Absorptive metasurface inaccuracies made of micropore and microslit panels in multiple structures

Abstract

Microperforated panels in forms of micropores/microslits (MPP/MSP) represent absorptive metasurfaces, which can achieve high absorption with single- or double-layer MSP/MPP absorbers. However, architectural acoustics practice usually requires broadband high absorption. Multiple metastructures in more than two layers become one of possible options. Multiple MPP/MSP metasurfaces arranged in layers inherently complicate the design process. This work applies a model-based Bayesian framework employing a potentially multilayered prediction model. The Bayesian framework involves two-levels of probabilistic inference to design a parsimonious number of layers as the higher level, quantitatively implementing Occam’s razor. Once the number of multilayers is selected, the MPP/MSP parameters are readily available within the Bayesian framework, so that the overall metastructure satisfies the design scheme. This work experimentally validates the designed prediction in comparison with the design scheme. However, manufacture inaccuracies may lead to unacceptable deviations. This work applies causation analysis based on a causal model to reveal causal uncertainties/inaccuracies. Using the causal model for causal inference, the Bayesian multilayer design can be satisfactorily validated. This paper discusses comparative investigations of metasurfaces made of micropores and microslits.