Scholarlink

<p>Organic waste represents a sustainable alternative to fossil fuels and contributes to emission reduction; however, its energy value must be critically evaluated. Most previous studies estimate theoretical methane potential (BMPₜₕ) and composition of individual residues but often ignore co-digestion variability. This work investigated co-digestion of cow paunch content (CPC) with mechanically pretreated rice husks (RH), a neglected combination. Raw RH displayed a BMPₜₕ of 396.55 mL CH₄/g VS, while CPC produced 211.0 mL CH₄/g VS. Mixtures (M1–M4) yielded 274.86–328.81 mL CH₄/g VS, with chemical formulas deviating from predicted linear effects. Gaussian process regression predicted BMPₜₕ using ultimate analysis with exceptional accuracy (R² = 0.9915, MAE = 4.72). Importance analysis identified carbon and nitrogen as dominant variables. These results reveal that co-digestion influences methane yield and substrate chemistry, highlighting the necessity of accounting for feedstock interactions when estimating BMPₜₕ. The novelty of this study lies in combining ML with stoichiometric modelling to capture non-linear blending effects, thereby establishing a superior predictive and optimization framework for biogas generation from heterogeneous organic wastes, while supporting reliable anaerobic digestion system design.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Characterization and gaussian process modelling of cow paunch content and mechanically pretreated rice husk co-substrates for enhanced biogas production

  • 期刊论文

  • 发表日期：

    2026年4月19日

查看全文
  • Isaac Mbir Bryant,
  • Samuel Nketia Boateng,
  • Richard Arthur,
  • Martina Francisca Baidoo

摘要

Organic waste represents a sustainable alternative to fossil fuels and contributes to emission reduction; however, its energy value must be critically evaluated. Most previous studies estimate theoretical methane potential (BMPₜₕ) and composition of individual residues but often ignore co-digestion variability. This work investigated co-digestion of cow paunch content (CPC) with mechanically pretreated rice husks (RH), a neglected combination. Raw RH displayed a BMPₜₕ of 396.55 mL CH₄/g VS, while CPC produced 211.0 mL CH₄/g VS. Mixtures (M1–M4) yielded 274.86–328.81 mL CH₄/g VS, with chemical formulas deviating from predicted linear effects. Gaussian process regression predicted BMPₜₕ using ultimate analysis with exceptional accuracy (R² = 0.9915, MAE = 4.72). Importance analysis identified carbon and nitrogen as dominant variables. These results reveal that co-digestion influences methane yield and substrate chemistry, highlighting the necessity of accounting for feedstock interactions when estimating BMPₜₕ. The novelty of this study lies in combining ML with stoichiometric modelling to capture non-linear blending effects, thereby establishing a superior predictive and optimization framework for biogas generation from heterogeneous organic wastes, while supporting reliable anaerobic digestion system design.