• We are available for your help 24/7
  • Email: info@isindexing.com, submission@isindexing.com

Paper Details

Neural network modeling and optimization of ?-amylase production from spoiled starch rich vegetables

Satish Babu Rajulapati1 and Lakshmi Narasu. M 2

Journal Title:Journal of Chemical, Biological and physical sciences

Modeling and Optimization of process variables for the improvement of ?-amylase production in the specially made starch medium by the cultivation of Aspergillus Niger was performed using artificial neural networks and Genetic Algorithm (G.A). Cultivation of Aspergillus Niger was conducted in submerged fermentation in the starch medium. Initially, the effect of incubation time (12-72 hours), pH (4-8), Temperature (25-450C), starch concentration (5- 25 mg/mL) and Inoculum size (5-25 v/v%) on two objectives i.e total amount of crude enzyme and its activity was examined. It is observed that process variables were found to have a significant influence on the enzyme production, and it was further modeled and optimized using neural networks and Genetic Algorithm (GA). Predicted values were compared between statistical model and neural networks model. neural network model predicted more accurately than a statistical model. Then multi objective optimization was done with the genetic algorithm. Optimal process variables found from Genetic Algorithm (GA) were- Incubation time (50hrs), pH (6.5), Temperature (340C), Starch Concentration (17 mg/mL) and Inoculum size (25%). At these variables, maximum concentration of protein and its activity was observed as 9.09 mg/ml and 96.4 IU/mL respectively.