Water Systems Identification and Control
The thesis is concerned with the application of the state-of-art techniques for System Identification to the study case of River Basins and/or in Water Transportation Systems.
The common way scientists model Water Systems is via deterministic white-box models. This methodology gave good results and powerful commercial software is currently available, e.g. Mike11 by DHI for River Basin Monitoring and Forecast (http://www.dhisoftware.com/Products/WaterResources/MIKE11.aspx).
Recent literature shows however that introduction of stochastic models and of less physics-based models might prove more accurate and more robust.
The ideal candidate has a strong background in Control Theory, Functional Analysis and Identification Theory. S/He will learn some of the most advanced techniques for Nonparametric Identification. S/He will develop methods for River Basins Identification, Water Network Leakage recognition and Water Network Control. In case of River Basins Identification, real data from the Po River Basin, North Italy, are available.