Dennis Román | Project Portfolio

Aircraft Icing (FPCS Lab)

FPCS Lab Icing Verification Series

Consolidated FPCS Lab view covering IPW-style end-to-end pipeline execution and RG-15 reproduction checks, focused on verification discipline, run reproducibility, and comparison-ready outputs.

Concise summary for portfolio; details available on request.

ANSYS FluentFENSAP-ICEPythonSLURMMATLABStarCCMEnsightParaviewANSYSOpenFOAMJavaC++
IPW setup geometry

IPW-style geometry and setup context used for the end-to-end icing verification flow.

When

  • FPCS Lab involvement: Jun 2025–Jan 2026 (icing verification and IPW-style pipeline work).
    • Summer Research Intern: June–July 2025
    • Undergraduate Research Assistant: August 2025–January 2026

Why this is one project

  • This project builds systems progressively: starting with simple 2D cases, advancing to more complex 3D cases, implementing automation on those 3D cases, and ensuring procedures remain consistent on increasingly complex 3D geometries.

Context anchor

This work sits inside the broader multi-physics and HPC-focused research environment of the Flow Physics and Computational Science Lab at Georgia Tech.

Case index

  • IPW pipeline verification: End-to-end staged workflow from mesh/flow through DROP3D/ICE3D and post wrapper. Key output: comparison-ready artifact package for IPW-style review workflows.
  • RG-15 IPW2 reproduction: Bin sensitivity and grid-family trend behavior under a 2D-intent study strategy. Key output: stable trend interpretation with explicit sensitivity tracking across metrics.

Case notes

IPW pipeline verification case

  • Focused on repeatability across stage handoffs and deterministic run structure.
  • Built for auditability: artifacts and metrics are exported in consistent package layouts.

RG-15 IPW2 reproduction case

  • Focused on trend reproduction confidence, not just one final curve match.
  • Used level-based and bin-based comparisons to expose sensitivity in key outputs.

What this demonstrates quickly

  • Ability to structure complex icing workflows into repeatable, reviewable execution systems.
  • Verification mindset: convergence and comparison quality are treated as first-class outputs.
  • Communication discipline: outputs are organized to support technical decisions, not only visual plots.

Validation Signals

  • Convergence and comparison artifacts captured per run group

    pass

  • Cross-case metric consistency preserved for A/B interpretation

    pass

Reproducibility Signals

  • Template-driven run generation

    Journal/sbatch templates enforce consistent run structure between stages.

  • Deterministic artifact packaging

    Post wrapper and exports produce repeatable cuts, thickness, MCCS, and convergence outputs.

  • Scoped public detail

    Concise summary for portfolio; details available on request.