caa a22 4500 378927744 CHVBK 20180305123621.0 cr unu---uuuuu 161128e20040115xx s 000 0 eng 10.2202/1542-6580.1128 doi (NATIONALLICENCE)gruyter-10.2202/1542-6580.1128 Combined Mechanistic and Empirical Modelling [Elektronische Daten] [Belmiro Duarte, P. M. Saraiva, C. C. Pantelides] We consider a hybrid modelling approach based on the combination of prior knowledge, in the form of mechanistic models, with tools for the extraction of knowledge from operating data: the first component captures first-principles system behavior features, while the second one accounts for the differences found between mechanistic predictions and real data. The empirical modelling methods tested include backpropagation Artificial Neural Networks (ANN), Multivariate Adaptive Regressive Splines (MARS) and Regression Analysis (RA). In the proposed hybrid structure, mechanistic predictions provide additional inputs for the empirical module. To evaluate the performance of the various combinations of modules and hybrid approaches, we consider two simulated case studies (involving a CSTR with a reversible reaction and a fed-batch penicillin fermentation process). In all of the above, our hybrid structure based on a mechanistic module together with an empirical component has outperformed other approaches, and the mechanistic model/MARS combination resulted in lowest overall prediction errors. ©2011 Walter de Gruyter GmbH & Co. KG, Berlin/Boston Hybrid Modelling nationallicence Gray Box Modelling nationallicence Multivariate Adaptive Regression Splines nationallicence Duarte Belmiro ISEC, Polytechnic Institute of Coimbra, bduarte@isec.pt aut Saraiva P. M. DEQ, Universidade de Coimbra, pas@eq.uc.pt aut Pantelides C. C. Imperial College, c.pantelides@imperial.ac.uk aut International Journal of Chemical Reactor Engineering De Gruyter 2/1(2004-01-15) 2:1 2004 2 ijcre https://doi.org/10.2202/1542-6580.1128 text/html Onlinezugriff via DOI 1 research article jats NATIONALLICENCE 856 40 https://doi.org/10.2202/1542-6580.1128 text/html Onlinezugriff via DOI NATIONALLICENCE 700 1- Duarte Belmiro ISEC, Polytechnic Institute of Coimbra, bduarte@isec.pt aut NATIONALLICENCE 700 1- Saraiva P. M. DEQ, Universidade de Coimbra, pas@eq.uc.pt aut NATIONALLICENCE 700 1- Pantelides C. C. Imperial College, c.pantelides@imperial.ac.uk aut NATIONALLICENCE 773 0- International Journal of Chemical Reactor Engineering De Gruyter 2/1(2004-01-15) 2:1 2004 2 ijcre CC0 http://creativecommons.org/publicdomain/zero/1.0 nationallicence BK010053 XK010053 XK010000 NATIONALLICENCE NATIONALLICENCE NL-gruyter