Recurrent neural network modeling applied to expanded bed adsorption chromatography of chitosanases produced by Paenibacillus ehimensis

详细信息    查看全文

• 作者：Carlos Eduardo de Araú ; jo Padilha^a ; Carlos Alberto de Araú ; jo Padilha^b ; Domingos Fabiano de Santana Souza^a ; Jackson Araú ; jo de Oliveira^a ; Gorete Ribeiro de Macedo^a ; Everaldo Silvino dos Santos^a ; ^{everaldo@eq.ufrn.br} ; ^{everaldo.eq@gmail.com}
• 关键词：Adsorption ; Downstream processing ; Enzyme activity ; Chitosanases ; Modeling ; NARX
• 刊名：Chemical Engineering Research and Design
• 出版年：2017
• 出版时间：January 2017
• 年：2017
• 卷：117
• 期：Complete
• 页码：24-33
• 全文大小：1194 K
• 卷排序：117

文摘

Hydrodynamic experiments showed a stable expanded bed in presence of feedstock containing particulate. Yield and purification factor were little affected by cell presence. A neural network built using few neurons avoided overfitting. NARXs showed better performance to adjust the data of an expanded bed adsorption protocol than MLPs.