Experimental and numerical evaluation of the force due to the impact of a dam-break wave on a structure

Flood events caused by the collapse of dams or river levees can have damaging consequences on buildings and infrastructure located in prone areas. Accordingly, a careful prediction of the hydrodynamic load acting on structures is important for flood hazard assessment and potential damage evaluation. However, this represents a challenging task and requires the use of suitable mathematical models. This paper investigates the capability of three different models, i.e. a 2D depth-averaged model, a 3D Eulerian two-phase model, and a 3D Smoothed Particle Hydrodynamics (SPH) model, to estimate the impact load exerted by a dam-break wave on an obstacle. To this purpose, idealised dam-break experiments were carried out by generating a flip-through impact against a rigid squat structure, and measurements of the impact force were obtained directly by using a load cell. The dynamics of the impact event was analyzed and related to the measured load time history. A repeatability analysis was performed due to the great variability typically shown by impact phenomena, and a confidence range was estimated. The comparison between numerical results and experimental data shows the capability of 3D models to reproduce the key features of the flip-through impact. The 2D modelling based on the shallow water approach is not entirely suitable to accurately reproduce the load hydrograph and predict the load peak values; this difficulty increases with the strength of the wave impact. Nevertheless, the error in the peak load estimation is in the order of 10% only, thus the 2D approach may be considered appropriate for practical applications. Moreover, when the shallow water approximation is expected to work well, 2D results are comparable with the experimental data, as well as with the numerical predictions of far more sophisticated and computationally demanding 3D solvers. All the numerical models overestimate the falling limb of the load hydrograph after the impact. The SPH model ensures good evaluation of the long-time load impulse. The 2D shallow water solver and the 3D Eulerian model are less accurate in predicting the load impulse but provide similar results. A sensitivity analysis with respect to the model parameters allows to assess model uncertainty. Finally, the experimental data collected have been made available online as supplementary material for validation purposes. (C) 2014 Elsevier Ltd. All rights reserved.