Interval uncertainty propagation in metamaterial beams using machine learning based optimization

Metamaterials have recently emerged in the search for lightweight noise and vibration solutions. One of their appealing properties control engineering is ability to create stop bands, which are frequency ranges without free wave propagation. These bands arise from sub-wavelength addition identically tuned resonators or on a host structure. However, when manufacturing metamaterials, variability material geometry inevitably introduced. On one hand, metamaterial attenuation performance can deteriorate due variability, while other even broaden typically narrowband performance. During early phases design, often little information concerning inherent available, it would be desirable able assess input uncertainty effects metamaterials. This work focuses numerical assessment impact uncertainties resonator beams. To this end, machine learning based optimization strategy, so-called Bayesian optimization, employed developed non-intrusive propagation approach. enables very efficient evaluation upper lower bounds performance, with uncertain parameters defined as interval variables.