变转速下基于广义解调算法的滚动轴承故障诊断
|
摘要
变转速条件下故障轴承的冲击间隔会相应的发生改变,导致以包络分析为代表以恒转速为前提的故障诊断方法失效。阶比分析因其在消除频谱模糊方面的有效性,成为处理变转速故障轴承信号最为常规的方法。然而,上述方法在对信号重采样的过程中存在幅值误差、包络畸变以及计算效率低等问题。为此,从滚动轴承的振动特性出发,提出了无需角域重采样的基于广义解调算法的滚动轴承故障诊断方法。整个算法主要包括五部分:(1)利用快速谱峭度算法确定最优带通滤波参数,并对原始振动信号进行滤波;(2)根据转速脉冲信号计算并拟合转速曲线;(3)通过转频方程以及滚动轴承的故障特征系数确定广义解调算法所需要的相位函数;(4)根据相位函数对滤波信号进行广义解调,对解调信号进行快速傅里叶变换(Fast Fourier Transform,FFT)获取解调信号的频谱图;(5)观察频谱图中的峰值,更改故障特征系数重复步骤(3)-(4),最终确定轴承故障类型。仿真及实测的故障轴承信号分析证明了新算法对变转速下滚动轴承故障诊断的有效性。
The intervals of the faulty rolling element bearing impulses will not appear periodically under variable rotational speeds,which lead to the envelope analysis method,which is based on the assumption of constant speed,no longer applicable.The order tracking is the most familiar algorithm to process time-varying speed bearing signals because of its effectiveness in eliminating the fuzzy spectrum.However,it also has defects such as amplitude error and envelope deformation.Hence,based on the characteristics of rolling element bearing,a method of rolling element bearing fault diagnosis is proposed by the generalized demodulation method without angular domain resampling.The method consists of six main steps:( a) signal filtering via optimal bandpass filter parameters obtained by fast kurtosis algorithm;( b) to obtain the speed curve based on the rotational speed pulse signal;( c) to calculate the phase-function-based speed curve and fault characteristic coefficient;( d) to apply the generalized demodulation algorithm to the filter signal and obtain the frequency spectrum via fast Fourier transform( FFT);( e) to search the peak from the spectrum and then modify the fault characteristic coefficient and repeat steps( c)-( d) to identify the fault type of bearing.The studies of simulated and experimental faulty bearing signals indicate that the proposed method is effective to diagnose faulty bearing under time-varying rotational speed.
The intervals of the faulty rolling element bearing impulses will not appear periodically under variable rotational speeds,which lead to the envelope analysis method,which is based on the assumption of constant speed,no longer applicable.The order tracking is the most familiar algorithm to process time-varying speed bearing signals because of its effectiveness in eliminating the fuzzy spectrum.However,it also has defects such as amplitude error and envelope deformation.Hence,based on the characteristics of rolling element bearing,a method of rolling element bearing fault diagnosis is proposed by the generalized demodulation method without angular domain resampling.The method consists of six main steps:( a) signal filtering via optimal bandpass filter parameters obtained by fast kurtosis algorithm;( b) to obtain the speed curve based on the rotational speed pulse signal;( c) to calculate the phase-function-based speed curve and fault characteristic coefficient;( d) to apply the generalized demodulation algorithm to the filter signal and obtain the frequency spectrum via fast Fourier transform( FFT);( e) to search the peak from the spectrum and then modify the fault characteristic coefficient and repeat steps( c)-( d) to identify the fault type of bearing.The studies of simulated and experimental faulty bearing signals indicate that the proposed method is effective to diagnose faulty bearing under time-varying rotational speed.
引文
[1]徐亚军,于德介,孙云嵩,等.滚动轴承故障诊断的阶比多尺度形态学解调方法[J].振动工程学报,2013,26(2):252—259.Xu Ya-jun,Yu De-jie,Sun Yun-song,et al.Roller bearing fault diagnosis using order multi-scale morphology demodulation[J].Journal of Vibration Engineering,2013,26(2):252—259.
[2]DING Xiaoxi.HE Qingbo,LUO Nianwu.A fusion feature and its improvement based on locality preserving projections for rolling element bearing fault classification[J].Journal of Sound and Vibration,2015,335:367—383.
[3]ZIMROZ R,BARTELMUS W,BARSZCZ T,et al.Diagnostics of bearings in presence of strong operating conditions non-stationarity-a procedure of load-dependent features processing with application to wind turbines bearings[J].Mechanical Systems and Signal Processing,2014,46:16—27.
[4]BAI M,HUANG J,HONG M,et al.Fault diagnosis of rotating machinery using an intelligent order tracking system[J].Journal of Sound and Vibration,2005,280(3-5):699—718.
[5]WANG K S,HEYNS P S.The combined use of order tracking techniques for enhanced Fourier analysis of order components[J].Mechanical Systems and Signal Processing,2011,25(3):803—811.
[6]GUO Yu,LIU Tingwei,NA Jing,et al.Envelope order tracking for fault detection in rolling element bearings[J].Journal of Sound and Vibration,2012,331:5644—5654.
[7]SAAVEDRA P N,RODRIGUEZ C G.Accurate assessment of computed order tracking[J].Shock and Vibration,2006,13(1):13—21.
[8]程卫东,王天杨,Wang Jin-jiang,等.包络变形对阶比分析结果的影响及消除方法[J].振动工程学报,2015,28(3):470—477.Cheng Wei-dong,Wang Tian-yang,Wang Jin-jiang,et al.Effects and its elimination method of envelop deformation on order analysis[J].Journal of Vibration Engineering,2015,28(3):470—477.
[9]OLHEDE S,WALDEN A T.A generalized demodulation approach to time-frequency projections for multicomponent signals[J].Proceedings of the Royal Society A,2005,461(2059):2159—2179.
[10]CHENG Junsheng,YANG Yu,YU Dejie.Application of the improved generalized demodulation time–frequency analysis method to multi-component signaldecomposition[J].Signal Processing,2009,89:1205—1215.
[11]皮维,于德介,彭富强.基于多尺度线调频基稀疏信号分解的广义解调方法及其在齿轮故障诊断中的应用[J].机械工程学报,2010,46(15):59—64.PI Wei,YU De-jie,PENG Fu-qiang.Generalized demodulation method based on multi-scale chirplet and sparse signal decomposition and its application to gear fault diagnosis[J].Journal of Mechanical Engineering,2010,46(15):59—64.
[12]FENG Zhipeng,CHU Fulei,ZUO M J.Time-frequency analysis of time-varying modulated signals based on improved energy separation by iterative generalized demodulation[J].Journal of Sound and Vibration,2011,330:1225—1243.
[13]RANDALL R B,ANTONI J.Rolling element bearing diagnostics-A tutorial[J].Mechanical Systems and Signal Processing,2011,25(2):485—520.
[14]ANTONI J.The spectral kurtosis:A useful tool for characterizing non-stationary signals[J].Mechanical Systems and Signal Processing,2006,20(2):282—307.
[15]ANTONI J.Fast computation of the kurtogram for the detection of transient faults[J].Mechanical Systems and Signal Processing,2007,21:108—124.
[16]王天杨,李建勇,程卫东.基于低次故障特征阶比系数的变转速滚动轴承等效转频估计算法[J].机械工程学报,2015,51(3):121—128.Wang Tian-yang,Li Jian-yong,Cheng Wei-dong.Equivalent rotational frequency estimation algorithm of faulty rolling bearing under varying rotational speed based on the lower fault characteristic order coefficient[J].Journal of Mechanical Engineering,2015,51(3):121—128.
[2]DING Xiaoxi.HE Qingbo,LUO Nianwu.A fusion feature and its improvement based on locality preserving projections for rolling element bearing fault classification[J].Journal of Sound and Vibration,2015,335:367—383.
[3]ZIMROZ R,BARTELMUS W,BARSZCZ T,et al.Diagnostics of bearings in presence of strong operating conditions non-stationarity-a procedure of load-dependent features processing with application to wind turbines bearings[J].Mechanical Systems and Signal Processing,2014,46:16—27.
[4]BAI M,HUANG J,HONG M,et al.Fault diagnosis of rotating machinery using an intelligent order tracking system[J].Journal of Sound and Vibration,2005,280(3-5):699—718.
[5]WANG K S,HEYNS P S.The combined use of order tracking techniques for enhanced Fourier analysis of order components[J].Mechanical Systems and Signal Processing,2011,25(3):803—811.
[6]GUO Yu,LIU Tingwei,NA Jing,et al.Envelope order tracking for fault detection in rolling element bearings[J].Journal of Sound and Vibration,2012,331:5644—5654.
[7]SAAVEDRA P N,RODRIGUEZ C G.Accurate assessment of computed order tracking[J].Shock and Vibration,2006,13(1):13—21.
[8]程卫东,王天杨,Wang Jin-jiang,等.包络变形对阶比分析结果的影响及消除方法[J].振动工程学报,2015,28(3):470—477.Cheng Wei-dong,Wang Tian-yang,Wang Jin-jiang,et al.Effects and its elimination method of envelop deformation on order analysis[J].Journal of Vibration Engineering,2015,28(3):470—477.
[9]OLHEDE S,WALDEN A T.A generalized demodulation approach to time-frequency projections for multicomponent signals[J].Proceedings of the Royal Society A,2005,461(2059):2159—2179.
[10]CHENG Junsheng,YANG Yu,YU Dejie.Application of the improved generalized demodulation time–frequency analysis method to multi-component signaldecomposition[J].Signal Processing,2009,89:1205—1215.
[11]皮维,于德介,彭富强.基于多尺度线调频基稀疏信号分解的广义解调方法及其在齿轮故障诊断中的应用[J].机械工程学报,2010,46(15):59—64.PI Wei,YU De-jie,PENG Fu-qiang.Generalized demodulation method based on multi-scale chirplet and sparse signal decomposition and its application to gear fault diagnosis[J].Journal of Mechanical Engineering,2010,46(15):59—64.
[12]FENG Zhipeng,CHU Fulei,ZUO M J.Time-frequency analysis of time-varying modulated signals based on improved energy separation by iterative generalized demodulation[J].Journal of Sound and Vibration,2011,330:1225—1243.
[13]RANDALL R B,ANTONI J.Rolling element bearing diagnostics-A tutorial[J].Mechanical Systems and Signal Processing,2011,25(2):485—520.
[14]ANTONI J.The spectral kurtosis:A useful tool for characterizing non-stationary signals[J].Mechanical Systems and Signal Processing,2006,20(2):282—307.
[15]ANTONI J.Fast computation of the kurtogram for the detection of transient faults[J].Mechanical Systems and Signal Processing,2007,21:108—124.
[16]王天杨,李建勇,程卫东.基于低次故障特征阶比系数的变转速滚动轴承等效转频估计算法[J].机械工程学报,2015,51(3):121—128.Wang Tian-yang,Li Jian-yong,Cheng Wei-dong.Equivalent rotational frequency estimation algorithm of faulty rolling bearing under varying rotational speed based on the lower fault characteristic order coefficient[J].Journal of Mechanical Engineering,2015,51(3):121—128.