Hostname: page-component-7b7567c8c-fvjsw Total loading time: 0 Render date: 2026-11-16T06:23:48.891Z Has data issue: false hasContentIssue false

Turbulent scalar flux in inclined jets in crossflow: counter gradient transport and deep learning modelling

Published online by Cambridge University Press:  16 November 2020

Pedro M. Milani*
Affiliation:
Mechanical Engineering Department, Stanford University, Stanford, CA 94305, USA
Julia Ling
Affiliation:
Citrine Informatics, Redwood City, CA 94061, USA
John K. Eaton
Affiliation:
Mechanical Engineering Department, Stanford University, Stanford, CA 94305, USA
*
†Email address for correspondence: pmmilani@stanford.edu

Abstract

A cylindrical and inclined jet in crossflow is studied under two distinct velocity ratios, $r=1$ and $r=2$, using highly resolved large eddy simulations. First, an investigation of turbulent scalar mixing sheds light onto the previously observed but unexplained phenomenon of negative turbulent diffusivity. We identify two distinct types of counter gradient transport, prevalent in different regions: the first, throughout the windward shear layer, is caused by cross-gradient transport; the second, close to the wall right after injection, is caused by non-local effects. Then, we propose a deep learning approach for modelling the turbulent scalar flux by adapting the tensor basis neural network previously developed to model Reynolds stresses (Ling et al., J. Fluid Mech., vol. 807, 2016a, pp. 155–166). This approach uses a deep neural network with embedded coordinate frame invariance to predict a tensorial turbulent diffusivity that is not explicitly available in the high-fidelity data used for training. After ensuring analytically that the matrix diffusivity leads to a stable solution for the advection diffusion equation, we apply this approach in the inclined jets in crossflow under study. The results show significant improvement compared to a simple model, particularly where cross-gradient effects play an important role in turbulent mixing. The model proposed herein is not limited to jets in crossflow; it can be used in any turbulent flow where the Reynolds averaged transport of a scalar is considered.

Information

Type
JFM Papers
Copyright
© The Author(s), 2020. Published by Cambridge University Press
Figure 0

Figure 1.$(a)$ Schematic of the inclined jet in crossflow. The coordinate frame origin is located at the centre of the hole as it meets the bottom wall. $(b)$ Three-dimensional view of the complete simulation domain, with plenum feed inlet plane shown in red, outlet plane in blue and walls in grey. Figures adapted from Milani et al. (2019a).

Figure 1

Figure 2.Scalar concentration results $c$ on the centreplane ($z=0$, a,c) and in an axial plane located at $x/D=2$ (b,d). Colour contours show an instantaneous snapshot of $c$ and lines indicate isocontours of mean concentration, $\bar {c}=0.2, 0.4, 0.6, 0.8$.

Figure 2

Figure 3.Vertical profiles at $x/D=5$ and $z/D=0$ from the $r=1$ (a,b) and $r=2$ (c,d) LES. (a,c) Streamwise turbulent scalar flux and mean concentration gradient. (b,d) Mean streamwise velocity and concentration. The grey area shows a region of implied negative diffusivity in the streamwise direction. $(a)$$r=1$. $(b)$$r=1$. $(c)$$r=2$. $(d)$$r=2$.

Figure 3

Figure 4.Vertical profiles at $x/D=5$ and $z/D=0$ of $\overline {u'c'}$ from the LES. Different averages are performed: unconditional or conditioned on high/low values of $|v'|$ at the centre of the grey band. $(a)$$r=1$. $(b)$$r=2$.

Figure 4

Figure 5.Vertical profiles of different components of the $\overline {u'c'}$ production term (marked as I in (3.1)). The profiles are shown at $x/D=5$, $z/D=0$ and the grey band represents the area of negative diffusivity. Terms are non-dimensionalized by $U_c$ and $D$. $(a)$$r=1$. $(b)$$r=2$.

Figure 5

Figure 6.Vertical profiles at the centreplane of the $y$-component of scalar gradient and turbulent scalar flux. The grey band indicates regions where both have the same sign; for $r=2$ and $x/D=5$, no consistent region of counter gradient transport is observed (thus, no grey band), but $\overline {v'c'}$ is very near zero close to the wall. $(a)$$r=1$. $(b)$$r=2$.

Figure 6

Figure 7.Terms from (3.1) in the channel centreline, non-dimensionalized using $D$ and $U_c$. For $r=1$, the zoomed in version shows $0 < y/D < 0.25$. For $r=2$, it shows $0 < y/D < 0.3$. The grey band indicates counter gradient transport in the vertical direction. $(a)$$r=1$, $x/D=2$. $(b)$$r=1$, $x/D=5$. $(c)$$r=2$, $x/D=2$. $(d)$$r=2$, $x/D=5$.

Figure 7

Figure 8.Counter gradient transport in the $r=1$ case. (a,c,e) Centreplanes ($z=0$) and (b,d,f) axial planes at $x/D=2$. Lines indicate isocontours of $\bar {c}=0.2, 0.4, 0.6, 0.8$. (a,b) Contains colour contours of $\theta$ in degrees. (c,d) Shows, in red, places where $\overline {u'c'} ({\partial \bar {c}}/{\partial x}) > 0$ (with arrows highlighting type 1 transport). (e,f) Identifies regions where $\overline {v'c'} ({\partial \bar {c}}/{\partial y}) > 0$ (with arrows highlighting type 2 transport).

Figure 8

Figure 9.Counter gradient transport in the $r=2$ case. Same description as figure 8.

Figure 9

Figure 10.Schematic of TBNN-s architecture. Blue cells represent inputs to the network, green cells represent the hidden layers containing tuneable parameters and red cells denote different levels of outputs.

Figure 10

Figure 11.Angle between modelled and true turbulent scalar flux in degrees, given in $\theta ^{TBNN\text {-}s}$. (a,c) Centreplane ($z=0$) and (b,d) axial planes at $x/D=2$; black lines indicate isocontours of $\bar {c}=0.2, 0.4, 0.6, 0.8$.

Figure 11

Figure 12.Dimensionless components of the diffusivity matrix $\boldsymbol{\mathsf{D}}$ predicted by the TBNN-s in $r=1$ and $r=2$ cases. (a,c,e,g) Centreplane ($z=0$) and (b,d,f,h) axial planes at $x/D=2$; black lines indicate isocontours of $\bar {c}=0.2, 0.4, 0.6, 0.8$.

Figure 12

Figure 13.Mean scalar concentration contours in the $r=1$ case, with (a,c,e) showing centreplane ($z=0$) and (b,d,f) showing axial ($y$–$z$) planes at $x/D=2$. Isocontours at $\bar {c}=0.2, 0.4, 0.6, 0.8$ are also shown. (a,b) Contains LES results, (c,d) shows results from baseline GDH model, and (e,f) presents results from the TBNN-s.

Figure 13

Figure 14.Mean scalar concentration contours in the $r=2$ case, with same legend as figure 13.

Figure 14

Figure 15.Vertical profiles of $\bar {c}$ at the centre spanwise position ($z=0$) calculated in three distinct streamwise locations. $(a)$$r=1$. $(b)$$r=2$.

Figure 15

Figure 16.Horizontal profiles of $\bar {c}$ at $x/D=5$ plotted for two distinct vertical stations. $(a)$$r=1$. $(b)$$r=2$.

Figure 16

Figure 17.Colour contours of the basis coefficients $g^{(i)}$ predicted by the TBNN-s model in the $r=2$ jet. (a,c,e,g) Centreplane ($z=0$) and (b,d,f,h) axial planes at $x/D=2$; black lines indicate isocontours of $\bar {c}=0.2, 0.4, 0.6, 0.8$.

Figure 17

Figure 18.Colour contours of the basis coefficients $g^{(i)}$ predicted by the TBNN-s model in the $r=1$ jet. (a,c,e,g) Centreplane ($z=0$) and (b,d,f,h) axial planes at $x/D=2$; black lines indicate isocontours of $\bar {c}=0.2, 0.4, 0.6, 0.8$.

Figure 18

Figure 19.Magnitude of turbulent scalar flux components relative to the equivalent mean advection component in the $r=1$ jet. Black lines showing isocontours at $\bar {c}=0.2, 0.4, 0.6, 0.8$ are also shown. (a,b) The streamwise component $| \overline {u'c'} / (\bar {u} \bar {c}) |$ and (c,d) the wall-normal component $| \overline {v'c'} / (\bar {v} \bar {c}) |$.

Figure 19

Figure 20.Vertical profiles of the turbulent scalar flux in the $r=1$ jet at the centreplane ($z/D=0$) and two different streamwise locations. Lines compare LES results to the baseline model (GDH with $Sc_t=0.85$) and to the TBNN-s model trained on the $r=2$ jet (as discussed in § 4.3). $(a)$ The streamwise component $\overline {u'c'}$ of the flux, and $(b)$ the wall-normal component $\overline {v'c'}$.