Research On Parametric Array Loudspeaker Based On Nonlinear Inverse Control Algorithm

Parametric acoustic array is a popular research direction in nonlinear acoustics.At present,parametric acoustic array is widely used in underwater communication.Based on the good directivity of signal in water,parametric array acoustic can complete the tasks of signal transmission and target detection.When it is applied in the air,along with the nonlinear effect of the air,the sound will be seriously distorted during the transmission process.The purpose of this paper is to propose a method for optimizing the distortion and directivity of speakers.Berktay’s far-field solution only considers the second harmonics due to the nonlinear action of air,and ignores other higher harmonics.KZK equation,the nonlinear acoustic equation,is an accurate acoustic model,but its higher-order analytical solution is difficult to calculate.Neural networks have the property of approximating nonlinear functions.Based on this characteristic of neural network,the acoustic model of parametric array based on neural network is proposed.The nonlinear inverse control is used to optimize the signal distortion.The main research contents of this paper are as follows.In this paper,the directivity of loudspeaker array of different sizes is simulated in3 d simulation software.The influence of the number of array elements on directivity is studied,the directivity of different array loudspeakers is compared and the most suitable design scheme is selected.At the same time,a new method to optimize the directivity of sound is also proposed.The simulation model of parametric acoustic array is established.Acoustic models of parametric array are established by nonlinear autoregressive neural network with exogenous inputs and error back-propagation neural network.And then the inverse system model is built in MATLAB/SIMULINK to carry out nonlinear inverse control of the system.Single frequency,dual frequency,and speech signal were processed by inverse control system.The results show that the inverse control processing completed by the neural network has an obvious effect on the restoration of the signal.Meanwhile,the neural network with a better control effect is further optimized.The directivity test of loudspeaker array is carried out.The directivity test is carried out by measuring sound pressure levels at all angles with a speaker array containing 64 elements.According to the experimental data,the sound produced by the array has good directivity,and the sound pressure at the main lobe is more concentrated.Parametric array loudspeaker can be used in many applications because of the good directivity of the sound.