Research On Noise Control Technology In Aircraft Cabin Based On Multi-Channel ANC Algorithm

As the utilization rate of airplanes increases in modern society,the noise inside aircraft cabins has gradually gained attention.Due to limitations of passive noise control(PNC)technology,active noise control(ANC)technology began to be used for controlling noise inside aircraft cabins.In addition,since a large cabin space requires the construction of a multi-channel ANC system for ideal noise reduction,the computational complexity of active noise control algorithms increases rapidly with the number of channels.Therefore,this thesis studies how to improve the noise reduction effect of a multi-channel ANC system in aircraft cabins without increasing too much computational complexity.Firstly,addressing the fact that there are many narrowband components in the noise inside aircraft cabins,this paper introduces the notch filter FXLMS algorithm based on the multi-channel FXLMS algorithm to improve the noise reduction effect of the algorithm on narrowband noise.Subsequently,a variable step-size adaptive filter with momentum term is proposed to improve the convergence speed and reduce the steadystate error of the algorithm.To address the issue of increased computational complexity,delay of the reference signal is used instead of filter-x signal calculation,while the notch filter FXLMS algorithm is replaced by the notch delay LMS algorithm.In addition,the FXLMS algorithm is replaced by the filter-e signal,and the step size change error data is used as the basis for changing the iterative control number,proposing a local iterative adaptive filter weight coefficient algorithm where fewer adaptive filter weight coefficients are updated when the error is small.Secondly,noise signals are constructed for different parts of the algorithms and simulations are conducted using MATLAB to preliminarily verify the effects of the above-mentioned optimization methods.The optimized algorithm and multi-channel FXLMS algorithm are used in a noise reduction experiment on turboprop aircraft cabin noise in MATLAB,and the results show that the optimized algorithm has an average noise reduction improvement of about 5d B,while the computational complexity of the filter weight coefficient update is 70.74% of the previous amount.Finally,an active noise control hardware experimental platform was built and tested using turboprop aircraft cabin noise.The experiment compared the noise reduction effects of the optimized multi-channel ANC algorithm and the multi-channel FXLMS algorithm.The results show that the optimized algorithm has an initial convergence speed improvement of about 300% and a final noise reduction improvement of about 4d B,while the computational complexity of the filter weight coefficient update is 72.72% of the previous amount.