Control of a Batch Polymerization System Using Hybrid Neural Network - First Principle Model

In this work, the utilization of neural network in hybrid with first principle models for modelling and control of a batch polymerization process was investigated. Following the steps of the methodology, hybrid neural network (HNN) forward models and HNN inverse model of the process were first devel...

Full description

Saved in:
Bibliographic Details
Published in:Canadian journal of chemical engineering 2007-12, Vol.85 (6), p.936-945
Main Authors: Wei, Ng Cheah, Hussain, Mohamed Azlan, Wahab, Ahmad Khairi Abdul
Format: Article
Language:English
Subjects:
Applied sciences
batch polymerization
Chemical engineering
Exact sciences and technology
first principle model
hybrid neural networks
model-based control
modelling
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this work, the utilization of neural network in hybrid with first principle models for modelling and control of a batch polymerization process was investigated. Following the steps of the methodology, hybrid neural network (HNN) forward models and HNN inverse model of the process were first developed and then the performance of the model in direct inverse control strategy and internal model control (IMC) strategy was investigated. For comparison purposes, the performance of conventional neural network and PID controller in control was compared with the proposed HNN. The results show that HNN is able to control perfectly for both set points tracking and disturbance rejection studies. On a étudié dans ce travail l'utilisation d'un réseau neuronal hybride avec des modèles basés sur les premiers principes dans le but de modéliser et contrôler un procédé de polymérisation discontinu. Au niveau méthodologique, on a d'abord mis au point des modèles anticipatifs à réseau neuronal hybride (HNN) et des modèles inverses à HNN pour le procédé, puis on a étudié la performance du modèle en stratégie de contrôle inverse directe et en stratégie de contrôle de modèle interne (IMC). À des fins de comparaison, la performance du contrôleur de réseau neuronal et de PID a été comparée au HNN proposé. Les résultats montrent que le HNN est capable de contrôler parfaitement en ce qui a trait aussi bien au suivi des points de consigne que du rejet des perturbations.
ISSN:0008-4034
1939-019X
DOI:10.1002/cjce.5450850616