An advanced evolutionary strategy with an adaptive penalty function for mixed-discrete structural optimisation

Modular three-dimensional steel frames are prefabricated structures with slender elements and dominating compressive forces. Their analysis includes different non-linearities and the optimisation variables are often mixed-discrete or topological. The optimisation of these systems requires the most advanced solution techniques but also a systematic preprocessing, a sophisticated non-linear structural analysis and an automated handling of stress- and displacement-constraints as well as further side-constraints resulting from the underlying codes of practice. In this contribution, a (μ+1)-evolutionary strategy combined with an adaptive penalty function and a special selection scheme is presented to solve this optimisation problem.