AeroLab Studio

A desktop application for aerogel-fibre research: every equation and measurement from three papers on thermoplastic-polyurethane aerogels, applied by hand or automatically, with publication figures exported straight into OriginLab and a formatted Excel workbook.

Built from:

• Omranpour et al. (2024a) — TPU aerogels: structure, fatigue and strain–life
• Omranpour et al. (2024b) — TPU/silica composite aerogels: thermal and cyclic behaviour
• Omranpour et al. (2025) — nanofibrous aerogel fibres: spinning, structure, properties

User guide: AeroLab_Studio_User_Guide.docx in this folder — step by step, with screenshots (docs/USER_GUIDE.md is the same material as plain text).

---

What it does

59 equations	Every relation in the three papers, with unit conversion on every input and uncertainty propagation by central differences.
11 measurements	Stress–strain, cyclic loading, S–N, strain–life, density scaling, TGA, DSC, thermal properties, BET, Herschel–Bulkley, amplitude sweep.
Automatic import	CSV, TSV, TXT, DAT, Excel, JSON, NumPy archives. Instrument preambles, unit rows, European decimals and multi-sheet workbooks are all handled; columns are matched to the measurement by name.
Origin export	Styled workbooks (Long Name / Units / Comments) and graphs with publication colours and symbols, saved as a .opju project with PNG exports. No Origin on the machine? It writes a LabTalk package that rebuilds the same project elsewhere.
Excel export	Summary, one sheet per dataset with its curves, native Excel charts, the validation table, and a provenance sheet recording versions, files and settings.
Self-validation	26 cases recompute numbers printed in the papers. Two are recorded as disagreements with the papers, with the reasoning stated, rather than quietly fudged.

---

Install and run

The built application (no Python needed)

Run AeroLabStudio-Setup.exe, or unzip the portable folder and run AeroLabStudio.exe.

From source

pip install -r requirements.txt
python main.py

Python 3.11 or newer. Tested on Python 3.14 with PySide6 6.11.

OriginLab support is optional. With Origin installed and pip install originpro pywin32, the app drives it directly; without, it writes a script package instead.

---

A five-minute tour

1. Data → Load demo data. Eleven datasets from the papers appear, deliberately messy: instrument preambles, a units row, European decimals.
2. Pick one. The preview shows what was parsed; the measurement is guessed from the headers and the file name, and each column is mapped. Change anything you disagree with.
3. Analyse this dataset. The results, the graph and the settings appear under Measurements. Change a setting and re-run — it is free to try, and Ctrl+Z undoes it.
4. Plots overlays several runs on one pair of axes.
5. Export writes the Excel workbook, or sends everything to Origin.
6. Validation shows the equation library recomputing the papers' own numbers.

Ctrl+K opens a command palette with every action in the app.

---

Layout

aerolab/
  core/
    units.py          ~28 dimensions, exact conversion factors
    equations.py      59 equations, each with inputs, units and a reference
    curves.py         the analysers: fits, peaks, integrals
    measurements.py   what each experiment needs, and what it plots
    validation.py     26 cases recomputing published values
  io/
    readers.py        delimiter sniffing, unit rows, Excel, JSON, NPZ
    excel.py          the formatted workbook with native charts
    origin.py         live Origin automation, and the LabTalk fallback
  viz/
    style.py          the palette and symbol set, shared with Origin
    figures.py        matplotlib rendering
  ui/
    app.py            main window, navigation, command palette
    state.py          the session, with undo/redo
    widgets.py        cards, tiles, toasts, plot canvas, formula rendering
    pages/            one file per page
  resources/samples/  the eleven demo datasets
tests/                pytest suite
docs/USER_GUIDE.md    the longer guide

---

Testing

pip install pytest
pytest tests -v

The suite checks the equation library against the papers, the analysers against synthetic data with known answers, the importer against deliberately awkward files, the Excel and Origin writers, and a headless smoke test of the interface.

---

Building the application

pip install pyinstaller
python build_assets/make_icon.py          # draws the icon and version resource
pyinstaller AeroLabStudio.spec --noconfirm

dist/AeroLabStudio/ is a self-contained folder. To build the installer, install Inno Setup and compile installer/AeroLabStudio.iss.

---

Two things the papers and this program disagree about

Both are shown on the Validation page rather than hidden:

1. Effusivity of NFT-4. Figure 6e plots 38.2 W·s^0.5/(m²·K). Equation 10 applied to the k and α plotted in the same figure gives ≈62.6. Worth asking the authors.
2. Draw ratio of the final spinning step. The stated take-up speed is below the mean nozzle exit speed, so there is little draw-down; the sub-100 µm fibre diameter comes mostly from solvent loss and shrinkage.

---

Author

Ezzeldien Yousef — ezz.yousef@mail.utoronto.ca MPML, Microcellular Plastics Manufacturing Laboratory Department of Mechanical & Industrial Engineering, University of Toronto

The same details are on the application's About page, with a copy-to-clipboard button.