https://orcid.org/0000-0001-7744-330X
![Sample our Economics, Finance,Business & Industry journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5931/TOC-Subject-Sample-2021-Economics-Finance-Business-Industry.jpg"})
ABSTRACT
Energy is an essential bedrock, which plays a high impact role in the running of domestic and industrial activities. Most energy used for these activities is majorly from conventional sources, which after combustion result in ecological imbalance, climatic affray, health hazards, and degradation of natural resources. Therefore, the quest for eco-friendly energy has made researchers to investigate on alternative energy, such as biogas. This review study presents a comprehensive analysis of various biomass used for biogas production considering the effects that co-digestion of these materials has on biogas yield, as well as the technology involved. It further evaluated the applicability of artificial intelligence for modeling and optimization of the anaerobic digestion process including the blend ratios, process parameters and so on. These indices determine the percentage methane yield from biomaterial. The review effort revealed that methane content of biomaterials digested without pre-treatment varies from 3.6 ± 0.7 to 443.55 ± 13.68 while the yield from biomaterials pre-treated using various methods varies from 301.38 mL
/g
to 0.73–5.87 L/week. Anaerobic digestion of the blends of cow dung, mango pulp, and Chromolaena odorata was reportedly necessary, as information is scantily available on it. The modeling of the resulting experimental data using different machine learning techniques such as an adaptive-neuro-fuzzy inference system and ANFIS for predicting biogas yield is a major information gathered in this study. The AI models reviewed have high correlation factors ranging from 0.8700 to 0.9998. This information gathered in this paper will motivate the production of useful fuel to complement the existing energy sources while offering a near-term and practical means for reduction of environmental pollution.
Nomenclature
| Acronym | = | Full Meaning |
| ANFIS | = | Adaptive-Neuro-Fuzzy Inference System |
| ANN | = | Artificial Neural Network |
| C/N ratio | = | Carbon Nitrogen Ratio |
| COD | = | Chemical Oxygen Demand |
| CST | = | Capillary Suction Time |
| CSTR | = | Continuously Stirred Tank Reactors |
| °C | = | Degree Celsius |
| DEX | = | Design Expert |
| DM | = | Data Mining |
| FIS | = | Fuzzy Inference System |
| FL | = | Fuzzy-Logic |
| GA | = | Generic Algorithm |
| MJ/m3 | = | Mega Joules per Meters Cube |
| mLd−1 | = | Millilitres per Day |
| mLg−1 | = | Millilitres per Gram |
| MLP Network | = | Multilayer Perceptron, Network |
| MVR | = | Multivariable Regression |
| NaOH | = | Sodium Hydroxide |
| OLR | = | Organic Loading Rate |
| PRB | = | Population Reference Bureau |
| R2-Value | = | Coefficient of Determination |
| RMSE | = | Root Means Square Error |
| R-Value | = | Coefficient of Correlation |
| TPOMW | = | Two-Phase Olive Mill Wastes |
| VFA | = | Volatile Fatty Acids |
| VS | = | Volatile Solid |
| Wt. | = | Weight |
Disclosure statement
No potential conflict of interest was reported by the author(s).