基于多传感器信息融合方法的刀具破损识别
|
摘要
针对铣削过程监控中多目标状态源存在同频干扰的问题,基于经验模态分解和独立分量分析提出了一种多通道信号盲源分离算法,以声音传感器及振动传感器为信号检测元件,利用多传感器信息融合技术对铣削加工过程中刀具破损监测相关技术问题进行了详细分析。通过设计多齿铣削试验,将所采集的声音信号与振动加速度信号进行了对比分析,并对声振信号进行EMD-ICA分析。研究表明:①切削声音信号和Y轴方向上的振动加速度信号处在同一个频段;②多传感器信息融合监测方式能消除监测信号中存在的背景噪声及目标状态相互干扰的问题,提取出混合信号中与刀具破损状态相关的故障特征频率成分,为刀具破损识别提供依据。
An information coordination from multi-sensors approach based on empirical mode decomposition(EMD) and independent component analysis(ICA) was presented to deal with the blind source separation(BSS) problem of cutting sound signals and vibration signals in the process of face milling.EMD method was used to extract all intrinsic mode functions(IMF) in the sound and vibration signals which had been acquired from face milling processes,then deal with those IMFs using FastICA,and can obtain a lot of independent components.Analysis result shows that ① the main frequency of the sound signal and the(Y) axis direction component of the vibration acceleration signal are at the same frequency band ② this method can extract the characteristic frequency components related to tool breakage from mixed signals.
An information coordination from multi-sensors approach based on empirical mode decomposition(EMD) and independent component analysis(ICA) was presented to deal with the blind source separation(BSS) problem of cutting sound signals and vibration signals in the process of face milling.EMD method was used to extract all intrinsic mode functions(IMF) in the sound and vibration signals which had been acquired from face milling processes,then deal with those IMFs using FastICA,and can obtain a lot of independent components.Analysis result shows that ① the main frequency of the sound signal and the(Y) axis direction component of the vibration acceleration signal are at the same frequency band ② this method can extract the characteristic frequency components related to tool breakage from mixed signals.
引文
[1]Mannan MA,Kassim AA,Jing M(2000)Application of im-age and sound analysis techniques to monitor the condition ofcutting tools.Pattern Recogn Lett 21(11):969-979.
[2]C.H.Jun,S.H.Suh,Statistical tool breakage detectionschemes based on vibration signals in NC milling,Interna-tional Journal of Machine Tools&Manufacture 39(1999)1733-1746.
[3]I.Marinescu,D.A.Axinte,A critical analysis of effective-ness of acoustic emission signals to detect tool and workpiecemalfunctions in milling operations,International Journal ofMachine Tools&Manufacture,48(2008)1148-1160.
[4]E.Kuljanic,M.Sortino,TWEN,a method based on cuttingforces—monitoring tool wear in face milling,InternationalJournal of Machine Tools&Manufacture,45(2005):29-34.
[5]Shao,H.,Wang,H.L.,Zhao,X.M.A cutting power mod-el for tool wear mentoring in Milling[J].International Journalof Machine Tools&Manufacture.2004,44:1503-1509.
[6]胡秋.CIMS环境下刀具状态监测研究回顾与展望[J].机床与液压,2003(6):17-18.
[7]史恒,李桂林,王伟,等.基于总体经验模式分解的地震信号随机噪声消除[J].地球物理学进展,2011,26(1):71-78.
[8]HUANG N E,SHEN Z,S LONG R,et al.The empiricalmode decomposition and Hilbert spectrum for nonlinear andnon-stationary time series analysis[J].Proceedings of theRoyal Society of London,1998,454:903-995.
[9]Cardoso,J.F.and A.Souloumiac,Blind beamforming fornon-Gaussian signals.IEE Proceedings,Part F:Radar andSignal Processing,1993,140(6):362-370.
[10]Bell,A.J.and T.J.Sejnowski,An information-maximiza-tion approach to blind separation and blind deconvolution.Neural computation,1995,7(6):1129-1159.
[11]Hyvrinen,A.and E.Oja,Independent component analy-sis:Algorithms and applications.Neural Networks,2000,13(4-5):411-430.
[2]C.H.Jun,S.H.Suh,Statistical tool breakage detectionschemes based on vibration signals in NC milling,Interna-tional Journal of Machine Tools&Manufacture 39(1999)1733-1746.
[3]I.Marinescu,D.A.Axinte,A critical analysis of effective-ness of acoustic emission signals to detect tool and workpiecemalfunctions in milling operations,International Journal ofMachine Tools&Manufacture,48(2008)1148-1160.
[4]E.Kuljanic,M.Sortino,TWEN,a method based on cuttingforces—monitoring tool wear in face milling,InternationalJournal of Machine Tools&Manufacture,45(2005):29-34.
[5]Shao,H.,Wang,H.L.,Zhao,X.M.A cutting power mod-el for tool wear mentoring in Milling[J].International Journalof Machine Tools&Manufacture.2004,44:1503-1509.
[6]胡秋.CIMS环境下刀具状态监测研究回顾与展望[J].机床与液压,2003(6):17-18.
[7]史恒,李桂林,王伟,等.基于总体经验模式分解的地震信号随机噪声消除[J].地球物理学进展,2011,26(1):71-78.
[8]HUANG N E,SHEN Z,S LONG R,et al.The empiricalmode decomposition and Hilbert spectrum for nonlinear andnon-stationary time series analysis[J].Proceedings of theRoyal Society of London,1998,454:903-995.
[9]Cardoso,J.F.and A.Souloumiac,Blind beamforming fornon-Gaussian signals.IEE Proceedings,Part F:Radar andSignal Processing,1993,140(6):362-370.
[10]Bell,A.J.and T.J.Sejnowski,An information-maximiza-tion approach to blind separation and blind deconvolution.Neural computation,1995,7(6):1129-1159.
[11]Hyvrinen,A.and E.Oja,Independent component analy-sis:Algorithms and applications.Neural Networks,2000,13(4-5):411-430.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心