xref: /libCEED/examples/fluids/index.md (revision a62415e8b4a45f4517abd31b18d1fb676e45053b)
1(example-petsc-navier-stokes)=
2
3# Compressible Navier-Stokes mini-app
4
5This example is located in the subdirectory {file}`examples/fluids`.
6It solves the time-dependent Navier-Stokes equations of compressible gas dynamics in a static Eulerian three-dimensional frame using unstructured high-order finite/spectral element spatial discretizations and explicit or implicit high-order time-stepping (available in PETSc).
7Moreover, the Navier-Stokes example has been developed using PETSc, so that the pointwise physics (defined at quadrature points) is separated from the parallelization and meshing concerns.
8
9## Running the mini-app
10
11```{include} README.md
12:start-after: inclusion-fluids-marker
13```
14## The Navier-Stokes equations
15
16The mathematical formulation (from {cite}`shakib1991femcfd`) is given in what follows.
17The compressible Navier-Stokes equations in conservative form are
18
19$$
20\begin{aligned}
21\frac{\partial \rho}{\partial t} + \nabla \cdot \bm{U} &= 0 \\
22\frac{\partial \bm{U}}{\partial t} + \nabla \cdot \left( \frac{\bm{U} \otimes \bm{U}}{\rho} + P \bm{I}_3 -\bm\sigma \right) - \rho \bm{b}  &= 0 \\
23\frac{\partial E}{\partial t} + \nabla \cdot \left( \frac{(E + P)\bm{U}}{\rho} -\bm{u} \cdot \bm{\sigma} - k \nabla T \right) - \rho \bm{b} \cdot \bm{u} &= 0 \, , \\
24\end{aligned}
25$$ (eq-ns)
26
27where $\bm{\sigma} = \mu(\nabla \bm{u} + (\nabla \bm{u})^T + \lambda (\nabla \cdot \bm{u})\bm{I}_3)$ is the Cauchy (symmetric) stress tensor, with $\mu$ the dynamic viscosity coefficient, and $\lambda = - 2/3$ the Stokes hypothesis constant.
28In equations {eq}`eq-ns`, $\rho$ represents the volume mass density, $U$ the momentum density (defined as $\bm{U}=\rho \bm{u}$, where $\bm{u}$ is the vector velocity field), $E$ the total energy density (defined as $E = \rho e$, where $e$ is the total energy including thermal and kinetic but not potential energy), $\bm{I}_3$ represents the $3 \times 3$ identity matrix, $\bm{b}$ is a body force vector (e.g., gravity vector $\bm{g}$),  $k$ the thermal conductivity constant, $T$ represents the temperature, and $P$ the pressure, given by the following equation of state
29
30$$
31P = \left( {c_p}/{c_v} -1\right) \left( E - {\bm{U}\cdot\bm{U}}/{(2 \rho)} \right) \, ,
32$$ (eq-state)
33
34where $c_p$ is the specific heat at constant pressure and $c_v$ is the specific heat at constant volume (that define $\gamma = c_p / c_v$, the specific heat ratio).
35
36The system {eq}`eq-ns` can be rewritten in vector form
37
38$$
39\frac{\partial \bm{q}}{\partial t} + \nabla \cdot \bm{F}(\bm{q}) -S(\bm{q}) = 0 \, ,
40$$ (eq-vector-ns)
41
42for the state variables 5-dimensional vector
43
44$$
45\bm{q} =        \begin{pmatrix}            \rho \\            \bm{U} \equiv \rho \bm{ u }\\            E \equiv \rho e        \end{pmatrix}        \begin{array}{l}            \leftarrow\textrm{ volume mass density}\\            \leftarrow\textrm{ momentum density}\\            \leftarrow\textrm{ energy density}        \end{array}
46$$
47
48where the flux and the source terms, respectively, are given by
49
50$$
51\begin{aligned}
52\bm{F}(\bm{q}) &=
53\underbrace{\begin{pmatrix}
54    \bm{U}\\
55    {(\bm{U} \otimes \bm{U})}/{\rho} + P \bm{I}_3 \\
56    {(E + P)\bm{U}}/{\rho}
57\end{pmatrix}}_{\bm F_{\text{adv}}} +
58\underbrace{\begin{pmatrix}
590 \\
60-  \bm{\sigma} \\
61 - \bm{u}  \cdot \bm{\sigma} - k \nabla T
62\end{pmatrix}}_{\bm F_{\text{diff}}},\\
63S(\bm{q}) &=
64 \begin{pmatrix}
65    0\\
66    \rho \bm{b}\\
67    \rho \bm{b}\cdot \bm{u}
68\end{pmatrix}.
69\end{aligned}
70$$ (eq-ns-flux)
71
72### Finite Element Formulation (Spatial Discretization)
73
74Let the discrete solution be
75
76$$
77\bm{q}_N (\bm{x},t)^{(e)} = \sum_{k=1}^{P}\psi_k (\bm{x})\bm{q}_k^{(e)}
78$$
79
80with $P=p+1$ the number of nodes in the element $e$.
81We use tensor-product bases $\psi_{kji} = h_i(X_0)h_j(X_1)h_k(X_2)$.
82
83To obtain a finite element discretization, we first multiply the strong form {eq}`eq-vector-ns` by a test function $\bm v \in H^1(\Omega)$ and integrate,
84
85$$
86\int_{\Omega} \bm v \cdot \left(\frac{\partial \bm{q}_N}{\partial t} + \nabla \cdot \bm{F}(\bm{q}_N) - \bm{S}(\bm{q}_N) \right) \,dV = 0 \, , \; \forall \bm v \in \mathcal{V}_p\,,
87$$
88
89with $\mathcal{V}_p = \{ \bm v(\bm x) \in H^{1}(\Omega_e) \,|\, \bm v(\bm x_e(\bm X)) \in P_p(\bm{I}), e=1,\ldots,N_e \}$ a mapped space of polynomials containing at least polynomials of degree $p$ (with or without the higher mixed terms that appear in tensor product spaces).
90
91Integrating by parts on the divergence term, we arrive at the weak form,
92
93$$
94\begin{aligned}
95\int_{\Omega} \bm v \cdot \left( \frac{\partial \bm{q}_N}{\partial t} - \bm{S}(\bm{q}_N) \right)  \,dV
96- \int_{\Omega} \nabla \bm v \!:\! \bm{F}(\bm{q}_N)\,dV & \\
97+ \int_{\partial \Omega} \bm v \cdot \bm{F}(\bm q_N) \cdot \widehat{\bm{n}} \,dS
98  &= 0 \, , \; \forall \bm v \in \mathcal{V}_p \,,
99\end{aligned}
100$$ (eq-weak-vector-ns)
101
102where $\bm{F}(\bm q_N) \cdot \widehat{\bm{n}}$ is typically replaced with a boundary condition.
103
104:::{note}
105The notation $\nabla \bm v \!:\! \bm F$ represents contraction over both fields and spatial dimensions while a single dot represents contraction in just one, which should be clear from context, e.g., $\bm v \cdot \bm S$ contracts over fields while $\bm F \cdot \widehat{\bm n}$ contracts over spatial dimensions.
106:::
107
108### Time Discretization
109For the time discretization, we use two types of time stepping schemes through PETSc.
110
111#### Explicit time-stepping method
112
113  The following explicit formulation is solved with the adaptive Runge-Kutta-Fehlberg (RKF4-5) method by default (any explicit time-stepping scheme available in PETSc can be chosen at runtime)
114
115  $$
116  \bm{q}_N^{n+1} = \bm{q}_N^n + \Delta t \sum_{i=1}^{s} b_i k_i \, ,
117  $$
118
119  where
120
121  $$
122  \begin{aligned}
123     k_1 &= f(t^n, \bm{q}_N^n)\\
124     k_2 &= f(t^n + c_2 \Delta t, \bm{q}_N^n + \Delta t (a_{21} k_1))\\
125     k_3 &= f(t^n + c_3 \Delta t, \bm{q}_N^n + \Delta t (a_{31} k_1 + a_{32} k_2))\\
126     \vdots&\\
127     k_i &= f\left(t^n + c_i \Delta t, \bm{q}_N^n + \Delta t \sum_{j=1}^s a_{ij} k_j \right)\\
128  \end{aligned}
129  $$
130
131  and with
132
133  $$
134  f(t^n, \bm{q}_N^n) = - [\nabla \cdot \bm{F}(\bm{q}_N)]^n + [S(\bm{q}_N)]^n \, .
135  $$
136
137#### Implicit time-stepping method
138
139  This time stepping method which can be selected using the option `-implicit` is solved with Backward Differentiation Formula (BDF) method by default (similarly, any implicit time-stepping scheme available in PETSc can be chosen at runtime).
140  The implicit formulation solves nonlinear systems for $\bm q_N$:
141
142  $$
143  \bm f(\bm q_N) \equiv \bm g(t^{n+1}, \bm{q}_N, \bm{\dot{q}}_N) = 0 \, ,
144  $$ (eq-ts-implicit-ns)
145
146  where the time derivative $\bm{\dot q}_N$ is defined by
147
148  $$
149  \bm{\dot{q}}_N(\bm q_N) = \alpha \bm q_N + \bm z_N
150  $$
151
152  in terms of $\bm z_N$ from prior state and $\alpha > 0$, both of which depend on the specific time integration scheme (backward difference formulas, generalized alpha, implicit Runge-Kutta, etc.).
153  Each nonlinear system {eq}`eq-ts-implicit-ns` will correspond to a weak form, as explained below.
154  In determining how difficult a given problem is to solve, we consider the Jacobian of {eq}`eq-ts-implicit-ns`,
155
156  $$
157  \frac{\partial \bm f}{\partial \bm q_N} = \frac{\partial \bm g}{\partial \bm q_N} + \alpha \frac{\partial \bm g}{\partial \bm{\dot q}_N}.
158  $$
159
160  The scalar "shift" $\alpha$ scales inversely with the time step $\Delta t$, so small time steps result in the Jacobian being dominated by the second term, which is a sort of "mass matrix", and typically well-conditioned independent of grid resolution with a simple preconditioner (such as Jacobi).
161  In contrast, the first term dominates for large time steps, with a condition number that grows with the diameter of the domain and polynomial degree of the approximation space.
162  Both terms are significant for time-accurate simulation and the setup costs of strong preconditioners must be balanced with the convergence rate of Krylov methods using weak preconditioners.
163
164More details of PETSc's time stepping solvers can be found in the [TS User Guide](https://petsc.org/release/docs/manual/ts/).
165
166### Stabilization
167We solve {eq}`eq-weak-vector-ns` using a Galerkin discretization (default) or a stabilized method, as is necessary for most real-world flows.
168
169Galerkin methods produce oscillations for transport-dominated problems (any time the cell Péclet number is larger than 1), and those tend to blow up for nonlinear problems such as the Euler equations and (low-viscosity/poorly resolved) Navier-Stokes, in which case stabilization is necessary.
170Our formulation follows {cite}`hughesetal2010`, which offers a comprehensive review of stabilization and shock-capturing methods for continuous finite element discretization of compressible flows.
171
172- **SUPG** (streamline-upwind/Petrov-Galerkin)
173
174  In this method, the weighted residual of the strong form {eq}`eq-vector-ns` is added to the Galerkin formulation {eq}`eq-weak-vector-ns`.
175  The weak form for this method is given as
176
177  $$
178  \begin{aligned}
179  \int_{\Omega} \bm v \cdot \left( \frac{\partial \bm{q}_N}{\partial t} - \bm{S}(\bm{q}_N) \right)  \,dV
180  - \int_{\Omega} \nabla \bm v \!:\! \bm{F}(\bm{q}_N)\,dV & \\
181  + \int_{\partial \Omega} \bm v \cdot \bm{F}(\bm{q}_N) \cdot \widehat{\bm{n}} \,dS & \\
182  + \int_{\Omega} \nabla\bm v \tcolon\left(\frac{\partial \bm F_{\text{adv}}}{\partial \bm q}\right) \bm\tau \left( \frac{\partial \bm{q}_N}{\partial t} \, + \,
183  \nabla \cdot \bm{F} \, (\bm{q}_N) - \bm{S}(\bm{q}_N) \right) \,dV &= 0
184  \, , \; \forall \bm v \in \mathcal{V}_p
185  \end{aligned}
186  $$ (eq-weak-vector-ns-supg)
187
188  This stabilization technique can be selected using the option `-stab supg`.
189
190- **SU** (streamline-upwind)
191
192  This method is a simplified version of *SUPG* {eq}`eq-weak-vector-ns-supg` which is developed for debugging/comparison purposes. The weak form for this method is
193
194  $$
195  \begin{aligned}
196  \int_{\Omega} \bm v \cdot \left( \frac{\partial \bm{q}_N}{\partial t} - \bm{S}(\bm{q}_N) \right)  \,dV
197  - \int_{\Omega} \nabla \bm v \!:\! \bm{F}(\bm{q}_N)\,dV & \\
198  + \int_{\partial \Omega} \bm v \cdot \bm{F}(\bm{q}_N) \cdot \widehat{\bm{n}} \,dS & \\
199  + \int_{\Omega} \nabla\bm v \tcolon\left(\frac{\partial \bm F_{\text{adv}}}{\partial \bm q}\right) \bm\tau \nabla \cdot \bm{F} \, (\bm{q}_N) \,dV
200  & = 0 \, , \; \forall \bm v \in \mathcal{V}_p
201  \end{aligned}
202  $$ (eq-weak-vector-ns-su)
203
204  This stabilization technique can be selected using the option `-stab su`.
205
206In both {eq}`eq-weak-vector-ns-su` and {eq}`eq-weak-vector-ns-supg`, $\bm\tau \in \mathbb R^{5\times 5}$ (field indices) is an intrinsic time scale matrix.
207The SUPG technique and the operator $\frac{\partial \bm F_{\text{adv}}}{\partial \bm q}$ (rather than its transpose) can be explained via an ansatz for subgrid state fluctuations $\tilde{\bm q} = -\bm\tau \bm r$ where $\bm r$ is a strong form residual.
208The forward variational form can be readily expressed by differentiating $\bm F_{\text{adv}}$ of {eq}`eq-ns-flux`
209
210$$
211\begin{aligned}
212\diff\bm F_{\text{adv}}(\diff\bm q; \bm q) &= \frac{\partial \bm F_{\text{adv}}}{\partial \bm q} \diff\bm q \\
213&= \begin{pmatrix}
214\diff\bm U \\
215(\diff\bm U \otimes \bm U + \bm U \otimes \diff\bm U)/\rho - (\bm U \otimes \bm U)/\rho^2 \diff\rho + \diff P \bm I_3 \\
216(E + P)\diff\bm U/\rho + (\diff E + \diff P)\bm U/\rho - (E + P) \bm U/\rho^2 \diff\rho
217\end{pmatrix},
218\end{aligned}
219$$
220
221where $\diff P$ is defined by differentiating {eq}`eq-state`.
222
223:::{dropdown} Stabilization scale $\bm\tau$
224A velocity vector $\bm u$ can be pulled back to the reference element as $\bm u_{\bm X} = \nabla_{\bm x}\bm X \cdot \bm u$, with units of reference length (non-dimensional) per second.
225To build intuition, consider a boundary layer element of dimension $(1, \epsilon)$, for which $\nabla_{\bm x} \bm X = \bigl(\begin{smallmatrix} 2 & \\ & 2/\epsilon \end{smallmatrix}\bigr)$.
226So a small normal component of velocity will be amplified (by a factor of the aspect ratio $1/\epsilon$) in this transformation.
227The ratio $\lVert \bm u \rVert / \lVert \bm u_{\bm X} \rVert$ is a covariant measure of (half) the element length in the direction of the velocity.
228A contravariant measure of element length in the direction of a unit vector $\hat{\bm n}$ is given by $\lVert \bigl(\nabla_{\bm X} \bm x\bigr)^T \hat{\bm n} \rVert$.
229While $\nabla_{\bm X} \bm x$ is readily computable, its inverse $\nabla_{\bm x} \bm X$ is needed directly in finite element methods and thus more convenient for our use.
230If we consider a parallelogram, the covariant measure is larger than the contravariant measure for vectors pointing between acute corners and the opposite holds for vectors between oblique corners.
231
232The cell Péclet number is classically defined by $\mathrm{Pe}_h = \lVert \bm u \rVert h / (2 \kappa)$ where $\kappa$ is the diffusivity (units of $m^2/s$).
233This can be generalized to arbitrary grids by defining the local Péclet number
234
235$$
236\mathrm{Pe} = \frac{\lVert \bm u \rVert^2}{\lVert \bm u_{\bm X} \rVert \kappa}.
237$$ (eq-peclet)
238
239For scalar advection-diffusion, the stabilization is a scalar
240
241$$
242\tau = \frac{\xi(\mathrm{Pe})}{\lVert \bm u_{\bm X} \rVert},
243$$ (eq-tau-advdiff)
244
245where $\xi(\mathrm{Pe}) = \coth \mathrm{Pe} - 1/\mathrm{Pe}$ approaches 1 at large local Péclet number.
246Note that $\tau$ has units of time and, in the transport-dominated limit, is proportional to element transit time in the direction of the propagating wave.
247For advection-diffusion, $\bm F(q) = \bm u q$, and thus the SU stabilization term is
248
249$$
250\nabla v \cdot \bm u \tau \bm u \cdot \nabla q = \nabla_{\bm X} v \cdot (\bm u_{\bm X} \tau \bm u_{\bm X}) \cdot \nabla_{\bm X} q .
251$$ (eq-su-stabilize-advdiff)
252
253where the term in parentheses is a rank-1 diffusivity tensor that has been pulled back to the reference element.
254See {cite}`hughesetal2010` equations 15-17 and 34-36 for further discussion of this formulation.
255
256For the Navier-Stokes and Euler equations, {cite}`whiting2003hierarchical` defines a $5\times 5$ diagonal stabilization $\mathrm{diag}(\tau_c, \tau_m, \tau_m, \tau_m, \tau_E)$ consisting of
2571. continuity stabilization $\tau_c$
2582. momentum stabilization $\tau_m$
2593. energy stabilization $\tau_E$
260
261The Navier-Stokes code in this example uses the following formulation for $\tau_c$, $\tau_m$, $\tau_E$:
262
263$$
264\begin{aligned}
265
266\tau_c &= \frac{C_c \mathcal{F}}{8\rho \trace(\bm g)} \\
267\tau_m &= \frac{C_m}{\mathcal{F}} \\
268\tau_E &= \frac{C_E}{\mathcal{F} c_v} \\
269\end{aligned}
270$$
271
272$$
273\mathcal{F} = \sqrt{ \rho^2 \left [ \left(\frac{2C_t}{\Delta t}\right)^2
274+ \bm u \cdot (\bm u \cdot  \bm g)\right]
275+ C_v \mu^2 \Vert \bm g \Vert_F ^2}
276$$
277
278where $\bm g = \nabla_{\bm x} \bm{X}^T \cdot \nabla_{\bm x} \bm{X}$ is the metric tensor and $\Vert \cdot \Vert_F$ is the Frobenius norm.
279This formulation is currently not available in the Euler code.
280
281For Advection-Diffusion, we use a modified version of the formulation for Navier-Stokes:
282
283$$
284\tau = \left [ \left(\frac{2 C_t}{\Delta t}\right)^2
285+ \frac{\bm u \cdot (\bm u \cdot  \bm g)}{C_a}
286+ \frac{\kappa^2 \Vert \bm g \Vert_F ^2}{C_d} \right]^{-1/2}
287$$
288for $C_t$, $C_a$, $C_d$ being some scaling coefficients.
289Otherwise, $C_a$ is set via `-Ctau_a` and $C_t$ via `-Ctau_t`.
290
291In the Euler code, we follow {cite}`hughesetal2010` in defining a $3\times 3$ diagonal stabilization according to spatial criterion 2 (equation 27) as follows.
292
293$$
294\tau_{ii} = c_{\tau} \frac{2 \xi(\mathrm{Pe})}{(\lambda_{\max \text{abs}})_i \lVert \nabla_{x_i} \bm X \rVert}
295$$ (eq-tau-conservative)
296
297where $c_{\tau}$ is a multiplicative constant reported to be optimal at 0.5 for linear elements, $\hat{\bm n}_i$ is a unit vector in direction $i$, and $\nabla_{x_i} = \hat{\bm n}_i \cdot \nabla_{\bm x}$ is the derivative in direction $i$.
298The flux Jacobian $\frac{\partial \bm F_{\text{adv}}}{\partial \bm q} \cdot \hat{\bm n}_i$ in each direction $i$ is a $5\times 5$ matrix with spectral radius $(\lambda_{\max \text{abs}})_i$ equal to the fastest wave speed.
299The complete set of eigenvalues of the Euler flux Jacobian in direction $i$ are (e.g., {cite}`toro2009`)
300
301$$
302\Lambda_i = [u_i - a, u_i, u_i, u_i, u_i+a],
303$$ (eq-eigval-advdiff)
304
305where $u_i = \bm u \cdot \hat{\bm n}_i$ is the velocity component in direction $i$ and $a = \sqrt{\gamma P/\rho}$ is the sound speed for ideal gasses.
306Note that the first and last eigenvalues represent nonlinear acoustic waves while the middle three are linearly degenerate, carrying a contact wave (temperature) and transverse components of momentum.
307The fastest wave speed in direction $i$ is thus
308
309$$
310\lambda_{\max \text{abs}} \Bigl( \frac{\partial \bm F_{\text{adv}}}{\partial \bm q} \cdot \hat{\bm n}_i \Bigr) = |u_i| + a
311$$ (eq-wavespeed)
312
313Note that this wave speed is specific to ideal gases as $\gamma$ is an ideal gas parameter; other equations of state will yield a different acoustic wave speed.
314
315:::
316
317Currently, this demo provides three types of problems/physical models that can be selected at run time via the option `-problem`.
318{ref}`problem-advection`, the problem of the transport of energy in a uniform vector velocity field, {ref}`problem-euler-vortex`, the exact solution to the Euler equations, and the so called {ref}`problem-density-current` problem.
319
320### Subgrid Stress Modeling
321
322When a fluid simulation is under-resolved (the smallest length scale resolved by the grid is much larger than the smallest physical scale, the [Kolmogorov length scale](https://en.wikipedia.org/wiki/Kolmogorov_microscales)), this is mathematically interpreted as filtering the Navier-Stokes equations.
323This is known as large-eddy simulation (LES), as only the "large" scales of turbulence are resolved.
324This filtering operation results in an extra stress-like term, $\bm{\tau}^r$, representing the effect of unresolved (or "subgrid" scale) structures in the flow.
325Denoting the filtering operation by $\overline \cdot$, the LES governing equations are:
326
327$$
328\frac{\partial \bm{\overline q}}{\partial t} + \nabla \cdot \bm{\overline F}(\bm{\overline q}) -S(\bm{\overline q}) = 0 \, ,
329$$ (eq-vector-les)
330
331where
332
333$$
334\bm{\overline F}(\bm{\overline q}) =
335\bm{F} (\bm{\overline q}) +
336\begin{pmatrix}
337    0\\
338     \bm{\tau}^r \\
339     \bm{u}  \cdot \bm{\tau}^r
340\end{pmatrix}
341$$ (eq-les-flux)
342
343More details on deriving the above expression, filtering, and large eddy simulation can be found in {cite}`popeTurbulentFlows2000`.
344To close the problem, the subgrid stress must be defined.
345For implicit LES, the subgrid stress is set to zero and the numerical properties of the discretized system are assumed to account for the effect of subgrid scale structures on the filtered solution field.
346For explicit LES, it is defined by a subgrid stress model.
347
348(sgs-dd-model)=
349#### Data-driven SGS Model
350
351The data-driven SGS model implemented here uses a small neural network to compute the SGS term.
352The SGS tensor is calculated at nodes using an $L^2$ projection of the velocity gradient and grid anisotropy tensor, and then interpolated onto quadrature points.
353More details regarding the theoretical background of the model can be found in {cite}`prakashDDSGS2022` and {cite}`prakashDDSGSAnisotropic2022`.
354
355The neural network itself consists of 1 hidden layer and 20 neurons, using Leaky ReLU as its activation function.
356The slope parameter for the Leaky ReLU function is set via `-sgs_model_dd_leakyrelu_alpha`.
357The outputs of the network are assumed to be normalized on a min-max scale, so they must be rescaled by the original min-max bounds.
358Parameters for the neural network are put into files in a directory found in `-sgs_model_dd_parameter_dir`.
359These files store the network weights (`w1.dat` and `w2.dat`), biases (`b1.dat` and `b2.dat`), and scaling parameters (`OutScaling.dat`).
360The first row of each files stores the number of columns and rows in each file.
361Note that the weight coefficients are assumed to be in column-major order.
362This is done to keep consistent with legacy file compatibility.
363
364:::{note}
365The current data-driven model parameters are not accurate and are for regression testing only.
366:::
367
368##### Data-driven Model Using External Libraries
369
370There are two different modes for using the data-driven model: fused and sequential.
371
372In fused mode, the input processing, model inference, and output handling were all done in a single CeedOperator.
373Conversely, sequential mode has separate function calls/CeedOperators for input creation, model inference, and output handling.
374By separating the three steps to the model evaluation, the sequential mode allows for functions calling external libraries to be used for the model inference step.
375This however is slower than the fused kernel, but this requires a native libCEED inference implementation.
376
377To use the fused mode, set `-sgs_model_dd_use_fused true`.
378To use the sequential mode, set the same flag to `false`.
379
380(differential-filtering)=
381### Differential Filtering
382
383There is the option to filter the solution field using differential filtering.
384This was first proposed in {cite}`germanoDiffFilterLES1986`, using an inverse Hemholtz operator.
385The strong form of the differential equation is
386
387$$
388\overline{\phi} - \nabla \cdot (\beta (\bm{D}\bm{\Delta})^2 \nabla \overline{\phi} ) = \phi
389$$
390
391for $\phi$ the scalar solution field we want to filter, $\overline \phi$ the filtered scalar solution field, $\bm{\Delta} \in \mathbb{R}^{3 \times 3}$ a symmetric positive-definite rank 2 tensor defining the width of the filter, $\bm{D}$ is the filter width scaling tensor (also a rank 2 SPD tensor), and $\beta$ is a kernel scaling factor on the filter tensor.
392This admits the weak form:
393
394$$
395\int_\Omega \left( v \overline \phi + \beta \nabla v \cdot (\bm{D}\bm{\Delta})^2 \nabla \overline \phi \right) \,d\Omega
396- \cancel{\int_{\partial \Omega} \beta v \nabla \overline \phi \cdot (\bm{D}\bm{\Delta})^2 \bm{\hat{n}} \,d\partial\Omega} =
397\int_\Omega v \phi \, , \; \forall v \in \mathcal{V}_p
398$$
399
400The boundary integral resulting from integration-by-parts is crossed out, as we assume that $(\bm{D}\bm{\Delta})^2 = \bm{0} \Leftrightarrow \overline \phi = \phi$ at boundaries (this is reasonable at walls, but for convenience elsewhere).
401
402#### Filter width tensor, Δ
403For homogenous filtering, $\bm{\Delta}$ is defined as the identity matrix.
404
405:::{note}
406It is common to denote a filter width dimensioned relative to the radial distance of the filter kernel.
407Note here we use the filter *diameter* instead, as that feels more natural (albeit mathematically less convenient).
408For example, under this definition a box filter would be defined as:
409
410$$
411B(\Delta; \bm{r}) =
412\begin{cases}
4131 & \Vert \bm{r} \Vert \leq \Delta/2 \\
4140 & \Vert \bm{r} \Vert > \Delta/2
415\end{cases}
416$$
417:::
418
419For inhomogeneous anisotropic filtering, we use the finite element grid itself to define $\bm{\Delta}$.
420This is set via `-diff_filter_grid_based_width`.
421Specifically, we use the filter width tensor defined in {cite}`prakashDDSGSAnisotropic2022`.
422For finite element grids, the filter width tensor is most conveniently defined by $\bm{\Delta} = \bm{g}^{-1/2}$ where $\bm g = \nabla_{\bm x} \bm{X} \cdot \nabla_{\bm x} \bm{X}$ is the metric tensor.
423
424#### Filter width scaling tensor, $\bm{D}$
425The filter width tensor $\bm{\Delta}$, be it defined from grid based sources or just the homogenous filtering, can be scaled anisotropically.
426The coefficients for that anisotropic scaling are given by `-diff_filter_width_scaling`, denoted here by $c_1, c_2, c_3$.
427The definition for $\bm{D}$ then becomes
428
429$$
430\bm{D} =
431\begin{bmatrix}
432    c_1 & 0        & 0        \\
433    0        & c_2 & 0        \\
434    0        & 0        & c_3 \\
435\end{bmatrix}
436$$
437
438In the case of $\bm{\Delta}$ being defined as homogenous, $\bm{D}\bm{\Delta}$ means that $\bm{D}$ effectively sets the filter width.
439
440The filtering at the wall may also be damped, to smoothly meet the $\overline \phi = \phi$ boundary condition at the wall.
441The selected damping function for this is the van Driest function {cite}`vandriestWallDamping1956`:
442
443$$
444\zeta = 1 - \exp\left(-\frac{y^+}{A^+}\right)
445$$
446
447where $y^+$ is the wall-friction scaled wall-distance ($y^+ = y u_\tau / \nu = y/\delta_\nu$), $A^+$ is some wall-friction scaled scale factor, and $\zeta$ is the damping coefficient.
448For this implementation, we assume that $\delta_\nu$ is constant across the wall and is defined by `-diff_filter_friction_length`.
449$A^+$ is defined by `-diff_filter_damping_constant`.
450
451To apply this scalar damping coefficient to the filter width tensor, we construct the wall-damping tensor from it.
452The construction implemented currently limits damping in the wall parallel directions to be no less than the original filter width defined by $\bm{\Delta}$.
453The wall-normal filter width is allowed to be damped to a zero filter width.
454It is currently assumed that the second component of the filter width tensor is in the wall-normal direction.
455Under these assumptions, $\bm{D}$ then becomes:
456
457$$
458\bm{D} =
459\begin{bmatrix}
460    \max(1, \zeta c_1) & 0         & 0                  \\
461    0                  & \zeta c_2 & 0                  \\
462    0                  & 0         & \max(1, \zeta c_3) \\
463\end{bmatrix}
464$$
465
466#### Filter kernel scaling, β
467While we define $\bm{D}\bm{\Delta}$ to be of a certain physical filter width, the actual width of the implied filter kernel is quite larger than "normal" kernels.
468To account for this, we use $\beta$ to scale the filter tensor to the appropriate size, as is done in {cite}`bullExplicitFilteringExact2016`.
469To match the "size" of a normal kernel to our differential kernel, we attempt to have them match second order moments with respect to the prescribed filter width.
470To match the box and Gaussian filters "sizes", we use $\beta = 1/10$ and $\beta = 1/6$, respectively.
471$\beta$ can be set via `-diff_filter_kernel_scaling`.
472
473### *In Situ* Machine-Learning Model Training
474Training machine-learning models normally uses *a priori* (already gathered) data stored on disk.
475This is computationally inefficient, particularly as the scale of the problem grows and the data that is saved to disk reduces to a small percentage of the total data generated by a simulation.
476One way of working around this to to train a model on data coming from an ongoing simulation, known as *in situ* (in place) learning.
477
478This is implemented in the code using [SmartSim](https://www.craylabs.org/docs/overview.html).
479Briefly, the fluid simulation will periodically place data for training purposes into a database that a separate process uses to train a model.
480The database used by SmartSim is [Redis](https://redis.com/modules/redis-ai/) and the library to connect to the database is called [SmartRedis](https://www.craylabs.org/docs/smartredis.html).
481More information about how to utilize this code in a SmartSim configuration can be found on [SmartSim's website](https://www.craylabs.org/docs/overview.html).
482
483To use this code in a SmartSim *in situ* setup, first the code must be built with SmartRedis enabled.
484This is done by specifying the installation directory of SmartRedis using the `SMARTREDIS_DIR` environment variable when building:
485
486```
487make SMARTREDIS_DIR=~/software/smartredis/install
488```
489
490#### SGS Data-Driven Model *In Situ* Training
491Currently the code is only setup to do *in situ* training for the SGS data-driven model.
492Training data is split into the model inputs and outputs.
493The model inputs are calculated as the same model inputs in the SGS Data-Driven model described {ref}`earlier<sgs-dd-model>`.
494The model outputs (or targets in the case of training) are the subgrid stresses.
495Both the inputs and outputs are computed from a filtered velocity field, which is calculated via {ref}`differential-filtering`.
496The settings for the differential filtering used during training are described in {ref}`differential-filtering`.
497
498The SGS *in situ* training can be enabled using the `-sgs_train_enable` flag.
499Data can be processed and placed into the database periodically.
500The interval between is controlled by `-sgs_train_write_data_interval`.
501There's also the choice of whether to add new training data on each database write or to overwrite the old data with new data.
502This is controlled by `-sgs_train_overwrite_data`.
503
504The database may also be located on the same node as a MPI rank (collocated) or located on a separate node (distributed).
505It's necessary to know how many ranks are associated with each collocated database, which is set by `-smartsim_collocated_database_num_ranks`.
506
507(problem-advection)=
508## Advection-Diffusion
509
510A simplified version of system {eq}`eq-ns`, only accounting for the transport of total energy, is given by
511
512$$
513\frac{\partial E}{\partial t} + \nabla \cdot (\bm{u} E ) - \kappa \nabla E = 0 \, ,
514$$ (eq-advection)
515
516with $\bm{u}$ the vector velocity field and $\kappa$ the diffusion coefficient.
517In this particular test case, a blob of total energy (defined by a characteristic radius $r_c$) is transported by two different wind types.
518
519- **Rotation**
520
521  In this case, a uniform circular velocity field transports the blob of total energy.
522  We have solved {eq}`eq-advection` applying zero energy density $E$, and no-flux for $\bm{u}$ on the boundaries.
523
524- **Translation**
525
526  In this case, a background wind with a constant rectilinear velocity field, enters the domain and transports the blob of total energy out of the domain.
527
528  For the inflow boundary conditions, a prescribed $E_{wind}$ is applied weakly on the inflow boundaries such that the weak form boundary integral in {eq}`eq-weak-vector-ns` is defined as
529
530  $$
531  \int_{\partial \Omega_{inflow}} \bm v \cdot \bm{F}(\bm q_N) \cdot \widehat{\bm{n}} \,dS = \int_{\partial \Omega_{inflow}} \bm v \, E_{wind} \, \bm u \cdot \widehat{\bm{n}} \,dS  \, ,
532  $$
533
534  For the outflow boundary conditions, we have used the current values of $E$, following {cite}`papanastasiou1992outflow` which extends the validity of the weak form of the governing equations to the outflow instead of replacing them with unknown essential or natural boundary conditions.
535  The weak form boundary integral in {eq}`eq-weak-vector-ns` for outflow boundary conditions is defined as
536
537  $$
538  \int_{\partial \Omega_{outflow}} \bm v \cdot \bm{F}(\bm q_N) \cdot \widehat{\bm{n}} \,dS = \int_{\partial \Omega_{outflow}} \bm v \, E \, \bm u \cdot \widehat{\bm{n}} \,dS  \, ,
539  $$
540
541(problem-euler-vortex)=
542
543## Isentropic Vortex
544
545Three-dimensional Euler equations, which are simplified and nondimensionalized version of system {eq}`eq-ns` and account only for the convective fluxes, are given by
546
547$$
548\begin{aligned}
549\frac{\partial \rho}{\partial t} + \nabla \cdot \bm{U} &= 0 \\
550\frac{\partial \bm{U}}{\partial t} + \nabla \cdot \left( \frac{\bm{U} \otimes \bm{U}}{\rho} + P \bm{I}_3 \right) &= 0 \\
551\frac{\partial E}{\partial t} + \nabla \cdot \left( \frac{(E + P)\bm{U}}{\rho} \right) &= 0 \, , \\
552\end{aligned}
553$$ (eq-euler)
554
555Following the setup given in {cite}`zhang2011verification`, the mean flow for this problem is $\rho=1$, $P=1$, $T=P/\rho= 1$ (Specific Gas Constant, $R$, is 1), and $\bm{u}=(u_1,u_2,0)$ while the perturbation $\delta \bm{u}$, and $\delta T$ are defined as
556
557$$
558\begin{aligned} (\delta u_1, \, \delta u_2) &= \frac{\epsilon}{2 \pi} \, e^{0.5(1-r^2)} \, (-\bar{y}, \, \bar{x}) \, , \\ \delta T &= - \frac{(\gamma-1) \, \epsilon^2}{8 \, \gamma \, \pi^2} \, e^{1-r^2} \, , \\ \end{aligned}
559$$
560
561where $(\bar{x}, \, \bar{y}) = (x-x_c, \, y-y_c)$, $(x_c, \, y_c)$ represents the center of the domain, $r^2=\bar{x}^2 + \bar{y}^2$, and $\epsilon$ is the vortex strength ($\epsilon$ < 10).
562There is no perturbation in the entropy $S=P/\rho^\gamma$ ($\delta S=0)$.
563
564(problem-shock-tube)=
565
566## Shock Tube
567
568This test problem is based on Sod's Shock Tube (from{cite}`sodshocktubewiki`), a canonical test case for discontinuity capturing in one dimension. For this problem, the three-dimensional Euler equations are formulated exactly as in the Isentropic Vortex problem. The default initial conditions are $P=1$, $\rho=1$ for the driver section and $P=0.1$, $\rho=0.125$ for the driven section. The initial velocity is zero in both sections. Symmetry boundary conditions are applied to the side walls and wall boundary conditions are applied at the end walls.
569
570SU upwinding and discontinuity capturing have been implemented into the explicit timestepping operator for this problem. Discontinuity capturing is accomplished using a modified version of the $YZ\beta$ operator described in {cite}`tezduyar2007yzb`. This discontinuity capturing scheme involves the introduction of a dissipation term of the form
571
572$$
573\int_{\Omega} \nu_{SHOCK} \nabla \bm v \!:\! \nabla \bm q dV
574$$
575
576The shock capturing viscosity is implemented following the first formulation described in {cite}`tezduyar2007yzb`. The characteristic velocity $u_{cha}$ is taken to be the acoustic speed while the reference density $\rho_{ref}$ is just the local density. Shock capturing viscosity is defined by the following
577
578$$
579\nu_{SHOCK} = \tau_{SHOCK} u_{cha}^2
580$$
581
582where,
583
584$$
585\tau_{SHOCK} = \frac{h_{SHOCK}}{2u_{cha}} \left( \frac{ \,|\, \nabla \rho \,|\, h_{SHOCK}}{\rho_{ref}} \right)^{\beta}
586$$
587
588$\beta$ is a tuning parameter set between 1 (smoother shocks) and 2 (sharper shocks. The parameter $h_{SHOCK}$ is a length scale that is proportional to the element length in the direction of the density gradient unit vector. This density gradient unit vector is defined as $\hat{\bm j} = \frac{\nabla \rho}{|\nabla \rho|}$. The original formulation of Tezduyar and Senga relies on the shape function gradient to define the element length scale, but this gradient is not available to qFunctions in libCEED. To avoid this problem, $h_{SHOCK}$ is defined in the current implementation as
589
590$$
591h_{SHOCK} = 2 \left( C_{YZB} \,|\, \bm p \,|\, \right)^{-1}
592$$
593
594where
595
596$$
597p_k = \hat{j}_i \frac{\partial \xi_i}{x_k}
598$$
599
600The constant $C_{YZB}$ is set to 0.1 for piecewise linear elements in the current implementation. Larger values approaching unity are expected with more robust stabilization and implicit timestepping.
601
602(problem-density-current)=
603
604## Gaussian Wave
605This test case is taken/inspired by that presented in {cite}`mengaldoCompressibleBC2014`. It is intended to test non-reflecting/Riemann boundary conditions. It's primarily intended for Euler equations, but has been implemented for the Navier-Stokes equations here for flexibility.
606
607The problem has a perturbed initial condition and lets it evolve in time. The initial condition contains a Gaussian perturbation in the pressure field:
608
609$$
610\begin{aligned}
611\rho &= \rho_\infty\left(1+A\exp\left(\frac{-(\bar{x}^2 + \bar{y}^2)}{2\sigma^2}\right)\right) \\
612\bm{U} &= \bm U_\infty \\
613E &= \frac{p_\infty}{\gamma -1}\left(1+A\exp\left(\frac{-(\bar{x}^2 + \bar{y}^2)}{2\sigma^2}\right)\right) + \frac{\bm U_\infty \cdot \bm U_\infty}{2\rho_\infty},
614\end{aligned}
615$$
616
617where $A$ and $\sigma$ are the amplitude and width of the perturbation, respectively, and $(\bar{x}, \bar{y}) = (x-x_e, y-y_e)$ is the distance to the epicenter of the perturbation, $(x_e, y_e)$.
618The simulation produces a strong acoustic wave and leaves behind a cold thermal bubble that advects at the fluid velocity.
619
620The boundary conditions are freestream in the x and y directions. When using an HLL (Harten, Lax, van Leer) Riemann solver {cite}`toro2009` (option `-freestream_riemann hll`), the acoustic waves exit the domain cleanly, but when the thermal bubble reaches the boundary, it produces strong thermal oscillations that become acoustic waves reflecting into the domain.
621This problem can be fixed using a more sophisticated Riemann solver such as HLLC {cite}`toro2009` (option `-freestream_riemann hllc`, which is default), which is a linear constant-pressure wave that transports temperature and transverse momentum at the fluid velocity.
622
623## Vortex Shedding - Flow past Cylinder
624This test case, based on {cite}`shakib1991femcfd`, is an example of using an externally provided mesh from Gmsh.
625A cylinder with diameter $D=1$ is centered at $(0,0)$ in a computational domain $-4.5 \leq x \leq 15.5$, $-4.5 \leq y \leq 4.5$.
626We solve this as a 3D problem with (default) one element in the $z$ direction.
627The domain is filled with an ideal gas at rest (zero velocity) with temperature 24.92 and pressure 7143.
628The viscosity is 0.01 and thermal conductivity is 14.34 to maintain a Prandtl number of 0.71, which is typical for air.
629At time $t=0$, this domain is subjected to freestream boundary conditions at the inflow (left) and Riemann-type outflow on the right, with exterior reference state at velocity $(1, 0, 0)$ giving Reynolds number $100$ and Mach number $0.01$.
630A symmetry (adiabatic free slip) condition is imposed at the top and bottom boundaries $(y = \pm 4.5)$ (zero normal velocity component, zero heat-flux).
631The cylinder wall is an adiabatic (no heat flux) no-slip boundary condition.
632As we evolve in time, eddies appear past the cylinder leading to a vortex shedding known as the vortex street, with shedding period of about 6.
633
634The Gmsh input file, `examples/fluids/meshes/cylinder.geo` is parametrized to facilitate experimenting with similar configurations.
635The Strouhal number (nondimensional shedding frequency) is sensitive to the size of the computational domain and boundary conditions.
636
637Forces on the cylinder walls are computed using the "reaction force" method, which is variationally consistent with the volume operator.
638Given the force components $\bm F = (F_x, F_y, F_z)$ and surface area $S = \pi D L_z$ where $L_z$ is the spanwise extent of the domain, we define the coefficients of lift and drag as
639
640$$
641\begin{aligned}
642C_L &= \frac{2 F_y}{\rho_\infty u_\infty^2 S} \\
643C_D &= \frac{2 F_x}{\rho_\infty u_\infty^2 S} \\
644\end{aligned}
645$$
646
647where $\rho_\infty, u_\infty$ are the freestream (inflow) density and velocity respectively.
648
649## Density Current
650
651For this test problem (from {cite}`straka1993numerical`), we solve the full Navier-Stokes equations {eq}`eq-ns`, for which a cold air bubble (of radius $r_c$) drops by convection in a neutrally stratified atmosphere.
652Its initial condition is defined in terms of the Exner pressure, $\pi(\bm{x},t)$, and potential temperature, $\theta(\bm{x},t)$, that relate to the state variables via
653
654$$
655\begin{aligned} \rho &= \frac{P_0}{( c_p - c_v)\theta(\bm{x},t)} \pi(\bm{x},t)^{\frac{c_v}{ c_p - c_v}} \, , \\ e &= c_v \theta(\bm{x},t) \pi(\bm{x},t) + \bm{u}\cdot \bm{u} /2 + g z \, , \end{aligned}
656$$
657
658where $P_0$ is the atmospheric pressure.
659For this problem, we have used no-slip and non-penetration boundary conditions for $\bm{u}$, and no-flux for mass and energy densities.
660
661## Channel
662
663A compressible channel flow. Analytical solution given in
664{cite}`whitingStabilizedFEM1999`:
665
666$$ u_1 = u_{\max} \left [ 1 - \left ( \frac{x_2}{H}\right)^2 \right] \quad \quad u_2 = u_3 = 0$$
667$$T = T_w \left [ 1 + \frac{Pr \hat{E}c}{3} \left \{1 - \left(\frac{x_2}{H}\right)^4  \right \} \right]$$
668$$p = p_0 - \frac{2\rho_0 u_{\max}^2 x_1}{Re_H H}$$
669
670where $H$ is the channel half-height, $u_{\max}$ is the center velocity, $T_w$ is the temperature at the wall, $Pr=\frac{\mu}{c_p \kappa}$ is the Prandlt number, $\hat E_c = \frac{u_{\max}^2}{c_p T_w}$ is the modified Eckert number, and $Re_h = \frac{u_{\max}H}{\nu}$ is the Reynolds number.
671
672Boundary conditions are periodic in the streamwise direction, and no-slip and non-penetration boundary conditions at the walls.
673The flow is driven by a body force determined analytically from the fluid properties and setup parameters $H$ and $u_{\max}$.
674
675## Flat Plate Boundary Layer
676
677### Laminar Boundary Layer - Blasius
678
679Simulation of a laminar boundary layer flow, with the inflow being prescribed
680by a [Blasius similarity
681solution](https://en.wikipedia.org/wiki/Blasius_boundary_layer). At the inflow,
682the velocity is prescribed by the Blasius soution profile, density is set
683constant, and temperature is allowed to float. Using `weakT: true`, density is
684allowed to float and temperature is set constant. At the outlet, a user-set
685pressure is used for pressure in the inviscid flux terms (all other inviscid
686flux terms use interior solution values). The wall is a no-slip,
687no-penetration, no-heat flux condition. The top of the domain is treated as an
688outflow and is tilted at a downward angle to ensure that flow is always exiting
689it.
690
691### Turbulent Boundary Layer
692
693Simulating a turbulent boundary layer without modeling the turbulence requires
694resolving the turbulent flow structures. These structures may be introduced
695into the simulations either by allowing a laminar boundary layer naturally
696transition to turbulence, or imposing turbulent structures at the inflow. The
697latter approach has been taken here, specifically using a *synthetic turbulence
698generation* (STG) method.
699
700#### Synthetic Turbulence Generation (STG) Boundary Condition
701
702We use the STG method described in
703{cite}`shurSTG2014`. Below follows a re-description of the formulation to match
704the present notation, and then a description of the implementation and usage.
705
706##### Equation Formulation
707
708$$
709\bm{u}(\bm{x}, t) = \bm{\overline{u}}(\bm{x}) + \bm{C}(\bm{x}) \cdot \bm{v}'
710$$
711
712$$
713\begin{aligned}
714\bm{v}' &= 2 \sqrt{3/2} \sum^N_{n=1} \sqrt{q^n(\bm{x})} \bm{\sigma}^n \cos(\kappa^n \bm{d}^n \cdot \bm{\hat{x}}^n(\bm{x}, t) + \phi^n ) \\
715\bm{\hat{x}}^n &= \left[(x - U_0 t)\max(2\kappa_{\min}/\kappa^n, 0.1) , y, z  \right]^T
716\end{aligned}
717$$
718
719Here, we define the number of wavemodes $N$, set of random numbers $ \{\bm{\sigma}^n,
720\bm{d}^n, \phi^n\}_{n=1}^N$, the Cholesky decomposition of the Reynolds stress
721tensor $\bm{C}$ (such that $\bm{R} = \bm{CC}^T$ ), bulk velocity $U_0$,
722wavemode amplitude $q^n$, wavemode frequency $\kappa^n$, and $\kappa_{\min} =
7230.5 \min_{\bm{x}} (\kappa_e)$.
724
725$$
726\kappa_e = \frac{2\pi}{\min(2d_w, 3.0 l_t)}
727$$
728
729where $l_t$ is the turbulence length scale, and $d_w$ is the distance to the
730nearest wall.
731
732
733The set of wavemode frequencies is defined by a geometric distribution:
734
735$$
736\kappa^n = \kappa_{\min} (1 + \alpha)^{n-1} \ , \quad \forall n=1, 2, ... , N
737$$
738
739The wavemode amplitudes $q^n$ are defined by a model energy spectrum $E(\kappa)$:
740
741$$
742q^n = \frac{E(\kappa^n) \Delta \kappa^n}{\sum^N_{n=1} E(\kappa^n)\Delta \kappa^n} \ ,\quad \Delta \kappa^n = \kappa^n - \kappa^{n-1}
743$$
744
745$$ E(\kappa) = \frac{(\kappa/\kappa_e)^4}{[1 + 2.4(\kappa/\kappa_e)^2]^{17/6}} f_\eta f_{\mathrm{cut}} $$
746
747$$
748f_\eta = \exp \left[-(12\kappa /\kappa_\eta)^2 \right], \quad
749f_\mathrm{cut} = \exp \left( - \left [ \frac{4\max(\kappa-0.9\kappa_\mathrm{cut}, 0)}{\kappa_\mathrm{cut}} \right]^3 \right)
750$$
751
752$\kappa_\eta$ represents turbulent dissipation frequency, and is given as $2\pi
753(\nu^3/\varepsilon)^{-1/4}$ with $\nu$ the kinematic viscosity and
754$\varepsilon$ the turbulent dissipation. $\kappa_\mathrm{cut}$ approximates the
755effective cutoff frequency of the mesh (viewing the mesh as a filter on
756solution over $\Omega$) and is given by:
757
758$$
759\kappa_\mathrm{cut} = \frac{2\pi}{ 2\min\{ [\max(h_y, h_z, 0.3h_{\max}) + 0.1 d_w], h_{\max} \} }
760$$
761
762The enforcement of the boundary condition is identical to the blasius inflow;
763it weakly enforces velocity, with the option of weakly enforcing either density
764or temperature using the the `-weakT` flag.
765
766##### Initialization Data Flow
767
768Data flow for initializing function (which creates the context data struct) is
769given below:
770```{mermaid}
771flowchart LR
772    subgraph STGInflow.dat
773    y
774    lt[l_t]
775    eps
776    Rij[R_ij]
777    ubar
778    end
779
780    subgraph STGRand.dat
781    rand[RN Set];
782    end
783
784    subgraph User Input
785    u0[U0];
786    end
787
788    subgraph init[Create Context Function]
789    ke[k_e]
790    N;
791    end
792    lt --Calc-->ke --Calc-->kn
793    y --Calc-->ke
794
795    subgraph context[Context Data]
796    yC[y]
797    randC[RN Set]
798    Cij[C_ij]
799    u0 --Copy--> u0C[U0]
800    kn[k^n];
801    ubarC[ubar]
802    ltC[l_t]
803    epsC[eps]
804    end
805    ubar --Copy--> ubarC;
806    y --Copy--> yC;
807    lt --Copy--> ltC;
808    eps --Copy--> epsC;
809
810    rand --Copy--> randC;
811    rand --> N --Calc--> kn;
812    Rij --Calc--> Cij[C_ij]
813```
814
815This is done once at runtime. The spatially-varying terms are then evaluated at
816each quadrature point on-the-fly, either by interpolation (for $l_t$,
817$\varepsilon$, $C_{ij}$, and $\overline{\bm u}$) or by calculation (for $q^n$).
818
819The `STGInflow.dat` file is a table of values at given distances from the wall.
820These values are then interpolated to a physical location (node or quadrature
821point). It has the following format:
822```
823[Total number of locations] 14
824[d_w] [u_1] [u_2] [u_3] [R_11] [R_22] [R_33] [R_12] [R_13] [R_23] [sclr_1] [sclr_2] [l_t] [eps]
825```
826where each `[  ]` item is a number in scientific notation (ie. `3.1415E0`), and `sclr_1` and
827`sclr_2` are reserved for turbulence modeling variables. They are not used in
828this example.
829
830The `STGRand.dat` file is the table of the random number set, $\{\bm{\sigma}^n,
831\bm{d}^n, \phi^n\}_{n=1}^N$. It has the format:
832```
833[Number of wavemodes] 7
834[d_1] [d_2] [d_3] [phi] [sigma_1] [sigma_2] [sigma_3]
835```
836
837The following table is presented to help clarify the dimensionality of the
838numerous terms in the STG formulation.
839
840| Math                                           | Label    | $f(\bm{x})$?   | $f(n)$?   |
841| -----------------                              | -------- | -------------- | --------- |
842| $ \{\bm{\sigma}^n, \bm{d}^n, \phi^n\}_{n=1}^N$ | RN Set   | No             | Yes       |
843| $\bm{\overline{u}}$                            | ubar     | Yes            | No        |
844| $U_0$                                          | U0       | No             | No        |
845| $l_t$                                          | l_t      | Yes            | No        |
846| $\varepsilon$                                  | eps      | Yes            | No        |
847| $\bm{R}$                                       | R_ij     | Yes            | No        |
848| $\bm{C}$                                       | C_ij     | Yes            | No        |
849| $q^n$                                          | q^n      | Yes            | Yes       |
850| $\{\kappa^n\}_{n=1}^N$                         | k^n      | No             | Yes       |
851| $h_i$                                          | h_i      | Yes            | No        |
852| $d_w$                                          | d_w      | Yes            | No        |
853
854#### Internal Damping Layer (IDL)
855The STG inflow boundary condition creates large amplitude acoustic waves.
856We use an internal damping layer (IDL) to damp them out without disrupting the synthetic structures developing into natural turbulent structures. This implementation was inspired from
857{cite}`shurSTG2014`, but is implemented here as a ramped volumetric forcing
858term, similar to a sponge layer (see 8.4.2.4 in {cite}`colonius2023turbBC` for example). It takes the following form:
859
860$$
861S(\bm{q}) = -\sigma(\bm{x})\left.\frac{\partial \bm{q}}{\partial \bm{Y}}\right\rvert_{\bm{q}} \bm{Y}'
862$$
863
864where $\bm{Y}' = [P - P_\mathrm{ref}, \bm{0}, 0]^T$, and $\sigma(\bm{x})$ is a
865linear ramp starting at `-idl_start` with length `-idl_length` and an amplitude
866of inverse `-idl_decay_rate`. The damping is defined in terms of a pressure-primitive
867anomaly $\bm Y'$ converted to conservative source using $\partial
868\bm{q}/\partial \bm{Y}\rvert_{\bm{q}}$, which is linearized about the current
869flow state. $P_\mathrm{ref}$ is defined via the `-reference_pressure` flag.
870
871### Meshing
872
873The flat plate boundary layer example has custom meshing features to better resolve the flow when using a generated box mesh.
874These meshing features modify the nodal layout of the default, equispaced box mesh and are enabled via `-mesh_transform platemesh`.
875One of those is tilting the top of the domain, allowing for it to be a outflow boundary condition.
876The angle of this tilt is controlled by `-platemesh_top_angle`.
877
878The primary meshing feature is the ability to grade the mesh, providing better
879resolution near the wall. There are two methods to do this; algorithmically, or
880specifying the node locations via a file. Algorithmically, a base node
881distribution is defined at the inlet (assumed to be $\min(x)$) and then
882linearly stretched/squeezed to match the slanted top boundary condition. Nodes
883are placed such that `-platemesh_Ndelta` elements are within
884`-platemesh_refine_height` of the wall. They are placed such that the element
885height matches a geometric growth ratio defined by `-platemesh_growth`. The
886remaining elements are then distributed from `-platemesh_refine_height` to the
887top of the domain linearly in logarithmic space.
888
889Alternatively, a file may be specified containing the locations of each node.
890The file should be newline delimited, with the first line specifying the number
891of points and the rest being the locations of the nodes. The node locations
892used exactly at the inlet (assumed to be $\min(x)$) and linearly
893stretched/squeezed to match the slanted top boundary condition. The file is
894specified via `-platemesh_y_node_locs_path`. If this flag is given an empty
895string, then the algorithmic approach will be performed.
896
897## Taylor-Green Vortex
898
899This problem is really just an initial condition, the [Taylor-Green Vortex](https://en.wikipedia.org/wiki/Taylor%E2%80%93Green_vortex):
900
901$$
902\begin{aligned}
903u &= V_0 \sin(\hat x) \cos(\hat y) \sin(\hat z) \\
904v &= -V_0 \cos(\hat x) \sin(\hat y) \sin(\hat z) \\
905w &= 0 \\
906p &= p_0 + \frac{\rho_0 V_0^2}{16} \left ( \cos(2 \hat x) + \cos(2 \hat y)\right) \left( \cos(2 \hat z) + 2 \right) \\
907\rho &= \frac{p}{R T_0} \\
908\end{aligned}
909$$
910
911where $\hat x = 2 \pi x / L$ for $L$ the length of the domain in that specific direction.
912This coordinate modification is done to transform a given grid onto a domain of $x,y,z \in [0, 2\pi)$.
913
914This initial condition is traditionally given for the incompressible Navier-Stokes equations.
915The reference state is selected using the `-reference_{velocity,pressure,temperature}` flags (Euclidean norm of `-reference_velocity` is used for $V_0$).
916