Prediction and optimization of ethanol concentration in biofuel production using fuzzy neural network

Abstract

In recent years, producing economic al biofuels especially bio - ethanol from lignocellulosic materials has been wi dely considered. Fermentation is an important step in ethanol production process. Fermentation process is completely nonlinear and depends on some parameter s such as temperature, sugar content, and PH. One of the difficulties in producing biomass is finding t he optimum point of the inter related parameters in the fermentation step . In this study , an elaborate prediction Neuro - Fuzzy model was built to predict the bio - ethanol production from corn stover . Also , particle swarm optimization ( PSO) method was used to optimize the three studied parameters: temperature, glucose content, and fermentation time . The attained c orrelation c oefficient (0.99), and root mean square error ( 0.637 ) for model validation show the reliability of the model. Optimization of the model shows the optimum f ermentation time and required temperature quantities , 6 9 . 39 hours and 34 .5 0 ͦ C, respectively. The good result for ANFIS modeling on fermentation process in bio - ethanol pr oduction from corn stover shows that this method can be used to investigate more about other biomass lignocellulos sources.