Control via neural networks of absorption columns in the production process of ethanol by fermentation

• Grant number: 05/02536-9
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: February 01, 2006
• End date: December 31, 2008
• Field of knowledge: Engineering - Chemical Engineering
• Principal Investigator: Ana Maria Frattini Fileti
• Grantee: Eduardo Eyng
• Host Institution: Faculdade de Engenharia Química (FEQ). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil

Abstract

During the fermentation process, some of ethanol produced is lost by evaporation, reducing the yield. An absorption column may be used to recover the product lost. This equipment is also applied in carbon dioxide treatment, at the dryer stage. Moreover, a strong control system is required to secure that a satisfactory performance will be achieved. In this work a feedforward controller based on inverse neural network model is proposed. Its supposed to control the gas composition at the out of a plate absorption column. This technique is applied on two situations: recovering of the amount of ethanol that was lost by evaporation and drying of carbon dioxide current. The control system performance must be compared with the PI or PID one. This comparison may be done with the performance of a stronger control strategy, like the Dynamic Matrix Control. Furthermore, the influence on the controllers performance of dead time, as uncertainties and noises on the sensors must be evaluated. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

EDUARDO, EYNG; FILETI, ANA M. F.. Control of absorption columns in the bioethanol process: Influence of measurement uncertainties. ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, v. 23, n. 2, p. 271-282, MAR 2010. (05/02536-9)

EYNG, EDUARDO; DA SILVA, FLAVIO VASCONCELOS; PALU, FERNANDO; FRATTINI FILETI, ANA MARIA. Neural network based control of an absorption column in the process of bioethanol production. Brazilian Archives of Biology and Technology, v. 52, n. 4, p. 961-972, JUL-AUG 2009. (05/02536-9)

Academic Publications

(References retrieved automatically from State of São Paulo Research Institutions)

EYNG, Eduardo. Feedforward-feedback control applied to the absorption columns of the ethanol production process by fermentation. 2008. Doctoral Thesis - Universidade Estadual de Campinas (UNICAMP). Faculdade de Engenharia Química Campinas, SP.