Computational Design of a nature-inspired architectural structure using the concepts of self-similar and random fractals

•Self-similar and random fractals were used for designing a canopy structure.•The IFS and the Diamond Square Algorithm were used for generating fractals.•Fractal dimension was used as a design variable for developing a crinkled roof.•Finite element analyses show that the crinkled roof is stronger than a flat roof.•Increasing fractal dimension increases the stiffness and weight of crinkled roof.

This paper aims to explore the scope of applying the concept of fractal geometry in the field of architecture and construction. There are mainly two different types of fractals – self-similar fractal and random fractal. In this paper, both types of fractals are used to design a nature-inspired architectural structure with the strategy of exploring the potency of fractal geometry as a geometric framework that can offer new structural forms. Based on the mathematical formulations of self-similar fractal shape and random fractal shape, tree-inspired branching supports and natural terrain inspired unsmooth crinkled roof are modeled using the algorithms of Iterated Function System and Midpoint Displacement (Diamond Square Algorithm) method respectively. Fractal dimensions are calculated to assess the visual complexity of the roof surface and branching supports. Finite element analysis is performed to assess the structural strength of the model with respect to changing of fractal dimensions.