Mirror.cl File Reference

Mirroring process for the symmetry boundary condition. More...

#include "resources/Scripts/types/types.h"

Include dependency graph for Mirror.cl:

Functions
__kernel void	drop (__global int *imove, __global vec *r, unsigned int N, vec symmetry_r, vec symmetry_n, vec domain_max)
	Remove the particles at the other side of the mirror.
__kernel void	detect (const __global int *imove, const __global vec *r_in, __global unsigned int *imirror, unsigned int N, vec symmetry_r, vec symmetry_n)
	Detect the particles to be mirrored.
vec_xyz	reflection (vec_xyz u, vec_xyz n)
__kernel void	feed (__global int *imove, __global int *iset, __global unsigned int *imirror, __global unsigned int *imirror_invperm, __global float *m, __global vec *normal, __global vec *tangent, __global vec *r_in, __global vec *u_in, __global vec *dudt_in, __global float *rho_in, __global float *drhodt_in, unsigned int N, unsigned int nbuffer, vec symmetry_r, vec symmetry_n)
	Mirror the particles marked with a flag imirror = 1.

Detailed Description

Mirroring process for the symmetry boundary condition.

Function Documentation

detect()

__kernel void detect	(	const __global int *	imove,
		const __global vec *	r_in,
		__global unsigned int *	imirror,
		unsigned int	N,
		vec	symmetry_r,
		vec	symmetry_n
	)

Detect the particles to be mirrored.

The mirroring particles (the ones close enough to the symmetry plane) will be marked with imirror = 1.

Parameters

imove	Moving flags. imove > 0 for regular fluid particles. imove = 0 for sensors. imove < 0 for boundary elements/particles.
r_in	Position \( \mathbf{r} \).
imirror	0 if the particle has not been mirrored, 1 otherwise.
N	Number of particles.
symmetry_r	Position of the symmetry plane.
symmetry_n	Normal of the symmetry plane. It is assumed as normalized.

drop()

__kernel void drop	(	__global int *	imove,
		__global vec *	r,
		unsigned int	N,
		vec	symmetry_r,
		vec	symmetry_n,
		vec	domain_max
	)

Remove the particles at the other side of the mirror.

Parameters

imove	Moving flags. imove > 0 for regular fluid particles. imove = 0 for sensors. imove < 0 for boundary elements/particles.
r	Position \( \mathbf{r} \).
N	Number of particles.
symmetry_r	Position of the symmetry plane.
symmetry_n	Normal of the symmetry plane. It is assumed as normalized.
domain_max	Top-right-back corner of the computational domain.

feed()

__kernel void feed	(	__global int *	imove,
		__global int *	iset,
		__global unsigned int *	imirror,
		__global unsigned int *	imirror_invperm,
		__global float *	m,
		__global vec *	normal,
		__global vec *	tangent,
		__global vec *	r_in,
		__global vec *	u_in,
		__global vec *	dudt_in,
		__global float *	rho_in,
		__global float *	drhodt_in,
		unsigned int	N,
		unsigned int	nbuffer,
		vec	symmetry_r,
		vec	symmetry_n
	)

Mirror the particles marked with a flag imirror = 1.

Parameters

imove	Moving flags. imove > 0 for regular fluid/solid particles. imove = 0 for sensors. imove < 0 for boundary elements/particles.
iset	Index of the set of particles.
imirror	0 if the particle should not be mirrored, 1 otherwise.
imirror_invperm	Permutation to find the index of the particle in the list of particles to become split.
m	Mass, \( m \).
normal	Normal, \( \mathbf{n} \).
tangent	Tangent, \( \mathbf{t} \).
r_in	Position \( \mathbf{r} \).
u_in	Velocity \( \mathbf{u} \).
dudt_in	Velocity rate of change \( \frac{d \mathbf{u}}{d t} \).
rho_in	Density \( \rho \).
drhodt_in	Density rate of change \( \frac{d \rho}{d t} \).
N	Number of particles.
nbuffer	Number of available buffer particles.
symmetry_r	Position of the symmetry plane.
symmetry_n	Normal of the symmetry plane. It is assumed as normalized.

Here is the call graph for this function:

reflection()

vec_xyz reflection	(	vec_xyz	u,
		vec_xyz	n
	)

Reflection vector.

The deflection vector is defined as follows:

\( \mathbf{v}(\mathbf{u}, \mathbf{n}) = -2 \left( \mathbf{u} \cdot \mathbf{n} \right) \mathbf{n} \)

where \(\mathbf{u}\) is the vector to become deflected, \(\mathbf{n}\) is the reflection plane normal, and \(\mathbf{v}\) is the deflection vector, which added to the original vector returns its reflected version.

Note

The input vector should be relative to the symmetry plane. That's important in case of position vectors, from which an arbitrary point of the plane should be substracted