Experimental validation of a CFD model using a narrow wave flume

Induced swell is characterized in an experimental wave flume that is used to validate the corresponding computational model. The experiments and the numerical simulations are performed in water at several depths (h [m] of 0.2, 0.1, and 0.07), using a piston-type wave maker at set amplitudes (0.015 < A(p) [m] < 0.15), accelerations-decelerations (0.3 < a(p) < 1.0) and average velocities (0.03 < U [m/s] < 0.3) that control the propagation velocity, the period and the wavelength of the waves. The physical effects are modelled with a 213 computational model (STAR-CCM + v11.02) with a mesh of around 630,000 cells of different adaptive sizes, depending on the region under consideration. The physical model is based on a two-phase Eulerian Volume of Fluid unsteady model, accounting for gravity and surface tension, that characterizes turbulence with a k-epsilon model. A user-defined function, based on the period and the amplitude of the vertical paddle in the wave maker, describes the cyclic motion of the linear induction motor. Both the experimental and the computational results are analyzed taking the validity limits of various wave theories as a reference (Le Mehaute). As a result, the experiments are classified within the intermediate water depth regime that corresponds to the second-order Stokes' wave theory. In addition, both the wave propagation velocity and the period are represented as a function of the wavelength and compared with the analytical solutions from the wave theories. The experimental and the computational test campaign yielded results that confirmed the validity of the computational model and that defined the most appropriate conditions for a high-quality CFD simulation.