The paper presents the modelling results of the batch alcoholic fermentation bioreactor using both the first principle and the Artificial Neural Networks (ANN) approach. For the nonlinear fermentation process the first principle model validated with experimental data considers for the biomass the Monod and for substrate the Bovée and Strehaiano models, including the temperature influence. It has been used for the ANN model development. A design and training methodology is proposed for statistical modelling based on artificial neural networks. Comparison between the two models is performed revealing the incentives of the ANN modelling method for reducing the computational time and sparing the computer resources.
|Number of pages||8|
|Journal||Studia Universitatis Babes-Bolyai Chemia|
|Publication status||Published - Dec 1 2009|
All Science Journal Classification (ASJC) codes