Sunday, May 4, 2008
2-51
Development of Software Sensors for Real Time Operation: Application for Bioethanol Production
Elmer Ccopa Rivera1, Nádson Murilo Nascimento Lima1, Rafael Ramos de Andrade1, Rubens Maciel Filho2, and Aline Carvalho da Costa1. (1) School of Chemical Engineering, State University of Campinas, Laboratory of Optimization, Design and Advanced Control, Cidade Universitária Zeferino Vaz, CEP: 13081-970, Campinas - SP, Brazil, (2) Department of Chemical Process – School of Chemical Engineering, State University of Campinas, Cidade Universitária, P.O. Box 6066, Campinas, Brazil
One of the main difficulties in biotechnological processes nowadays is the lack of robustness of the operation in the presence of fluctuations in the quality of raw material, variations of dominant microorganisms and deviations in temperature, which cause inherent process variability. This can be avoided by frequent adjustments in the operational conditions and control settings of the process, which requires an efficient monitoring with reliable sensors.
The results obtained in this work have shown that it is possible to infer the key variables of an ethanol fermentation process (concentration of biomass, substrate, and bioethanol) from secondary measurements, such as pH, turbidity, CO2 flow rate and temperature. Two alternatives were considered for the development of the software sensor, a Multilayer Perceptron Neural Network and a Takagi-Sugeno fuzzy model.
The application of the two software sensors on experimental data provided a reasonable description of the concentration trajectories, providing real time information of the key process variables.
The results obtained in this work have shown that it is possible to infer the key variables of an ethanol fermentation process (concentration of biomass, substrate, and bioethanol) from secondary measurements, such as pH, turbidity, CO2 flow rate and temperature. Two alternatives were considered for the development of the software sensor, a Multilayer Perceptron Neural Network and a Takagi-Sugeno fuzzy model.
The application of the two software sensors on experimental data provided a reasonable description of the concentration trajectories, providing real time information of the key process variables.
See more of Poster Session One
See more of General Submissions
See more of The 30th Symposium on Biotechnology for Fuels and Chemicals (May 4 -- 7, 2008)
See more of General Submissions
See more of The 30th Symposium on Biotechnology for Fuels and Chemicals (May 4 -- 7, 2008)