Study On Sound Absorption Of Ultra-low Density Plant Fiber Material

Sound absorbing material has attracted more and more attention nowadays with the increasingly serious pollution of noise. The ultra-low density plant fiber material which density falls in between 0.01-0.12 g/cm3 is a kind of environmentally friendly, multifunctional biomass material. It has the following advantages:impact resistance, heat insulation, flame retardant, corrosion prevention, and sound absorption and so on. The internal material forms a three-dimensional network structure, and there are a lot of tiny and interconnected pores forming complex structures between the plant fibers. In this paper, based on the theoretical and experimental research to the sound absorption characteristics of the ultra-low density plant fiber material, the sound absorption property prediction model is established in order to reveal the sound absorption mechanism of the material. And the relationship between the sound absorption property and the material structure parameters is deduced. This study has some significance to the actual engineering design and application.In this paper, the acoustical absorption coefficient of the ultra-low density plant fiber material was measured by standing wave tube method, and the porosity and pore size of the material was computed by image analysis method. The effect of the factors such as density, porosity, pore size, thickness and the cavity behind the material to the sound absorption property of material was discussed. The results showed that the material density, porosity and pore size which either is too big or too small will affect the sound absorption property. There exists an optimum value under certain conditions. Increasing the thickness of the material will conducive to improve the sound absorption property, but the sound absorption property will not be improved obviously when the thickness increases to a certain value. Setting the cavity behind the material will help to improve the sound absorption property of low frequency, and the sound absorption property will be improved with the increasing thickness of cavity behind the material.On the basis of experiments, sound absorption property prediction models of the ultra-low density plant fiber material were established in the following two ways:(1).The mathematical model of the material’s sound absorption property was set up according to the circular tube theory model of porous materials. The acoustical absorption coefficients which calculated by mathematical model were compared with the experimental test ones. The result suggested that the mathematical model had a good predictive ability to the acoustical absorption coefficient of the ultra-low density plant fiber material. In addition, this model can correctly reflect the impact of material structure property variables for the sound absorption property. The mathematical model also can calculate the corresponding thickness and pore size when the material obtains the best sound absorbing property. (2). The prediction model based on the BP neural network theory was built, which took the structural parameters (such as thickness, porosity and pore size) of the ultra-low density plant fiber material as the input parameters and took the average absorption coefficient as the output parameter. The paper studied the following problems during the modeling process:how to build the network structure of the model; how to extract and pre-process the sample data; how to select the transfer function and the training function; how to configure the learning rate and the target error. The average absorption coefficients which exported by the BP neural network model were compared with the experimental test ones. The result indicated that this BP neural network model also has good prediction accuracy for the sound absorption property of the ultra-low density plant fiber material.