Implement Bernstein's breakdown for the perturbed potential energy#213
Implement Bernstein's breakdown for the perturbed potential energy#213parkjk wants to merge 59 commits into
Conversation
…the develop branch.
…meter and set the wanted number of solution component then this will make Solution{sol_num}.bin file containing the {sol_num}th solution on Wt. If there aren't any sol_num then this will work just as same as before.
…ializes sol_num to 1, allowing for better control of solution components.
| @@ -0,0 +1 @@ | |||
| snu_offshoot No newline at end of file | |||
| bin_bal2=f ! Binary output for bal_flag functions | ||
|
|
||
| netcdf_out=t ! Replicate ascii dcon.out information in a netcdf file | ||
| bernstien_flag=t |
There was a problem hiding this comment.
Please document things as you add features and flags
| ideal_flag=t | ||
| contour_flag=t | ||
|
|
||
| sol_num=1 |
|
|
||
| c----------------------------------------------------------------------- | ||
| c subprogram 2. bernstein_shear. | ||
| c computes local magnetic shear S(ψ, θ) from Eq. (29) of GPEC Notes. |
There was a problem hiding this comment.
Are these notes available to everyone? Can you add them to the repo or cite a publication?
|
@parkjk This branch appears to have completed a clear and finite feature addition. I recommend it be cleaned up no and merged into the develop branch for everyone to enjoy. Not doing so risks sliding into a never ending state of adding features but never finishing them rigorously - which could lead to a permanent split of the SNU code from the rest of the GPEC source. |
… section to documentation
…n with convention
Add recon_flag3 path computing the |C|^2 decomposition of delta-W_p: - gpeq_recon3: A=|Q|^2/mu0, B=2Re(V*.Q)/mu0 (V computed directly from the equilibrium current, not via C-Q), Cpar=K2 xi_n^2, Iperp term, plus optional per-theta surf_extra(0:mthsurf,22) field pack. - gpout_recon3: 30-column surface output (gated by recon_out) and a C-free dW_p = 1/2(A+B+Iperp-dst1-dst3) via gpeq_dst (dst2 cancels). - Unify recon1/2/3 terminal+log output (dW_p, dW_p-normalize, GPEC ep, DCON ep); fix recon1 DCON ep normalization to use dcon_fac. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Add gpec_recon_terms_solX.out (per-point C-form energy densities: c2_den, c2_psi/theta/zeta_den, dst1/2/3_den) from gpout_recon, the recon1 analog of recon3_terms for 2 dW_p = C2/mu0 - dst1 - dst2 - dst3. gpeq_epf / gpeq_dst gain optional per-theta density out-args; their theta loops now run 0..mthsurf with an itheta<mthsurf integration guard so the flux-surface integrals are unchanged. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
main.tex section 4.1: add "Verification: C as an effective field" passage (proof that (curl C).grad psi = 0; Eulerian vs Lagrangian frames; effective current j_eff = curl C / mu0 by analogy with Ampere's law) and Figure 1 (C_eulerian_lagrangian_n1.png, force-added since *.png is gitignored). Also reframe the V-component block as the correction term, drop section 6.1 (dcon_fac moved to the section-6 intro), and rename sections 4.1/4.2. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Add "!docs/tex/recon/C_eulerian_lagrangian_n1.png" so the figure included by docs/tex/recon/main.tex is tracked without `git add -f`. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- A.11: replace the proportionality with an exact equality; expose the 1/(B^2|grad psi|^2) prefactor and that the numerator vanishes by force balance mu0 j x B = mu0 p' grad psi. - A.2 intro: correct which V-products enter H4. Only Y*.V, U*.V, and |V|^2 (with the -mu0 K scalar) form H4; X*.V feeds the C block and Z*.V = mu0 p' xi_psi'^* xi_psi feeds the E block. - Remove redundant "Input Variables and Parameters" table. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Appendix A.1 presented V only through its components A_2,A_3, which read backwards (the components are the *result* of expanding the physical field). Fix the framing: - State up front that V is the edge current rotated into the surface, V = (xi_psi/|grad psi|^2) mu0 j x grad psi -- a concrete (psi,theta)-grid field -- and that A_2,A_3 are just its Jacobian-weighted contravariant components. - Add the scalar-triple-product rule (A.4): W . V = xi_psi (a mu0 j^zeta - b mu0 j^theta) for W = a v_2 + b v_3, i.e. the cofactor identity in coordinate-free form. - Replace the repeated per-product cofactor derivations for X*.V, Z*.V, U*.V with one-line references to (A.4); drop the now-redundant cofactor intermediate equation. No change to results; presentation only. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
gpout_recon: dump gpec_recon_qvmn_sol*.out = covariant Q_i (bare delta-B = bv*_mn, captured right after gpeq_epf before gpeq_K/gpeq_dst). Effective field is cvmn = C_i = Q_i + V_i; cwmn is the contravariant J*C^i. The quantitative QvsC figure built its (curl Q).grad(psi) panel from cwmn (contravariant) fed into the covariant curl operator -> spurious RMS 3.8e-2. Rebuilt from qvmn (covariant Q_i): RMS 2.1e-3 (Q pierces) vs 2.1e-12 (C tangent), matching the C-verify diagnostic; both panels now use covariant components as the caption states. Q-vs-C table + caption updated; figure swapped C_eulerian_lagrangian_n1.png -> QvsC_quant_n1.png (+ gitignore exception). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The xi_flag/xifile namelist entries + declarations were WIP and never wired up (idcon_inject_wt was never called). Removed from gpec.f, gpglobal.F, and input/gpec.in to avoid future conflicts. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
|
@jaesun57, is this branch ready to merge? If not, what is needed and what it the expected timeline? I'd like to not let it live too much longer, as the mission will creep. PR's are best if finite so they do not diverge overmuch from the develop branch. |
|
@logan-nc I'm mainly working on the recon branch right now, and it will take about two more weeks to finish. In the meantime, I'll make sure to keep both recon and snu_offshoot branches updated with the develop branch so they don't diverge too much. As for snu_offshoot, since I didn't create it myself, I'll need to check with @parkjk to see what else needs to be done there. |
|
Thanks for the response @jaesun57. What is the recon branch? Please open a Pull Request for it and make a clear statement of the intent as the opening comment. You can mark the PR and "draft" and that tells people not to bother reviewing the code yet, but it is important to signal what you are doing as it can impact what others are doing (we don't want multiple people accidentally doing the same thing or conflicting things). It looks like you branched off of Similarly, it would be great if you could coordinate with the SNU team to edit the opening comment of this branch with a description of the finite objective, which will clarify whether it is ready to review and merge or not. We don't want infinitely diverging parallel develop branches. Thanks! |
|
Here is AI's summary of the branch: Branch objective (finite)Add a selectable solution-component parameter What it does: Definition of done: It is ready to merge once Questions needing clarification (@parkjk / @jaesun57)
|
Bring 107 develop commits (RDCON/RMATCH resistive work, harvest removal, positive-q sign conventions, install dependency regeneration) into the recon energy-decomposition branch. Conflict resolutions: - gpglobal.F / gpec.f: keep both sides (recon covariant arrays, gpout_* types, recon_* flags + develop bwp_mn_rmatch, bpsi, use_res_spot, singthresh_slayer_inpr_prof) - gpeq.f: keep recon's new subroutines (gpeq_c .. gpeq_recon3); take recon-side subprogram numbering - out_ahg2msc / vac_memory: adopt develop's refactor (moved gpec -> equil module); dropped recon's stale gpec-side copies - solovev gpec.in: take develop's example settings - CLAUDE.md: keep deleted (recon removed it from VCS; local copy retained) Builds clean: make gpec succeeds, gpec executable produced. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
Your points are good and well taken, though this touches on a big picture conversation we should have separately (a bit too much to get into here). Anyway, @jaesun57 has already committed to opening a PR within the next two weeks, so I'd ask for a little patience on that front. Once it's up, she'll follow the practice closely. |
Match input/match.in comment to match/match.f; sol_num=1 selects the most unstable DCON ideal eigenfunction. Addresses snu_offshoot review Q2. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Drop unintended net-diff leftovers from abandoned bernstein experiments: input/dcon.in kin_flag t (was f), dcon/dcon.F xi' comment typo, equil/equil_out.f whitespace. Addresses snu_offshoot review Q3. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
@logan-nc ,All addressed and pushed to Q1 – sol_num: Yes, this is the objective. Q2 – index wording (fixed in 311922c): Standardized on "most unstable" in Q3 – leftovers (reverted to develop in 08ba8bd):
All four leftover diffs vs That said, I still have a few other in-progress changes to wrap up, so I'll do |
|
Thanks @jaesun57 ! @parkjk this looks like a clean feature add rn. Our options are,
Either way, I'm happy to chat about the big picture separately via slack, zoom, etc. or in person Aug 10. |
|
Option 1 sounds good. @jaesun57: Would this be a representing PR name at the moment? "Implement Bernstein's breakdown for the perturbed potential energy" or would you reserve it? Let me know the PR name otherwise and tell me when. Then I will change the name, merge pull, and let you go from there. |
parkjk
changed the title
4 participants
…the develop branch.