EasyFEA - Easy Finite Element Analysis

Overview

EasyFEA is a user-friendly Python library that simplifies finite element analysis. It is flexible and supports different types of simulation without requiring users to handle complex PDE formulations. The library currently supports four specific simulation types:

1. ElasticSimu (static and dynamic): Examples in /examples/Elastic, /examples/Dynamic and /examples/Contact.
2. BeamSimu (static Euler-Bernoulli): Examples in /examples/Beam.
3. ThermalSimu (stationary and transient): Examples in /examples/Thermal.
4. PhaseFieldSimu: (quasi-static phase field) Examples in /examples/PhaseField.

All examples are available here.

For each simulation, users create a mesh and a model. Once the simulation has been set up, defining the boundary conditions, solving the problem and visualizing the results is straightforward.

Numerous examples of mesh creation are available in the examples/Meshes folder.

The simplest and quickest introduction is shown below and is available in examples/HelloWorld.py.

from EasyFEA import (Display, Mesher, ElemType,
                     Materials, Simulations)
from EasyFEA.Geoms import Point, Domain

# ----------------------------------------------
# Mesh
# ----------------------------------------------
L = 120 # mm
h = 13

domain = Domain(Point(), Point(L,h), h/5*2)
mesh = Mesher().Mesh_2D(domain, [], ElemType.QUAD4, isOrganised=True)

# ----------------------------------------------
# Simulation
# ----------------------------------------------
E = 210000 # MPa
v = .3
F = -800 # N

mat = Materials.Elas_Isot(2, E, v, planeStress=True, thickness=h)

simu = Simulations.ElasticSimu(mesh, mat)

nodesX0 = mesh.Nodes_Conditions(lambda x,y,z: x==0)
nodesXL = mesh.Nodes_Conditions(lambda x,y,z: x==L)

simu.add_dirichlet(nodesX0, [0]*2, ["x","y"])
simu.add_surfLoad(nodesXL, [F/h/h], ["y"])

simu.Solve()

# ----------------------------------------------
# Results
# ----------------------------------------------
Display.Plot_Mesh(mesh)
Display.Plot_BoundaryConditions(simu)
Display.Plot_Result(simu, 'uy', plotMesh=True)
Display.Plot_Result(simu, 'Svm', plotMesh=True, ncolors=11)

Display.plt.show()

Installation

EasyFEA can be easily installed from PyPI using pip, compatible with Python versions 3.9 through 3.12:

pip install EasyFEA

Dependencies

EasyFEA uses several libraries such as NumPy and Gmsh - as such, the following projects are required dependencies of EasyFEA:

• numpy - Fundamental package for scientific computing with Python.
• gmsh (>= 4.12) - Three-dimensional finite element mesh generator.
• scipy - Fundamental package for scientific computing in Python.
• matplotlib - Plotting package.
• pyvista - Plotting package.
• numba - Compiling Python code using LLVM.
• pandas (3.9 <= Python <= 3.12) - Powerful data structures for data analysis.
• imageio and imageio[ffmpeg] - Library for reading and writing a wide range of image, video, scientific, and volumetric data formats.

For detailed information on installing numba, refer to the Numba Installation Guide.

Optional Dependencies

EasyFEA includes a few optional dependencies for reducing resolution time or for performing DIC:

• pypardiso (Python > 3.8 & Intel oneAPI) - Library for solving large systems of sparse linear equations.
• petsc and petsc4py - Python bindings for PETSc.
• opencv-python - Computer Vision package.

Naming conventions

EasyFEA uses Object-Oriented Programming (OOP) with the following naming conventions:

• PascalCasing for classes
• camelCasing for properties
• Snake_Casing or Snake_casing for functions/methods

In this library, objects can contain both public and private properties or functions.

Private parameters or functions are designated by a double underscore, such as __privateParam. In addition, parameters or functions beginning with an underscore, such as _My_Function are accessible to advanced users, but should be used with caution.

Contributing

EasyFEA is an emerging project with a strong commitment to growth and improvement. Your input and ideas are invaluable to me. I welcome your comments and advice with open arms, encouraging a culture of respect and kindness in our collaborative journey towards improvement.

To learn more about contributing to EasyFEA, please consult the Contributing Guide.

Citing EasyFEA

If you are using EasyFEA as part of your scientific research, please contribute to the scientific visibility of the project by citing it as follows.

Matthieu Noel. EasyFEA: a user-friendly Python library that simplifies finite element analysis. 2024, https://hal.science/hal-04571962

Bibtex:

@softwareversion{noel:hal-04571962v1,
  TITLE = {{EasyFEA: a user-friendly Python library that simplifies finite element analysis}},
  AUTHOR = {Noel, Matthieu},
  URL = {https://hal.science/hal-04571962},
  NOTE = {},
  INSTITUTION = {{Universit{\'e} Gustave Eiffel}},
  YEAR = {2024},
  MONTH = Apr,
  SWHID = {swh:1:dir:ffb0e56fe2ce8a344ed27df7baf8f5f1b58700b5;origin=https://github.com/matnoel/EasyFEA;visit=swh:1:snp:88527adbdb363d97ebaee858943a02d98fc5c23c;anchor=swh:1:rev:ee2a09258bfd7fd60886ad9334b0893f4989cf35},
  REPOSITORY = {https://github.com/matnoel/EasyFEA},
  LICENSE = {GNU General Public License v3.0 or later},
  KEYWORDS = {Finite element analyses ; Computational Mechanics ; Numerical Simulation ; Phase field modeling of brittle fracture ; Linear elasticity ; Euler-Bernoulli beam ; DIC - Digital Image Correlation ; User friendly ; Object oriented programming ; Mesh Generation},
  HAL_ID = {hal-04571962},
  HAL_VERSION = {v1},
}

License

Copyright (C) 2021-2024 Université Gustave Eiffel. All rights reserved.

EasyFEA is distributed under the terms of the GNU General Public License v3.0 or later, see LICENSE.txt and CREDITS.md for more information.