| Title |
Tikslinio produkto koncentracijos prognozavimo modelių tyrimas periodinėse su pamaitinimu E. coli kultivacijose / |
| Translation of Title |
Investigation of concentration estimation models for target product in E. coli fed-batch bio-synthesis. |
| Authors |
Kemėšis, Benas |
| Full Text |
|
| Pages |
49 |
| Keywords [eng] |
biotechnology ; target product ; mathematical models |
| Abstract [eng] |
Biotechnological industrial processes aim to produce as much high-quality desired (target) product as possible. However, it is impossible to directly assess target product concentration during cell cultivations. To solve this problem, engineers use software sensors that employ mathematical models which use information from physical measurements and state variables captured during the process. The paper describes and implements mathematical models with different structures to estimate the concentration of the target protein during fed-batch cultivations of E. Coli cell cultures. Mechanistic models described in the literature are implemented, together with developed empirical and hybrid models. Paper presents the procedure for identifying model parameters for mechanistic models and training algorithms for empirical and hybrid models based on artificial neural networks. The models are tested with experimental data and their results are compared according to selected criteria. |
| Dissertation Institution |
Kauno technologijos universitetas. |
| Type |
Master thesis |
| Language |
Lithuanian |
| Publication date |
2022 |
