Skip to content

fix(statics): smear the attachment drag in a cross current - #110

Merged
SMI-Lab-Inha merged 2 commits into
devfrom
fix/attachment-polish-current
Nov 8, 2026
Merged

SMI-Lab-Inha merged 2 commits into
devfrom
fix/attachment-polish-current

Conversation

@SMI-Lab-Inha

Copy link
Copy Markdown
Owner

In a current across its plane, on a frictional seabed, the static solve of a finite-EI line with ATTACHMENTS reached the discrete attachment state only through the polish fallbacks (pseudo-transient continuation, line search, halved steps), and in some cases failed. One example is a lazy wave with 25 modules in 0.3 m/s across its plane: its first polish step stalled at load fraction 0.125 after 200 Newton iterations.

Cause

  • The smeared start: it spreads each attachment's net weight over the elements beside its node, but it did not spread the attachment's current drag.
  • The polish: it then applied the whole module drag to a state without it, about 1.6 kN lateral against 2.6 kN of cable drag. The grounded run slid metres sideways through the sharply regularised friction, and the damped Newton solves diverged.

Fix

  • smear_attachment_drag: in a current with a component across the chord's vertical plane, the attachments' CdA and CdAx are added to the elements' drag coefficients, using the line's own Morison law and the same split as the weight.
  • Polish blending: attachment_static_polish blends those coefficients back to the bare ones with the load fraction, as it does the weights.
  • In-plane currents: these keep the former path. Smearing there stopped a torsion-restrained lazy wave on a coarse mesh at the grounded-run compression check, while the former path had no fallbacks anywhere in the sweep.
  • Tolerances: none changed.
  • Deck syntax: no change.

Evidence

The sweep covers 351 decks: speed 0–0.5 m/s × heading 0/45/90° × 10/25/40 modules × 1/2/4× refinement, over three deck families.

family before OK/fallback/fail after
lazy_wave_buoyancy_modules.dat (frictionless) 117/0/0 117/0/0
frictional touchdown deck, 4 m/10 m grading 77/32/8 111/0/6
the same deck on 2 m elements 60/57/0 117/0/0
  • Six remaining failures: these predate this PR and stop in the smeared stage on a strongly graded mesh, not in the polish. They are a separate item, followed up separately.
  • Equilibrium: on 4× reference meshes the end forces are identical to the printed digits. Mid-water force balance holds to 3e-8–2e-7 of the largest load, and the out-of-plane deflection follows the current.
  • Examples: all 54 examples/*.dat are byte-identical in every result file.

Tests

  • New attachment_polish_current: 16 frictional decks plus the torsion deck at 45°, all requiring no fallback note, and 4 force-balance checks.
    • Before the fix: fails 15 gates.
    • After the fix: passes in 12 s Release, 55 s Debug.
  • attachment_cross_current: now requires no fallback on any case. Previously it required the fallback note.
  • Full CTest: Release 366/366 on the default and Prescott kernels; Debug for the attachment tests.

@SMI-Lab-Inha

Copy link
Copy Markdown
Owner Author

@codex review

@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Nov 8, 2026 •

Copy link
Copy Markdown
Codex Review Summary: ✅ Completed

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-11-08T10:35:55.329362Z f9d2523 Manual request
ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

@read-the-docs-community

read-the-docs-community Bot commented Nov 8, 2026 •

Copy link
Copy Markdown

Documentation build overview

📚 CableDyn | 🛠️ Build #35731451 | 📁 Comparing f9d2523 against latest (e8f88f5)

  🔍 Preview build  

5 files changed · ± 5 modified

± Modified

@chatgpt-codex-connector

Copy link
Copy Markdown

Codex Review: Didn't find any major issues. Already looking forward to the next diff.

Reviewed commit: f027e3b6b6

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

The static solve of a finite-EI line with ATTACHMENTS finds the equilibrium
with each attachment's net weight smeared over the elements beside its node,
then moves the loads to the node in the attachment polish. The smeared start
carried the weight but not the attachments' current drag, so the polish applied
the whole drag to a state without it. In a current across the line's vertical
plane that drag pushes the line out of the plane; on a frictional seabed the
grounded run slid metres sideways and the damped Newton solves diverged from
the first load step, so the discrete state was reached only through the
pseudo-transient and line-search fallbacks, or not at all (a lazy wave with 25
modules on 2 m elements in 0.3 m/s across its plane; 89 of 234 variants of
the frictional lazy wave over current speed, heading, module count and mesh
refinement fell back, and a lazy wave restrained in torsion in 0.3 m/s at
45 deg stopped).

In a current across the plane the attachments' drag areas are now smeared over
the same elements as their weight (per-metre drag coefficients of those
elements), and the polish moves the smeared drag to the nodes with the weight.
Its load steps then move the line by millimetres and converge in a few Newton
iterations; refined reference meshes give the same equilibrium to the printed
digits. A current along the plane keeps the former path, which the polish
handles without fallbacks, and every in-plane or still-water deck is
bit-identical.

A new test, attachment_polish_current, sweeps speed, heading, module count and
refinement on the frictional lazy wave (no fallback note, deflection toward the
current), solves the lazy wave restrained in torsion at 45 deg, and checks the
global force balance of a mid-water lazy wave through the C API.
attachment_cross_current now requires the refined deck to converge without its
fallbacks.

Signed-off-by: Jae Hoon Seo <jaehoon.seo@inha.ac.kr>
@SMI-Lab-Inha

Copy link
Copy Markdown
Owner Author

@codex review

@SMI-Lab-Inha
SMI-Lab-Inha force-pushed the fix/attachment-polish-current branch from f027e3b to 52bd7d4 Compare November 8, 2026 09:11
@chatgpt-codex-connector

Copy link
Copy Markdown

Codex Review: Didn't find any major issues. You're on a roll.

Reviewed commit: 52bd7d4cc6

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

The [Unreleased] entries of the open changes are added together in one
change, so that they do not conflict with each other.

Signed-off-by: Jae Hoon Seo <jaehoon.seo@inha.ac.kr>
@SMI-Lab-Inha

Copy link
Copy Markdown
Owner Author

@codex review

@chatgpt-codex-connector

Copy link
Copy Markdown

Codex Review: Didn't find any major issues. Hooray!

Reviewed commit: f9d2523328

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

@SMI-Lab-Inha
SMI-Lab-Inha merged commit 7e3938c into dev Nov 8, 2026
29 of 32 checks passed
@SMI-Lab-Inha
SMI-Lab-Inha deleted the fix/attachment-polish-current branch November 9, 2026 08:55
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant