scBESO: An In-House Topology Optimization Software Integrating Open-Source Tools
Please login to view abstract download link
Topology optimization is a computational design methodology for generating material-efficient structures within a given design domain and a set of boundary conditions. Among the various optimization algorithms available, the Bidirectional Evolutionary Structural Optimization (BESO) method [1] is a robust and computationally efficient algorithm, with intuitive material addition and removal strategies. In this context, scBESO is an in-house developed topology optimization software that implements the BESO method to perform structural and frequency optimizations. This software is capable of performing both 2D and 3D large scale simulations and integrates open-source tools for its pre- and post-processing stages. The pre-processing stage, which includes geometry modelling, mesh generation and application of boundary conditions, uses Gmsh [2] program and its native file format, while the post-processing stage, responsible for visualizations, animations and data extraction uses Paraview [3] and its VTK file format. Structural optimizations are formulated as compliance minimization problems and solved using a Preconditioned Conjugate Gradient (PCG) solver. Its performance was benchmarked using the MBB beam problem, comprising 5.184M elements (15,856,203 DOFs) and solved using a multi-CPU approach. Frequency optimizations are formulated as a frequency maximization problem, solved using a Lanczos solver, and validated experimentally on a cantilever beam [4]. In the future, scBESO will be released as open-source software for educational and academic use.
