Multi-objective optimization of asymmetric acoustic transmission with periodical structure.

Asymmetric acoustic wave propagation is important for control and manipulation of the acoustic wave signals in various devices. However, previous approach to find optimal asymmetric acoustic transmission (AAT) is through repeatedly adjusting the geometrical parameters, thus causing time-consuming. Here we propose a study on the multi-objective optimization of the AAT, aiming to achieve the widest working frequency range (fr) and the highest transmittance peak (η) with regard to the design variables. For this purpose, the Radial Basis Function (RBF) neural work and finite element (FE) method are applied to obtain the valuable data in the procedure. Furthermore, local sensitivity analysis of design parameters on the fr and η are analyzed. Ultimately, the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) is adapted for getting the Pareto-optimal solutions. The optimization results show great improvement for the overall performance of the AAT, which could be potentially significant in designing various sound devices.