Abstract
We developed a robust approach for real-time levee condition monitoring based on combination of data-driven methods (one-side classification) and finite element analysis. It was implemented within a flood early warning system and validated on a series of full-scale levee failure experiments organised by the IJkdijk consortium in August-September 2012 in the Netherlands. Our approach has detected anomalies and predicted levee failures several days before the actual collapse. This approach was used in the UrbanFlood decision support system for routine levee quality assessment and for critical situations of a potential levee breach and inundation. In case of emergency, the system generates an alarm, warns dike managers and city authorities, and launches advanced urgent simulations of levee stability and flood dynamics, thus helping to make informed decisions on preventive measures, to evaluate the risks and to alleviate adverse effects of a flood.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
