| With the rapid development of my country’s infrastructure level,the construction machinery used in infrastructure is constantly updated and iterated.The requirements for equipment have been improved step by step from satisfying work practicability to improving operating comfort,and noise is an important factor affecting operating comfort.This paper studied the active noise reduction technology for the noise of the excavator cab,mainly improved the online secondary path modeling method,and designed a dual-channel active noise control system suitable for the excavator cab.The main contents of this paper are as follows:(1)The theoretical derivation of the adaptive filter is carried out in the noise reduction structure.The noise reduction algorithm is derived from the most basic least mean square(LMS)algorithm,and then the filter x least mean square(Fx LMS)algorithm is proposed,which considers The effect of secondary paths.In addition,the step size factor in the update formula in the noise reduction algorithm is analyzed.The order of the filter and the step factor of variable step size are simulated by MATLAB,and finally the filter order L is 16,the parameter α in the variable step factor is 7,and the parameter β is 0.02.(2)The classical Eriksson online secondary path modeling method is analyzed,and its theoretical derivation is carried out.It is found that this structure has some defects,mainly because the white noise signal used for online secondary path modeling can interfere with the original noise reduction system.Since the online secondary path modeling requires the participation of white noise signals,this interference cannot be eliminated.Aiming at this defect,a new online secondary path modeling method is proposed based on Eriksson’s online secondary path modeling method.An amplitude adjustment module is added after the white noise signal.The gain coefficient of the amplitude adjustment is mainly determined according to the first-order forward difference of the square value of the error signal at intervals of 50 times.According to the change trend of the error signal,the amplitude of the white noise is adjusted Adjustment,the amplitude of white noise can be reduced to a small value after the system converges,and its interference to the control system can be reduced as much as possible.In addition,the method of variable step size is adopted for the update algorithm of the control filter and the identification filter,which can speed up the convergence of the algorithm.Through MATLAB simulation,Eriksson’s online secondary path modeling method and the improved method are compared.The simulation results show that the overall noise reduction of Eriksson’s method is about 8d B,and the noise reduction can reach about 10 d B at the frequency of 371 Hz where the sound pressure is the largest;the overall noise reduction of the improved method is about 11 d B,and can reach the frequency of 371 Hz.The amount of noise reduction is about 14 d B.And when the algorithm iterates to 10,000 times,the algorithm can also re-converge by interfering with the secondary path.(3)The multi-channel active noise control algorithm is studied,the theoretical formula of the general multi-channel algorithm is deduced,and the dual-channel Fx LMS algorithm is analyzed.Combined with the specific noise reduction component arrangement in the excavator cab,a dual-channel online secondary path modeling structure suitable for the excavator cab is designed,which simplifies the influence of different secondary sound sources on different error microphones.Through MATLAB simulation,the comparison The noise reduction effect of the single-channel online secondary path modeling algorithm and the dualchannel online secondary path modeling algorithm is presented.The simulation results show that the noise reduction of the dual-channel algorithm at a single microphone is close to that of the single-channel algorithm,but the dual-channel algorithm can achieve noise reduction at two error microphones,while the single-channel structure cannot.(4)The hardware equipment and software environment of the excavator cab active noise control experiment are introduced,including DSP chips,audio chips and data transmission protocols.The noise reduction components are placed in the cab of the excavator,and the improved online quadratic path modeling algorithm is tested.The experimental results show that the Eriksson method can achieve a noise reduction of about 2.9d B,and can achieve the maximum sound pressure at the frequency of 371 Hz.Reduced by about 8d B;the improved method can achieve a noise reduction of about 4.17 d B,which can be reduced by about 9d B at a frequency of 371 Hz.Finally,a dual-channel online secondary path modeling experiment is done,and compared with the single-channel,the experimental results show that the noise reduction of the dual-channel structure at the error microphone 1 and error microphone 2 is 3.9d B and 3.8d B,respectively;The noise reduction of the channel structure at error microphone 1 and error microphone 2 are4 d B and 1.1d B,respectively.The dual-channel structure has a wide range of noise reduction.Because of the interference between multiple secondary sound sources,the single-point noise reduction effect will be lower than that of the single-channel,but compared with the reduced noise reduction amount,the overall noise reduction effect brought by the increase of the noise reduction range is better. |
