Comparison of methods to estimate the scour downstream of a ski jump. (June 2017)

• Record Type: Journal Article
• Authors: Castillo, Luis G.; Carrillo, José M.
• Is Part Of: International journal of multiphase flow. Volume 92(2017)
• Page Start: 171
• Page End: 180
• Publication Date: 2017-06
• Subjects: Flow-3D -- Impingement jets -- Plunge pools -- Scour -- Ski jump -- Turbulence models; Multiphase flow -- Periodicals; Écoulement polyphasique -- Périodiques; Multiphase flow; Periodicals; 620.1064
• DOI: 10.1016/j.ijmultiphaseflow.2017.03.006
• Languages: English
• ISSNs: 0301-9322

Abstract

Highlights: The scour is studied with four complementary procedures: physical model, empirical formulations, erosion potential semi-empirical formulation and CFD simulations. The results are close to the physical model. Each method has its limitations. The results demonstrate the suitability of crossing methodologies to solve complex phenomena. The choice of the turbulence model and of the bed load coefficient are of great importance. Abstract: This study analyzes the expected depth and scour shape in the Toachi River (Ecuador) as a result of the construction of the Toachi Dam in order to avoid problems with progressive scour. The dam has a maximum height of 59 m to the foundations. The free surface weir ends in a ski jump and has been designed to operate with a peak discharge of 1213 m 3 /s. As each method has its limitations, the scour is studied with four complementary procedures: 1) a 1:50 Froude scale similitude laboratory model used as validation case; 2) 36 empirical formulae derived from models and prototypes; 3) a semi-empirical methodology based on pressure fluctuations-erodibility index; and 4) FLOW-3D numerical simulations. The expected scour depth is 6.65 m for the design flow. The results are close to the physical model. In the numerical analysis, turbulence is treated using the three RANS approaches. The choice of the turbulence model and the bed load coefficient in the Meyer-Peter & Müller formula are of great importance. The best results were obtained using the RNG k- ε turbulence model and a bed load coefficient below 7.