Examples

The repository ships end-to-end studies under examples/. Each script is self-contained and runs in the examples environment (julia --project=examples). The User Guide's worked-examples chapter walks through the physics and the code of the main studies in detail.

Cone calorimeter

ScriptWhat it models
06_cone_calorimeter.jlPMMA slab under a 50 kW/m² cone heat flux with Beer–Lambert in-depth absorption, single-step Arrhenius kinetics, and gas release.
06_cone_calorimeter_surface_absorption.jlThe same PMMA case with the opaque surface-absorption radiation model, for comparison against the semi-transparent variant.
07_douglas_fir_cone.jlDouglas fir with multi-step kinetics, moisture, Darcy–Fick transport, and ALE mesh motion.
08_pressure_treated_wood_cone.jlPressure-treated wood with moisture, mapped from a ThermaKin calibration.
macfp/macfp_gasification.jlMaCFP gasification benchmark with parallel solves.

Transport and physics demonstrations

ScriptWhat it models
08_darcy_fick_demo.jlPressure-driven Darcy–Fick gas transport through a charring, wood-like material.
moisture_hos_comparison.jlEffect of the moisture heat-of-sorption model on drying and pyrolysis.
convergence_douglas_fir.jlGrid-convergence study over mesh resolutions (uses Distributed).

Comparison and validation studies

ScriptWhat it models
ptw_comparison.jlPressure-treated wood results against ThermaKin output (includes a ThermaKin output parser).
df_sensitivity_comparison.jlInter-code sensitivity comparison for the Douglas fir case.
literature_comparison/literature_comparison.jlRuns the cone cases across literature property sets via serialisable run specs.

Inverse analysis

inverse_analysis/ fits white-pine component conductivities and kinetics to cone-calorimeter data: forward model, objective, optimisation driver, uncertainty quantification, and plotting/diagnostic helpers. See the inverse-analysis walkthrough and the sensitivity-analysis chapter for the underlying pattern.

Sensitivity analysis

sensitivity_analysis/ contains forward uncertainty propagation via Latin Hypercube sampling (sensitivity_douglas_fir.jl), global Sobol indices via polynomial chaos expansion (sensitivity_sobol.jl), and a PCE degree-selection study (compare_degrees.jl), together with write-ups of the results.

Running the examples

# from the repository root
julia --project=examples examples/06_cone_calorimeter.jl

The examples environment adds Plots and other post-processing dependencies on top of the package itself. Scripts that use Distributed (the convergence and sweep studies) document their worker setup at the top of the file.