Problem Statement

Extracting interpretable modal structure from turbulent simulations remains challenging in high-dimensional datasets.

Why It Matters

Modal decomposition provides useful insight for reduced-order modeling and control-oriented analysis.

Methods / Numerical + ML Setup

• Performed modal analysis on spectral-element simulated turbulent datasets.
• Compared modal behavior across representative simulation conditions.

Key Results

• Reported modal trends linked to flow-structure organization.
• Contributed to analysis workflows relevant for CFD-informed control research.

Toolchain

• Spectral element solver outputs
• Python analysis
• Scientific visualization

Challenges and Lessons

• Handling large turbulent datasets with stable post-processing pipelines.
• Interpreting modal outputs in physically meaningful terms.

Future Work

• Link modal signatures to control-policy state representations.
• Use modal features in RL state abstraction experiments.