Research on the Effect of Parameter Selection on Engine Noise Control in FXLMS Algorithm

Research on the Effect of Parameter Selection on Engine Noise Control in FXLMS Algorithm

In order to reduce the induction noise of automobile engine, an active control system model of engine induction noise is established with the FXLMS algorithm, the off-line identification method is used to identify the secondary channel, and the engine speed signal is used to construct the sound source reference signal to avoid the interference from secondary sources to the reference signal. The influence of filter length and convergence factor on the convergence speed and stability of the system is studied by simulation, the average power of the noise signal and the error signal is calculated, the noise variation at different engine speeds is analyzed, and the noise reduction effect and system stability are evaluated. The results show that when the filter order L is 20 and the convergence factor is 0.01, the engine induction noise can be effectively reduced, the noise reduction can reach 15dB, and the system stability can be improved at the same time. This paper optimizes the convergence factor and the filter order in the FXLMS algorithm, and can effectively control the induction noise in the car and improve the indoor sound quality of the car. It has important reference significance for practical engineering applications.