Methodology for optimizing composite design via biological pattern generation mechanisms

Abstract

Mechanistic capabilities found in nature have influenced a variety of successful functional designs in engineering. However, the unique combinations of mechanical properties found in natural materials have not been readily adapted into synthetic materials, due to a disconnect between biological principles and engineering applications. Current biomimetic material approaches tend to involve mimicking nature's microstructure geometries or mimicking nature's adaptive design process through brute force element-by-element composite optimization techniques. While the adaptive approach promotes the generation of application-specific microstructure geometries, the element-by-element optimization techniques encompass a large design space that is directly related to the number of elements in the system. In contrast, a novel methodology is proposed in this paper that merges biological pattern generation mechanisms observed in the Gray-Scott model, with an evolutionary-inspired genetic algorithm to create adaptive bio-inspired composite geometries optimized for stiffness and toughness. The results reveal that this methodology significantly reduces the optimization parameter space from tens of thousands of parameters to only four or five. In addition, the resultant composite geometries improved upon the overall combination of stiffness and toughness by a factor of 13, when compared to published brute force element-by-element techniques.