Artificial neural networks to infer biomass and product concentration during the production of penicillin G acylase from Bacillus megaterium

Abstract

BACKGROUND: Production of microbial enzymes in bioreactors is a complex process including such phenomena as metabolic networks and mass transport resistances. The use of neural networks (NNs) to infer the state of bioreactors may be an interesting option that may handle the nonlinear dynamics of biomass growth and protein production.

Experimental and neural model analysis of styrene removal from polluted air in a biofilter

“…A detailed study on the effect of internal network parameters on the performance of back propagation networks19 and the procedure involved in selecting the best network topology has been described elsewhere 22. However, in most instances, literature suggests the use of a trial and error approach where the performance goal is set by the user 29. The best network architecture was chosen (Table 2) based on the maximum predictability of the network for the test data by analyzing values of the coefficient of determination 3…”

Performance prediction of a RPF-fired boiler using artificial neural networks

“…A detailed study on the effect of internal parameters on the performance of back propagation networks and the procedure involved in selecting the best network topology has been described elsewhere . The best network architecture (Table ) was based on the correlation coefficient ( R value) and was achieved by a vigorous trial and error approach , by keeping some training parameters constant and by slowly moving the other parameters over a wide range of values. As a first step of the modeling procedure, studies were initiated to identify the influence of both T c and N h for obtaining the suitable network topology.…”

Comparative analysis on the application of neuro‐fuzzy models for complex engineered systems: Case study from a landfill and a boiler

“…The selection of the training data was done randomly. The best network architecture was selected based on the R and MAPE values, which were achieved by trial‐and‐error approach (Silva, Pinotti, Cruz, Giordano, & Giordano, ). As a first step of the modelling procedure for NF network, the influence of both T c (training count/epochs/iteration) and N h (hidden layer node) was identified for obtaining the suitable network topology.…”

“…The selection of the training data was done randomly. The best network architecture was selected based on the R and MAPE values, which were achieved by trial-and-error approach (Silva, Pinotti, Cruz, Giordano, & Giordano, 2008). As a first step of the modelling proce-…”