Neural network modelling of fermentation processes

For the purpose of dynamic systems modeling, it was proposed to include feedback connections or delay elements in the classical feed-forward neural network structure so that the present output of the neural network depends on its previous values. These delay elements can be connected to the hidden and/or output neurons of the main neural network. Each delay element gets a value of a state variable at a past time instant and keeps this value during a single sampling period. The groups of delay elements record the values of the state variables for a given time period in the past. Changing the number of the delay elements, which belongs to one group, a shorter or a longer time period in the past can be accounted for. Thus, the connection weights determine the influence of the past process states on the present state in a similar way as it is in the time delay kernel or cause-effect relation membership function (CER-MF) models. Specific feed-forward neural networks with time delay connections are used to solve the problem of neural network chemostat modeling as well as specific kinetic rates modeling. The weights of the feedback connections obtained during model training are discussed as the points of a time delay kernel or as the strength levels in a CER model (the points in the CER-MF). The corresponding changes in these weights with the changing time period in the past are shown.

show abstract

“…6 Several neural network models of the process are trained, [13][14][15] and the best ones are discussed here.…”

Section: Resultsmentioning

confidence: 99%

Memory effects description by neural networks with delayed feedback connections

Koprinkova

Patarinska

Petrova

2004

Int. J. Intell. Syst.

View full text Add to dashboard Cite

show abstract

“…Also, the variation of the adjustment weights of DNN remains in a bounded region providing small estimation errors with respect to experimentally measured ones. Figures 4,5,6,7,8,9,10,and 11 show the evolution of the estimate and experimentally measured states (the biomass concentration) starting from different initial states. After 5 h of the evaluation, these curves have a very similar pattern.. As for the substrate concentration, the process was performed such a way that the concentration remained near to zero throughout the 12 h of evolution.…”

Section: Resultsmentioning

confidence: 99%

“…Static (back-propagation) NN were effectively applied in biotechnology's environment identi®cation for the prediction of fermentation process variables [6,7,8] and for the modeling of the fermentation process [9,10]. Because of the complexity and high nonlinearity of analyzed systems, it is required to obtain the kinetic parameter estimates of the bioreactors [11,12], which obviously demands the construction of on-line models in continuous or in discrete time [13,14].…”

Section: Static and Dynamic Nnmentioning

confidence: 99%

Identification of a fed-batch fermentation process: comparison of computational and laboratory experiments

Cabrera¹,

Poznyak²,

T.³

et al. 2002

Bioprocess and Biosystems Engineering

View full text Add to dashboard Cite

To identify a Saccharomyces cerevisiae fed-batch fermentation process, it is suggested that a differential (dynamic in continuous time) neural network (DNN) be implemented. A priori information on the dynamic equations and their parameters is not required to realize this approach. Based on real data, obtained from laboratory experiments, this DNN of a simple structure (six neurons) is shown to have a high capability of identifying this process (to reproduce the input±output relation) after several working hours of learning.

show abstract

“…These studies indicate that neural network models are more efficient in describing industrial applications than conventional linear models for the identification and prediction of non-linear dynamic systems [6].…”

Section: Introductionmentioning

confidence: 99%

Dynamic neural networks as a tool for the online optimization of industrial fermentation

Becker

Enders

Delgado

2002

Bioprocess and Biosystems Engineering

View full text Add to dashboard Cite

A system for online optimization of industrial fermentation based on a model with dynamic neural networks is described. The developed dynamic neural network, consisting of adapted neurons to consider the process dynamics, can model the complex, non-linear fermentation of beer in order to predict the future process. The predicted trajectories of gravity, pH, and diacetyl are in agreement with the experimental data measured at an automated pilot fermenter. It was possible to predict the future course of the batch fermentation as soon as 12 h of process data were available. In combination with the variational principle, the process model was used to optimize productivity. The temperature trajectory is optimized using a cost functional, including technical and technological conditions of the brewery in order to reduce the process time by steady product quality. The results show a reduction of the process time of up to 20%, which leads to an increase in utilization capacity.

show abstract

Neural network modelling of fermentation processes

Cited by 17 publications

References 13 publications

Memory effects description by neural networks with delayed feedback connections

Memory effects description by neural networks with delayed feedback connections

Identification of a fed-batch fermentation process: comparison of computational and laboratory experiments

Dynamic neural networks as a tool for the online optimization of industrial fermentation

Contact Info

Product

Resources

About