The modeling and tracing of chemical reactions in Water Distribution Networks (WDNs) is essential for maintaining drinking water quality. The aim of potential BA/MA theses or projects is to investigate or develop data-driven methods (e.g. Machine learning methods such as Deep Learning) for modeling and tracing chemicals such as chlorine (reactions) in WDNs. Depending on the student’s interest and the required number of credit points, different objectives are possible such as:

• Detecting chemical events such as contamination
• Forecasting and tracing of chemical concentrations For those objectives, the focus could be on the creation of new benchmark scenarios, implementation and evaluation of existing methods from the literature on (simulated) benchmark scenarios, or development and evaluation of new methods (e.g. neural network architectures). Furthermore, the investigation (empirically or formally) of other aspects such as robustness is also possible.

It is also possible to work on this as part of a paid student/research assistant job.

Keywords: Deep Learning, Critical Infrastructure, Water Distribution Networks

Literature

1. https://www.sciencedirect.com/science/article/abs/pii/S1367578823000159