Abstract
© The Author(s), under exclusive licence to Springer Nature B.V. 2025.Anaerobic digestion (AD) is widely considered as a potential process for renewable energy production from organic waste materials. However, the efficiency of biogas production via AD is limited by slow microbial activity, process instability, and variability in feedstock composition. Recent advancements include incorporating biochar, which physically improves environmental conditions (e.g., enhancing microbial colonization, reducing inhibitors). Despite these benefits, exploring optimal operational conditions remains challenging due to heterogeneous experimental goals. To address this, Machine Learning (ML), a critical subset of Artificial Intelligence (AI), distinctly contributes to AD by enabling accurate prediction of optimal operational parameters, real-time process monitoring, and adaptive management of process complexities, which biochar alone cannot provide. The integration of ML-driven predictive capabilities with both in-silico (computational) and in-situ (experimental) platforms significantly enhances traditional biotechnological methods. This review explicitly outlines the contributions of ML, rooted in AI, in optimizing biochar-enhanced AD operations, particularly through forecasting optimal conditions for enhanced bioenergy recovery. Furthermore, it highlights ML’s role in exploiting cost-performance improvements beyond the physical benefits of biochar. Finally, current knowledge gaps related specifically to ML-assisted optimization of biochar applications in AD are discussed.
NETL
NSTL
Springer Nature