[Gate-users] variance reduction techniques: small field simulations

[G Tom]

I ended up using directional splitting and biasing where the radius is
roughly the area of the opening of the collimator.
I am currently testing the BiasingFactor that gives me reasonable results
compared to no biasing.

Thanks GT



/process/em/setDirectionalSplitting true
/process/em/setDirectionalSplittingTarget 0 0 0 mm # x, y, z components of
center
/process/em/setDirectionalSplittingRadius {SplittingRadius} mm

/process/em/setSecBiasing eBrem world {BiasingFactor} 100 MeV
/process/em/setSecBiasing Rayl world {BiasingFactor} 100 MeV
/process/em/setSecBiasing phot world {BiasingFactor} 100 MeV
/process/em/setSecBiasing compt world {BiasingFactor} 100 MeV
/process/em/setSecBiasing annihil world {BiasingFactor} 100 MeV

On Thu, Feb 25, 2021 at 2:26 PM G Tom <gregthom992 at gmail.com> wrote:

> HI all,
>
> I am trying to improve my simulations with  variance reduction techniques.
>
> The simulation uses particles from a phase space source just above the
> jaws. However, because of the size of the jaws (5cmx5cm opening) and SRS
> cone (10mm) , only a small fraction of particles make it through the jaws
> to the tungsten SRS cone aperture below. Very few photons actually reach
> the dose measurement box below.
>
> What are some ideas to improve the statistics in the dose measurement
> volume ? One way is to choose particles that have a higher probability  of
> reaching that cone opening. How is that implemented in Gate ? Are there
> other VRTs in GATE that will help here?
>
> Below is my setup with the jaws and cone. I show a few photons (green)
> stopped by the upper  jaws before making it to the SRS cone (blue/Grey)
> below.
>
> Thanks a lot
>
>
>
> [image: image.png]
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opengatecollaboration.org/pipermail/gate-users/attachments/20210303/89fb0853/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 6863 bytes
Desc: not available
URL: <http://lists.opengatecollaboration.org/pipermail/gate-users/attachments/20210303/89fb0853/attachment.png>