给水管网生物膜是含氮消毒副产物（DBPs）的重要反应前体物。生物膜群落结构和活性状态的演替变化，改变了胞外聚合物（EPS）的产量与特性，进而影响了DBPs的生成。然而目前，给水管网生物膜变化特征对含氮DBPs生成的影响机制以及不同特性EPS与含氮DBPs生成的关联规律仍不明晰。本项目拟通过实地采样、室内实验和模型模拟等手段，阐明群落构成与活性特征对EPS产量、分子组成及结构的作用规律；利用高分辨质谱和光谱联合技术，辨识EPS中的典型含氮DBPs前体物，探究其生成DBPs的种类、路径及动力学；进而基于人工智能技术，构建生物膜群落/活性特征-EPS特性-含氮DBPs生成的深度网络，识别影响含氮DBPs生成的主控特征因子，揭示EPS与含氮DBPs生成的关联规律，实现对DBPs浓度的时空刻画。研究成果将有助于深入理解生物膜特征与DBPs生成的内在机制，为保障饮用水安全提供科学依据和技术支持。

The biofilm in drinking water distribution system provides various nitrogenous disinfection by-products (N-DBPs) formation precursors. The biofilm shows the succession of changes in community structure and activity, which determines the production and characteristics of extracellular polymeric substances (EPS) and would further affect the formation of highly toxic nitrogenous DBPs. However, the mechanism induced by the changes in the biofilm characteristics on the N-DBPs production, as well as the relationship between EPS and N-DBPs formation are remains unclear. This study intends to clarify the effects of the relative abundance and activity of dominant bacteria (species) on EPS yields, molecular composition and structure by means of field sampling, laboratory experiments and model simulation. The combination of high-resolution mass spectrometry and spectroscopy is used to identify and analyze the typical N-DBPs precursors in EPS, and explore the types, pathway and kinetics of DBPs formation. Then, the network structure between biofilm community/activity, EPS characteristics and N-DBPs formation is constructed by applying artificial intelligence technology, identifying the key biofilm characteristic factors, characterizing the links between EPS and N-DBPs formation, predicting the spatial-temporal DBPs concentrations. The results of this proposed project will reveal the internal relationship between biofilm and DBPs formation, and provide theoretical basis and technical support for ensuring the safety of drinking water.