Penalty "boundary conditions"

A BC which appears as a/forcing": the mini version. More...

Files
file	penalty_mini.f
	Penalty for ABL user code of nek5000.

Functions/Subroutines
subroutine	pen_update ()
	Update penalty. More...
subroutine	pen_forcing (ix, iy, iz, ieg)
	Compute penalty forcing. More...
subroutine	pen_fmask_get ()
	Get 1D projection, array mapping and forcing smoothing. More...
subroutine	pen_fterms_get (ifreset)
	Generate forcing terms: pen_k_len and pen_fstab. More...
subroutine	pen_register ()
	Register penalty module. More...
subroutine	pen_init ()
	Initilise penalty module. More...
logical function	pen_is_initialised ()
	Check if module was initialised. More...
subroutine	local_cart_dy (us, u, N, e, Dt)
	Assuming the mesh is cartesian, compute exact y-derivative for an element, by operating. More...
subroutine	local_cart_dy_pseudo (us, u, N, e, Dt)
	Assuming the mesh is cartesian, compute pseudo(?) y-derivative along y-axis for an element, by operating. More...
subroutine	cart_diff_y (uy, u, ifexact)

Detailed Description

A BC which appears as a/forcing": the mini version.

Function/Subroutine Documentation

cart_diff_y()

subroutine cart_diff_y	(	real, dimension(lxyz,1)	uy,
		real, dimension(lxyz,1)	u,
		logical	ifexact
	)

Compute exact / pseudo y-derivative in a Cartesian vel. mesh

Parameters

[out]	uy
[in]	u
[in]	ifexact	Compute exact or pseudo-derivative

Here is the call graph for this function:

local_cart_dy()

subroutine local_cart_dy	(	real, dimension(0:n,0:n,0:n)	us,
		real, dimension (0:n,0:n,0:n,1)	u,
		integer	N,
		integer	e,
		real, dimension(0:n,0:n)	Dt
	)

Assuming the mesh is cartesian, compute exact y-derivative for an element, by operating.

\[ \mathbf{u} D^T \]

Parameters

[out]	us
[in]	u,N,e,Dt

Here is the call graph for this function:

local_cart_dy_pseudo()

subroutine local_cart_dy_pseudo	(	real, dimension(0:n,0:n,0:n)	us,
		real, dimension (0:n,0:n,0:n,1)	u,
		integer	N,
		integer	e,
		real, dimension(0:n,0:n)	Dt
	)

Assuming the mesh is cartesian, compute pseudo(?) y-derivative along y-axis for an element, by operating.

\[ Dt \mathbf{u} \]

Parameters

[out]	us
[in]	u,N,e,Dt

Here is the call graph for this function:

pen_fmask_get()

subroutine pen_fmask_get

Get 1D projection, array mapping and forcing smoothing.

This routine is just a simple version supporting only lines paralles to z axis. In future it can be generalised. The subroutine initializes pen_prj, pen_map and pen_npoint

See also

Schlatter and Örlü, “Turbulent Boundary Layers at Moderate Reynolds Numbers.” pg. 12

Remarks

This routine uses global scratch space CTMP0 and CTMP1

pen_forcing()

subroutine pen_forcing	(	integer	ix,
		integer	iy,
		integer	iz,
		integer	ieg
	)

Compute penalty forcing.

Parameters

[in,out]	ffx,ffy,ffz	forcing; x,y,z component
[in]	ix,iy,iz	GLL point index
[in]	ieg	global element number

Todo:

add ffy component and rotation

Here is the call graph for this function:

Here is the caller graph for this function:

pen_fterms_get()

subroutine pen_fterms_get

(

logical

ifreset

)

Generate forcing terms: pen_k_len and pen_fstab.

This used to also initializes the pen_famp array. This was to allow for spatio-temporally varying forcing, which has been removed in this version.

Parameters

[in]

ifreset

reset flag to reinitialize geometry term pen_k_len

Here is the call graph for this function:

Here is the caller graph for this function:

pen_init()

subroutine pen_init

Initilise penalty module.

Note

This routine should be called in frame_usr_init

pen_is_initialised()

logical function pen_is_initialised

Check if module was initialised.

Returns

pen_is_initialised

pen_register()

subroutine pen_register

Register penalty module.

Note

This routine should be called in frame_usr_register

Here is the call graph for this function:

pen_update()

subroutine pen_update

Update penalty.