Artificial Neural Network Modeling in Pretreatment of Garden Biomass for Lignocellulose Degradation

摘要

In this work, Artificial Neural Network (ANN) model was developed for the garden biomass pretreatment process from the available experimental data of previous work. ANN model was studied to compare the results obtained through Response Surface Methodology (RSM), both graphically and numerically. The influence of process variables such as reaction temperature, Fe2+ concentration and H2O2 concentration on lignocellulose degradation of garden biomass were investigated by this model. The ANN analysis using Matlab, R2012a has shown promising results for this nonlinear system with the correlation coefficient of 0.9663 for cellulose and 0.9699 for lignin, indicating good fit. ANN found to be the effective tool for modeling the experimental data of lignin and cellulose degradation from pretreated biomass and showed a good match with experimental data than RSM. We found that reaction temperature 50 degrees C, Fe2+ concentration 250ppm and H2O2 concentration 10000ppm are the optimum conditions for maximum lignin and cellulose degradation i.e. 47.688% of cellulose and 57.529% of lignin from pretreated garden biomass.