On‐line prediction of fermentation variables using neural networks

Thibault, Jules; Breusegem, V. Van; Chéruy, A.

doi:10.1002/bit.260361009

Cited by 204 publications

(61 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The strategy when using neural networks to solve a given problem is to choose the network architecture best adapted to the problem, and a learning process whereby representative examples of the knowledge to be acquired are shown to the network, which self-organizes to integrate, within its structure, the information being presented by adjusting accordingly the synaptic strength of the neural connections. Learning usually requires showing to the network examples of the learning set several times [7]. The success in obtaining a reliable and robust network depends strongly on the choice of process variables involved, as well as the available sets of data and the domain used for training purposes.…”

Section: Neural Network Modelsmentioning

confidence: 99%

Optimization of Fischer‐Tropsch Synthesis Using Neural Networks

Fernandes

2006

Chem Eng & Technol

View full text Add to dashboard Cite

Fischer-Tropsch synthesis is an important chemical process for the production of liquid fuels and olefins. Optimization of hydrocarbon products such as diesel and gasoline produced by Fischer-Tropsch synthesis usually requires the knowledge of the complex polymerization mechanism and the kinetic parameters associated with it in order to optimize production. The Fischer-Tropsch reaction mechanism is still not fully understood, making optimization a hard task. In this work, a neural network was used in substitution to the reaction mechanism to optimize diesel and gasoline production based on few experimental data for the reaction. The neural network has yielded satisfactory predictions of the product distribution (with prediction errors lower than 5 %) and the optimum operating conditions for gasoline and diesel production were found for a commercial iron based catalyst.

show abstract

Section: Neural Network Modelsmentioning

confidence: 99%

Optimization of Fischer‐Tropsch Synthesis Using Neural Networks

Fernandes

2006

Chem Eng & Technol

View full text Add to dashboard Cite

show abstract

“…The time series was expressed by a differential delay equation which was proposed by Mackey and Glass. The ability of neural networks to identify and predict different bioprocesses such as the dynamic fermentation systems has also been demonstrated by Thibault et al [7] The dynamic neural network was used for the on-line prediction of key variables of a CSTR-mode fermentation process. With neural networks to identify the detection of ethanol from the brewing fermentation process have also been performed [8].…”

Section: Backpropagation Neural Networkmentioning

confidence: 99%

Backpropagation neural network predictive control and control scheme comparison of 2,3-butanediol fermentation by Klebsiella oxytoca

Syu

Hou

1999

Bioprocess Engineering

View full text Add to dashboard Cite

A backpropagation neural network (BPN) was applied for the control study of 2,3-butanediol fermentation (2,3-BDL) carried by Klebsiella oxytoca. The measurements of cell mass and glucose were not included in the network models, instead, only the on-line measured product concentrations from the MIMS (membrane introduction mass spectrometer) were involved. Oxygen composition was chosen to be the control variable for this fermentation system for the formation of 2,3-BDL is regulated by oxygen. Oxygen composition was directly correlated to the measured product concentrations. A twodimensional (number of input nodes by number of data sets) moving window to supply data for on-line, dynamic learning of this fermentation system was applied. The input nodes of the networks were also properly selected. Two neural network control schemes for this 2,3-BDL fermentation were discussed and compared in this work. Fermentations often exist time delay due to the measurement and their slow reaction nature. Hence, the order of time delay for the network controller was also investigated.1 Introduction 2,3-BDL is the major product from the fermentation by Klebsiella oxytoca. As a secondary metabolite, it can be produced at an optimal yield under a partially anaerobic condition [1]. Some other metabolites such as acetate, ethanol, and acetoin are also secreted as the end products from this fermentation. The major metabolic pathway of this fermentation was shown in our previous work [2]. The fermentation products shown in that work are only limited to those small molecules that secrete extracellularly. The distribution of product formation in 2,3-BDL fermentation can be governed by the amount of supplied oxygen. Therefore, the secreted product concentrations can be directly correlated to the oxygen supply. The on-line measurement of four products, acetic acid, acetoin, ethanol, and 2,3-BDL, was successfully performed by a mass spectrometer modi®ed by the insertion of a membrane probe [3]. In general, glucose and cell mass are two important measurements involved in fermentation related studies. However, these two measurements will not be referred to in this study. The dissolved oxygen (DO) in the fermentation broth re¯ects the oxygen availability to the cells. Therefore, the DO level of the fermentation can be regulated by aeration rate or oxygen composition. In this study, oxygen composition is used as the control input variable.Instead of a static model, dynamic identi®cation with adaptively adjusted parameters was applied to the control system of this fermentation. The control system is performed by the process model and the controller model. The process model is obtained from system identi®cation which was proceeded dynamically in this work. Therefore, a precise process model (or system identi®cation) and a good controller model (control law) are two major components for a control system. In general, the model from system identi®cation is combined with controller model to become a control system so that the future control input comma...

show abstract

“…Based on earlier work (Yong et al, 1992), an exponential feed rate was employed. While the reactor model included segregational instability of the plasmid-bearing cells, it was extended to mimic large scale fermentations by adding 10% Gaussian noise as done by Thibault et al (1990).…”

Section: Applicationmentioning

confidence: 99%

Preliminary screening of neural network configurations for bioreactor applications

Patnaik

1996

Biotechnol Tech

View full text Add to dashboard Cite

A methodology and a computer code have been devised to perform a preliminary analysis of six types of neural networks commonly employed for bioreactor problems. Both static and time-varying data can be analysed, and the values of the parameters and/or sampling times can be chosen according to the system behavior. The results help to select a suitable network configuration for detailed training and application. This is illustrated for a fed-batch fermentation to produce recombinant [Ggalactosidase.

show abstract

On‐line prediction of fermentation variables using neural networks

Cited by 204 publications

References 13 publications

Optimization of Fischer‐Tropsch Synthesis Using Neural Networks

Optimization of Fischer‐Tropsch Synthesis Using Neural Networks

Backpropagation neural network predictive control and control scheme comparison of 2,3-butanediol fermentation by Klebsiella oxytoca

Preliminary screening of neural network configurations for bioreactor applications

Contact Info

Product

Resources

About