Research On Active Noise Control Algorithm Based On Hybrid Structure

Industrial equipment has been widely used in various occasions of production and life,and the noise generated by these equipments has become a prominent problem affecting people’s life and has received wide attention.In this context,active noise control(ANC)technology has emerged.In many noise scenarios,ANC technology has achieved good noise reduction results.In the development process of ANC technology,there are still some key issues.The broadband and narrowband hybrid structure and the feedforward feedback hybrid structure are common algorithmic structures in the ANC field.In linear environment,the broadband and narrowband hybrid system does not have an ideal narrowband noise reduction effect,the feedforward and feedback hybrid system has a limited ability to cancel incoherent noise,and has a poor ability to resist impulsive interference.In the nonlinear environment,neither of the two conventional hybrid systems has the ability to adapt to the nonlinear environment.Facing the complex noise environment in the nonlinear environment,the noise reduction ability of the conventional hybrid system is weak.Aiming at the above problems,this paper investigates the new hybrid structure ANC algorithm.In order to cancel the broadband and narrowband mixed noise signals in a linear environment,the IHANC system is proposed in this paper.In this paper,a new weight vector updating formula is proposed,which can be adjusted by adjusting the parameters to adjust the updating formula,and by applying the new formula to the HANC system,the IHANC system is constituted.The noise reduction performance of this system is significantly improved and its attenuation ability for narrowband signals is optimized.In order to attenuate the mixed broadband and narrowband noise in nonlinear environment and to improve the nonlinear adaptation ability of the algorithm,this paper proposes the IHFLANN system.Based on the HFLANN system,this paper applies the wavelet packet structure to the broadband subsystem,which improves the noise reduction performance of the broadband subsystem,and applies the FLANN filter to the narrowband subsystem,which gives the narrowband subsystem the ability to adapt to the nonlinear environment.In addition,this paper adds the FELMS algorithm and M-max selector to the IHFLANN system to form the SIHFLANN system,which reduces the computational complexity of the algorithm.In the linear environment,in order to cancel the incoherent noise and to improve the impulsive resistance of the algorithm,the MVC-HANC system is proposed in this paper.In this paper,the weight vector update formula of the FXMVC algorithm is added to the CHANC system,the error signal is compressed,at the same time,the sigmoid function is used to constrain the reference signal,and the exponential function is used to construct the variable step function,which improves the robustness of the CHANC system by the above three signal processing methods.In the impulsive environment,as the attenuation performance of the feedforward subsystem improves,the performance of the feedback subsystem is also enhanced.In order to attenuate the incoherent noise at the error microphone in nonlinear environment and to improve the nonlinear adaptation capability of the algorithm,the DWT-HANC system is proposed in this paper.In this paper,the FLANN filter is introduced into the CHANC system,which gives it the ability to adapt to the nonlinear environment.In addition,this paper applies the DWT structure to the feedforward subsystem and the feedback subsystem,which effectively improves the overall noise reduction performance of the system.In order to make the algorithm have the ability to resist impulsive noise,this paper proposes a new type of error signal compression function and variable step size strategy.Finally,this paper introduces an error signal separation subsystem to solve the error signal coupling problem between the feedforward subsystem and the feedback subsystem,which further improves the overall noise reduction performance of the system.In a nonlinear environment,in order to counteract the broadband and narrowband mixed noise and to make the ANC algorithm resistant to impulsive noise and incoherent noise,the FLANN-ECHANC system is proposed in this paper.In this paper,the C-IFXtanh LMS algorithm is formed on the basis of the convex combination algorithm by adding the error signal compression function and the reference signal constraint function,which makes the convex combination algorithm resistant to impulses.In this paper,the C-IFXtanh LMS algorithm is used to replace the broadband subsystem in the HANC system,and the SRE subsystem is added to form the Convex-HANC system,which gives the HANC system the ability to deal with the broadband and narrowband mixed noise in the impulsive environment.In this paper,the feedback structure is added to the Convex-HANC system to form the EC-HANC system,which gives the HANC system the ability to resist incoherent noise.In this paper,the FLANN structure is added to the EC-HANC system to form the FLANN-ECHANC system,which gives it the ability to adapt to the nonlinear environment.In this paper,dSPACE is used as a controller to build an open space active noise control test platform.On the basis of this test platform,this paper formulates the test program of ANC algorithm and builds the dSPACE test model of the ANC algorithm proposed in this paper on the software platform.In this paper,the simulated noise and real vehicle noise are used as the reference noise signals,and the ANC algorithm is verified in order to verify the actual noise reduction effect of the algorithm.The EC-HANC system proposed in this paper has the better ability to deal with complex noise in the linear environment,and the ANR results of the system can reach-1.3～-1.1 dB when using the simulated noise test,and when using the vehicle noise test.The ANR results of the system can reach-1.8～-1.2 dB.The ANC algorithm test results prove that the proposed algorithm has better noise reduction performance.