Linearizing control of a binary distillation column based on a neuro-estimator

被引：8

作者：

González, J

Aguilar, R

Alvarez-Ramírez, J

Fernández, G

Barrón, M

机构：

[1] Univ Autonoma Metropolitana Azcapotzalco, Div Ciencias Bas & Ingn, Mexico City 07300, DF, Mexico

[2] Univ Autonoma Metropolitana Iztapalapa, Dept Ingn Procesos & Hidraul, Iztapalapa, Mexico

[3] Univ Iberoamer, Dept Matemat, Mexico City, DF, Mexico

来源：

ARTIFICIAL INTELLIGENCE IN ENGINEERING | 1999年 / 13卷 / 04期

关键词：

neural networks; adaptive control; uncertainty estimation; distillation column;

D O I：

10.1016/S0954-1810(99)00018-7

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

In this work, the LV-control problem in binary distillation columns is addressed. With least prior knowledge, a linear reference model with unknown terms is obtained. The time variations of the unknown terms are estimated using two on-line trained perceptrons. These estimates are subsequently used to design a feedback linearizing-like controller. The closed-loop behavior is analyzed through numerical examples. The resulting controller shows robustness against external disturbances and set-point changes. (C) 1999 Elsevier Science Ltd. All rights reserved.

引用

页码：405 / 412

页数：8

共 22 条

[1]

BRDYS MA, 1996, 13 TRIEN WORLD C IFA, P151

[2] ADAPTIVE-CONTROL OF NONLINEAR-SYSTEMS USING NEURAL NETWORKS [J].