Repository navigation
feat(torsion): torque-tension coupling of finite-EI cables - #68
Conversation
The helical armour of a power cable or a wire rope couples stretch and twist: T = EA eps + K tau, M = GJ tau + K eps. The 14-column LINE TYPES row's GJ column now takes GJ|K (finite-EI, linear-elastic, K^2 < EA GJ). With the torque still uniform along a line, tau = (M - K eps)/GJ follows the strain pointwise and the condensed torsion stays exact. Two changes carry the coupling: - the cable takes the axial stiffness EA - K^2/GJ of a section free to twist (build_finite_section_mesh); - a line restrained in torsion twists by Theta_eff = Theta - sum (K/GJ)(l_e - L_e). CD_HermiteTorsion_Line gains an optional per-element couple = K/GJ and adds the stretched element length's value, gradient and Hessian, so the statics, the dynamics, the rank-one tangent, the stability test, the host loads and the body equilibrium all take it. The coupling twist has no 2 pi ambiguity, so a first static evaluation unwraps the holonomy alone about its hint. The node and segment tensions add the element's (K/GJ) M, which the kernel's EA - K^2/GJ leaves out. The twist channels, the range rows and CurvX/CurvY sum the element twists M L/GJ - (K/GJ)(l - L) (new CD_HermiteCable_Dyn_Element_Twist). The two-moving-end compatibility route refuses the coupling. test_torsion_coupling: - the kernel's effective-twist gradient and Hessian match finite differences (1.7e-10, 9.8e-11); - a stretched straight cable gives T = EA eps, M = K eps between rigid ends, T = (EA - K^2/GJ) eps with a free end, and T = EA eps + K Phi/L, M = GJ Phi/L + K eps with a 90 deg pretwist, to the printed digits; - a 20 s stretch-and-roll march ends on the static values to 1e-7; - GJ|0 is the uncoupled cable bit for bit; - K^2 >= EA GJ stops by name. The full suite passes and all 52 example decks are bit-identical. The Python deck reader mirrors the grammar and checks, and the builder takes gj=(GJ, K). Signed-off-by: Jae Hoon Seo <jaehoon.seo@inha.ac.kr>
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: a25bf4ce53
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".
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: a25bf4ce53
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".
Under torque-tension coupling the kernel carries EA - K^2/GJ, so its axial resultant is not the section's tension: that adds (K/GJ) M_t. Every place that reads the resultant of a line restrained in torsion now adds it: - the dynamic tensile_safety audit (dyn_step_core): the element means and the lower-bound shortcut take the coupling force at the converged torque (the step's twist is now evaluated, read-only, before the audit); - the final static mesh-safety diagnosis after the torsion stage (CD_HermiteCable_Resolution_Metrics gains an optional per-element axial_offset, also in its smoothed element means and end resultants); - the element recovery (cable_axial_tension: OutputRecovery element node tensions, segment tensions, range rows) and the static element profile's axial resultant extremes, through the new public CD_HermiteCable_Dyn_Coupling_Tension. The OpenFAST aggregate preflight evaluated the twist without the coupling and compared it against the coupled committed twist: a stretched coupled cable (-8 rad of coupling twist) had every update refused by the pi/2 bound. It now passes the coupling as the step does. New gates in torsion_coupling: the element recovery and the element profile read EA eps between rigid ends; a cable stretched by 1e-4 and twisted by -45 deg reads T = EA eps + K Phi/L = -2.14 kN and tensile_safety True refuses it in the statics and in a march rolled there, while the march rolled the other way passes; the coupled aggregate accepts an update at the unchanged boundary. All five fail on the unfixed code. Full suite passes; all 53 example decks are bit-identical. Signed-off-by: Jae Hoon Seo <jaehoon.seo@inha.ac.kr>
|
@codex review |
|
Codex Review: Didn't find any major issues. More of your lovely PRs please. 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". |
The helical armour of a power cable or a wire rope couples stretch and
twist: T = EA eps + K tau, M = GJ tau + K eps. The 14-column LINE TYPES
row's GJ column now takes GJ|K (finite-EI, linear-elastic, K^2 < EA GJ).
With the torque still uniform along a line, tau = (M - K eps)/GJ follows
the strain pointwise and the condensed torsion stays exact. Two changes
carry the coupling:
twist (build_finite_section_mesh);
sum (K/GJ)(l_e - L_e). CD_HermiteTorsion_Line gains an optional
per-element couple = K/GJ and adds the stretched element length's
value, gradient and Hessian, so the statics, the dynamics, the rank-one
tangent, the stability test, the host loads and the body equilibrium
all take it.
The coupling twist has no 2 pi ambiguity, so a first static evaluation
unwraps the holonomy alone about its hint. The node and segment tensions
add the element's (K/GJ) M, which the kernel's EA - K^2/GJ leaves out.
The twist channels, the range rows and CurvX/CurvY sum the element twists
M L/GJ - (K/GJ)(l - L) (new CD_HermiteCable_Dyn_Element_Twist). The
two-moving-end compatibility route refuses the coupling.
test_torsion_coupling:
differences (1.7e-10, 9.8e-11);
ends, T = (EA - K^2/GJ) eps with a free end, and T = EA eps + K Phi/L,
M = GJ Phi/L + K eps with a 90 deg pretwist, to the printed digits;
The full suite passes and all 53 example decks are bit-identical. The
Python deck reader mirrors the grammar and checks, and the builder takes
gj=(GJ, K).