Nonlinear Model Predictive Control of a Distillation Column Using Wavenet Based Hammerstein Model

K., Ramesh and Anwaruddin, Hisyam and N., Aziz and S. R., Abd Shukor (2012) Nonlinear Model Predictive Control of a Distillation Column Using Wavenet Based Hammerstein Model. Engineering Letters, 20 (4). pp. 1-6. ISSN 1816-093X (Print); 1816-0948 (Online). (Published)

	PDF Nonlinear_model_predictive_control.pdf Restricted to Repository staff only Download (674kB) | Request a copy
Preview	PDF fkksa-2012-ramesh-Nonlinear Model Predictive Control of a Distillation.pdf Download (21kB) | Preview

Abstract

Distillation columns are fairly complex multivariable systems and needs to be controlled close to optimum operating conditions because of economic incentives. Nonlinear Model Predictive Control (NMPC) scheme is one of the best options to be explored for proper control of distillation columns. In the present work, a new wavenet based Hammerstein model NMPC has been developed to control distillation column. An experimentally validated equilibrium model was used as plant model in nonlinear system identification and in NMPC. Two multiple-input-single-output (MISO) wavenet based Hammerstein models are developed to model the dynamics of the distillation column. The nonlinear model parameters were estimated using iterative prediction error minimization method. The Unscented Kalman Filter (UKF) was used to estimate the state variables in NMPC and the NLP problem was solved using sequential quadratic programming (SQP) method. The closed loop control studies have indicated that the performance of developed NMPC scheme was good in controlling the distillation column

Item Type:	Article
Uncontrolled Keywords:	Distillation column; Nonlinear model predictive control; Sequential quadratic programming; Wavenet based; Hammerstein model
Subjects:	Q Science > QD Chemistry
Faculty/Division:	Faculty of Chemical & Natural Resources Engineering
Depositing User:	Ms. Hazima Anuar
Date Deposited:	17 May 2016 02:20
Last Modified:	11 Jan 2018 01:04
URI:	http://umpir.ump.edu.my/id/eprint/6772
Download Statistic:	View Download Statistics

Actions (login required)

View Item