Previous research on the use of Multi-Objective Genetic Algorithms (MOGAs) for the shape optimization of concert spaces (Mendez Echenagucia ) shows the acoustic potential of freeform continuous surfaces with both convex and concave areas and the difficulty of achieving high standards with synclastic shapes. Thrust Network Analysis (Block 2009) is capable of generating compression-only shapes that go beyond synclastic surfaces, thus presenting an opportunity for the use of funicular structures in concert auditoria.
This researchs uses multi-Objective Genetic Algorithms (MOGAs) for the acoustical optimization of compression-only shell structures. The MOGA combines ray-tracing simulation for the study of the acoustical quality of the spaces with TNA for the generation of freeform funicular shapes. The method is exemplified in the case study of the multipurpose room in Barranquilla, Colombia.
The MOGA employed in this research uses force densities in the TNA framework as design parameters to control the shape of the ceiling. In other words, the geometry is not controlled with a parametric model that moves control points to modify the shape but rather by modifying the forces and boundary conditions of a TNA model. In doing so the MOGA generates exclusively compression-only shapes.