Merge fkpl collisions#149

Merged

mrhardman merged 338 commits intomasterfrom merge_fkpl_collisions

Dec 14, 2023

+10,412-2,174

This pull request is big! We're only showing the most recent 250 commits

Commits on Aug 2, 2023

Integration of K(z) from z = 0 to 1.
mrhardman
committed
Add test of integral of K(z)/sqrt(1-z) from z = 0 to 1. This test seems to show a minimum achievable error of ~ 1e-7 that may be due to the limits of available precision.
mrhardman
committed

Commits on Aug 3, 2023

Test subtracting the divergence in poorly converging toy problem.
mrhardman
committed

Commits on Aug 4, 2023

Reverse order of integration dummy coordinate for easier understanding of printout, and addition of L2 norm for diagnostic tests
mrhardman
committed

Commits on Aug 7, 2023

Commits on Aug 8, 2023

Correct integration method to only put divergence-accommodating grid where there is a divergence. The previous method put the divergence accommodating grid wherever ielement_vpa or ielement_vperp w…
mrhardman
committed

Commits on Aug 9, 2023

Commits on Aug 10, 2023

Further test and variable scope fixes.
mrhardman
committed

Commits on Aug 11, 2023

Commits on Aug 14, 2023

Commits on Aug 15, 2023

Commits on Aug 19, 2023

Commits on Aug 24, 2023

Revert quadrature in single point test script to use the same quadrature within a 2D element, for all Lagrange polynomials
Michael Hardman
committed

Commits on Aug 30, 2023

Commits on Aug 31, 2023

Commits on Sep 1, 2023

Plots for diagnosing simulations with the FP collision operator.
Michael Hardman
committed
Plot of vth vs time and plot of the L2 norm of f - f_M to test how the distribution approaches the Maxwellian distribution.
Michael Hardman
committed

Commits on Sep 6, 2023

Commits on Sep 7, 2023

Commits on Oct 1, 2023

Commits on Oct 3, 2023

Changes to 1) make the type of ad-hoc conserving model a string-option input and 2) modify the conserving models to impose the boundary conditions before calculating the moments. This latter change…
Michael Hardman
committed

Commits on Oct 5, 2023

Diagnosis of the symmetric_matrix_inverse function (commented out).
Michael Hardman
committed
Log scale plot of L2 norm of f - f_Maxwellian.
Michael Hardman
committed
Refactor conserving terms application so that the moments are forced to be those of the pdf_out that enters the collision operator routine, not the those of pdf_in (which is the pdf at the start of…
Michael Hardman
committed
Remove commented code.
Michael Hardman
committed

Commits on Oct 6, 2023

Fix broken if statement that ensures that the numerical diffusion is not applied in dimensions of size 1.
Michael Hardman
committed
Upgrade calculus tests to test both matrix and FFT based differentiation, revealing a possible bug.
Michael Hardman
committed

Commits on Oct 9, 2023

Attempt to construct script with 2D matrix assembly
Michael Hardman
committed
Almost working (modulo BC problems) 2D FEM test
Michael Hardman
committed
Make sure that weights are shared memory arrays, correct typo so that logfC does not overwrite a used dummy array. Ensure that a "none" option is supported for conserving terms.
Michael Hardman
committed

Commits on Oct 11, 2023

Development trying to understand BC problems with mass matrix.
Michael Hardman
committed

Commits on Oct 12, 2023

Further iterative changes trying to understand the problem.
Michael Hardman
committed
Cleaned version of 2D FEM mass matrix inversion with d2(.)/dvpa2 and d(.)/dvperp2 test differentiations to confirm a correct implementation.
Michael Hardman
committed
Successful Laplacian solve for H_Maxwellian using exact boundary data.
Michael Hardman
committed
Successful calculation of G via two Laplacian solves using exact boundary data. Addition of test functions to fokker_planck.jl
Michael Hardman
committed
Add test of calculation of d2Gdvpa2 for a Maxwellian distribution.
Michael Hardman
committed

Commits on Oct 16, 2023

Partially successful implementation of calculation of dGdvperp with exact boundary data. Convergence at the expected rate appears to be obtained for ngrid = 2, but for ngrid > 1 a Gibbs phenomena a…
Michael Hardman
committed
Add calculation of d2Gdvperpdvpa by elliptic solve.
Michael Hardman
committed
Introduce dummy array for RHS source in elliptic solve.
Michael Hardman
committed
Added calculation of dHdvpa by elliptic solve using exact boundary data.
Michael Hardman
committed
Added calculation of dHdvperp by elliptic solve using exact boundary data.
Michael Hardman
committed
Move calculatiof 1D elemental matrices to outermost part of assembly loop.
Michael Hardman
committed
Fix bug in KJ matrix for elliptic solve of dGdvperp, dGdvperpdvpa, dHdvperp. These elliptic solves now pass the test.
Michael Hardman
committed
Diagnostic of 1D element matrices in comments.
Michael Hardman
committed
output timing data
Michael Hardman
committed
Added calculation of d2Gdvperp2 = 2 H - d2Gdvpa2 - (1/vperp)dGdvperp, using weak method to deal with (1/vperp) factor.
Michael Hardman
committed
Refactor plotting output to use a generic funciton.
Michael Hardman
committed

Commits on Oct 17, 2023

Commits on Oct 18, 2023

Commits on Oct 19, 2023

Commits on Oct 20, 2023

Commits on Oct 21, 2023

Commits on Oct 23, 2023

Added structs for passing around the resulting errors from the tests.
Michael Hardman
committed
Remembered sqrt() in L2 norm definition and reorder functions so that body of script is in one location.
Michael Hardman
committed
Refactor the script to produce plots of the maximum and L2 norm errors as a function of input nelement.
Michael Hardman
committed
Added facility for testing the timing data for the initialisation of the test arrays and weights, and the time to carry out the collision operator evaluation.
Michael Hardman
committed
Added expected curves to the timing data plot and corrected typo in error data struct. Split plots of errors into overall result (C), and coefficients.
Michael Hardman
committed
Change labels back to using epsilon notation.
Michael Hardman
committed

Commits on Oct 27, 2023

Addition of print statements for error data. Correct typo so that nelement_vpa = 2*nelement_vperp in nelement scan.
Michael Hardman
committed

Commits on Oct 31, 2023

Correct t/cref labels to t/Lref/cref.
Michael Hardman
committed
Add weak-form collision operator to time evolving code. Plausbile behaviour is observed on a single core. MPI issues may be present on runs using multiple cores.
Michael Hardman
committed
Correct diagnostic L2 norm to have the previously missing sqrt().
Michael Hardman
committed
Corrected bugs in the usage of the shared-memory formalism.
Michael Hardman
committed

Commits on Nov 1, 2023

Commits on Nov 2, 2023

Commits on Nov 3, 2023

Commits on Nov 6, 2023

Commits on Nov 7, 2023

Commits on Nov 8, 2023

Commits on Nov 9, 2023

Included ad-hoc numerical conserving terms in "optimized" weak-form Fokker-Planck operator to see if this improves H-theorem properties.
mrhardman
committed

Commits on Nov 10, 2023

Commits on Nov 11, 2023

Changes suggested by @johnomotani to remove boundary conditions from mass matrix inversion when carrying out differentiation, and including explicit boundary terms. Here this is handled with explic…
mrhardman
committed

Commits on Nov 13, 2023

Series of interlinked changes that have the effect of significantly improving the numerical H-theorem property of the weak-form Fokker-Planck operator. The changes are experimental and do not suppo…
mrhardman
committed

Commits on Nov 14, 2023

Commits on Nov 15, 2023

Commits on Nov 16, 2023

Commits on Nov 17, 2023

Addition of a time-evolution test of dF/dt = C[F,F], using data from the cheapest possible relaxation simulation (the input file is also recorded in this commit) for the given initial conditions. T…
mrhardman
committed

Commits on Nov 19, 2023

Two changes: first, using @johnomotani 's suggestion of manually replacing the input Dict with the input Dict constructed from the test input.toml. Second, increase the time step size of the test to…
mrhardman
committed

Commits on Nov 21, 2023

Commits on Nov 22, 2023

Commits on Nov 23, 2023

Commits on Nov 24, 2023

Commits on Nov 25, 2023

Commits on Nov 26, 2023

Commits on Nov 27, 2023

Commits on Nov 29, 2023

Commits on Nov 30, 2023

Support ppari for manufactured solutions test with periodic boundary conditions in z.
mrhardman
committed

Commits on Dec 1, 2023

Commits on Dec 4, 2023

Update src/file_io.jl

mrhardman
and
johnomotani
authored

Commits on Dec 10, 2023

Commits on Dec 11, 2023

Commits on Dec 12, 2023

Commits on Dec 13, 2023