Modelling the fate of organic carbon and micropollutants in Biological Active Granular Activated Carbon Filters

由于有机微污染物对环境和人类健康的潜在威胁，其在接收水体中的存在引起了人们的极大关注。众所周知，污水处理厂的污水是有机污染物的一个主要来源；作为回应，目前正在讨论法律框架，并正在研究各种减少管理资源项目的技术。颗粒活性炭（GAC）过滤器已经成为一种从污水处理厂污水中去除OMP的合适技术。除了吸附去除外，gac过滤器还具有生物去除有机物和omp的能力。控制omp吸附和生物还原的现象以及这两种机制之间的协同作用非常重要。然而，对这些过程的调查是非常复杂的。一方面，污水处理厂出水是多组分混合物，难以表征；另一方面，GAC、生物膜、omp和有机物之间发生的各种相互作用，在实验上测量或跟踪是相当复杂的。数学模型是克服这些实验障碍、分析各种场景并最终支持进一步实验设计的强大工具。利用实验室和中试规模的数据，建立了第一个能够令人满意地描述生物活性gac过滤器中溶解有机碳去除的数学模型。本项目旨在改进和扩展该模型，为模型的进一步应用提供新的关键特性。具体来说，将检验三个假设：(i)是否有可能将孔径分布引入模型？孔径分布是表征不同gac类型的关键参数，因此在模型中实现孔径分布至关重要。然而，传统方法需要的参数难以确定。（ii）包含n循环的微生物群落能否提高模型的解释力？有实验证据表明，OMPs的生物降解与硝化物的活性有关，该项目旨在实施共同代谢过程，并评估其对全球模型结果的影响。（iii）如何将个别的管理废物纳入模型，以及如何令人满意地再现它们的行为？对单个omp去除的预测是高度相关的，因此，四个omp将典型地实施到模型中，并用作其他感兴趣的omp去除的代理。由于omp的机械描述容易变得复杂，因此建模方法将机械建模与机器学习方法相结合作为支持。

The presence of Organic Micropollutants (OMPs) in receiving waterbodies is of great concern, due to their potential threat to the environment and human health. Wastewater Treatment Plants (WWTP) effluents are known to be one major source of OMPs; as a response, legal frameworks are currently under discussion and various technologies for the reduction of OMPs are investigated. Granular Activated Carbon (GAC) filters have established themselves as a suitable technology for OMP removal from WWTP effluents. Alongside adsorptive removal GAC-filters have the ability to also biologically remove organic matter and OMPs. The phenomena governing adsorptive and biological reduction of OMPs as well as the synergies between these two mechanisms are of great importance. The investigation of these processes is, however, highly complex. On the one hand, WWPT effluents are multi-component mixtures and difficult to characterize; on the other hand, the various interactions that take place between the GAC, the biofilm, the OMPs and the organic matter, is rather complicated to measure or track experimentally. Mathematical models are powerful tools to overcome such experimental barriers, to analyze various scenarios and eventually to support the design of further experiments. Using lab and pilot-scale data, a first mathematical model capable of satisfactorily describing the dissolved organic carbon removal in a biological active GAC-filter has been developed. This project aims to improve and extend this model with new key features that are required for further application of the model. In particular, three hypotheses will be tested: (i) Is it possible to introduce the pore size distribution to the model? Pore size distribution is key parameter for the characterization of different GAC-types, thus its implementation into the model is vital. Nevertheless the conventional approaches require parameters that are difficult to determine. (ii) Could a microbial community including the N-cycle improve the explanatory power of the model? With experimental evidence that relates biodegradation of OMPs to the activity of nitrifiers, the project aims to implement co-metabolic processes and to assess their effect on the global modelling results. (iii) How can individual OMPs be included into the model and their behavior be satisfactory reproduced? The prediction of the removal of individual OMPs is of high relevance and therefore, four OMPs will exemplarily be implemented into the model and used as proxy for the removal of other OMPs of interest. Since the mechanistic description of the OMPs may easily get complicated, the modeling approach will combine mechanistical modeling with machine learning methods as support.