Machine Learning-Assisted Optimization of Mixed Carbon Source Compositions for High-Performance Denitrification

• DOI: 10.1021/acs.est.4c01743
• Publication Date: 2024-06-20T14:32:49Z
• Authors: Yuan Pan; Tian-Wei Hua; Rui-Zhe Sun; Ying-Ying Fu; Zhi-Chao Xiao; Jin Wang; Han-Qing Yu
• Keywords: Carbon source; Carbon fibers

Appropriate mixed carbon sources have great potential to enhance denitrification efficiency and reduce operational costs in municipal wastewater treatment plants (WWTPs). However, traditional methods struggle efficiently select the optimal mixture due variety of compositions. Herein, we developed a machine learning-assisted high-throughput method enabling WWTPs rapidly identify optimize sources. Taking local WWTP as an example, source data set was established via employed train learning model. The composition types inoculated sludge served input variables. XGBoost algorithm predict total nitrogen removal rate microbial growth, thereby aiding assessment potential. predicted exhibited enhanced over single both kinetic experiments long-term reactor operations. Model feature analysis shows that cumulative effect interaction among individual significantly overall Metagenomic reveals increased diversity complexity denitrifying bacterial ecological networks WWTPs. This work offers efficient for compositions provides new insights into mechanism behind under supply multiple