基于人耳听觉特性瞬态信号的提取方法研究
|
摘要
针对瞬态信号提取问题,利用人耳听觉系统识别机制,提出一种在单通道信号条件下自动提取信号源中瞬态信号的方法。该方法首先对信号进行带通滤波,并通过逆序滤波实现相位调整,继而通过包络分析获得各滤波信号极大和极小值及其各对应的时间点,最终利用极值的幅值和时间,得到同步性和瞬态性线索。通过这2类信息,可以在时频域中筛选出符合瞬态成分特点的滤波信号所对应的时频段,最终完成瞬态成分的波形生成。该方法仅需1路信号源就可准确地提取出故障产生的瞬态信号,从而判断出设备的故障源,且该方法比盲源分离等方法能快速实现在线检测。
In view of the extraction of transient signals, by using the identification mechanism of the human hearing system, the paper proposed a method of automatically extracting the transient signals from the signal source in the case of single-channel signal. In the method, the band pass filtration of the signal was conducted first, and the phase regulation was realized via inverse sequence filtration. And then, the maximum value and the minimum value of the filtered signal as well as their corresponding time points were obtained after envelope analysis. Finally, the clues of synchronization and transient were obtained by using the amplitude and time of the extreme value. Through the two kinds of clue information, the time frequency zone corresponding to the filtered signal that conformed to the characteristics of the transient component was identified from the time frequency domain, thus the waveform of the transient component generated. For the method, one channel of signal was only needed to accurately extract the transient signal due to fault so as to judge the fault source.Compared with the blind source separation method, the method rapidly realized the online detection.
In view of the extraction of transient signals, by using the identification mechanism of the human hearing system, the paper proposed a method of automatically extracting the transient signals from the signal source in the case of single-channel signal. In the method, the band pass filtration of the signal was conducted first, and the phase regulation was realized via inverse sequence filtration. And then, the maximum value and the minimum value of the filtered signal as well as their corresponding time points were obtained after envelope analysis. Finally, the clues of synchronization and transient were obtained by using the amplitude and time of the extreme value. Through the two kinds of clue information, the time frequency zone corresponding to the filtered signal that conformed to the characteristics of the transient component was identified from the time frequency domain, thus the waveform of the transient component generated. For the method, one channel of signal was only needed to accurately extract the transient signal due to fault so as to judge the fault source.Compared with the blind source separation method, the method rapidly realized the online detection.
引文
[1]PELED R,BRAUN S,ZACKSENHOUSE M.A blind deconvolution separation of multiple sources,with application to bearing diagnostics[J].Mechanical Systems and Signal Processing,2005,19(6):1181-1195.
[2]季忠,黄捷,秦树人.提升小波在齿轮箱故障诊断中的应用[J].振动、测试与诊断,2010,30(3):291-294.
[3]王磊,彭圆,林正青,等.听觉外周计算模型在水中目标分类识别中的应用[J].电子学报,2012,5(1):199-203.
[4]李允公,张金萍,戴丽,等.基于听觉神经同步振荡网络的时频结构描述方法研究[J].机械工程学报,2012,48(5):55-62.
[5]AURELIE B C,CONSTANZE M,ROBERT T K.Load effects in auditory selective attention:evidence for distinct facilitation and inhibition mechanisms[J].NeuroImage,2010,50(1):277-284.
[6]罗跃嘉,姜扬,程康.认知神经科学教程[M].北京:北京大学出版社,2006:66-80.
[7]李允公,刘杰,张金萍.基于实测冲击响应的转子碰摩故障特征提取方法[J].机械工程学报,2007,43(4):224-228.
[8]WEINTRAUB M.A theory and computational model of auditory monaural sound separation[D].State of California:Stanford University,1985:33-36.
[9]CARLYON R P.How the brain separates sounds[J].Trends in Cognitive Sciences,2004,8(1):465-470.
[10]TAO L.Real-valued discret gabor transforms for discret signal and image representation chinese[J].Journal of Electronics,2001,10(4):444-449.
[2]季忠,黄捷,秦树人.提升小波在齿轮箱故障诊断中的应用[J].振动、测试与诊断,2010,30(3):291-294.
[3]王磊,彭圆,林正青,等.听觉外周计算模型在水中目标分类识别中的应用[J].电子学报,2012,5(1):199-203.
[4]李允公,张金萍,戴丽,等.基于听觉神经同步振荡网络的时频结构描述方法研究[J].机械工程学报,2012,48(5):55-62.
[5]AURELIE B C,CONSTANZE M,ROBERT T K.Load effects in auditory selective attention:evidence for distinct facilitation and inhibition mechanisms[J].NeuroImage,2010,50(1):277-284.
[6]罗跃嘉,姜扬,程康.认知神经科学教程[M].北京:北京大学出版社,2006:66-80.
[7]李允公,刘杰,张金萍.基于实测冲击响应的转子碰摩故障特征提取方法[J].机械工程学报,2007,43(4):224-228.
[8]WEINTRAUB M.A theory and computational model of auditory monaural sound separation[D].State of California:Stanford University,1985:33-36.
[9]CARLYON R P.How the brain separates sounds[J].Trends in Cognitive Sciences,2004,8(1):465-470.
[10]TAO L.Real-valued discret gabor transforms for discret signal and image representation chinese[J].Journal of Electronics,2001,10(4):444-449.