about 50% of the current topology optimisation software utilise pure density-based methods (e.g. SIMP or BESO) and about one third of the market share belongs to hybrid approaches, whereby the majority includes also density-based methods. The remainder are level set and other evolutionary algorithms (not BESO).
Optistruct, Inspire or COMSOL employ e.g. level set and density-based methods, whereas Abaqus, ANSYS or Nastran centre solely on the latter.
The removal of material and the layout of the final topology in the density-based methods is linked with the strain energy in the individual elements whereas the topology derived from levels set methods is defined by the shape functions i.e. iso-contour lines.
SIMP (Solid Isotropic Material with Penalization) and BESO (Bi-directional Evolutionary structural Optimization) are among the most used methods to handle topology optimization in commercial FE codes. SIMP uses an interpolation scheme based on density for the Young modulus and uses sensitivity analysis to understand wheather or not the material can be removed from the design. BESO also heavly rely on sensitivity and can add or remove material from the design. The optimization is guided by the objective function that one defines, so they "remove" elements according to the particular functional you want to minimize/maximize. This can be strain energy, mutual potential energy but also compliance with certain failure criteria and so forth.
about 50% of the current topology optimisation software utilise pure density-based methods (e.g. SIMP or BESO) and about one third of the market share belongs to hybrid approaches, whereby the majority includes also density-based methods. The remainder are level set and other evolutionary algorithms (not BESO).
Optistruct, Inspire or COMSOL employ e.g. level set and density-based methods, whereas Abaqus, ANSYS or Nastran centre solely on the latter.
The removal of material and the layout of the final topology in the density-based methods is linked with the strain energy in the individual elements whereas the topology derived from levels set methods is defined by the shape functions i.e. iso-contour lines.
ANSYS topological optimization capability was present both in Mechanical APDL and Workbench in the past. ANSYS software has built-in topology optimization module to model, analyze and perform topology optimization with two options available: optimality criteria (OC) approach (default choice) and sequential convex programming (SCP) approach.
Because the capabilities were limited, those tools were improved (see theory manuals) and moved to ANSYS ACT Extensions for Topological Optimization in ANSYS Mechanical for versions >17.0 and can be used with workbench based ANSYS Mechanical Solver.
When available for the problem, the optimality criteria (OC) approach is the fastest.
Topology optimisation module is available in commercial softwares like Ansys, Comsol, SIMULIA-Tosca etc. Popular methods for topology optimisation are Density method, Level set method and Evolutionary Structural Optimisation method. In density method, the parameters of physics like young's modulus or thermal conductivity is defined as a function of 'material density' of that finite element cell. Then by running the optimiser and by restricting the density to take either 0 or 1 value, topology optimisation is carried out. If the material density becomes 1 that represents a solid region and if it becomes 0 then it represents a void region.
So these topology optimisation methods, internally use a gradient based optimisation algorithm and the commonly used algorithms are Method of Moving Asymptotes, Sequential Quadratic Programming, Optimality Criteria method etc.
Comsol has MMA, SQP and SNOPT optimisation algorithms for Density Method.
FE Software almost used density based approach like SIMP and for another approach you want like SQP, BA, AA, level set etc, you must write their codes in FORTRAN or any FE software or in matlab. In optisruct (Altair Hyper-works) level-set method also used.
We have recently published a review article titled "Review on design and structural optimisation in additive manufacturing: Towards next-generation lightweight structures" which can be obtained from DOI: 10.1016/j.matdes.2019.108164. Here, we provide - among others - an overview on the use of topology optimisation from an academic and industrial perspective. This includes a summary of today's software landscape, elucidating some of their capabilities and features (see also supplementary information). We hope this provides valuable insight on this matter.