Optimization of Graded Metamaterials for Control of Energy Transmission Using a Genetic Algorithm

• DOI: 10.48550/arxiv.2306.10628
• Publication Date: 2023-01-01
• Authors: Joshua Morris; Weidi Wang; Thomas Plaisted; Christopher J. Ha...; Alireza V. Amirkhizi
• Keywords: Maxima and minima

Optimization of functionally graded metamaterial arrays with a high dimensional and continuous geometric design space is cumbersome could be accelerated via machine learning tools. Mechanical metamaterials can manipulate acoustic or ultrasonic waves by introducing large dispersive attenuative effects near their natural frequency. In this work structures are designed optimized to combine the energy attenuation performance number unit cells varying frequency responses reduce interlayer mismatch effects. through genetic algorithm avoids many local minima related dimensionality but requires iterations. A reduced order model (ROM) applied that reproduce transmission response traditionally calculated FEM, in fraction time. Pairing GA ROM together, an array 6 (with total 18 independent variables) have stop bands extended width sharper boundaries. Symmetric determined optimal configurations. Measured 3D printed features projected onto solutions quantify effect printing uncertainty on performance. Repeatability error $\pm~20$ $\mu$m mean depth band factor $10^2$ introduce small shifts center width. Proposed methods improve resolution accessible points space, sensitivity uncertainty, add freedom include out-of-plane perforations constituent materials using tunable filled resin systems.