In this work, the fermentation of “ricotta cheese whey” for the production of ethanol was simulated by means of a multiple hybrid neural model (HNM), obtained by coupling neural network approach to mass balance equations for lactose (substrate), ethanol (product) and biomass.AHNMrepresents an alternative method that may allow predicting the behaviour of complex systems, such as biotechnological processes, in a more efficient way. Some well-assessed phenomena, in fact, are described by a fundamental theoretical approach; some others, being very difficult to interpret, are analysed by means of rather simple “cause–effect” models, based on artificial neural networks. The experimental data, necessary to develop the model, were collected during batch fermentation runs. For all the proposed networks, the inputs were chosen as the operating variables with the highest influence on reaction rate. Simulation results showed the ability of the developed model to represent the process dynamics. The HNM was capable of an accurate representation of the system behaviour by predicting biomass, lactose and ethanol concentration profiles with an average error percentage lower than 10%. Moreover, the hybrid approach showed the ability to limit error propagation into the models that can be caused by the purely black-box nature, typical of neural networks.

A hybrid neural approach to model batch fermentation of "ricotta cheese whey" to ethanol

CURCIO, Stefano;CALABRO', Vincenza;
2010-01-01

Abstract

In this work, the fermentation of “ricotta cheese whey” for the production of ethanol was simulated by means of a multiple hybrid neural model (HNM), obtained by coupling neural network approach to mass balance equations for lactose (substrate), ethanol (product) and biomass.AHNMrepresents an alternative method that may allow predicting the behaviour of complex systems, such as biotechnological processes, in a more efficient way. Some well-assessed phenomena, in fact, are described by a fundamental theoretical approach; some others, being very difficult to interpret, are analysed by means of rather simple “cause–effect” models, based on artificial neural networks. The experimental data, necessary to develop the model, were collected during batch fermentation runs. For all the proposed networks, the inputs were chosen as the operating variables with the highest influence on reaction rate. Simulation results showed the ability of the developed model to represent the process dynamics. The HNM was capable of an accurate representation of the system behaviour by predicting biomass, lactose and ethanol concentration profiles with an average error percentage lower than 10%. Moreover, the hybrid approach showed the ability to limit error propagation into the models that can be caused by the purely black-box nature, typical of neural networks.
2010
Grey-box models; Artificial neural networks; Batch fermentation; Modeling
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.11770/153861
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 42
  • ???jsp.display-item.citation.isi??? 38
social impact