Optimization of Biogas Output From Biomaterials via Python and ANN, and Its Digester Design and Fabrication

ABSTRACT Due to the instability of the energy sector leading to the high cost of purchasing non‐biomass fuel, there is a need to investigate an ideal biomass fuel that can be utilized in place of non‐biomass fuel and whose biomaterials are within reach. Still, the issues/complexity of manual data processing minimize the optimal biogas generated. Thus, ANN and Python coding techniques were applied to digest the Dacryodes edulis seeds with spoiled cooked egg and stew‐rice digestate in an anaerobic unit to maximize the output of biomass gas (biogas) generated. Different mixing ratios of substrate‐to‐inoculum of 1.09:1, 1.50:1, 2.50:1, 3.50:1, and 3.91:1 were deployed against 20.86, 25, 35, 45, and 49.14 days Hydraulic Reaction Time. The outcome demonstrates that ANN generated a maximum biogas output of 6.43 L with a mixing ratio of 3.5:1 over 45 days, comparable to 6.38 L via Python coding. Additionally, the ANN and the Python yielded model R 2 (coefficient of determination) values of 0.9909 and 1, respectively. Compared to ANN, the plots and data produced via Python coding are superior. The Gas Chromatography–Mass Spectroscopy (GC–MS) analysis of the produced biomass gas revealed 65.67% biomass methane and 24.21% carbon dioxide (CO 2 ) by volume, both within a reliable range for a sustainable biomass energy solution. Furthermore, the biogas flashpoint is reported as −178°C, indicating a significant degree of flammability, and the designed and fabricated digester met the requirement. Thus, the synthesized biomass gas optimal parameters can be deployed into the fabricated digester for domestic usage at a cheaper rate.