Ilya Igorevich Mokhov

Flood early warning system: design, implementation and computational modules

We present a prototype of the flood early warning system (EWS) developed within the UrbanFlood FP7 project. The system monitors sensor networks installed in flood defenses (dikes, dams, embankments, etc.), detects sensor signal abnormalities, calculates dike failure probability, and simulates possible scenarios of dike breaching and flood propagation. All the relevant information and simulation results are fed into an interactive decision support system that helps dike managers and city authorities to make informed decisions in case of emergency and in routine dike quality assessment. In addition to that, a Virtual Dike computational module has been developed for advanced research into dike stability and failure mechanisms, and for training the artificial intelligence module on signal parameters induced by dike instabilities. This paper describes the UrbanFlood EWS generic design and functionality, the computational workflow, the individual modules, their integration via the Common Information Space middleware, and the first results of EWS monitoring and performance benchmarks.

Time-frequency methods for structural health monitoring.

Detection of early warning signals for the imminent failure of large and complex engineered structures is a daunting challenge with many open research questions. In this paper we report on novel ways to perform Structural Health Monitoring (SHM) of flood protection systems (levees, earthen dikes and concrete dams) using sensor data. We present a robust data-driven anomaly detection method that combines time-frequency feature extraction, using wavelet analysis and phase shift, with one-sided classification techniques to identify the onset of failure anomalies in real-time sensor measurements. The methodology has been successfully tested at three operational levees. We detected a dam leakage in the retaining dam (Germany) and “strange” behaviour of sensors installed in a Boston levee (UK) and a Rhine levee (Germany).

Artificial intelligence and finite element modelling for monitoring flood defence structures

We present a hybrid approach to monitoring the stability of flood defence structures equipped with sensors. This approach combines the finite element modelling with the artificial intelligence for real-time signal processing and anomaly detection. This combined method has been developed for the UrbanFlood early warning system and successfully tested on a large-scale sea dike during a simulated strong storm with very high water level. The artificial intelligence module detects the onset of dike instability after being trained on the data from the Virtual Dike finite element simulation.

An Approach for Real-time Levee Health Monitoring Using Signal Processing Methods

We developed a levee health monitoring system within the UrbanFlood project funded under the EU 7th Framework Programme. A novel real-time levee health assessment Artificial Intelligence system is developed using data-driven methods. The system is implemented in the UrbanFlood early warning system. We present the application of dedicated signal processing methods for detection of leakage through the water retaining dam and subsequent analysis of the measurements collected from one of the UrbanFlood pilot levees at the Rhine river in Germany.

Combining Data-Driven Methods with Finite Element Analysis for Flood Early Warning Systems

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.