State inference in bioreactors through neural network consortium

• Grant number: 06/04446-0
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: November 01, 2006
• End date: October 31, 2007
• Field of knowledge: Engineering - Chemical Engineering - Chemical Process Industries
• Principal Investigator: Roberto de Campos Giordano
• Grantee: Daniel Seino Ferreira
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil

Abstract

This project focuses on the application of computational intelligence for the on-line state inference of bench-sacle, stirred-and-aerated-tank biorreactors. A softsensor architecture (consortium of MLP – MultiLayer Perceptron nerual networks) will be implemented and validated. For two cultivation processes. The applicant already has experience in the automation of this type of process. From 09/05 to 08/06 he had an ITI-CNPq scholarship (CNPq Proc. 120247/2005-4), also supervised by Prof. Roberto C. Giordano. During this period, he has been trained in the implementation of algorithms for supervision and control, using LabVies (national Instruments), and in the integration of this system with the hardware in two bioreactors (in DEQ/UFSCar and Butantan Institute). Thus, his training for a broader project such as the present one has already taken place. During this project, the computational tools necessary to infer cell, substrate and product concentrations during the cultivation of two bacteria will be developed and implemented. The studied microorganisms are: Bacillus megaterium, producing the enzyme peniciline G acylase (PGA), in colaboration with the Laboratory for Development of Enzymatic Processes in DEQ/UFSCar, coordenated by Profa. Raquel Giordano; procs. 98/10860-5, 99/01326-8, 04/12482-0, 04/06984-3; and of Streptococcus pneumoniae, sinthetizing capsules of polyssacharide for human pneumococic vaccines (in cooperation with the Fermentation Laboratory of the Biotechnology Group of Instituto Butantan, coordenated by Dra. Martha Tanizaki, procs. 98/10577, 03/10125-3). For validation of the algorithms proposed herein, new experiments will be employed – carried on by the teams fo these projects. Both Instituto Butantan and DEQ/UFSCar have bioreactors with all the necessary instrumentation, monitores by LabView. Thus, this work is integrated to two FAPESP projects: “Computational intelligence applied to the monitoring and control of bioreactors during the cultivation of wild and recombinant microorganisms for the production of PGA”, coordinated Prof. Roberto Giordano (Laboratory for Deveolpement and Automation of Bioprocesses, LaDABio, DEQ/UFSCar; proc. 05/02044-9); and “Scale-up of bioprocesses for production of vaccines”, a PRONEX-FAPESP project, (proc. 03/10125-3), coordenated by Dra. Martha Tanizaki (Instituto Butantan), which team of researchers includes Prof. Roberto Giordano.