Optimization of the hydrolysis of lignocellulosic residues by using radial basis functions modeling and particle swarm optimization

摘要: The concentrations of glucose and total reducing sugars obtained by chemical hydrolysis three different lignocellulosic feedstocks were maximized. Two response surface methodologies applied to model the amount produced: (1) classical quadratic least-squares fit (QLS), (2) artificial neural networks based on radial basis functions (RBF). results applying RBF more reliable better statistical parameters obtained. Depending type biomass, Improvements in between 35% 55% when comparing coefficients determination (R²) computed for both QLS methods. Coupling models with particle swarm optimization calculate global desirability function, allowed perform multiple optimization. predicted optimal conditions confirmed carrying out independent experiments.