Files
FESADev/phases/1-linear-static-mitc4/step13.md
T
2026-05-01 02:40:19 +09:00

72 lines
2.5 KiB
Markdown

# Step 13: linear-static-analysis-path
## Sprint Contract
### Objective
Implement the `LinearStaticAnalysis` Template Method path from parsed Domain to solved state and written `U`/`RF` results.
### Required Reading
- /AGENTS.md
- /PROGRESS.md
- /PLAN.md
- /docs/README.md
- /docs/HARNESS_ENGINEERING.md
- /docs/PRD.md
- /docs/ARCHITECTURE.md
- /docs/ADR.md
- /docs/NUMERICAL_CONVENTIONS.md
- /docs/RESULTS_SCHEMA.md
- /docs/VERIFICATION_PLAN.md
- /phases/1-linear-static-mitc4/step8.md
- /phases/1-linear-static-mitc4/step12.md
### Scope
- Implement the linear static analysis strategy using the documented common flow: validate, build model, number DOFs, assemble, constrain, solve, reconstruct, recover RF, update state, write results.
- Store mutable physical quantities and iteration/result state in `AnalysisState`.
- Route solving through the linear solver interface.
- Emit minimum Phase 1 result fields `U` and `RF`.
### Allowed Files
- `include/**`
- `src/**`
- `tests/**`
- `PLAN.md`
- `PROGRESS.md`
- `phases/1-linear-static-mitc4/index.json`
### Explicit Non-Goals
- Do not implement nonlinear, dynamic, thermal, composite, or contact workflows.
- Do not bypass `DofManager`, assembler, solver interface, or result writer boundaries.
- Do not compare against stored Abaqus references in this step; that is Step 14.
### Tests To Write First
- End-to-end small-model test using a deterministic element or minimal fixture where expected `U` and `RF` are known.
- Singular-system failure test through the analysis path.
- Result-emission test for `U` and `RF` fields and component order.
### Reference Artifacts
- None required for this orchestration step.
### Acceptance Commands
```bash
python scripts/validate_workspace.py
```
### Evaluator Checklist
- Strategy + Template Method architecture is visible and not hardwired to only one future analysis type.
- `AnalysisState` owns mutable solve/result data.
- `RF` uses full-vector recovery.
- Results match the documented step/frame/field layout.
### Handoff Requirements
- Record analysis behavior, tests, and validation in `PROGRESS.md`.
- Update `PLAN.md` if Phase 1 end-to-end risks remain.
- Update the matching phase index entry: use `completed` with a one-line `summary`, or `blocked` with `blocked_reason`, or `error` with `error_message`.
## Verification
1. Run the acceptance command.
2. Inspect the analysis path for architecture-boundary shortcuts.
## Do Not
- Do not make the first end-to-end path depend on local Abaqus execution.