Large Eddy Simulation (LES) is an effective intermediate approach between DNS and RANS, capable of simulating flow features which cannot be handled with RANS such as significant flow unsteadiness and strong vortex-acoustic couplings, and providing higher accuracy than RANS at reasonable cost but still typically an order of magnitude more expensive. Desirable modeling choices involve selecting a...

The actuator disc-RANS model has widely been used in wind and tidal energy to predict the wake of a horizontal axis turbine. The model is appropriate where large-scale effects of the turbine on a flow are of interest, for example, when considering environmental impacts, or arrays of devices. The accuracy of the model for modelling the wake of tidal stream turbines has not been demonstrated, and...

The paper evaluates a new method to prescribe synthesized turbulent inlet boundary conditions. The method can also be used when prescribing turbulent fluctuations at an interface between (U)RANS and LES regions when the flow enters the LES region. When making LES, DES or hybrid LES-RANS a precursor channel DNS is often used. The disadvantage of this method is that it is difficult to re-scale th...

Nowadays there are wide possibilities to perform numerical studies of turbulent natural convection flows on the base of 3D unsteady formulations [1]. Direct Numerical Simulation (DNS) is the most attractive and reliable approach for getting a detailed knowledge on convection [2, 3]. However, DNS applications are practically limited by the case of simplified geometry and/or the Rayleigh numbers ...

Computational modeling is an important tool to understand and stabilize transient turbulent fluid flow in the continuous casting of steel to minimize defects. The current work combines the predictions of two steady Reynolds-averaged Navier–Stokes (RANS) models, a ‘‘filtered’’ unsteady RANS model, and two large eddy simulation (LES) models with ultrasonic Doppler velocimetry (UDV) measurements i...

Deep learning (DL)-based Reynolds stress with its capability to leverage values of large data can be used to close Reynolds-averaged Navier-Stoke (RANS) equations. Type I and Type II machine learning (ML) frameworks are studied to investigate data and flow feature requirements while training DL-based Reynolds stress. The paper presents a method, flow features coverage mapping (FFCM), to quantif...

Numerical shape optimization for aerodynamic problems can lead to improved designs than what is possible by conventional methods. Gradient-based optimization methods require the development of adjoint codes which can be a lengthy exercise and is also not very mature for RANS-based approaches. Furthermore, they may lead to locally optimal solution only, which is not desirable, especially in a pr...

T h e covariance ma t r i x of a rotationally s y m m t r i c shape i s a scalar ident i ty ma t r i x . In this paper, w e apply this property of covariance m a t r i x t o deskew the skewed shape of rotational s y m m e t r y . T h e parameters of the deskew t rans format ion m a t r i x are solved by lett ing the covariance ma t r i x of t he t rans formed shape be equal t o a scalar ident i...

In this work, we present a hybrid solver coupling a high-order LES solver with Cartesian mesh and a RANS solver with body-fitting mesh. This hybrid solver is developed within a dual-mesh consistent hybrid framework, where LES and RANS simulations for the same flow are conducted simultaneously on different computational domains and different meshes. In the LES, an immersed boundary method with r...