APPLICATION OF RESPONSE SURFACE METHODOLOGY (RSM) AND ARTIFICIAL NEURAL NETWORK (ANN) FOR ACHIEVING DESIRE BA IN THE BIOTRANSFORMATION OF BENZALDEHYDE USING FREE CELLS OF SACCHAROMYCES CEREVISAE AND THE EFFECT OF Β-CYCLODEXTRIN

Adepoju, T. F. and Olawale, O. and Ojediran, J. O. and Layokun, S. K. (2014) APPLICATION OF RESPONSE SURFACE METHODOLOGY (RSM) AND ARTIFICIAL NEURAL NETWORK (ANN) FOR ACHIEVING DESIRE BA IN THE BIOTRANSFORMATION OF BENZALDEHYDE USING FREE CELLS OF SACCHAROMYCES CEREVISAE AND THE EFFECT OF Β-CYCLODEXTRIN. International Journal of Sustainable Energy and Environmental Research, 3 (1). pp. 62-79.

[img] Text
IJSEER-2014-3(1)-62-79.pdf

Download (1MB)

Abstract

This work dwells on the production of benzene alcohol (BA) from the biotransformation of benzaldehyde using free cells of Saccharomyces cerevisae and effects of β-Cyclodextrin. Meanwhile, the properties of BA produced was evaluated. The effects of five variables considered in this research work were evaluated using RSM and ANN. The root mean square error, the coefficient of determination, the adjusted coefficient of determination and the predicted values were used to compare the performance of the RSM and ANN models. The RMSE and R2 of RSM and ANN were 2.00 and 0.0739; 0.9898 and 0.99206, respectively. The R2 adj. and the predicted values of RSM and ANN were found to be 0.98416 and 0.9889 and 327.259 mg/100 ml and 351.50 mg/100 ml. The quality of BA showed that at room temperature, BA was colourless liquid with density 1.030 kg/dm3, the boiling point and refractive index was found to be 204 ± 2 0C and 1.5453, respectively. The results indicated the ANN model to have higher predictive capability than RSM model. Thus, the ANN methodology presents a better alternative than the RSM model. The quality of produced BA was found to be in line with Analytic grade values.

Item Type: Article
Subjects: S Agriculture > S Agriculture (General)
Depositing User: ELDER OGUNTAYO SUNDAY ADEBISI
Date Deposited: 30 Nov 2018 17:12
Last Modified: 17 Apr 2019 11:53
URI: https://eprints.lmu.edu.ng/id/eprint/1523

Actions (login required)

View Item View Item