International Journal of Recent Trends in Engineering & Research

online ISSN

Artificial Neural Network(ANN) and response Surface Methodolgy (RSM) of extraction of pectin from sweet lemon peels by microbial protopectinase.

Publication Date : 03/11/2017

DOI : 10.23883/IJRTER.2017.3476.GJOXZ

Author(s) :

Dr. Kshama W.Murarkar , Dr. Pratrtima Shastri.

Volume/Issue :
Volume 3
Issue 10
(11 - 2017)

Abstract :

Pectin is an important byproduct of fruit and vegetable processing industry. Conventional process for solubilization of protopectin involves hot acid extraction, which generates acid effluent and causes degradation of pectin in the process. Application of microbial protopectinases (PPase) for solubilization of protopectin is expected to enhance the quality and recovery of pectin. Standardization of process parameters including to achieve maximum recovery needs extensive experimentation. Various parameters including enzyme source, enzyme concentration, treatment time, and composition waste influence the recovery. Artificial Neural Network (ANN) and Response surface methodology (RSM) are innovative tools for modeling and prediction of such complex of biotechnology processes. In the present work, experiments were carried out to standardize the optimum conditions for pectin extraction using microbial protopectinase (PPase) from Kluveromyces marxianus-MTCC 188 and Kluveromyces wickerhamii –MTCC 455 (pH 5.0 at 300C). ANN and RSM models was developed using elite-ANN© and Minitab 511 software to predict pectin yield as a function of enzyme concentration and time of treatment. Optimum showed excellent predictability with R2> 0.95 for both training and test data sets of ANN and R2> 0.90 for RSM.

No. of Downloads :




Traffic Stats

Total Visits : 4,165