Application of computational fluid dynamics to predict hydrodynamic downpull on high head gates

Abstract

Purpose High head gates are commonly used in hydropower plants for flow regulation and emergence closure. Hydrodynamic downpull can be a critical parameter in design of the lifting mechanism. The purpose of this paper is to show that a simplified two-dimensional (2D) computational fluid dynamics solution can be used in the prediction of the downpull force on the gate lip by comparison of computed results to experimentally measured data. Design/methodology/approach In this study, ANSYS FLUENT CFD software was used to obtain 2D numerical solution for the flow field around a generic gate model located in a power intake structure which was previously used in an experimental study. Description of the flow domain, computational grid resolution, requirements on setting appropriate boundary conditions and methodology in describing downpull coefficient are discussed. Total number of 245 simulations for variable gate lip geometry and gate openings were run. The downpull coefficient evaluated from the computed pressure field as function of gate opening and lip angle are compared with the experimental results. Findings The computed downpull coefficient agrees well with the previous experimental results, except one gate with small lip angle where a separation bubble forms along the lip, which is responsible from this deviation. It is observed that three-dimensional (3D) effects are confined to the large gate openings where downpull is minimum or even reversed. Research limitations/implications In large gate openings, three dimensionality of the flow around gate slots plays an important role and departure from 2D solutions become more pronounced. In that case, one might need to perform a 3D solution instead. Practical implications This paper presents a very fast and accurate way to predict downpull force on high head gates in the absence of experimental data. Originality/value An extensive amount of simulations are run within the scope of this study. It is shown that knowing its limitations, 2D numerical models can be used to calculate downpull for a wide range of gate openings without the need of expensive experimental models.