Neural network designs for poly- ?-hydroxybutyrate production optimization under simulated industrial conditions

Patnaik, Pratap R.

doi:10.1007/s10529-005-1775-0

Cited by 20 publications

(29 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While the recurrent neurons and information recycle improve the accuracy of an Elman network, they reduce its speeds of convergence during training and responses to unseen situations. Nevertheless, the Elman network is the best choice, as previous results [8,34,39,40] and the present investigation have shown.…”

Section: Kinetic Modelssupporting

confidence: 62%

“…Sometimes it is helpful to include some bias neurons in these layers so as help the network ride over local minima or shallow troughs in its quest for a global optimum. Obviously the neurons and the flow of signals between them may be arranged in different configurations, and previous studies [8,34,39,40] have demonstrated that an Elman neural network is the most suitable for a reactor with finite dispersion. Such a network has the architecture shown in Fig.…”

Section: Kinetic Modelsmentioning

confidence: 99%

“…This approach has been adopted by many earlier studies of the PHB system [1,34,39,40] and other fermentations [42][43][44], which found it useful to explore bioreactor dynamics by generating data by solving a model with laboratory-scale kinetics and nonideal features that mimic real conditions. Finite dispersion in these applications and in the present work was included by employing mass balance equations with dispersion terms expressed by Pe [8,45].…”

Section: Data Generationmentioning

confidence: 99%

See 2 more Smart Citations

Cognitive optimization of microbial PHB production in an optimally dispersed bioreactor by single and mixed cultures

Patnaik

2008

Bioprocess Biosyst Eng

View full text Add to dashboard Cite

Cognitive (or intelligent) models are often superior to mechanistic models for nonideal bioreactors. Two kinds of cognitive models--cybernetic and neural--were applied recently to fed-batch fermentation by Ralstonia eutropha in a bioreactor with optimum finite dispersion. In the present work, these models have been applied in simulation studies of co-cultures of R. eutropha and Lactobacillus delbrueckii. The results for both cognitive and mechanistic models have been compared with single cultures. Neural models were the most effective for both types of cultures and mechanistic models the least effective. Simulations with co-culture fermentations predicted more PHB than single cultures with all three types of models. Significantly, the predicted enhancements in PHB concentration by cognitive methods for mixed cultures were four to five times larger than the corresponding increases in biomass concentration. Further improvements are possible through a hybrid combination of all three types of models.

show abstract

Section: Kinetic Modelssupporting

confidence: 62%

Section: Kinetic Modelsmentioning

confidence: 99%

Section: Data Generationmentioning

confidence: 99%

See 1 more Smart Citation

Cognitive optimization of microbial PHB production in an optimally dispersed bioreactor by single and mixed cultures

Patnaik

2008

Bioprocess Biosyst Eng

View full text Add to dashboard Cite

show abstract

“…On the other hand, increased complexity, linked to a lack of detailed knowledge of the metabolism, limit their applicability. This limitation can be overcome to some extent through the use of artificial neural networks, which have been applied to model a great number of systems, including PHB synthesis [31]. They do not require a rigorous mathematical description but rather rely on an exhaustive training methodology, with solutions coming from a heuristic search.…”

Section: Though Many Bacteria Have Been Isolated and Characterised Asmentioning

confidence: 99%

Building a predictive model for PHB production from glycerol

Rivero

Sun

Theodoropoulos

et al. 2016

Biochemical Engineering Journal

View full text Add to dashboard Cite

Poly-3-hydroxybutyrate (PHB) is a biodegradable biopolyester with plastic like properties, which on its own or as part of a heteropolymer, finds application in everyday products, competing directly with fossil fuel based plastics in terms of physical and mechanical properties. In nature, PHB is produced as an energy reservoir for the host cell, when environmental conditions limit growth. It is this inherent condition for PHB synthesis (i.e. an environment unsuitable for growth) that challenges design of conventional batch production systems. Balance between growth (driven by nitrogen availability) and PHB production (enhanced by an excess of carbon) is the critical aspect for consideration in such designs. However, selecting the best operating conditions is not obvious for this system and so a systematic approach has been used in this paper, utilising simulations based on a purpose built model to supplement experimental studies.The interaction between the carbon and nitrogen sources (glycerol and ammonium sulphate respectively) was carefully evaluated and incorporated into a low-structured model able to describe the dynamics of substrate consumption and product accumulation during C.necator DSM 545 cultivation at small scale. The kinetic parameters thus determined have been assumed to be constant, fixed accordingly, and the model used to predict the fermentation profiles for different operating conditions. Results showed good agreement with experimental data, supporting the efficacy of this approach. The dual utilization of multiple substrates suggests there is a system capacity to which both growth and PHB production contribute and that sets the maximum total biomass concentration. A logistic type term added to both growth and product rate equations enabled the effective decoupling of cell proliferation and PHB accumulation for a wide range of scenarios. In this way, the combination of predictive modelling and experimental verification potentially reduces, by a significant amount, the number of experiments required to establish operational targets such as specific growth rate and productivity as well as identifying often sought criteria such as optimum C:N ratio. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 List of symbols:Half saturation constant for glycerol (g/L)Inhibition constant for PHB production (g/L)Half saturation constant for nitrogen (g/L)Half saturation constant for PHB (g/L)Carrying capacity (g/L)Hill coefficient (-)Nitrogen concentration (g/L)Yield of PHB based on glycerol (g/g)Yield of non-PHB biomass based on nitrogen (g/g)Yield of non-PHB biomass based on glycerol (g/g)

show abstract

“…The method combines "first principles models" with modular artificial neural networks trained by the EM (Expectation-Maximization) algorithm (for details the reader should consult the works of Jordan and Jacobs 148 and Xu et a1. 149 ). The modular networks were used in order to treat the 'cells system' as a "highly complex network of metabolic reactions organized in modular pathways".…”

Section: Neural Network and Hybrid Models In Modelling Of Phas Biosymentioning

confidence: 99%

Mathematical Modelling as a Tool for Optimized PHA Production

Novák

Koller

Braunegg

et al. 2015

Chem.Biochem.Eng.Q.

View full text Add to dashboard Cite

c ARENA -Association for resource efficient and sustainable technologies, Inffeldgasse 21b, 8010 Graz, AustriaThe potential of poly(hydroxyalkanoates) (PHAs) to replace conventional plastic materials justifies the increasing attention they have drawn both at lab-scale and in industrial biotechnology.The improvement of large-scale productivity and biochemical/genetic properties of producing strains requires mathematical modeling and process/strain optimization procedures. Current models dealing with structurally diversified PHAs, both structured and unstructured, can be divided into formal kinetic, low-structured, dynamic, metabolic (high-structured), cybernetic, neural networks and hybrid models; these attempts are summarized in this review. Characteristic properties of specific groups of models are stressed in light of their benefit to the better understanding of PHA biosynthesis, and their applicability for enhanced productivity. Unfortunately, there is no single type of mathematical model that expresses exactly all the characteristics of producing strains and/or features of industrial-scale plants; in addition, the different requirements for modelling of PHA production by pure cultures or mixed microbial consortia have to be addressed. Therefore, it is crucial to sophisticatedly adapt and fine-tune the modelling approach accordingly to actual processes, as the case arises. For "standard microbial cultivations and everyday practices", formal kinetic models (for simple cases) and "low-structured" models will be appropriate and of great benefit. They are relatively simple and of low computational demand.To overcome the specific weaknesses of different established model types, some authors use hybrid models. Here, satisfying compromises can be achieved by combining mechanistic, cybernetic, and neural and computational fluid dynamics (CFD) models. Therefore, this hybrid modelling approach appears to constitute the most promising solution to generate a holistic picture of the entire PHA production process, encompassing all the benefits of the original modelling strategies. Complex growth media require a higher degree of model structuring. For scientific purposes and advanced development of industrial equipment in the future, real systems will be modelled by highly organized hybrid models.All solutions related to modelling PHA production are discussed in this review.

show abstract

Neural network designs for poly- ?-hydroxybutyrate production optimization under simulated industrial conditions

Cited by 20 publications

References 20 publications

Cognitive optimization of microbial PHB production in an optimally dispersed bioreactor by single and mixed cultures

Cognitive optimization of microbial PHB production in an optimally dispersed bioreactor by single and mixed cultures

Building a predictive model for PHB production from glycerol

Mathematical Modelling as a Tool for Optimized PHA Production

Contact Info

Product

Resources

About