与解耦特性的ITAE指标最优为目标,对滤波器参数进行寻优,仿真结果证明,控制良好,抗干扰能力强,且易于在工程中DCS的应用。4.针对胖系统研究了两种内模控制方法:主控量法和广义逆法。仿真结果表明,控制效果良好。本文率先进行了多变量系统的IMC-PID设计研究,并给出了可行性的方法,是一个创新。
【Abstract】 In reality,many chemical process is a complex multivariable instability, and time-delay system,and there is greater coupling,while the traditional method of PID control can not be solved.In this paper,the all-pole similarity is proposed for the time-delay of IMC-PID system,and we study typical multi-variable systems for the chemical process study and propose two design methods which based on the v-norm decoupling of the new multi-variable IMC design and optimization.NLJ optimization algorithm has been used in this two IMC method.For fat systems,two kinds of IMC design methods have been proposed:main manipulated variable method and generalized inverse method.As follows:1.The all-pole similarity is proposed for the time-delay of IMC-PID system,put into internal model control for the first time,applied to IMC-PID of the transformation process,and derived from the IMC-PID control parameters calculated General formula to facilitate the realization of simulation programming.The simulation results show that the entire pole approximation method similar to the high level of effective systems of control and still has strong robustness. 2.It is proposed two methods of v-norm decoupling multi-variable IMC design:multi-variable v-norm decoupling IMC design and multi-variable IMC design with v-norm decoupling system.They were carried out systematic analysis of decoupling,derived from IMC controller in detail,and calculate the general formula of controller.The different v-norm IMC-PID controller can be got.So formed a relatively complete set of analysis and control design of the system3.NLJ optimization algorithm has been used in this two IMC method, and improves a comprehensive tracking features and decoupling of ITAE optimal as the goal for the filter parameters optimization.And put them into simulation.The results are good,which haveconfirmed these two design methods very good versatility and the optimized algorithm validity.4.For the non-square systems,two kinds of IMC design methods have been proposed:main manipulated variable method and generalized inverse method.The simulation results show that effective control.This article was the first multi-variable system of the IMC-PID design research,and gives the feasibility of methods,which is an innovative.
【Key words】 internal model control; multivariable systems; decoupling control; time delays; NLJ optimization;
- 【网络出版投稿人】 北京化工大学 【网络出版年期】2008年 11期
- 【分类号】TP273
- 【被引频次】21
- 【下载频次】700