多变量非线性系统的模糊内模控制  

Fuzzy Internal Model Control of Nonlinear Multivariable Systems

作　　者：靳其兵[1] 林艳春[1] 袁琴[1] 赵大力[1] 

机构地区：[1]北京化工大学自动化研究所,北京100029

出　　处：《计算机仿真》2007年第2期134-136,190,共4页Computer Simulation

基　　金：北京市科技新星资助项目(2004B08);中国石油化工股份有限公司资助项目(X503014)

摘　　要：大多数的先进控制器是基于线性模型的,它们对化学工业中常见的非线性过程的控制效果并不能达到最优。因此,考虑使用非线性模型,以使控制性能获得改善。用基于T-S模型的自适应模糊聚类辨识算法对系统进行辨识。T-S模型是用线性的方程来描述非线性系统,从而利于求出模型的逆。而模型逆又是IMC的关键一步,因此选用这种基于T-S模糊模型的控制器(FIMC)来实现对非线性多变量系统的控制。对2输入2输出的非线性系统进行仿真,结果表明FIMC在多变量系统中可以实现好的控制。Internal model control（IMC） is one of the advanced control methods ,and is based on models. However, the majority of the IMC internal models are linear models and the controllers＇ performance in controlling the nonlinear processes common in the chemical industry is not so efficient. Thus a nonlinear model is adopted to improve control performance. An adaptive fuzzy clustering identification algorithm is employed to get the T - S fuzzy model of a system. This model is capable of giving an accurate representation of a nonlinear system. A fuzzy internal model control strategy is proposed. A novel inversion method is presented which allows the fuzzy T - S model to be inverted and used within the IMC structure. Simulation results are presented showing the fuzzy IMC performance on a nonlinear multivariable system. The performance of the fuzzy IMC in the tests is found to be satisfactory.

关 键 词：内模控制 模型辨识 模糊聚类 非线性系统 多变量控制 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]