Machine learning in fermentative biohydrogen production: Advantages, challenges, and applications

Hydrogen can be produced in biological processes that are environmentally friendly using various organic waste and biomass as feedstock. However, the complexity of the biological process limits the predictability and reliability, which hinders the scale-up and dissemination. This article reviewed current edge research and perspectives on the application of machine learning in biohydrogen technology. Several machine learning algorithems have been implemented recently for modeling the nonlinear and complex relationships among operational and performance parameters in biohydrogen production as well as predicting the process performance and microbial population dynamics. The reinforced machine learning method exhibited precise state prediction and retrieved the underlying kinetics effectively. The machine-learning based prediction was improved by using microbial sequencing data as input parameters. Further research on machine learning would be promising to derive a process control tool to maintain reliable hydrogen production performance and to connect the missing nodes in the process performance and microbial population.