| Active noise control technology is the most important means to reduce low frequency noise,However,in the process of propagation,the sound wave of the loudness radiated by the noise source in engineering will show the signal distortion,and the nonlinear higher harmonics are excited.At the same time,the characteristics of different noise reduction fields and the structure of complex noise reduction system also introduce the interference of unknown nonlinear factors in the sound transmission channel,the performance of active noise control system is seriously affected.To solve this problem,this paper adopts the methods of theoretical derivation,signal simulation and indoor noise reduction experiment.Based on the analysis of the distortion propagation of engineering noise and the characteristics of nonlinear propagation channel,an active noise control method based on linear average control and nonlinear function expansion combination is designed.In addition,the modeling and analysis of the secondary propagation path in the noise reduction system are improved.At last,an adaptive variable-step iterative algorithm based on tracking error is proposed to update the control coefficient of the system and improve the noise reduction performance of the system.In this paper,the reference signal of the noise source is extended to the high-dimensional space by the orthogonal trigonometric function,and is used as a new input vector of the noise reduction system,so as to enhance the nonlinear expression of the distortion characteristics of the noise source signal by the noise reduction controller.At the same time,the mapping relation between the noise source signal and the high-order harmonics excited by the noise propagation channel and the sum and difference frequency sound waves caused by the interaction of different frequencies is established to control the nonlinear factors in the process of noise reduction.The FIR linear average controller is combined with that to construct the nonlinear composite structure feedforward active noise controller.Through comparative analysis,the expression mode of the extended structure of orthogonal trigonometric function has better noise reduction and cost performance compared with other extended structures of orthogonal function and polynomial.When modeling the secondary path excited by the classical white noise signal generator,the error increment of the noise reduction system is caused,and the output power of the whitenoise is adjusted by the constraint of error tracking.In order to reduce the influence of the generator and the inherent structure of the system on the noise reduction performance,the correction control of the error transmission channel is established to meet the demand of high-precision modeling of the signal transmission channel.In order to balance the contradiction between the convergence rate and steady state performance of the whole noise reduction system,the hyperbolic tangent function is used as the expected ideal functional relation between the residual error noise signal and the iteration step size of the system.The influence proportion relation of the coefficient factor of the function is analyzed and an adaptive step size control coefficient updating algorithm for noise reduction system is established.The active noise control system of the combined structure is used to carry out noise reduction experiments on local static areas in bounded space,and various engineering noises such as air compressor and ball mill are collected as target objects,to simulate the noise environment of the strong acoustic quantity in the field,the system has achieved effective low-frequency energy control and significant attenuation of the excited high-order harmonics.The noise reduction performance of the system is stable for noise signals below 500 Hz,and it has a good ability to identify the channel of sound propagation in bounded space.The conclusion of active noise control system based on composite structure is verified. |
