Aiming at the problem of difficulty in detecting the faults of sound source during driving,the received signal was affected by Doppler effect,which caused the problem of frequency shift,frequency band spread and amplitude modulation,and in addition,signal was in a low signal-noise ratio(SNR)which caused by strong environmental noise,which seriously affectd the extraction of the target sound signal,a fault detection method for train bearings based on Doppler signal correction and short-step long-wave beamforming was proposed.Firstly,based on the relative motion relationship between sound source and microphone and Morse acoustic theory,the interpolation time series and amplitude reduction formulas were obtained to recover the distorted signal modulated by Doppler effect.Then,a short-step beamforming method based on microphone array was proposed,which intercepted the signal in segments and enhanced each segment in its expected direction.Finally,the effectiveness of this method under different signal-to-noise ratio was studied by simulation.The rail edge fault detection experiment was designed,and the rolling bearing with inner ring crack defect was analyzed experimentally.Combining with simulation and experiment,the effectiveness of this method was verified.The research result shows that comparing with the original low SNR signal collected by 8 microphones,the signal is corrected in both time domain and frequency domain,and the SNR is increased by 11.5 dB.The proposed method can recover the Doppler distortion signal and effectively extract the fault characteristics of mobile sound sources under strong interference environment.
关键词
移动声源/多普勒效应/滚动轴承/故障诊断/多普勒信号校正/短步长波束形成方法/信噪比
Key words
mobile sound source/Doppler effect/rolling bearing/fault diagnosis/Doppler signal correction/short-step long wave beamforming method/signal-noise ratio(SNR)
维普
万方数据