Title: Inferential state estimator design for a batch distillation column

Authors: Amiya K. Jana

Addresses: Department of Chemical Engineering, Birla Institute of Technology and Science-Pilani, Rajasthan 333 031, India

Abstract: This paper presents the design of an inferential state estimator for a batch distillation column. This observation scheme consists of a sensor model and a non-linear state estimator. The sensor model is derived to compute the liquid phase compositions from readily available temperature measurements since the composition analysers are expensive, difficult to maintain and give delayed response. The design approach of Reduced-Order Closed-Loop Estimator (ROCLE) with varying the number of measurement inputs has been studied first, followed by the design of Short-Cut Closed-Loop Estimator (SCCLE) for the batch rectifier. Satisfactory composition estimates and fast convergence are obtained through the several simulation experiments. The estimation method has confirmed its ability to reject the effects of error in estimated state vector initialisation, measurement and modelling noises, unmeasured disturbance and uncertainty. The lower computational requirements and simple design make the estimator attractive for online use.

Keywords: nonlinear state estimators; sensor modelling; batch distillation column; liquid phase compositions; temperature measurements; simulation.

DOI: 10.1504/IJMIC.2007.014620

International Journal of Modelling, Identification and Control, 2007 Vol.2 No.2, pp.75 - 87

Published online: 21 Jul 2007 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article