Classification of fermentation process models using recurrent neural networks
The authors propose the use of recurrent neural networks for the classification of batch fermentation data. A recurrent neural network uses temporal information on the state variables together with the time values. The procedure consists of using time as a supplementary input and the self-normalization of the training data. The time input permits the use of data measured at non-constant intervals of time, and by a parameter sensitivity study it is shown that the self-normalization assigns to a family of models a unique representation. This representation is used to train the recurrent neural network instead of the family itself, and thus the dimensions of the network and the training time are reduced. The trained neural network is used for the classification of real lactic fermentation data. The model which best describes the dynamics of the process is selected from several possible models. The results obtained show the good capacity of the neural structure to recognize the best model.
- A recurrent neural network for controlling a fed-batch fermentation of B. t
- Recurrent neural networks for time series classification
- Application of radial basis function and feedforward artificial neural networks to the Escherichia coli fermentation process
- scientific article; zbMATH DE number 1928674
- Neural network modelling of continuous microbial cultivation accounting for the memory effects
- Application of radial basis function and feedforward artificial neural networks to the Escherichia coli fermentation process
- Recurrent neural networks for time series classification
- Multiple proportion case-basing driven CBRE and its application in the evaluation of possible failure of firms
- Time accounting artificial neural networks for biochemical process models
- Neural network modelling of continuous microbial cultivation accounting for the memory effects
- scientific article; zbMATH DE number 2079800 (Why is no real title available?)
- A recurrent neural network for controlling a fed-batch fermentation of B. t
- scientific article; zbMATH DE number 1928674 (Why is no real title available?)
- Solving evolutionary problems using recurrent neural networks
This page was built for publication: Classification of fermentation process models using recurrent neural networks
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3152435)
