Moving bed biofilm reactor for blackwater treatment: insights into pollutant removal, microbial communities, and water quality prediction through machine learning
Jiao Xu , Zhulin Lai , Wei Zhang , Tongcai Liu , Shaoze Xiao , Libin Yang , Zhenjiang Yu , Xuefei Zhou
Front. Environ. Sci. Eng. ›› 2025, Vol. 19 ›› Issue (8) : 102
Moving bed biofilm reactor for blackwater treatment: insights into pollutant removal, microbial communities, and water quality prediction through machine learning
Effective treatment of blackwater is critical for sustainable water management and environmental protection. This study investigated the performance of a novel two-stage anoxic-oxic moving bed biofilm reactor (A/O-MBBR) over an operational period of 82 d to enhance the treatment efficiency of blackwater. With an HRT of 25.5 h, the MBBR achieved removal rates of 94.4% for COD, 99.7% for NH3-N, 84.0% for TN, and 74.6% for TP. Even at reduced HRT, the system maintained consistently high removal efficiencies for both COD and TN, highlighting its robust performance under varying operational conditions. This study underscored the superior nitrification activity of attached biofilm compared to the suspended biomass. Predominant microbial genera identified within the biofilm included Thiothrix, Azospira, Acinetobacter, and Thauera genera, which played a critical role in nutrient removal processes. Notably, at low operational temperatures ranging from 8 to 15 °C, facultative anaerobic species contributed significantly to sustaining nitrogen removal efficiencies, hence demonstrating the adaptability of the microbial community to varied environmental conditions. Furthermore, an advanced machine learning model, eXtreme Gradient Boosting (XGBoost), was developed and applied to predict pollutant concentrations across different A/O-MBBR chambers. The model exhibited exceptional predictive accuracy, highlighting the potential of integrating computational intelligence with biological treatment systems to optimize wastewater treatment processes.
Blackwater / MBBR / Nitrification / Microbial analysis / Machine learning
● A novel two-stage A/O-MBBR was established to enhance blackwater treatment. | |
● The excellent pollutant removal rates were achieved when the HRT was 25.5 h. | |
● The attached biofilm was found to play a crucial role in removal of nitrogen. | |
● The system has superior robustness to shortened HRT and low temperature. | |
● Precise prediction by ML model supported the refinement of blackwater treatment. |
| [1] |
|
| [2] |
|
| [3] |
|
| [4] |
|
| [5] |
|
| [6] |
|
| [7] |
|
| [8] |
|
| [9] |
|
| [10] |
|
| [11] |
|
| [12] |
|
| [13] |
|
| [14] |
|
| [15] |
|
| [16] |
|
| [17] |
|
| [18] |
|
| [19] |
|
| [20] |
|
| [21] |
|
| [22] |
SEPA (2002). Water and Wastewater Monitoring and Analysis Method. 4th Edition. Beijing: China Environmental Science Press (in Chinese) |
| [23] |
|
| [24] |
|
| [25] |
|
| [26] |
|
| [27] |
|
| [28] |
|
| [29] |
|
| [30] |
|
| [31] |
|
| [32] |
|
| [33] |
|
| [34] |
|
| [35] |
|
| [36] |
|
| [37] |
|
| [38] |
|
| [39] |
|
| [40] |
|
| [41] |
|
| [42] |
|
Higher Education Press 2025
Supplementary files
/
| 〈 |
|
〉 |