Scholarlink

<p>The development of biogas production as a sustainable and effective alternative has been fuelled by the growing need for renewable energy sources. Thus, the growing global demand for renewable energy has demanded more research interest in anaerobic digestion (AD). AD is among the simplest ways to boost biogas production productivity. Thus, AD is considered one of the best renewable energy options because it not only manages organic waste effectively but also recovers valuable renewable energy—such as methane—from food waste (FW) and garden or yard waste (YW). Through this process, AD provides a dual benefit: sustainable waste treatment and clean energy production. This results in increased nutrient bioavailability and biogas yield. An important challenge that restricts the acceptance and implementation of AD is the quality and quantity of biogas generated from the process. One practical approach to overcome this challenge is the optimisation of the process parameters. In this study, Methane yield efficiency can be enhanced by optimising the operational environment, utilising response surface methodology (RSM) and artificial neural network (ANN). The findings were compared for anaerobic co-digestion of food waste and yard waste based on statistical parameters in predicting the methane yield. Performance metrics such as R², MSE, and RMSE show that the RSM model outperformed the trained ANN model. The RSM approach demonstrated higher accuracy (R² = 0.983, MSE = 143.68, RMSE = 11.98) in predicting methane yield, whereas the ANN model showed lower predictive performance (R² = 0.938, MSE = 783.05, RMSE = 27.98). The individual and the interaction effect of pH, treatment temperature, and C/N ratio on methane yield were investigated by the RSM model. The obtained optimum condition (pH = 7.56, treatment temperature = 41.96&#xa0;°C, and C/<i>N</i> = 26.43) showed a methane yield of 535 mL/gVS<sub>fed</sub>. A comprehensive energy balance analysis was performed for this AD process, and a net positive energy balance of 16.811&#xa0;kJ/gVS was confirmed. This study revealed that the most effective modelling technique for predicting methane and highlights the importance of pH, treatment temperature and C/N ratio on AD. The results of the analysis showed that the RSM outperformed the ANN in terms of accuracy and prediction error. It is determined that RSM and ANN work together to provide useful models for forecasting particle size limitations in a multiple-input parameter system with a small error rate and without trying any experiments in a short amount of time. Thus, the present study provides a comprehensive assessment of the optimal conditions for the solid-state co-digestion of yard waste and food waste, which contributes towards sustainable biowaste management and resource recovery.</p> Graphical Abstract <p></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

Modelling and Optimisation of Methane Yield From Anaerobic Co-digestion of Food and Yard Waste Using RSM and ANN Approaches

  • 期刊论文

  • 发表日期：

    2026年6月27日

查看全文
  • Satchidananda Mishra,
  • Sagarika Panigrahi,
  • Rahul Das,
  • Sagar Aditya,
  • Anindita Mitra,
  • Krishna Pada Bauri,
  • Abhijeet Das,
  • Arun Kumar Shukla

摘要

The development of biogas production as a sustainable and effective alternative has been fuelled by the growing need for renewable energy sources. Thus, the growing global demand for renewable energy has demanded more research interest in anaerobic digestion (AD). AD is among the simplest ways to boost biogas production productivity. Thus, AD is considered one of the best renewable energy options because it not only manages organic waste effectively but also recovers valuable renewable energy—such as methane—from food waste (FW) and garden or yard waste (YW). Through this process, AD provides a dual benefit: sustainable waste treatment and clean energy production. This results in increased nutrient bioavailability and biogas yield. An important challenge that restricts the acceptance and implementation of AD is the quality and quantity of biogas generated from the process. One practical approach to overcome this challenge is the optimisation of the process parameters. In this study, Methane yield efficiency can be enhanced by optimising the operational environment, utilising response surface methodology (RSM) and artificial neural network (ANN). The findings were compared for anaerobic co-digestion of food waste and yard waste based on statistical parameters in predicting the methane yield. Performance metrics such as R², MSE, and RMSE show that the RSM model outperformed the trained ANN model. The RSM approach demonstrated higher accuracy (R² = 0.983, MSE = 143.68, RMSE = 11.98) in predicting methane yield, whereas the ANN model showed lower predictive performance (R² = 0.938, MSE = 783.05, RMSE = 27.98). The individual and the interaction effect of pH, treatment temperature, and C/N ratio on methane yield were investigated by the RSM model. The obtained optimum condition (pH = 7.56, treatment temperature = 41.96 °C, and C/N = 26.43) showed a methane yield of 535 mL/gVSfed. A comprehensive energy balance analysis was performed for this AD process, and a net positive energy balance of 16.811 kJ/gVS was confirmed. This study revealed that the most effective modelling technique for predicting methane and highlights the importance of pH, treatment temperature and C/N ratio on AD. The results of the analysis showed that the RSM outperformed the ANN in terms of accuracy and prediction error. It is determined that RSM and ANN work together to provide useful models for forecasting particle size limitations in a multiple-input parameter system with a small error rate and without trying any experiments in a short amount of time. Thus, the present study provides a comprehensive assessment of the optimal conditions for the solid-state co-digestion of yard waste and food waste, which contributes towards sustainable biowaste management and resource recovery.

Graphical Abstract