Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion CMakeLists.txt
Original file line number Diff line number Diff line change
Expand Up @@ -2747,7 +2747,8 @@ if(Python3_Interpreter_FOUND)
LABELS "python;driver;integration;static"
ENVIRONMENT "${_cabledyn_python_test_environment}")
# A hang-off through a tapered bend stiffener follows the planar elastica (scipy solve_bvp)
# within 2 % of the root moment and 1 cm; skipped where scipy is not installed.
# within 2 % of the root moment and 1 cm, and node by node on a sweep of element sizes within
# 0.5 % of the local curvature on 0.1 m elements; skipped where scipy is not installed.
add_test(NAME driver_bend_stiffener
COMMAND ${Python3_EXECUTABLE} ${CMAKE_CURRENT_SOURCE_DIR}/tests/check_bend_stiffener.py
$<TARGET_FILE:cabledyn>)
Expand Down
14 changes: 10 additions & 4 deletions VALIDATION.md
Original file line number Diff line number Diff line change
Expand Up @@ -78,7 +78,7 @@ article's methods, are compared in
| **L3-6d** | Moving-touchdown dynamic contact | OrcaFlex 11.6d installed | curv 0.48/0.58%, tension 0.74/0.90/0.80%, TDP 0.08 m, migration 2–3 m/cycle | curv <3–5%, tension <2–3%, TDP excursion bounds <1.5 m | Pass |
| **L3-6e** | Out-of-plane / 3D whirl response | OrcaFlex 11.6d | 3D curv 0.52%, interior Y env ≤1.0 mm, sag ≤22 mm, tension 0.14/0.17/0.12% | curv <3%, env <5 cm, tension <2–3% | Pass |
| **L3-7** | 1-hour VolturnUS-S mooring dynamic | OrcaFlex 11.6d | `FairTen` mean 0.27%, std 0.26%, min/max 0.50/0.43%, Fourier 0.27/1.14% | ≤5% | Pass |
| **L3-8** | Bend stiffener at a clamped hang-off (5 m tapered polyurethane) | OrcaFlex 11.6d (stiffener attachment) / planar elastica | curvature 0.74% of root; cable moment 0.74%, section moment 0.18%, deflection angle 0.003° | <1.5% / <2%, 0.05° | Pass |
| **L3-8** | Bend stiffener at a clamped hang-off (5 m tapered polyurethane) | OrcaFlex 11.6d (stiffener attachment) / planar elastica | curvature 0.51% of root; cable moment 0.18%, section moment 0.18%, deflection angle 0.003°; local curvature 0.24% (0.1 m elements) | <1.5% / <2%, 0.05°, <0.5% | Pass |
| **L3-H** | Holcombe static lazy-wave experiment | measured marker geometry | position RMSE 1.07–1.89% of suspended length; peak marker-curvature error 0.34–5.54% | end-restraint bracket | Pass |
| **L3-B** | Bergdahl chain experiment, 30 orbit conditions | measured upper-end force | RMSE 1.58 N; MAPE 3.11%; Pearson 0.993; Lin concordance 0.989 | full published matrix | Pass |
| **L-fig** | Along-arc curvature overlay (CableDyn / OrcaFlex / MoorDyn-F ±4×) | 3-code committed record | CableDyn peak 0.11% from OrcaFlex | <0.5% | Pass |
Expand Down Expand Up @@ -275,13 +275,19 @@ initialisation.
polyurethane stiffener (0.5 m cylinder of 0.60 m, cone to 0.22 m, bore 0.20 m, E 50 MPa;
16 times the cable's EI at the root), on 0.05 m elements. The static curvature over the
stiffener and 3 m beyond follows OrcaFlex 11.6d with the stiffener given as a line attachment
within **0.74% of the root curvature** (root 0.4523 against 0.4522 1/m;
within **0.51% of the root curvature** (root 0.4523 against 0.4522 1/m;
[`orcaflex_bend_stiffener.py`](https://github.com/SMI-Lab-Inha/CableDyn/blob/main/validation/scripts/orcaflex_bend_stiffener.py)).
Against an independent collocation solution of the planar extensible elastica with
`EI(s)` and the sleeve's weight, the cable's own moment follows within **0.74%** of its root
`EI(s)` and the sleeve's weight, the cable's own moment follows within **0.18%** of its root
value, cable and sleeve moments add up to the section moment within **0.18%**, the node
positions within **0.6 mm**, and the end force's deflection from the clamp axis within
**0.003°** (32.37°).
**0.003°** (32.37°). Node by node, relative to the local value floored at 0.1% of the root
curvature, the curvature follows the elastica within **0.97 / 0.49 / 0.24 / 0.18%** on
0.2 / 0.15 / 0.1 / 0.05 m elements (gates 1 / 1 / 0.5 / 0.5%), under either output recovery,
and under the stiffener the element peaks of `.elements.out`, read through `bend_check`,
within **the same figures** on the same gates; 0.12% of it at the root is
the order-`T/EA` difference between the elastica's bending law and the element energy's,
not mesh error.
- **L-fig curvature overlay**: `l3_lazywave_dynamic` writes each site's along-arc
curvature profile.
[`validation/scripts/curvature_profile_overlay.py`](https://github.com/SMI-Lab-Inha/CableDyn/blob/main/validation/scripts/curvature_profile_overlay.py)
Expand Down
5 changes: 4 additions & 1 deletion doc/driver_format.md
Original file line number Diff line number Diff line change
Expand Up @@ -1099,7 +1099,10 @@ Behaviour:
- The element properties are piecewise constant, one value per deck element, so the deck mesh
must resolve the taper: an initialisation note names a stiffener that spans fewer than 10
elements and the element length that resolves it. Split the section at the stiffener tip and
mesh the part under it finely (about 0.1–0.2 m elements for a typical 3–8 m stiffener).
mesh the part under it finely (about 0.1–0.2 m elements for a typical 3–8 m stiffener). The
node outputs and the element peaks of `.elements.out` do not take these element values: they
read `EI(s)` at their own arc, and at a real step of the section (the tip, rows that do not
match, two line types) the softer side's.
- The stiffener root takes the moment only where the end is held in rotation: declare the end
`Rigid`, or a rotational spring, in [END CONNECTIONS](#end-connections-optional-finite-ei-lines).
A stiffener at a pinned end prints a note.
Expand Down
5 changes: 4 additions & 1 deletion doc/modeling.rst
Original file line number Diff line number Diff line change
Expand Up @@ -91,7 +91,10 @@ sleeve's bending stiffness, mass, buoyancy and drag to the elements it covers
the hang-off direction. A pinned end carries no moment, so the stiffener then takes none;
- **the mesh under the stiffener.** One stiffness per element: split the section at the
stiffener tip and mesh the part under it with elements of about 0.1–0.2 m, at least ten over
the stiffener (the driver notes fewer);
the stiffener (the driver notes fewer). The node outputs and the element peaks of
``.elements.out`` take the stiffness at their own arc, so on that mesh the curvature follows a
much finer one within about 0.2 % (``lazy_wave_bend_stiffener.dat``), and shortening the
elements further gains little;
- **the material.** The sleeve is linear elastic. Polyurethane stiffness depends on temperature
and strain rate; use the modulus for the design temperature and check the result at the
ends of its range.
Expand Down
3 changes: 2 additions & 1 deletion doc/options.rst
Original file line number Diff line number Diff line change
Expand Up @@ -492,7 +492,8 @@ Finite-EI cable controls
:math:`\mathbf F\cdot\mathbf t`, the moment the neighbouring elements carry, and
curvature = moment / ``EI`` (see :ref:`theory-output-recovery`). ``element`` reports the
length-weighted element-mean axial force :math:`EA(\lvert\mathbf r'\rvert - 1)`, the
larger one-sided exact curvature, and the axial resultant at the element midpoint as the
larger one-sided exact curvature (on an element with a bend stiffener sleeve, the recovered
moment over ``EI(s)``), and the axial resultant at the element midpoint as the
segment tension, the definitions of the releases before 0.2.0. End forces, positions, ``TDP``
channels and the material-strain checks (``tensile_safety``, compression notes) are the
same under both
Expand Down
17 changes: 11 additions & 6 deletions doc/outputs.rst
Original file line number Diff line number Diff line change
Expand Up @@ -289,7 +289,9 @@ B)``, header row, units row, data (12 columns):
- unstretched arc length of the curvature maximum
* - ``PeakCurvature``
- ``(1/m)``
- maximum curvature over the element
- maximum curvature over the element; under a bend stiffener the curvature of the section,
the moment over the stiffness ``EI(s)`` along the element, not of the interpolation
(:ref:`theory-bend-stiffeners`)
* - ``BendMomentAtPeak``
- ``(N.m)``
- ``EI × PeakCurvature``, with the line type's ``EI`` (the cable's share under a bend
Expand Down Expand Up @@ -684,10 +686,12 @@ Line-node channels
- curvature at node ``J``. ``EI = 0`` lines: discrete curvature of the circle through
the node and its two neighbours (end nodes take the adjacent interior value).
Cubic-Hermite lines: ``BendMom<L>N<J>`` / ``EI``, the larger of the two sides at a
section junction, and the exact geometric curvature where ``EI`` is below
``max(|w| L0^2)/12`` times 1 m (:ref:`theory-output-recovery`) (``OutputRecovery
section junction (under a bend stiffener the section moment over the stiffness ``EI(s)``
at the node, :ref:`theory-bend-stiffeners`), and the exact geometric curvature where
``EI`` is below ``max(|w| L0^2)/12`` times 1 m (:ref:`theory-output-recovery`) (``OutputRecovery
element``: exact curvature of the continuous
centreline, the larger of the two one-sided element values at an interior node). The
centreline, the larger of the two one-sided element values at an interior node; the
recovered moment over ``EI(s)`` on an element with a bend stiffener sleeve). The
library routines ``CD_HermiteCable_Dyn_Curvature`` and ``CD_HFMF_Curvature`` return the
same value; ``CD_HermiteCable_Dyn_Curvature_Exact`` and ``CD_HFMF_Curvature_Exact`` the
exact geometric traces
Expand All @@ -709,8 +713,9 @@ Line-node channels
(one-sided at a line end, where it is the end connection moment);
``OutputRecovery element``: ``EI`` times the larger one-sided exact curvature. 0 on
``EI = 0`` lines. Under a bend stiffener (``STIFFENERS``) it is the cable's own share,
the line type's ``EI`` times the curvature; the stiffener's share is ``StfMom<L>N<J>``
and the two add up to the section moment
the line type's ``EI`` times the curvature, which is the section moment over the
stiffness ``EI(s)`` at the node; the stiffener's share is ``StfMom<L>N<J>`` and the two
add up to the section moment
- N·m
* - ``StfMom<L>N<J>``
- the share of the section moment a bend stiffener carries at node ``J``, the section
Expand Down
56 changes: 46 additions & 10 deletions doc/theory.rst
Original file line number Diff line number Diff line change
Expand Up @@ -178,11 +178,38 @@ part of the element a segment covers. With these one-per-element values the tape
by the mesh, as in a segmented line model; the driver notes a stiffener that spans fewer than
ten elements.

The recovered section moment :math:`M` (:ref:`theory-output-recovery`) is shared in proportion
to the stiffnesses, since cable and sleeve take the same curvature :math:`\kappa = M/EI_e`: the
cable carries :math:`EI_c\,\kappa`, reported as ``BendMom``, and the sleeve
:math:`(EI_e - EI_c)\,\kappa`, reported as ``StfMom``. At a node between a covered and a bare
element the side of the larger curvature, the bare one, gives the values. The deflection angle
The node outputs do not use :math:`EI_e`. It stands for the whole element, so under a taper it
steps at every node although the section does not, and the softer side would overstate the
curvature by about :math:`h\lvert EI'\rvert/(2 EI)`, first order in the element length
:math:`h`. A node takes :math:`EI(s)` at its own arc instead, from the stiffener rows, one limit
from either side. The section moment :math:`M` (:ref:`theory-output-recovery`) is continuous and
second-order accurate, so the curvature :math:`\kappa = M/EI(s)` is too. Cable and sleeve take
that curvature and share the moment in proportion to the stiffnesses: the cable carries
:math:`EI_c\,\kappa`, reported as ``BendMom``, and the sleeve
:math:`(EI(s) - EI_c)\,\kappa`, reported as ``StfMom``.

The two limits differ only at a real step of the section: at the bare tip, between stiffener
rows whose end diameter, bore or modulus do not match, and between line types of different
:math:`EI`. There the side of the larger curvature gives the values, the bare one at the tip;
no threshold on the stiffness ratio is involved. A node past a tip inside an element is bare.

An element's interpolation bends under :math:`EI_e`, so on an element with a sleeve its curvature
is the curvature of no section: :math:`EI_e` times it is the linear fit of the moment over the
element, so at an element's end it is off by the same first-order step against :math:`EI(s)`,
high where the curvature falls toward the tip and low where it rises, and :math:`EI_e` times it
is low by :math:`h^2 M''/12` against :math:`M`. Two outputs that read the interpolation elsewhere read
the section here. The peak of such an element in ``.elements.out`` is the largest
:math:`\lvert M(u)\rvert/EI(s(u))` along it, with :math:`M(u)` the quadratic through the
recovered moments of the end nodes that keeps the mean of the fit; a row end inside the element,
or the tip, is a step of :math:`EI(s)` that the search keeps. Below the stiffness that resolves
the recovered moment (:ref:`theory-output-recovery`) both the nodes and the peak fall back on
the interpolation's curvature times :math:`EI_e/EI(s)`. ``OutputRecovery element`` reports
the recovered moment over :math:`EI(s)` on such an element, the same value as the equilibrium
recovery, and the exact trace on every other element. The geometric traces themselves stay
available as ``CD_HermiteCable_Dyn_Curvature_Exact``, and the mesh diagnostics
(:math:`h\kappa`) read them.

The deflection angle
``FairDefl``/``AnchDefl`` is :math:`\arccos(\hat{\mathbf F}\cdot\hat{\mathbf d})` between the
end force :math:`\mathbf F` and the direction :math:`\mathbf d` the end connection holds (into
the line), the angle through which the stiffener turns the cable.
Expand All @@ -192,10 +219,17 @@ The planar check ``driver_bend_stiffener`` compares a hang-off clamped through a
collocation solution of the extensible elastica
:math:`\theta' = M/EI(s)`, :math:`M' = -(1 + \varepsilon)(\cos\theta\,n_z - \sin\theta\,n_x)`,
:math:`\mathbf n' = (0, w(s))`: on 0.05 m elements the cable's moment over the stiffener and
3 m beyond follows the elastica within 0.8 % of its root value, cable and sleeve moments add up
3 m beyond follows the elastica within 0.2 % of its root value, cable and sleeve moments add up
to the section moment within 0.2 % at the root, the deflection angle agrees within 0.003°, and
the node positions within 1 mm. On the same hang-off, the curvature along the stiffener follows
OrcaFlex 11.6d with its stiffener given as a line attachment within 0.8 % of the root curvature.
the node positions within 1 mm. Node by node, the curvature and the cable's moment follow the
elastica within 0.24 % of their local value on 0.1 m elements and 0.18 % on 0.05 m, 0.97 % on
0.2 m, under either output recovery, and the peaks of ``.elements.out`` under the stiffener
as closely. The local value is floored at 0.1 % of the root curvature, as the curvature changes sign
near the tip, and part of the difference is not mesh error: the elastica takes the moment
proportional to the rotation per unstretched length, the element energy the curvature of the
stretched line, a difference of order :math:`T/EA` that is 0.12 % at this root. On the same
hang-off, the curvature along the stiffener follows OrcaFlex 11.6d with its stiffener given as a
line attachment within 0.6 % of the root curvature.

.. _theory-output-recovery:

Expand Down Expand Up @@ -277,7 +311,8 @@ it, and its tangent rows are :math:`\mathbf{m}` there. An interior node reports
\kappa_j = M_j / EI ,

with :math:`\mathbf{t}_j` the unit nodal tangent, :math:`\mathbf{m}_j` the mean of the two
one-sided values, and at a section junction the larger of :math:`M_j/EI` on the two sides. The
one-sided values, and at a section junction the larger of :math:`M_j/EI` on the two sides (under
a bend stiffener :math:`EI` is the stiffness at the node, :ref:`theory-bend-stiffeners`). The
recovered moment carries an error that does not depend on :math:`EI`, of the order of the largest
consistent-load moment :math:`w L_0^2/12` of an element (a section junction leaves a jump of that
order between the two one-sided moments), so over a soft section dividing by :math:`EI` amplifies
Expand Down Expand Up @@ -325,7 +360,8 @@ at most 0.45 N of 9.4 kN and moves the bend moment away from :math:`EI\kappa`.

``OPTIONS`` ``OutputRecovery element`` restores the element definitions: the length-weighted
mean of the element-mean axial forces :math:`EA(\lvert\mathbf{r}'\rvert - 1)` at an interior
node, the larger one-sided exact curvature, and the midpoint axial resultant as the segment
node, the larger one-sided exact curvature (the section's on an element with a bend stiffener
sleeve, :ref:`theory-bend-stiffeners`), and the midpoint axial resultant as the segment
tension. The material-strain checks (``tensile_safety``, the static compression notes) judge
:math:`EA\varepsilon` under either setting.

Expand Down
Loading
Loading