Skip to content

Datasets

Overview

This section collects curated datasets developed within the MAO research group for:

  • design-space dimensionality reduction
  • surrogate-based optimization
  • time-series modeling and system identification

All datasets are publicly available on :contentReference[oaicite:1]{index=1} and are part of the MAO research group collection:

👉 https://zenodo.org/communities/cnr-inm-mao/

The collection ensures consistency, traceability, and long-term accessibility of all published datasets.


Dataset Categories

🔷 Dimensionality Reduction Benchmarks

Datasets designed for design-space compression, PME-based methods, and optimization workflows.

These datasets share a common structure:

  • geometry (D)
  • design variables (U)
  • optional physics fields
  • performance outputs

Available datasets

  • Bio-inspired underwater glider
  • RAE2822 airfoil
  • DTMB 5415 hull

👉 Explore dimensionality reduction datasets


🌊 Time-Series & Dynamics Datasets

Datasets focused on temporal evolution of physical systems, including ship motions and wave-induced responses.

These datasets are suitable for:

  • system identification
  • reduced-order modeling
  • machine learning for time-dependent phenomena

Available datasets

  • Ship motion time-series dataset

👉 Explore time-series datasets


Reproducibility

All datasets include:

  • full metadata
  • variable definitions
  • preprocessing details
  • standardized formats

They are designed for:

  • fair benchmarking
  • cross-method comparison
  • integration with PME-toolkit workflows

Citation

Please cite the corresponding Zenodo records when using these datasets.