Research On Active Noise Control Based On Sound Quality

The interior noise is an important factor that affect the ride comfort of a vehcile,with the improvement of people’s living quality,aiming at sound pressure level of noise control can not meet the requirements of comfort.Sound quality reflect people’s subjective feeling of noise,the research and improving the sound quality of interior noise is of positive significance for improving the ride comfort of the vehcile.In this paper,the steady-state noise in the driver’s right ear under the uniform speed condition of a domestic SUV was taken as the research object.Based on the established sound quality prediction model of interior noise,the active control method is used to selectively offset the noise and explore the active control scheme of interior noise quality.The main research contents are as follows:（1）The subjective evaluation of noise samples was carried out by annoyance level.Using GA to optimize the initial weights of BP neural network,a high-precision GA-BP sound quality prediction model was established with the objective psychoacoustic parameters as input and the subjective annoyance level as output.Through weight analysis,it was found that the loudness and roughness were the main factors affecting the sound quality,which provided the objective evaluation basis of acoustic quality for active noise control and determined the objective psychoacoustic parameters to be controlled.（2）After theoretical analysis,the Filter-Error Least Mean Square（FELMS）algorithm for active control of sound quality was selected.The construction method of the reference signal was studied,and the off-line modeling of the secondary path was carried out,and the adaptive noise active control Simulink simulation model based on FELMS algorithm was built.Through the simulation analysis,considering the convergence,stability and control effect of sound quality,the two key parameters of the length and convergence factor of the adaptive filter were determined.It lays a theoretical foundation for the subsequent active control simulation of sound quality.（3）Considering the influence of the controlled frequency band on the filtering performance,a method for finding the optimal passband of the residual filter based on the contribution of the critical band to the controlled parameter was proposed and The relationship between the controlled parameters and the control frequency band was studied.First,based on the loudness which has the greatest influence on the sound quality,the controlled frequency band was sequentially increased according to the magnitude of the special loudness.The simulation results of the active control of the sample show that when the control of loudness is optimal,the level of objective annoyance level was significantly reduced;Comprehensive consideration of loudness and roughness,according to the weight proportion of GA-BP sound quality prediction model,the weighted parameter W_L-R,which has a greater correlation with level of annoyance,was obtained after weighted processing.The calculation formula of special weighted parameter in the critical band was derived,and the controlled frequency band was sequentially increased according to the special weighed parameter.The results show that when the weighted parameter reaches the optimal value,the decrease of objective annoyance level was better,which proves the correctness of active control of sound quality based on weighted parameter W_L-R.（4）Under the four working conditions of the car running at a constant speed,the acoustic quality active control test was carried out on the driver’s right ear area.The results show that the loudness and roughness of the noise achieved good control effect.The two methods of subjective and objective evaluations test result show that the annoyance level was reduced by 1.12².57,and proves the feasibility and validity of the research in this paper to improve the interior noise quality,which can provide reference for the active control of interior noise quality.