Research On Hybrid Active Noise Control Algorithm In Vehicle Based On Momentum Term

The development of the automobile industry has improved people’s quality of life.However,the noise poses a significant risk to the physical and mental health of drivers and passengers.The low-frequency noise in the automobile is distinctly characterized by broadband and narrowband mixture,with the majority of the noise coming from the engine and the road.Passive noise control(PNC)technology can be used to eliminate med-high frequency noise,and the noise reduction effect on low-frequency noise is not satisfactory.Active noise control(ANC)technology effectively makes up for the shortcomings of PNC technology and can achieve effective noise reduction of low-frequency noise in the vehicle.Broadband-narrowband hybrid active noise control system handles both the broadband and narrowband components of noise and has great potential for reducing noise levels in cab.This research focus on improving the comprehensive performance of the broadband-narrowband hybrid ANC system,and study the broadband ANC algorithm,narrowband ANC algorithm,and broadband-narrowband hybrid ANC algorithm applicable to in-cab environment noise reduction.The details of the study are as follows:Firstly,to improve the convergence performance and noise reduction performance of the broadband ANC system,a variable momentum normalized filtered-X least mean square(VMNFXLMS)algorithm is proposed.The momentum term technique and the normalized variable step algorithm are introduced into the ANC algorithm to speed up the convergence of the algorithm by incorporating past weight information in the process of adaptive weight coefficient update and making the convergence coefficients adjust adaptively with the reference signal power.In addition,in order to solve the problem of elevated computation of the algorithm brought by the momentum term technique and variable step size algorithm,we further propose the set membership VMNFXLMS(SM-VMNFXLMS)algorithm,which achieves the reduction of computational complexity by selectively updating the weight coefficients of the filter.To address the problems of slow convergence and poor tracking performance of the traditional notch FXLMS algorithm,we propose a variable momentum smoothed filtered-X least mean square(VM-SFXLMS)algorithm.By introducing a smoothing factor to preprocess the engine speed signal,the instability of the narrowband ANC system caused by speed fluctuation is eliminated.A variable momentum factor is constructed using the error signal,and the convergence process of the narrowband ANC system is accelerated by additional terms.The simulation results prove that the VM-SFXLMS algorithm can improve the convergence speed and tracking performance of the narrowband ANC system while ensuring the steady-state performance.Secondly,the SM-VMNFXLMS algorithm and VM-SFXLMS algorithm are applied to the broadband-narrowband hybrid active noise control system to improve the noise reduction effect of the ANC system on the interior noise.In order to make the broadband-narrowband hybrid active noise control system have the noise reduction capability in the impulse disturbance environment such as the car crossing the speed bump or driving on the rough road,an impulse detector is set in the ANC system to detect the impulse noise in real time in order to shut down the narrowband related subsystem in time,which enhances the robustness and adaptability of the broadband-narrowband hybrid ANC system.Thirdly,in order to improve the ANC system operation efficiency,we propose a parameter matching method based on the Improved Whale Optimization Algorithm(IWOA)algorithm,using the evaluation index of the ANC algorithm as the fitness function,which can efficiently and precisely match the parameters with better noise reduction performance compared with the trial-and-error method.Lastly,we conducted noise data acquisition tests under the car driving conditions at constant speed and over speed bumps,and identified the sound paths inside the car.Based on MATLAB software,we conducted noise reduction simulations for in-cab noise,and the results demonstrated that the improved broadband-narrowband hybrid ANC algorithm can cancel both the low-frequency broadband noise and the target order noise in the vehicle,and can achieve an average noise reduction of 10-12 d B.In this paper,a d SPACE-based hardware-in-the-loop testbed is built to conduct real-time experiments to verify the practicality of the improved algorithm.The experimental results show that the improved broadband-narrowband hybrid ANC algorithm can achieve 14-16 d B of noise reduction at the second-order frequency of the engine,and achieve an average noise reduction of about 5 d B.