Skip to content

fix(statics): the attachment polish accepts the line's residual floor - #124

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

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

Conversation

@SMI-Lab-Inha

Copy link
Copy Markdown
Owner

A finite-EI line with discrete ATTACHMENTS failed its static solve with "||R||=5e-8 > tol=1e-8; homotopy stopped" in 12 decks of the robustness campaign. In one of them the polish failed at a residual seven times lower than the 2.9e-7 at which the smeared start had already been accepted.

Cause

The line's own static solve accepts a residual floor at the coupled tolerance of 1e-6. The attachment polish solved every load step only at the tight tolerance, and its fallbacks stall on the same floor. Its load steps therefore halved until the run failed.

Fix

When a polish solve stalls, it is retried from the same state at the coupled tolerance before the fallbacks run. The run then prints a note giving the residual reached. Solves that meet the tight tolerance, and coupled runs, are unchanged.

Evidence

  • New attachment_polish_floor: a cable clamped in a tight bend at its anchor, carrying one clump, over a grid of mass and arc position. Before the fix all 9 grid points fail. After it they converge with the floor note and balance the weights to 1e-5.
  • Suites: CTest 368/368 on both kernels; pytest passes.
  • Examples: all 54 are identical.
  • Campaign decks: with the other two fixes, all 12 now converge.

The CHANGELOG entry goes into the batch CHANGELOG PR.

The finite-EI static solve of a line accepts a residual floor above the tight
static tolerance at the coupled static tolerance 1e-6 and notes it. The
discrete-attachment polish that follows it solved every load step at the
tight tolerance only: a solve that stalled at a few times 1e-8 (a fine or
locally refined mesh, a long line) went on to the pseudo-transient and
merit fallbacks, which stall on the same floor, and the load steps were
halved down to 1/1024 until the static start failed. A suspended cable
clamped at its anchor in a tight bend with one clump failed for every clump
mass and arc position tried, the polish stalling at 3e-8 to 6e-8 where the
smeared equilibrium had been accepted at 2.9e-7.

A polish solve whose Newton iteration stalls above the tight tolerance is
now repeated from the same state at the coupled static tolerance before the
next update rule, as the line's own solve is, and the run notes the residual
of the discrete equilibrium when it lies above the tight tolerance. Solves
that meet the tight tolerance are unchanged, and a coupled run, whose static
tolerance is already 1e-6, takes the same solves as before.

The new attachment_polish_floor test runs the clamped cable with a 400 or
1600 kg clump at 450 or 760 m and requires each run to converge without a
fallback note, its end forces balancing the line and clump weights to 1e-5.

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

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 2026-11-08T15:20:28.096986Z 58b9f78 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

Copy link
Copy Markdown

Documentation build overview

📚 CableDyn | 🛠️ Build #35735091 | 📁 Comparing 58b9f78 against latest (e8f88f5)

  🔍 Preview build  

10 files changed · ± 10 modified

± Modified

@chatgpt-codex-connector

Copy link
Copy Markdown

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

Reviewed commit: 58b9f78c0f

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 a045075 into dev Nov 8, 2026
22 checks passed
@SMI-Lab-Inha
SMI-Lab-Inha deleted the fix/attachment-polish-floor 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