Process modelling development through artificial neural networks and hybrid models

Citation
Lfm. Zorzetto et al., Process modelling development through artificial neural networks and hybrid models, COMPUT CH E, 24(2-7), 2000, pp. 1355-1360
Citations number
7
Categorie Soggetti
Chemical Engineering
Journal title
COMPUTERS & CHEMICAL ENGINEERING
ISSN journal
00981354 → ACNP
Volume
24
Issue
2-7
Year of publication
2000
Pages
1355 - 1360
Database
ISI
SICI code
0098-1354(20000715)24:2-7<1355:PMDTAN>2.0.ZU;2-E
Abstract
Developing fully mechanistic models for bioprocess is expensive and time-co nsuming. On the other hand, using pure 'black-box' approaches can lead to a misuse of available information, because there are aspects of the process that can be accurately described by simple equations as, for example, mass balances. This work analyses the use of different types of 'black-box' and hybrid models to outline the dynamics of a batch beer production. The hybri d models, combine mechanistic equations with 'black-box' techniques (reserv ed only for the unclear parts of the system), in order to achieve an effici ent use of the available information. The hybrid models can also be called 'grey-box' approaches. To generate the hybrid models, different level of in formation is introduced into the 'black-box' models, allowing for an intere sting model performance comparison in the end. Results demonstrate that the 'black-box' models present a good performance in the range of process cond itions used to develop them. However, the inclusion of mechanistic knowledg e into the hybrid models increase the model extrapolative capability. In th is work, artificial neural networks (ANN) are used as the main technique fo r both the 'black-box' models and the 'black-box' parts in the hybrid model s. (C) 2000 Elsevier Science Ltd. All rights reserved.