| Autor: |
Kirilova, Elisaveta, Yankova, Sophia, Ilieva, Bilyana, Vaklieva-Bancheva, Natasha |
| Předmět: |
|
| Zdroj: |
Journal of Chemical Technology & Metallurgy; 2014, Vol. 49 Issue 5, p473-478, 6p |
| Abstrakt: |
In this study a new approach for the modeling of a biotransformation process of crude glycerol to formation of 1,3-propanediol by the help of bacteria Pseudomonas denitrificans 1625 is proposed. Results obtained from the experimental studies of the process are presented for the first time. For the purpose of modeling dynamic Nonlinear AutoRegressive with eXogenous inputs - NARX Artificial Neural Networks are developed. They are used for prediction of the change in the time of the concentration of the substrate and the product as a function of their previous values. The models are trained and validated with real data samplings, obtained from parallel batch experiments carried out at three different initial concentrations of the substrate selected for this purpose. The obtained results show an acceptable coincidence between the measured and the predicted from the model values. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
