engleski

Modeling of the pyruvate production with Escherichia coli: Comparison of mechanistic and neural network- based models

Three different models: unstructured mechanistic black-box model, input/output neural network based model and externally recurrent neural network model were used to describe pyruvate production process from glucose and acetate using genetically modified Escherichia coli YYC202 ldhA::Kan strain. The experimental data was used from batch and fed-batch experiments described recently [1, 2, 23, 24]. The neural networks was built using experimental data obtained in the fed-batch pyruvate production experiments with constant glucose feed rate. Validation of models was performed using experimental results obtained in the batch and the fed-batch pyruvate production experiments with constant acetate feed rate. Dynamics of substrates and products concentration changes were estimate using two neural network-based models for biomass and pyruvate. It was shown that neural networks could be used for the modeling of complex microbial fermentation processes even for the conditions in which mechanistic unstructured models cannot be applied.

pyruvate ; Escherichia coli ; modeling ; unstructured “ black-box” model ; input/output neural network-based model ; externally recurrent neural network model

nije evidentirano