Repository navigation
feat(viv): cross-flow VIV on the coupled routes - #60
Conversation
MoorDyn-F runs its VIV model inside OpenFAST; CableDyn refused Cl > 0 on the aggregate route (OpenFAST, FAST.Farm and the standalone runner of a mixed EI = 0 / finite-EI deck). The aggregate now installs the lift on both line kinds, and the phase follows the host's fluid field. The in-memory correction rewind already carried the VIV state; the OpenFAST checkpoint now does too: an EI = 0 line ends its span of the state mirror with its phases, running mean squares and averaging time, and a cable's mirror carries the same block before its torsion block. A restart reloads them with their rollback and recovery copies. A riser and a cable side by side in a 0.15 m/s host current lock in at A/D 0.665 at 0.292 Hz (MoorDyn-F 0.662 at 0.293 Hz); a checkpoint rebuild continues bit for bit, and one without the VIV reload does not. Signed-off-by: Jae Hoon Seo <jaehoon.seo@inha.ac.kr>
|
@codex review |
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 39b3322e25
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".
Without VIVrmsWindow an EI = 0 line averages over MoorDyn's 500 steps of the deck dtM, but a cable took 500 steps of the step it was built with, which a coupled host adjusts from the deck dtM to its DT: deck dtM 0.001 s at DT 0.01 s gave 0.5 s and 5 s in one run. The cable now takes the deck dtM too, and the step only when the deck has none. Standalone runs are unchanged (their cables are built at the deck dtM). The options reference now lists the window on every route. Signed-off-by: Jae Hoon Seo <jaehoon.seo@inha.ac.kr>
|
@codex review |
|
Codex Review: Didn't find any major issues. 👍 Reviewed commit: ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
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". |
MoorDyn-F runs its VIV model inside OpenFAST, but CableDyn refused
Cl > 0on the aggregate route. That route serves OpenFAST, FAST.Farm and the standalone runner of a mixedEI = 0/ finite-EI deck. This PR lifts the refusal on all three.Changes
install_viv), as the standalone Hermite march does. ItsEI = 0lines already built it throughinit_one_model. The phase follows the host's fluid field (CD_AGG_SetFluidFields).x%states) now carries the VIV state: the phase and the two running mean squares per node, then the averaging time.EI = 0line ends its span with this block. The block is sized byCD_System_Line_VIV_MirrorSizeand goes throughCD_Get/Set_System_Line_VIV_State, which sit over the model-levelCD_Get/Set_Model_VIV_State.CD_HFMF_VIVMirrorSize).VIVrmsWindow, both line kinds average over MoorDyn's 500 steps of the deckdtM, also when OpenFAST steps CableDyn at adtMadjusted to itsDT. With no deckdtM, they average over 500 coupling steps.CD_Init_Deck_HermiteCableapplies no fluid field, so it still refusesCl > 0. Its message now points to the aggregate.DeckFiledrops the coupled/mixed refusal. On the coupled route it does not ask for water in the deck, since the host supplies it.Checks
viv_coupledtest (22 s). AnEI = 0riser and a finite-EI cable (EI 1 N·m²) sit side by side on the coupled aggregate in a 0.15 m/s host current: each 100 m long at 50 kN, 50 segments.Lock-in over 60–120 s:
With
Cl = 0both lines stay still (A/D below 4e-5).Checkpoint restart, using the shell's rebuild recipe: it continues bit for bit. A rebuild without the VIV reload diverges.
A correction rewind repeats bit for bit.
Averaging window: deck
dtM0.001 s at a 0.01 s coupling step gives both line kinds 0.5 s; with no deckdtM, both get 5 s.CompMooring = 5. This ran on the IEA-15 MW mixed deck withCl 0.8on the lazy-wave cable, a SeaState 0.5 m/s current across it, no waves, 60 s. It was checkpointed at 30 s and restarted.Cl = 0run.Cl > 0on these routes.Remaining VIV refusals: a finite-EI line beside bodies, rods or free points (next PR), and a finite-EI line with two moving ends (the Cosserat compatibility element, which has no VIV).