Simulating and optimizing the integrating process of anaerobic digestion and aerobic composting using machine learning

Hao Peng, Lan Mu, Yingjin Song, Hong Su, Yadong Ge, Zhanjun Cheng, Rundong Li, Guanyi Chen,
Simulating and optimizing the integrating process of anaerobic digestion and aerobic composting using machine learning,
Sustainable Energy Technologies and Assessments,
Volume 83,
2025,
104671,
ISSN 2213-1388,
https://doi.org/10.1016/j.seta.2025.104671.
(https://www.sciencedirect.com/science/article/pii/S2213138825005028)
Abstract: This study employed machine learning models to simulate and optimize the integrated anaerobic digestion (AD) and aerobic composting (AC) process for treating rural organic solid waste. The M−ADM1 and Random Forest models were combined to provide a comprehensive prediction of the system dynamics, and the carbon offset was used as a key evaluation indicator. Both the models demonstrated high predictive accuracy. The R2 and RMSE of the AD model reached 0.967 and 0.037, respectively. The R2 and RMSE of the AC model reached 0.955 and 0.0037, respectively. Major carbon offsets were achieved by substituting methane for natural gas and converting solid digestate into compost to replace chemical fertilizers. The total carbon offset increased markedly between 6–16 days, reaching a peak of 39.73 g CO2-eq on Day 16 prior to AD stabilization. These results underscored the significant potential of the integrated AD-AC process in reducing carbon emissions and provided valuable insights for operational optimization.
Keywords: Machine learning; Rural organic solid wastes; Anaerobic digestion; Aerobic composting; Carbon emission