期刊
BIORESOURCE TECHNOLOGY
卷 343, 期 -, 页码 -出版社
ELSEVIER SCI LTD
DOI: 10.1016/j.biortech.2021.126111
关键词
Biohydrogen; Dark fermentation; Machine learning; Process modelling; Wastewater treatment
资金
- University of Technology Sydney
- UTS President's Scholarship
- 111 Project [D18012]
This study used machine learning (ML) procedures to model and analyze H-2 production from wastewater during dark fermentation, and found that gradient boosting machine (GBM), support vector machine (SVM), random forest (RF), and AdaBoost were the most appropriate models. By optimizing the models with grid search and analyzing them deeply with permutation variable importance (PVI), the research identified the relative importance of process variables.
Dark fermentation process for simultaneous wastewater treatment and H-2 production is gaining attention. This study aimed to use machine learning (ML) procedures to model and analyze H-2 production from wastewater during dark fermentation. Different ML procedures were assessed based on the mean squared error (MSE) and determination coefficient (R-2) 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 prom-ising performances in predicting H-2 production with high R-2 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.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据