Scholarlink

<p>This study examines the efficacy of microbial fuel cells (MFCs) in treating landfill leachate while generating bioenergy. The MFCs achieved substantial reductions in chemical oxygen demand (62%) and nitrogen concentrations (78%), highlighting their role in sustainable waste management. Anammox bacteria and microbial consortia were identified as critical in optimizing nitrogen and COD removal. Experimental findings revealed an optimal power density (PD) of 283.38&#xa0;mW/m<sup>2</sup> and a coulombic efficiency (CE) of 24.98% under specific conditions: NH4-N concentration of 800&#xa0;mg/L, COD of 1325&#xa0;mg/L, pH 6.88, and voltage of 528.59&#xa0;mV. Using advanced machine learning models—CatBoost, XGBoost, Adaboost, and multiple linear regression—we predicted power density with high precision (R-squared: 0.92, RMSE: 0.05, MAE: 0.08). CatBoost emerged as the most accurate model, achieving an R-squared value of 0.999 and demonstrating robust predictive capabilities. A linear relationship between power density and COD concentration showed the importance of maintaining balanced organic matter levels for efficient MFC operation. This research bridges data-driven modelling and environmental engineering, providing practical strategies to enhance MFC performance. By integrating machine learning with bioelectrochemical systems, it offers actionable insights for sustainable leachate treatment, renewable energy production, and eco-friendly waste management practices.</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

Optimizing leachate treatment and energy generation in microbial fuel cells through advanced machine learning

  • 期刊论文

  • 发表日期：

    2025年2月18日

查看全文
  • M. R. Houmsi,
  • A. Ishaq,
  • S. J. Mohammad,
  • Z. T. Jagun

摘要

This study examines the efficacy of microbial fuel cells (MFCs) in treating landfill leachate while generating bioenergy. The MFCs achieved substantial reductions in chemical oxygen demand (62%) and nitrogen concentrations (78%), highlighting their role in sustainable waste management. Anammox bacteria and microbial consortia were identified as critical in optimizing nitrogen and COD removal. Experimental findings revealed an optimal power density (PD) of 283.38 mW/m2 and a coulombic efficiency (CE) of 24.98% under specific conditions: NH4-N concentration of 800 mg/L, COD of 1325 mg/L, pH 6.88, and voltage of 528.59 mV. Using advanced machine learning models—CatBoost, XGBoost, Adaboost, and multiple linear regression—we predicted power density with high precision (R-squared: 0.92, RMSE: 0.05, MAE: 0.08). CatBoost emerged as the most accurate model, achieving an R-squared value of 0.999 and demonstrating robust predictive capabilities. A linear relationship between power density and COD concentration showed the importance of maintaining balanced organic matter levels for efficient MFC operation. This research bridges data-driven modelling and environmental engineering, providing practical strategies to enhance MFC performance. By integrating machine learning with bioelectrochemical systems, it offers actionable insights for sustainable leachate treatment, renewable energy production, and eco-friendly waste management practices.