Flow and sediment transport in a sharp river bend using a 3D-RANS model

Flow dynamics and sediment transport in a river bend have recently been studied using experimental and numerical investigations. A three-dimensional numerical modeling model named NaysCUBE was used in this study to describe the flow pattern and process of sediment transport in a sharp river bend as a complement to the prior work of the physical hydraulic model. The model uses the RANS equation to simulate flow where a fully complex 3D flow is governed. Despite the limitations of the RANS model, NaysCUBE well reproduces the flow pattern and turbulence phenomena in a movable bed channel with sharp curvature. Compared with data from a prior experiment, the morphological adjustment is simulated sufficiently. The three-dimensional flow structures are useful for determining the appropriate countermeasures for local scouring and riverbank protection.

Aerodynamic Characteristics of a Single Airfoil for Vertical Axis Wind Turbine Blades and Performance Prediction of Wind Turbines

The design of wind turbines requires a deep insight into their complex aerodynamics, such as dynamic stall of a single airfoil and flow vortices. The calculation of the aerodynamic forces on the wind turbine blade at different angles of attack (AOAs) is a fundamental task in the design of the blades. The accurate and efficient calculation of aerodynamic forces (lift and drag) and the prediction of stall of an airfoil are challenging tasks. Computational fluid dynamics (CFD) is able to provide a better understanding of complex flows induced by the rotation of wind turbine blades. A numerical simulation is carried out to determine the aerodynamic characteristics of a single airfoil in a wide range of conditions. Reynolds-averaged Navier–Stokes (RANS) equations and large-eddy simulation (LES) results of flow over a single NACA0012 airfoil are presented in a wide range of AOAs from low lift through stall. Due to the symmetrical nature of airfoils, and also to reduce computational cost, the RANS simulation is performed in the 2D domain. However, the 3D domain is used for the LES calculations with periodical boundary conditions in the spanwise direction. The results obtained are verified and validated against experimental and computational data from previous works. The comparisons of LES and RANS results demonstrate that the RANS model considerably overpredicts the lift and drag of the airfoil at post-stall AOAs because the RANS model is not able to reproduce vorticity diffusion and the formation of the vortex. LES calculations offer good agreement with the experimental measurements.

Numerical investigation of unsteady cavitating turbulent flows around a three-dimensional hydrofoil using stress-blended eddy simulation

The mechanism of flow instability, which involves complex gas–liquid interactions and multiscale vortical structures, is one of the hot research areas in cavitating flow. The role of turbulence modeling is crucial in the numerical investigation of unsteady flow characteristics. Although large-eddy simulation (LES) has been used as a reliable numerical method, it is computationally costly. In this work, we used a hybrid Reynolds-averaged Navier–Stokes (RANS) and LES model, that is, stress-blended eddy simulation (SBES), to improve the prediction capability for the cloud cavitating flow. Our hybrid approach introduces a shielding function to integrate the RANS model with the LES applied only regionally, such as to large-scale separated flow regions. The results showed that the periodic shedding of cavity growth, break off, and collapse around a three-dimensional Clark-Y hydrofoil was reproduced in accordance with experimental observations. The lift/drag coefficients, streamwise velocity profiles, and cavity patterns obtained by the SBES model were in better agreement with the experimental data than those obtained by the modified RANS model. The re-entrant jet dynamics responsible for the break off of the attached cavity were discussed. Further analysis of vorticity transportation indicated that the stretching and dilatation terms dominated the development of vorticity around the hydrofoil. In conclusion, the SBES model can be used to predict cavitating turbulent flows in practical engineering applications.

Improving the methodology for calculation of energy losses in axial turbine cascades

Представлена оценка неопределенности прогнозирования потерь энергии в решетках профилей осевых турбин. В сравнении с экспериментальными данными рассмотрены эмпирическая модель ЦИАМ и метод CFD анализа в рамках RANS модели. Геометрические и режимные параметры решеток профилей варьируются в широком диапазоне. Результаты CFD расчета отличаются существенно в зависимости от модели турбулентности. Наименьшая неопределенность получена для модели рейнольдсовых напряжений RSM. Определено выборочное стандартное относительное отклонение для анализируемой базы данных. Применительно к CFD расчету данное отклонение составило 18,6%, применительно к эмпирической модели ЦИАМ 46,4%. Разработана эмпирическая модель коррекции потерь полученных по результатам CFD анализа с моделью турбулентности RSM. Корректирующая функция включает в себя геометрические и режимные параметры решеток и особенности течения в межлопаточном канале (всего 14 параметров). Использование разработанного подхода позволило снизить неопределённость прогнозирования потерь в 2 раза. В результате работы выборочное стандартное относительное отклонение предсказания потерь для рассматриваемой базы решеток профилей составило 9,3%. Estimation of the uncertainty in predicting profile losses using various models was performed. In comparison with the experimental data, empirical model of CIAM and method of CFD analysis are considered. RANS models are used. The geometric and operating parameters of the analyzed turbine cascades vary over a wide range. Turbulence models strongly influence loss prediction uncertainty. The smallest uncertainty was obtained using the RSM turbulence model. The sample standard deviation for the considered turbine cascades base was determined. The deviation for CFD analysis is 18.6%. For the empirical model of CIAM the deviation is 46.4%. The new empirical model has been created to correct the results of calculating losses according to the RANS model using the RSM turbulence model. The corrective function takes into account the influence of the geometric and operating parameters of the turbine cascades and the features of the airfoil flow (14 parameters in total). The developed approach allows reducing the uncertainty in the estimation of losses according to the RANS model by 2 times. As a result, the sample standard deviation in the prediction of losses is 9.3% for the considered turbine cascades base.

Run-up of breaking focused waves on a beach studied experimentally in a large scale facility and numerically using hybrid FNPT-RANS model

<p>The feasibility of generating long period waves using a piston-type wavemaker was comprehensively demonstrated in [1] for the Großer Wellenkanal (GWK) at 1:100 scale. These included regular waves, elongated solitons, N-waves as well as time-series pertaining to earthquake tsunamis (namely the 2004 Indian Ocean and 2011 Tohoku tsunamis).</p><p>In the companion paper [2], the aforementioned long-period waves were simulated using fully nonlinear potential theory (FNPT) and the Korteweg-de Vries (KdV) equations and compared with GWK measurements. It was established that the FNPT and KdV models accurately predicted long-distance evolution of these waves as well as dispersion-induced splitting of elongated solitons and N-waves to the form of undular bores. In addition, the run-up characteristics of the “2009 Samoa tsunami” record and elongated solitons were also studied in [2]. The semi-numerical procedure adopted in [2] for run-up estimation was limited to long non-breaking waves. However, the experimental data collection included also violently breaking focused waves of a smaller period.</p><p>In the present work, we apply IITM-RANS3D to simulate the run-up characteristics of breaking focused waves based on experiments carried out in the GWK using a 1:6 slope. An in-house, Reynolds-Averaged Navier-Stokes model (IITM-RANS3D) has been recently developed and hybridized with the in-house potential code IITM-FNPT2D. The RANS framework allows for complete description of breaking wave hydrodynamics and FNPT-hybridization ensures energy preservation for long-duration wave simulations. The run-up problem is considered as a multiphase flow where the beach is physically modelled as a high-viscosity fluid. The simulations would provide valuable insight into the run-up characteristics of breaking bores at large scale as viscous and aeration effects are fully accounted for in the RANS model.</p><p><strong>REFERENCES</strong></p><p>[1] S. Schimmels, V. Sriram, I. Didenkulova, Tsunami generation in a large scale experimental facility, Coastal Engineering 110, 32-41 (2016).</p><p>[2] V. Sriram, I. Didenkulova, A. Sergeeva, S. Schimmels, Tsunami evolution and run-up in a large scale experimental facility, Coastal Engineering 111, 1-12 (2016). </p>