AI-Driven Optimization of Bio-Energy Systems: Models for Resource Assessment and Emission Reduction

Résumé fourni par la source

The increasing complexity of bio-energy systems is a reason for the need of advanced analytical methods to enhance resource utilization, process stability and environmental performance. AI methods are popular in this domain, but many papers neglect concerns around data quality, interpretability, scalability and the low generalization potential of models toward plants and feedstocks different from those they were trained on. This paper intends to offer a systematic review on AI techniques available for biomass resource assessment, conversion-process optimization, and supply-chain planning and emission management. The review is structured adopting a rigorous review approach focusing on model, data set, optimization framework and hybrid method developed in the scope of bio-energy value chain. Highlights – The key findings are that AI improves the prediction of biomass availability and biogas/syngas yields, of feedstock properties and emission behavior; surrogate and hybrid models result in expedited simulation time and facilitate real-time decision making. The review also highlights an emerging trend with digital twins, remote sensing with application of machine learning, and federated learning in multi-plant optimization. These findings also have important implications for researchers, engineers and policy-makers who aim to develop robust low emissions bio-energy systems that are economically feasible.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.