Sound Wave Separation Method In Three Dimensional Acoustic Field With Single Layer Microphone Array

Near-field Acoustic Holography(NAH)can be used to reconstruct acoustic field distribution,and it has been widely used for sound source identification and localization.But,a limiting condition given by NAH that all the sound sources must be on the same side of the measurement array.When there are sound sources on both sides of the microphone array,researchers have proposed sound wave separation methods based on NAH to separate sound waves.The existing sound wave separation methods can be divided into three different kinds,according to the different front-end devices: the first,double layer microphone array based sound wave separation method,that is using a double layer microphone array to obtain sound pressure in the near field of the target source;the second,double layer particle velocity transducer array based sound wave separation method,that is using a double layer particle velocity transducer array to obtain particle velocity of the media in the near field of the target source;the third,single layer Pressure-Velocity(P-U)transducer array based sound wave separation method,that is using a Pressure-Velocity transducer array to obtain sound pressure and particle velocity at the same time and the same position.All of the above methods use a double measured quantity as the input quantity,in other words,they either use a double layer array to obtain same kind acoustic quantity or use a single layer array to gather different kinds of data.Compared with the measurement method of the single physical quantity and single measurement array in NAH,the existing measurement method is not only much more burdensome and expensive but also very hard to carry out.It's an obstacle that must be overcome on the way to industrial applications.Thus,a sound wave separation method with single physical quantity and single measurement array is presented.In this method,a single layer microphone array is used to obtain acoustic information of the sound field.The superposition of spherical harmonics in different orders is used to approximate the sound pressure distribution in the proposed sound wave separation method.So,the measurements array with all kinds of geometry shape can be used to obtain sound field information.The main work of this dissertation is as follows:(1)The development and features of the existing NAH methods,which is the theoretical fundamentals of the three-dimensional sound wave separation method is introduced firstly.Then the state-of-the-art of the three-dimensional sound wave separation methods are discussed,and the background and significance of the selected topic are presented.In order to know more about the superposition theory of the spherical harmonics in different orders,some simulation analysis based on the superposition theory is given.(2)With modeling of acoustic field and theoretical derivation,according the different method of data processing,the superposition-approach and twice-reconstruction based sound wave separation methods with single layer microphone array are proposed,respectively.In the method,sound field distribution is depicted with superposition of finite spherical harmonics,and a single layer microphone array is used to obtain sound field data.And the sound pressure distribution of the target source is obtained after sound wave separation.(3)Using the superposition-approach based sound field separation method,three cases are investigated by numerical simulation: firstly,the target source and interference source are radially oscillating sphere and transversely oscillating rigid sphere,respectively;secondly,the target source and interference source are radially oscillating sphere and forced vibration simple supported plate,respectively;thirdly,both of the target source and interference source are transversely oscillating rigid spheres.The relationship between the frequency and reconstruction accuracy is studied,and the separation method is also examined in an anechoic chamber with two symmetrical speakers.(4)For the cases that high resolution and phase information are not necessary,a sound field separation method with 5-points array is proposed,which is simplified from superposition-approach based sound field separation method.In the method,a five-microphones probe is used to achieve acoustic field data.The method is examined numerically in sound field with spherical sound sources,and it is also examined in an anechoic chamber with mostly spherical and combined speakers.(5)With the twice-reconstruction based sound field separation method,the sound field generated by a radially oscillating sphere and a transversely oscillating rigid sphere is examined by numerical simulation.Firstly,the method is compared with superposition-approach based sound field separation method,then some parameters are investigated,that is the relative source strength,the size of the measurement array,the shape of the measurement array,etc.In the experiments,an acoustic field with asymmetric sound sources is investigated firstly,and then a similar frustum of a square pyramid array is designed and manufactured aiming to improve reconstruction accuracy.At last,the separated results by the designed array are compared with the results by the planar grid array in an anechoic chamber.(6)Experimental verification of the sound wave separation method in water medium.An underwater sound wave separation prototype system is developed based on the acoustic field distribution features of enclosure space of water.The prototype system includes planar hydrophone arrays,a reverberation water tank,an anechoic water tank,a data acquisition module,and a data post processing system,etc.A reverberant field is examined by the twice-reconstruction based sound field separation method(the hydrophone is microphone for water medium),and the separated sound field distribution is compared with the directly-measured sound pressure in the anechoic water tank.Due to the limiting of experimental conditions,the experimental verification of the prototype system is carried out at the frequency span from 4 kHz to 8 kHz.For the sound field separation in lower frequency span,it can be remedied by changing the position of the transducers and the size of the measurement array.Finally,a summary of the full text is given.The main conclusions and innovational points are presented.And also some advices are given for the future research for sound wave separation based on single layer microphone array.