The production of ethanol from this lignocellulosic biomass

Application of neural network for modeling the saccharification of biomass by concentrated acid hydrolysis has been investigated. Due to a more complex and unknown kinetics of the investigated reaction, the proposed approach based on application of neural networks is an efficient and accurate tool to solve modeling problem. The Influence of the main pretreatment variables including temperature, processing time, solid contents, and acid concentration on fermentable sugar generation was investigated. Levels for pretreatment temperature (65, 80, 90oC), process time (120, 180, 240 minute), solid content (5, 10, 15%) and concentration of sulfuric acid (20, 40, 60%) were selected. Glucose were analyzed by HPLC and modeled by two layers neural network with different topology. Two layers neural network with different neurons was studied. Despite using simple data and a simple neural network, high accuracy was achieved. The trained MLP that has one hidden layer with 11 Log-sig neurons has a lower average relative error for test data. Therefore, it is selected as the most suitable network to simulate the concentrated acid hydrolysis process .Keywords—biomass, glucose, Acid sulfuric, concentrated acid hydrolysis, neural network