Machine learning modeling and analysis of biohydrogen production from wastewater by dark fermentation process

• Published in: Bioresource technology Vol. 343; p. 126111
• Main Authors: Hosseinzadeh, Ahmad, Zhou, John L., Altaee, Ali, Li, Donghao
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.01.2022
• Subjects: acetates; Biohydrogen; butyrates; Dark fermentation; ethanol; fermentation; hydrogen production; Machine learning; Process modelling; support vector machines; wastewater; Wastewater treatment; Process modelling; Biohydrogen; Dark fermentation; Wastewater treatment; Machine learning
• ISSN: 0960-8524; 1873-2976; 1873-2976
• DOI: 10.1016/j.biortech.2021.126111

[Display omitted] •Machine learning models developed for H2 production by dark fermentation.•GBM, SVR, RF and AdaBoost methods were robust models for the process.•All four models demonstrated close agreement with results (R2 ≥ 90%).•PVI was able to assess the relative importance of the inputs.•Acetate, butyrate, ethanol, Fe and Ni showed high importance in decreasing order. Dark fermentation process for simultaneous wastewater treatment and H2 production is gaining attention. This study aimed to use machine learning (ML) procedures to model and analyze H2 production from wastewater during dark fermentation. Different ML procedures were assessed based on the mean squared error (MSE) and determination coefficient (R2) to select the most robust models for modeling the process. The research showed that gradient boosting machine (GBM), support vector machine (SVM), random forest (RF) and AdaBoost were the most appropriate models, which were optimized by grid search and deeply analyzed by permutation variable importance (PVI) to identify the relative importance of process variables. All four models demonstrated promising performances in predicting H2 production with high R2 values (0.893, 0.885, 0.902 and 0.889) and small MSE values (0.015, 0.015, 0.016 and 0.015). Moreover, RF-PVI demonstrated that acetate, butyrate, acetate/butyrate, ethanol, Fe and Ni were of high importance in decreasing order.