haoyuanli93 / xraysimulation

This is my new simulation repo.

Warning:

My x-y-z coordinate convention is different from the SLAC convention!

In my simulation, the z axis is the beam propagation direction.

The y axis is the horizontal transverse direction. This the the x axis in SLAC convension.

The x axis is the vertical transverse direction. This is the y axis in SLAC convension.

XRaySimulation

This is my new simulation repo.

The reason that I am abandoning the previous repo is that the previous one is too complicated and contains too much information and too many functions.

I need to simplify the repo and make it more maintainable.

In this new repo, no effort has been made to include Laue and forward Bragg diffraction. Later in the future, I may include those case, however.

Essential To-Do list

1. Add unit test module
2. Release a stable version.
3. Add functions to calculate quantities with prism
4. The one thing that I am most unsatisfied about this simulation is that different functions are not in most appropriate position. This makes it not easy to start a new simulation. I want to re-arrange the functions to make the location of each function more reasonable.

Not so essential to-do list

1. Change the auto-alignment behavior
2. Change the crystal property grab behavior.

Some thoughts

I am currently using a lot of Kohzu motors in the simulation. However, I am not sure if I want to implement sub classes of kohzu motor models or not. The good thing is that I can shorten the code. The bad thing is that I need to create a lot of layers of code.

The most important feature of this repo is that I believe it is more readable than those well-developed repos. I will choose the one that can perserve this.

Notice

Optimization

In this repo, not all the functions are optimized. Functions in the GPU folder are optimized to some extend. Functions in the MultiDevice.py file are not optimized.

Fresnal diffraction

Previously, I have explicitly considered the Fresnel diffraction. However, I just realized that I do not need such things because in my simulation, the propagation of each monochromatic components follow the more universal Maxwell equations.

Therefore, in this version, I remove the Fresnel diffraction functions.

Later, I might add a function to handle a pure propagation directly.