作者: Guilherme A. Moreira , Gabriela A. Micheloud , Alejandro J. Beccaria , Héctor C. Goicoechea
DOI: 10.1016/J.BEJ.2006.12.025
关键词: Desirability function 、 Pulp and paper industry 、 Multiple response optimization 、 Effluent 、 Sugar cane 、 Response surface methodology 、 Artificial neural network 、 Wastewater 、 Mathematics 、 Environmental engineering 、 Bacillus thuringiensis 、 Biotechnology 、 Bioengineering 、 Biomedical engineering
摘要: Abstract An experimental mixture design coupled with data analysis by means of both response surface methodology (RSM) and artificial neural networks (ANNs) followed multiple optimization through a desirability function, was applied to the production δ-endotoxins from Bacillus thuringiensis var. kurstaki . The composition culture medium defined testing three regional effluents: milky effluent, beer wastewater sugar cane molasses. Both RSM ANNs accomplished goal pursued in this work, predicting optimal effluents. provided more reliable results due complexity models be fitted. selected blend was: 74%, 26% 0%, respectively for each above-mentioned