Amaranthus hybridus waste solid biofuel: comparative and machine learning studies

Area of Science:

• Agricultural Science
• Renewable Energy
• Materials Science

Background:

• Fossil fuel depletion and environmental concerns necessitate sustainable energy alternatives.
• Agro-waste management poses challenges, requiring innovative disposal and utilization strategies.
• Bio-briquettes offer a promising renewable energy source derived from biomass.

Purpose of the Study:

• To develop and evaluate bio-briquettes from Amaranthus hybridus waste using cassava starch as a binder.
• To assess the impact of delignification on the combustion and physicochemical properties of the bio-briquettes.
• To optimize the bio-briquette production process using machine learning models.

Main Methods:

• Preparation of alkali-treated (TAHB) and untreated (UAHB) Amaranthus hybridus bio-briquettes with cassava starch binder.
• Characterization using Fourier Transform Infrared Spectroscopy (FTIR), Scanning Electron Microscopy (SEM), and Energy-Dispersive X-ray Fluorescence (EDXRF).
• Combustion and physicochemical parameter evaluation, alongside optimization using Adaptive Neuro-Fuzzy Inference System (ANFIS) and Fuzzy C-Means (FCM) clustering.

Main Results:

• Delignification significantly reduced lignin content in TAHB (11.47%) compared to UAHB (12.31%).
• Calorific value increased notably in TAHB (12.53 MJ kg⁻¹) versus UAHB (10.43 MJ kg⁻¹).
• FTIR and SEM confirmed morphological and structural changes, while EDXRF indicated low Potential Toxic Elements (PTEs). Machine learning models demonstrated high prediction accuracy (RMSE, MAE, MAPE).

Conclusions:

• Alkali treatment and delignification of Amaranthus hybridus waste enhance bio-briquette quality for cleaner combustion.
• The developed bio-briquettes are environmentally friendly and possess improved energy content, suitable as sustainable solid biofuels.
• Machine learning models effectively predict bio-briquette efficiency, aiding process optimization for renewable energy production.