基于包络解调滤波的滚动轴承复合故障诊断  

Rolling bearing multi-fault diagnosis based on envelope demodulation filter algorithm

作　　者：刘东东[1] 程卫东[1] 温伟刚[1] 万广通[1] LIU Dongdong;CHENG Weidong;WEN Weigang;WAN Guangtong(School of Mechanical Electronic and Control Engineering,Beijing Jiaotong University,Beijing 100044,China)

机构地区：[1]北京交通大学机械与电子控制工程学院,北京100044

出　　处：《中南大学学报（自然科学版）》2018年第4期881-887,共7页Journal of Central South University:Science and Technology

基　　金：国家自然科学基金资助项目(51275030);中央高校基本科研业务费专项资金资助项目(2016JBM051)~~

摘　　要：针对在变转速工况下进行滚动轴承复合故障诊断时,由于故障间的相互作用使得多种故障特征混叠在一起,彼此干扰,造成误判﹑漏判等问题,以及时域滤波过程中共振频带参数难以获得这一问题,提出基于包络解调滤波的滚动轴承复合故障诊断方法。首先,使用多项式函数拟合轴承转速,并根据轴承各部分的故障特征频率系数和转速信息计算瞬时故障特征频率(IFCF)趋势线;其次,根据各部分IFCF趋势线的拟合函数,构造解调算法的相位函数,对原始信号经过Hilbert变换得到的包络信号解调;最后,以各部分IFCF拟合函数的初始值为中心频率构造窄带巴特沃斯带通滤波器,对解调信号进行滤波处理,并对滤波后的信号进行快速傅里叶变换(FFT)得到包络解调滤波信号的频谱。研究结果表明:该算法使轴承IFCF的能量集中在初始值的位置,使用巴特沃斯带通滤波器便可以在包络信号中提取解调后的故障信息。该方法既克服了转速变化引起的"频率模糊"现象,又避免了时域滤波中共振频带难以确定的问题。仿真算例和应用实例证明了该算法的有效性。Considering that erroneous judgment and false negative occur in bearing multi-fault diagnosis because of interaction between different faults,and that it is difficult to obtain resonance band filtering parameters under the condition of variable speed and load,the method of rolling bearing multi-fault diagnosis based on envelope demodulation filter algorithm was proposed.Firstly,according to fault characteristic frequency coefficient of each part of the bearing and polynomial fitting function of rotating speed,instantaneous fault characteristic frequency(IFCF)trend was calculated.Secondly,phase functions were constructed based on fitting functions of the obtained IFCF trends,and the envelope signal obtained by Hilbert transform was demodulated.Finally,filtered signal was obtained from the narrow band Butterworth band-pass filter in which the initial value of IFCFs was used as centre frequency,and then frequency spectrum of demodulation filtering signal was obtained by fast Fourier transform(FFT).The results show that the algorithm makes energy of IFCFs concentrate in the position of initial value,and band-pass filter can be used to extract demodulated fault information in envelope signal.The proposed method not only overcomes the“frequency ambiguity”phenomenon caused by the change of speed,but also avoids the difficulty of determining resonance band in time domain filtering.The effectiveness of the proposed method is validated by simulation and experimental results.

关 键 词：变转速 滚动轴承 复合故障诊断 解调算法 瞬时故障特征频率 巴特沃斯带通滤波器 

分 类 号：TH165[机械工程—机械制造及自动化]