Andrea Del Gaudio

Dam break waves cause huge damages, human fatalities, high economic losses, and ecological and environmental issues in the surrounded landscape. Dam breaks could be modelled as downstream waves that are generated by a sudden release of water stored behind a wall. The resulting wave is defined as surge, if it flows on a dry bed, or a bore, if it propagates on a wet bed. Tsunamis, tidal bores, landslides, and the collapse of an artificial embarkment can generate similarly unsteady flows and drastic consequences. It is necessary to take into account that in real conditions the flow of a dam break does not consist only of water and often contains some impurities and particles, or in the case of avalanche a different physical state. Since these differences the fluid under consideration will have a different movement pattern and can be described as a non-Newtonian flow. Only a few studies showed in-field measurement of dam break flows, and often the available data are post-forensic engineering reconstructions of the event. Real field manifestations are indeed rare, and it is very difficult to collect data and visual observations during these occurrences. Moreover, some regions of the world have not yet experienced tsunamis or dam break waves (landslides, avalanche, etc.) in recent times, and thus there are no field observations on which policymakers can base and develop safety procedures to protect residences (Lynett, n.d.). A step beyond is to characterize the impact of these waves against structures in order to develop adequate countermeasures to protect the areas at risk. Scaled laboratory experiments of dam break waves have a crucial role in understanding hydrodynamic evolution and in validating numerical models. Chanson observed in 2009 that the numerical models implemented were validated by a limited data set. In particular, only few studies report an extensive data set (water depth, pressure and velocity field evolution in time) of the main features of the flow. On the other hand numerical models are paramount to test real case simulations and predicting future scenarios, as the characterization of the areas at risk. Moreover they are an useful tool to design structural countermeasures to protect the areas at risk. A fully 3D numerical model, that is extensive validated by experimental results, and can predict Newtonian and Non-Newtonian dam break flows is the main aim of the research project.

The main purpose of this research is to analyse the implementation of a fully 3-D numerical model that solves the Navier-Stokes equations in order to predict dam break waves impacting against structures, and to characterize the hydrodynamic and the forces extent by the waves on obstacles. The numerical model will be validated by extensive experimental campaigns that will be carried out at Hydraulic laboratory of the University of Naples Federico Secondo. The experimental set-up have been designed ad hoc to understand the main hydrodynamic features of dam break flows as water depth, velocity front celerity and pressures extent on the obstacles.

First of all, a more in-depth analysis of the hydro-dynamic of Newtonian and Non-Newtonian dam break waves against obstacles is required. Moreover, this examination may verify the influence of the wave dynamic on the pressure distribution on the obstacle in order to quantify the reliability and the error that eventually is committed by shallow water depth integrated model in predicting these kind of phenomena. The fully 3-D numerical model will include a turbulence closure model Large Eddies simulation. Once the numerical model will be extensively validated in both cases with a Newtonian fluid (water) and a non-Newtonian fluid (CMC).

Shallow water numerical models can predict force time series for newtonian dam break waves as well as non newtonian rheology. The wall shear stress is a very interesting parameter to monitoring. The fully 3D numerical simulations with LES model show an increas in the wall shear stresses for non newtonian rheology compered with water as well as a slower front propagation in time.