Step 1: research

Read First

/AGENTS.md
/docs/ABAQUS_SUBROUTINE_AGENT_DESIGN.md
/docs/ARCHITECTURE.md
/docs/ADR.md
/.codex/skills/abaqus-subroutine-research/SKILL.md
/.codex/skills/fem-theory-query/SKILL.md
/docs/requirements/uel-3d-euler-beam.md
/docs/research/README.md

Task

Create /docs/research/uel-3d-euler-beam-research.md.

Research only the evidence needed to support a two-node 3D Euler-Bernoulli beam UEL implementation and validation plan. Separate:

Source-backed facts.
Inferences made from those facts.
Applicability limits for this implementation.

Cover at minimum:

Abaqus/Standard UEL responsibilities for RHS, AMATRX, SVARS, ENERGY, NDOFEL, NNODE, MCRD, LFLAGS, PROPS, COORDS, U, DU, and JPROPS.
Euler-Bernoulli 3D beam stiffness matrix in local coordinates for axial, torsion, bending about local 2 and local 3 axes.
Local-to-global transformation for two-node beam elements, including orientation-vector edge cases.
Residual sign convention needed by Abaqus UEL.
Verification benchmarks appropriate for no-Abaqus tests and external Abaqus CSV comparison.

Prefer official Abaqus documentation and reliable FEM references. Public example repositories may be used only as learning references, not acceptance evidence, unless license/version/provenance are documented.

Do not create or modify Fortran source, tests, reference artifacts, or implementation plans in this step.

Tests To Write First

None. This is a research artifact step.

Acceptance Criteria

python -m unittest discover -s scripts -p "test_*.py"
python scripts/validate_reference_artifacts.py
python scripts/validate_workspace.py

The step is complete only when the research document cites sources and clearly marks facts, inferences, and applicability limits.

Validation Notes

Abaqus execution remains external to this repository.
Update /phases/uel-3d-euler-beam/index.json step 1 with completed, error, or blocked and a concrete summary or reason.

Forbidden

Do not add JavaScript/TypeScript/npm fallback.
Do not run Abaqus analyses from this project.
Do not generate reference CSVs.
Do not create Fortran production source.
Do not break existing tests.

2.2 KiB Raw Blame History