Research On Adaptive Active Noise Control Technology In Passenger Car Based On Notch Delay LMS Algorithm

When the traditional fuel vehicle is in idle speed or low speed driving condition,the noise of the high sound level in the cabin mainly comes from the low frequency order noise generated by the engine,and the traditional passive noise reduction methods such as sound absorption and sound insulation have a poor effect on low frequency noise,which results in the vehicle Active Noise Control(ANC)technology.However,the efficiency of acquiring the number of the secondary path delay points in the current ANC technology is too low,which limits the application of the delay idea in the secondary path.At the same time,the traditional notch LMS algorithm requires a large amount of computation and costly controllers,which limits the popularity of the vehicle ANC technology.Therefore,it is important to improve the efficiency of acquiring the secondary path delay points and reduce the computational complexity of the traditional notch LMS algorithm.In this paper,a new time-delay-point estimation method combining additional-randomnoise method and the time-delay-estimation method is proposed for identifying the number of delay points of the secondary path,which is intended to improve the efficiency of acquiring the number of delay points corresponding to the target frequency band in the secondary path.In addition,a notch delay LMS algorithm is proposed based on the traditional notch LMS method.It takes the delay operation of the reference signal to replace the convolution operation between the unit impulse response function in the secondary path and the reference signal in the traditional notch LMS algorithm,which is intended to greatly reduce the calculation amount of the algorithm.The proposed method acquiring the number of the delay points of the secondary path is verified by building the algorithm simulation model with MATLAB/Simulink software.The results demonstrate that the estimated number of the secondary path delay points is almost errorfree compared with true values;Besides,a comparison between the notch delay LMS algorithm and the traditional notch LMS algorithm are conducted by simulating analysis,and it is found that the former ensuring the noise reduction effect is far faster than the latter.In the next step,the bench test of ANC hardware platform system with dSPACE controller is built,and then the ANC system with DSP controller is designed on the vehicle.Based on the experimental conditions of the platform and the vehicle,the notch delay LMS algorithm and the traditional notch LMS algorithm are further compared and analyzed.The result showed that the traditional notch LMS algorithm with sampling frequency of 8000 Hz cannot be implemented on the existing processor hardware resources,due to the excessive computational cost.But the current hardware resources are surplus when running the notch delay LMS algorithm.Moreover,the notch delay LMS algorithm obtains a total noise reduction of more than 7.4dBA in the range of 20~200Hz under the idle condition of the passenger car.In addition,it also achieves significant attenuation of the acceleration roaring sound under the acceleration condition of 1300 rpm to 3000 rpm.Which leading a linear growth trend of the sound pressure level of the engine acceleration noise,meeting the requirements of the engine acceleration sound quality.Therefore,the proposed notch delay LMS algorithm in this paper can greatly reduce the computational resources and save the controller cost while ensuring the noise reduction of the ANC system.