experimental
ANSYS_skill
A validation layer for automated ANSYS Mechanical runs. I built it to record the stage each run reached and the result evidence behind it; a cantilever benchmark matches Euler-Bernoulli theory within 0.223%.
Writing about this project
-
A result object is not an evaluated result
Why automated ANSYS Mechanical work needs separate geometry, setup, solver-input, solved, evaluated, and credibility gates.
-
A narrow Mechanical delivery protocol
A stepwise Discovery-to-Workbench-to-Mechanical path with explicit success signals and recovery actions.
-
Failure taxonomy for automated Mechanical handoffs
Seven ways an ANSYS workflow can look complete while its solver, evaluation, delivery, or exported evidence is still incomplete.