Anaerobic digestion integrated with microbial electrolysis cell to enhance biogas production and upgrading in situ.

Anaerobic digestion (AD) is an effective and applicable technology for treating organic wastes to recover bioenergy, but it is limited by various drawbacks, such as long start-up time for establishing a stable process, the toxicity of accumulated volatile fatty acids and ammonia nitrogen to methanogens resulting in extremely low biogas productivities, and a large amount of impurities in biogas for upgrading thereafter with high cost. Microbial electrolysis cell (MEC) is a device developed for electrosynthesis from organic wastes by electroactive microorganisms, but MEC alone is not practical for production at large scales. When AD is integrated with MEC, not only can biogas production be enhanced substantially, but also upgrading of the biogas product performed in situ. In this critical review, the state-of-the-art progress in developing AD-MEC systems is commented, and fundamentals underlying methanogenesis and bioelectrochemical reactions, technological innovations with electrode materials and configurations, designs and applications of AD-MEC systems, and strategies for their enhancement, such as driving the MEC device by electricity that is generated by burning the biogas to improve their energy efficiencies, are specifically addressed. Moreover, perspectives and challenges for the scale up of AD-MEC systems are highlighted for in-depth studies in the future to further improve their performance.

Developing high-dimensional machine learning models to improve generalization ability and overcome data insufficiency for mixed sugar fermentation simulation.

Biorefinery can be promoted by building accurate machine learning models. This work proposed a strategy to enhance model's generalization ability and overcome insufficient data conditions for mixed sugar fermentation simulation. Multiple inputs single output models, using initial glucose, initial xylose, and time together as inputs, have higher generalization ability than single input single output models with time as sole input in predicting glucose, xylose, ethanol, or biomass separately. Multiple inputs multiple outputs models, integrating outputs, enhanced model accuracy and resulted in an average R2 at 0.99. To overcome data insufficiency conditions, consensus yeast (CY) model, through consolidating data from 4 yeasts, obtained R2 at 0.90. By adjusting the pretrained CY model, the model can save more than 50% data and get R2 at 0.95 and 0.93 for yeast and bacterial fermentation simulation. The strategy can expand the application range and save costs of data curation for ANN models.