Predicting the Sound Absorption Performance of Warp-Knitted Spacer Fabrics via an Artificial Neural Network System

In this study, the sound absorption performance of warp-knitted spacer fabrics (WKSFs) was investigated. WKSFs with different angles of connecting yarns between two surfaces and different numbers of layers were fabricated. The effect of porosity and the number of layers on the sound absorption coefficient (SAC) of WKSFs were measured by the impedance tube method at the frequency range of 0-6100 Hz. The results indicated that increasing the angle of connecting yarns and the number of layers will enhance the SAC of WKSFs. An artificial neural network (ANN) model was also used to predict the effect of knit structure and the number of layers on the SAC of WKSFs at different frequencies. It is found that the ANN model provided an accurate and reliable prediction of SAC for different WKSF structures with a high value of correlation coefficient (more than 0.99%). The obtained results showed that the developed high-precision ANN model would be a helpful and powerful tool for modeling and predicting sound absorption performance of fibrous acoustic materials.

Automated summary

The sound absorption performance of warp-knitted spacer fabrics (WKSFs) was studied. WKSFs with different angles of connecting yarns and numbers of layers were fabricated. The effect of porosity and the number of layers on the sound absorption coefficient (SAC) of WKSFs was measured, and an artificial neural network (ANN) model was used to accurately predict the SAC for different WKSF structures.