Scholarlink

<p>The development of sustainable biodiesel production systems requires the integration of non-edible feedstocks, waste-derived heterogeneous catalysts, and advanced modeling approaches capable of capturing nonlinear process behavior. This study presents an integrated experimental and intelligent modeling framework for biodiesel production from <i>Carica papaya</i> seed oil using a calcium-rich heterogeneous catalyst derived from papaya peel waste, enabling simultaneous valorization of biomass residues as both feedstock and catalyst. The catalyst was synthesized via calcination and KOH activation, resulting in a substantial increase in BET surface area (from 1.2&#xa0;m² g⁻¹ to 150.5&#xa0;m² g⁻¹) and pore volume (from 0.005&#xa0;cm³ g⁻¹ to 0.25&#xa0;cm³ g⁻¹), indicating enhanced surface basicity and porosity. A Central Composite Design (CCD) comprising 32 experimental runs was employed to evaluate the effects of methanol-to-oil ratio, catalyst loading, reaction temperature, reaction time, and agitation speed on biodiesel yield and fuel properties. The experimental data were modeled using Response Surface Methodology (RSM), Artificial Neural Networks (ANN), and Adaptive Neuro-Fuzzy Inference System (ANFIS). Analysis of variance (ANOVA) confirmed that methanol-to-oil ratio and reaction temperature are the most statistically significant factors (<i>p</i> &lt; 0.05), with significant quadratic effects indicating nonlinear system behavior. The quadratic RSM model demonstrated good predictive capability for biodiesel yield (R² = 0.9451), while ANN and ANFIS models provided superior accuracy in capturing complex nonlinear relationships. Hybrid optimization integrating Genetic Algorithm (GA), Grey Wolf Optimizer (GWO), and Grasshopper Optimization Algorithm (GOA) revealed that ANFIS-based models consistently converged to stable optimal solutions, demonstrating robustness and reduced sensitivity to optimizer selection. Under optimized conditions, biodiesel yield reached 98%, with kinematic viscosity (5.23&#xa0;mm² s⁻¹) and cetane index (68) satisfying ASTM D6751 specification. GC–MS analysis confirmed the formation of fatty acid methyl esters, validating successful transesterification. Overall, this study establishes a robust and sustainable framework for biodiesel production through dual biomass valorization and intelligent hybrid optimization, with strong statistical validation and significant potential for industrial application and circular bioeconomy advancement.</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

Hybrid intelligent optimization of biodiesel production from papaya seed oil using waste papaya peel–derived catalyst

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月21日

查看全文
  • Emmanuel Nnali-Uroh,
  • Chizoo Esonye,
  • Dominic Okechukwu Onukwuli,
  • Matthew Chukwudi Menkiti,
  • Sunday Uzochukwu John,
  • Yakubu Yunusa

摘要

The development of sustainable biodiesel production systems requires the integration of non-edible feedstocks, waste-derived heterogeneous catalysts, and advanced modeling approaches capable of capturing nonlinear process behavior. This study presents an integrated experimental and intelligent modeling framework for biodiesel production from Carica papaya seed oil using a calcium-rich heterogeneous catalyst derived from papaya peel waste, enabling simultaneous valorization of biomass residues as both feedstock and catalyst. The catalyst was synthesized via calcination and KOH activation, resulting in a substantial increase in BET surface area (from 1.2 m² g⁻¹ to 150.5 m² g⁻¹) and pore volume (from 0.005 cm³ g⁻¹ to 0.25 cm³ g⁻¹), indicating enhanced surface basicity and porosity. A Central Composite Design (CCD) comprising 32 experimental runs was employed to evaluate the effects of methanol-to-oil ratio, catalyst loading, reaction temperature, reaction time, and agitation speed on biodiesel yield and fuel properties. The experimental data were modeled using Response Surface Methodology (RSM), Artificial Neural Networks (ANN), and Adaptive Neuro-Fuzzy Inference System (ANFIS). Analysis of variance (ANOVA) confirmed that methanol-to-oil ratio and reaction temperature are the most statistically significant factors (p < 0.05), with significant quadratic effects indicating nonlinear system behavior. The quadratic RSM model demonstrated good predictive capability for biodiesel yield (R² = 0.9451), while ANN and ANFIS models provided superior accuracy in capturing complex nonlinear relationships. Hybrid optimization integrating Genetic Algorithm (GA), Grey Wolf Optimizer (GWO), and Grasshopper Optimization Algorithm (GOA) revealed that ANFIS-based models consistently converged to stable optimal solutions, demonstrating robustness and reduced sensitivity to optimizer selection. Under optimized conditions, biodiesel yield reached 98%, with kinematic viscosity (5.23 mm² s⁻¹) and cetane index (68) satisfying ASTM D6751 specification. GC–MS analysis confirmed the formation of fatty acid methyl esters, validating successful transesterification. Overall, this study establishes a robust and sustainable framework for biodiesel production through dual biomass valorization and intelligent hybrid optimization, with strong statistical validation and significant potential for industrial application and circular bioeconomy advancement.