Question of SmoothIndicator and AnalyticalIdentity

[Bobbie]

Dear community

When I’m learning multi-phase and multi-component examples, I found SmoothIndicator and AnalyticalIdentity are used together to achieve the distinction of different lattice densities (such as the following codes in the case of microFluidcs2d). How is this achieved?

AnalyticalConst2D<T,T> zero ( 0. );
AnalyticalConst2D<T,T> one ( 1. );
IndicatorCuboid2D<T> ind1(xl1+dx, ny, {xl1/T(2), ny/T(2)});
SmoothIndicatorCuboid2D<T,T> section1( ind1, 0. );
IndicatorCuboid2D<T> ind2(xl2 + xl3, ny, {xl1 + (xl2 + xl3)/T(2), ny/T(2)});
SmoothIndicatorCuboid2D<T,T> section2( ind2, 0. );
AnalyticalIdentity2D<T,T> c1( section1 );
AnalyticalIdentity2D<T,T> c2( section2 );
AnalyticalIdentity2D<T,T> rho( one );
AnalyticalIdentity2D<T,T> phi( c1 – c2 );
AnalyticalIdentity2D<T,T> psi( rho – c1 – c2 );

In openlb src, I have found the detailed description of SmoothIndicatorCuboid2D, which implements a smooth cuboid in 2D with an _epsilon sector. But I’m confused whether this SmoothIndicatorCuboid2D is referring to this cuboid region or something else?

[TimBingert]

Hi Bobbie,

all the SmoothIndicator* are (in this particular case) scalar fields that can be used on any lattice to initialize all kinds of variables in your simulation. The AnalyticalIdentity operator is able to perform certain calculations between those fields such as addition, subtraction and multiplication. This is also what you can see in the code of your post.

Hope this helps
Kind regards
Tim

[Author]

Posts