Simulation object has no attribute fields

[Aratopn]

Hello,

I am trying to simulate T.Charoy et al benchmark (DOI 10.1088/1361-6595/ab46c5).
Among other things, it involves recalculating the potential after solving the Poisson equation, taking into account the emission plane. I was planning to use the sim.get.fields() function for PICMI, but I encountered an error:
AttributeError: 'Simulation' object has no attribute 'fields'

I also tried to run the test Examples/Tests/embedded_boundary_python_api but I encountered the same error.

Did I miss something during the build? Or do the fields now need to be accessed in some other way?

Thank you for your time.

[RemiLehe]

Hi @Aratopn,

Thanks for reaching out.
The sim.fields attribute is a very recent change in WarpX, and I think that it is actually not yet in latest release. (It should be in the December release though.)
For now, if you want to use this feature: could you make sure that you are compiling WarpX from source, using the latest version of the development branch from Github?

[Aratopn]

Hi @RemiLehe,

Thank you for pointing out the problem. The problem was indeed the warpx version. I was able to access the fields using: sim.extension.warpx.multifab("phi_fp", level=0)[...].

I would like to ask for advice on this. The mentioned benchmark uses the following approach:

1. First, the Poisson equation is solved with the boundary conditions $U(0,y)=200V$ and $U(Lx,y)=0V$.
2. Next, the potential on the emission plane $\overline{U_e}$, averaged in the $y$ direction, is determined.
3. Then, the potential distribution is recalculated according to the expression: $\phi(x,y)'=\phi(x,y)-\frac{x}{x_e}\overline{U_e}$.

Here, $x_e$ is the emission plane position.

I implemented this using a callback function:

def updatePhi():
    phi = sim.extension.warpx.multifab("phi_fp", level=0)[...]
    phi_emi_mean = np.mean(phi[i_emi])
    sim.extension.warpx.multifab("phi_fp", level=0)[...] -= (X_grid / x_emi) * phi_emi_mean

"""
some code
"""

callbacks.installbeforeEsolve(updatePhi)

X_grid is precalculated numpy meshgrid with x positions for all nodes.
x_emi is the emission plane position.
i_emi is the index corresponding to x_emi.

I'm not very good at measuring computational efficiency, but it seems to me this approach could significantly slow down the simulation. Perhaps you can recommend a more efficient solution.

Thank you for your time.

[RemiLehe]

Hi @Aratopn,

This approach seems fine to me, in terms of performance.
If you are running on GPU, you might want to use xp, _ = load_cupy() instead of np, as done here:
https://github.com/BLAST-WarpX/warpx/blob/development/Examples/Physics_applications/capacitive_discharge/inputs_base_1d_picmi.py#L415
In addition, instead of using -=, you can use Saxpy, as used here:
https://github.com/BLAST-WarpX/warpx/blob/development/Examples/Physics_applications/spacecraft_charging/inputs_test_rz_spacecraft_charging_picmi.py#L54

You can always check whether the callback function takes a significant amount of time by looking at the table of timing at the end of a simulation. (The time spent in callback functions is under warpx_py_.)