Author(s): Arnau Bayon Barrachina

Linked Author(s): Arnau Bayón

Keywords: CFD; RANS; Hydraulic jump; VOF; K-epsilon; K-omega;

Abstract: Hydraulic structures often require to dissipate energy so that the flow can be smoothly restituted to the natural streambed. This is one of the key aspects of spillway and stilling basin design. In most practical cases, a hydraulic jump is chosen to deal with this energy surplus. Unfortunately, hydraulic jumps are highly turbulent, involving three-dimensional structures and showing extremely complex aeration patterns. Hence, resorting to computational fluid dynamics (CFD) and/or experimental modeling becomes a must. Regardless the modeling strategy used, a fully satisfactory in-depth analysis of the flow behavior in hydraulic jumps completely free from scale effects has not been achieved so far. At least, for the usual range of Froude numbers in hydraulic structure design. The research presented herein provides a detailed analysis of the hydraulic jump behavior, comparing the results obtained with different well known RANS turbulence models. The main goal is to assess the effect of turbulence model choice on the outcome accuracy. To this end, a 3D model is set up using OpenFOAM and three different unsteady RANS turbulence models, namely Standard k-ε, RNG k-ε and K-ω SST. A VOF approach coupled with an interface compression algorithm is used to deal with flow aeration phenomena. A structured mesh of cubic cells is employed and a convergence analysis to ensure grid independence of results is conducted.
Several relevant variables describing the hydraulic jump behavior are analyzed, including sequent depth ratios, water free surface profiles and hydraulic jump roller lengths. It is shown that for the same scenario and flow conditions, the obtained results present certain differences depending on the RANS model. Experimental data, along with previously published results from other authors, are used for model validation.

DOI: https://doi.org/10.3850/38WC092019-0918

Year: 2019