Applied deep learning to predict urban drainage system status based on radar data at DHI

This project focuses on developing a new method to predict status of an urban drainage system (water level and flow). In particular, emphasis will be on predicting whether the inflow to a waste water treatment plant is below or above a critical threshold. If the candidate has a particular interest in explainability of deep learning models, emphasis can be put on this aspect, possibly including an interactive dashboard to allow exploration of the basis of model predictions and extracted features, instead of or in addition to optimizing classification performance.

Skills: Deep learning experience and Python programming are central to succeed in this project.

Attention: Often you need a pre-approval from your university or study counsellor, to be sure that projects or thesis found on Future Finder can be accepted as part of your education. Please contact the right entity in due time to ensure that you're picking the right project.