Open AccessModeling of Sugar Crystallization through Knowledge Integration
Author(s) -
Georgieva P.,
Feyo de Azevedo S.,
Gonçalves M.J.,
Ho P.
Publication year - 2003
Publication title -
engineering in life sciences
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.547
H-Index - 57
eISSN - 1618-2863
pISSN - 1618-0240
DOI - 10.1002/elsc.200390019
Subject(s) - artificial neural network , process (computing) , population , black box , computer science , process modeling , data mining , artificial intelligence , machine learning , biological system , engineering , work in process , biology , operations management , demography , sociology , operating system
This paper reports on the comparison of three modeling approaches that were applied to a fed batch evaporative sugar crystallization process. They are termed white box, black box, and grey box modeling strategies, which reflects the level of physical transparency and understanding of the model. White box models represent the traditional modeling approach, based on modeling by first principles. Black box models rely on recorded process data and knowledge collected during the normal process operation. Among various tools in this group an artificial neural networks (ANN) approach is adopted in this paper. The grey box model is obtained from a combination of first principles modeling, based on mass, energy and population balances, with an ANN to approximate three kinetic parameters ‐‐ crystal growth rate, nucleation rate and the agglomeration kernel. The results have shown that the hybrid modeling approach outperformed the other aforementioned modeling strategies.