Mechanical characterization and design of square honeycombs with the aid of additive manufacturing and AI

Abstract

Metamaterials offer a viable mean to attain targeted mechanical characteristics tailored to particular loading conditions. Aperiodic metamaterials provide higher tailorability of mechanical behavior by providing a customizable deformation mode, properties, and mechanical response. Artificial intelligence has enhanced metamaterial design by discerning correlations between parameters and mechanical characteristics. This work studies two types of gradation on square honeycombs: wall thickness and wall angle. The studied gradation characteristics were wall inclination, pattern distribution, and direction. Fifteen designs were proposed, each combining different gradation characteristics. The designs were additively manufactured with PLA on an FFF 3D printer and experimentally tested under compression. The effects of the gradation characteristics on the mechanical response, mechanical properties, and deformation mode were analyzed. The results confirmed the influence of gradation on the mechanical behavior of the structures. The gradation characteristics influence specific properties or responses, such as a 30% energy absorption difference between graded honeycombs with aligned and not aligned walls. The metamodel evolutionary optimizer (MEVO) algorithm was used to assist in the design of a tailored square honeycomb with an angle gradation to minimize the displacement of a designated point in the structure. The algorithm was tested on multiple nonconventional loading scenarios to prove its versatility.