HaomingLuo/AgentFEM-ThickCylinder-Elasticity

AgentFEM · Pressurized thick cylinders

Axisymmetric thick-cylinder elasticity with a closed-form Lame displacement reference for every sample. A compact verification and learning baseline.

256 independently solved parameter sets · full meshes and fields · 192 / 32 / 32 split · CC BY 4.0

Built with AgentFEM. A small, reproducible engineering dataset for surrogate learning, field prediction and numerical-method experiments.

A first experiment

Plot radial displacement through the wall and compare it with the plane-strain Lamé solution. This is a useful first check for units, field loading and mesh handling before attempting a harder surrogate-learning task.

Load the test split · Inspect verification evidence · Try the classroom cylinder exercise

Make your first verification plot

Download plot_lame.py, then run:

python -m pip install numpy h5py matplotlib huggingface_hub
python plot_lame.py

The script downloads only the test HDF5 file at a pinned revision, plots one recorded FEM sample against the plane-strain Lamé solution, and prints the radial-displacement relative nodal L2 error. No AgentFEM installation or GPU is needed. To use an existing download, run python plot_lame.py --data test.h5.

Try another HDF5 sample ID with --sample ID. The analytic displacement is u_r = (1+nu)/E * [(1-2nu)*A*r + B/r], where A=p*a²/(b²-a²) and B=A*b². Coordinates and displacement are stored in metres; the figure converts them to millimetres and micrometres. This reads recorded results, not a new simulation. These are plane-strain results: using the plane-stress formula would give a different answer.

Physical problem

Long-cylinder plane strain represented by a 0.02 m meridian strip. Axial displacement is constrained throughout. Inner pressure p is applied at r=a; outer pressure is zero. Axisymmetric solid elasticity, not a thin-shell approximation.

Quadratic quadrilateral displacement elements; 24 radial by 2 axial subdivisions.

A deliberately simple analytic-solvable baseline, not a surrogate challenge requiring a neural operator. No corrosion, thermal loading, yielding, end-cap effects, or service-life assessment.

What is included

• train.h5, validation.h5, test.h5: one group per sample ID, geometry, connectivity, point fields and cell fields. Gzip-compressed without floating-point downcasting.
• HDF5 inputs/: boundary-node masks and, for heat, the prescribed Gaussian source evaluated at mesh nodes. These are model inputs, not predicted fields or integrated nodal loads.
• Corresponding JSONL files: browsable parameters and engineering observables; they point to HDF5 groups, not serialized full fields.
• quality.json: checks, dependency versions, convergence evidence.
• baselines.json, baseline.py: held-out scalar mean/nearest-neighbor baselines with train-only normalization; no neural-model claims.
• generate.py, quality.py, load_example.py: generation, audit and minimal loading workflows.

SI units: coordinates and U in m; stress S and MISES in Pa; strain E dimensionless; temperature in K. coordinates are undeformed reference nodes. Connectivity follows the AgentFEM XDMF export order: Triangle (heat), Triangle_6 (notch), or Quadrilateral_9 (cylinder). Cell stress is projected and unsmoothed. See generated XDMF for native export semantics.

Tensor components are flattened row-major: 2×2 for plane stress (x,y), 3×3 for axisymmetry (r,theta,z). Displacement U uses padded three-component storage; active components are (x,y) for the plate and (r,z) for the cylinder. The HDF5 files contain mesh connectivity, not an assumed common image grid.

Parameters, in generator order: inner_radius_m, outer_radius_m, young_pa, poisson, internal_pressure_pa.

Observed sample ranges (exact values are in JSONL; intended sampling bounds are in generate.py):

Parameter	Minimum	Maximum
inner_radius_m	0.020144	0.0799171
outer_radius_m	0.0265325	0.214178
young_pa	7.02559e+10	2.0998e+11
poisson	0.220324	0.339834
internal_pressure_pa	1.07221e+06	1.99466e+07

Quality evidence

Every sample checked against the plane-strain Lame radial displacement solution. Maximum relative nodal L2 error: 4.247e-07.

All published field arrays are finite; all parameter sets and geometry/source-layout parameter tuples are unique. Splits use one fixed random permutation after scrambled Sobol sampling (seed 20260923; split seed 20260924). All splits are in-distribution; no claim of unseen-family generalization. One row is one independently solved case, not a frame or a rescaled duplicate.

Generic continuous material ranges are synthetic, not calibrated material grades. Dataset generation does not imply experimental validation. These are original parameterized examples, not certified NAFEMS benchmarks or copies of commercial solver datasets.

Load and reproduce

python load_example.py --repo HaomingLuo/AgentFEM-ThickCylinder-Elasticity --split test
# In the pinned AgentFEM/FEniCSx environment:
python generate.py cylinder --count 256

Loading requires only h5py, numpy and huggingface_hub; FEniCSx is needed only to regenerate FEM results. Generation writes a local generated/ directory and uses one MPI rank. The heat manufactured-solution check is available via python generate.py heat --validate.

Attribution

Haoming Luo, Pressurized thick cylinders — AgentFEM parametric FEM dataset, 2026. Please cite this repository URL and the commit revision used. Data: CC BY 4.0. Example code: Apache-2.0.