ARL Topologies vs FEniTop

ARL Topologies is free, open-source software for topology optimization, published by USArmyResearchLab as a United States government project. It is intended for research and general use, and runs self-hosted from the command line with an XML input file. The program separates design representation, optimization method, and objective function. Representations include two- and three-dimensional element-based density discretization with SIMP interpolation, plus nodal-based density discretization with Heaviside projection and filtering. Inputs can be ASCII ExodusII or GMSH meshes, or rectangular pixel and voxel regions. Available optimizers include optimality criterion, method of moving asymptotes, BFGS through NLopt, and a genetic algorithm. Its implemented objective performs static structural finite element analysis. Density results can be post-processed to STL, while VTK, GMSH, and Matlab are output formats. A shared-library objective function allows extension to other problems. Building requires a C++11 compiler and listed libraries.

FEniTop is free, open-source software for topology optimization based on FEniCSx. It handles two- and three-dimensional problems on structured or unstructured meshes, with varied element geometries, types, orders, and quadrature degrees. Users express partial differential equations in weak form, while automatic differentiation supports finite-element and sensitivity analysis. PETSc through petsc4py provides linear and nonlinear solvers and preconditioners. Parallel implementations of Optimality Criteria and Method of Moving Asymptotes update design variables. A Helmholtz-type PDE filter targets large-scale parallel problems, including large filter radii. FEniTop can run from laptops to distributed computing clusters. Optimized designs export as XDMF files for ParaView post-processing. Repository examples cover beams in 2D and 3D, a 2D disk, a 3D shell, and a 2D compliant mechanism. The software is licensed under GPL-3.0. Setup instructions specify FEniCSx 0.7.3, PyVista, and Xvfb; Windows 11 instructions use Ubuntu under WSL and Docker Desktop.