Prediction of bioethanol production from lignocellulosic compounds in an anaerobic digester with pneumatic-mechanical agitator by neural network

In this study, an anaerobic digester with pneumatic digester (PMD) was designed to ferment lignocellulosic compounds of wheat and rice straw and was compared with mechanical digester (MD). Ultrasonic pretreatment was performed with time values of 30 min, acid concentration of 1.905 %v/v and bagasse load of 6%. Then acid hydrolysis was performed in 45 min, acid concentration of 2.04 %v/v and temperature of 148.4 °C. Then biomass was poured for fermentation in PMD and MD digesters and the fermentation process took place within 20 days and the sampling sequence was every 5 days. According to the results, the pneumatic digester was able to produce 38.66% more bioethanol than the mechanical digester. H2S, CO and O2 compounds were also measured during fermentation. In the pneumatic digester, the production of H2S and O2 was lower than in the mechanical digester, but the amount of CO production in the pneumatic digester was significantly higher. Also, the amount of produced compounds created 50 fermentation predicted days was examined by the neural network and compared with the actual results. According to the results, it is possible to predict the trend of changes in fermentation compounds by the neural network and make a definite decision to ferment the straw, wheat and rice compounds. Finally, it can be said that with the help of pneumatic digestion system, the amount of harmful compounds produced during the process can be controlled and the amount of bioethanol production can be increased.