Modelling sound absorption properties of broom fibers using artificial neural networks

The use of broom to produce fibers has ancient roots. The Greeks appreciated its resistance to water and for this reason they used it to manufacture sailing ropes. But broom fiber was also appreciated for its sound absorption qualities. In this study, a new methodology was developed for the numerical modeling of the acoustic behavior of broom fibers. First, the characteristics of the different varieties of broom were examined and the procedures for processing the samples to be analyzed were described. Subsequently, the results of the measurements of the following acoustic properties of the material were analyzed: air flow resistance, porosity, and sound absorption coefficient. Finally, the results of the numerical modeling of the acoustic coefficient were reported using an algorithm based on artificial neural networks. The results obtained are compared with a model based on linear regression. The model based on neural networks showed high values of the Pearson correlation coefficient (0.989), indicating a high number of correct predictions.