Add calculation of force error with anisotropic pressure tensor #395
Conversation
Codecov Report
Additional details and impacted files@@ Coverage Diff @@
## master #395 +/- ##
==========================================
+ Coverage 94.63% 94.65% +0.01%
==========================================
Files 79 79
Lines 18756 18872 +116
==========================================
+ Hits 17750 17863 +113
- Misses 1006 1009 +3
|
f0uriest
changed the title
| @@ -1303,4 +1303,5 @@ def body(i, mins): | |||
| "electron_density": "ne_l", | |||
| "ion_temperature": "Ti_l", | |||
| "atomic_number": "Zeff_l", | |||
| "anisotropy": "a_lmn", | |||
There was a problem hiding this comment.
So to do anisotropy, you have to give a regular pressure profile + a new anisotropy profile? I thought we could do this by just generalizing the pressure profile (something like p_l would become p_lmn)
There was a problem hiding this comment.
that would be anisotropic scalar pressure (ie, scalar pressure but a function of r,t,z). What we actually want is an anisotropic pressure tensor, hence the need for the new profile to represent the difference between the parallel and perpendicular pressures (see the notes linked above for a full description)
| @@ -62,6 +63,27 @@ def test_SOLOVEV_results(SOLOVEV): | |||
| np.testing.assert_allclose(theta_err, 0, atol=1e-4) | |||
|
|
|||
|
|
|||
| @pytest.mark.regression | |||
| @pytest.mark.solve | |||
| def test_SOLOVEV_anisotropic_results(SOLOVEV): | |||
There was a problem hiding this comment.
This is good, but it doesn't actually test that solving with non-zero anisotropy works correctly. Do we have any benchmark cases we could use for that?
There was a problem hiding this comment.
Not really. Frank and I are working on comparing to near axis stuff but still in progress.
Resolves #312