Urban water systems
The research in water distribution has focused on improving the efficiency of systems through discolouration risk management, and leak identification and reduction. The work has been supported by developing tailored mathematical optimization tools for pressure management, and associated design problems. Similarly, by devising new mixed-integer optimisation and graph theoretic algorithms, approaches were presented for determining optimal configurations for self-cleaning water distribution systems and optimal placement of pressure reducing valves and isolation valves. Simulation and practical testing of leak localization with different demand models and monitoring systems (including remote sensing) led to novel insights for model selection and the design of leak localization campaigns.
Our research into urban hydrology has shown that hydrological response in urban systems is sensitive to small-scale rainfall variability, in space and especially in time, indicating that the resolution of current operational rainfall radar products is insufficient for urban flood prediction. Analysis of flood observations datasets based on citizens’ reports has shown that proximity to drainage outlet and variability in impervious surface cover, play a significant role in explaining flood occurrence, while the role of variability in rainfall intensity could not be confirmed at the available data resolution. We innovatively analysed insurance claims of flood damage and found that failure of in-house storm water drainage facilities, caused the largest number of damage claims. Claims associated with flooding from urban drainage systems, led to significantly higher damage amounts than failure of in-house drainage systems.