Topology optimization of multi-material acoustic metamaterials for low-frequency and broadband sound absorption

• This paper uses a non-gradient topology optimization method to design multi-material acoustic metamaterials. • We propose a data filtering method for eliminating the checkerboard phenomenon. • The adaptive fuzzy fitness granulation method is introduced to reduce the algorithm’s calculation time. • Compared with porous materials, the absorption bandwidth of the final optimized result increased by 47.19%. • The optimized algorithm is proved to be feasible by experiments. This paper uses a non-gradient topology optimization method for designing multi-material acoustic metamaterials with low-frequency and broadband sound absorption performance. The method combines a genetic algorithm and a topology optimization method to design appropriate cavities and rigid structures with irregular shapes. The finite element method is used to calculate the sound absorption coefficients, and its accuracy is verified using experimental methods. A cell filtering rule suitable for three-phase materials is proposed to eliminate the checkerboard phenomenon. A data processing method with a penalty mechanism and an adaptive fuzzy fitness granulation method are introduced to improve the convergence effect and the convergence speed of the algorithm. The research results indicate that structures with similar sound absorption performance are generated under different initial structures and iteration loops. The porous material matrices with poor sound absorption performance can also obtain topology structures with good sound absorption performance. Compared with the two-phase material topology structures, the three-phase material topology structure has better sound absorption performance. The final optimized structure designed by a three-stage optimization strategy exhibits excellent low-frequency and broadband sound absorption performance, which remains stable at different incidence angles. The research method in this paper provides some ideas for multi-material topology optimization.