I need a concise, fully reproducible finite-element model of a corroded steel pipe built in ANSYS Mechanical APDL and driven entirely from PyMAPDL. The end goal is to predict corrosion patterns, so every step—from geometry creation to post-processing—has to be callable from Python and suitable for later machine-learning use. Scope • Simulate uniform corrosion only. • Account for the pipe’s type of material, its key mechanical properties, and any previous corrosion history that could influence the stress field. • Keep the mesh lean but numerically stable so the job runs quickly on a standard workstation. Deliverables • A clean, well-commented PyMAPDL script that: – Generates or imports the corroded geometry, – Assigns material data via tables or JSON, – Sets boundary conditions and loads, – Solves for stresses/strains, and – Exports nodal and elemental results to CSV or HDF5 for downstream ML training. • Short README describing the workflow and how to swap material datasets. • Two or three sample result plots (stress contour, deformation) rendered directly from the script. Acceptance I will run your script on a fresh ANSYS install; it should execute without manual clicks and reproduce the figures you provide. Feel free to suggest minor refinements—mesh density, element type, or solver settings—so long as they stay aligned with the primary goal of rapid corrosion-pattern prediction. If you have prior PyMAPDL automation or pipeline-to-ML experience, mention it; that will move your proposal to the top of the list.