Application of Acoustical Method to Characterize Nonwoven Material

Abstract

Some structural parameters, such as tortuosity, viscous and thermal characteristic lengths, are difficult to obtain through direct measurements. Existing indirect methods, i.e. acoustical method, make it possible to estimate these parameters. This paper presents an application of acoustical inversion methods for estimating structural parameters of polyester nonwoven materials. A four-microphone impedance tube was used to measure sound reflection and transmission coefficients. The inversion methods used in this research are least square and Bayesian approaches. The least square method was achieved via Nelder-Mead algorithm. The Bayesian inversion process was conducted with Metropolis-Hastings algorithm and maximum a posteriori. Inversed parameters from two methods as well as front and back sides of nonwoven fabric were compared. Moreover, inversed porosity and airflow resistivity were compared with measured values. The results show that the least square and Bayesian methods has a good agreement. Estimated parameters from Bayesian method were selected for further analysis. A sizable differences on thermal characteristic length were found by comparing the values from two sides, while the differences are relatively small for other parameters. It is also found that the inversed porosity, tortuosity are reasonable. The results suggest that the acoustical inversion methods can be used to accurately characterize polyester fibrous materials.