Fast python-based yields simulation technique for producing LHe yields
This is an early developmental version of a yields simulation package. Further development will continue.
In the terminal, simply use:
pip install HeST==0.2.0
Then in your python scripts you can use:
from HeST import HeST
which will get you all of the functions in HeST
You'll need to clone the code from git:
git clone [email protected]:spice-herald/HeST.git
Then from within the newly cloned "HeST" directory, install with:
pip install .
This is recommended simply so that the user has access to the source files, see how things are defined, and make custom changes.
A jupyter notebook, "ExampleUsage.ipynb" has been included for an example of how to generate events.
Detector geometries require light collection efficiency (LCE) and QP evaporation (QPE) maps to speed up the yields generation. Example detector geometry (Amherst and LBNL designs) have been included. The "map_generation" directory has example scripts for how these maps were generated. It's recommended to run these in an environment where you can submit jobs with many nodes, such as NERSC or Great Lakes.
To run these scripts, verify that LCEmap_2DmapFromZ.py and QPEmap_2DmapFromZ.py are using the proper detector geometry, and the X,Y binning is
set properly.
Then, make sure that the script processMapsInParallel.py uses the right Z binning. Running
python processMapsInParallel.py LCEmap_2DmapFromZ
will submit one 2D map-making job for each z-bin. You can then use the script Merge2DMaps.py to combine the parallel outputs into a single numpy array.
If you need help or have suggestions, reach out to me, Greg Rischbieter via slack or at [email protected]
So currently, I'm focusing purely on the quasiparticle modeling, as this is the most important area of development, and the most interesting physics that we can get from this sim.
The question is how does this whole thing work: Everything is an array. Each particle exists as an index at every array. alive: determines whether or not the particle is alive (1 if it is, 0 if it is dead). X, Y, Z: position X1, Y1, Z1: Position of intersection with wall. You can think of the sim as constantly calculating intersections with walls, so we always need a starting and ending location.
dx, dy, dz: These are the direction vectors.
Velocity: velocity, momentum: momentum, energy: energy.
Everything is one-to-one.
living_indices: the indices of each living particle.
Each particle has all of it's attributes in this chain of arrays, all of which are connected via this 1-1 nature. We let the sim run until the sum of the alive array is 0, meaning that all quasiparticles are dead.