Modeling the pressure strain correlation in turbulent flows using deep neural networks

The pressure strain correlation plays a critical role in the Reynolds stress transport modeling. Accurate modeling of the pressure strain correlation leads to the proper prediction of turbulence stresses and subsequently the other terms of engineering interest. However, classical pressure strain correlation models are often unreliable for complex turbulent flows. Machine learning–based models have shown promise in turbulence modeling, but their application has been largely restricted to eddy viscosity–based models. In this article, we outline a rationale for the preferential application of machine learning and turbulence data to develop models at the level of Reynolds stress modeling. As an illustration, we develop data-driven models for the pressure strain correlation for turbulent channel flow using neural networks. The input features of the neural networks are chosen using physics-based rationale. The networks are trained with the high-resolution DNS data of turbulent channel flow at different friction Reynolds numbers (Reλ). The testing of the models is performed for unknown flow statistics at other Reλ and also for turbulent plane Couette flows. Based on the results presented in this article, the proposed machine learning framework exhibits considerable promise and may be utilized for the development of accurate Reynolds stress models for flow prediction.

Evaluation of the aerodynamic performance of the counter rotating turbo fan COBRA by means of experimental and numerical data

In the present study, steady numerical simulations performed on the counter rotating turbo fan (CRTF) COBRA are compared with experimental data carried at the CIAM C-3A test-bench in Moscow. For this purpose, a systematic analysis of the measurement uncertainties was performed for the global aerodynamic performances of the CRTF, namely, the massflow, the total pressure ratio, the isentropic efficiency, as well as the torque ratio applied on both fan rows. Several numerical models are investigated to highlight their effects on the aforementioned predicted quantities. Differences in modeling consist in grid resolutions and the use of two turbulence models popular in the turbomachinery community. To match as much as possible the experiment running conditions, the performance map of the CRTF is simulated using the exact measured speed ratio and massflow. The comparisons show good estimations of the numerical simulation over the entire performance map. The main differences between the turbulence models occur at part-speed close to stall conditions. More surprisingly at aerodynamic design point, the importance of the turbulence modeling on the predicted torque ratio has been pointed out.

Reynolds' Turbulence Solution

This study revisits the Reynolds-averaged Navier--Stokes equations (RANS) and finds that the existing literature is erroneous regarding the primary unknowns and the number of independent unknowns in the RANS. The literature claims that the Reynolds stress tensor has six independent unknowns, but in fact the six unknowns can be reduced to three that are functions of the three velocity fluctuation components, because the Reynolds stress tensor is simply an integration of a second-order dyadic tensor of flow velocity fluctuations rather than a general symmetric tensor. This difficult situation is resolved by returning to the time of Reynolds in 1895 and revisiting Reynolds' averaging formulation of turbulence. The study of turbulence modeling could focus on the velocity fluctuations instead of on the Reynolds stress. An advantage of modeling the velocity fluctuations is, from both physical and experimental perspectives, that the velocity fluctuation components are observable whereas the Reynolds stress tensor is not.

Study the influence of ship length ratio on ship to ship interaction on moving in parallel on their hydrodynamic characteristics

Суда в некоторых случаях эксплуатации могут двигаться в непосредственной близости друг от друга. Такой сценарий обычно связан с изменением полей давления и скорости вблизи корпуса судов, в результате чего возникают гидродинамические силы и моменты взаимодействия, которые сильно зависит от относительной длины. В этой статье была проведена серия систематических расчётов на двух корпусах KVLCC2, движущихся на большой глубине в безветренную погоду с одинаковой постоянной малой скоростью, не превышающей 4 уз., чтобы исследовать влияние отношения длин на силы и моменты гидродинамического взаимодействия. OpenFOAM, пакет CFD с открытым исходным кодом использовался для организации и проведения расчётов. Метод осреднения по Рейнольдсу уравнений Навье-Стокса (RANS) применялся для моделирования турбулентности. Хорошо известная модель турбулентности использовалась для замыкания уравнений Навье-Стокса. Числовые результаты, касающиеся поля скоростей и гидродинамического следа за судами, были обработаны, проанализированы, сопоставлены и показали хорошее согласование с экспериментальными результатами. Ships, during the lightering operations, are forced to sail in a close position to each other, such a scenario generally associates with a change in the pressure and velocity fields surrounding their hulls, as a result, interaction hydrodynamic forces and moments are generated which are strongly related to the relative length of the interacted ships. In this paper, a series of systematic computations were performed on two KVLCС2 hulls advancing in deep and calm water with the same constant low speed (full scale speed 4kt) in order to investigate the influence of the length ratio on the hydrodynamic interaction forces and moments during the lightering operation. OpenFOAM, an open-source CFD packet was used for carrying out the simulations, Reynolds Averaged Navier-Stokes (RANS) method was used for turbulence modeling and the well-known turbulent model k-ω SST was used to close RANS equations. Numerical results have been post-processed, analyzed, compared and found to be of a good agreement with the experimental results. The velocity fields and wake were presented and analyzed.

Numerical analysis of the equipment position influence on the premises thermal regime under gas infrared emitter operation and mixed convection conditions

The analysis of mathematical modeling results on premise heating by a gas infrared emitter (GIE) during supply and exhaust ventilation operation is presented in this article. The model is presented in a one-dimensional non-stationary setting for an incompressible medium with allowances for the radiant heat flux transfer between opaque solid surfaces in the air. The air is transparent to thermal radiation. The traditional k-ε model is used for turbulence modeling. The possibility for creating comfortable conditions in the area of a horizontal surface with different heights, imitating laboratory equipment, is analyzed.