Lehrinhalte
The digital transition of urban water networks towards more data-driven and intelligent systems represents a primary opportunity to tackle the challenges posed by increasing population, urbanisation, and changing climate conditions. As the data-driven transformation reaches into the economy and society, ever-increasing amounts of data are generated by machines or processes based on emerging technologies, such as the Internet of Things (IoT), connected systems, and advanced modelling. While digital disruption has already transformed a number of other industries globally, the water sector has only recently embraced the digital transformation. This is the key to developing suitable adaption strategies that, relying on better information than in the past, support management and decision-making actions to plan adaptation strategies that enhance the resilience of urban water systems under uncertain future climate and social scenarios.
The phenomenon of digitalization of urban water system will be analysed, starting from basic notions of modelling water distribution networks, and then adding more focus on Data Science and Artificial Intelligence approaches to modelling and controlling such networks. The course will be structured around the main topic of modelling and management of water distribution networks. In addition, other sub-topics will be touched during the course, enabling the students to get an overview of the different elements of modern urban water systems, acquire knowledge about best technologies, learn how to identify anomalies (e.g., leakages) in the normal operation of these systems, and get insights on the role and influence of human behaviours in such systems.
The following CORE TOPICS will be covered:
1. Mathematical modelling and hydraulic modelling of water distribution networks (WDN)
2. Model calibration, validation, and performance assessment
3. Water demand modelling
4. Simulation and optimisation of WDN operations.
In tackling the above topics, notions on the following DS and AI techniques will be introduced:
1. Time series analysis
2. Clustering techniques
3. Control theory and optimisation
4. Artificial neural networks.
Additionally, one module of this course - the Journal Club - will be focused on developing skills for reading and analysing scientific literature.
During the project activity, the students will be actively fostered to develop own solutions for a sample problems introduced during the guided practical lab activities. These activities will rely on open available datasets.
Assessment includes final quizzes, a final presentation, and a short project report. The lecture will be given in English and will include lectures by international guest speakers.
