scholarly journals Characterization of Model-Based Uncertainties in Incompressible Turbulent Flows by Machine Learning

2018 ◽  
Author(s):  
Mustafa Usta ◽  
Ali Tosyali
Keyword(s):  
Machine Learning ◽  
Turbulent Flows ◽  
Three Dimensional ◽  
Turbulence Models ◽  
Machine Learning Algorithms ◽  
Mean Flow ◽  
Flow Properties ◽  
Significant Discrepancy ◽  
Model Based ◽  
Flow Features

This work determines the inaccuracy of using Reynolds averaged Navier Stokes (RANS) turbulence models in transition to turbulent flow regimes by predicting the model-based discrepancies between RANS and large eddy simulation (LES) models. Then, it incorporates the capabilities of machine learning algorithms to characterize the discrepancies which are defined as a function of mean flow properties of RANS simulations. First, three-dimensional CFD simulations using k-omega Shear Stress Transport (SST) and dynamic one-equation subgrid-scale models are conducted in a wall-bounded channel containing a cylinder for RANS and LES, respectively, to identify the turbulent kinetic energy discrepancy. Second, several flow features such as viscosity ratio, wall-distance based Reynolds number, and vortex stretching are calculated from the mean flow properties of RANS. Then the discrepancy is regressed on these flow features using the Random Forests regression algorithm. Finally, the discrepancy of the test flow is predicted using the trained algorithm. The results reveal that a significant discrepancy exists between RANS and LES simulations, and ML algorithm successfully predicts the increased model uncertainties caused by the employment of k-omega SST turbulence model for transitional fluid flows.

Turbulence Measurements Downstream of a Turbine Cascade at Different Incidence Angles and Pitch-Chord Ratios

1993 ◽  
Author(s):  
Vincenzo Dossena ◽  
Antonio Perdichizzi ◽  
Marina Ubaldi ◽  
Pietro Zunino
Keyword(s):  
Turbulent Flow ◽  
Reynolds Stress ◽  
Incidence Angle ◽  
Three Dimensional ◽  
Turbulence Models ◽  
Turbine Cascade ◽  
Mean Flow ◽  
Flow Measurements ◽  
Flow Features ◽  
Turbulent Flow Structure

An experimental investigation on a linear turbine cascade has been carried out to study the effects induced by incidence angle and pitch-chord ratio variations on the three-dimensional turbulent flow downstream of the cascade. Previous mean flow measurements have shown how these parameters influence the energy losses and the secondary velocity field. Now detailed hot wire measurements have been performed on a plane located at 22 per cent of an axial chord downstream of the trailing edge, in order to determine the distribution of all the six Reynolds stress tensor components, for three incidence conditions (i = −30, 0, +30 deg) and for three pitch-chord ratios (s/c = 0.58, 0.72, 0.87). Significant changes of the turbulent flow structure, interesting magnitude and distribution of the Reynolds stress components, have been observed for all the considered test conditions. The analysis of the results shows the correlation between the mean flow features and the turbulent quantities and the relationship between the energy loss production and the blade loading variation. The presented data are also suitable for assessing the behaviour of turbulence models in complex 3D flows, on design and off-design conditions.

Direct numerical simulation of separated flow in a three-dimensional diffuser

2010 ◽  
Vol 650 ◽  
pp. 307-318 ◽  
Author(s):  
JOHAN OHLSSON ◽  
PHILIPP SCHLATTER ◽  
PAUL F. FISCHER ◽  
DAN S. HENNINGSON
Keyword(s):  
Numerical Simulation ◽  
Direct Numerical Simulation ◽  
Turbulent Flows ◽  
Three Dimensional ◽  
Turbulence Models ◽  
Separated Flow ◽  
Spectral Element ◽  
Mean Flow ◽  
Development Section ◽  
Inflow Condition

A direct numerical simulation (DNS) of turbulent flow in a three-dimensional diffuser at Re = 10000 (based on bulk velocity and inflow-duct height) was performed with a massively parallel high-order spectral element method running on up to 32768 processors. Accurate inflow condition is ensured through unsteady trip forcing and a long development section. Mean flow results are in good agreement with experimental data by Cherry et al. (Intl J. Heat Fluid Flow, vol. 29, 2008, pp. 803–811), in particular the separated region starting from one corner and gradually spreading to the top expanding diffuser wall. It is found that the corner vortices induced by the secondary flow in the duct persist into the diffuser, where they give rise to a dominant low-speed streak, due to a similar mechanism as the ‘lift-up effect’ in transitional shear flows, thus governing the separation behaviour. Well-resolved simulations of complex turbulent flows are thus possible even at realistic Reynolds numbers, providing accurate and detailed information about the flow physics. The available Reynolds stress budgets provide valuable references for future development of turbulence models.

An explicit algebraic Reynolds stress model for incompressible and compressible turbulent flows

2000 ◽  
Vol 403 ◽  
pp. 89-132 ◽  
Author(s):  
STEFAN WALLIN ◽  
ARNE V. JOHANSSON
Keyword(s):  
Turbulent Flows ◽  
Reynolds Stress ◽  
Compressible Flows ◽  
Three Dimensional ◽  
Turbulence Models ◽  
Reynolds Stress Model ◽  
Reynolds Stresses ◽  
Mean Flow ◽  
New Developments ◽  
The Individual

Some new developments of explicit algebraic Reynolds stress turbulence models (EARSM) are presented. The new developments include a new near-wall treatment ensuring realizability for the individual stress components, a formulation for compressible flows, and a suggestion for a possible approximation of diffusion terms in the anisotropy transport equation. Recent developments in this area are assessed and collected into a model for both incompressible and compressible three-dimensional wall-bounded turbulent flows. This model represents a solution of the implicit ARSM equations, where the production to dissipation ratio is obtained as a solution to a nonlinear algebraic relation. Three-dimensionality is fully accounted for in the mean flow description of the stress anisotropy. The resulting EARSM has been found to be well suited to integration to the wall and all individual Reynolds stresses can be well predicted by introducing wall damping functions derived from the van Driest damping function. The platform for the model consists of the transport equations for the kinetic energy and an auxiliary quantity. The proposed model can be used with any such platform, and examples are shown for two different choices of the auxiliary quantity.

Marketing customer response scoring model based on machine learning data analysis

2020 ◽  
pp. 1-11
Author(s):  
Tang Yan ◽  
Li Pengfei
Keyword(s):  
Machine Learning ◽  
Data Analysis ◽  
Data Extraction ◽  
Machine Learning Algorithms ◽  
Customer Relationship ◽  
Customer Data ◽  
Modeling And Analysis ◽  
Scoring Model ◽  
Model Based ◽  
Learning Data

In marketing, problems such as the increase in customer data, the increase in the difficulty of data extraction and access, the lack of reliability and accuracy of data analysis, the slow efficiency of data processing, and the inability to effectively transform massive amounts of data into valuable information have become increasingly prominent. In order to study the effect of customer response, based on machine learning algorithms, this paper constructs a marketing customer response scoring model based on machine learning data analysis. In the context of supplier customer relationship management, this article analyzes the supplier’s precision marketing status and existing problems and uses its own development and management characteristics to improve marketing strategies. Moreover, this article uses a combination of database and statistical modeling and analysis to try to establish a customer response scoring model suitable for supplier precision marketing. In addition, this article conducts research and analysis with examples. From the research results, it can be seen that the performance of the model constructed in this article is good.

scholarly journals Assessing biases, relaxing moralism: On ground-truthing practices in machine learning design and application

2021 ◽  
Vol 8 (1) ◽  
pp. 205395172110135
Author(s):  
Florian Jaton
Keyword(s):  
Machine Learning ◽  
William James ◽  
A Priori ◽  
Learning Algorithms ◽  
Three Dimensional ◽  
Ground Truth ◽  
Machine Learning Algorithms ◽  
Ground Truthing ◽  
Set Up ◽  
The Moment

This theoretical paper considers the morality of machine learning algorithms and systems in the light of the biases that ground their correctness. It begins by presenting biases not as a priori negative entities but as contingent external referents—often gathered in benchmarked repositories called ground-truth datasets—that define what needs to be learned and allow for performance measures. I then argue that ground-truth datasets and their concomitant practices—that fundamentally involve establishing biases to enable learning procedures—can be described by their respective morality, here defined as the more or less accounted experience of hesitation when faced with what pragmatist philosopher William James called “genuine options”—that is, choices to be made in the heat of the moment that engage different possible futures. I then stress three constitutive dimensions of this pragmatist morality, as far as ground-truthing practices are concerned: (I) the definition of the problem to be solved (problematization), (II) the identification of the data to be collected and set up (databasing), and (III) the qualification of the targets to be learned (labeling). I finally suggest that this three-dimensional conceptual space can be used to map machine learning algorithmic projects in terms of the morality of their respective and constitutive ground-truthing practices. Such techno-moral graphs may, in turn, serve as equipment for greater governance of machine learning algorithms and systems.

Characteristics of turbulence in a face-centred cubic porous unit cell

2019 ◽  
Vol 873 ◽  
pp. 608-645 ◽  
Author(s):  
Xiaoliang He ◽  
Sourabh V. Apte ◽  
Justin R. Finn ◽  
Brian D. Wood
Keyword(s):  
Pore Space ◽  
Three Dimensional ◽  
Jet Impingement ◽  
Unit Cell ◽  
Mean Flow ◽  
Flow Topology ◽  
Turbulence Structures ◽  
Integral Scales ◽  
Flow Features ◽  
Acceleration And Deceleration

Direct numerical simulations (DNS) are performed in a triply periodic unit cell of a face-centred cubic (FCC) lattice covering the unsteady inertial, to fully turbulent, flow regimes. The DNS data are used to quantify the flow topology, integral scales, turbulent kinetic energy (TKE) transport and anisotropy distribution in the tortuous geometry. Several unique flow features are observed within this low porosity configuration, where the mean flow undergoes strong acceleration and deceleration regions with presence of three-dimensional helical motions, weak wake-like structures behind spheres, stagnation and jet-impingement-like flows together with merging and spreading jets in the main pore space. The jet-impingement and weak wake-like flow structures give rise to regions with negative total TKE production. Unlike flows in complex shaped ducts, the turbulence intensity levels in the cross-stream directions are found to be larger than those in the streamwise direction. Furthermore, due to the compact nature and confined geometry of the FCC packing, the turbulent integral length scales are estimated to be less than 10 % of the bead diameter even for the lowest Reynolds number studied, indicating the absence of macroscale turbulence structures for this configuration. This finding suggests that even for the highly anisotropic flow within the pore, the upscaled flow statistics are captured well by the representative volumes defined by the unit cell.

Improved Prediction of Turbomachinery Flows Using Near-Wall Reynolds-Stress Model

2001 ◽  
Vol 124 (1) ◽  
pp. 86-99 ◽  
Author(s):  
G. A. Gerolymos ◽  
J. Neubauer ◽  
V. C. Sharma ◽  
I. Vallet
Keyword(s):  
Experimental Data ◽  
Turbulent Flows ◽  
Reynolds Stress ◽  
Turbulence Models ◽  
Reynolds Stress Model ◽  
Stress Model ◽  
Reynolds Stresses ◽  
Mean Flow ◽  
Flow Equations ◽  
Near Wall

In this paper an assessment of the improvement in the prediction of complex turbomachinery flows using a new near-wall Reynolds-stress model is attempted. The turbulence closure used is a near-wall low-turbulence-Reynolds-number Reynolds-stress model, that is independent of the distance-from-the-wall and of the normal-to-the-wall direction. The model takes into account the Coriolis redistribution effect on the Reynolds-stresses. The five mean flow equations and the seven turbulence model equations are solved using an implicit coupled OΔx3 upwind-biased solver. Results are compared with experimental data for three turbomachinery configurations: the NTUA high subsonic annular cascade, the NASA_37 rotor, and the RWTH 1 1/2 stage turbine. A detailed analysis of the flowfield is given. It is seen that the new model that takes into account the Reynolds-stress anisotropy substantially improves the agreement with experimental data, particularily for flows with large separation, while being only 30 percent more expensive than the k−ε model (thanks to an efficient implicit implementation). It is believed that further work on advanced turbulence models will substantially enhance the predictive capability of complex turbulent flows in turbomachinery.

Experimental Study of Flow Inside a Centrifugal Fan Using Magnetic Resonance Velocimetry

2019 ◽  
Author(s):  
Davis W. Hoffman ◽  
Laura Villafañe ◽  
Christopher J. Elkins ◽  
John K. Eaton
Keyword(s):  
Three Dimensional ◽  
Leading Edge ◽  
Suction Side ◽  
Reynolds Numbers ◽  
Centrifugal Fan ◽  
Mean Flow ◽  
Magnetic Resonance Velocimetry ◽  
Flow Features ◽  
Outlet Flow ◽  
The Mean

Abstract Three-dimensional, three-component time-averaged velocity fields have been measured within a low-speed centrifugal fan with forward curved blades. The model investigated is representative of fans commonly used in automotive HVAC applications. The flow was analyzed at two Reynolds numbers for the same ratio of blade rotational speed to outlet flow velocity. The flow patterns inside the volute were found to have weak sensitivity to Reynolds number. A pair of counter-rotating vortices evolve circumferentially within the volute with positive and negative helicity in the upper and lower regions, respectively. Measurements have been further extended to capture phase-resolved flow features by synchronizing the data acquisition with the blade passing frequency. The mean flow field through each blade passage is presented including the jet-wake structure extending from the blade and the separation zone on the suction side of the blade leading edge.

Numerical Investigations of the Gas Flow Inside the Bassoon

2012 ◽  
Vol 134 (4) ◽  
Author(s):  
Andreas Richter
Keyword(s):  
Gas Flow ◽  
Three Dimensional ◽  
Musical Instrument ◽  
Mean Flow ◽  
Steady State Flow ◽  
Plane Wave Solution ◽  
Pressure Time ◽  
Flow Features ◽  
Order Of Magnitude ◽  
Three Dimensional Simulations

This work is devoted to the numerical investigation of the gas flow inside a bassoon while it is played. The digitized geometry for the simulations is taken from measurements using laser scan techniques in combination with image processing. Pressure time series measured at the bell and reed were used to define adequate boundaries. Additional pressure measurements inside the musical instrument helped to validate the calculations. With this approach, it was possible to model the characteristics of a bassoon which plays the lowest note. The results of the three-dimensional simulations showed that the acoustic velocities and the underlying mean flow exhibit the same order of magnitude. The calculations indicate that the flow in curved sections such as the crook and the 180 deg bend is considerably different from a steady-state flow. For example, in bends the time-averaged flow features chains of small, alternating vortex pairs, and the pressure distribution differs significantly from a plane wave solution.

Sign in / Sign up

Export Citation Format

Share Document