Optimization algorithm of sound absorbent performance of sintered metal fiber materials based on deep learning

Taking sintered metal fiber porous materials as the research object, several optimization objective functions with broadband sound absorption characteristics are constructed, and their respective characteristics are analyzed and discussed. Finally, neural network and other algorithms are used to optimize the sound absorbent performance of sintered metal fiber materials. Through optimization calculation, The variation rules of macroscopic acoustic parameters such as porosity, tortuosity factor, static flow resistance and static flow resistance of sintered metal fiber materials with different thicknesses are obtained, which can provide reference for engineering application of sintered metal fiber materials. The theoretical model is further expanded to the theoretical model of high-temperature sound absorption of gradient porous metal fiber by applying the acoustic impedance transfer formula, and the multilayer gradient material structure is optimized by combining with the optimization algorithm. The results show that the sound absorption effect of the material at high temperature is slightly worse than that at normal temperature. Porosity and fiber diameter of porous materials have significant influence on the sound absorbent performance of materials. With the increase of porosity or fiber diameter, the sound absorption coefficient of materials decreases in the low frequency band and increases in the middle frequency band. After structural optimization design, under the same conditions, the sound absorbent performance of multi-layer gradient metal fiber material is obviously better than that of single-layer structure, and it has good sound absorption effect.