Abaqus代理

Installation

Install from source:

git clone https://github.com/Tomsabay/abaqus_agent.git
cd abaqus_agent
pip install -e ".[dev,mcp]"

Optional extras:

pip install -e ".[llm]"  # Anthropic / OpenAI planners
pip install -e ".[all]"  # dev + mcp + llm

Safety And Deployment

All generated or processed workflows are intended to run locally in the user's own Abaqus-licensed environment.

The recommended commercial deployment model is BYOL:

• customer-local runner
• customer-owned Abaqus license
• local artifacts and ODBs
• optional consulting, report templates, private recipes, and team runner

Do not run third-party Abaqus workloads as a hosted SaaS without explicit legal review of the relevant Dassault Systemes license terms.

Quick Start

Run tests that do not require Abaqus:

pytest tests/ -v

Check whether the current machine is ready for real Abaqus validation:

abaqus-agent validate env --json
abaqus-agent validate env --expected-release 2026 --strict --out validation-preflight.md
abaqus-agent validate env --workdir runs --runner-json '{"cpus":4,"mp_mode":"threads","timeout_seconds":900}' --json
abaqus-agent validate record --environment "Windows 11" --abaqus "Abaqus 2021" --workflow "cantilever" --result PASS --evidence "status=COMPLETED"

Export an offline report from a run directory, capsule.json, or result.json:

abaqus-agent report export runs/my_run --template client_summary --out report.html
abaqus-agent report export runs/my_run --template client_summary --out report.pdf
abaqus-agent report export runs/my_run --template engineering_delivery --out delivery.html
abaqus-agent report export runs/my_run --out report.zip

PDF export is optional and renders the standalone HTML report through Playwright:

pip install "abaqus-agent[pdf]"
playwright install chromium

The web UI's Report panel can also load the same offline source path and render the report without starting a new analysis run.

Validate public benchmark specs without Abaqus:

python run_benchmark.py --dry-run

Run one full Abaqus case on a machine with Abaqus installed:

python agent/orchestrator.py cases/cantilever/spec.yaml \
  cases/cantilever/expected.json \
  cases/cantilever/runner.json

Use an existing .inp as a first-class input:

meta:
  abaqus_release: "2024"
  model_name: "CustomerModel"
geometry:
  type: custom_inp
  inp_path: model.inp
material:
  name: Placeholder
  E: 210000
  nu: 0.3
analysis:
  solver: standard
  step_type: Static
bc_load: {}
outputs:
  kpis:
    - name: U_tip
      type: nodal_displacement

Create an experiment capsule from an .inp:

abaqus-agent capsule init --from-inp model.inp --out runs/model_capsule

from capsule.store import init_from_inp

capsule = init_from_inp("model.inp", "runs/model_capsule")
print(capsule["run_id"])

Evaluate physics contracts:

from contracts import evaluate_contracts

result = evaluate_contracts(
    [
        {"name": "deflects_down", "type": "direction", "kpi": "U_tip", "direction": "negative"},
        {"name": "stress_margin", "type": "range", "kpi": "MISES_MAX", "max": 250.0},
    ],
    {"U_tip": -0.002, "MISES_MAX": 210.0},
)

Diagnose solver logs:

abaqus-agent doctor Job-1.msg Job-1.sta

from doctor import diagnose_logs

diagnosis = diagnose_logs(paths=["Job-1.msg", "Job-1.sta"])

Compare KPI results:

abaqus-agent diff runs/baseline runs/candidate --out diff.md
abaqus-agent diff runs/baseline runs/candidate --tolerances-json '{"MISES": 0.20}' --out diff.md

Search local case memory:

abaqus-agent memory search runs/ --query too_many_attempts --json
abaqus-agent memory search runs/ --similar-to runs/candidate --kpi U_tip --out memory.md

from simdiff import diff_runs

diff = diff_runs("runs/baseline", "runs/candidate")

Normalize an ODB Lens KPI recipe and render a KPI report:

kpis:
  - name: max_mises
    source: odb
    field: S
    invariant: MISES
    region: set:CRITICAL_ZONE
    reducer: max

abaqus-agent lens normalize kpis.yaml --out _kpi_spec.json
abaqus-agent lens report result.json --recipe kpis.yaml --out kpi_report.md