scholarly journals Revising of the Near Ground Helicopter Hover: The Effect of Ground Boundary Layer Development

2021 ◽  
Vol 11 (21) ◽  
pp. 9935
Author(s):  
Theologos Andronikos ◽  
George Papadakis ◽  
Vasilis Riziotis ◽  
Spyros Voutsinas
Keyword(s):  
Boundary Layer ◽  
Flow Field ◽  
Hybrid Model ◽  
Computational Cost ◽  
Thrust Coefficient ◽  
Rotor Wake ◽  
Strongly Coupled ◽  
Entire Flow ◽  
And Performance ◽  
Cfd Method

The interaction of a helicopter rotor with the ground in hover flight is addressed numerically using a hybrid Eulerian–Lagrangian CFD model. When a helicopter takes off or lands, its wake interferes with the ground. This interaction, depending on the height-to-rotor diameter ratio, causes the altering of the rotor loading and performance as compared to the unconstrained case and gives rise to the development of a complex outwash flow field in the surrounding of the helicopter. The present study aims to characterize the interactional phenomena occurring in the early stages of the rotor wake development and in particular the interference of the starting vortex with the ground boundary layer and the effect of this interaction in the motion of the vortex in the rotor outwash flow. The hybrid CFD method employed combines a standard URANS compressible finite volume solver, the use of which is restricted to confined grids around solid bodies, and a Lagrangian approximation of the entire flow field in which conservation equations are solved in their material form, disctretized using particle representation of the flow quantities. The two methods are strongly coupled to each other through an appropriate iterative scheme. The main advantage of the proposed methodology is that it can conveniently handle complex configurations with several bodies that move independently from one another, with affordable computational cost. In this paper, thrust coefficient predictions of the hybrid model are compared to predictions of a free wake code and to experimental data indicating that consistent prediction of the rotor load requires the inclusion of the ground boundary layer in the analysis. Moreover, detailed comparisons of the rotor wake evolution predicted by the hybrid model are presented.

Subsonic Turbulent Flow Past a Planar Fence

1979 ◽  
Vol 101 (3) ◽  
pp. 373-375
Author(s):  
M. L. Agarwal ◽  
P. K. Pande ◽  
Rajendra Prakash
Keyword(s):  
Boundary Layer ◽  
Turbulent Boundary Layer ◽  
Flow Field ◽  
Boundary Conditions ◽  
Mean Flow ◽  
Governing Equations ◽  
Flow Past ◽  
Entire Flow ◽  
The Mean ◽  
Shape And Size

The mean flow past a fence submerged in a turbulent boundary layer is numerically simulated. The governing equations have been simplified by neglecting the convective effects of turbulence and solved numerically using experimental boundary conditions. The information obtained includes the shape and size of the upstream and downstream separation bubbles and the streamline pattern in the entire flow field. General agreement between the simulated and the experimental flow field was found.

scholarly journals Influence of coupled boundary layer suction and bowed blade on flow field and performance of a diffusion cascade

2017 ◽  
Vol 30 (1) ◽  
pp. 249-263 ◽  
Author(s):  
Zhiyuan Cao ◽  
Bo Liu ◽  
Ting Zhang ◽  
Xiqiong Yang ◽  
Pingping Chen
Keyword(s):  
Boundary Layer ◽  
Flow Field ◽  
Boundary Layer Suction ◽  
And Performance

A Hybrid RANS-LES Modeling Methodology Sensitized to Transitional and Curvature/Rotation Effects

2015 ◽  
Author(s):  
Varun Chitta ◽  
D. Keith Walters
Keyword(s):  
Boundary Layer ◽  
Hybrid Model ◽  
Computational Cost ◽  
Navier Stokes ◽  
Test Cases ◽  
New Model ◽  
Eddy Viscosity Model ◽  
Modeling Methodology ◽  
Flow Configurations ◽  
Layer Flow

This paper presents a new hybrid model that seeks to combine the strengths of Reynolds–averaged Navier-Stokes (RANS) and large eddy simulation (LES) methods. The new model is based on a recently proposed version of a dynamic hybrid RANS-LES (DHRL) framework that addresses several deficiencies inherent in most current hybrid models, including explicit grid dependence, boundary layer model stress depletion, and delayed shear layer breakdown. The DHRL framework is highly generalized, allowing coupling of any desired LES model with any given RANS model. In this study, a recently proposed four-equation eddy-viscosity model (EVM) capable of predicting both flow transition from laminar-to-turbulent (T) and rotation and/or streamline curvature (RC) effects is used for the RANS component, and a monotonically-integrated LES (MILES) scheme is used for the LES component. The new model (DHRL with T-RC effects) is implemented into a commercial computational fluid dynamics (CFD) code and investigated against three different flow configurations. The test cases include nonrotating and rotating channel flow, zero-pressure-gradient (ZPG) boundary layer flow over a flat plate, and flow over a circular cylinder. Results obtained from the numerical simulations indicate that the new hybrid model produces significant levels of turbulent fluctuations in the flowfield and successfully resolves T-RC effects. The results show improved accuracy compared to RANS models and are obtained at a significant reduction of computational cost compared to full LES models. Further investigation on complex test cases is warranted before commenting on the accuracy and potential usage of the model as a practical tool.

scholarly journals Emotion Recognition on Edge Devices: Training and Deployment

Sensors ◽  
2021 ◽  
Vol 21 (13) ◽  
pp. 4496
Author(s):  
Vlad Pandelea ◽  
Edoardo Ragusa ◽  
Tommaso Apicella ◽  
Paolo Gastaldo ◽  
Erik Cambria
Keyword(s):  
Emotion Recognition ◽  
Language Processing ◽  
Computational Cost ◽  
Sequential Learning ◽  
High Quality ◽  
Fast Training ◽  
Online Sequential Learning ◽  
And Performance ◽  
Resource Constrained Devices ◽  
Constrained Devices

Emotion recognition, among other natural language processing tasks, has greatly benefited from the use of large transformer models. Deploying these models on resource-constrained devices, however, is a major challenge due to their computational cost. In this paper, we show that the combination of large transformers, as high-quality feature extractors, and simple hardware-friendly classifiers based on linear separators can achieve competitive performance while allowing real-time inference and fast training. Various solutions including batch and Online Sequential Learning are analyzed. Additionally, our experiments show that latency and performance can be further improved via dimensionality reduction and pre-training, respectively. The resulting system is implemented on two types of edge device, namely an edge accelerator and two smartphones.

scholarly journals Prediction of 3D Cardiovascular hemodynamics before and after coronary artery bypass surgery via deep learning

2021 ◽  
Vol 4 (1) ◽  
Author(s):  
Gaoyang Li ◽  
Haoran Wang ◽  
Mingzi Zhang ◽  
Simon Tupin ◽  
Aike Qiao ◽  
...  
Keyword(s):  
Deep Learning ◽  
Artery Bypass ◽  
Computational Cost ◽  
Specific Model ◽  
Arterial System ◽  
Patient Specific ◽  
Cardiovascular Hemodynamics ◽  
Before And After ◽  
Deep Learning Network ◽  
Cfd Method

AbstractThe clinical treatment planning of coronary heart disease requires hemodynamic parameters to provide proper guidance. Computational fluid dynamics (CFD) is gradually used in the simulation of cardiovascular hemodynamics. However, for the patient-specific model, the complex operation and high computational cost of CFD hinder its clinical application. To deal with these problems, we develop cardiovascular hemodynamic point datasets and a dual sampling channel deep learning network, which can analyze and reproduce the relationship between the cardiovascular geometry and internal hemodynamics. The statistical analysis shows that the hemodynamic prediction results of deep learning are in agreement with the conventional CFD method, but the calculation time is reduced 600-fold. In terms of over 2 million nodes, prediction accuracy of around 90%, computational efficiency to predict cardiovascular hemodynamics within 1 second, and universality for evaluating complex arterial system, our deep learning method can meet the needs of most situations.

Sign in / Sign up

Export Citation Format

Share Document