Inverse design of periodic cavities in anechoic coatings with gradient changes of radii and distances via a conditional generative adversarial network.

Anechoic coatings are usually applied to underwater targets, such as submarine shells, to reduce the detection distance of enemy active sonar. The main challenge is obtaining low-frequency and broadband sound absorption characteristics through the design of material parameters and geometric structures. In this study, the low-frequency and broadband sound absorption performance characteristics of anechoic coatings were assessed. Design research of the material parameters and cavity geometry structures of anechoic coatings was conducted through deep learning. An inverse design method based on a conditional generative adversarial network (cGAN) was proposed to address the difficulties in quantitatively designing variable radius and distance gradient parameters. A dataset comprising 86,400 sets of material and structural parameters and corresponding sound absorption coefficients was constructed to train and test the cGAN model. The optimal model was obtained after 360 epochs of training. A case study is provided to improve the broadband sound absorption performance of a three-band sound absorber, demonstrating the effectiveness of the proposed cGAN model in the structural design of gradient combined cavities in anechoic coatings. The results revealed that reducing the radius of the first cavity while increasing the radii of the second and third cavities, as well as the elastic modulus and loss factor, can enable the coupling of and resonance between the first and second cavities, thereby increasing the sound absorption performance within the 2640 to 6000 Hz frequency band. The cGAN model can be used to efficiently, accurately and quantitatively design the structures of gradient combined cavities and materials in anechoic coatings.