Prediction of biogas production from chemically treated co-digested agricultural waste using artificial neural network

• Published in: Fuel (Guildford) Vol. 280; p. 118573
• Main Author: Almomani, Fares
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 15-11-2020; Elsevier BV
• Subjects: Agricultural biomass; Agricultural wastes; Agrochemicals; Algorithms; Anaerobic digestion; Anaerobic processes; Animal wastes; Artificial neural networks; Biodegradability; Biodegradation; Biofuel potential; Biogas; Cattle manure; Chemical composition; Chemical treatment; Co-digestion; Cost analysis; Kinetic model; Lignin degradation; Moisture content; Neural networks; Neutrons; Operating temperature; Pretreatment; Sodium bicarbonate; Solid wastes; Substrates; Waste treatment; Water content; Kinetic model; Biofuel potential; Agricultural biomass; Co-digestion; Lignin degradation
• ISSN: 0016-2361; 1873-7153
• DOI: 10.1016/j.fuel.2020.118573

[Display omitted]
•Biofuel production from agricultural wastes and cow manure was studied at 35–55 °C.•Chemical treatment of substrate enhanced anaerobic digestion (AD) efficiency.•Artificial neural network (ANN) predicts the AD biogas production at high accuracy.•Co-digestion 70:30 wt% of AWs to CD produced 334.4. NL/kgVS of biogas.•Alkalinity treatment with 1.0 g NaHCO3/gVS improved methane production by 29.7%.
The present study evaluates the effect of co-digestion of agricultural solid wastes (ASWs), cow manure (CM), and the application of chemical pre-treatment with NaHCO3 on the performance of anaerobic digestion (AD) process. An Artificial neural network (ANN) algorithm was developed to model and optimize the cumulative methane production (CMP) from ASWs, CM, and their mixture under mesophilic and thermophilic conditions. The results demonstrated that co-digestion of ASWs with CM with a ratio of 70% to 30% produced the highest CMP of 334 ± 4 NL/kgVS in comparison with 230 ± 10 NL/kgVS for mono-digested substrate. The CMP was the highest for the substrate with moisture content (%MC) in the range of 34% to 48%, and it decreased for %MC > 50%. The chemical treatment with NaHCO3 improved the biodegradability of the substrate and increased the CMP by at least 43% with reference to the untreated substrate. An ANN model consists of three layers, 15 neutrons and 260 epochs accurately predict the CMP with 99.1% of data within ±10% deviation of the mean experimental value. The developed model can be used to forecast the CMP as a function of operating temperature, the substrate composition, and chemical dose, and can be used for scaling-up and cost analysis purposes.