The cellulase production by Trichoderma viride was optimized using artificial intelligence-based techniques under solid state fermentation. In this study, a back propagation network was designed with Levenberg-Marquardt training algorithm, and the tangent sigmoid and pure linear functions were used as the transfer functions in the hidden and output layers of the ANN, respectively. An artificial neural network coupling genetic algorithms was used to optimize the process parameters, which include the mass ratio of wheat straw to wheat bran, moisture content and fermentation time. The ultimate process parameters of optimization were mass ration of wheat straw to wheat bran 2.9, moisture content 69.6 percent, and fermentation time 123.3h. Further test experiment showed that the final cellulase activity can reach to 11.62 U/g, which was the highest value among all the experimental results. This result indicates that the genetic algorithm based on a neural network model is a better optimization method for cellulase production in solid state fermentation. To improve the cellulase production, a mixed culture system of Trichoderma viride and Aspergillus niger was also developed. The cellulase activity increased by 7.40 percent with the addition of Aspergillus niger at 72h.
Integration of Artificial Neural Network Modeling and Genetic Algorithm Approach for Enrichment of Laccase Production in Solid State Fermentation by Pleurotus ostreatus
