Article Preview
TopIntroduction
Various methods to optimize the acoustic environment to specific design criteria that meets a pre-conceived design proposal exist, using various mathematical and geometrical prediction methods that are adapted to indoor environments (Ismail, 2013) or external propagation modeling (Ismail & Oldham, 2002; 2003; 2005). However approaches to adapt to a certain acoustic requirement used a search method that alters the design to reach an optimal target performance. Through this approach acoustic notions are used to drive the morphogenetic process of the building design. These search methods could be Neural Networks, Simulated Annealing and Genetic Algorithms (Brownlee, 2011). Genetic Algorithm developed by John Holland in the 1960’s and 1970’s (Holland, 1992) has been chosen by (Isak et al, 2012) to switch from a conventional ‘model-simulate’ approach to a ‘generative-model’, more on GA could be found in the work of Fogel (1997; 2000) and Goldberg (2002).
In their work the three essential operational parameters namely a)the fitness function, b)Algorithm variables and c)Genotype and phenotype, are set according to acoustic optimum design aspect, in this case Sabine’s equation for RT, and the geometric parameters of the site and the internal building shell. They found employing the GA model with an organizational algorithm, that acoustic aspects can be used in the morphogenetic of the building shape, and to generate spatial volumes and expressive surfaces.
Based on performance-driven morphogenetic (Banda, 2010) Pablo developed a modular acoustic ceiling based on the design of Helmholtz Resonators (Ismail, 2013), the results of a physical model measurement positively supported the modeling process.
In another attempt to make architectural design acoustically driven, (Peters, 2009) introduced possibly a wider spectrum of solutions, namely absorbers, resonators, and diffusers that were integrated in his modeling environment. He concluded that if calculation routines of acoustic prediction software such as ODEON (Lynge, 1999) could be accessed, this could form an exciting new performance driven architectures if integrated as a plug-in in CAD. Brady further demonstrated in his work (Peters & Brady, 2011) how his model was used in terms of six parametric design strategies to design a school project in Copenhagen, Denmark. It was practically demonstrated that the acoustic performance can guide morphology and material strategies selection in the design, and through this approach performance can be shape linked. In further extension to his work (Brady et al, 2011) he designed a responsive acoustic surface that was designed with fixed acoustic criteria for a decided sonic experience. In this work the parametric design model was linked to fabrication machinery, which required to waiver some design ideas, since the complexity of linking shape, performance and fabrication formed a competing demand and the balance could not be easily achieved.
It is obvious that the previous work demonstrates the linking of acoustic performance and the morphological aspects of the building shape. Whether in some cases external shapes were performance driven or internal shapes to reach a decided sonic experience. For all the cases outlined the resultant is a fixed shape that responds to a single ended acoustic performance environment. However in most cases in open plan design and multipurpose halls in buildings there is a need to optimize the sonic experience to suit a changing use of a multipurpose element.