Banca de DEFESA: MAYARA REZENDE CARVALHO
Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.STUDENT : MAYARA REZENDE CARVALHO
DATE: 19/02/2021
TIME: 14:09
LOCAL: meet.google.com/aij-knvj-ffj
TITLE:
EXTRATIVE DISTILLATION PROCESS OF HYDRATED ETHANOL: OPTIMIZATION VIA RESPONSE SURFACE METHODOLOGY AND MONITORING VIA PRINCIPAL COMPONENT ANALYSIS USING COMPUTER SIMULATION BY ASPEN PLUS
KEY WORDS:
Extractive distillation; Static simulation; Optimization; Response surface methodology; Dynamic simulation; Monitoring of continuous processes; Fault detection; Principal component analysis
PAGES: 94
BIG AREA: Engenharias
AREA: Engenharia Química
SUBÁREA: Processos Industriais de Engenharia Química
SPECIALTY: Processos Bioquímicos
SUMMARY:
The present work aimed the optimization and the monitoring of the extractive distillation process to obtain anhydrous ethanol, through computer simulation in the software Aspen Plus®. For optimization, a static process model was implemented, and for monitoring, a dynamic closed-loop model. Anhydrous ethanol is important because it is used as an additive in gasoline to mainly increase the octane rating of this fuel. One challenge is to obtain ethanol with content above 99% purity, for use in the transport sector. Currently, in Brazil, one of the most used processes for obtaining anhydrous ethanol is extractive distillation. To optimize this system, initially, a full factorial design 22 with 5 central points was used in order to investigate the effects of solvent temperature and reflux ratio, in addition to the interaction between both factors, on the concentration of ethanol in the top product and reboiler heat duty of the extractive column. The curvature was also verified, given the domain of the factors. The central composite rotational design (CCRD), which is part of the response surface methodology (RSM), was used to obtain the optimum points of operation of the process, given the simultaneous evaluation of both response variables. It was found that the best operating point for the solvent temperature would be 80° C, and the reflux ratio, 0.31. The values of the ethanol content in the top product and the reboiler heat duty are 99.6% and 1472kW, respectively. Regarding the monitoring of the process, a set of usual faults in extractive distillation columns was simulated. Step disturbances and variations in the standard deviation, with different magnitudes, in input and intermediate variables were evaluated. To detect these failures, the multivariate statistical technique called Principal Component Analysis (PCA) was applied, using the T2 of Hotteling and Q. PCA is a dimensionality reduction technique, commonly applied to problems monitoring continuous chemical processes. The degree of difficulty in detecting the failures was discussed, and this characteristic was correlated with the operation of extractive distillation columns.
BANKING MEMBERS:
Presidente - 1811284 - EDSON ROMANO NUCCI
Interna - 1810303 - PATRICIA DA LUZ MESQUITA
Interno - 1742695 - JUAN CANELLAS BOSCH NETO
Externo à Instituição - GUSTAVO MATHEUS DE ALMEIDA - UFMG
Externo à Instituição - ANTONIO JOSÉ GONÇALVES DA CRUZ