Multidisciplinary design optimisation as strategy for building design

Abstract

In this research project, an approach is proposed for applying multidisciplinary design optimisation on building design problems. The software used in the approach are Rhinoceros and Grasshopper for the parametric configuration of the geometry, the plug-in Octopus driving the optimisation process with its multi-objective genetic algorithm and a custom post-processing tool PPOR for comparing and filtering different results and investigating the trade-offs between different objectives. To test the functionality of the proposed approach, it was applied on a relevant case study. The large span steel roof of the Ukrainian Donbass Arena was chosen as test case. Focus was given on the structural and acoustic performance of the roof. The structural performance was described by the structural mass of the roof, whereas the acoustic deviation throughout the stadium was measured for the acoustic assessment. The plug-in Karamba was used for the structural analysis and a custom ray-tracing tool, SPL-Estimator, was developed for the acoustic analysis. A parametric model of the football stadium was set up, in which the geometry of the roof was described by eleven design variables.