Optimization Design Of Low-frequency Sound Absorber Based On Metamaterials

With the improvement of living standards,our tolerance for noise pollution is decreasing daily.Noise pollution not only affects normal life and work,but also causes continuous damage to people's physical and mental health.Researchers have been looking for a variety of technical methods to reduce noise pollution,which can be summarized as methods such as suppressing sources intensity,obstructing transmission,and increasing absorption.Limited by the Theory of linear correspondence,it is difficult for traditional absorber materials such as porous materials and micro-perforated plates to effectively absorb low-frequency sound waves.Moreover,because low-frequency sound waves have great ability of diffraction and penetration,how to effectively vanish low-frequency noise in the air is always a difficult problem.On the other hand,with the improvement of comprehensive national strength of our country,the national maritime security strategy has become increasingly important.As the only effective carrier of energy and information in the marine environment,soundwave is an important means for communication and detection.Therefore,the anechoic coating has great strategic value in many occasions such as sonar detection and submarine stealth.Traditional anechoic coating generally adopts rubber or polyurethane,but due to the larger wavelength than that in air sound and the complicated working environment,traditional absorbers can only get a better absorption at high frequencies of megahertz or even tens of megahertz.How to achieve the absorption of sub-kilohertz low-frequency underwater sounds by a subwavelength thickness is a difficult problem to be solved urgently.In recent years,the research enthusiasm of acoustic metamaterials has been increasing.Through the resonance of the internal structure,low-frequency acoustic energy can be highly localized in subwavelength metamaterials,which provides the possibility for further absorption of sound.Therefore,whether in the field of airborne sound absorption or waterborne sound absorption,acoustic metamaterials have shown broad application prospects.In this context,this thesis studies deep subwavelength absorbers based on acoustic metamaterials in two fields: airborne and waterborne.The specific contents are as follows:The first chapter briefly introduces the basic concepts and development history of acoustic metamaterials,followed by an overview of the current research status of absorbers in airborne and waterborne.In the second chapter,for the 2D cylindrical wave sound field,a ventilated airborne sound absorber based on a curled spatial resonator is proposed.Since most of the literature of sound-absorbing are based on the plane wave assumption,this chapter firstly deduces the theory absorption of the cylindrical wave sound field.Secondly,on the basis of the equivalent medium model,using the transfer matrix method and the finite element method,the theoretical analysis method of the absorption performance of the toroidal acoustic metamaterial is studied.And then,based on the coiled space resonator,a subwavelength circular ring absorber is proposed,and the structural parameters are optimized with the help of genetic algorithm.In the end,a deep subwavelength absorption more than 93% can be achieved with a thickness of 30 mm at 303 Hz,and its wavelength-thickness ratio is ?/d?38.In the third chapter,we firstly propose a subwavelength ultrathin sound absorber.And based on the deep neural network,a forward predictive neural network is trained,which can quickly predict the absorption spectrum of a specific sound absorber,and its calculation speed is three orders of magnitude higher than that of the transfer matrix method.On the basis of combining the forward predictive neural network,the backward design neural network is further proposed,which can reverse design the specific target spectrum and directly give the corresponding geometric parameters.Verified by the matrix transfer method,both networks show a high accuracy.The research results show that the deep neural network can be used to quickly predict the sound absorption performance of acoustic absorbers with complex structures,and it can also be used to quickly reverse design of absorber with target absorption spectrum.In the four chapter,we propose an ultrathin underwater sound absorber based on pentamode material(PM)and rubber-metal resonance(RMR),which achieves deep subwavelength absorption at extremely low sub-kilohertz frequencies.With a thickness of 63 mm,an average sound absorption coefficient more than 87.8% is achieved in the frequencies band from 760 Hz to 920 Hz(the wavelength is 30.9 to 25.5 times the material thickness).In the composite materials,the PM layer acts as an energy converter,which can completely convert the incident underwater sound into its internal mechanical vibrations.When the frequency of the incident wave is close to the resonance frequency of the RMR on the back side of the PM,the energy of the mechanical vibration will be highly localized in the RMR,and then be converted into heat and dissipated by the the intermolecular friction and relaxation effects inside the viscoelastic rubber.So as to achieve subwavelength absorption of underwater sound.In addition to the energy converter,the PM itself will also produce a kind of string-like vibration due to the introduction of fixed boundary conditions,which will also drive the RMR to produce secondary vibrations,thereby absorbing mechanical vibration energy.Finally,we also show the robustness of the composite materials to a wide range of oblique incident angles.The fifth chapter summarizes the work and innovation point of the thesis,and puts forward the prospects for the future development of the work.