Overview
As many people work, learn, study and live in open plan spaces, it has become important to understand how the acoustic properties of these spaces influence the perception of the final architectural solution. In particular, the design of spaces such as theatres, concert halls and art spaces is strongly related to the performance criteria and linked to the programmatic, aesthetic and acoustic requirements. Therefore, it is vital for the architects/designers to integrate, at an early stage, the acoustic performance parameters such as reverberation time, decay time, initial time, centre time or lateral fraction coefficient, which are all essential parameters for these building typologies.
Looking at the design evolution of these spaces it would be noticeable that the design is developed following a consecutive order; from initial design objectives to an acoustic evaluation with feedback on acoustic performance usually given after completion of the design, and limited interdisciplinary exchange informing design iterations that would become a parameter to other disciplines. In addition, the understanding of spatial and acoustic performance reviews architecture as a “cultural expression that derives its lifespan from the reflective ability to address a change”.
During the last decade, the system logic of computational design has fostered a number of interdisciplinary approaches, which investigate performative strategies such as of their emergence in architecture and biology or of (self) formations shared between architecture and structural engineering. Consequently, a performative design emerges exploring a reverse mapping process of a given or desired performance requirements towards a formal representation realised in geometry and materiality. Considering this, the role of the designer shifts towards a design system, which explores an area of possible design iterations that meet the required performance criteria and “unfold initially not entirely anticipated performance capacities”.
Specifically in the context of performative cultures, “the acoustic consequences of generative and structural form variations open architectural space to material interaction, human perception and affect”. The current project investigates how the acoustic properties of a space can be used as a form generator and inform spatial concepts whilst the acoustic performative criteria are being satisfied. The project’s key goals are to design, to deploy and to evaluate simulations along with interfacing digital technologies in a performance environment. Thus, the research addresses synergetic qualities of the physical and the digital for a culturally and experientially rich environment.
Acoustic Volumes
Algorithim Design
The Genetic Algorithm’s conceptual construct, developed by John Holland in the 1960’s and 1970’s, mimics the evolutionary processes in nature by populations, reproduction and heredity with the inherent ability for the designer to alter several parameters within the method such as population size, crossover technique and mutation rate. Many literature sources can be found on the subject by e.g. John Holland (1992), David Fogel (1997, 2000), David Goldberg and Kumura Sastry (2002, 2005) illustrating not only the subject’s diversity on application, but also its growing importance as a probabilistic solver for singular and multi-objective problems.
The basic principle of an acoustic evolutionary design system is relatively simple. It is based on a repetitive process, which creates geometry, assigns acoustic properties to the geometry, calculates the acoustic behavior, evaluates the acoustic quality of the created individuals and manipulates geometry and properties until acceptable results are produced. The design explorer is basically a twofold process, in which the geometry-synthesis-algorithm creates geometry based on a genetic code using tetrahedrons in Processing.
The evaluation process, on the other hand, assigns quality values to each individual primarily based on acoustic criteria. Optimisation of acoustic aspects within the design process asks for a fitness function, which searches a design specific intention that can be described as a number - as a target for the algorithm. To prove the concept, the acoustic evaluation has been limited to reverberation time, RT60 as outlined in Sabine’s equation.
The main parameters that affect the performance of the genetic algorithm are the ‘population size’ ( - the amount of genomes that can be selected and reproduced from), the cross over technique ( - how the information from each genome is paired to become the next generation’s offspring) and the mutation rate ( - the percentage of how often a random alteration to a genome occurs). The genotype - the evolutionary algorithm - controls the phenotypic behavior and progression which, within this work, can be observed in the evolving volume that is geometrically restrained within an x,y,z-domain.
Properties related to the acoustic behavior have to be defined within the genetic code, in order to enable a subsequent analysis of the acoustical performance. The material related properties, which are relevant for acoustic performance, are the absorption value a. However, instead of varying the value in the preliminary simulations, the geometric representation of a triangular reflector system has been explored. The main aim has been to maximise the reflection count between the reflectors without sending the sound back into the space.
Schematic of Genetic Algorithm
Form Evolution
Almost every shape can be approximated to a desired precision with simplicial complexes, which are mathematically defined as instances of a k-simplex. Almost every arbitrary surface, even if it is double curved, can be approximated with a grid of the 2-simplexes. A tetrahedron is an instance of such a k-simplex which means that every volume can be more or less approximated with tetrahedrons.
Simplex
Acoustics
Reverberation Time
Sabine’s equation describing the amount of time it takes
for the sound pressure to decrease to 60 dB after the
sound source is terminated, RT60
RT60 = Ta = 0.16* V / Sa
where:
V: the volume of the room in m³,
S: the total surface area of the room in m²,
a: the average absorption coefficient of the room surfaces,
and the product
Sa: the total absorption in sabins
Acoustic Surfaces
Non-orthogonal reflection
R = 2N(N∙L)-L
where:
R = the reflection vector
N = the surface normal
L = the incident vector
N∙L = the dot product of
the surface normal
and the incident
vector
Processing Simulations
Design Outcomes
Population =100 & 4 genes, RT60 = 2s, a = 30%, mutation = 5% - Minimise Sabine’s equation & Without geometric restrictions, Outcome = 4 points
Population =100 & 4 genes, RT60 = 2s, a = 30%, mutation = 5% - Minimise Sabine’s equation & Maximise distances between points, Outcome = Tetrahedron
Population =100 & 4 genes, RT60 = 2s, a = 30%, mutation = 5% - Minimise Sabine’s equation & Maximise distances between points, Outcome = Tetrahedron with extended design domain
Population =100 & 5 genes, RT60 = 2s, a = 30%, mutation = 5% - Minimise Sabine’s equation & Maximise distances between points, Outcome = Polyhedron - 2 Tetrahedra
Population =100 & 6 genes, RT60 = 2s, a = 30%, mutation = 5% - Minimise Sabine’s equation & Maximise distances between points, Outcome = Polyhedron - 3 Tetrahedra
Population =100 & 7 genes, RT60 = 2s, a = 30%, mutation = 5% - Minimise Sabine’s equation & Maximise distances between points, Polyhedron - 4 Tetrahedra
